SOCIAL CAPITAL

SOCIAL CAPITAL

INTRODUCTION
Social capital has been given a number of different definitions, many of them refer to manifestations of social capital rather than to social capital itself. There are therefore numerous definitions of social capital found in the literature. 'Social capital can be defined simply as the existence of a certain set of informal values or norms shared among members of a group that permit cooperation among them' (Fukuyama 1997).Social capital is an instantiated informal norm that promotes cooperation between two or more individuals. The norms that constitute social capital can range from a norm of reciprocity between two friends, family members, all the way up to complex and elaborately articulated doctrines like Christianity, Muslim or Confucianism. They must be instantiated in an actual human relationship.
Social capital refers to the norms and networks that enable collective action. Increasing evidence shows that social cohesion is critical for poverty alleviation and sustainable human and economic development. The term social capital captures the idea that social bonds and social norms are important for sustainable livelihoods. Aspects of social structure and organization act as resources for individuals to use and realize their personal interests. Local institutions are effective because they permit the social group to carry on their daily lives with a minimum of repetition and costly negotiation (Bromley, 1993).
TYPES OF SOCIAL CAPITAL
Bonding Social Capital.
Bonding is horizontal, among equals within a community and localized, found among people who live in the same or adjacent communities. It is of thick trust (wider radius of trust)
Bridging Social Capital. Bridging social capital is vertical between communities, which extends to individuals and organizations that are more distant. It is of thin trust (a narrow radius of trust).,
IMPORTANCE OF SOCIAL CAPITAL
Social capital has facilitated a series of very important empirical investigations and theoretical debates which have stimulated reconsideration of the significance of human relations, of networks, of organizational forms for the quality of life and of developmental performance. As it lowers the costs of working together, social capital facilitates co-operation. People have the confidence to invest in collective activities, knowing that others will also do so. They are also less likely to engage in free private actions that result in negative impacts, such as resource degradation.
SOURCES OF SOCIAL CAPITAL
There are a number of key sources of social capital in the context of social and economic development.
1. Families: As the main source of economic and social welfare for its members, the family is the first building block in the generation of social capital for the larger society (Bubolz 1998, Hogan 1998). In addition to influencing the human capital development of children, the family's internal and external relationships model behaviors that are transmitted via children to future relationships.
Strong family systems are often found in places where the rule of law is weak, like Italy (Gambetta 1993) and India (Milner 1994). Communities: Social interactions among neighbors, friends and groups generate social capital and the ability to work together for a common good. This is especially important for the poor as social capital can be used as a substitute for human and physical capital.
2. Firms: Building and sustaining efficient organizations like firms demands trust and a common sense of purpose, i.e., social capital. Social capital benefits firms by reducing transactions costs, but can also have negative effects for a firm and society.
3. Civil Society: Social capital is crucial to the success of any non-governmental organization because it provides opportunities for participation and gives voice to those who may be locked out of more formal avenues to affect change.
4. Public Sector: That is the state and its institutions, is central to the functioning and welfare of any society. Good governance, a commitment to the welfare of citizens and the protection of their rights, fair and accountable institutions, well-established rule of law and citizen involvement all foster social and economic development. Considerable evidence links the type and effectiveness of a country’s public sector to society’s level of social cohesion, including the definition of civic duty and level of commitment to it (Esping-Andersen 1994, Putnam 1993).
5. Ethnicity: Ethnic relations come up frequently in discussions of social capital. Whether it is immigration, micro enterprise development, tribal favoritism or racial conflict, ethnic ties are a clear example of how actors who share common values and culture can band together for mutual benefit.
6. Gender: Social networks of impoverished women are important for women to obtain income and other necessities. Gender is a social construct placing cultural significance onto sexual identity. Since women are typically the primary care givers, they serve a critical role in the early development of social capital in a society. The individual’s capacity to trust has roots in the mother-child relationship (Picciotto 1998).
CENTRAL ASPECTS OF SOCIAL CAPITAL
Four central aspects have been identified (Pretty and Ward, 2001): i) relations of trust; ii) reciprocity and exchanges; iii) common rules, norms and sanctions; iv) connectedness, networks and groups.

i) Relations of trust
Trust lubricates co-operation. It reduces the transaction costs between people, and so liberates resources. Instead of having to invest in monitoring others, individuals are able to trust them to act as expected. This saves money and time. It can also create a social obligation - by trusting someone this engenders reciprocal trust. There are two types of trust: the trust we have in individuals whom we know; and the trust we have in those we do not know, but which arises because of our confidence in a known social structure. Trust takes time to build, but is easily broken (Gambetta, 1988; Fukuyama, 1995), and when a society is pervaded by distrust, cooperative arrangements are unlikely to emerge (Baland and Platteau, 1998).
ii) Reciprocity and exchanges
Reciprocity and exchanges also increase trust. There are two types of reciprocity (Coleman, 1990; Putnam, 1993). Specific reciprocity refers to simultaneous exchanges of items of roughly equal value; and diffuse reciprocity refers to a continuing relationship of exchange that at any given time may be unrequited, but over time is repaid and balanced. Again, this contributes to the development of long-term obligations between people, which can be an important part of achieving positive environmental outcomes (Platteau, 1997).
iii) Common rules, norms and sanctions
Common rules, norms and sanctions are the mutually agreed or handed-down norms of behaviour that place group interests above those of individuals. They give individuals the confidence to invest in collective or group activities, knowing that others will do so too. Individuals can take responsibility and ensure their rights are not infringed. Mutually-agreed sanctions ensure that those who break the rules know they will be punished.
These are sometimes called the rules of the game (Taylor, 1982), or the internal morality of a social system (Coleman, 1990), the cement of society (Elster, 1989), or the basic values that shape beliefs (Collins and Chippendale, 1991). They reflect the degree to which individuals agree to mediate or control their own behaviour. Formal rules are those set out by authorities, such as laws and regulations, while informal ones are those individuals use to shape their own everyday behaviour. Norms are, by contrast, preferences and indicate how individuals should act; rules are stipulations of behaviour with positive and/or negative sanctions. A high social capital implies high `internal morality’, with individuals balancing individual rights with collective responsibilities (Etzioni, 1995).
iv) Connectedness (networks and groups)
Connectedness, networks, and groups and the nature of relationships are a vital aspect of social capital. There may be many different types of connection between groups (trading of goods, exchange of information, mutual help, provision of loans, common celebrations, such as prayer, marriages, funerals). They may be one-way or two-way, and may be long-established (and so not responsive to current conditions), or subject to regular update.
High social capital implies a likelihood of multiple membership of organisations and links between groups. It is possible to imagine a context with large numbers of organisations, but each protecting its own interests with little cross-contact. Organisational density may be high, but inter-group connectedness low (Cernea, 1993). A better form of social capital implies high organisational density and cross-organisational links.
Connectedness, therefore, has five elements:
1) Local connections – strong connections between individuals and within local groups and communities.
2) Local-local connections – horizontal connections between groups within communities or between communities, which sometimes become platforms and new higher-level institutional structures.
3) Local-external connections – vertical connections between local groups and external agencies or organizations, being one-way (usually top-down) or two-way.
4) External-external connections – horizontal connections between external agencies, leading to integrated approaches for collaborative partnerships.
5) External connections–strong connections between individuals within external agencies.
CIVIL SOCIETY AND SOCIAL CAPITAL
Social Capital within a Non-Governmental Organization (NGO). Trust and willingness to cooperate allows people to form groups and associations, which facilitate the realization of shared goals.
Social Capital and Civil Society Can Promote Welfare and Economic Development. When the state is weak or not interested, civil society and the social capital it engenders can be a crucial provider of informal social insurance and can facilitate economic development.
Social Capital Across Sectors. State, market and civil society can increase their effectiveness by contributing jointly to the provision of welfare and economic development. The success of this synergy is based on complementary rather than substitutable inputs, trust, freedom of choice and incentives of parties to cooperate (Evans 1996, Ostrom 1996).
ADVANTAGES OF SOCIAL CAPITAL (BENEFITS FIRMS)
Social Capital reduces transaction costs. Firms benefit from social capital because it facilitates cooperation and coordination which minimize transaction costs, such as negotiation and enforcement, imperfect information and layers of unnecessary bureaucracy. Reciprocal, interdependent relationships--models of social capital--embody enforcement.
Social Capital provides a competitive edge. Efficiency gains in time and information allow more resources to be devoted to producing and marketing a better product at a higher volume. Trust based relations between economic agents have been seen as part of the competitive advantage of manufacturing enterprises.
Social capital promotes greater coordination among individuals and between departments. Teamwork can enhance efficiency and quality in small companies as well as multinational corporations. Social capital within and beyond the firm improve morale and enhance productivity (Smith 1994).
Trust is the foundation for cooperation between enterprises. Businesses banding together in a joint effort are able to establish deeper relationships with one another, which can be accessed in the future for other business projects beyond the scope of the original group.
Social capital enhances the importance of cross-sectoral partnerships for sustainable business as well as sustainable development.
Social capital affects the types of firms that are successful within a society (Fukuyama 1995, La Porta et al. 1997) and creates an enabling environment for private sector development.
DISADVANTAGE OF SOCIAL CAPITAL
Social capital can exclude outsiders. The same social ties which enable community members to work together can exclude outsiders.
Community pressure may be harmful to individuals. Enforceable group norms are not necessarily good for community members. Traditions can stifle individual growth and creativity, members who do not comply with norms and their families can be ridiculed or ousted from the community
Social ties may be harmful to communities and beyond. Communities with a lot of social capital, particularly if organized along ethnic lines or religious lines, can be harmful to each other and to society.
CONLUSION
Social capital, qualifies as capital because require some investment of time and effort and sometimes money. Social capital is similar to other forms of capital in that it can be invested with the expectation of future returns. The consequences of social capital are capital in nature because capital suggests something that is durable or long lasting and suggests something that retains its identity even after repeated use, something that can be used up, destroyed, maintained, or improved. Social capital is different from other forms of capital in that it resides in social relationships whereas other forms of capital can reside in the individual. Further, social capital cannot be traded by individuals on an open market like other forms of capital, but is instead embedded within a group. It is clear from the literature that social capital has both similarities and dissimilarities with neocapital theories and is certainly quite dissimilar from classical theory of capital.
REFERENCES:
Adler and Kwon. 1999 “Is Social Capital Really Capital. The Capital Debate”
Bourdieu. 1986 “The Social Construction of Literacy”
Gant, John, Casey Ichniowski, and Kathryn Shaw. 2002. 'Social capital and organisational change in high-involvement and traditional work organisations." Journal of Economics and Management 11: 289-328Gant et al. 2002; Glaeser
Glaeser, Edward L, David Laibson, and Bruce Sacerdote. 2002. 'An economic approach to social capital.' The Economic Journal 112: 437-458.
Robinson, David. 2000. "Social Capital in Action." Social Policy Journal of New Zealand: 185.
Robison, Lindon J., A. Allan Schmid, and Marcelo E. Siles. 2002. "Is social capital really capital?" Review of Social Economy 60: 1-24.

DEVELOPMENT&CLIMATE CHANGE IN TANZANIA

Organisation for Economic Co-operation and Development 2003
Organisation de Coopération et de Développement Economiques
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
ENVIRONMENT DIRECTORATE
DEVELOPMENT CO-OPERATION DIRECTORATE
Working Party on Global and Structural Policies
Workin
Organisation for Economic Co-operation and Development 2003
Organisation de Coopération et de Développement Economiques
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
ENVIRONMENT DIRECTORATE
DEVELOPMENT CO-OPERATION DIRECTORATE
Working Party on Global and Structural Policies
Working Party on Development Co-operation and Environment
DEVELOPMENT AND CLIMATE CHANGE
IN TANZANIA:
FOCUS ON MOUNT KILIMANJARO
by
Shardul Agrawala, Annett Moehner, Andreas Hemp, Maarten
van Aalst, Sam Hitz, Joel Smith, Hubert Meena,
Stephen M. Mwakifwamba, Tharsis Hyera
and Obeth U. Mwaipopo
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
2
Copyright OECD, 2003
Application for permission to reproduce or translate all or part of this material should be addressed to the
Head of Publications Service, OECD, 2 rue André Pascal, 75775 Paris, Cedex 16, France.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
3
FOREWORD
This document is an output from the OECD Development and Climate Change project, an activity being
jointly overseen by the Working Party on Global and Structural Policies (WPGSP) of the Environment
Directorate, and the Network on Environment and Development Co-operation of the Development Cooperation
Directorate (DAC-Environet). The overall objective of the project is to provide guidance on how
to mainstream responses to climate change within economic development planning and assistance policies,
with natural resource management as an overarching theme. Insights from the work are therefore expected
to have implications for the development assistance community in OECD countries, and national and
regional planners in developing countries.
This document has been authored by Shardul Agrawala and Annett Moehner. It draws upon four primary
consultant inputs that were commissioned for this country study: “Climate Impacts and Responses in
Mount Kilimanjaro” by Andreas Hemp (University of Bayreuth, Germany); “Review of Development
Plans, Strategies, Assistance Portfolios, and Select Projects Potentially Relevant to Climate Change in
Tanzania” by Maarten van Aalst of Utrecht University, The Netherlands; “Analysis of GCM Scenarios and
Ranking of Principal Climate Impacts and Vulnerabilities in Tanzania” by Stratus Consulting, Boulder,
USA (Sam Hitz and Joel Smith); and “Development and Climate Change in Tanzania” by the Center for
Energy, Environment, Science and Technology (CEEST), Dar es Salaam, Tanzania (Hubert Meena,
Stephen M. Mwakifwamba, Tharsis Hyera, and Obeth U. Mwaipopo).
In addition to delegates from WPGSP and DAC-Environet, comments from Tom Jones, Jan Corfee-
Morlot, Georg Caspary, and Remy Paris of the OECD Secretariat are gratefully acknowledged. Tomoko
Ota and Martin Berg provided project support at various times during the project. Shardul Agrawala would
like to acknowledge inputs on Kilimanjaro ice-field and regional climate patterns from Lonnie Thompson
(Ohio State University, USA), Jeanne Altman (Princeton University, USA), Douglas Hardy (University of
Massachusetts, USA) and Georg Kaser (University of Innsbruck, Austria). Andreas Hemp acknowledges
support for prior fieldwork in the Kilimanjaro from 1996-2002 from the Deutsche
Forschungsgemeinschaft, the UNEP project “Global Trends in Africa: the Case of Mt. Kilimanjaro” that
forms the basis for several major findings of the Kilimanjaro case study, and support from the Tanzanian
Commission for Science and Technology, the Chief Park Wardens of Kilimanjaro National Park, to the
Catchment Forest officers and to Mr. Mushi, Moshi. The Secretariat and Maarten van Aalst would like to
acknowledge several members of the OECD DAC who provided valuable materials on country strategies
as well as specific projects. Stratus Consulting would like to acknowledge inputs from Tom Wigley at the
National Center for Atmospheric Research (NCAR).
This document does not necessarily represent the views of either the OECD or its Member countries. It is
published under the responsibility of the Secretary General.
Further inquiries about either this document or ongoing work on sustainable development and climate
change should be directed to Shardul Agrawala of the OECD Environment Directorate:
shardul.agrawala@oecd.org, or Georg Caspary of the OECD Development Co-operation Directorate:
georg.caspary@oecd.org.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
4
TABLE OF CONTENTS
FOREWORD................................................................................................................................................. 3
EXECUTIVE SUMMARY ............................................................................................................................ 6
LIST OF ACRONYMS ................................................................................................................................. 8
1. Introduction ..................................................................................................................................... 9
2. Country background........................................................................................................................ 9
3. Climate: baseline climatology and climate change scenarios......................................................... 11
3.1 Current climate ......................................................................................................................... 11
3.2 Climate change and sea level rise projections .......................................................................... 12
4. Overview of impacts, vulnerabilities and adaptation responses..................................................... 14
4.1 Agriculture ............................................................................................................................... 15
4.2 Forests...................................................................................................................................... 15
4.3 Water resources........................................................................................................................ 16
4.4 Coastal resources ...................................................................................................................... 16
4.5 Human health........................................................................................................................... 17
4.6 Energy, industry and transport.................................................................................................. 17
4.7 Overview of adaptation responses ............................................................................................ 17
5. Attention to climate concerns in donor activities ........................................................................... 19
5.1 Donor activities affected by climate risks................................................................................. 20
5.2 Attention to climate risks in donor strategies............................................................................ 24
5.3 Climate risks in selected development programs and projects ................................................. 25
6. Attention to climate concerns in national planning........................................................................ 26
6.1 National Action Plan on Climate Change................................................................................. 27
6.2 National communications to international environmental agreements ..................................... 27
6.3 Poverty Reduction Strategy Paper (PRSP) ............................................................................... 28
6.4 Other national policies of relevance to climate change ............................................................ 28
7. Climate change and Mount Kilimanjaro ........................................................................................ 29
7.1 Climate, glaciers, and hydrology .............................................................................................. 30
7.2 Ecosystems, biodiversity and land tenure on Mount Kilimanjaro ............................................ 32
7.3 Climatic trends on Mount Kilimanjaro ..................................................................................... 35
7.4 Potential impacts of climatic changes: glacier retreat............................................................... 38
7.5 Potential impacts of climatic changes: enhancement of fire risk.............................................. 39
7.6 Other threats to the Mount Kilimanjaro ecosystem.................................................................. 45
7.7 Scenarios for 2020 with respect to fire impact ......................................................................... 46
7.8 Climate risks in perspective: shrinking glaciers versus enhanced fire risk............................... 47
8. Policy responses for Mount Kilimanjaro........................................................................................ 47
8.1 Policy responses to the shrinking ice cap.................................................................................. 47
8.2 Policy responses to general environmental threats ................................................................... 48
8.3 Policy responses to enhanced fire risk ...................................................................................... 48
8.4 Promotion of ecosystem friendly livelihood opportunities....................................................... 52
9. Concluding remarks ....................................................................................................................... 53
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
5
9.1 Differentiated adaptation strategy.................................................................................................... 54
9.2 Climate change and donor portfolios............................................................................................... 54
9.3 Attention to climate change concerns in national planning............................................................. 55
9.4 Climate risks in perspective on Mount Kilimanjaro........................................................................ 55
9.5 Policy responses for Mount Kilimanjaro......................................................................................... 56
APPENDIX A: PREDICTIVE ERRORS FOR SCENGEN ANALYSIS FOR TANZANIA...................... 57
APPENDIX B: LIST OF PURPOSE CODES INCLUDED IN THE SELECTION OF CLIMATEAFFECTED
PROJECTS, ORGANIZED BY THE DAC SECTOR CODE................................................ 58
APPENDIX C: REVIEW OF SELECTED DONOR STRATEGIES FOR TANZANIA............................ 59
APPENDIX D: REVIEW OF SELECTED DEVELOPMENT PROJECTS/PROGRAMS......................... 63
APPENDIX E: SOURCES FOR DOCUMENTATION .............................................................................. 65
REFERENCES ............................................................................................................................................ 67
Boxes
Box 1. A brief description of MAGICC/SCENGEN ............................................................................... 12
Box 2. Creditor Reporting System (CRS) Database ................................................................................ 21
Box 3. Flora of Mount Kilimanjaro.......................................................................................................... 32
Box 4. Fauna of Mount Kilimanjaro ........................................................................................................ 34
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
6
EXECUTIVE SUMMARY
This report presents the integrated case study for Tanzania carried out under an OECD project on
Development and Climate Change. The report is structured around a three-tiered framework. First, recent
climate trends and climate change scenarios for Tanzania are assessed, and key sectoral impacts are
identified and ranked along multiple indicators to establish priorities for adaptation. Second, donor
portfolios in Tanzania are analyzed to examine the proportion of donor activities affected by climate risks.
A desk analysis of donor strategies and project documents as well as national plans is conducted to assess
the degree of attention to climate change concerns in development planning and assistance. Third, an indepth
analysis is conducted for climate change impacts and response strategies for Mount Kilimanjaro – a
critical ecosystem, biodiversity hotspot, and source of freshwater. This part of the analysis draws upon
extended field research by a case study consultant in collaboration with national and international partners.
Analysis of recent climate trends reveals that climate change poses significant risks for Tanzania.
While projected changes in precipitation are uncertain, there is a high likelihood of temperature increases
as well as sea level rise. Climate change scenarios across multiple general circulation models show
increases in country averaged mean temperatures of 1.3°C and 2.2°C projected by 2050 and 2100, which
are broadly consistent, though lower than, projections used in Tanzania’s Initial National Communication.
The sectors potentially impacted by climate change include agriculture, forests, water resources, coastal
resources, human health, as well as energy, industry and transport.
While uncertainties in climate change and impact projections pose a challenge for anticipatory
adaptation in any country, Tanzania’s case has several specific characteristics that might suggest the need
for a differentiated adaptation strategy. First, the climate change projections which form the basis of
national assessments rely on an older generation of climate models which project higher temperature
increases than more recent models analyzed in the present study. Updating of climate scenarios and impact
projections through the use of multiple and more recent models might therefore be advisable prior to the
formulation of aggressive (and potentially expensive) adaptation responses. A second characteristic feature
of Tanzania is that certain sectors such as agriculture and water resources are projected to experience both
negative and positive impacts under climate change – for example, while production of maize is projected
to decline, the production of two cash crops (coffee and cotton) is projected to increase. The implication
for adaptation therefore may be to not only cushion adverse impacts, but also to harness positive
opportunities. A third key characteristic is that unlike most other countries where the need for adaptation
relies on projections of future impacts, some discernible trends in climate and attendant impacts are already
underway in Tanzania. Such impacts – as is the case of the Kilimanjaro ecosystem - argue for more
immediate adaptation responses as opposed to a “wait and see” strategy.
Tanzania receives close to a billion US dollars of Official Development Assistance (ODA)
annually. Analysis of donor portfolios in Tanzania using the OECD-World Bank Creditor Reporting
System (CRS) database reveals that between 12-25% of development assistance (by aid amount) or 20-
30% of donor projects (by number) are in sectors potentially affected by climate risks. However, these
numbers are only indicative at best, given that any classification based on sectors suffers from oversimplification
– the reader is referred to the main report for a more nuanced interpretation. Donor and
government documents generally do not mention climate change explicitly, although frequent references
are made to the impacts of climate variability and their linkages to economic performance. There is
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
7
however considerable synergy between priorities of at least some national plans and measures that might
be required for climate change adaptation, such as water conservation, improving agricultural resilience,
and forest conservation. However, some of these goals (such as water conservation) had been articulated,
though not successfully implemented in previous plans. Therefore, a key obstacle facing “mainstreaming”
is not synergies at the level of planning documents, but rather the successful implementation of such plans.
The in-depth sector analysis focuses on the climate change impacts and policy responses on the
Mount Kilimanjaro ecosystem. Glaciers on Mount Kilimanjaro have been in a general state of retreat on
account of natural causes for over a hundred and fifty years. Due to a decline in precipitation coupled with
a local warming trend that has been recorded in the second half of the twentieth century Kilimanjaro’s ice
cap is now projected to vanish entirely by as early as 2020. The symbolism of this loss is indeed
significant, and furthermore the loss of the ice cap would also imply that valuable records of past climates
contained in its ice cores would also be irreplaceably destroyed. From a physical and socio economic
perspective however, this analysis concludes that the impact of the loss of the ice cap is likely to be very
limited. Much more significant is the enhancement in the intensity and risk of forest fires on Mount
Kilimanjaro as a consequence of the increase in temperatures and a concomitant decline in precipitation
over the past several decades. Forest fires have resulted in the replacement of the fog intercepting
subalpine forest belt by low lying shrub which has already seriously impacted the hydrological balance of
the mountain as fog intercepting cloud forests play a key role in the water budgets of high altitude drainage
basins. A continuation of current trends in climatic changes, fire frequency, and human influence could
result in the loss of most of the remaining subalpine Erica forests in a matter of years. With this, Mount
Kilimanjaro will have lost its most effective water catchment. Among the more immediate adaptation
responses identified by this report are institutional measures such as the inclusion of the forest belt into the
Kilimanjaro National Park and the creation of a paramilitary ranger group to deter logging, as well as better
investments in early warning systems, particularly the purchase of one or two aircraft for aerial
surveillance. There is also a need to limit cross-border migration of big game from neighbouring Amboseli,
which is adding to the stress on the Kilimanjaro ecosystem. In addition to short term solutions there is a
critical need to develop a comprehensive and holistic development plan focusing on fire-risk and forest
destruction, livelihood needs of the local population as well as on conservation strategies to ensure the long
term sustainability of the valuable resources of the Kilimanjaro ecosystem.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
8
LIST OF ACRONYMS
AfDB
AMA
asl
AVVA
CBO
CCCM
CEEST
CERES-Maize
COMPACT
CRS
DAC
DFID
EACC
FAO
FFYP
GCA
GCM
GDP
GEF
GHG
GMBA
GNP
GNI
IDA
IFAD
IPCC
KINAPA
MW
NEAP
NEP
NCAA
NGO
ODA
PRSP
SIDA
TAF
TANAPA
TFYP
UN
UNCB
UNCCD
UNCED
UNDAF
UNDP
UNEP
UNESCO
UNF
UNFCCC
USAID
USCSP
WHO
WNHS
African Development Bank
African Mountain Association
Above Sea Level
Aerial Videotape-assisted Vulnerability Analysis
Community Based Organization
Canadian Climate Centre Model
Centre for Energy, Environment, Science and Technology
Crop Environment Resource Synthesis model
Community Management of Protected Areas Conservation Project
Creditor Reporting System of the OECD/World Bank
Development Assistance Committee
Department for International Development
East African Coastal Current
Food and Agriculture Organization of the United Nations
First Five Year Plan
Game Controlled Areas
General Circulation Model
Gross Domestic Product
Global Environment Facility
Greenhouse Gases
Global Mountain Biodiversity Assessment
Gross National Product
Gross National Income
International Development Assistance
International Fund for Agricultural Development
Intergovernmental Panel on Climate Change
Kilimanjaro National Park
Mega Watt
National Environmental Action Plan
National Environmental Policy
Ngorongoro Conservation Authority Area
Non Governmental Organization
Official Development Assistance
Poverty Reduction Strategy Papers
Swedish International Development Agency
Tanzanian Association of Foresters
Tanzanian National Parks
Third Five Year Plan
United Nations
United Nations Convention on Biodiversity
United Nations Convention to Combat Desertification
United Nations Conference on Environment and Development
United Nations Development Assistance Framework
United Nations Development Programme
United Nations Environment Programme
United Nations Educational, Scientific and Cultural Organization
United Nations Foundation
United Nations Framework Convention on Climate Change
The United States Agency for International Development
United States Country Studies Program
World Health Organization
World Natural Heritage Sites
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
9
1. Introduction
This report presents the integrated case study for Tanzania for the OECD Development and
Climate Change Project, an activity jointly overseen by the Working Party on Global and Structural
Policies (WPGSP), and the Working Party on Development Co-operation and Environment (WPENV).
The overall objective of the project is to provide guidance on how to mainstream responses to climate
change within economic development planning and assistance policies, with natural resource management
as an overarching theme. The Tanzania case study was conducted in parallel with five other country case
studies1 in Africa, Latin America, and Asia and the Pacific.
Each case study is based upon a three-tiered framework for analysis (Agrawala and Berg 2002).
1. Review of climate trends and scenarios at the country level based upon an examination of results
from seventeen recent general circulation models, as well as empirical observations and results
published as part of national communications, country studies, and scientific literature. These
projections are then used in conjunction with knowledge of socio-economic and sectoral
variables to rank key sectoral and regional impacts on the basis of a number of parameters. The
goal of this tier is to present a framework to establish priorities for adaptation.
2. Review of economic, environmental, and social plans and projects of both the government and
international donors that bear upon the sectors and regions identified as being particularly
vulnerable to climate change. The purpose of this analysis is to assess the degree of exposure of
current development activities and projects to climate risks, as well as the degree of current
attention by the government and donors to incorporating such risks in their planning. This section
will review donor portfolios and projects, as well as development priorities of the Government of
Tanzania to determine the degree of attention to potential risks posed by climate change on
relevant sectors.
3. In-depth analyses at a thematic, sectoral, regional or project level on how to incorporate climate
responses within economic development plans and projects, again with a particular focus on
natural resource management. In the case of Tanzania this case study provides an overview of
critical impacts and mainstreaming challenges for a number of sectors. This is followed by an indepth
analysis on Mount Kilimanjaro – a UNESCO World Heritage Site and also a critical
ecosystem and source of freshwater resources for Tanzania. The analysis on climate change
impacts and response strategies for the Mount Kilimanjaro ecosystem draws upon field research
over an extended period by a case study consultant in collaboration with national and
international partners.
2. Country background
Tanzania is located in East Africa, on the Indian Ocean bordered by Kenya to the north and
Mozambique to the south (Figure 1). It has an area of 945,000 km2 which includes the three major coastal
islands of Mafia, Pemba, and Zanzibar, and a coastline that is about 800 km long. The geography is
characterized by plains along the coast, a central plateau, and highlands in the north and south. The
northwest of the country encompasses approximately one-half of Lake Victoria, the second largest body of
freshwater in the world, and the western and southwestern borders abut the comparably massive Lake
Tanganyika and Lake Nyasa. Elevations range from sea level to the highest point in Africa, the glaciated
peak of Kilimanjaro at 5,895 m, the expansive slopes of which constitute one of the unique ecosystems of
1 Bangladesh, Egypt, Uruguay, Fiji, and Nepal.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
10
Africa. Tanzania also includes the Serengeti, the site of one of the last major terrestrial mammalian
migrations in the world and a prominent tourist destination.
Figure 1. Map of Tanzania
Tanzania is one of the poorest countries in the world with a GNI per capita of only US $ 280
(World Bank 2002). Gross national income per capita over the period 1994-2000 stood at about US$270
compared to US$470 for sub-Saharan Africa in general. Some 42% of the total population and 50% of the
rural population live below the poverty line, according to a 1993 survey, with 20% of the entire population
surviving on less than US$1 per day (World Bank, 2002). Based on the same 1993 survey, the Gini
Coefficient2 for Tanzania is 0.38, with the poorest 10% accounting for 2.8% of the national income and the
richest 10% accounting for 30.1%. According to World Bank estimates, Tanzania’s population in 2000 was
33.7 million, and growing at 1.8% a year. Average Life-expectancy is only 43.1 years (World Bank 2002).
While an overwhelming proportion of the population still lives in rural areas, by the late 1990s, 27.8% of
the country’s population lived in an urban setting, up from only 10.1% in 1975. Tanzania’s economy is
heavily dependent on agriculture, which accounts for nearly one-half of GDP, employs 80% of the work
force, and provides 85% of exports (World Bank, 2002). Topography and climatic conditions, however,
limit cultivated crops to only 4% of the land area. Industry has traditionally been limited to the processing
of agricultural products and light consumer goods. However, with a significant infusion of funds from the
World Bank, International Monetary Fund, and bilateral donors, growth over the last decade has featured
2 The Gini coefficient is a number between zero and one that measures the degree of inequality in the
distribution of income in a given society. The coefficient would register zero inequality for a society in
which each member received exactly the same income and it would register a coefficient of one (maximum
inequality) if one member got all the income and the rest got nothing.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
11
an increase in industrial production and a substantial increase in output of minerals (CIA, 2002). Private
sector growth and investment have also increased and, coupled with donor aid and liberal macroeconomic
policies, should support continued growth of about 5% (World Bank, 2002).
Economic growth could play an important role in increasing the capacity of a country like
Tanzania to adapt to climate change. However, the current state of its infrastructure and educational system
is likely an impediment to Tanzania’s ability to cope effectively with climatic risks. In 2000, only 4.2% of
Tanzania’s road network was paved, compared to 16.5% for low income countries in general. Further,
while 37% of tertiary level students were enrolled in science and engineering programs between 1987 and
1997, gross tertiary enrolment stood at only 0.66% by 1997 (World Bank, 2002). Similarly, gross
secondary enrolment was 6.5%. Adult literacy was 24.9% in 2000. Figure 2 provides an indication of how
Tanzania compares to other low income countries in terms of four key indices of development. On all four
measures of development, Tanzania ranks considerably below the average for low income countries.
Figure 2. Development diamond for Tanzania
Tanzania
Low-income group
Development diamond
Life expectancy
Access to improved water source
GNI
per
capita
Gross
primary
enrollment
Source: World Bank 2002
3. Climate: baseline climatology and climate change scenarios
This section briefly reviews projections of temperature and precipitation change for Tanzania
from climate models, and then provides a synthesis of key climate change impacts and vulnerabilities.
3.1 Current climate
Tanzania’s climate ranges from tropical to temperate in the highlands. Average annual
precipitation over the entire nation is 1,042 mm. Average temperatures range between 17°C and 27°C,
depending on location. Natural hazards include both flooding and drought. Within the country, altitude
plays a large role in determining rainfall pattern, with higher elevations receiving more precipitation.
Generally speaking, the total amount of rainfall is not very great. Only about half the country receives
more than 762 mm annually (Mwandosya et al., 1998). Tanzania’s precipitation is governed by two rainfall
regimes. Bimodal rainfall, comprised of the long rains of Masika between March-May and short rains of
Vuli between October-December, is the pattern for much of the northeastern, northwestern (Lake Victoria
basin) and the northern parts of the coastal belt. A unimodal rainfall pattern, with most of the rainfall
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
12
during December-April, is more typical of most of the southern, central, western, and southeastern parts of
the country.
3.2 Climate change and sea level rise projections
Changes in area averaged temperature and precipitation over Tanzania were assessed using
outputs from over a dozen recent (post 1995) GCMs which are processed using a new version of
MAGICC/SCENGEN. MAGICC/SCENGEN is briefly described in Box 1. First, results for Tanzania for
17 GCMs developed since 1995 were examined. Next, 11 of 17 models which best simulate current
climate over Tanzania were selected. The models were run with the IPCC B2 SRES scenario (Nakicenovic
and Swart 2000)3.
Box 1. A brief description of MAGICC/SCENGEN
MAGICC/SCENGEN is a coupled gas-cycle/climate model (MAGICC) that drives a spatial climate-change
scenario generator (SCENGEN). MAGICC is a Simple Climate Model that computes the mean global surface air
temperature and sea-level rise for particular emissions scenarios for greenhouse gases and sulphur dioxide (Raoer et
al., 1996). MAGICC has been the primary model used by IPCC to produce projections of future global-mean
temperature and sea level rise (see Houghton et al., 2001). SCENGEN is a database that contains the results of a
large number of GCM experiments. SCENGEN constructs a range of geographically-explicit climate change scenarios
for the world by exploiting the results from MAGICC and a set of GCM experiments, and combining these with
observed global and regional climate data sets. SCENGEN uses the scaling method of Santer et al. (1990) to produce
spatial pattern of change from an extensive data base of atmosphere ocean GCM – AOGCM (atmosphere ocean
general circulation models) data. Spatial patterns are “normalized” and expressed as changes per 1°C change in
global-mean temperature. The greenhouse-gas and aerosol components are appropriately weighted, added, and
scaled up to the actual global-mean temperature. The user can select from a number of different AOGCMs for the
greenhouse-gas component. For the aerosol component there is currently only a single set of model results. This
approach assumes that regional patterns of climate change will be consistent at varying levels of atmospheric
greenhouse gas concentrations. The MAGICC component employs IPCC Third Assessment Report (TAR) science
(Houghton et al., 2001). The SCENGEN component allows users to investigate only changes in the mean climate state
in response to external forcing. It relies mainly on climate models run in the latter half of the 1990s.
Source: National Communications Support Program Workbook
The spread in temperature and precipitation projections of these 11 GCMs for various years in
the future provides an estimate of the degree of agreement across various models for particular projections.
More consistent projections across various models will tend to have lower scores for the standard
deviation, relative to the value of the mean. The results of the MAGICC/SCENGEN analysis for Tanzania
are shown in Table 1.
3 The IPCC SRES B2 scenario assumes a world of moderate population growth and intermediate level of
economic development and technological change. SCENGEN estimates a global mean temperature
increase of 0.8 °C by 2030, 1.2 °C by 2050, and 2 °C by 2100 for the B2 scenario.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
13
Table 1. GCM estimates of temperature and precipitation changes4
Year
Temperature change (°C)
mean (standard deviation)
Precipitation change (%)
mean (standard deviation)
Tanzania Annual JJA5 SON6 DJF7 MAM8 Annual JJA SON DJF MAM
2030 0.9 (0.20) 1.0
(0.21)
.8
(0.17)
.8
(0.30)
0.9
(0.30)
4.1
(5.05)
-2.4
(7.98)
3.9
(10.04)
6.6
(8.06)
2.2
(5.34)
2050 1.3 (0.28) 1.5
(0.31)
1.2
(0.25)
1.1
(0.43)
1.3
(0.44)
5.9
(7.30)
-3.5
(11.53)
5.6
(14.51)
9.6
(11.64)
3.1
(7.72)
2100 2.2 (0.49) 2.6
(0.54)
2.1
(0.43)
1.9
(0.75)
2.3
(0.77)
10.2
(12.70)
-6.0
(20.07)
9.7
(25.27)
16.7
(20.27)
5.4
(13.44)
The results indicate that mean annual temperatures are projected to rise by 2.2 C by 2100, with
somewhat higher increases (2.6 °C) over June, July and August, and lower values (1.9 °C) for December,
January, February. Low standard deviations relative to the mean indicate good agreement across the 11
models. The Initial National Communication of Tanzania (2003) projects a temperature increase between
3-5 °C under doubling of carbon dioxide, which is benchmarked to the year 2075. The lower estimates of
MAGICC/SCENGEN are likely from the use of more recent scenarios (SRES) and multiple (17), more
recent (post 1995) GCMs with a better treatment of aerosols in the MAGICC/SCENGEN analysis. The
Tanzania National Communication meanwhile relied on four earlier generation models (primarily the
UK1989), as well as older (unspecified) emissions scenarios. Both sets of analyses however show
temperature increases, and furthermore the patterns of seasonal temperature increase are consistent.
Specifically, greater warming is projected for the cooler months (June-August) compared to the warmer
months (December-February).
In terms of precipitation meanwhile, according to the MAGICC/SCENGEN analysis annual
precipitation over the whole country is projected to increase by 10% by 2100, although seasonal declines
of 6% are projected for June, July and August, and increasers of 16.7% for December, January, February.
However, high standard deviations are indicative of low confidence in these projections across the various
models. Furthermore, the precipitation regimes across Tanzania vary considerably, as discussed in the
preceding section. Therefore country averaged values for precipitation, as is done in the
MAGICC/SCENGEN analysis, are of limited utility9. The Tanzania Initial National Communication does
offer greater regional specificity – although the results should be interpreted with caution as they do not
include an uncertainty analysis and rely on one or two older climate models. Under a doubling carbon
dioxide scenario some parts of Tanzania are projected to experience increases in annual rainfall, while
4 This analysis uses a combination of the 11 best SCENGEN models (BMRCTR98, CCSRTR96,
CERFTR98, CSI2TR96, CSM_TR98, ECH3TR95, ECH4TR98, GFDLTR90, HAD2TR95, HAD3TR00,
PCM_TR00) based on their predictive error for annual precipitation levels. Errors were calculated for each
of the models, and for an average of the 17 models. Each model was ranked by its error score, which was
computed using the formula 100*[(MODEL-MEAN BASELINE / OBSERVED) - 1.0]. Error scores
closest to zero are optimal. The error score for an average of the 17 models was 30%, and the error score
for an average of the 11 models was 21%. See the appendix for details.
5 June July August
6 September October November
7 December January February
8 March-April-May
9 A higher resolution analysis across multiple GCMs was beyond the scope of this study.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
14
precipitation is projected to decline in other areas (see Figure 3). However, the timing of these changes
might vary from location to location as well. The National Vulnerability and Adaptation Assessment of
Tanzania (Mwandosaya et al. 1998, which is the bases for the Initial National Communication of 2003)
estimates that northern and southeastern sectors of the country would experience an increase in rainfall
ranging from between 5% and 45% under doubling of carbon dioxide. The central, western, southwestern,
southern, and eastern parts of the country might experience a decrease in rainfall of 10% to 15%. The
southern highlands might similarly experience a decrease of 10%, which could alter the suitability of this
area for maize cultivation. Seasonal patterns in possible changes in rainfall could be complex. For instance,
the northeastern sector might experience an increase of 25%-60% in the short rains and an increase of 20-
45% in the long rains. Or, the north coastal region might get an increase of 0-20% in the short rains and a
decrease of 0-10% in the long rains. In the unimodal region, rainfall might decrease between 0% and 25%
in central regions during October, November, and December, but increase by 15% in March, April, and
May. Finally, the southeastern sector could get between 5 and 45% increase in rainfall during the first three
months of the season and in increase of 10-15% during the last three months.
Figure 3. Change in mean annual rainfall (in %) under 2XCO2
Source: Mwandosaya et al. 1998
The Tanzania National Vulnerability and Adaptation Assessment (1998) as well as the Initial
national Communication (2003) do not include sea-level rise scenarios for Tanzania’s 800 km coastline.
Tide gauge records in Tanzania cover only a very short period of time with some missing data. Instead, a
coastal vulnerability assessment is conducted under two arbitrary sea level rise scenarios 50cm and 1m,
coupled with aerial videotape assisted mapping of coastal topography, resources, and land use. Given the
most recent IPCC assessment (the Third Assessment Report), the 50cm scenarios roughly falls in the
middle, and the 1m scenarios a little beyond the upper estimate of the range of global sea level rise (9cm-
88cm) projected to occur by 2100.
4. Overview of impacts, vulnerabilities and adaptation responses
Given the large size and widely different climatology and climate change projections and impacts
across Tanzania, a national priority ranking might conflate intra-sectoral or sub-national positive and
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
15
negative effects of climate change, and thereby produce misleading results. Therefore, this synthesis
highlights the spectrum of possible sectoral and or regional impacts, and identifies critical impacts and
vulnerabilities, but without an aggregate sectoral or regional ranking. The section concludes with a
discussion of adaptation strategies and priorities for adaptation.
4.1 Agriculture
Agriculture is clearly the most important sector of the Tanzanian economy. It comprised 45.1%
of GDP in 2000 (World Bank, 2002). Upwards of 80% of the population of the country relies directly on
agriculture of one sort or another for their livelihood. Only 3.3% of the cropland was irrigated as of 1999
(World Bank, 2002). The three most important crops are: maize, coffee and cotton – with maize being a
major food staple, coffee a major cash crop grown in large plantations (and contributing significantly to the
GNI), while cotton is another cash crop grown largely by smallholder farmers.
Estimates of the affect of climate change on maize yields are available from model runs of the
Crop Environment Resource Synthesis model (CERES-Maize) (Jones and Kiniry, 1986). In general,
simulation results show that maize yields were lower, a result of higher temperatures and, where
applicable, decreased rainfall. The average yield decrease over the entire country was 33%, but simulations
produced decreases as high as 84% in the central regions of Dodoma and Tabora. Yields in the
northeastern highlands decreased by 22% and in the Lake Victoria region by 17%. The southern highland
areas of Mbeya and Songea were estimated to have decreases of 10-15%. These results suggest that climate
change may significantly influence future maize yields in Tanzania, reducing them in all zones that were
studied, relative to baseline levels. These reductions are due mainly to increases in temperature that shorten
the length of the growing season and to decreases in rainfall. Consequently, the continued reliance on
maize as a staple crop over wide areas of the country could be at risk. The two cash crops on the other hand
are projected to experience increases in yield (Tanzania Initial National Communication 2003). For
Lyamunugu, located within an area of bimodal rainfall, coffee yields are expected to increase by 18%, and
for Mbozi, where rainfall is unimodal, the coffee yield is expected to increase by 16%. These yield
estimates depend critically on estimates of change in precipitation. The potential impacts of climate change
on cotton production in Tanzania parallel that for coffee. The agriculture sector thus may have both
negative and positive impacts that could partially offset each other. However, maize production in
particular might require particular attention for adaptation and mainstreaming responses, given that it is a
critical food crop.
4.2 Forests
Tanzania has about 338,000 km2 under forest cover, which represents about 44% of the total land
area. These forests are an important source of fuel wood and other products for large numbers of
Tanzanians. Furthermore, many of Tanzania’s 43 threatened mammal species, 33 threatened bird species,
and prodigious biodiversity depend on its forests (World Bank 2002). Under climate change most of the
forests across Tanzania are projected to shift towards drier regimes: from subtropical dry forest, subtropical
wet forest, and subtropical thorn woodland to tropical very dry forest, tropical dry forest, and small areas
of tropical moist forest respectively (Tanzania Initial National Communication 2003). Much of this
projected change in distribution is attributed to an increase in ambient temperatures and a decline in
precipitation in forested regions of the country.
Current assessments of climate change impacts on forests in Tanzania however do not explicitly
account for the potential effects of climate change on disturbances such as fire. The Kilimanjaro region
deserves particular attention. In addition to the well-known glacier retreat and eventual disappearance of
the ice cap, there might be major changes in the extent, distribution, and species composition of the forests
on the Kilimanjaro as a consequence of changes in fire regimes. There is indication that intensification of
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
16
fire risk as a result of warmer and drier conditions might already be underway. Continued loss of the
montane forest belt (which collects a significant amount of water from fog entrapment) from fire
intensification would lead to a significant reduction of water yields with serious regional implications,
affecting sectors such as agriculture and livestock as well. These issues as well as possible responses on the
Kilimanjaro are the focus of in-depth analysis later in this report.
4.3 Water resources
Like the agriculture sector, climate change is projected to have both positive and negative
consequences for Tanzania’s water-resources, specifically for the three major river basins: Ruvu, Pangani,
and Rufiji. The Ruvu basin, of particular importance because it is upstream of Tanzania’s major population
center, Dar es Salaam, could experience a 10% decrease in runoff according to the Initial National
Communication (2003). The Pangani basin supplies water to the Tanga, Kilimanjaro, and Arusha regions,
supporting a number of economically important activities. These include the Arusha Chini sugar
plantations in the Kilimanjaro region, the lower Moshi irrigation scheme, the Handeni District water
supply, and a number of important power stations. For the Pagani River, there is some seasonal variation
with runoff projected to increase in some months runoff and decrease in others, with annual basin runoff
decreasing by an estimated 6%. However, the Kikuletwa River, also within the Pagani Basin, is projected
to decrease in all months, with annual reductions of 9%. The Rufiji basin meanwhile is a large catchment
in the south of the country, focused on the Great Ruaha River, which is economically important to the
nation in part because of the hydropower it generates at Mtera Dam and Kidatu Dam. The national
assessment of vulnerability and adaptation (Mwandosaya et al. 1998) projects increases in annual runoff of
5% and 11% at Mtera and Kidatu, respectively, most coming in the period from November to March. All
these estimates however are based on scenarios from a single GCM, and should be interpreted with some
caution. Real uncertainties exist concerning present and future withdrawals for irrigation, changed land
use, and urbanization. Nevertheless, decreases in runoff could potentially have serious affects on
socioeconomic activities in the regions of Dar es Salaam, Morogoro, Tanga, Coast, and Kilimanjaro. Dar
es Salaam might be particularly vulnerable because it is the largest industrial, commercial, and
administrative city in Tanzania.
4.4 Coastal resources
The coastline of Tanzania is about 800 km long and the coastal zone varies in width from 20 km
to 70 km gradually rising to a plateau. Tanzania has relatively limited coastal lowlands, but there are
extensive coastal wetlands, some important cities (Dar es Salaam), a number of important islands (such as
Zanzibar), and a delta — the Rufigi River (Mwaipopo 2001). The main coastal features include mangrove
forests and swamps, coral reefs, sand and mudflats, tidal marshes, woodland, and sisal and cashew nut
estates. Mangrove forests in particular represent an important economic resource for coastal people,
supplying firewood and timber for the construction of fishing boats, and providing feeding, breeding, and
nursery grounds for a number of fish species and a variety of insects, birds, and small animals. The highest
densities of population that might be threatened are found near Dar es Salaam and the islands of Zanzibar
and Pemba.
The Initial National Communication of Tanzania (2003) considers scenarios of both 0.5 m and 1
m sea level rise over the next century. Maps with a 2 m and 20 m contour were used and it was assumed
that land rises linearly from sea level to these contours. The 0.5 m and 1 m contours and the land area they
represent were approximated. Estimates of land lost to erosion were also produced with the aid of aerial
videotape-assisted vulnerability analysis. Total land-loss is estimated to be 247 km2 and 494 km2 for 0.5
and 1 meters of sea level rise respectively. According to this analysis the Dar es Salaam region would be
vulnerable with values of structures at risk estimated to total US$ 48 million for a 0.5 m sea level rise and
US$82 million for a 1 m rise (Tanzania Initial National Communication 2003).
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
17
4.5 Human health
Climate plays an important role in the geographical distribution and seasonal abundance of vector
species that are responsible for the transmission of a number of human diseases. Changes in temperature,
precipitation, humidity, and wind patterns will directly affect vector species’ reproduction, development,
and longevity. The distribution of vector borne diseases in the human population is also limited by
temperature in many regions where the climate is too cold for parasite survival (Martens et al. 1999). Of
the various vector borne diseases malaria in particular is a major public health concern in Tanzania. It
accounts for 16.7% of all reported deaths in Tanzania and is one of the leading causes of morbidity in all
regions, ranging from 24.4% in Rukwa regions to 48.9% in Dar es Salaam (Tanzania Initial National
Communication 2003). Also, the problem of malaria is getting worse because of growing parasite
resistance to first line anti-malarial drugs and mosquito resistance to insecticides. Malaria is endemic in
most of Tanzania even under the current climate. However, many population centers are located in areas
where malaria transmission is currently only epidemic or nonexistent. Most of these centers are located in
the central highlands region (e.g., Mbeya, Njombe, Iringa, and Arusha), where cooler temperatures prevent
or interrupt the transmission of malaria. These areas are of particular concern in considering a warmer
climate. Increased temperatures might open new areas to seasonal or year-around transmission. The
vulnerability of highland populations to an increase in the endemicity of transmission of malaria, or of any
of Tanzania’s population to climate change induced health risks, will depend strongly on the evolution of
control methods and the ability of Tanzania to afford such measures (Tol and Dowlatabadi, 2002).
4.6 Energy, industry and transport
Climate change may also have direct and/or indirect effects on Tanzania’s energy, industry, and
transportation sectors. Among the direct effects, an increase in temperatures would likely increase energy
demands for cooling. Areas projected to have declines in precipitation and or stream flow are also likely to
face increased demands for purposes such as irrigation. However, as highlighted by the discussion on
climate change scenarios, the projections for changes in precipitation remain highly uncertain. The
Tanzania country study also projects decline in stream flow in two key river basins (as discussed in Section
4.3), which will not only increase energy demands for irrigation, but more significantly adversely impact
energy supply, given that these two basins are significant contributors to Tanzania’s hydroelectric
generation. Transportation infrastructure such as railways, roads, pipelines and ports may also be at risk
from impacts of climate change (particularly sea level rise), but specific vulnerability analyses are lacking.
Other potential impacts of climate change on energy supply include the vulnerability of the Songo Songo
and Mnazi Bay natural gas reserves to sea level rise.
4.7 Overview of adaptation responses
While uncertainties in climate change and impacts projections are a characteristic feature that
poses a challenge for anticipatory adaptation for any country, Tanzania’s case has several characteristics
that might argue for a differentiated adaptation strategy. First, the climate change projections on which all
national impact and vulnerability assessments are based (all the way to the Initial National Communication
of 2003) rely on a limited number of older generation of climate models and scenarios, circa early 1990s
which has several implications for assessment of impact and adaptation options. For example, an analysis
based on more recent climate models conducted as part of this study concludes that the magnitudes of
temperature increases projected for Tanzania might be somewhat lower (though the trends are broadly
consistent) with the projections used in the National Assessment of Vulnerability and Adaptation. Thus,
information on impacts might need updating in Tanzania prior to the formulation of aggressive adaptation
responses, more so than in other countries where projections might be based on more recent models.
Second, some key sectors are projected to experience both positive and negative impacts under climate
change – for example, while production of maize is projected to decline, the production of two key cash
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
18
crops (coffee and cotton) is projected to increase. Similarly, while stream-flows are projected to decline in
two of three key river basins (Ruvu and Pangani), they are projected to increase in the third (Rufiji). The
implication for adaptation therefore might be to not only cushion adverse impacts, but also to harness
positive opportunities. Finally, a third key characteristic is that unlike most other countries where the need
for adaptation relies largely on projections of future impacts, there might be some discernible trends in
climate and attendant impacts already underway in Tanzania. This might argue for more immediate
adaptation measures in the case of such impacts as opposed to a “wait and see” strategy.
For all the above reasons, there might be a need for a differentiated adaptation strategy across
various sectors and regions depending upon the certainty of projections, the mix of beneficial and adverse
impacts, and the urgency and timing of such impacts. For the case of agriculture a key portfolio of
adaptation responses would involve measures that boost maize production: increased irrigation, increased
use of manure and fertilizer, and better use of management tools including climate information. These
measures are discussed in Tanzania’s Initial National Communication. However, given that the production
of the country’s two cash crops (coffee and cotton) is projected to increase under the same climate
scenarios, another adaptation response – which is not discussed in Tanzania’s Initial National
Communication - might involve a strategic shift over the medium to the long term from maize towards
these cash crops.
With regard to human health, the spread of malaria to the population centers in the highlands as a
result of rising temperatures is a key concern. Much of Tanzania however is already malaria endemic, so
policy responses might need to be driven by the additionality of the disease burden, and not necessarily the
existence of the risk itself. Most roll back malaria programs function in the reactive mode (antimalarial
drugs, spraying of insecticides, and elimination of breeding sites), while in the cases of highland areas
precautionary adaptation to prevent or limit the spread of malaria to these regions might be ideal.
For coastal resources meanwhile a key priority is to construct regional sea level rise scenarios,
that not only incorporate local topography (as has been done) but also subsidence rates. Lacking such
specific information, and given the long time-scales at which sea level rise will manifest itself, an initial set
of adaptation priorities should ideally focus on no regrets measures in particularly low lying or otherwise
vulnerable areas including urban areas as well as coastal wetlands and mangroves, such as the Mafia Island
Marine Park, the Menai Bay Conservation Area, and the Misali Island Conservation Area. Coastal zones
adaptation priorities may also be synergistic with several ongoing government-donor initiatives including
the Conservation of Lowland Coastal Forests Project, the Sustainable Dar es Salaam Project and the
Tanzania Coastal Management Partnership.
No regrets adaptation - specifically water and energy conservation – could be a viable initial
priority for adaptation in water resources, where stream-flow is projected to decline in two critical river
basins (Ruvu and Pangani), affecting water use and hydroelectricity generation. The Tanzania Initial
National Communication identifies privatization (as is already the case for Dar es Salaam) as a key
adaptation response to promote efficient water use. This measure however may have equity effects as it
may result in an increase in price of water making it unaffordable to the poor. Further, given that roughly
half of the water in Dar es Salaam is lost to leakage, a second key no regrets response would be leakage
prevention – although it would require significant new capital investment and regular maintenance of water
infrastructure. Third, given that streamflow in a third river basin – the Rufiji – is projected to increase,
another adaptation strategy may revolve around water transfer from Rufiji to Dar es Salaam which relies on
the Pangani. However, given that streamflow projections are based upon the results from one water balance
model (and dated climate scenarios), and the fact that the costs and environmental impacts of inter-basin
transfers are yet to be analyzed, such a response may not be advisable at this time.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
19
With regard to the energy, industry and transportation sectors, the Tanzania National Action Plan
(1997) has conducted a hierarchical screening of potential adaptation responses. Given that climate change
is projected to impact energy demands (through rise in temperature), as well as particular sources of energy
supply (hydro and to some extent natural gas fields located in coastal areas), these adaptation options focus
on either demand side management or on the promotion of energy supply sources that are not impacted by
climate change. Figure 4 shows the results of various adaptation responses from this hierarchical screening.
A majority of these measures are no-regrets. However some measures – particularly a fuel switch to
kerosene – may run into conflict with greenhouse gas mitigation, as they may imply a switch away from an
energy source of lower carbon intensity (natural gas and hydro). Therefore, synergy between mitigation
and adaptation responses, as well as with other development priorities must also be considered in screening
adaptation measures.
Figure 4. Screening of adaptation measures in the energy, industry and transportation sectors
Ranking of Adaptation Measures
0
10
20
30
40
50
60
70
80
90
100
Measures
Weighted Points
Power load shedding
Fuel switch to kerosene
Fuel switch to other fuels
Interconnection
Emergency Power Plants
Minihydro and geothermal plants
Demand Side Management
Efficiency in industry
Change of Products
Import off-road vehicles
Railway Maintenance
Road maintenance
Source: National Action Plan on Climate Change 1997
One area where the need for adaptation may be immediate is the Kilimanjaro ecosystem where
climatic changes are likely already contributing to significant impacts on the natural and human system,
including the intensification of fire risk, in part a consequence of observed changes in temperature and
precipitation patterns, and to a lesser extent the retreat of the ice cap. The causes and implications of these
impacts, as well as potential responses to them and the potential synergies and conflicts with
environmental and development priorities are investigated in-depth later in Section 8.
5. Attention to climate concerns in donor activities
Tanzania receives large amounts of donor aid, in the order of one billion US$ per year, which is
equivalent to about 11% of its GNI. The largest donors, in terms of overall investments, are the World
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
20
Bank (IDA), Japan, and the United Kingdom. Figure 5 displays the distribution of this aid by development
sector and by donor.
Figure 5. Development aid to Tanzania (1998-2000
The following sections highlight the possible extent of climate risks to development investments
in Tanzania and examine to what extent current and future climate risks are factored in to development
strategies and plans, as well as individual development projects 10. Given the large quantity of strategies
and projects, our analysis is limited to a selection. This selection was made in three ways (i) a direct
request to all OECD DAC members to submit documentation of relevant national and sectoral strategies,
as well as individual projects (ii) a direct search for some of the most important documents (including for
instance national development plan/PRSP, submissions to the various UN conventions, country and sector
strategies from multilateral donors like the World Bank and UNDP, and some of the larger projects in
climate-sensitive sectors), and (iii) a pragmatic search (by availability) for further documentation that
would be of interest to our analysis (mainly in development databases and on donors’ external websites).
Hence, the analysis is not comprehensive, and its conclusions are not necessarily valid for a wider array of
development strategies and activities. Nevertheless, the authors feel confident that this limited set allows
an identification of some common patterns and questions that might be relevant for development planning.
5.1 Donor activities affected by climate risks
This section explores the extent to which development activities in Tanzania are affected by
climate risks, which gives an indication of the importance of climate considerations in development
planning. The extent to which climate risks affect development activities in Tanzania can be gauged by
examining the sectoral composition of the total aid portfolio, which is analyzed here using the World
10 The phrase “climate risk” or “climate-related risk” is used here for all risks that are related to climatic
circumstances, including weather phenomena and climate variability on various timescales. In the case of
Tanzania, these risks include the effects of seasonal climate anomalies, including droughts, as well as
trends therein due to climate change, and risks due to sea level rise. “Current climate risks” refer to climate
risks under current climatic conditions, and “future climate risks” to climate risks under future climatic
conditions, including climate change and sea level rise.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
21
Bank/OECD Creditor Reporting System (CRS) database (Box 2). Development activities in sectors such as
agriculture, infectious diseases, or water resources could clearly be affected by current climate variability
and weather extremes, and consequently also by changing climatic conditions. At the other end of the
spectrum, development activities relating to education, gender equality, and governance reform are much
less directly affected by climatic circumstances.
Box 2. Creditor Reporting System (CRS) Database
The Creditor Reporting System (CRS) comprises of data on individual aid activities on Official Development
Assistance (ODA) and official aid (OA). The system has been in existence since 1967 and is sponsored and operated
jointly by the OECD and the World Bank. A subset of the CRS consists of individual grant and loan commitments (from
6000 to 35000 transactions a year) submitted by DAC donors (23 members) on a regular basis. Reporters are asked to
supply (in their national currency), detailed financial information on the commitment to the developing country such as:
terms of repayment (for loans), tying status and sector allocation. The secretariat converts the amounts of the projects
into US dollars using the annual average exchange rates.
In principle, the sectoral selection should include all development activities that may be designed
differently depending on whether or not climate risks are taken into account. In that sense, the label
“affected by climate risks” has two dimensions. It includes projects that are at risk themselves, such as an
investment that could be destroyed by flooding. But it also includes projects that affect the vulnerability of
other natural or human systems. For instance, new roads might be fully weatherproof from an engineering
standpoint (even for climatic conditions in the far future), but they may also trigger new settlements in
high-risk areas, or it may have a negative effect on the resilience of the natural environment, thus exposing
the area to increased climate risks. These considerations should be taken into account in project design and
implementation. Hence, these projects are also affected by climate risks. A comprehensive evaluation of
the extent to which development activities are affected by climate change would require detailed
assessments of all relevant development projects as well as analysis of site specific climate change
impacts, which was beyond the scope of this analysis. This study instead assesses activities affected by
climate risks on the basis of CRS purpose codes (see Appendix B, which identifies “the specific area of the
recipient’s economic or social structure which the transfer is intended to foster”)11, 12.
Clearly, any classification that is based solely on sectors suffers from oversimplification. In
reality, there is a wide spectrum of exposure to climate risks even within particular sectors. For instance,
rain-fed agriculture projects may be much more vulnerable than projects in areas with reliable irrigation. At
the same time, the irrigation systems themselves may also be at risk, further complicating the picture.
Similarly, most education projects would hardly be affected by climatic circumstances, but school
buildings in flood-prone areas may well be at risk. Without an in-depth examination of risks to individual
projects, it is impossible to capture such differences. Hence, the sectoral classification only provides a
rough first sense about the share of development activities that may be affected by climate risks.
To capture some of the uncertainty inherent in the sectoral classification, the share of
development activities affected by climate change was calculated in two ways, a rather broad selection, and
a more restrictive one. The first selection includes projects dealing with infectious diseases, water supply
11 Each activity can be assigned only one such code; projects spanning several sectors are listed under a
multi-sector code, or in the sector corresponding to the largest component.
12 The OECD study “Aid Activities Targeting the Objectives of the Rio Conventions, 1998-2000” provides a
similar, but much more extensive database analysis. It aimed to identify the commitments of ODA that
targeted to objectives of the Rio Conventions. For this purpose, a selection was made of those projects in
the CRS database that targeted the Conventions as either their “principal objective”, or “significant
objective”.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
22
and sanitation, transport infrastructure, agriculture, forestry and fisheries, renewable energy and
hydropower13, tourism, urban and rural development, environmental protection, food security, and
emergency assistance. The second classification is more restricted. First of all, it excludes projects related
to transport and storage. In many countries, these projects make up a relatively large share of the
development portfolio, simply due to the large size of individual investments (contrary to investments in
softer sectors such as environment, education and health). At the same time, infrastructure projects are
usually designed on the basis of detailed engineering studies, which should include attention at least to
current climate risks to the project.14 Moreover, the second selection excludes food aid and emergency
assistance projects. Except for disaster mitigation components (generally a very minor portion of
emergency aid), these activities are generally responsive and planned at short notice. The treatment of risks
is thus very different from well-planned projects intended to have long-term development benefits.
Together, the first and the second selection give an indication of the range of the share of climate-affected
development activities.
In addition, the share of emergency-related activities was calculated. This category includes
emergency response and disaster mitigation projects, as well as flood control. The size of this selection
gives an indication of the development efforts that are spent on dealing with natural hazards, including,
often prominently, climate and weather related disasters.
The implications of this classification should not be overstated. If an activity falls in the “climateaffected”
basket, which does not mean that it would always need to be redesigned in the light of climate
change or even that one would be able to quantify the extent of current and future climate risks. Instead,
the only implication is that climate risks could well be a factor to consider among many other factors to be
taken into account in the design of development activities. In some cases, this factor could be marginal. In
others, it may well be substantial. In any case, these activities would benefit from a consideration of these
risks in their design phase. Hence, one would expect to see some attention being paid to them in project
documents, and related sector strategies or parts of national development plans. Figures 6 and 7 show the
results of these selections, for the three years 1998, 1999, and 200015.
13 Traditional power plants are not included. Despite their long lifetime, these facilities are so localized
(contrary to, e.g., roads and other transport infrastructure) that climate risks will generally be more limited.
Due to the generally large investments involved in such plants, they could have a relatively large influence
on the sample, not in proportion with the level of risk involved.
14 Note however, that they often lack attention to trends in climate records, and do not take into account
indirect risks of infrastructure projects on the vulnerability of natural and human systems.
15 The three-year sample is intended to even out year-to-year variability in donor commitments. At the time
of writing, 2000 was the most recent year for which final CRS data were available. Note that coverage of
the CRS is not yet complete. Overall coverage ratios were 83% in 1998, 90% in 1999, and 95% in 2000.
Coverage ratios of less than 100% mean that not all ODA/OA activities have been reported in the CRS. For
example, data on technical co-operation are missing for Germany and Portugal (except since 1999), and
partly missing for France and Japan. Some aid extending agencies of the United States prior to 1999 do
not report their activities to the CRS. Greece, Luxembourg and New Zealand do not report to the CRS.
Ireland has started to report in 2000. Data are complete on loans by the World Bank, the regional banks
(the Inter-American Development Bank, the Asian Development Bank, the African Development Bank)
and the International Fund for Agricultural Development. For the Commission of the European
Communities, the data cover grant commitments by the European Development Fund, but are missing for
grants financed from the Commission budget and loans by the European Investment Bank (EIB). For the
United Nations, the data cover the United Nations Children's Fund (UNICEF) since 2000, and a significant
proportion of aid activities of the United Nations Development Programme (UNDP) for 1999. No data are
yet available on aid extended through other United Nations agencies. Note also that total aid commitments
in the CRS are not directly comparable to the total ODA figures in Figure 5, which exclude most loans.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
23
Figure 6. Share of aid amounts committed to activities affected by climate risk and to emergency in Tanzania
(1998-2000
d a rk : a ffe c te d b y
c lim a te r is k s
(h ig h e s tim a te )
26%
74%
dark: affected by
climate risks
(low estimate)
12%
88%
dark: emergency
activities
97
%
3%
Figure 7. Share (by number) committed to activities affected by climate risk and to emergency activities in
Tanzania (1998-2000
dark: affected by
climate risks
(high estimate)
31
%
69
%
dark: affected by
climate risks
(high estimate)
21
%
79
%
dark: emergency
activities
96
%
4%
Emergency projects make up 3 to 4% of all activities. In monetary terms, between one-eighth and
a quarter of all development activities in Tanzania could be affected by climate change. By number, the
shares are higher, between about 20 and 30 percent16. In addition to providing insight in the sensitivity of
16 Note that the number of activities gives a less straightforward indication than the dollar amounts. First of
all, activities are listed in the CRS in each year when a transfer of aid has occurred. Hence, when a donor
disburses a particular project in three tranches, that project counts three times in our three-year sample. If
the financing for a similar three-year project is transferred entirely in the first year, it only counts once.
Secondly, the CRS contains a lot of non-activities, including items like “administrative costs of donors”.
Moreover, some bilateral donors list individual consultant assignments as separate development activities.
In most cases, such transactions will fall outside of the “climate-affected” category. Hence, the share of
climate-affected activities relative to the total number of activities (which is diluted by these non-items) is
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
24
development activities in Tanzania as a whole, the classification also gives a sense of the relative exposure
of various donors. These results are listed in Table 2 and 3 (again in the years 1998, 1999, and 2000).
Table 2. Relative shares of CRS activities, by total disbursed amounts, for the top-five donors in Tanzania
(1998-2000)
Amounts of activities
(millions US$)
Activities affected by
climate risks
(high estimate)
Activities affected by
climate risks
(low estimate)
Emergency activities
Donor
Amoun
t % Donor
Amoun
t % Donor Amount % Donor
Amoun
t %
Total 2916
100
% Total 761 100% Total 356 100% Total 81
100
%
UK 524 18%
CEC/ED
F 134 18%
German
y 50 14% USA 31 39%
IDA 453 16% Denmark 81 11% UK 41 12% AfDF 13 16%
Japan 326 11% Germany 72 9% Japan 38 11% UK 11 13%
CEC/EDF 264 9% Japan 71 9% IFAD 33 9% CEC/EC 8 10%
Denmark 215 7% UK 58 8% Norway 32 9% Sweden 7 8%
Table 3. Relative shares (by number) of CRS activities for the top-five donors in Tanzania (1998-2000)
Numbers of activities
Activities affected by
climate risks
(high estimate)
Activities affected by
climate risks
(low estimate)
Emergency activities
Donor Number % Donor Number % Donor
Numbe
r % Donor Number %
Total 1745 100% Total 536 100% Total 369 100% Total 76 100%
Sweden 232 13% Ireland 72 13% Ireland 64 17% Switzerl. 16 21%
UK 222 13% UK 67 13% UK 42 11% Sweden 15 20%
Norway 210 12% Norway 53 10% Norway 37 10% UK 14 18%
Ireland 191 11% Sweden 46 9% Sweden 31 8% Norway 6 8%
Germany 124 7% Germany 36 7%
Germany 30 8% Finland 5 7%
Given the extensive share of development activities in Tanzania that could be affected by climate
risks, one would assume that these risks are reflected in development plans and a large share of
development projects. The following sections examine to which extent this is the case.
5.2 Attention to climate risks in donor strategies
Tanzania regularly suffers from various climate-related hazards, including droughts that have
substantial effects on economic performance and poverty. Many development plans and projects recognize
this influence, and Tanzania’s climate even turns up in the context of economic analyses. However, few of
the development plans and projects that were reviewed take these risks into account. Given that current
climate risks are already being neglected, it comes as no surprise that climate change is often ignored
lower. On the other hand, the shares by total amount tend to be dominated by structural investments (which
tend to be more costly than projects in sectors such as health, education, or environmental management).
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
25
altogether. In the few cases where climate change does receive attention, the focus is on mitigation, rather
than adaptation.
Several donor strategies recognize Tanzania’s dependence on favourable weather, and the
linkages between poverty, drought, and food security. For instance, the AfDB Country Strategy Paper
highlights the impact of weather on economic performance: “growth rates have been fluctuating from year
to year reflecting the vulnerability of the economy to external shocks. Although strong growth was
registered in FY 1996/97 (4.2 percent), it declined to 3.3 percent in FY 1997/98 due to the adverse impact
of the drought on agricultural output. The drought was followed by the El-Nino floods late 1997 and early
1998, which destroyed some of the crops and damaged roads, thereby, disrupting internal movement of
agricultural commodities as well as export shipments.” IFAD’s Country Strategic Opportunities Paper
estimates that the country has a structural food deficit of about 700 tons, with imports rising to up to 1.5
million tons in times of flood or drought. This vulnerability cannot be attributed to weather conditions
alone. For instance, the AfDB paper notes that less than 20% of the irrigation potential is utilized,
unnecessarily exposing agricultural production to droughts. “While droughts have contributed to water
supply problems, the underlying factors include weak institutional capacity in the sector, poor water
resource management, and the dilapidated condition of the water schemes and distribution networks in the
rural and urban areas resulting from the under-funding of maintenance and rehabilitation.” The UN
Development Assistance Framework (UNDAF) also highlights Tanzania’s vulnerability to climate-related
disasters, due to natural and human factors: “Natural and man-made disasters erode the coping capacity of
the vulnerable population especially in drought-prone areas. There have been poor rains in Central
Tanzania for the last three years, and traditional coping strategies are breaking down as land pressure
increases. These types of shocks have become a frequent phenomenon in Tanzania in recent years. Floods
and droughts, epidemics and crop pests, environmental damage and economic instabilities, have all had
their effects on people’s capacity to meet their basic needs and subsequently their ability to survive and
pursue their development ambitions and potential”.
Despite these strong linkages between climate and economic performance, as well as the
relationships between droughts, environmental degradation and poverty, none of the donor strategies even
mentions climate change. Attention to current hazards, particularly droughts, varies from donor to donor.
Some of them, including SIDA, Ireland Aid and the EU, do not explicitly recognize the impact of current
climate-related risks on the success of development investments. Others, such as DFID and IFAD, have
components that aim to address Tanzania’s vulnerability to such risks.
In 2001, a joint “Emergency Consolidated Appeal for the Drought in Tanzania” was launched by
a number of UN agencies. Instead of just addressing short-term relief, the appeal intended to address the
underlying causes of the chronic droughts, including early warning systems, and drought mitigation
measures in Rural and Agricultural Development Strategies. Despite the longer-term focus of the appeal,
climate change was not considered.
5.3 Climate risks in selected development programs and projects
None of the (relatively few) development projects that were reviewed paid attention to the risks
associated with climate change. For instance, a World Bank forest conservation and management project,
which explicitly addresses climate change through carbon uptake, does not address current or future
climate-related risks. A regional GEF-funded World Bank project to improve the long-term environmental
management of Lake Victoria, does not consider the potential impacts of climate change on the water
resources and ecosystems at stake, and a USAID-sponsored coastal management partnership neglects sea
level rise in its analysis of integrated coastal zone management options.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
26
6. Attention to climate concerns in national planning
Since attaining its independence in 1961, Tanzania has been addressing its development process
through long, medium and short-term development plans and programs, which are developed by the
Planning Commission in the Ministry of Planning and Privatization. See Table 4 for an overview on
Tanzania’s planning history. The latest medium-term program is the so-called Three Year Rolling Plan and
Forward Budget, which rolls on an annual basis and has been in place since 1993/94 up to the present. The
major macroeconomic and sectoral policy objectives and cross-sectoral issues included in Tanzania’s plan
are poverty alleviation, population, science and technology as well as environmental protection.
Besides, Tanzania also embarks on long term planning, the latest being the National
Development Vision 2025, which aims for economic prosperity, equity, self-reliance, the transformation
from a rural based agricultural economy to a more diversified and industrialized one, as well as
sustainability by the year 2025. Despite the Vision’s long time horizon, climate change is not mentioned. It
neither discusses climate-related risks, nor strategies to mitigate or to adapt to them (such as irrigation,
reforestation, and crop diversification). Similarly, the shorter-term (5-year) Tanzania Assistance Strategy
(“a medium-term framework for promoting local ownership and development partnerships”) does not
discuss climate change either. However, climate-related risks, mainly floods and droughts, feature
prominently. Besides specific attention to disaster preparedness activities, the plan also advocates the
integration of disaster mitigation in Tanzania’s development plans.
Table 4. Tanzania’s main planning documents
National Plans Period
National Development Plans
Three Year Plan
First Five Year Plan
Second Five Year Plan
Third Five Year Plan
First Union Five Year Plan
Second Union Five Year Plan
Three Year Rolling Plan and Forward Budget (rolls annually)
1961-1963/64
1964/65-1968/69
1969/70-1973/74
1976/77-1980/81
1981/82-1985/86
1988/89-1992/93
1993/94 to date
Emergency Plans
National Economic Survival Programme
Structural Adjustment Programme
Economic Recovery Programme (ERP-I)
Economic Recovery Programme (ERP-II)
1982
1983-85
1986/87-1988/89
1989/90-1991/92
Long Term Perspective Plans
15 Year Development Plan
20 Year Development Plan
National Development Vision 2025
1964-1980
1981-2000
1998-2025
In 1997 Tanzania developed a first National Action Plan on Climate Change, which contained an
inventory of emissions by source and removal by sinks of greenhouse gases based on 1990 data. Besides
the Action Plan, various studies focusing on technological and other options for mitigating greenhouse
gases in Tanzania as well as on the assessment of vulnerabilities and possible adaptation measures have
been completed. Tanzania has also signed or ratified a number of multilateral environmental agreements,
and has a number of national level environmental and sectoral plans that intersect with responses that may
be required to manage climate variability and long term climate change.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
27
6.1 National Action Plan on Climate Change
The National Action Plan on Climate Change was developed in 1997 and has different objectives
for various timeframes:
6.1.1 Short term program (Year 1 - 2)
In the beginning efforts should be undertaken to raise awareness of possible impacts stemming
from climate change on various social and economic activities. The overall aim of these meetings would be
to explore possibilities of how current activities or sectoral plans could complement climate change
mitigation options. Besides, there is a need to analyze the effects of governmental macroeconomic policies
in relation to climate change.
6.1.2 Medium term program (Year 2 - 5)
In the medium term, projects already internalizing climate change aspects, especially those
reducing GHG emissions, should be supported. Support will either be sought from internal such as the
Government budget or from external sources. In addition, climate change aspects should be included into
the educational curriculum, preferably starting at secondary school level. Also, the Government should
start introducing environmental economic instruments such as fiscal measures (pollution taxes, input taxes,
product taxes, import tariffs, royalties , land user taxes, tax differentiation etc), property rights (ownership
right, user right, and development rights], and performance bonds (land reclamation bond, waste delivery
bond, environmental performance bond, etc.) as incentives to increase environmental conservation.
6.1.3 Long term program (Year 10 - 20)
In the long-term, large projects in the energy and transport sector should be undertaken. In
addition, adaptation measures to cope with a rising sea level and its adverse effects on coastal
infrastructures should be implemented.
6.2 National communications to international environmental agreements
Tanzania is a party to various international environmental agreements, including the UNFCCC,
UNCCD, and UNCBD. Tanzania recently submitted its Initial National Communication to the UN FCCC
(in July 2003), and is currently preparing a National Adaptation Programme of Action (NAPA).
While Tanzania’s National Report to the UN Convention on Biodiversity does not mention
climate change at all, its First National Report to the UN Convention to Combat Desertification refers only
to climate change mitigation mainly through the diversification of Tanzania’s energy resources. The
Second National Report to the UNCCD, however, does highlight the linkages between climate change and
desertification. It also notes that desertification programs have been quite successful, not only in terms of
awareness raising, but also by mainstreaming desertification concerns in national and sectoral plans and
policies.
Tanzania’s National Report to World Summit on Sustainable Development (2002) refers to the
national vulnerability and adaptation assessment, and explicitly lists agriculture, water resources, forestry,
grasslands, livestock, coastal resources and wildlife and biodiversity as vulnerable to climate change.
Nevertheless, adaptation receives very little attention (except in agriculture, where further work is
planned), in sharp contrast to mitigation, which is discussed extensively. Several components of potential
climate change adaptation strategies are included in efforts to address current-day vulnerability to climaterelated
risks, including better water management, for instance in the context of irrigation development, and
research on drought-resistant, high-yield crops.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
28
6.3 Poverty Reduction Strategy Paper (PRSP)
Although Tanzania’s PRSP recognizes the grave impact of weather and climate hazards on
development, and particularly on the poor, it neglects climate change. The important impact of climaterelated
risks, however, is clearly recognized. For instance, stakeholder groups that were interviewed in
preparation for the poverty strategy voiced their worries: “A major concern of the poor is their
vulnerability to unpredictable events. In Tanzania, famine often results from either floods or drought. Since
the mid-1990s, Tanzania has in fact experienced a series of adverse weather conditions, which undermined
food security. […] There is, therefore, a growing need for safety-nets.” In response, the PRSP lists a
number of activities to reduce this vulnerability, including early warning systems, irrigation, better food
supply systems, development of drought resistant crops, facilitation of the provision of adequate, safe and
clean water to the rural areas from 48.5% population coverage in 2000 to 85% by 2010, promotion of the
use of rainwater harvesting and sustained efforts in reforestation as well as sustained efforts in adaptation.
The PRSP progress report, which was published a year later, notes that agricultural growth has
been lagging behind expectation “owing to adverse weather and the collapse of export prices”. Despite this
observation, the report’s response to this lagging growth includes no direct measures to reduce
vulnerability to climate risks, not even the ones mentioned in the original PRSP a year earlier. Similarly,
these options are also neglected in the World Bank/IMF Joint Staff Assessments of the PRSP and the PRSP
progress report, suggesting that climate-related risks do not get much attention in the PRSP oversight
process.
6.4 Other national policies of relevance to climate change
Tanzania has put in place a number of environmental and sectoral policies and plans especially
during the 1990s, which are intended to increase its ability to cope with current environmental problems as
well as with additional risks posed by climate change. The following paragraphs discuss some of the most
relevant policies.
The National Environmental Action Plan (NEAP) of 1994 was a first step towards incorporating
environmental concerns into national planning and development. NEAP identified six priority
environmental concerns, namely land degradation; lack of accessible, good quality water for both urban
and rural inhabitants; pollution; loss of wildlife habitats; deterioration of marine and freshwater systems;
and deforestation. In order to address these issues the National Environmental Policy (NEP) was
promulgated in December 1997 to provide a framework for mainstreaming environmental considerations
into the decision-making processes in Tanzania. Though NEP does not pay explicit attention to climate
change, the primary environmental issues brought forward include many of the concerns that would be
addressed by no-regrets climate change adaptation measures. In particular, the NEP highlights the
importance of integrating environmental management in several sectoral programs and policies.
A particularly strong example of such integration is found in the agriculture sector, which is
crucial for food security and the eradication of rural poverty. The NEP, for example, proposes “the
improvement of land husbandry through soil erosion control and soil fertility improvement; the
minimization of encroachment in public lands including forests, woodlands, wetlands, and pastures; the
strengthening of environmentally sound use, monitoring, registration and management of agrochemicals;
as well as the improvement in water use efficiency in irrigation, including control of water logging and
salinization.” In addition, the forestry section of NEP is most explicit in giving attention to cross-sectoral
environmental issues: “the main objective is the development of sustainable regimes for soil conservation
and forest protection, taking into account the close linkages between desertification, deforestation,
freshwater availability, climate change, and biological diversity.” The only other paragraph in the NEP
that relates to climate change reads as follows “The need to undertake climate studies in order to come up
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
29
with mitigation options is stressed. In view of Tanzania’s vulnerability to climate variations, an assessment
of impacts of climate change and climate variations will be undertaken. In this regard strategies will be
evolved to ensure that options which are pursued do not unduly sacrifice national development endeavors.”
Similarly, the 1998 National Forestry Policy (NFP), which is a review of the 1953 one, gives no
direct references to climate change despite the vulnerability of Tanzanian forests to changed climatic
conditions. One of the main objectives of the NFP is to ensure ecosystem stability through conservation of
forest biodiversity, water catchments, and soil fertility. The policy states that new forest reserves for
conservation will be established in areas of high biodiversity value and that biodiversity conservation and
management will be included in the management plans for all protected forests. This policy is a great
departure from the traditional forestry approach of command-and-control by involving communities and
other stakeholders through joint management agreements.
Likewise, despite the criticality of climate change impacts on water resources the new National
Water Policy (NAWAPO), which has been approved by the Tanzanian cabinet in July 2002, does not
explicitly mention the issue. Nonetheless, the NAWAPO is participatory, multi-sectoral, river-basin based
and tries to integrate land use with water use and water quality as well as quantity. The four key issues in
the revised policy are 1) the demand respond approach, which leads to community ownership and
management of water and sanitation facilities; 2) private sector participation; 3) integration of water supply
and sanitation and finally 4) decentralization of service delivery from central government to district
councils.
7. Climate change and Mount Kilimanjaro
Mount Kilimanjaro derives its name from the Swahili words Kilima Njaro meaning “shining
mountain”, a reference to its legendary ice cap. It is the retreat of this ice cap, arguably linked to rising
temperatures, that has made the Kilimanjaro a prominent symbol of the impacts of global climate change.
Beyond the symbolism of the ice cap Kilimanjaro is also a hot spot of biodiversity with nearly 3000 plant
species and providing a range of critical ecosystem services to over one million local inhabitants who
depend on it for their livelihoods, as well as to the broader region that depends on water resources that
originate at the Kilimanjaro. The Kilimanjaro ecosystem is also subject to wide ranging impacts that may
be more directly attributable to changes in temperature and precipitation patterns, and which may have far
greater significance than the melting of the ice cap itself. This in-depth analysis has two objectives: (i) to
provide an overview on the impact of climatic changes on Mt. Kilimanjaro and on the resulting impacts on
the environment, ecosystems and on the human population; and (ii) to describe adaptation responses to
reduce or manage some of the most critical impacts and their synergy or conflict with other environmental
and development priorities.
Mt. Kilimanjaro is located 300 km south of the equator in Tanzania, on the border with Kenya. It
is the highest mountain in Africa, a huge strato-volcano (ca. 90 by 60 km), composed of three single peaks,
Kibo, Mawenzi and Shira that reach respectively an altitude of 5,895, 5,149 and 3,962 meters (Figure 8).
Kilimanjaro is also the world’s highest free standing mountain, looming 5,000 meters above an open
undulating plain that averages around 1,000 meters above sea level. The morphology of the upper areas of
Mt. Kilimanjaro is formed by glaciers which reached down to an altitude of 3000 m above sea level (asl)
during the ice age (Downie & Wilkinson 1972, Hastenrath 1984).
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
30
Figure 8. Mount Kilimanjaro
7.1 Climate, glaciers, and hydrology
Mt. Kilimanjaro is characterized by a typical equatorial day-time climate. Due to its near-equator
location, it experiences two distinct rainy seasons: the long rains from March to May forming the main
rainy season; and the short, but heavy rains centered on the month of November of the small rainy season.
The driest period falls into the months from July to October, while April and May are the wettest months.
However, rainfall and temperature vary with altitude and exposure due to the dominant wind blowing from
the Indian Ocean. Annual rainfall reaches a maximum of around 3,000 mm at 2,100 meters on the central
southern slope in the lower part of the forest belt, clearly exceeding precipitation on other East African
high mountains (HEMP 2001a). Higher up at 2,400, 2,700 and 3,000 meters, approximately 90, 70 and 50%
respectively of this maximum precipitation has been observed. The northern slopes, on the leeward side of
the mountain, receive much less annual rainfall (Figure 9).
The mean annual temperature in Moshi township (813 m) is 23.4°C (Walter et al. 1975). It
decreases to 9.2°C at an altitude of 3100 m, 5.0°C at 4000 m (HEMP, unpub. data) and –7.1°C on top of the
Kibo peak at about 5800 m (Thompson et al. 2002), with a lapse rate of about 0.6°C per 100 m increase in
altitude. The climate in the alpine belt above 3500-4000 m is characterized by extremes, with nightly frosts
and intense sunshine during daytime all year round (HEDBERG 1964).
Kilimanjaro represents a rare instance of the occurrence of glaciers in equatorial regions and like
the glaciers of Rwenzori and Mt. Kenya these are a relic of the colder and wetter climatic conditions of the
region during the Pleistocene (Downie & Wilkinson 1972). At present permanent ice exists only on Kibo -
covering an area of 2.6 km2 (Thompson et al. 2002). Yet, the distribution of moraines reaching down to an
altitude of 3000 m indicates that a much greater area of the mountain was formerly covered by ice (Downie
& Wilkinson 1972, Hastenrath 1984).
Mt. Kilimanjaro is a critical water catchment for both Tanzania and Kenya. High rainfall and
extensive forests give Mt. Kilimanjaro its high catchment value. The southern and the south-eastern forest
slopes form the main upper catchments of the Pangani River, one of Tanzania’s largest rivers, which drains
into the Indian Ocean near Tanga. Although the greater aridity of the northern slopes is reflected in a
sparser network of valleys on this side, in their shallower cross-section and in the general absence of
running water above 3000 m (Downie & Wilkinson 1972), the north-western slopes form the catchment of
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
31
the Tsavo River, a tributary of the Galana River, one of Kenya’s major rivers. The Amboseli National Park
in Kenya also depends on the hydrology of Mt. Kilimanjaro17.
Figure 9. Annual precipitation on Mt. Kilimanjaro
Source: Hemp 2001a
Mt. Kilimanjaro’s ice cap is relatively small in comparison to its height and surface area and its
contribution in developing water sources must be assumed to be equally slight (Ramsay 1965)18. Very few
streams originate in this zone and most of these have small flows. In contrast, the montane forest belt
between 1600 and 3100 m provides most of the water (96%) coming from the mountain (Ramsay 1965). In
17 Further afield in Kenya, it is likely that the mountain has an effect on Ol Turesh swamp and possibly
Mzima Springs, whose primary catchment is the Chyulu Hills.
18 Only Weru-Weru and Kikafu River, important branches in the headwaters of the Pangani River, are linked
by permanent streams to glaciers on the south-west edge of Kibo. The relatively few springs between the
ice cap and the forest belt indicate that the percolation of melt water downwards through the permeable
surface volcanic ash is small as well. Therefore - except below the glaciers of south-west Kibo - the valleys
above 3600 m are dry for a large part of the year (Downie & Wilkinson 1972).
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
32
this zone the rainfall is very high while evaporation losses are low due to an almost permanent cloud cover.
A great amount of water from this zone flows underground, directly to the savanna plains.
7.2 Ecosystems, biodiversity and land tenure on Mount Kilimanjaro
The Kilimanjaro Region consists of six administrative districts: Moshi Rural (1,713 km2), Moshi
Urban (58 km2), Hai (2,111 km2), Rombo (1,442 km2), Same (5,186 km2) and Mwanga (2,698 km2) of
which only the four former are immediately adjacent to Mt. Kilimanjaro. Regional headquarter is Moshi.
Most of the forest is part of the Mt. Kilimanjaro Forest Reserve (107,828 ha). The upper areas of Mt.
Kilimanjaro that lie above the 2,700 meters contour fall within Kilimanjaro National Park with 75,575 ha.
Mt. Kilimanjaro has a rich diversity of ecosystems, particularly of vegetation types that result mainly from
a large range in altitude and rainfall (summarized in Table 5). Due to the high diversity of its ecosystems,
Mt. Kilimanjaro is also very rich in fauna and flora, including about 2200 vascular plant species and 140
mammals. Details on the fauna and flora on Mt. Kilimanjaro are summarized in Box 3 and 4 respectively.
Kilimanjaro is one of the main agricultural regions of Tanzania contributing approximately 30%
of the country’s high quality Arabica coffee in 1985/1986 (O’KTING´ATI & KESSY 1991). In addition to
coffee the other cash crops are sugar cane, sisal, pyrethrum and cotton. Mt. Kilimanjaro is also important in
terms of food crops such as bananas, beans, rice and millet. Most of this activity on the southern and
(north) eastern slope of Mt. Kilimanjaro is performed by smallholders of the Chagga tribe, who use the
vegetation zones in various ways (see Table 6), depending on the climatic conditions (cp. HEMP et al.
1999). On the southern slopes of the mountain, the area below the montane forest was traditionally divided
into two zones. The upper part, the highland area of the irrigated banana belt in the submontane zone
(“kihamba” land), was permanently cultivated and inhabited by the Chagga for reasons of suitable climate
and defense against the Masai. The lower part, the “shamba” land of the colline savanna zone was
cultivated seasonally and provided annual crops like maize, beans and finger millet as well as fodder for
cattle.
Box 3. Flora of Mount Kilimanjaro
About 2,200 vascular plant species occur on Mt. Kilimanjaro (HEMP, unpub. data). These are 22% of the
approximately 10,000 vascular plant species of Tanzania (BRENAN 1978). Dissecting diversity into different types of
habitats or formations, the forest belt is the most important habitat in terms of species diversity on Mt. Kilimanjaro.
Nearly 900 species occur in the forests of Kilimanjaro, representing roughly 45% of the whole vascular flora (HEMP,
unpub. data). Besides the richness in epiphytes another prominent feature of the forests of Mt. Kilimanjaro is the
wealth of ferns, especially on the southern slope, due to the high humidity. 145 taxa of pteridophytes, constituting
roughly 35% of the pteridophyte flora of Tanzania, occur on the mountain, most of them (over 90%) in the forests
(HEMP 2001 a, b, 2002). The number of vascular plants capable of enduring the harsh climate conditions in the alpine
zone is rather small – together with the ericaceous subalpine zone Kilimanjaro harbours in its alpine belt only 350
species of vascular plants (HEMP, unpub. data), 13 of which are endemic to Kilimanjaro (HEDBERG 1961). About 600
species of bryophytes (of which 415 are mosses and 185 are liverworts) and approximately 120 lichens occur on Mt.
Kilimanjaro. 12 bryophytes are strict endemics. The richest belt for bryophytes is between 2100-4100 m (PÓCS 1991).
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
33
Table 5. Altitudinal zones and main vegetation units at Mount Kilimanjaro
Altitude
(meters) Main vegetation type
Altitudinal zone
according to Hemp (2001
a)
4400
Cushion vegetation (Helichrysum) 11
lower
Alpine
3800
Erica shrubland, Helichrysum cushion vegetation 10
upper
Erica shrubland, Erica excelsa forest, Hagenia-
Rapanea forest 9
middle
2800
Erica excelsa forest, Podocarpus forest, moorland 8
lower
Subalpine
2700
Podocarpus-Ocotea forest, Erica excelsa forest 7
upper
Ocotea-Podocarpus forest 6
middle
Agauria-Ocotea forest, Cassipourea forest 5
1600
Agauria-Ocotea forest, coffee-banana plantations,
Bulbostylis meadows 4
lower
Montane
1500
Coffee-banana plantations, Croton-Olea forest,
Hyparrhenia meadows 3
upper
900
Coffee-banana plantations, savanna bushland and
grassland, agriculture, pasture 2
lower
Submontane
800
700
Savanna bushland and grassland, agriculture,
pasture 1 Colline
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
34
Box 4. Fauna of Mount Kilimanjaro
GRIMSHAW et al. (1995) recorded about 140 species of mammals for Mt. Kilimanjaro, a number far exceeding the
diversity known for Mt. Kenya (GATHAARA 1999). Among them, 87 species are regarded as being pure forest species.
Black Rhinoceros is now extinct in the area, as possibly are reedbuck and klipspringer. Twenty four antelope species
are recorded in the area, as well as 25 species of carnivores and 7 species of primates. The forest is home to the
largest known population of Abbot’s duiker, which is globally threatened. There are also 25 species of bats
(Chiroptera).
SJÖSTEDT (1909) listed 405 bird species in his expedition report for Mt. Meru and Mt. Kilimanjaro, while GRIMSHAW
(1996) gives a number of 179 highland bird species inhabiting Mt. Kilimanjaro. In an ethno-zoological study, 82 bird
species were recorded on the southern slopes in the area of the Chagga home gardens, mostly from an altitude of
1400 m (Hemp et al. 1999) reflecting the high diversity of bird habitats. 4 bird species which are globally threatened
occur on Kilimanjaro. These are Lesser Kestrel, the Taita Falcon, the Corncrake and Abbot´s Starling. The
Madagascar Pond-Heron and the Pallid Harrier are near threatened species
418 reptile species are recorded for East Africa of which 302 are listed for Tanzania. The habitat range of 88
reptile species lies within Mt. Kilimanjaro (SPAWL et al. 2002). Thus Kilimanjaro harbours about 21% of the reptile fauna
of East Africa and 29% of Tanzania. The side-spotted dwarf gecko (Lygodactylus laterimaculatus) known only from Mt.
Kilimanjaro and the Taita Hills, and the Mt. Kilimanjaro two-horned Chameleon (Chameleo tavetanus) occurring on Mt.
Kilimanjaro and Mt. Meru, the adjacent North and South Pare Mts., and the Chyulu and Taita Hills in Kenya are locally
restricted species.
SJÖSTEDT recorded 1,310 species of beetles (Coleoptera), 594 Hymenoptera, 447 bugs and allies (Hemiptera),
and 537 butterflies and moths (Lepidoptera) species for the area including Mt. Meru, but with a main focus on Mt.
Kilimanjaro. The insect materials collected highlight the diversity of Mt. Kilimanjaro and the large number of endemic
species: 47 of the 107 known Homoptera species were endemic to the mountain, as well as 27 of the 57 recorded
Darkling beetles (Tenebrionidae). A high rate of endemism was also recorded for the Rove beetles (Staphylinidae,
39% endemism), the Scarab beetles (Scarabaeidae, 25% endemism) and the long-horned beetles (Cerambycidae,
36% endemism in the mountain among all species known in East Africa) (FORCHHAMMER & BREUNING 1986; HEMP &
WINTER 1999; HEMP, C., 2001). Grasshoppers and locusts (Saltatoria) have been well studied on Mt. Kilimanjaro; 140
species of Acridoidea have been collected around the mountain in the past 10 years (HEMP & HEMP, in press), which
represent 33% of the species found in entire Tanzania according to a list published by JOHNSEN & FORCHHAMMER
(1975). Together with the Ensifera, about 190 species of Saltatoria are recorded on the mountain, of which 12 species
are only known from Mt. Kilimanjaro localities (HEMP, C., in press), and three species are still un-described,
representing 8% endemism in this insect group.
439 species of Odonata are reported for East Africa of which 171 occur in Tanzania (CLAUSNITZER 2001).
Nevertheless, the number of dragonflies recorded for Tanzania is constantly growing with every field survey due to the
very poor original data base. There are 16 species restricted to Tanzania which means a share of 9% endemism of
dragonflies for Tanzania. Mt. Kilimanjaro alone harbours 85 species (20%), among them are 14 species typical for
montane areas (17%) (CLAUSNITZER, pers. comm.). In comparison to other montane habitats of volcanic origin in East
Africa, Kilimanjaro, though being the youngest, shows an unusual high diversity due to Eastern Arc species, which
reached Kilimanjaro via the adjacent North Pare Mountains. Thus, this particular insect group exemplary reflects the
high diversity of habitats on Kilimanjaro.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
35
Table 6. Land-use in the different vegetation zones of Mount Kilimanjaro
Altitude
(meters) Land-use Altitudinal zone
3200
2700
• South eastern slopes: forest border with tussock grasses and giant lobelias
are fringed by so-called moorland zones into Erica bushland at steeper
slopes
• Tussock grassland, although already situated in the Kilimanjaro National
Park (KINAPA), is in some areas cut by fodder collectors
• Bee hives were seen up to over 3000 m with bees sucking on Erica. Except
for the tourist climbing routes the afro-alpine zone of the National Park is
mostly undisturbed by direct human impact
Subalpine zone
1700
• Southern and eastern slope: half-mile forest strip ranges between the
plantation belt and the forest reserve; provides timber and firewood (mostly
pines, cypress and eucalyptus)
• Meadows reach far into the montane forest, especially along the rivers
• Forest strip grades into natural montane forest, which should be excluded as
“forest reserve” from any usage. Nevertheless, since the 1950s about 12%
of the forest was changed into cypress and pine plantations
• Northern, north eastern and western slope: large forest plantations
• Honey collectors also frequent the montane forest zone
• Special type of land use: Shamba (Taungay) system practices (allowing
local farmers to inter-crop annual agricultural crops – mainly potatoes,
carrots and cabbage – with tree seedlings in forest plantation areas until the
third year of tree growth. By the third year, the young tree canopy casts too
much shade for the normal growth of agricultural crops. At this point farmers
move out and are allocated another plot, if available)
Montane forest
1000
• Most intensively cultivated by the Chagga (population density 500 person
per km2)
• Tree layer provides firewood, fodder and shadow, banana trees (in about 25
varieties)
• Network of irrigation canals
• The Chagga live among their home gardens in single dwellings, villages as
such do not exist
• Livestock like cattle, goats, sheep and pigs, sometimes even chicken, are
kept in stalls
• Bee-keeping plays an important role (Two bee species are kept: the bigger,
stinging honey-bee Apis mellifera ssp. monticola resembling the European
honey-bee, and a small stingless bee of the genus Meliponula)
Submontane coffeebanana
zone
700
• Southern foothills: most areas planted with maize and beans
• North-eastern foothills: maize, finger millet (important ingredient of local
beer), pigeon peas, groundnuts and sunflowers
• Western and north western foothill: large farms owned by big companies or
the government growing mainly wheat
• East of Moshi: rice
• South of Moshi: sugar plantations
Colline savanna
zone
7.3 Climatic trends on Mount Kilimanjaro
Over the past millennium, equatorial East Africa has witnessed a series of contrasting climate
conditions19. A drastic climatic dislocation took place during the last two decades of the 19th century,
19 A significantly drier climate than today occurred during the “Medieval Warm Period” (~AD 1000-1270)
and a relatively wet climate during the so-called “Little Ice Age” (~AD 1270-1850), which was interrupted
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
36
manifested in a drop of lake levels and in the onset of glacier recession (Hastenrath 1984, 2001,
Verschuren et al. 2000, Nicholson 2000, Nicholson & Yin 2001)20. A decrease in the annual precipitation
of the order of 150 mm with attendant albedo and change of cloudiness during the last quarter of the 19th
century constitutes also the most likely cause of the retreat of the Lewis Glacier on Mt. Kenya. In contrast,
the continuation of ice retreat beyond the early decades of the 20th century – as is the case for Kilimanjaro
– has been favored by a warming trend (Kruss 1983). Further, weather patterns on the Kilimanjaro are
intricately linked to landscape characteristics (e. g. Altmann et al. 2002)21. During the past few decades
vast savanna woodlands have increasingly been turned to agricultural use and thousands of hectares of
forest cover on the mountain have been destroyed by logging and burning. Whether such reciprocal effects
caused by (mostly man-made) landscape changes or whether climatic changes are of higher influence on
the Kilimanjaro remains an open question.
The most striking and most easily recognizable evidence for a steady change in regional climatic
conditions on Mt. Kilimanjaro, directly influencing landscape characteristics, are the vanishing glaciers. As
there are no signs of an increasing volcanic activity on a major scale this phenomenon has to be linked to
climatic conditions. Also, the fact that such glacier retreat is coincident with similar patterns elsewhere
around the globe leads to the assumption that their causes are also of a global character (Kaser 1999). In
contrast to this direct climatic impact, there are other even larger landscape changes, which are linked
indirectly to changing weather conditions. During the last century not only were the glaciers melting
rapidly, but there was also a significant increase in number and intensity of wild fires on Mt. Kilimanjaro,
which are most likely caused by the same climatic changes and which are simultaneously enhanced by
human influence. Changing weather patterns influence not only landscape characteristics but also animal
distributions (cp. Altmann et al. 2002). A changing migration behavior and population dynamic of big
game has been observed in the forests of Mt. Kilimanjaro.
Analysis of proxy data reveals that annual precipitation decreased by 150 mm, this means a lapse
rate of 7.5 mm/year between about 1880 and 1900. Since 1935 there are actual daily rainfall records from
the Lyamungu Coffee Research Institute which is located at an altitude of 1200 meters in the submontane
cultivated zone on the southern slope of Mt. Kilimanjaro. The annualized values are shown in Figure 10.
by three episodes of several decades of persistent aridity more severe than any recorded drought of the
twentieth century (Verschuren et al. 2000).
20 This glacier recession was caused by enhanced solar radiation due to diminished cloud cover which
accompanied the reduced precipitation. The drastic drop of the water level of Lake Victoria from around
1880 to the turn of the century was caused by a reduction in annual precipitation of about 150-200 mm
(Hastenrath 1984). These data are apparently indicative of an important precipitation reduction throughout
an area exceeding East Africa (cp. the variations of the water level of Lake Chad (Street-Perrott & Perrott
1990), where severe droughts started from the year 1900), followed by little secular precipitation variation.
21 The role of temperature and rainfall in shaping the landscape has long been recognized. More recently both
empirical evidence and mathematical models have highlighted the reciprocal impact of landscape changes
on weather patterns.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
37
Figure 10. Annual precipitation at the Lyamungu Coffee Research Institute
Source: Lyamungu Coffee Research Institute
It appears that there is a decrease in precipitation since 1935 of about 11% or 177 mm (equivalent
to 2.6 mm/year) at Lyamungo or a lapse rate of 2.6 mm per year. If this rate is extrapolated back to the year
1900 this would mean an annual loss of over 400 mm compared with the situation before 1880.
This records from Lyamungu are consistent with a general reduction in rainfall throughout most
of Africa since 1950 (Nicholson 2000) and in the area of Kilimanjaro according to the maps presented by
Hay et al. (2002) for the time interval 1941-1995 between 1941-1960 and 1971-1995. In addition to the
decline in annual precipitation, the Lyamungo data also reveal that the number of dry months with less
than 30 mm increased, whereas wet months with more than 125 mm were stable. With regard to
temperature, the maps presented by Hay et al. (2002) indicate that spatially averaged temperatures in the
area of Kilimanjaro rose between 1951-1960, were stable or decreased slightly between 1960-1981, and
increased again between 1981-1995. While no time-series exists for a particular location on the mountain,
there is however a 25-year temperature record (from 1976) from the Amboseli region just to the north
(Altmann et al. 2002). This record shows daily temperatures increased dramatically throughout the same
25-year period. Mean daily maximum temperatures increased with a rate of 0.275 °C per annum, with
increases being greatest during the hottest months of February and March.
To summarize, available climate records reveal a declining trend in precipitation on the
Kilimanjaro at least since 1880. Although available data is not sufficient to infer temperature trends at
different altitudes on the mountain, a distinct overall warming trend has been observed for most of the
period since 1950 to present. Observations from neighboring Amboseli in fact indicate a local warming
rate of 0.275 °C per decade between 1976-2000, significantly higher than globally averaged warming.
Either of these trends – declining precipitation or increasing temperature – contribute to enhanced
glacier melting, as well as to enhanced fire risk22. Consistent with the pronounced decrease of precipitation
22 Decreased precipitation reduces cloud cover and therefore enhances the sunlight reaching the glacier,
causing it to melt faster. The effect of increased temperatures on glacier melting is self evident. With
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
38
at the end of the 19th century fires in the areas of the subalpine forests were documented by the first
Europeans on the mountain (Meyer 1890, Volkens 1897, Jaeger 1909). At the same time the glaciers
started to recede (Kaser et al. under review) – driven by such changes in precipitation23. In the following
decades the climatic situation was more stable while the glaciers changed more slowly (Kaser 1999).
Enhanced glacier melt and fire risk have both been empirically observed in recent decades. These effects
are consistent with the simultaneity of precipitation decline and temperature increase24 which has been
observed during the same time period.
7.4 Potential impacts of climatic changes: glacier retreat
The ice cap on the Kilimanjaro has been in a general state of retreat since the end of the Little Ice
Age around 1850. This retreat was driven by natural climatic shifts (particularly a decline in regional
precipitation), but appears to have accelerated due to the warming observed in the second half of the 20th
century. Later in 1976 the glaciers covered 4.2 km2 (Hastenrath & Greischar 1997) compared with only 2.6
km2 in 2000 (Thompson et al. 2002). Measurements taken in 2000-2001 on Kilimanjaro show that its
glaciers are not only retreating but also rapidly thinning (Thompson et al. 2002). Figure 11 shows the
diminishing extent of the glaciers on Kibo between 1962 and 2000. Over these 38 years, Kilimanjaro has
lost approximately 55 % of its glaciers. There is general consensus that the ice cap of Kilimanjaro will
have disappeared by the year 2020 for the first time in the surveyed period of over 11,000 years.
Figure 11. Development of the Kilimanjaro (Kibo) ice fields from 1912 to 2000
Source: Thompson et al. 2002
The symbolism of this loss notwithstanding, it is important to note that the impact of the
disappearance of the ice cap on the natural and human systems would be very limited. The present glaciers
of Kibo cover an area equivalent to 0.2% of the area covered by the forest belt on Mount Kilimanjaro.
regard to forests, drier and hotter conditions both contribute to enhanced inflammability of the forest,
thereby enhancing fire risk.
23 The causes of such changes in precipitation are likely natural and not linked to climate change.
24 The warming in recent decades is consistent with climate change.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
39
Only two rivers are directly linked by very small streams to the glaciers, while 90% of the precipitation is
tapped by the forest belt. Even if the glaciers have melted till 2020 there will still be precipitation on Kibo
feeding springs and rivers, although not so continuously and to a much lesser degree.
Therefore, contrary to the opinion expressed by Thompson et al. (2002) it is very unlikely that the
loss of the glaciers will have a major impact on the hydrology of the mountain. Kaser et al. (under review)
come to the same conclusion. Further, observations of dry river beds are not necessarily an indicator of
long term climatic changes or the impact of shrinking glaciers. Dried out rivers in some areas are much
more likely the result of forest destruction or of increasing water demands of the rapidly growing
population. Water diversion has in fact quadrupled in certain areas during the last 40 years (Sarmett &
Faraji 1991).
Today Kilimanjaro National Park (KINAPA) is a major tourist attraction in Tanzania and gains
the most foreign exchange of any National Park in Tanzania (Newmark & Nguye 1991). Most visitors are
mainly interested in reaching the summit of Kibo, known as Uhuru Peak, the highest point in Africa. Since
the establishment of the Park in 1972, the number of visitors of KINAPA has multiplied by five. Without
any doubt Mt. Kilimanjaro will lose part of its beauty with the inevitable loss of its glaciers. However, it
will still remain the highest mountain in Africa – and incentive enough to climb. Therefore, a decline in
tourist numbers is unlikely.
7.5 Potential impacts of climatic changes: enhancement of fire risk
A less publicized and possibly far more significant impact of climate change on Mount
Kilimanjaro is the intensification of fire risk and its attendant impacts on biodiversity as well as ecosystem
services. In theory rising temperatures should result in the upward migration of vegetation zones, as
observed in the Alps by GRABHERR & PAULI (1994). This effect however has been offset by the
intensification of fire risk as a result of warmer temperatures and declining precipitation. This risk is
particularly acute in the vast ericaceous belt in the upper reaches of the vegetation. Consequently, climatic
changes have actually pushed the upper forest line downward on the Kilimanjaro.
On Mt. Kilimanjaro fires are common in the colline savanna zone, in the (sub-) alpine zone and –
to a lesser degree – in the submontane and lower montane forest zone, whereas in the middle montane
forest zone – at least on the southern slope – fires are rare. Most of these fires are lit by man (often as a
maintenance tool), especially in the cultivated areas on the lower reaches of the mountain. The situation
however is different in the upper regions of the mountain where no grazing or agriculture exists above the
forest belt and logging in the upper forest zone is also rare25. Since climate change is the objective of this
analysis, man lit fires are of minor interest. Therefore the destructive role of fires in the forests and in the
alpine zone where climatic conditions play a more critical role are the focus of this discussion.
7.5.1 Elevation distribution of species richness and its relationship to fire
Fire variously influences species diversity, composition and vegetation structure in the different
altitudinal zones on Kilimanjaro (Hemp, in press). Figure 12 shows the species numbers of vascular plants
25 Although even these remote areas are not free from human influence, as population on the foothills has
increased enormously. Since 1895 population has multiplied by 20. As a result an increasing human
activity can be seen in all altitudinal zones and areas, promoted in particular by tourism. Since the
establishment of the park in 1972, the number of visitors of KINAPA has multiplied by five. Together with
porters, guides and tourists about 100.000 people visit the upper regions of Kilimanjaro per year. Such
increasing numbers of visitors have certainly effects on the environment. Thus, the (natural) impacts of the
changing climatic conditions are additionally enhanced by human influence.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
40
at 100 m elevation intervals between 700 and 4500 m. Vascular plants have their maximum (745 species)
in the 1300 - 1400 m interval in the area of the banana-coffee plantations of the submontane zone.
Figure 12. Absolute species numbers of ferns and of all occurring vascular plants on the southern slope
of Mount Kilimanjaro
Source: HEMP 2001a
As shown in Figure 12 species numbers are highest in moderately cultivated or disturbed areas
and not in natural, completely untouched areas. In this context the second peak at 2600 m at the lower
border of the subalpine zone is of interest. In this altitude fire starts to be an important ecological factor on
Mt. Kilimanjaro, creating a mosaic of different fire induced stages of forest, shrub and tussock grassland
communities. This high diversity in habitats - compared with the closed forest at lower altitudes and the
monotonous heath lands at higher altitudes – leads to a high diversity in species numbers. This trend is
enhanced by the occurrence of fire-tolerating species, which show the same bimodal altitudinal distribution
with a gap in the wettest central forest parts where fires are uncommon. They can therefore be regarded as
fire indicators.
7.5.2 Influence of fire on regeneration, composition and structure of forests
Forest fires are frequent in the subalpine zone and also, less frequent in the submontane and
lower montane zone between 1300 and 2000 m above sea level (asl). Fires in the submontane and lower
montane forests are mostly set by people. In these forests, fire changes species composition and structure
of the tree as well as the herb layer (HEMP, A. in press). This is of major importance for forest
regeneration, as the dense cover of bracken impedes the sprouting of trees. In the subalpine forests
between 2800-3000m asl fire causes sharp discontinuities in the floristic composition and structure26. Once
Erica excelsa has established, regeneration of a broad-leaved forest becomes more and more improbable
(HEMP & BECK 2001). If the frequency of fire becomes too high it degrades Erica excelsa forests into bush
lands in which E. excelsa is substituted by E. trimera and E. arborea. This Erica bush extends between
26 Giant heather (Erica excelsa) becomes dominant at this altitude forming dense mono-specific stands,
which border the Podocarpus and Juniperus forests without any transition (HEMP & BECK 2001).
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
41
3200 and 3900 m asl. Continuously high frequency of fires destroys this bush vegetation, ultimately
resulting in Helichrysum cushion vegetation.
7.5.3 Major impacts of fire on Mount Kilimanjaro’s ecosystem
On Mt. Kilimanjaro structure and composition of the subalpine vegetation is strongly influenced
by recurrent fires. Above 3200 m asl Erica excelsa forest is replaced today by Erica trimera and E.
arborea bush in most areas. But from field observations and historical descriptions (JAEGER 1909, KLUTE
1920) it can be assumed that the forest extended up to 3600 m in some areas of Kilimanjaro at the
beginning of the twentieth century while an open Erica forest was reported at altitudes of over 3900 m; this
is 800 m higher than today (HEMP & BECK 2001). On the south-eastern slopes at an altitude of 2800 m
Erica excelsa stands and the “moorland” tussock vegetation produce very abrupt boundaries. Tree-islands
consisting of a core of Podocarpus forest are surrounded by a fringe of Erica trees and various shrubs. In
this area, a mosaic of Podocarpus forest, Erica forest and subalpine grassland occur at the same altitude.
Substantial microclimatic differences can thus be ruled out as an explanation for this pattern. Rather,
recurrent fires may be the crucial factor pushing the forest back from the subalpine to lower and moister
regions.
The comparison of two classified Landsat images from 1976 and 2000 reveals enormous changes
in the upper vegetation zones of Mt. Kilimanjaro during the last 24 years (Figures 13 and 14)27.
Figure 13. Vegetation cover in the montane and alpine zone on Mount Kilimanjaro (1976)
27 It should also be mentioned that on Fig. 13 and 14 differences in glacier size are apparent. While in 1976
the glaciers covered 4.2 km2 (HASTENRATH & GREISCHAR 1997) in 2000 they have been shrunk to 2.6 km2
(THOMPSON et al. 2002).
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
42
Figure 14. Vegetation cover in the montane and alpine zone on Mount Kilimanjaro (2000)
In 1976, the Erica trimera bush, which today is depressed in the western and northern parts of the
mountain below 3400 m, reached up as a continuous belt to over 4100 m into an area which today is
covered by Helichrysum cushion vegetation. Erica forests covered nearly 5 times the current area (166 and
36 km2 respectively), extending in many places up to 3700 m. This equals a loss of 130 km2 or over 10%
of Kilimanjaro’s forest cover due to fire since 1976.
As discussed earlier, Erica vegetation is largely influenced and controlled by fire. The growing
influence of fire pushed down the forest line replacing Erica forests with Erica bush. Fire has also shifted
the upper border of Erica trimera bush by replacing it with Helichrysum cushion vegetation. The
Helichrysum cushion vegetation is not threatened by fire because its little biomass provides little fuel. In
addition, distances between vegetation patches and cushions are too high to allow fire to spread. When fire
reaches this vegetation zone it stops. Therefore, the upper line of this vegetation formation has been stable
for the examined 24 years. On the lower edge, however, fire was able to spread into the Erica bush zone. It
can be assumed that most of the Erica bush of the year 1976 as shown in Figure 13 has still been Erica
forest at the end of the 19th century while most of the Erica forest of 1976 was still broadleaved forest at
that respective time. This constitutes a loss of over 300 km2 of upper montane forest (or a third of the
present forest size) during the last 120 years. As a consequence, the ericaceous belt on Kilimanjaro with
the easily inflammable heathlands became larger, giving rise to more and bigger fires.
7.5.4 Socioeconomic impact of increasing fire intensity
The increase in fire intensity on the slopes of Mt. Kilimanjaro has very significant impacts on
both the natural and human systems that it sustains. The most direct impact is a significant decline in water
resources; other impacts include effects on farming and other activity, as well as a loss of biodiversity.
7.5.4.1 Water resources
The devastation of 13,000 ha of forests, mostly of Erica forest, in the upper reaches of the
Kilimanjaro since 1976 by fire has caused a serious disturbance in the water balance of the entire
mountain, given that the forest belt functions as the main water catchment area. Montane and subalpine
mossy or cloud forests are of great importance for watersheds in East Africa. They play an active and
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
43
important role in the protection of slopes against erosion by controlling the damaging effects of torrential
rainfall and regulating the outflow patterns of watercourses. In cloud forests about one third of the total
rainfall is absorbed by the dense epiphytic layer (PÓCS 1976). Destruction of these forests reduces the
function of the forest belt as a water filter and reservoir. Instead of remaining in the thick epiphytic
biomass, humus and upper soil of the forest, percolating slowly to the groundwater, rainwater flows off
quickly on the surface to the rivers eroding the soil and increasing the danger of floods on the foothills.
Another consequence of the quicker rate of rain flow is water shortage during periods without rain.
In addition to the function of filtering and storing water the upper montane and subalpine cloud
forests have a high potential of collecting cloud water (fog interception). Fog interception or fog deposition
refers in this case to the small cloud droplets that do not settle on horizontal surfaces and, thus, are not
collected in a rain gauge. Cloud water droplets are blown by the wind against the vegetation where they
coalesce to form large drops that run off and fall to the ground. Fog droplets have to be intercepted by the
vegetation and do not precipitate spontaneously (cp. CAVELIER et al. 1996, GLASOW & BOTT 1999).
Above 2000 m asl fog and mist occur nearly every day, above 2600 m asl every day. Thus, fog
interception increases with altitude, especially its relative share of water input. The amount depends on the
height and leaf area index of the vegetation providing wetting capacity for interception, the frequency of
fog, and exposure to the prevailing wind (CAVELIER & GOLDSTEIN 1989, CAVELIER et al. 1996, GLASOW
& BOTT 1999, ZIMMERMANN et al. 1999). Several studies suggest that fog can supply different amounts of
liquid water to tropical montane cloud forests. In some areas fog interception represents 99% of the water
input while in others only 3.5%. In general, fog interception is an important additional water source at sites
with regular and frequent occurrence of fog, contributing far more than one third of the bulk precipitation
in tropical montane forests (cp. e. g. CAVELIER & GOLDSTEIN 1989, JUVIK & NULLET 1993, CAVELIER et
al. 1996). In lower montane tropical rain forests an average of about 16% was measured (CAVELIER et al.
1996).
The following calculations are based on a comprehensive ecological and meteorological database
collected by a consultant to this report. A vegetation map was produced by analyzing over 1200 vegetation
plots. In addition, 16 meteorological stations along 4 transects inside the forest belt were established,
producing the first reliable weather data (rainfall, temperature, air humidity, radiation, wind speed etc.)
from this vegetation zone of Mt. Kilimanjaro. Using these data a map of mean annual rainfall and mean
annual temperature was created. According to the distribution of the different forest types, the annual
rainfall, the estimated amounts of cloud water collection and evapotranspiration (based on measured
vegetation density, altitude, climatic parameters and numbers given in literature e.g. LARCHER 1984,
CAVELLIER et al. 1997) the forest belt was divided in 11 eco-climatic zones. For the first time this
approach allows to estimate the water output of the 939 km2 of indigenous forest (excluding forest
plantations) of Mt. Kilimanjaro (Table 7).
Table 7. Hydrometrical data of the forest belt on Mount Kilimanjaro
Water Input Water Output
Rain (million m3) fog (million m3) Evapotranspiration (million
m3)
groundwater and streams
(million m3)
1,533.5 560.0 797.4 1296.1
73.3% 26.7% 38.1% 61.9%
The indigenous forests of Mt. Kilimanjaro receive 2093.5 million m3 water annually of which
73% is by rainfall and 27% by fog interception. The intercepted moisture has to be considered to be a net
gain, since the energy used in its evaporation from the leaf surfaces during fog-free periods would have
been used in transpiration of an equal amount of water from the soil (KERFOOT 1968). In contrast, half of
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
44
the amount of rainwater re-evaporates back to the air by evapo-transpiration. The circa 1,300 million m3 of
remaining water percolate into the groundwater or run off as surface flow into streams. The approximately
800 million m3 of water evapo-transpired are not lost for the ecosystem. The forest dampens the air,
leading to permanent high air humidity over the forest belt. This results in cloudiness and rain showers
even during the dry seasons. Therefore the forest stores water not only in its biomass and the forest soil,
but even in its surrounding air. This mechanism enhances the forest’s function as a water reservoir
regulating the outflow patterns of watercourses. Without such a permanent cloud cover over the forest
evapo-transpiration would be much higher (due to higher temperatures) and rain showers during the dry
season would be absent.
In his analysis of the value of East African forests in influencing climate and water supply
NICHOLSON (1936) estimates the condensing capacity of montane forests add up to at least 25% of the total
annual rainfall. This amount (in the case of the forests on Kilimanjaro equivalent to 383.4 million m3) has
to be added to the 560 million m3 water of fog interception, to get a more reliable impression about the
influence of the forest on the water balance. This gives 943.4 million m3 or a surplus of 146 million m3
water (nearly 10% of the rain water input) which forests on Kilimanjaro contribute more to the water
balance every year than comparable open areas. Table 7 further shows that fog interception is an important
factor in the hydrological balance of the mountain. About one quarter of the atmospheric water input in the
forests derives from this source. Without the cloud water collecting forests this water would be lost for the
mountain. If the surface and groundwater run-off is compared with only the “ordinary” precipitation, i. e.
rainfall, the role of forest for the water production becomes evident.28
Consequently, the loss of 13,000 ha of Erica forest since 1976 results in a water yield reduction
of about 58 million m3 of fog water annually. This number represents over 10% of the annual fog water
input of the entire forest belt or the equivalent of the annual drinking water demand of nearly three million
inhabitants on the mountain (this calculation is based on numbers given by UNITED REPUBLIC OF
TANZANIA & CES 2002). In this calculation, however, are neither the several 10,000 ha of destroyed
ericaceous bush land nor the montane forests, which have been lost due to logging activities included.
Since the Chagga with their irrigation system are highly dependent on a steady river discharge
changes to the water balance present a serious threat to their existence. During the dry seasons water
shortages especially on the lower foothills become increasingly common. Women and children have to
spend a big part of the day fetching water. Yet, the water demand grows rapidly. The hydrometric report of
the Hai district water supply Phase IV (UNITED REPUBLIC OF TANZANIA & CES 2002) referring to an
selected area on the south western, western and northern parts of the mountain presents the following
numbers: Currently, population in this area totals 132,258 inhabitants with a daily demand of 7,200 m³
water and it is expected to rise until 2015 to 162,570 inhabitants demanding daily about 8,900 m³ water.
Besides, the situation on Mt. Kilimanjaro affects the entire region. The Pangani River, one of
Tanzania’s largest rivers, provides water to the hydropower plants of Nyumba ya Mungu (8 MW), Hale
(17 MW) and Pangani Falls (66 MW), which generate some 20% of Tanzania’s total electricity output. A
water shortage during the dry periods would increase the number of power cuts which have already inhibit
economic prosperity. Fishing in Nyumba ya Mungu dam yields a maximum catch of approximately 4,000
tonnes annually. The river also supplies the large scale South-East Moshi rice scheme. Furthermore, the
28 Although in general water output by run off is lower than the input caused by rains, this relation varies
within the 11 distinguished eco-climatic zones. In the relatively dry submontane Croton-Calodendrum
forests the run-off counts only to 40% of the rainwater input. In higher altitudes with lower evapotranspiration
but higher fog interception this ratio turns: In the Juniperus forests above 2600 m the total
water run-off is 110% of the rainfall input. This is due to the additional water available from fog
deposition. In the Podocarpus, Hagenia and Erica forests this ratio lies even higher at 120%.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
45
southern slopes provide water to Arusha Chini sugarcane plantation. In Kenya, the Amboseli ecosystem
including the wetlands of Ol Tukai and Kimana, which support Masai pastoralists and an abundance of
wildlife, depend on the Kilimanjaro water supplies.
7.5.4.2 Other ecosystem services diminished by fire
Forest fires do not only reduce the water budget of the mountain, but they also directly and
indirectly destroy other goods and benefits. Forest fires burn huge amounts of precious wood including fire
wood, which people are allowed to collect, and timber, which people cut illegally. Besides, fires reduce the
beauty of the heathlands that attract tourists and destroy the flower trees for bees. Bee-keeping is important
on Mt. Kilimanjaro. An ethnobotanical study (HEMP 1999) showed that the Chagga make use of their plant
environment in a variety of ways. The plants serve as forage for households and agricultural purposes, and
many are used in medicinal applications either as drugs or for “magic” purposes. The montane forest is
home to many of such plants. In addition, repeated burning also modifies the nutrient balance of soils
(CRUTZEN & ANDREAE 1990).
7.6 Other threats to the Mount Kilimanjaro ecosystem
The climate related threats to the Kilimanjaro ecosystem need to be viewed in conjunction with
other stresses stemming from human activities as well as changed migration behavior and population
dynamics of big game. The results of a 2001 aerial survey (LAMBRECHTS et al. 2002) and the examination
ground data revealed that the forests of Mt. Kilimanjaro are heavily impacted by illegal logging of
indigenous trees in most areas below 2,500 metres on the western, southern and eastern slopes, and by the
establishment of forest villages in the western and northern slopes. Logging activities affect the entire
broadleaved mixed forests below an altitude of 2,500 metres on the southern slopes of Mt. Kilimanjaro.
The moist Ocotea forests which cover most of the southern slopes are subject to serious destruction due to
intensive illegal logging of camphor trees.29
In addition, large tracts of indigenous forests on the north-western and northern slopes have been
converted into forest plantation, using fast growing exotic tree species, such as pine and cypress. On the
north western slopes, the expansion of the forest plantations has reduced the indigenous forest belt to a
width of less than one kilometer. The majority of the clear felled compartments within the forest
plantations have not been replanted as required by the normal rotation management. To summarize, the
aerial survey revealed that the forest belt is threatened on its upper and lower border, thus shrinking on
both sides. This further exacerbates the adverse impacts on the water balance of the mountain.
Changing climate patterns not only influence landscape characteristics but also animal
distributions. The Kenyan Amboseli National Park is situated on the northern foothills of Mt. Kilimanjaro.
This area has experienced extensive habitat changes since the early 1960`s (ALTMANN 2002). These
include dramatic loss of tree and shrub cover which was partly caused by an increasing elephant population
and temperature changes. The “natural” landscape alterations are further enhanced by a steadily growing
Masai population on the whole northern foothill of Mt. Kilimanjaro. According to rangers of Kilimanjaro
29 During the survey, over 2,100 recently-logged camphor trees were counted. On the lower slopes bordering
the half-mile forest strip, there was no recent logging of camphor trees since these areas have already been
depleted. However, other indigenous tree species were targeted; some 4,300 recently-logged indigenous
trees were recorded. As a result, evidence of 57 landslides in the heavily impacted Ocotea forests was
recorded. To the east, above Marangu, 19 cleared fields have been opened up in the forest, and a large
number of livestock was seen up to 8 kilometers deep into the forest. There were fewer observations
recorded in the half-mile forest strip because this zone is virtually denuded of indigenous trees. Some areas
have been completely cleared. Logging activities also impact heavily the east and west sides of the
northern slopes; 574 recently-logged cedar trees were counted, as well as over 800 other indigenous trees.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
46
National Park elephant migration from the Amboseli National Park through the so-called “Kitendeni
corridor” into the forests of Kilimanjaro has increased. In addition, more elephant herds stay permanently
inside Kilimanjaro’s forests given the better conditions compared with the Amboseli basin.
A ground survey of the forests on the western and northern slopes of Kilimanjaro reveals that in
most places elephants and buffaloes are abundant. Besides former logging activities (the last sawmills
inside the indigenous forest were closed in the 1970´s) grazing patterns of big game cause a change in the
dense forest cover towards a mosaic of openings and patches of closed canopies. If the openings become
larger, forest regeneration is impeded. In the long term, this development will destroy the forest and change
it into a bush land with scattered trees with all the known disadvantages.
7.7 Scenarios for 2020 with respect to fire impact
Assuming that the observed trends in fire frequency continue in a linear mode the following
scenarios are probable. Regarding the upper forest line, most of the remaining subalpine Erica forests will
have disappeared within five years. As a result, Mt. Kilimanjaro will have lost its most effective water
catchment area. Compared with the situation of 2000, this means an annual loss of 16.2 million m3 fog
water. Subsequently, the upper forest line will retreat more slowly because on the one side mostly broadleaved
forests remain, which are to a much lesser degree inflammable and because on the other the lower
areas receive an increasing amount of precipitation. Nevertheless, an average retreat of the upper forest line
of about 100 m in altitude seems to be probable by 2020, when the glaciers will have melted. Forest
regeneration will completely be inhibited and regressive succession will prevail, as illustrated in Figure 15,
substituting increasing areas of Erica heathland with low layered Helichrysum cushion vegetation.
Figure 15. Forest succession after continued fires
Linear increasing temperature and decreasing precipitation in combination with increasing
logging activity will also result in more forest fires, which will heavily destroy the lower forest zone up to
an average altitude of 2000 m (for example closed forests will be replaced with an open bush that cannot
carry out the necessary ecological functions). These trends will cause a further shrinking and fragmentation
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
47
of the forest belt. Especially on the western and eastern but also on the south eastern slopes the forest zone
will be interrupted by large gaps with all the known disadvantages for wildlife and the ecological balance.
7.8 Climate risks in perspective: shrinking glaciers versus enhanced fire risk
With an average thickness of 30 m as indicated by ice core drilling (THOMPSON 2000,
THOMPSON et al. 2002) and likewise observations from KASER et al. (under review) the existing 2.6 km2 of
glaciers constitute a water volume of about 72 million m3. However, most of this water is not available for
the lowlands since most glacier ablation occurs as sublimation and the remaining melting water evaporates
immediately into the atmosphere (KASER et al. under review). If one quarter (or 18 million m3) of glacier
water would percolate into the rivers, an average annual water output of about 0.9 million m3 would result
until 2020, when the glaciers are predicted to have being melted. But even then, one can still expect
precipitation on Kibo to feed springs and rivers although not so continuously and to a lesser degree.
In contrast, Mt. Kilimanjaro receives 58.5 million m3 less water each year due to forest depletion
and vegetation changes incurred as a result of forest fires since 1976. The number is likely an
underestimate since the calculation assumes the timberline to remain stable for the next 20 years, which is
very unlikely. Moreover, the calculation did not include the several 10,000 ha of ericaceous bush land
which has been substituted by low Helichrysum cushion vegetation.
Summarizing, compared with roughly 1.3 billion m3 of water, contributed every year by the 1000
km2 of indigenous forest, the consequences of losing 2.6 km2 of glaciers providing an annual water output
of about 0.9 million m3, the loss of Mt. Kilimanjaro’s ice cap is negligible. Still, the melting glaciers are
certainly an alarming indicator of severe environmental changes on Mt. Kilimanjaro.
8. Policy responses for Mount Kilimanjaro
The preceding section has laid out the complex interaction between climatic and other stresses
that are causing significant changes in the Kilimanjaro ecosystem and adversely impacting the ecosystem
services it provides. While the most visible impact – glacier retreat – may only have limited consequences,
enhancement of fire risk that has resulted from climatic trends and human interference poses significant
threats not only to the viability of the ecosystem, but also neighboring regions through its critical influence
on regional water resources. Some of these changes (such as glacier retreat) may be inevitable, but others
can be managed to make the ecosystem more sustainable. However, this requires a comprehensive set of
policy responses that take into account the underlying demographic, environmental and climatic stresses.
This section starts with a brief discussion of policy responses to the shrinking ice cap, to the general
environmental threats facing the Kilimanjaro ecosystem, as well as to the enhancement of fire risk. Finally,
given that human livelihood choices might provide the trigger for forest fires, the section reviews alternate
livelihood strategies that might alleviate some of these stresses.
8.1 Policy responses to the shrinking ice cap
The melting of Kilimanjaro’s ice cap receives much attention. Articles in local, regional and
international newspapers described the results of ice core drilling by American scientists during the years
2000-2001. A study on climatic changes in the context of the receding ice level on Mt. Kilimanjaro was
chosen to be a topic among the proposed research priorities for the next years of TANAPA (Tanzanian
National Parks). However, there is obviously nothing that could be done by way of policy responses to
avoid or even delay its eventual loss.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
48
8.2 Policy responses to general environmental threats
The vulnerability of the Kilimanjaro to climate change can be alleviated at least partially by
reducing other environmental stresses on it. Since Mt. Kilimanjaro is a UNESCO World Natural Heritage
Site, the environmental problems of this unique volcano have attracted international attention and a number
of conservation projects are already under way. The United Nations Development Programme (UNDP) and
the United Nations Foundation (UNF) have jointly disbursed 264,000 US$ to the Tanzanian government
for running different environmental conservation projects and promoting eco-tourism on Mt. Kilimanjaro.
A first comprehensive inventory of threats - including wild fires – to Mt. Kilimanjaro was taken during an
aerial survey in September 2001.30 As a result of this survey it was decided by the Ministry of Natural
Resources and Tourism that the forest belt of Mt. Kilimanjaro and Mt. Meru will be taken away from the
Forest Department and included into Kilimanjaro National Park and Meru National park respectively. A
similar shift in management from Forest Department to Kenya Wildlife Service in 2000 on Mt. Kenya
resulting from an aerial survey had dramatic consequences. The illegal cutting dropped drastically.
Comparing the situation in 2002 with 1999 logging of camphor was reduced by 96%, logging of cedar by
73% and logging of other indigenous trees by 92% (LAMBRECHTS, pers. com.). It is therefore expected that
such a shift in management in the Kilimanjaro will have similar effects.
While these efforts are underway, several important challenges remain. One major threat is the
cross-border migration of big game from the Amboseli National Park. There is a need for the formulation
of a cross-border response between Tanzania and Kenya. An initial step would be to survey and count the
numbers of elephants and buffalos. Based on these figures further steps, including control or reduction
measures have to be taken into account. Also, the animals should be provided with adequate areas in the
Amboseli basin by restricting permanent settlements of the Masai.
8.3 Policy responses to enhanced fire risk
There are two general ways to cope with wild fires: first, reduction of fire risk and second,
fighting of fires. The main area of interest in this respect is the upper montane and subalpine zone on Mt.
Kilimanjaro, where fires are most common. The first aim is to protect the still existing upper montane and
subalpine forests from further destruction by fire. Second, since the potential climatic and historic tree line
is much higher than the actual fire-induced one it should be tried to increase the forest area and to push up
the actual forest line to areas that were formerly covered by forests.
8.3.1 Responses to forest destruction
While the natural montane forest on the Kilimanjaro has had protected status since the early
twentieth century, the cutting of indigenous trees continued to increase until 1984 when the severe forest
destruction led to the banning of all harvesting from the catchment forests on Kilimanjaro by a Presidential
Order. Prior to the ban local people were used to entering the reserve without restriction to utilize its
resources. Therefore, the new restrictions were not effective and encroachment activities have continued
illegally. Nevertheless, general awareness for the protection of Mt. Kilimanjaro’s natural resources
especially of its forests is high among local people, governmental and non-governmental institutions.
Everywhere Panda miti! (Plant trees!) stickers can be seen in offices, governmental cars and schools. The
government awards prizes to those villages that have planted the most trees. Unfortunately, no indigenous
30 The request for the aerial survey of the forests of Mt. Kilimanjaro was originally presented by UNDP/GEF
Small Grants Programme, New York. The objective was to identify the type, extent and location of the
threats to the forests and provide a baseline assessment for the newly developed Community Management
of Protected Areas Conservation Project (COMPACT). The aerial survey on Mt. Kilimanjaro was
supported by the Ministry of Natural Resources and Tourism and the Tanzania National Parks.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
49
trees are used for such competitions. The churches, which have great influence on people, also support
afforestation measures.
Many NGO’s like the Tanzania Association of Foresters (TAF) run reforestation projects on the
mountain. Some villages like Mbokomu and private institutions like the Maua Seminary do the same,
mostly without any support from the forest department in Moshi and sometimes even against the
authorities31. Another significant development was the initiation of the catchment forestry project in 1988.
The first phase of the project (1988-1992) focused on improving catchment forest management by
establishing an inventory of the forest by resurveying, replanting boundaries, mapping and through
reviving the management and protection activity. The second phase (1992-1996) tried to improve
management of the forest through boundary marking, mapping, policing and people’s participation. Efforts
are currently being made to involve the local communities in the management of the forest reserve.
Villages adjacent to the forest have now the responsibility to watch that there is no encroachment into the
forest. Village conservation committees are responsible for establishing tree nurseries, to organize patrols
into the forest, to mobilize the people for fire fighting and to control the entrance into the forest by issuing
permits. Timber that has been confiscated during the patrols becomes (partly) property of the village
(MISANA 1999). However, these activities of the forest department were not very successful, which
became apparent during the aerial survey of LAMBRECHTS et al. (2002).
In 2000, the GEF Small Grants Program implemented by UNDP, in collaboration with the United
Nations Foundation (UNF), launched the Community Management of Protected Areas Conservation
Project (COMPACT). The main objective of COMPACT is to demonstrate, by complementing and adding
value to existing conservation programs, how community-based initiatives can significantly increase the
effectiveness of biodiversity conservation in and around World Natural Heritage Sites (WNHS). The
project also aims at (i) enhancing the capacities of local organizations and NGOs whose existence and
future prospects are closely linked to these protected areas; (ii) increasing local awareness of, and concern
for, the protection of WNHS, (iii) promoting and supporting communication and cooperation among park
management personnel and other concerned groups, particularly local communities, (iv) increasing general
understanding of the synergies between community development and the role of globally significant
protected areas in contributing to sustainable development, and (v) drawing lessons from project
experience that can be shared widely at local, national and international levels.
Mount Kilimanjaro is one of six World Natural Heritage Sites on three continents participating in
COMPACT. A common methodology to prioritize COMPACT interventions at the six sites has been
developed. It involves a participatory approach to identify the main threats to the protected area, and to
assess the types of activities that may be carried out by local communities to address those threats while
improving their quality of life and livelihoods. This planning process involves a wide range of stakeholders
of Mt. Kilimanjaro: community-based organizations, local and national NGOs, local and national
authorities with management responsibilities of the mountain, and other programs and projects present in
the area. It is too early to assess the effect of COMPACT, although the expectation of the funding agencies
31 The Maua Seminary, a Franziscanian monastery leading a vocational school, is a good example for
possibilities and problems of private engagement in environmental projects. The Padres of this monastery
are very active replanting the whole valley of the Mue river inside the half-mile forest strip. The trees,
although paid and planted by private effort are the property of the government. To cut down expenses and
to get local people involved and interested in the project, they tried to use the Shamba (Taungya) system
practices as it is done in other forest plantations on the mountain. Many forest plantations in West and
North Kilimanjaro have usually been established by allowing local farmers to inter-crop annual agricultural
crops with tree seedlings in forest plantation areas until the third year of tree growth. But the Padres have
been unsuccessful fighting for eight years to get the permission from the forest office to use the Shamba
system. Since they cannot pay a lot, incentives for local people to cooperate are not very high and the
afforestation of the valley takes long.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
50
and the host government is that the project empowers local communities to participate effectively in
reversing extractive pressures that have adverse impacts on the mountain’s resources.
8.3.2 Responses to forest fires
Since most areas heavily affected by fire - Erica forests and bush lands, are located inside the
Kilimanjaro National Park (KINAPA), effective management of this park is one of the keystones to reduce
the fire risk on Mt. Kilimanjaro. During the 1997 fire outbreak, a contingent of 700 fire fighters including
the Tanzanian army was needed to extinguish the fire. A special fund of US$5,000 was subsequently set up
to fight fires on the mountain. Fundraising to collect the money has targeted various donors including the
business community, environmental institutions and other interested parties. However, since large amounts
are necessary, the Tanzanian government is seeking new donor funding to conserve forests on Mt.
Kilimanjaro (THE EAST AFRICAN, 9.10.2002).
One step towards fire prevention was already taken by the national park authorities by banning
camp fires. However as most fires are caused by pit-sawyers, poachers or honey gatherers more effort has
to be undertaken to cut down these illegal activities. A paramilitary ranger troop patrol the forests could
serve as an effective deterrent, as proven successful on Mt. Kenya.
The construction of open strips as fire breaks seems generally not suitable for the Kilimanjaro
due to the very difficult, inaccessible and steep slopes. One suitable area however exists on the south
eastern slopes where there is a plateau around 2700-2800 m with moorland vegetation, formed by tussock
grasses, occurring at the fringe of the forest. In this area grassland fires affecting the bordering forests are
very common and the construction of open strips to prevent fire from spreading into the forest seems to be
possible and effective. At the lower forest boundary, fire lines could be reactivated and cleared before the
dry seasons.
There is also a need for better forest fire early warning system on Mt. Kilimanjaro. One
possibility is the establishment of fire observation points such as fire towers on higher hills or ridges near
existing ranger posts or tourist camps and huts. The fire-fighting capabilities could be significantly boosted
with the provision of one or two small airplanes, as is the case for Mt. Kenya. A national park of the size,
topography and importance like Kilimanjaro cannot be managed properly in many respects (e. g.
observation of poachers) without such modern equipment. The fire fighting equipment also needs to be
supplemented by a suitably equipped task force.
As mentioned earlier, many NGO’s, some villages and private institutions run re- and
afforestation projects on the mountain. However, a large scale effort is missing. The forest department in
Moshi, which acts as the official initiator of such projects, has failed completely to protect the indigenous
forests. Nurseries are used as maize fields and illegal cutting of timber is not prohibited effectively. No
government supported tree nursery exists on the southern slopes. A first step might be the recent decision
to incorporate the forest belt into the National Park, since this institution should be able to employ well
paid rangers to take care for its resources. In addition, bearing in mind the financial resources – the annual
park income is US$ 6.5 million (20,000 park visitors stay on average 5 days paying a daily entrance fee of
US$ 65) park authorities should ensure that a certain fixed share of the income flows directly back to
Kilimanjaro.
The local people should also be involved and get benefits out of the forest to be more interested
in its protection. As shown on Mt. Kenya and some Kilimanjaro villages adjacent to the forest, people
should be given the responsibility to ensure that there is no encroachment into the forest, to organize
patrols and to mobilize people for fire fighting. These efforts should be compensated by rewarding the
participating villages with timber which has been confiscated during patrols.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
51
Likewise, the half-mile forest strip should be managed similarly by involving local people. This
strip of 8,769 ha on the southern and eastern slope, ranges between the plantation belt and the forest
reserve. It is meant to provide timber and firewood, and in some areas even pines, cypress and eucalyptus
have been planted. However, people use this strip mainly in an uncontrolled way to collect fodder for their
livestock. Sometimes the area even serves as pasture land. Since nobody feels responsible, this area is
highly degraded and not managed properly. This is quiet different from the situation in the beginning. The
half-mile forest strip was established resulting from a request of the Chagga Council in 1941 brought to the
colonial government in response to the need of unrestricted availability of forest products, which could not
be gained in the forest reserve. During the 20 years of managing the forest, the local people contributed
substantially in planting trees, demarcating the boundary and fighting fires. In return, they were allowed to
obtain forest products freely or at minimal costs (MISANA 1999). In 1962, however, management of the
strip was transferred to the District Council and later (1972) the central government took control of the
strip, which was then managed by the Forest Division, which restricted local people from collecting forest
products freely. This situation has not changed up to now, although the management of the strip has been
referred to the district councils again in 1987, together with the Forest Division.
Currently, a discussion has started to give the area back to the villages, which has been already
done in some areas. This may offer the possibility that the local population will take more care of the land
in the future than today. Since many fires originate in the agricultural land surrounding the forest and in the
half-mile forest strip, it would reduce the fire risk for the forest as well. However, it cannot be excluded
that this land would be abused as agricultural land and for settlement. Therefore, governmental control
would be necessary. In any case, this huge area - when properly used and planted with timber trees - has a
high potential to reduce the pressure from the indigenous forests.
Regarding replanting it appears to be a necessity to employ well-trained foresters, perhaps from
outside the country, to start forestation projects and to educate local foresters. Especially the choice of
suitable tree species is of importance. A variety of different species, not only the widely used exotic ones,
but also indigenous could be used. Riverine areas could exclusively be replanted with indigenous trees, and
existing natural forest patches in the half-mile forest strip situated mostly near rivers should not be
replaced with forest plantations.
Right now, the forest plantations in West and North Kilimanjaro are not managed properly.
During the aerial survey (LAMBRECHTS et al. 2002) it became evident that over 50% of the Shamba system
areas are not under tree growing, either because replanting was not successful or because it was not
undertaken at all. Since the production of timber is the primary goal and growing vegetables only a
secondary one, the share of tree-planted areas in the forest plantations has to be much higher, even though
this implies the removal of illegally erected villages inside the forest reserve.
The Chagga home gardens (vihamba) are an old and very sustainable way of land use that meets
several different demands. Besides crop production, the sparse tree layer provides people with fire wood,
fodder and timber. However, the high demand for wood and the introduction of coffee varieties that are
sun-tolerant endangers this effective system. In some areas of the mountain (e. g. on the eastern slopes) the
trees in the banana fields are very scattered or already missing. Therefore it seems to be necessary, in order
to reduce the pressure on the forest, to support the tree planting in the Chagga home gardens with their
unique agro forestry system. There could also be a program that rewards farmers to have a certain share of
their land covered with trees. As the banana belt is nearly as extensive as the forest reserve, this will
certainly have major effects in terms of forest protection and water balance. In combination with new
marketing and farming strategies for growing organic coffee using traditional methods an advertising
campaign should be started. The campaign should point out that the consumer buys high quality
ecologically grown coffee supporting not only sustainable land use and an old African cultural heritage but
he is also protecting the rain forest. A certain share of the coffee prize should be used to run this
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
52
environmental Chagga home garden program. Government programs and donor agencies should cover any
additional program costs instead of using financial resources for other expensive and in the long-term less
effective projects such as dairying, which are not suitable for the Chagga home gardens.
Finally, a comprehensive, holistic environmental development plan focusing on fire risk and
forest destruction while defining conservation strategies to ensure the long term sustainability of the
mountain should combine all the different requirements, constraints and aims under one leading guideline.
The high complexity of Mt. Kilimanjaro’s ecosystem requires the expertise of scientists familiar with the
biodiversity patterns and ecological conditions of the mountain for the preparation of such a report.
8.4 Promotion of ecosystem friendly livelihood opportunities
A key adaptation response to the threats facing the Kilimanjaro ecosystem is also to reduce
human and livelihood pressures that make it vulnerable to other stresses such as climatic change. Humans
have continuously occupied the slopes of Mt. Kilimanjaro for the last 2000 years (SCHMIDT 1989).
However, the population has multiplied by 20 during the 100 years since 1895. In general, the growth rate
is exponential, albeit it is slowly decreasing since 1978. The annual average population growth has been
2.1% between 1978 and 1988 and decreased to 1.6% between 1988 and 2002. In 1991 GAMASSA estimated
the doubling of the population within 39 years. The population increase was much higher in urban than in
rural areas. While the population has doubled between 1967 and 2002 on Kilimanjaro, the population in
Moshi town multiplied 5 times during that same period.
The overall population density for the four districts that comprise Mt Kilimanjaro region was 198
people per km2 in 2002. If population density would be based upon actual land availability, this number
would be approximately 331 people per km2. Most of the population is concentrated at an altitude between
1100 and 1800 m. Here, densities varying from 500 to 1000 people per km2 have been recorded in certain
places (TIMBERLAKE 1986, FAO 1986). From these data it is evident that every effort in environmental
protection, which ignores the demands of a still fast growing population, will fail. Therefore, it is necessary
to boost livelihood prospects in sectors that do not pose threat to the Kilimanjaro ecosystem.
Today, the bulk of development processes is departing from the mountain, although most of the
population still remains there. Manufacturing in the region has collapsed following the closure of most
leading factories. Even in the tourism market neighboring Arusha out-competes Moshi in the Kilimanjaro
region.
Agriculture, the livelihood for most residents, accounts for over 85% of the total regional income
with coffee being the main cash crop. Lately, however, coffee has become less profitable due to traditional
farming techniques and very low coffee prices. Today (March 2003) a farmer gets only 400 TSH
(equivalent to 0.4 US Dollar) for one kg of coffee and many farmers think about replacing their coffee
trees with other crops such as passion fruits. In the 1970s 35,000 tonnes of coffee were harvested annually
in the region whereas today only 12,000 to 15,000 tonnes are produced. Still, coffee accounts for over 60%
of the region’s income and authorities plan to raise coffee production to over 45,000 tonnes. The Coffee
Revival Programme, which was launched in 1998 aimed at producing 100,000 tonnes. The strategy
involves the formation of coffee revival committees and replacement of old coffee trees (THE EAST
AFRICAN, August 26-September 1, 2002). However, the question whether this could be the solution
remains open considering the over supply on the world market. A better way seems to be to raise the
quality. According to Tanzania Coffee Board officials (THE EAST AFRICAN, August 26-September 1, 2002)
new crop marketing and farming strategies aim at growing organic coffee through traditional methods
without any use of pesticides and artificial fertilizers.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
53
Despite its large cattle herds and successive government efforts to promote dairying, Tanzania is
a net importer of dairy products (MDOE & WIGGINS 1997). Since independence in 1961 the government of
Tanzania has tried to encourage more domestic milk production to achieve self-sufficiency. Since most
cattle are stall-fed, and fodder has to be collected and brought from remote areas, large scale dairying in the
Chagga home-gardens offers no alternative. Expensive dairy development projects therefore may not be
the right way to improve livelihood in the submontane banana zone in the long-term.
In terms of tourism Kilimanjaro National Park (KINAPA) is a major tourist attraction in
Tanzania and earns the most foreign exchange of any National Park in Tanzania (NEWMARK & NGUYE
1991). Most visitors are mainly interested in reaching the summit of Kibo, known as Uhuru Peak, the
highest point in Africa. In the year 1976 5,000 people tried to reach the summit (CARLÉ 1977). In the year
2002 this number had increased to 20,000 (Chief Park Warden KINAPA, pers. comm.). Since its
establishment in 1972, the number of visitors of KINAPA has multiplied by five. Today, about 100,000
people – porters and tourists combined - frequent the alpine areas of Kilimanjaro every year. Such
increasing numbers of visitors have certainly effects on the environment. Especially the alpine zone with
its highly specialized flora and fauna is a very sensitive ecosystem. Since a national park is meant for
nature protection, it appears that tourism has reached a level which should not be exceeded. Therefore,
alternatives to the mountain climbing tourism have to be explored.
During the last years a strong development of ecotourism could be observed world wide.
Ecotourism can not only help in protecting the environment especially on Mt. Kilimanjaro put also it
allows local population to participate in its economic potential. In the long run this type of tourism could
be an alternative to the usual climbing tourism on Mt. Kilimanjaro. Another possibility is one or two day
guided nature trips to the forest or to the lower alpine zone organized by the Kilimanjaro National Park.
Many tourists would prefer to visit only the lower vegetation zones of the park instead of climbing. A
special training program for guides should provide them with sufficient, basic knowledge about main
vegetation types, flora and fauna to explain the mountain ecosystem to interested tourists. For the
promotion of tourism it is of fundamental importance to generally improve tourist facilities and in
particular to raise the quality of Tanzanian tourist hotels.
To summarize, due to a rapidly growing population, the decline in coffee production, and the
collapse of manufacturing industry, the Kilimanjaro Region, which once has been one of Tanzania’s
leading economical areas, is now among the most poverty stricken. The region’s annual per capita income
is less than TSH 96,390 (US$ 96) (THE EAST AFRICAN, August 26-September 1, 2002). The most
promising economic alternatives for the region currently appear to be the promotion of high quality organic
coffee rather than necessarily increasing the quantity of production, the production of new cash crops such
as passion fruits and flowers, as well as the improvement of eco-tourism and improvement of tourism
infrastructure.
9. Concluding remarks
Climate change poses significant risks for Tanzania. While projected trends in precipitation are
uncertain (and may differ for various areas of the country), there is a high likelihood of year-round
temperature increase, as well as sea level rise. The sectors potentially impacted by climate change include
agriculture, forests, water resources, coastal resources, human health, and energy, industry and transport.
Given the low level of human development, extreme poverty, and high dependence on agriculture and
natural resources, Tanzania may be quite vulnerable to projected climatic changes.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
54
9.1 Differentiated adaptation strategy
While uncertainties in climate change and impacts projections pose a challenge for anticipatory
adaptation for any country, Tanzania’s case has several specific characteristics that may argue for a
differentiated adaptation strategy.
First, the climate change projections on which all national impact and vulnerability assessments
(all the way to the Initial National Communication of 2003) rely on an older generation of climate models
and scenarios (circa early 1990s). A preliminary analysis based on more recent climate models conducted
as part of this study concludes that temperature increases might be somewhat lower than (although broadly
consistent with) the estimates used in the National Communication and the National Climate Change
Action Plan. Updating of climate scenarios and impact projections through the use of multiple and more
recent models might therefore be advisable prior to the formulation of aggressive (and potentially
expensive) adaptation responses. This should not however affect “no regrets” adaptation measures such as
leakage prevention and water conservation.
A second characteristic feature of Tanzania is that certain sectors are projected to experience both
negative and positive impacts under climate change – for example, while production of maize is projected
to decline, the production of two key cash crops (coffee and cotton) which contribute significantly to the
GNI is projected to increase. Similarly, while stream-flow declines are projected to decline in two of three
key river basins (Ruvu and Pangani), they are projected to increase in the third (Rufiji). The implication for
adaptation therefore may be to not only cushion adverse impacts, but also to harness positive opportunities.
This suggests consideration of an enhanced portfolio of linked-adaptation responses – for example a
strategic shift from maize to cash crops over the medium term, and inter-basin transfers in the case of water
resources. Such strategic shifts however may entail economic and dislocation costs – and therefore require
careful screening, particularly with regard to their effects on equity and rural livelihoods. More rigorous
testing of particular crop and stream-flow projections may also be advisable prior to undertaking such
adaptation responses.
A third key characteristic is that unlike most other countries where the need for adaptation relies
on projections of future impacts, some discernible trends in climate and attendant impacts are already
underway in Tanzania. Such impacts – as is the case of the Kilimanjaro ecosystem - argue for more
immediate adaptation responses as opposed to a “wait and see” strategy.
9.2 Climate change and donor portfolios
Tanzania receives close to a billion dollars of development assistance annually. An analysis of
donor projects using the OECD/World Bank Creditor Reporting System (CRS) database reveals that
roughly 12 – 25% (in terms of investment dollars) and 20-30% (in terms of number of projects) of donor
portfolios in Tanzania may be potentially affected by climate change. This includes both activities in
sectors which may themselves be impacted by climate change, as well as those projects and other activities
which may influence the vulnerability of natural or human systems to climate change. These numbers are
only indicative at best, given that any classification based on sectors suffers from over-simplification.
Nevertheless, such measures can serve as a crude barometer to assess the degree to which particular
projects or development strategies may need to take climate change concerns into account. Several donor
strategies in fact already do make frequent references to the impacts of climate variability (such as El
Nino) and linkages between such events and economic performance. There is however as yet no explicit
reference to climate change. The (relatively few) development projects that were reviewed for this report
did not pay attention to the risks associated with climate change.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
55
9.3 Attention to climate change concerns in national planning
At the national level meanwhile Tanzania has a draft National Action Plan on Climate Change
since 1997 that highlights priorities on three time-scales (Short term 1-2 years; Medium term 2-5 years;
and Long term 10-20 years). The short term primarily focuses on capacity building through conferences;
the medium term flags “projects internalizing climate change aspects… especially those reducing GHG
emissions” and recommends the introduction of economic instruments to accomplish such goals; and the
long term identifies major infrastructure projects in energy, transportation, and coastal zones as priority
areas. While the sequencing appears reasonable, the plan remains short on specific details on how it may
be implemented. Tanzania’s recent National Communication to the UN Convention on Biodiversity, and
its report to the World Summit on Sustainable Development only make tangential references to climate
change. Its Poverty Reduction Strategy paper (PRSP) does explicitly recognize the significance of current
climatic impacts on the poor, although the potential links between climatic factors and performance of key
sectors (such as agriculture) are generally not discussed.
There is however considerable synergy between priorities of at least some national plans and the
measures that may be required for adaptation. Specifically, the National Environmental Policy which
emphasizes measures to improve the resilience of the agricultural sector, the National Water Policy that
highlights efficient water use and water conservation, and the National Forest Policy which highlights
forest conservation and biodiversity preservation. However, some of these goals (such as Water
Conservation) have been articulated in previous plans, but have not been successfully implemented.
Therefore, despite the obvious synergies between such policies and climate change adaptation, a key
obstacle facing successful “mainstreaming” is successful implementation.
9.4 Climate risks in perspective on Mount Kilimanjaro
The second half of this report discusses in-depth climate change impacts and policy responses on
the Mount Kilimanjaro ecosystem – Africa’s highest mountain and largest glacier, a biodiversity hotspot,
and a UNESCO World Heritage Site. Glaciers on Mount Kilimanjaro have been in a general state of retreat
on account of natural causes for over a hundred and fifty years. A decline in precipitation coupled with a
local warming trend that has been recorded in the second half of the twentieth century accelerated their
retreat, and the ice cap is projected to vanish entirely by as early as 2020. While the symbolism of this loss
is indeed significant, this analysis concludes that the impact of such a loss on the physical and socioeconomic
system is likely to be very limited. The present glaciers are already very small, and cover an area
which is only 0.2% of the forest belt on Mount Kilimanjaro. Glaciers do not feed any major rivers, and
even when they would have melted the mountain will still receive precipitation. Further, even without
glaciers Mount Kilimanjaro will remain the world’s highest free standing mountain and with Africa’s
highest peak. Therefore, it is unlikely that the loss of glaciers would have a significant long-term impact on
tourism. It must however be emphasised that ice-cores on the Kilimanjaro are a repository of paleoclimatic
records, and valuable climatic records would be irreplaceably lost with the loss of the ice cap.
The increase in temperatures and a concomitant decline in precipitation have also significantly
enhanced the intensity and risk of forest fires on the Kilimanjaro. Climatic changes have not shifted
vegetation zones upwards as in the case of other mountains, but on Mt. Kilimanjaro they have pushed the
upper forest line downward as a result of increase in forest fire risk and intensity on the upper fringes of
the forest. A whole vegetation zone, the ericaceous belt, has moved downwards since 1976 by several
hundred meters, substituting 13000 ha of forest. The replacement of the fog intercepting forest belt by low
lying shrub has already seriously impacted the hydrological balance of the mountain as fog intercepting
cloud forests play a key role in the water budgets of high altitude drainage basins. The decline of 13000 ha
of cloud forest since 1976 has already resulted in a reduction of the water yields of about 58 million m3 of
water every year. This constitutes about 10% of the annual fog water input of the whole forest belt. Not
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
56
included in this calculation are the several thousands of hectares of destroyed ericaceous bush land and the
loss of montane forests due to human activities such as logging. These impacts have implications that
extend beyond the region as it feeds the Pangani river, one of Tanzania’s largest, which is responsible for
20% of Tanzania’s electricity output.
Looking into the future, a continuation of current trends in climatic changes, fire frequency and
destructive human influence most of the remaining subalpine Erica forests could disappear within five
years. With this, Mt. Kilimanjaro will have lost its most effective water catchment area as fog interception
is of highest importance in the Erica forests. A further retreat of the upper forest line by about 100 m
altitude seems to be probable until 2020. Increasing logging activity in combination with a higher number
of forest fires is also expected to destroy the lower forest zone up to an average altitude of 2000 m. This
will result in a further shrinking and fragmentation of the forest belt.
9.5 Policy responses for Mount Kilimanjaro
While glacier retreat is inevitable and cannot even be delayed, forest fire risk can indeed be
reduced. Climate change only adds to the urgency of fire prevention and control, as well as forest
conservation activities on Mount Kilimanjaro. Among the measures identified by this report are
institutional measures such as the inclusion of the forest belt into the Kilimanjaro National Park and
creation of a paramilitary ranger group (as in Mount Kenya) to deter logging, as well as better investments
in early warning systems, particularly the purchase of one or two aircraft for aerial surveillance. There is
also a need to limit cross-border migration of big game from neighboring Amboseli, which is adding to the
stress on the Kilimanjaro ecosystem.
In addition to such piecemeal solutions there is an urgent need to better understand the livelihood
needs of the local population to engage them more successfully in conservation and fire-prevention efforts.
For example, an earlier policy response – the banning of camp-fires – did not have the desired effect
because most of the fires were actually being lit not by mountaineers, but by honey collectors. A more
sustainable solution therefore needs to identify viable livelihood opportunities that take some of the human
pressures away from the forest. Creative solutions to boost local incomes, such as provision of incentives
to switch to more lucrative specialty coffee production, may therefore be part of a package of responses
that may help reduce the pressures on activities like logging and honey collection. Finally, there is a critical
need to develop a comprehensive and holistic development plan focusing on fire-risk and forest destruction
as well as to identify conservation strategies to ensure the long term sustainability of the valuable resources
of Mount Kilimanjaro.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
57
APPENDIX A: PREDICTIVE ERRORS FOR SCENGEN ANALYSIS FOR TANZANIA
The table below shows the predictive error for annual precipitation levels for each SCENGEN
model for each country. Each model is ranked by its error score, which was computed using the formula
100*[(MODEL MEAN BASELINE / OBSERVED) - 1.0]. Error scores closest to zero are optimal. The six
models with the highest error scores from the estimation were dropped from the analysis.
Predictive errors for each SCENGEN model for Tanzania
Average error32 Minimum error Maximum error
Models to be kept for estimation
ECH3TR95 7% 3% 12%
ECH4TR98 13% 1% 27%
CCSRTR96 14% 5% 26%
HAD3TR00 18% 7% 30%
CERFTR98 22% 18% 25%
BMRCTR98 23% 2% 45%
HAD2TR95 24% 10% 42%
GFDLTR90 25% 1% 37%
CSI2TR96 32% 24% 40%
PCM_TR00 34% 7% 45%
CSM_TR98 35% 19% 57%
Models to be dropped from estimation
IAP_TR97 40% 7% 93%
GISSTR95 48% 4% 125%
LMD_TR98 63% 29% 100%
CCC1TR99 73% 53% 98%
W&M_TR95 94% 32% 136%
MRI_TR96 132% 94% 154%
32 SCENGEN outputs data for 5×5 degree grids. To estimate for an entire country, a 10×10 degree area was
used and the data output from the resulting four 5×5 grids were averaged. The maximum and minimum of
these four 5×5 grids are also reported.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
58
APPENDIX B: LIST OF PURPOSE CODES INCLUDED IN THE SELECTION OF CLIMATEAFFECTED
PROJECTS, ORGANIZED BY THE DAC SECTOR CODE
DAC
code
General sector name Purpose codes that are included in the selection
110 Education -
120 Health 12250 (infectious disease control)
130 Population -
140 Water supply and Sanitation
14000
14010
14015
14020 (water supply and sanitation – large systems)
14030 (water supply and sanitation – small systems)
14040 (river development)
14050 (waste management/disposal)
14081 (education/training: water supply and sanitation)
150 Government & civil society 15010 (economic & development policy/planning)
160 Other social infrastructure and
services
16330 (settlement) and
16340 (reconstruction relief)
210* Transport and storage All purpose codes
220 Communications -
230 Energy 23030 (renewable energy)
23065 (hydro-electric power plants)
[23067 (solar energy)]
23068 (wind power)
23069 (ocean power)
240 Banking and financial services -
250 Business and other services -
310 Agriculture, forestry, fishing All purpose codes
320 Industry, mining, construction -
330 Trade and tourism 33200 (tourism, general)
33210 (tourism policy and admin. management)
410 General environment protection 41000 (general environmental protection)
41010 (environmental policy and management)
41020 (biosphere protection)
41030 (biodiversity)
41040 (site preservation)
41050 (flood prevention/control)#
41081 (environmental education/training)
41082 (environmental research)
420 Women in development -
430 Other multi-sector 43030 (urban development)
43040 (rural development)
510 Structural adjustment -
520* Food aid excluding relief aid 52000 (dev. food aid/food security assist.)
52010 (food security programmes/food aid)
530 Other general programme and
commodity assistance
-
600 Action relating to debt -
700* Emergency relief 70000 (emergency assistance, general) #
710* Relief food aid 71000 (emergency food aid, general) #
71010 (emergency food aid) #
720* Non-food emergency and
distress relief
72000 (other emergency and distress relief) #
72010 (emergency/distress relief) #
910 Administrative costs of donors -
920 Support to NGOs -
930 Unallocated/unspecified -
* sector codes that are excluded in the second selection (low estimate).
# purpose codes that are included in the emergency selection
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
59
APPENDIX C: REVIEW OF SELECTED DONOR STRATEGIES FOR TANZANIA
C. 1 United National Development Program (UNDP)/United Nations Population Fund (UNPF)
Second country cooperation framework for the United Republic of Tanzania 2002-2006 (2001)
This cooperation framework focuses on governance and institutional aspects of poverty reduction, as well
as government services. Little attention is being paid to natural resources dimensions. Climate change,
current climate-related risks, or even food security in general, are not discussed.
C. 2 United Nations Development Assistance Framework (UNDAF) 2002-2006 (2001)
The UNDAF does not mention climate change. However, it recognizes the linkages between poverty and
degradation of natural resources. In particular, it mentions the increasing risk of desertification (with 60%
of Tanzania being composed of dry lands), partly caused by extensive deforestation. A fairly
comprehensive section on Tanzania’s vulnerability to natural hazards also highlights climate-related
concerns: “Natural and man-made disasters erode the coping capacity of the vulnerable population
especially in drought-prone areas. There have been poor rains in Central Tanzania for the last three years,
and traditional coping strategies are breaking down as land pressure increases. These types of shocks
have become a frequent phenomenon in Tanzania in recent years. Floods and droughts, epidemics and
crop pests, environmental damage and economic instabilities, have all had their effects on people’s
capacity to meet their basic needs and subsequently their ability to survive and pursue their development
ambitions and potential” In addition, the UNDAF observes a worrisome trend: “Some claim that during
recent years emergency preparedness has actually decreased and dependency on external support in these
kinds of situations has increased. Long term disaster management strategies to deal with predictable,
poverty related emergencies are needed to use available resources most effectively.” This general concern
is not yet translated into concrete activities in the UNDAF.
C. 3 United Nations Emergency Consolidated Appeal for the Drought in Tanzania 2001
This appeal illustrates Tanzania’s high vulnerability to climate variability: “The 1999/2000 rains were very
poor in many parts of northern and central Tanzania. This has resulted in abnormally low levels of food
production, particularly of the staple crop, maize grain, and has also caused a very poor cash crop harvest,
thereby further reducing the cash income of the drought-affected households. This has been highly
damaging to the household food security of many farming families in the semi-arid areas, who have
suffered a fourth consecutive year of poor harvests and low-income levels. This cumulative effect has
greatly undermined their purchasing power, forcing many of the poorest families to sell productive assets
in order to survive. The recurrent nature of these food crises exposes the underlying layer of core poverty”.
The appeal is intended to address Tanzania’s consecutive and chronic droughts, which affect the lives of
over 9 million people, almost 30% of the total population of Tanzania. About 1.3 million of these live in a
situation of total food insecurity. Instead of applying emergency measures year after year, the appeal
proposes a more fundamental approach, part of longer-term integrated development strategies, including
the Rural Development Strategy and the Agricultural Development Strategy which are currently under
development, as well as improved early warning systems. Despite this longer-term focus, shifts in risks, for
instance due to climate change, are not discussed. The implementation of the appeal, and particularly the
longer-term components, could involve up to eight members of the UN system: FAO, WFP, UNICEF,
ILO, UNIDO, IFAD, UNDP and the World Bank.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
60
C. 4 African Development Bank
Country Strategy Paper 1999-2001 (2000)
Country Economic Profile
In its macroeconomic analysis, the strategy notes that overall macroeconomic performance has been
satisfactory, but “growth rates have been fluctuating from year to year reflecting the vulnerability of the
economy to external shocks. Although strong growth was registered in FY 1996/97 (4.2 percent), it
declined to 3.3 percent in FY 1997/98 due to the adverse impact of the drought on agricultural output. The
drought was followed by the El-Nino floods late 1997 and early 1998, which destroyed some of the crops
and damaged roads, thereby, disrupting internal movement of agricultural commodities as well as export
shipments.” In response, the strategy underlines the need for “an aggressive export promotion drive and
continued diversification of the export base.” The direct causes of Tanzania’s vulnerability to natural
hazards are not analyzed in the macroeconomic analysis or in the sectoral sections.
In a section on poverty, the strategy highlights the links between poverty, drought, and food insecurity:
“Since the poor are entirely dependent on agriculture (mainly crops) for their livelihood, their incomes and
food consumption are vulnerable to droughts. Food insecurity is therefore a major feature of poverty,
especially in drought-prone areas of the country”. The strategy also notes that less than 20 percent of the
irrigation potential is utilized, unnecessarily exposing agricultural production to droughts. At the same time
however, the agriculture section of the strategy attributes poor agricultural performance mainly to limited
access to agricultural credits and weak extension services, poor transport infrastructure, limited use of
modern inputs (mainly due to high costs), weak agriculture planning and program implementation, and low
budgetary allocation to the sector. Dealing with droughts and irrigation are not mentioned here. Donor
support has in the past included, among others, small-scale irrigation and soil conservation. However, the
results have been mixed, mainly due to lack of counterpart funding, weak institutional capacity in the
Ministry of Agriculture, and the lack of a coherent sector framework.
Similarly, the strategy point to the underlying patterns causing water supply problems: “While droughts
have contributed to water supply problems, the underlying factors include weak institutional capacity in the
sector, poor water resource management, and the dilapidated condition of the water schemes and
distribution networks in the rural and urban areas resulting from the under-funding of maintenance and
rehabilitation.” Hence, the AfDB also focuses mainly on sector reform, operation efficiency, and
rehabilitation and expansion of existing facilities. In addition, it has adopted a river basin approach for
water management improvements. Finally, the strategy also notes the widespread environmental problems,
including land degradation, desertification, loss of biodiversity and wildlife, and the depletion of marine
and coastal resources. The environmental degradation is attributed to widespread poverty, high population
growth, and poor natural resource management practices. In that context, climatic factors (including
wildfires and flood and drought risks related to climate variability) are not discussed. Climate change is not
mentioned anywhere in the strategy.
The AfDB Country Economic Profile (from 1995) highlights the interrelationships between natural hazards
and natural resources management: “There is widespread consensus that one of the major problems facing
the nation is land degradation. This takes many forms: soil erosion, deforestation, bush fires and
overgrazing. The root cause often lies in the actions of the agricultural producers themselves. Land
degradation results in a loss of productivity in agriculture, land use conflicts, loss of biodiversity and
changes in water catchment areas which have led to both drought and floods”. These issues are highlighted
throughout the study, and illustrated by several examples. Deforestation in particular is highlighted as a
pressing problem, for biodiversity, but also floods and droughts. Bush fires are both a cause and
consequence. The profile also describes the environmental problems in coastal areas, including pollution,
but also clearance of mangrove forests and destruction of corals (particularly by dynamite fishing). Among
the consequences is a depletion of fishery resources.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
61
While the document reviews policy options to respond to the challenges, it mainly emphasizes the
implementation and further development of government policies that were already in place or under
consideration. Climate change is not discussed33.
C. 5 World Bank
Country Assistance Strategy (2000)
The effect of climate on the country’s performance is recognized by the fact that climatic conditions are
mentioned as part of the inputs to a (low-case) macro-economic scenario for Tanzania’s development. In
addition, the strategy states that “Tanzania is vulnerable to external shocks, commodity price changes and
droughts”, putting climatic conditions on a par with major economic issues that are discussed at length. In
addition, it describes that an infrastructure project has been restructured to address “El Nino damages”.
However, vulnerability to floods and drought is not mentioned, not as a risk to the Bank’s own projects,
nor as a development opportunity that could have been addressed by concrete activities. Climate change is
not mentioned.
C. 6 IFAD
Country Strategic Opportunities Paper (1998)
According to IFAD’s strategy paper “agriculture remains exposed to the vagaries of nature”. For instance,
the high growth in maize production (the main staple crop) is highly susceptible to weather conditions.
While the country has a structural food deficit of about 700 tons, imports rise to up to 1.5 million tons in
times of flood or drought. The main constraints to agricultural production are lack of irrigation,
unavailability of credit for the poorest segment of the population, and absence of an appropriate
institutional framework to support agricultural development activities. Donor support has been ineffective
due to poor counterpart funding from the government, cumbersome and centralized procedures, lack of
beneficiary participation and ownership, and lack of appropriate targeting criteria for women. IFAD aims
to address these sector-wide issues in order to make the agricultural sector more productive; at the same
time, this should contribute to a decrease in vulnerability to adverse weather conditions, particularly for the
smallholders who account for about 85% of the cultivable land. Climate-related risks to IFAD projects are
not discussed explicitly, although the report mentions the flood damage to irrigation schemes during the
1997/98 El Nino. Climate change is not mentioned.
C. 7 DFID
Country Strategy Paper (1999)
This country paper recognizes that Tanzania’s agriculture, accounting for half the GDP and 75% of
exports, is “highly vulnerable to climatic shocks”. In the past, DFID has provided substantial support in
response to natural disasters. The strategy for the coming years contains assistance to help protect poor
people’s livelihoods and strengthen the government’s capacity to prepare for and manage disasters.
However, climate risks to development investments and their outcomes are not recognized as a concern,
and climate change is not even mentioned.
C. 8 EU
Tanzania Strategy Paper for the Period 2001-2007 (2002)
This country strategy paper defines the priorities for EU assistance to Tanzania in the period 2001-2007.
The main sectors to be targeted are transport infrastructure and basic education. Further assistance will go
to governance and macro-economic support in line with the PRSP objectives. Ongoing programs in
agriculture, water & sewerage, and environment, will be continued. Despite the vulnerability of some of
these sectors, even current climate risks are mentioned only once, in the margins of an agriculture section.
33 It is noted that Tanzania had not yet ratified the UNFCCC, but was in the process of doing so.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
62
C. 9 Ireland Aid
Country Strategy Paper for Bilateral Aid Programme 2000 – 2002 (1999)
Ireland’s country strategy focuses on poverty reduction. While increasing food and livelihoods security are
among the key goals, weather and climate-related risks are not mentioned at all. In the coastal zone, Ireland
Aid is financing the Tanga Coastal Zone Conservation and Development Programme, managed by IUCN,
which aims to address issues like the destruction of coral reefs and mangrove swamps. Again, no reference
is made to climate change.
C. 10 JICA
Country Study for Japan’s Official Development Assistance to the United Republic of Tanzania
(1997)
Country Profile on Environment (1999)
The JICA country study recognizes the severe economic implications of climate risks in Tanzania:
“Several factors are considered directly responsible for the weakened economy. They include certain
political and economic policy choices made following independence, a rapidly growing population, climate
anomalies, and a deteriorating conditions for trade in the international market.” Recognizing the severe
pressures on Tanzania’s natural resource base, JICA aims to provide assistance to alleviate those pressures,
particularly in the forestry and water resources sectors. However, no attention is paid to interactions of
climate-related risks, poverty, and land degradation and water scarcity, or to ways to reduce that
vulnerability.
The JICA Country Profile on Environment gives a complete overview of environmental problems facing
Tanzania. It includes issues like desertification, deforestation and forest fires, but does not mention new
risks due to climate change. Even climate variability is largely ignored, for instance when discussing water
resources: “Tanzania is a well-watered country with moderate to good rainfall and with many rivers and
lakes. However, rainfall is seasonal and water is not readily available in the dry season.” The most
pressing problems, however, occur not in the average dry season, but in a dryer than normal period, due to
climate variability.
C. 11 SIDA
Tanzania Country Strategy 2001-2005 (2000)
The country strategy mentions that desertification, deforestation, and problems with coastal and marine
environments and urban settlements threaten Tanzania’s sustainable development. However, neither
climate change, nor even current climate risks that interact with these issues, are discussed.
C. 12 USAID Tanzania
Summary Strategic Plan for Environment and Natural Resources (1999)
Annual Report (2002)
This strategic plan provides an update of previous work of USAID in the area of environmentally
sustainable natural resources management. Its new focus will be on improved conservation of coastal
resources and wildlife in targeted areas. Climate change and sea level rise are not among the listed causes
of coastal degradation. Similarly, several possibly vulnerable biodiversity projects neglect climate risks.
Similarly, climate risks do not appear in USAID’s Annual Report for Tanzania.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
63
APPENDIX D: REVIEW OF SELECTED DEVELOPMENT PROJECTS/PROGRAMS
D.1 Projects dealing explicitly with climate related risks
D.1.1 US Country Studies Program
The US Country Studies Program supported several studies in Tanzania, on both mitigation and
vulnerability & adaptation. Results from these studies, which were performed by the Centre for Energy,
Environment, Science and Technology (CEEST) in Dar es Salaam, are reviewed in the Tier-1 component
of this project.
D.1.2 Draft National Action Plan on Climate Change in Tanzania (CEEST, 1998)
This Plan, developed with underlying materials from the US Country Study, gives a comprehensive
overview of Tanzania’s vulnerability to climate change in various sectors, and discusses both mitigation
and adaptation options. For adaptation, the focus is on no-regrets measures integrated in sectoral
development. Some adaptation options are proposed in Agriculture, Livestock, Forestry, Water Resources,
and Coastal Zones. No attention is paid to possible overlaps of adaptation and mitigation options, such as
in the forestry sector.
While this 1998 Draft Plan is comprehensive and detailed, it is unclear what its impact has been. It appears
that it has not been formally adopted by the Government. Moreover, its recommendations are not well
reflected in subsequent sectoral and national development plans.
D.1.3 GTZ (Energy and Transport Division) : Measures to Implement the UN FCCC: Technological
and other Options for the Mitigation of Greenhouse Gases in Tanzania (1995)
This somewhat older report discusses GHG mitigation options in Tanzania, including in the forestry sector.
This sector might allow for projects that integrate adaptation and mitigation goals, but these are not
discussed.
D.2 Other Development Programs and Projects
D.2.1 World Bank Forest Conservation and Management Project
Project Appraisal Document (2001)
Social and Environmental Considerations (2002)
This project focuses on the development of the forestry sector, and on biodiversity conservation in
Tanzania’s forests. The latter component, which is supported by the GEF and implemented jointly with
UNDP, focuses on the Eastern Arc forests, which are recognized as “biodiversity hotspots”. Besides their
biodiversity value, these forests provide local livelihoods, and are crucial as water catchment areas for
Tanzania’s water supply and hydroelectric power generation.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
64
The project also aims to contribute to carbon sequestration, partly by limiting forest fires: “The main
techniques for increasing carbon uptake in miombo is the reduction in fire frequency. Experiments in many
parts of Africa have shown that woody biomass and soil carbon both increase if fires are excluded.
Permanent fire exclusion is virtually impossible in the strongly seasonal miombo climate, but a reduction
in frequency is probably achievable at reasonable cost. This would simultaneously increase carbon dioxide
uptake and decrease the emission of methane and ozone precursors.”
While the project thus explicitly addresses climate change, current climate-related risks to the project itself
are not discussed, and possible risks due to climate change, including more frequent forest and direct
threats to biodiversity, are entirely ignored. It is unclear whether such considerations would have changed
the project design, which in its current form already contributes to a reduction in the vulnerability of these
valuable forests.
D.2.2 GEF/World Bank Lake Victoria Environmental Management Project (supplemental credit)
Project Information Document (2001)
Integrated Safeguards Sheet (2001)
This project addresses the management of Lake Victoria, and affects the three countries around the lake
(Uganda, Kenya, and Tanzania), with Tanzania acting as the regional coordinator. It had many
components, varying from community-level management, to watershed improvement, to hydrometeorological
monitoring. Given the far-reaching environmental issues at stake, the project aims to put
the region on a long-term path of better management of the Lake and its surrounding natural resources.
Despite this long-term focus, climatic changes, which might have strong effects on water resources and
ecosystems, are not considered.
D.2.3 GEF/UNDP
Aerial Survey of the Threats to Mt Kilimanjaro Forests (2002) [www.tz.undp.org]
This aerial survey is part of UNDP’s Community Management of Protected Areas Conservation Project
(COMPACT), which promotes community-based biodiversity conservation in and around World Heritage
Sites (such as Kilimanjaro). The main threats identified for the Kilimanjaro region were: logging of
indigenous trees, forest fires, and establishment of settlements. No specific attention was paid to issues
related to changing climatic circumstances.
D.2.4 USAID
Tanzania Coastal Management Partnership: Options for a national integrated coastal management
policy (undated.)
This report (prepared by Tanzania’s National Environment Management Council and the University of
Rhode Island/Coastal Resources Center and supported by USAID) analyzes options for integrated coastal
zone management in Tanzania. Climate change and sea level rise are not discussed. While an ICZM
approach would certainly contribute to the sustainable development of Tanzania’s coastal areas,
opportunities may be missed.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
65
APPENDIX E: SOURCES FOR DOCUMENTATION
Statistics
CRS database, OECD/World Bank http://www.oecd.org/htm/M00005000/M00005347.htm
Government Documents
PRSP related documents www.worldbank.org/prsp
Poverty Reduction Strategy Paper (PRSP) (2000)
PRSP progress report (2001)
PRSP Joint Staff Assessment (by IDA and IMF) (2001)
PRSP Progress Report Joint Staff Assessment (by IDA and IMF) (2001)
Other national strategies www.tzonline.org
Tanzania Assistance Strategy (A Medium Term Framework for Promoting Local Ownership and
Development Partnerships) Consultation draft, Ministry of Finance (2001)
Tanzania Development Vision 2025
National Environmental Policy (1997)
UN Conventions
UN Convention on Climate Change (UNFCCC) www.unfccc.int
UN Convention to Combat Desertification (UNCCD) www.unccd.int
Proposed National Action Programme (1999)
Second National Report (2002)
UN Convention on Biodiversity (UNCBD) www.biodiv.org
National Report (2001) www.biodiv.org
World Summit on Sustainable Development www.johannesburgsummit.org
National Report to The Earth Summit on Sustainable Development (2002)
Country Profile (2002)
Donor Agencies
AfDB www.afdb.org
Country Environmental Profile, Environmental and Social Policy Working Paper Series, no. 26
(1995); Country Strategy Paper 1999-2001 (2000)
DFID www.dfid.gov.uk
Country Strategy Paper (1999)
GEF/UNDP
Aerial Survey of the Threats to Mt Kilimanjaro Forests (2002)www.tz.undp.org
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
66
EU
Tanzania Strategy Paper for the Period 2001-2007 (2002)
IFAD www.ifad.org
Country Strategic Opportunities Paper (1998)
JICA www.jica.go.jp
Country Study for Japan’s Official Development Assistance to the United Republic of Tanzania
(1997)
Country Profile on Environment (1999)
SIDA
Tanzania Country Strategy 2001-2005 (2000)
UN
United Nations Emergency Consolidated Appeal for the Drought in Tanzania 2001 Development
Assistance Framework (UNDAF) 2002-2006 (2001)
UNDP www.undp.org.np
United National Development Programme (UNDP)/Population Fund (UNPF) Second country
cooperation framework for the United Republic of Tanzania (2002-2006) (2001)
UNEP www.unep.org
USAID www.usaid.gov
Summary Strategic Plan for Environment and Natural Resources (1999)
Annual Report (2002)
Tanzania Coastal Management Partnership: Options for a national integrated coastal management
policy (n.d.)
World Bank www.worldbank.org
Country Assistance Strategy (2000)
World Bank Forest Conservation and Management Project. Project Appraisal Document (2001),
Social and Environmental Considerations (2002)
GEF/World Bank Lake Victoria Environmental Management Project (supplemental credit), Project
Information Document (2001), Integrated Safeguards Sheet (2001)
US Country Studies Program
GTZ (Energy and Transport Division)
Measures to Implement the UN FCCC: Technological and other Options for the Mitigation of
Greenhouse Gases in Tanzania (1995)
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
67
REFERENCES
Agrawala, S. and M. Berg, 2002, “Development and Climate Change Project: Concept Paper on Scope and
Criteria for Case Study selection. COM/ENV/EPOC/DCD/DAC(2002)1/Final, OECD, Paris.
Altmann, J.; Alberts, S. C.; Altmann, S. A. and Roy, S. B. (2002): Dramatic change in local climate
patterns in the Amboseli basin, Kenya. In: Afr. J. Ecol. 40: 248-251.
Beck, E., Scheibe, R. & Senser, M. (1986): The vegetation of the Shira plateau and the western slopes of
Kibo (Mount Kilimanjaro, Tanzania). In: Phytocoenologia 11: 1-30.
Brenan, J. P. M. (1978): Some aspects of the phytogeography of tropical Africa. In: Annals of the Missouri
Botanical Garden 65: 437-478.
Brewin, D. R. (1965): Kilimanjaro Agriculture. In: Tanganyika Notes and Records 64: 115-117.
Carlé, W. (1977): Der Kilimandjaro. In: Jh. Ges. Naturkde. Württ. 132: 50-109.
Cavelier, J. & Goldstein, G. (1989): Mist and fog interception in elfin cloud forests in Columbia and
Venezuela. In: Journal of Tropical Ecology 5: 309-322.
Cavelier, J., Jaramillo, M., Solis, D. and León, D. d. (1997): Water balance and nutrient inputs in bulk
precipitation in tropical montane cloud forest in Panama. In: Journal of Hydrology 193: 83-96.
Cavelier, J., Solis, D. and Jaramillo, M. A. (1996): Fog interception in montane forests across the Central
Cordillera of Panamá. In: Journal of Tropical Ecology 12: 357-369.
CIA (2002): World Fact Book. Central Intelligence Agency, available on-line
http://www.cia.gov/cia/publications/factbook/ Accessed January 16, 2003.
Clausnitzer, V. (2001): Notes on the species diversity of East African Odontata, with a checklist of species.
In: Odonatologica 30: 49-66.
Crutzen, P. J. and Andreae, M. O. (1990): Biomass burning in the tropics: impact on atmospheric chemical
and biogeochemical cycles. In: Science 250: 1669-1678.
Diaz, H. F. and Graham, N. E. (1996): Recent changes in tropical freezing heights and the role of sea
surface temperature. In: Nature 383: 152-155.
Downie, C. and P. Wilkinson (1972): The Geology of Kilimanjaro, Sheffield.
Eva, H. and Lambin, E. F. (2000): Fires and land-cover change in the tropics: a remote sensing analysis at
the landscape scale. In: Journal of Biogeography 27: 765-776.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
68
FAO (1986): Rehabilitation of traditional schemes (Phase I), Tanzania. AG: TCP/URT/4522, Mission
Report, Rome.
Forchhammer, P., Breuning, S. (1986): Annotated list of Cerambycidae, subfamily Lamiinae collected in
Tanzania, Zambia and Botswana 1969-80 with description of 86 new taxa. In: Natura Jutlandica 22:
1-72.
Gaffen, D.J., Santer, B. D., Boyle, J. S., Christy, J. R., Graham, N. E. & Ross, R. J. (2000): Multidecadal
changes in the vertical temperature structure of the tropical troposphere. In: Science 287: 1242-1245.
Gamassa, D. M. (1991): Historical change in human population on Mount Kilimanjaro and its
implications. In: Newmark, W. D. (ed.): The Conservation of Mount Kilimanjaro. The IUCN
Tropical Forest Programme: 1-8.
Gathaara, G. N., Woodley, B and Lambrechts, C (1999): Aerial survey of the destruction of Mt. Kenya,
Imenti and Ngare Ndare Forest Reserve. Kenya Wildlife Service.
Glasow, R. v. and A. Bott (1999): Interaction of radiation fog with tall vegetation. In: Atmospheric
Environment 33: 1333-1346.
Grabherr, G. and Pauli, H. (1994): Climate effects on mountain plants. In: Nature 369: 448.
Grimshaw, J. M. (1996): Aspects of the ecology and biogeography of the forest of the northern slope of
Mt. Kilimanjaro, Tanzania. Thesis, University of Oxford.
Grimshaw, J. M.; Cordeiro, N. J. and Foley, C. A. H. (1995): The mammals of Kilimanjaro. In: Journal of
East African Natural History 84: 105-139.
Hastenrath, S. and Kruss, P. D. (1992): The dramatic retreat of Mount Kenya’s glaciers between 1963 and
1987: greenhouse forcing. In: Annals of Glaciology 16: 127-133.
Hastenrath, S. & L. Greischar (1997): Glacier recession on Kilimanjaro, East Africa, 1912-89. In: Journal
of Glaciology 43: 455-459.
Hastenrath, S. (1984): The glaciers of Equatorial East Africa, Solid Earth Sciences Library, Dordrecht,
Boston, Lancaster.
Hastenrath, S. (2001): Variations of East African climate during the past two centuries. In: Climatic
Change 50: 209-217.
Hay, S. I., Cox, J., Rogers, D. J., Randolphs, S. E., Stern, D. I., Shanks, G. D., Myers, M. F. and Snow, R.
W. (2002): Climate change and the resurgence of malaria in the East African highlands. In: Nature
415: 905-909.
Hedberg, O. (1961): The phytogeographical position of the afroalpine flora. In: Recent Advances in
Botany 1: 914-919.
Hedberg, O. (1964): Features of afroalpine plant ecology. In: Acta Phytogeographica Suecica 49: 1-144.
Hemp, A. & Beck, E. (2001): Erica excelsa as a fire-tolerating component of Mt. Kilimanjaro’s forests. In:
Phytocoenologia 31: 449-475.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
69
Hemp, A. (1999): An ethnobotanical study on Mt. Kilimanjaro. In: Ecotropica 5: 147-165.
Hemp, A. (2001a): Ecology of the pteridophytes on the southern slopes of Mt. Kilimanjaro. Part II: Habitat
selection. In: Plant Biology 3: 493-523.
Hemp, A. (2001b): Life form and strategies of forest ferns on Mt. Kilimanjaro. In: Gottsberger, G. &
Liede, S. (eds.): Life forms and dynamics in tropical forests. Disserationes Botanicae 346: 95-130.
Hemp, A. (2002): Ecology of the pteridophytes on the southern slopes of Mt. Kilimanjaro. Part I:
Altitudinal distribution. In: Plant Ecology 159: 211-239.
Hemp, A. (in press): The impact of fire on diversity, structure and composition of Mt. Kilimanjaro’s
vegetation. Proceedings of the GMBA Symposium 19-24 August in Moshi, Tanzania.
Hemp, A., Hemp, C. and Winter, J. C. (1999): Der Kilimanjaro – Lebensräume zwischen tropischer Hitze
und Gletschereis. Natur und Mensch 1998, Jahresmitteilungen der Nürnberger Naturhistorischen
Gesellschaft: 5-28.
Hemp, C. & Hemp, A. (in press): Saltatoria cenoses of high altitude grasslands on Mt. Kilimanjaro,
Tanzania (Orthoptera: Saltatoria). In: Ecotropica.
Hemp, C. (2001): Ethnozoological Research on invertebrates on Mt. Kilimanjaro, Tanzania. In: Ecotropica
7: 139-149.
Hemp, C. (in press): Comparison of the endemic Saltatoria fauna (Orthoptera) of Mt. Kilimanjaro, Mt.
Meru and the Monduli Range. In: Journal of East African Natural History.
Hemp, C., Winter, J. C. (1999): Ethnozoologische Feldforschung am Kilimanjaro. Arthropoda. In:
Bayreuther Forum Ökologie 64: 167-199.
IPCC (2001): Climate Change 2001. Impacts, Adaptation, and Vulnerability, Intergovernmental Panel on
Climate Change, Geneva.
Irion, R. (2001): The melting snows of Kilimanjaro. In: Science 291: 1690-1691.
Jaeger, F. (1909): Forschungen in den Hochregionen des Kilimandscharo. In: Mitteilungen aus den
Deutschen Schutzgebieten 22: 113-146; 161-196.
Johnsen, P. and Forchhammer, P. (1975): Checklist of the Acridomorpha of Tanzania. In: Natura
Jutlandica 18: 26-52.
Jones, C.A. and J.R. Kiniry (eds.) (1986): CERES-Maize. A Simulation Model of Maize Growth and
Development, Texas A&M Press, College Station.
Juvik, J. O. and Nullet, D. (1993): Relationships between rainfall, cloud-water interception, and canopy
throughfall in a Hawaian montane forest. In: Hamilton, L. S., Juvik, J. O. & Scatena, F. N. (eds.):
Tropical montane cloud forests. Proceedings of an international symposium at San Juan, Puerto
Rico, 31 May-5 June 1993: 102-125.
Kaser, G. (1999): A review of the modern fluctuations of tropical glaciers. In: Global and Planetary
Change 22: 93-103.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
70
Kaser, G. (2001): Glacier-climate interaction at low latitudes. In: Journal of Glaciology 47: 195-204.
Kaser, G., Hardy, D. R., Mölg, T. & Bradley, R. S. (under review): Modern glacier retreat on Kilimanjaro
as evidence of climate change: Observations and facts.
Kaser, G., Hastenrath, S. and Ames, A. (1996): Mass balance profiles on tropical glaciers. In: Zeitschrift
für Gletscherkunde und Glazialgeologie 32: 75-81.
Kerfoot, O. (1968): Mist precipitation on vegetation. In: Forestry Abstracts 29: 8-20.
Klute, F. (1920): Ergebnisse der Forschungen am Kilimandscharo 1912, Berlin.
Körner, C. (1999): Alpine Plant Life. Functional Plant Ecology of High Mountain Ecosystems. – Springer,
Berlin-Heidelberg.
Kruss, P.D. (1983): Climate Change in East Africa. A numerical simulation from the 100 years of terminus
record at Lewis Glacier, Mount Kenya. In: Zeitschrift für Gletscherkunde und Glazialgeologie 19:
43-60.
Lambrechts, C., Woodley, B., Hemp, A., Hemp, C. and Nnyiti, P. (2002): Aerial survey of the threats to
Mt. Kilimanjaro forests, UNDP, Dar es Salaam.
Larcher, W. (1984): Ökologie der Pflanzen, Ulmer, Stuttgart.
Martens, P.; R.S. Kovats; S. Nijhof; P. de Vries; M.T.J. Livermore; D.J. Bradley; J. Cox; and
A.J. McMichael (1999): Climate change and future populations at risk of malaria. In: Global
Environmental Change 9: 89-107.
Mbonile, M. J. (1999b): Population mobility and migration in Mount Kilimanjaro. In: Mount Kilimanjaro:
Land Use and Environmental Management. French Institute for Research in Africa. IFRA Les
Cahiers 16: 24-33.
Mdoe, N. and Wiggins, S. (1997): Returns to smallholder dairying in the Kilimanjaro region, Tanzania. In:
Agricultural Economics 17: 75-87.
Meyer, H. (1890): Ostafrikanische Gletscherfahrten. Forschungsreisen im Kilimandscharo-Gebiet, Leipzig.
Miehe, S. and G. Miehe (1994): Ericaceous Forests and Heathlands in the Bale Mountains of South
Ethiopia. Ecology and Man’s Impact. Hamburg.
Misana, S. B. (1999): Community utilization and management of forest resources on Mount Kilimanjaro.
In: Mount Kilimanjaro: Land Use and Environmental Management. French Institute for Research in
Africa. IFRA Les Cahiers 16: 48-58.
Mitchell, J.F.B.; C.A. Senior; and W.J. Ingram (1989): CO2 and Climate. A missing feedback? In: Nature
330: 238-240.
Mwaipopo, O.U. (2001): Implications of accelerated sea-level rise (ASLR) and climate change for
Tanzania. In: Proceedings of Third SURVAS Expert Workshop on: “African Vulnerability &
Adaptation to Impacts of Accelerated Sea-Level Rise (ASLR),” de la Vega-Leinert, A.C., Nicholls,
R.J., Hassan, A. and El-Raey, M. (eds.) Second workshop, Cairo, Egypt — 5th-8th November 2000,
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
71
Flood Hazard Research Centre, Middlesex University, London, available on-line at
www.survas.mdx.ac.uk
Mwandosya, M., B. Nyenzi, and M. Luhanga. (1998): The Assessment of Vulnerability and Adaptation to
Climate Change Impacts in Tanzania. Centre for Energy, Environment, Science and Technology,
Dar es Salaam, Tanzania.
Newmark W. D. and Nguye, P. A. (1991): Recreational impacts of tourism along the Mrangu route in
Kilimanjaro National Park. In: Newmark, W.D. (ed.): The Conservation of Mount Kilimanjaro. The
IUCN Tropical Forest Programme: 47-52.
Nicholson, J. W. (1936): The influence of forests on climate and water supply in Kenya. In: E. Afr. agric.
J. 2: 48-53, 164-170, 226-240.
Nicholson, S. E. (2000): The nature of rainfall variability over Africa on time scales of decades to
millennia. In: Global and Planetary Change 26: 137-158.
Nicholson, S.E. and X. Yin (2001): Rainfall conditions in Equatorial East Africa during the nineteenth
century as inferred from the record of Lake Victoria. In: Climatic Change 48: 387-398.
O'Kting´ati, A. and Kessy, J. F. (1991): The farming system on Mount Kilimanjaro. In: Newmark, W. D.
(ed): The Conservation of Mount Kilimanjaro. The IUCN Tropical Forest Programme: 71-80.
Pócs, T. (1976): The role of the epiphytic vegetation in the water balance and humus production of the rain
forests of the Uluguru Mountains, East Africa. In: Boissiera 24: 499-503.
Pócs, T. (1991): The significance of lower plants in the conservation of Mount Kilimanjaro. In: Newmark,
W. D. (ed.): The Conservation of Mount Kilimanjaro. The IUCN Tropical Forest Programme: 21-34.
Ramsay, J. C. (1965): Kilimanjaro - sources of water supplies. In: Tanganyika Notes and Records 64: 92-
94.
Sarmett, J. D. and Faraji, S. A. (1991): The hydrology of Mount Kilimanjaro: an eximination of dry season
runoff and possible factors leading to its decrease. In: Newmark, W. D. (ed.): The Conservation of
Mount Kilimanjaro. The IUCN Tropical Forest Programme: 53-70.
Schmidt, P. R. (1989): 8. Early exploitation and settlement in the Usambara Mountains. In: Hamilton, A.
C., Bensted-Smith, R. (1989): Forest conservation in the East Usambara Mountains Tanzania. The
IUCN Tropical Forest Programme: 75-78.
Sjöstedt, Y. (ed) (1909): Wissenschaftliche Ergebnisse der Schwedischen Zoologischen Expedition nach
dem Kilimanjaro, dem Meru und den umgebenden Massaisteppen Deutsch-Ostafrikas 1905-1906.
Spawls, S., Howell, K., Drewes, R. and Ashe, J. (2002): A field guide to the reptiles of East Africa.
Academic Press.
Street-Perrott, F. A. and Perrott, R. A. (1990): Abrupt climate fluctuations in the tropics: the influence of
Atlantic Ocean circulation. In: Nature 343: 607-612.
Thompson, L. G. (2000a): Ice core evidence for climate change in the tropics: implications for our future.
In: Quaternary Science Reviews 19: 19-35.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
72
Thompson. L. G. (2000b): Ice core paleoclimate study of east African Monsoon and ENSO variability
from the ice fields of Kilimanjaro. Unpubl. report.
Thompson, L. G., Mosley-Thompson, E., Davis, M. E., Henderson, K. A., Brecher, H. H., Zagorodnov, V.
S., Mashiotta, T. A., Lin, P.-N., Mikhalenko, V. N., Hardy, D. R. & Beer, J. (2002): Kilimanjaro ice
core records. Evidence of holocene climate change in tropical Africa. In: Science 298: 589-593.
Thompson, L. G., Mosley-Thompson, E., Davis, M. E., Lin, P.-N., Henderson, K. A., Cole-Dai, J., Bolzan,
J. F. and Liu, K.-B. (1995): Late glacial stage and Holocene tropical ice core records from Huacarán,
Peru. In: Science 269: 46-50.
Timberlake, L. (1986): Krisenkontinent Afrika. Hammer, Wuppertal.
Tol, R.S.J. and H. Dowlatabadi (2002): Vector-borne diseases, development, and climate change. In:
Integrated Environmental Assessment 2:173-181.
United Republic of Tanzania, Ministry of Water & CES, Consulting Engineers Salgitter GmbH (2002):
Hydrometic report of the Hai district water supply Phase IV (2002).
United Republic of Tanzania (2003): Initial Communication under the United Nations Framework
Convention on Climate Change (UNFCCC), Dar es Salaam, available on-line:
http://unfccc.int/resource/docs/natc/tannc1.pdf.
Verschuren, D., Laird, K. R. & Cumming, B. F. (2000): Rainfall and drought in equatorial east Africa
during the past 1,100 years. In: Nature 403: 410-414.
Volkens, G. (1897): Der Kilimandscharo. Darstellung der allgemeineren Ergebnisse eines
fünfzehnmonatigen Aufenthalts im Dschaggalande, Berlin.
Walther, H.; E. Harnickell and D. Mueller-Dombois (1975): Climate-diagram maps of the individual
continents and the ecological climatic regions of the earth.
World Bank. 2002. World Bank Development Indicators. On CD Rom. World Bank, Washington, DC.
Zimmermann L., Frühauf, C. and Bernhofer, Ch. (1999): The role of interception in the water budges of
spruce stands in the eastern Ore Mountains/Germany. In: Phy. Chem. Earth 24: 809-812.g Party on Development Co-operation and Environment
DEVELOPMENT AND CLIMATE CHANGE
IN TANZANIA:
FOCUS ON MOUNT KILIMANJARO
by
Shardul Agrawala, Annett Moehner, Andreas Hemp, Maarten
van Aalst, Sam Hitz, Joel Smith, Hubert Meena,
Stephen M. Mwakifwamba, Tharsis Hyera
and Obeth U. Mwaipopo
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
2
Copyright OECD, 2003
Application for permission to reproduce or translate all or part of this material should be addressed to the
Head of Publications Service, OECD, 2 rue André Pascal, 75775 Paris, Cedex 16, France.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
3
FOREWORD
This document is an output from the OECD Development and Climate Change project, an activity being
jointly overseen by the Working Party on Global and Structural Policies (WPGSP) of the Environment
Directorate, and the Network on Environment and Development Co-operation of the Development Cooperation
Directorate (DAC-Environet). The overall objective of the project is to provide guidance on how
to mainstream responses to climate change within economic development planning and assistance policies,
with natural resource management as an overarching theme. Insights from the work are therefore expected
to have implications for the development assistance community in OECD countries, and national and
regional planners in developing countries.
This document has been authored by Shardul Agrawala and Annett Moehner. It draws upon four primary
consultant inputs that were commissioned for this country study: “Climate Impacts and Responses in
Mount Kilimanjaro” by Andreas Hemp (University of Bayreuth, Germany); “Review of Development
Plans, Strategies, Assistance Portfolios, and Select Projects Potentially Relevant to Climate Change in
Tanzania” by Maarten van Aalst of Utrecht University, The Netherlands; “Analysis of GCM Scenarios and
Ranking of Principal Climate Impacts and Vulnerabilities in Tanzania” by Stratus Consulting, Boulder,
USA (Sam Hitz and Joel Smith); and “Development and Climate Change in Tanzania” by the Center for
Energy, Environment, Science and Technology (CEEST), Dar es Salaam, Tanzania (Hubert Meena,
Stephen M. Mwakifwamba, Tharsis Hyera, and Obeth U. Mwaipopo).
In addition to delegates from WPGSP and DAC-Environet, comments from Tom Jones, Jan Corfee-
Morlot, Georg Caspary, and Remy Paris of the OECD Secretariat are gratefully acknowledged. Tomoko
Ota and Martin Berg provided project support at various times during the project. Shardul Agrawala would
like to acknowledge inputs on Kilimanjaro ice-field and regional climate patterns from Lonnie Thompson
(Ohio State University, USA), Jeanne Altman (Princeton University, USA), Douglas Hardy (University of
Massachusetts, USA) and Georg Kaser (University of Innsbruck, Austria). Andreas Hemp acknowledges
support for prior fieldwork in the Kilimanjaro from 1996-2002 from the Deutsche
Forschungsgemeinschaft, the UNEP project “Global Trends in Africa: the Case of Mt. Kilimanjaro” that
forms the basis for several major findings of the Kilimanjaro case study, and support from the Tanzanian
Commission for Science and Technology, the Chief Park Wardens of Kilimanjaro National Park, to the
Catchment Forest officers and to Mr. Mushi, Moshi. The Secretariat and Maarten van Aalst would like to
acknowledge several members of the OECD DAC who provided valuable materials on country strategies
as well as specific projects. Stratus Consulting would like to acknowledge inputs from Tom Wigley at the
National Center for Atmospheric Research (NCAR).
This document does not necessarily represent the views of either the OECD or its Member countries. It is
published under the responsibility of the Secretary General.
Further inquiries about either this document or ongoing work on sustainable development and climate
change should be directed to Shardul Agrawala of the OECD Environment Directorate:
shardul.agrawala@oecd.org, or Georg Caspary of the OECD Development Co-operation Directorate:
georg.caspary@oecd.org.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
4
TABLE OF CONTENTS
FOREWORD................................................................................................................................................. 3
EXECUTIVE SUMMARY ............................................................................................................................ 6
LIST OF ACRONYMS ................................................................................................................................. 8
1. Introduction ..................................................................................................................................... 9
2. Country background........................................................................................................................ 9
3. Climate: baseline climatology and climate change scenarios......................................................... 11
3.1 Current climate ......................................................................................................................... 11
3.2 Climate change and sea level rise projections .......................................................................... 12
4. Overview of impacts, vulnerabilities and adaptation responses..................................................... 14
4.1 Agriculture ............................................................................................................................... 15
4.2 Forests...................................................................................................................................... 15
4.3 Water resources........................................................................................................................ 16
4.4 Coastal resources ...................................................................................................................... 16
4.5 Human health........................................................................................................................... 17
4.6 Energy, industry and transport.................................................................................................. 17
4.7 Overview of adaptation responses ............................................................................................ 17
5. Attention to climate concerns in donor activities ........................................................................... 19
5.1 Donor activities affected by climate risks................................................................................. 20
5.2 Attention to climate risks in donor strategies............................................................................ 24
5.3 Climate risks in selected development programs and projects ................................................. 25
6. Attention to climate concerns in national planning........................................................................ 26
6.1 National Action Plan on Climate Change................................................................................. 27
6.2 National communications to international environmental agreements ..................................... 27
6.3 Poverty Reduction Strategy Paper (PRSP) ............................................................................... 28
6.4 Other national policies of relevance to climate change ............................................................ 28
7. Climate change and Mount Kilimanjaro ........................................................................................ 29
7.1 Climate, glaciers, and hydrology .............................................................................................. 30
7.2 Ecosystems, biodiversity and land tenure on Mount Kilimanjaro ............................................ 32
7.3 Climatic trends on Mount Kilimanjaro ..................................................................................... 35
7.4 Potential impacts of climatic changes: glacier retreat............................................................... 38
7.5 Potential impacts of climatic changes: enhancement of fire risk.............................................. 39
7.6 Other threats to the Mount Kilimanjaro ecosystem.................................................................. 45
7.7 Scenarios for 2020 with respect to fire impact ......................................................................... 46
7.8 Climate risks in perspective: shrinking glaciers versus enhanced fire risk............................... 47
8. Policy responses for Mount Kilimanjaro........................................................................................ 47
8.1 Policy responses to the shrinking ice cap.................................................................................. 47
8.2 Policy responses to general environmental threats ................................................................... 48
8.3 Policy responses to enhanced fire risk ...................................................................................... 48
8.4 Promotion of ecosystem friendly livelihood opportunities....................................................... 52
9. Concluding remarks ....................................................................................................................... 53
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
5
9.1 Differentiated adaptation strategy.................................................................................................... 54
9.2 Climate change and donor portfolios............................................................................................... 54
9.3 Attention to climate change concerns in national planning............................................................. 55
9.4 Climate risks in perspective on Mount Kilimanjaro........................................................................ 55
9.5 Policy responses for Mount Kilimanjaro......................................................................................... 56
APPENDIX A: PREDICTIVE ERRORS FOR SCENGEN ANALYSIS FOR TANZANIA...................... 57
APPENDIX B: LIST OF PURPOSE CODES INCLUDED IN THE SELECTION OF CLIMATEAFFECTED
PROJECTS, ORGANIZED BY THE DAC SECTOR CODE................................................ 58
APPENDIX C: REVIEW OF SELECTED DONOR STRATEGIES FOR TANZANIA............................ 59
APPENDIX D: REVIEW OF SELECTED DEVELOPMENT PROJECTS/PROGRAMS......................... 63
APPENDIX E: SOURCES FOR DOCUMENTATION .............................................................................. 65
REFERENCES ............................................................................................................................................ 67
Boxes
Box 1. A brief description of MAGICC/SCENGEN ............................................................................... 12
Box 2. Creditor Reporting System (CRS) Database ................................................................................ 21
Box 3. Flora of Mount Kilimanjaro.......................................................................................................... 32
Box 4. Fauna of Mount Kilimanjaro ........................................................................................................ 34
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
6
EXECUTIVE SUMMARY
This report presents the integrated case study for Tanzania carried out under an OECD project on
Development and Climate Change. The report is structured around a three-tiered framework. First, recent
climate trends and climate change scenarios for Tanzania are assessed, and key sectoral impacts are
identified and ranked along multiple indicators to establish priorities for adaptation. Second, donor
portfolios in Tanzania are analyzed to examine the proportion of donor activities affected by climate risks.
A desk analysis of donor strategies and project documents as well as national plans is conducted to assess
the degree of attention to climate change concerns in development planning and assistance. Third, an indepth
analysis is conducted for climate change impacts and response strategies for Mount Kilimanjaro – a
critical ecosystem, biodiversity hotspot, and source of freshwater. This part of the analysis draws upon
extended field research by a case study consultant in collaboration with national and international partners.
Analysis of recent climate trends reveals that climate change poses significant risks for Tanzania.
While projected changes in precipitation are uncertain, there is a high likelihood of temperature increases
as well as sea level rise. Climate change scenarios across multiple general circulation models show
increases in country averaged mean temperatures of 1.3°C and 2.2°C projected by 2050 and 2100, which
are broadly consistent, though lower than, projections used in Tanzania’s Initial National Communication.
The sectors potentially impacted by climate change include agriculture, forests, water resources, coastal
resources, human health, as well as energy, industry and transport.
While uncertainties in climate change and impact projections pose a challenge for anticipatory
adaptation in any country, Tanzania’s case has several specific characteristics that might suggest the need
for a differentiated adaptation strategy. First, the climate change projections which form the basis of
national assessments rely on an older generation of climate models which project higher temperature
increases than more recent models analyzed in the present study. Updating of climate scenarios and impact
projections through the use of multiple and more recent models might therefore be advisable prior to the
formulation of aggressive (and potentially expensive) adaptation responses. A second characteristic feature
of Tanzania is that certain sectors such as agriculture and water resources are projected to experience both
negative and positive impacts under climate change – for example, while production of maize is projected
to decline, the production of two cash crops (coffee and cotton) is projected to increase. The implication
for adaptation therefore may be to not only cushion adverse impacts, but also to harness positive
opportunities. A third key characteristic is that unlike most other countries where the need for adaptation
relies on projections of future impacts, some discernible trends in climate and attendant impacts are already
underway in Tanzania. Such impacts – as is the case of the Kilimanjaro ecosystem - argue for more
immediate adaptation responses as opposed to a “wait and see” strategy.
Tanzania receives close to a billion US dollars of Official Development Assistance (ODA)
annually. Analysis of donor portfolios in Tanzania using the OECD-World Bank Creditor Reporting
System (CRS) database reveals that between 12-25% of development assistance (by aid amount) or 20-
30% of donor projects (by number) are in sectors potentially affected by climate risks. However, these
numbers are only indicative at best, given that any classification based on sectors suffers from oversimplification
– the reader is referred to the main report for a more nuanced interpretation. Donor and
government documents generally do not mention climate change explicitly, although frequent references
are made to the impacts of climate variability and their linkages to economic performance. There is
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
7
however considerable synergy between priorities of at least some national plans and measures that might
be required for climate change adaptation, such as water conservation, improving agricultural resilience,
and forest conservation. However, some of these goals (such as water conservation) had been articulated,
though not successfully implemented in previous plans. Therefore, a key obstacle facing “mainstreaming”
is not synergies at the level of planning documents, but rather the successful implementation of such plans.
The in-depth sector analysis focuses on the climate change impacts and policy responses on the
Mount Kilimanjaro ecosystem. Glaciers on Mount Kilimanjaro have been in a general state of retreat on
account of natural causes for over a hundred and fifty years. Due to a decline in precipitation coupled with
a local warming trend that has been recorded in the second half of the twentieth century Kilimanjaro’s ice
cap is now projected to vanish entirely by as early as 2020. The symbolism of this loss is indeed
significant, and furthermore the loss of the ice cap would also imply that valuable records of past climates
contained in its ice cores would also be irreplaceably destroyed. From a physical and socio economic
perspective however, this analysis concludes that the impact of the loss of the ice cap is likely to be very
limited. Much more significant is the enhancement in the intensity and risk of forest fires on Mount
Kilimanjaro as a consequence of the increase in temperatures and a concomitant decline in precipitation
over the past several decades. Forest fires have resulted in the replacement of the fog intercepting
subalpine forest belt by low lying shrub which has already seriously impacted the hydrological balance of
the mountain as fog intercepting cloud forests play a key role in the water budgets of high altitude drainage
basins. A continuation of current trends in climatic changes, fire frequency, and human influence could
result in the loss of most of the remaining subalpine Erica forests in a matter of years. With this, Mount
Kilimanjaro will have lost its most effective water catchment. Among the more immediate adaptation
responses identified by this report are institutional measures such as the inclusion of the forest belt into the
Kilimanjaro National Park and the creation of a paramilitary ranger group to deter logging, as well as better
investments in early warning systems, particularly the purchase of one or two aircraft for aerial
surveillance. There is also a need to limit cross-border migration of big game from neighbouring Amboseli,
which is adding to the stress on the Kilimanjaro ecosystem. In addition to short term solutions there is a
critical need to develop a comprehensive and holistic development plan focusing on fire-risk and forest
destruction, livelihood needs of the local population as well as on conservation strategies to ensure the long
term sustainability of the valuable resources of the Kilimanjaro ecosystem.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
8
LIST OF ACRONYMS
AfDB
AMA
asl
AVVA
CBO
CCCM
CEEST
CERES-Maize
COMPACT
CRS
DAC
DFID
EACC
FAO
FFYP
GCA
GCM
GDP
GEF
GHG
GMBA
GNP
GNI
IDA
IFAD
IPCC
KINAPA
MW
NEAP
NEP
NCAA
NGO
ODA
PRSP
SIDA
TAF
TANAPA
TFYP
UN
UNCB
UNCCD
UNCED
UNDAF
UNDP
UNEP
UNESCO
UNF
UNFCCC
USAID
USCSP
WHO
WNHS
African Development Bank
African Mountain Association
Above Sea Level
Aerial Videotape-assisted Vulnerability Analysis
Community Based Organization
Canadian Climate Centre Model
Centre for Energy, Environment, Science and Technology
Crop Environment Resource Synthesis model
Community Management of Protected Areas Conservation Project
Creditor Reporting System of the OECD/World Bank
Development Assistance Committee
Department for International Development
East African Coastal Current
Food and Agriculture Organization of the United Nations
First Five Year Plan
Game Controlled Areas
General Circulation Model
Gross Domestic Product
Global Environment Facility
Greenhouse Gases
Global Mountain Biodiversity Assessment
Gross National Product
Gross National Income
International Development Assistance
International Fund for Agricultural Development
Intergovernmental Panel on Climate Change
Kilimanjaro National Park
Mega Watt
National Environmental Action Plan
National Environmental Policy
Ngorongoro Conservation Authority Area
Non Governmental Organization
Official Development Assistance
Poverty Reduction Strategy Papers
Swedish International Development Agency
Tanzanian Association of Foresters
Tanzanian National Parks
Third Five Year Plan
United Nations
United Nations Convention on Biodiversity
United Nations Convention to Combat Desertification
United Nations Conference on Environment and Development
United Nations Development Assistance Framework
United Nations Development Programme
United Nations Environment Programme
United Nations Educational, Scientific and Cultural Organization
United Nations Foundation
United Nations Framework Convention on Climate Change
The United States Agency for International Development
United States Country Studies Program
World Health Organization
World Natural Heritage Sites
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
9
1. Introduction
This report presents the integrated case study for Tanzania for the OECD Development and
Climate Change Project, an activity jointly overseen by the Working Party on Global and Structural
Policies (WPGSP), and the Working Party on Development Co-operation and Environment (WPENV).
The overall objective of the project is to provide guidance on how to mainstream responses to climate
change within economic development planning and assistance policies, with natural resource management
as an overarching theme. The Tanzania case study was conducted in parallel with five other country case
studies1 in Africa, Latin America, and Asia and the Pacific.
Each case study is based upon a three-tiered framework for analysis (Agrawala and Berg 2002).
1. Review of climate trends and scenarios at the country level based upon an examination of results
from seventeen recent general circulation models, as well as empirical observations and results
published as part of national communications, country studies, and scientific literature. These
projections are then used in conjunction with knowledge of socio-economic and sectoral
variables to rank key sectoral and regional impacts on the basis of a number of parameters. The
goal of this tier is to present a framework to establish priorities for adaptation.
2. Review of economic, environmental, and social plans and projects of both the government and
international donors that bear upon the sectors and regions identified as being particularly
vulnerable to climate change. The purpose of this analysis is to assess the degree of exposure of
current development activities and projects to climate risks, as well as the degree of current
attention by the government and donors to incorporating such risks in their planning. This section
will review donor portfolios and projects, as well as development priorities of the Government of
Tanzania to determine the degree of attention to potential risks posed by climate change on
relevant sectors.
3. In-depth analyses at a thematic, sectoral, regional or project level on how to incorporate climate
responses within economic development plans and projects, again with a particular focus on
natural resource management. In the case of Tanzania this case study provides an overview of
critical impacts and mainstreaming challenges for a number of sectors. This is followed by an indepth
analysis on Mount Kilimanjaro – a UNESCO World Heritage Site and also a critical
ecosystem and source of freshwater resources for Tanzania. The analysis on climate change
impacts and response strategies for the Mount Kilimanjaro ecosystem draws upon field research
over an extended period by a case study consultant in collaboration with national and
international partners.
2. Country background
Tanzania is located in East Africa, on the Indian Ocean bordered by Kenya to the north and
Mozambique to the south (Figure 1). It has an area of 945,000 km2 which includes the three major coastal
islands of Mafia, Pemba, and Zanzibar, and a coastline that is about 800 km long. The geography is
characterized by plains along the coast, a central plateau, and highlands in the north and south. The
northwest of the country encompasses approximately one-half of Lake Victoria, the second largest body of
freshwater in the world, and the western and southwestern borders abut the comparably massive Lake
Tanganyika and Lake Nyasa. Elevations range from sea level to the highest point in Africa, the glaciated
peak of Kilimanjaro at 5,895 m, the expansive slopes of which constitute one of the unique ecosystems of
1 Bangladesh, Egypt, Uruguay, Fiji, and Nepal.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
10
Africa. Tanzania also includes the Serengeti, the site of one of the last major terrestrial mammalian
migrations in the world and a prominent tourist destination.
Figure 1. Map of Tanzania
Tanzania is one of the poorest countries in the world with a GNI per capita of only US $ 280
(World Bank 2002). Gross national income per capita over the period 1994-2000 stood at about US$270
compared to US$470 for sub-Saharan Africa in general. Some 42% of the total population and 50% of the
rural population live below the poverty line, according to a 1993 survey, with 20% of the entire population
surviving on less than US$1 per day (World Bank, 2002). Based on the same 1993 survey, the Gini
Coefficient2 for Tanzania is 0.38, with the poorest 10% accounting for 2.8% of the national income and the
richest 10% accounting for 30.1%. According to World Bank estimates, Tanzania’s population in 2000 was
33.7 million, and growing at 1.8% a year. Average Life-expectancy is only 43.1 years (World Bank 2002).
While an overwhelming proportion of the population still lives in rural areas, by the late 1990s, 27.8% of
the country’s population lived in an urban setting, up from only 10.1% in 1975. Tanzania’s economy is
heavily dependent on agriculture, which accounts for nearly one-half of GDP, employs 80% of the work
force, and provides 85% of exports (World Bank, 2002). Topography and climatic conditions, however,
limit cultivated crops to only 4% of the land area. Industry has traditionally been limited to the processing
of agricultural products and light consumer goods. However, with a significant infusion of funds from the
World Bank, International Monetary Fund, and bilateral donors, growth over the last decade has featured
2 The Gini coefficient is a number between zero and one that measures the degree of inequality in the
distribution of income in a given society. The coefficient would register zero inequality for a society in
which each member received exactly the same income and it would register a coefficient of one (maximum
inequality) if one member got all the income and the rest got nothing.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
11
an increase in industrial production and a substantial increase in output of minerals (CIA, 2002). Private
sector growth and investment have also increased and, coupled with donor aid and liberal macroeconomic
policies, should support continued growth of about 5% (World Bank, 2002).
Economic growth could play an important role in increasing the capacity of a country like
Tanzania to adapt to climate change. However, the current state of its infrastructure and educational system
is likely an impediment to Tanzania’s ability to cope effectively with climatic risks. In 2000, only 4.2% of
Tanzania’s road network was paved, compared to 16.5% for low income countries in general. Further,
while 37% of tertiary level students were enrolled in science and engineering programs between 1987 and
1997, gross tertiary enrolment stood at only 0.66% by 1997 (World Bank, 2002). Similarly, gross
secondary enrolment was 6.5%. Adult literacy was 24.9% in 2000. Figure 2 provides an indication of how
Tanzania compares to other low income countries in terms of four key indices of development. On all four
measures of development, Tanzania ranks considerably below the average for low income countries.
Figure 2. Development diamond for Tanzania
Tanzania
Low-income group
Development diamond
Life expectancy
Access to improved water source
GNI
per
capita
Gross
primary
enrollment
Source: World Bank 2002
3. Climate: baseline climatology and climate change scenarios
This section briefly reviews projections of temperature and precipitation change for Tanzania
from climate models, and then provides a synthesis of key climate change impacts and vulnerabilities.
3.1 Current climate
Tanzania’s climate ranges from tropical to temperate in the highlands. Average annual
precipitation over the entire nation is 1,042 mm. Average temperatures range between 17°C and 27°C,
depending on location. Natural hazards include both flooding and drought. Within the country, altitude
plays a large role in determining rainfall pattern, with higher elevations receiving more precipitation.
Generally speaking, the total amount of rainfall is not very great. Only about half the country receives
more than 762 mm annually (Mwandosya et al., 1998). Tanzania’s precipitation is governed by two rainfall
regimes. Bimodal rainfall, comprised of the long rains of Masika between March-May and short rains of
Vuli between October-December, is the pattern for much of the northeastern, northwestern (Lake Victoria
basin) and the northern parts of the coastal belt. A unimodal rainfall pattern, with most of the rainfall
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
12
during December-April, is more typical of most of the southern, central, western, and southeastern parts of
the country.
3.2 Climate change and sea level rise projections
Changes in area averaged temperature and precipitation over Tanzania were assessed using
outputs from over a dozen recent (post 1995) GCMs which are processed using a new version of
MAGICC/SCENGEN. MAGICC/SCENGEN is briefly described in Box 1. First, results for Tanzania for
17 GCMs developed since 1995 were examined. Next, 11 of 17 models which best simulate current
climate over Tanzania were selected. The models were run with the IPCC B2 SRES scenario (Nakicenovic
and Swart 2000)3.
Box 1. A brief description of MAGICC/SCENGEN
MAGICC/SCENGEN is a coupled gas-cycle/climate model (MAGICC) that drives a spatial climate-change
scenario generator (SCENGEN). MAGICC is a Simple Climate Model that computes the mean global surface air
temperature and sea-level rise for particular emissions scenarios for greenhouse gases and sulphur dioxide (Raoer et
al., 1996). MAGICC has been the primary model used by IPCC to produce projections of future global-mean
temperature and sea level rise (see Houghton et al., 2001). SCENGEN is a database that contains the results of a
large number of GCM experiments. SCENGEN constructs a range of geographically-explicit climate change scenarios
for the world by exploiting the results from MAGICC and a set of GCM experiments, and combining these with
observed global and regional climate data sets. SCENGEN uses the scaling method of Santer et al. (1990) to produce
spatial pattern of change from an extensive data base of atmosphere ocean GCM – AOGCM (atmosphere ocean
general circulation models) data. Spatial patterns are “normalized” and expressed as changes per 1°C change in
global-mean temperature. The greenhouse-gas and aerosol components are appropriately weighted, added, and
scaled up to the actual global-mean temperature. The user can select from a number of different AOGCMs for the
greenhouse-gas component. For the aerosol component there is currently only a single set of model results. This
approach assumes that regional patterns of climate change will be consistent at varying levels of atmospheric
greenhouse gas concentrations. The MAGICC component employs IPCC Third Assessment Report (TAR) science
(Houghton et al., 2001). The SCENGEN component allows users to investigate only changes in the mean climate state
in response to external forcing. It relies mainly on climate models run in the latter half of the 1990s.
Source: National Communications Support Program Workbook
The spread in temperature and precipitation projections of these 11 GCMs for various years in
the future provides an estimate of the degree of agreement across various models for particular projections.
More consistent projections across various models will tend to have lower scores for the standard
deviation, relative to the value of the mean. The results of the MAGICC/SCENGEN analysis for Tanzania
are shown in Table 1.
3 The IPCC SRES B2 scenario assumes a world of moderate population growth and intermediate level of
economic development and technological change. SCENGEN estimates a global mean temperature
increase of 0.8 °C by 2030, 1.2 °C by 2050, and 2 °C by 2100 for the B2 scenario.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
13
Table 1. GCM estimates of temperature and precipitation changes4
Year
Temperature change (°C)
mean (standard deviation)
Precipitation change (%)
mean (standard deviation)
Tanzania Annual JJA5 SON6 DJF7 MAM8 Annual JJA SON DJF MAM
2030 0.9 (0.20) 1.0
(0.21)
.8
(0.17)
.8
(0.30)
0.9
(0.30)
4.1
(5.05)
-2.4
(7.98)
3.9
(10.04)
6.6
(8.06)
2.2
(5.34)
2050 1.3 (0.28) 1.5
(0.31)
1.2
(0.25)
1.1
(0.43)
1.3
(0.44)
5.9
(7.30)
-3.5
(11.53)
5.6
(14.51)
9.6
(11.64)
3.1
(7.72)
2100 2.2 (0.49) 2.6
(0.54)
2.1
(0.43)
1.9
(0.75)
2.3
(0.77)
10.2
(12.70)
-6.0
(20.07)
9.7
(25.27)
16.7
(20.27)
5.4
(13.44)
The results indicate that mean annual temperatures are projected to rise by 2.2 C by 2100, with
somewhat higher increases (2.6 °C) over June, July and August, and lower values (1.9 °C) for December,
January, February. Low standard deviations relative to the mean indicate good agreement across the 11
models. The Initial National Communication of Tanzania (2003) projects a temperature increase between
3-5 °C under doubling of carbon dioxide, which is benchmarked to the year 2075. The lower estimates of
MAGICC/SCENGEN are likely from the use of more recent scenarios (SRES) and multiple (17), more
recent (post 1995) GCMs with a better treatment of aerosols in the MAGICC/SCENGEN analysis. The
Tanzania National Communication meanwhile relied on four earlier generation models (primarily the
UK1989), as well as older (unspecified) emissions scenarios. Both sets of analyses however show
temperature increases, and furthermore the patterns of seasonal temperature increase are consistent.
Specifically, greater warming is projected for the cooler months (June-August) compared to the warmer
months (December-February).
In terms of precipitation meanwhile, according to the MAGICC/SCENGEN analysis annual
precipitation over the whole country is projected to increase by 10% by 2100, although seasonal declines
of 6% are projected for June, July and August, and increasers of 16.7% for December, January, February.
However, high standard deviations are indicative of low confidence in these projections across the various
models. Furthermore, the precipitation regimes across Tanzania vary considerably, as discussed in the
preceding section. Therefore country averaged values for precipitation, as is done in the
MAGICC/SCENGEN analysis, are of limited utility9. The Tanzania Initial National Communication does
offer greater regional specificity – although the results should be interpreted with caution as they do not
include an uncertainty analysis and rely on one or two older climate models. Under a doubling carbon
dioxide scenario some parts of Tanzania are projected to experience increases in annual rainfall, while
4 This analysis uses a combination of the 11 best SCENGEN models (BMRCTR98, CCSRTR96,
CERFTR98, CSI2TR96, CSM_TR98, ECH3TR95, ECH4TR98, GFDLTR90, HAD2TR95, HAD3TR00,
PCM_TR00) based on their predictive error for annual precipitation levels. Errors were calculated for each
of the models, and for an average of the 17 models. Each model was ranked by its error score, which was
computed using the formula 100*[(MODEL-MEAN BASELINE / OBSERVED) - 1.0]. Error scores
closest to zero are optimal. The error score for an average of the 17 models was 30%, and the error score
for an average of the 11 models was 21%. See the appendix for details.
5 June July August
6 September October November
7 December January February
8 March-April-May
9 A higher resolution analysis across multiple GCMs was beyond the scope of this study.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
14
precipitation is projected to decline in other areas (see Figure 3). However, the timing of these changes
might vary from location to location as well. The National Vulnerability and Adaptation Assessment of
Tanzania (Mwandosaya et al. 1998, which is the bases for the Initial National Communication of 2003)
estimates that northern and southeastern sectors of the country would experience an increase in rainfall
ranging from between 5% and 45% under doubling of carbon dioxide. The central, western, southwestern,
southern, and eastern parts of the country might experience a decrease in rainfall of 10% to 15%. The
southern highlands might similarly experience a decrease of 10%, which could alter the suitability of this
area for maize cultivation. Seasonal patterns in possible changes in rainfall could be complex. For instance,
the northeastern sector might experience an increase of 25%-60% in the short rains and an increase of 20-
45% in the long rains. Or, the north coastal region might get an increase of 0-20% in the short rains and a
decrease of 0-10% in the long rains. In the unimodal region, rainfall might decrease between 0% and 25%
in central regions during October, November, and December, but increase by 15% in March, April, and
May. Finally, the southeastern sector could get between 5 and 45% increase in rainfall during the first three
months of the season and in increase of 10-15% during the last three months.
Figure 3. Change in mean annual rainfall (in %) under 2XCO2
Source: Mwandosaya et al. 1998
The Tanzania National Vulnerability and Adaptation Assessment (1998) as well as the Initial
national Communication (2003) do not include sea-level rise scenarios for Tanzania’s 800 km coastline.
Tide gauge records in Tanzania cover only a very short period of time with some missing data. Instead, a
coastal vulnerability assessment is conducted under two arbitrary sea level rise scenarios 50cm and 1m,
coupled with aerial videotape assisted mapping of coastal topography, resources, and land use. Given the
most recent IPCC assessment (the Third Assessment Report), the 50cm scenarios roughly falls in the
middle, and the 1m scenarios a little beyond the upper estimate of the range of global sea level rise (9cm-
88cm) projected to occur by 2100.
4. Overview of impacts, vulnerabilities and adaptation responses
Given the large size and widely different climatology and climate change projections and impacts
across Tanzania, a national priority ranking might conflate intra-sectoral or sub-national positive and
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
15
negative effects of climate change, and thereby produce misleading results. Therefore, this synthesis
highlights the spectrum of possible sectoral and or regional impacts, and identifies critical impacts and
vulnerabilities, but without an aggregate sectoral or regional ranking. The section concludes with a
discussion of adaptation strategies and priorities for adaptation.
4.1 Agriculture
Agriculture is clearly the most important sector of the Tanzanian economy. It comprised 45.1%
of GDP in 2000 (World Bank, 2002). Upwards of 80% of the population of the country relies directly on
agriculture of one sort or another for their livelihood. Only 3.3% of the cropland was irrigated as of 1999
(World Bank, 2002). The three most important crops are: maize, coffee and cotton – with maize being a
major food staple, coffee a major cash crop grown in large plantations (and contributing significantly to the
GNI), while cotton is another cash crop grown largely by smallholder farmers.
Estimates of the affect of climate change on maize yields are available from model runs of the
Crop Environment Resource Synthesis model (CERES-Maize) (Jones and Kiniry, 1986). In general,
simulation results show that maize yields were lower, a result of higher temperatures and, where
applicable, decreased rainfall. The average yield decrease over the entire country was 33%, but simulations
produced decreases as high as 84% in the central regions of Dodoma and Tabora. Yields in the
northeastern highlands decreased by 22% and in the Lake Victoria region by 17%. The southern highland
areas of Mbeya and Songea were estimated to have decreases of 10-15%. These results suggest that climate
change may significantly influence future maize yields in Tanzania, reducing them in all zones that were
studied, relative to baseline levels. These reductions are due mainly to increases in temperature that shorten
the length of the growing season and to decreases in rainfall. Consequently, the continued reliance on
maize as a staple crop over wide areas of the country could be at risk. The two cash crops on the other hand
are projected to experience increases in yield (Tanzania Initial National Communication 2003). For
Lyamunugu, located within an area of bimodal rainfall, coffee yields are expected to increase by 18%, and
for Mbozi, where rainfall is unimodal, the coffee yield is expected to increase by 16%. These yield
estimates depend critically on estimates of change in precipitation. The potential impacts of climate change
on cotton production in Tanzania parallel that for coffee. The agriculture sector thus may have both
negative and positive impacts that could partially offset each other. However, maize production in
particular might require particular attention for adaptation and mainstreaming responses, given that it is a
critical food crop.
4.2 Forests
Tanzania has about 338,000 km2 under forest cover, which represents about 44% of the total land
area. These forests are an important source of fuel wood and other products for large numbers of
Tanzanians. Furthermore, many of Tanzania’s 43 threatened mammal species, 33 threatened bird species,
and prodigious biodiversity depend on its forests (World Bank 2002). Under climate change most of the
forests across Tanzania are projected to shift towards drier regimes: from subtropical dry forest, subtropical
wet forest, and subtropical thorn woodland to tropical very dry forest, tropical dry forest, and small areas
of tropical moist forest respectively (Tanzania Initial National Communication 2003). Much of this
projected change in distribution is attributed to an increase in ambient temperatures and a decline in
precipitation in forested regions of the country.
Current assessments of climate change impacts on forests in Tanzania however do not explicitly
account for the potential effects of climate change on disturbances such as fire. The Kilimanjaro region
deserves particular attention. In addition to the well-known glacier retreat and eventual disappearance of
the ice cap, there might be major changes in the extent, distribution, and species composition of the forests
on the Kilimanjaro as a consequence of changes in fire regimes. There is indication that intensification of
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
16
fire risk as a result of warmer and drier conditions might already be underway. Continued loss of the
montane forest belt (which collects a significant amount of water from fog entrapment) from fire
intensification would lead to a significant reduction of water yields with serious regional implications,
affecting sectors such as agriculture and livestock as well. These issues as well as possible responses on the
Kilimanjaro are the focus of in-depth analysis later in this report.
4.3 Water resources
Like the agriculture sector, climate change is projected to have both positive and negative
consequences for Tanzania’s water-resources, specifically for the three major river basins: Ruvu, Pangani,
and Rufiji. The Ruvu basin, of particular importance because it is upstream of Tanzania’s major population
center, Dar es Salaam, could experience a 10% decrease in runoff according to the Initial National
Communication (2003). The Pangani basin supplies water to the Tanga, Kilimanjaro, and Arusha regions,
supporting a number of economically important activities. These include the Arusha Chini sugar
plantations in the Kilimanjaro region, the lower Moshi irrigation scheme, the Handeni District water
supply, and a number of important power stations. For the Pagani River, there is some seasonal variation
with runoff projected to increase in some months runoff and decrease in others, with annual basin runoff
decreasing by an estimated 6%. However, the Kikuletwa River, also within the Pagani Basin, is projected
to decrease in all months, with annual reductions of 9%. The Rufiji basin meanwhile is a large catchment
in the south of the country, focused on the Great Ruaha River, which is economically important to the
nation in part because of the hydropower it generates at Mtera Dam and Kidatu Dam. The national
assessment of vulnerability and adaptation (Mwandosaya et al. 1998) projects increases in annual runoff of
5% and 11% at Mtera and Kidatu, respectively, most coming in the period from November to March. All
these estimates however are based on scenarios from a single GCM, and should be interpreted with some
caution. Real uncertainties exist concerning present and future withdrawals for irrigation, changed land
use, and urbanization. Nevertheless, decreases in runoff could potentially have serious affects on
socioeconomic activities in the regions of Dar es Salaam, Morogoro, Tanga, Coast, and Kilimanjaro. Dar
es Salaam might be particularly vulnerable because it is the largest industrial, commercial, and
administrative city in Tanzania.
4.4 Coastal resources
The coastline of Tanzania is about 800 km long and the coastal zone varies in width from 20 km
to 70 km gradually rising to a plateau. Tanzania has relatively limited coastal lowlands, but there are
extensive coastal wetlands, some important cities (Dar es Salaam), a number of important islands (such as
Zanzibar), and a delta — the Rufigi River (Mwaipopo 2001). The main coastal features include mangrove
forests and swamps, coral reefs, sand and mudflats, tidal marshes, woodland, and sisal and cashew nut
estates. Mangrove forests in particular represent an important economic resource for coastal people,
supplying firewood and timber for the construction of fishing boats, and providing feeding, breeding, and
nursery grounds for a number of fish species and a variety of insects, birds, and small animals. The highest
densities of population that might be threatened are found near Dar es Salaam and the islands of Zanzibar
and Pemba.
The Initial National Communication of Tanzania (2003) considers scenarios of both 0.5 m and 1
m sea level rise over the next century. Maps with a 2 m and 20 m contour were used and it was assumed
that land rises linearly from sea level to these contours. The 0.5 m and 1 m contours and the land area they
represent were approximated. Estimates of land lost to erosion were also produced with the aid of aerial
videotape-assisted vulnerability analysis. Total land-loss is estimated to be 247 km2 and 494 km2 for 0.5
and 1 meters of sea level rise respectively. According to this analysis the Dar es Salaam region would be
vulnerable with values of structures at risk estimated to total US$ 48 million for a 0.5 m sea level rise and
US$82 million for a 1 m rise (Tanzania Initial National Communication 2003).
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
17
4.5 Human health
Climate plays an important role in the geographical distribution and seasonal abundance of vector
species that are responsible for the transmission of a number of human diseases. Changes in temperature,
precipitation, humidity, and wind patterns will directly affect vector species’ reproduction, development,
and longevity. The distribution of vector borne diseases in the human population is also limited by
temperature in many regions where the climate is too cold for parasite survival (Martens et al. 1999). Of
the various vector borne diseases malaria in particular is a major public health concern in Tanzania. It
accounts for 16.7% of all reported deaths in Tanzania and is one of the leading causes of morbidity in all
regions, ranging from 24.4% in Rukwa regions to 48.9% in Dar es Salaam (Tanzania Initial National
Communication 2003). Also, the problem of malaria is getting worse because of growing parasite
resistance to first line anti-malarial drugs and mosquito resistance to insecticides. Malaria is endemic in
most of Tanzania even under the current climate. However, many population centers are located in areas
where malaria transmission is currently only epidemic or nonexistent. Most of these centers are located in
the central highlands region (e.g., Mbeya, Njombe, Iringa, and Arusha), where cooler temperatures prevent
or interrupt the transmission of malaria. These areas are of particular concern in considering a warmer
climate. Increased temperatures might open new areas to seasonal or year-around transmission. The
vulnerability of highland populations to an increase in the endemicity of transmission of malaria, or of any
of Tanzania’s population to climate change induced health risks, will depend strongly on the evolution of
control methods and the ability of Tanzania to afford such measures (Tol and Dowlatabadi, 2002).
4.6 Energy, industry and transport
Climate change may also have direct and/or indirect effects on Tanzania’s energy, industry, and
transportation sectors. Among the direct effects, an increase in temperatures would likely increase energy
demands for cooling. Areas projected to have declines in precipitation and or stream flow are also likely to
face increased demands for purposes such as irrigation. However, as highlighted by the discussion on
climate change scenarios, the projections for changes in precipitation remain highly uncertain. The
Tanzania country study also projects decline in stream flow in two key river basins (as discussed in Section
4.3), which will not only increase energy demands for irrigation, but more significantly adversely impact
energy supply, given that these two basins are significant contributors to Tanzania’s hydroelectric
generation. Transportation infrastructure such as railways, roads, pipelines and ports may also be at risk
from impacts of climate change (particularly sea level rise), but specific vulnerability analyses are lacking.
Other potential impacts of climate change on energy supply include the vulnerability of the Songo Songo
and Mnazi Bay natural gas reserves to sea level rise.
4.7 Overview of adaptation responses
While uncertainties in climate change and impacts projections are a characteristic feature that
poses a challenge for anticipatory adaptation for any country, Tanzania’s case has several characteristics
that might argue for a differentiated adaptation strategy. First, the climate change projections on which all
national impact and vulnerability assessments are based (all the way to the Initial National Communication
of 2003) rely on a limited number of older generation of climate models and scenarios, circa early 1990s
which has several implications for assessment of impact and adaptation options. For example, an analysis
based on more recent climate models conducted as part of this study concludes that the magnitudes of
temperature increases projected for Tanzania might be somewhat lower (though the trends are broadly
consistent) with the projections used in the National Assessment of Vulnerability and Adaptation. Thus,
information on impacts might need updating in Tanzania prior to the formulation of aggressive adaptation
responses, more so than in other countries where projections might be based on more recent models.
Second, some key sectors are projected to experience both positive and negative impacts under climate
change – for example, while production of maize is projected to decline, the production of two key cash
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
18
crops (coffee and cotton) is projected to increase. Similarly, while stream-flows are projected to decline in
two of three key river basins (Ruvu and Pangani), they are projected to increase in the third (Rufiji). The
implication for adaptation therefore might be to not only cushion adverse impacts, but also to harness
positive opportunities. Finally, a third key characteristic is that unlike most other countries where the need
for adaptation relies largely on projections of future impacts, there might be some discernible trends in
climate and attendant impacts already underway in Tanzania. This might argue for more immediate
adaptation measures in the case of such impacts as opposed to a “wait and see” strategy.
For all the above reasons, there might be a need for a differentiated adaptation strategy across
various sectors and regions depending upon the certainty of projections, the mix of beneficial and adverse
impacts, and the urgency and timing of such impacts. For the case of agriculture a key portfolio of
adaptation responses would involve measures that boost maize production: increased irrigation, increased
use of manure and fertilizer, and better use of management tools including climate information. These
measures are discussed in Tanzania’s Initial National Communication. However, given that the production
of the country’s two cash crops (coffee and cotton) is projected to increase under the same climate
scenarios, another adaptation response – which is not discussed in Tanzania’s Initial National
Communication - might involve a strategic shift over the medium to the long term from maize towards
these cash crops.
With regard to human health, the spread of malaria to the population centers in the highlands as a
result of rising temperatures is a key concern. Much of Tanzania however is already malaria endemic, so
policy responses might need to be driven by the additionality of the disease burden, and not necessarily the
existence of the risk itself. Most roll back malaria programs function in the reactive mode (antimalarial
drugs, spraying of insecticides, and elimination of breeding sites), while in the cases of highland areas
precautionary adaptation to prevent or limit the spread of malaria to these regions might be ideal.
For coastal resources meanwhile a key priority is to construct regional sea level rise scenarios,
that not only incorporate local topography (as has been done) but also subsidence rates. Lacking such
specific information, and given the long time-scales at which sea level rise will manifest itself, an initial set
of adaptation priorities should ideally focus on no regrets measures in particularly low lying or otherwise
vulnerable areas including urban areas as well as coastal wetlands and mangroves, such as the Mafia Island
Marine Park, the Menai Bay Conservation Area, and the Misali Island Conservation Area. Coastal zones
adaptation priorities may also be synergistic with several ongoing government-donor initiatives including
the Conservation of Lowland Coastal Forests Project, the Sustainable Dar es Salaam Project and the
Tanzania Coastal Management Partnership.
No regrets adaptation - specifically water and energy conservation – could be a viable initial
priority for adaptation in water resources, where stream-flow is projected to decline in two critical river
basins (Ruvu and Pangani), affecting water use and hydroelectricity generation. The Tanzania Initial
National Communication identifies privatization (as is already the case for Dar es Salaam) as a key
adaptation response to promote efficient water use. This measure however may have equity effects as it
may result in an increase in price of water making it unaffordable to the poor. Further, given that roughly
half of the water in Dar es Salaam is lost to leakage, a second key no regrets response would be leakage
prevention – although it would require significant new capital investment and regular maintenance of water
infrastructure. Third, given that streamflow in a third river basin – the Rufiji – is projected to increase,
another adaptation strategy may revolve around water transfer from Rufiji to Dar es Salaam which relies on
the Pangani. However, given that streamflow projections are based upon the results from one water balance
model (and dated climate scenarios), and the fact that the costs and environmental impacts of inter-basin
transfers are yet to be analyzed, such a response may not be advisable at this time.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
19
With regard to the energy, industry and transportation sectors, the Tanzania National Action Plan
(1997) has conducted a hierarchical screening of potential adaptation responses. Given that climate change
is projected to impact energy demands (through rise in temperature), as well as particular sources of energy
supply (hydro and to some extent natural gas fields located in coastal areas), these adaptation options focus
on either demand side management or on the promotion of energy supply sources that are not impacted by
climate change. Figure 4 shows the results of various adaptation responses from this hierarchical screening.
A majority of these measures are no-regrets. However some measures – particularly a fuel switch to
kerosene – may run into conflict with greenhouse gas mitigation, as they may imply a switch away from an
energy source of lower carbon intensity (natural gas and hydro). Therefore, synergy between mitigation
and adaptation responses, as well as with other development priorities must also be considered in screening
adaptation measures.
Figure 4. Screening of adaptation measures in the energy, industry and transportation sectors
Ranking of Adaptation Measures
0
10
20
30
40
50
60
70
80
90
100
Measures
Weighted Points
Power load shedding
Fuel switch to kerosene
Fuel switch to other fuels
Interconnection
Emergency Power Plants
Minihydro and geothermal plants
Demand Side Management
Efficiency in industry
Change of Products
Import off-road vehicles
Railway Maintenance
Road maintenance
Source: National Action Plan on Climate Change 1997
One area where the need for adaptation may be immediate is the Kilimanjaro ecosystem where
climatic changes are likely already contributing to significant impacts on the natural and human system,
including the intensification of fire risk, in part a consequence of observed changes in temperature and
precipitation patterns, and to a lesser extent the retreat of the ice cap. The causes and implications of these
impacts, as well as potential responses to them and the potential synergies and conflicts with
environmental and development priorities are investigated in-depth later in Section 8.
5. Attention to climate concerns in donor activities
Tanzania receives large amounts of donor aid, in the order of one billion US$ per year, which is
equivalent to about 11% of its GNI. The largest donors, in terms of overall investments, are the World
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
20
Bank (IDA), Japan, and the United Kingdom. Figure 5 displays the distribution of this aid by development
sector and by donor.
Figure 5. Development aid to Tanzania (1998-2000
The following sections highlight the possible extent of climate risks to development investments
in Tanzania and examine to what extent current and future climate risks are factored in to development
strategies and plans, as well as individual development projects 10. Given the large quantity of strategies
and projects, our analysis is limited to a selection. This selection was made in three ways (i) a direct
request to all OECD DAC members to submit documentation of relevant national and sectoral strategies,
as well as individual projects (ii) a direct search for some of the most important documents (including for
instance national development plan/PRSP, submissions to the various UN conventions, country and sector
strategies from multilateral donors like the World Bank and UNDP, and some of the larger projects in
climate-sensitive sectors), and (iii) a pragmatic search (by availability) for further documentation that
would be of interest to our analysis (mainly in development databases and on donors’ external websites).
Hence, the analysis is not comprehensive, and its conclusions are not necessarily valid for a wider array of
development strategies and activities. Nevertheless, the authors feel confident that this limited set allows
an identification of some common patterns and questions that might be relevant for development planning.
5.1 Donor activities affected by climate risks
This section explores the extent to which development activities in Tanzania are affected by
climate risks, which gives an indication of the importance of climate considerations in development
planning. The extent to which climate risks affect development activities in Tanzania can be gauged by
examining the sectoral composition of the total aid portfolio, which is analyzed here using the World
10 The phrase “climate risk” or “climate-related risk” is used here for all risks that are related to climatic
circumstances, including weather phenomena and climate variability on various timescales. In the case of
Tanzania, these risks include the effects of seasonal climate anomalies, including droughts, as well as
trends therein due to climate change, and risks due to sea level rise. “Current climate risks” refer to climate
risks under current climatic conditions, and “future climate risks” to climate risks under future climatic
conditions, including climate change and sea level rise.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
21
Bank/OECD Creditor Reporting System (CRS) database (Box 2). Development activities in sectors such as
agriculture, infectious diseases, or water resources could clearly be affected by current climate variability
and weather extremes, and consequently also by changing climatic conditions. At the other end of the
spectrum, development activities relating to education, gender equality, and governance reform are much
less directly affected by climatic circumstances.
Box 2. Creditor Reporting System (CRS) Database
The Creditor Reporting System (CRS) comprises of data on individual aid activities on Official Development
Assistance (ODA) and official aid (OA). The system has been in existence since 1967 and is sponsored and operated
jointly by the OECD and the World Bank. A subset of the CRS consists of individual grant and loan commitments (from
6000 to 35000 transactions a year) submitted by DAC donors (23 members) on a regular basis. Reporters are asked to
supply (in their national currency), detailed financial information on the commitment to the developing country such as:
terms of repayment (for loans), tying status and sector allocation. The secretariat converts the amounts of the projects
into US dollars using the annual average exchange rates.
In principle, the sectoral selection should include all development activities that may be designed
differently depending on whether or not climate risks are taken into account. In that sense, the label
“affected by climate risks” has two dimensions. It includes projects that are at risk themselves, such as an
investment that could be destroyed by flooding. But it also includes projects that affect the vulnerability of
other natural or human systems. For instance, new roads might be fully weatherproof from an engineering
standpoint (even for climatic conditions in the far future), but they may also trigger new settlements in
high-risk areas, or it may have a negative effect on the resilience of the natural environment, thus exposing
the area to increased climate risks. These considerations should be taken into account in project design and
implementation. Hence, these projects are also affected by climate risks. A comprehensive evaluation of
the extent to which development activities are affected by climate change would require detailed
assessments of all relevant development projects as well as analysis of site specific climate change
impacts, which was beyond the scope of this analysis. This study instead assesses activities affected by
climate risks on the basis of CRS purpose codes (see Appendix B, which identifies “the specific area of the
recipient’s economic or social structure which the transfer is intended to foster”)11, 12.
Clearly, any classification that is based solely on sectors suffers from oversimplification. In
reality, there is a wide spectrum of exposure to climate risks even within particular sectors. For instance,
rain-fed agriculture projects may be much more vulnerable than projects in areas with reliable irrigation. At
the same time, the irrigation systems themselves may also be at risk, further complicating the picture.
Similarly, most education projects would hardly be affected by climatic circumstances, but school
buildings in flood-prone areas may well be at risk. Without an in-depth examination of risks to individual
projects, it is impossible to capture such differences. Hence, the sectoral classification only provides a
rough first sense about the share of development activities that may be affected by climate risks.
To capture some of the uncertainty inherent in the sectoral classification, the share of
development activities affected by climate change was calculated in two ways, a rather broad selection, and
a more restrictive one. The first selection includes projects dealing with infectious diseases, water supply
11 Each activity can be assigned only one such code; projects spanning several sectors are listed under a
multi-sector code, or in the sector corresponding to the largest component.
12 The OECD study “Aid Activities Targeting the Objectives of the Rio Conventions, 1998-2000” provides a
similar, but much more extensive database analysis. It aimed to identify the commitments of ODA that
targeted to objectives of the Rio Conventions. For this purpose, a selection was made of those projects in
the CRS database that targeted the Conventions as either their “principal objective”, or “significant
objective”.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
22
and sanitation, transport infrastructure, agriculture, forestry and fisheries, renewable energy and
hydropower13, tourism, urban and rural development, environmental protection, food security, and
emergency assistance. The second classification is more restricted. First of all, it excludes projects related
to transport and storage. In many countries, these projects make up a relatively large share of the
development portfolio, simply due to the large size of individual investments (contrary to investments in
softer sectors such as environment, education and health). At the same time, infrastructure projects are
usually designed on the basis of detailed engineering studies, which should include attention at least to
current climate risks to the project.14 Moreover, the second selection excludes food aid and emergency
assistance projects. Except for disaster mitigation components (generally a very minor portion of
emergency aid), these activities are generally responsive and planned at short notice. The treatment of risks
is thus very different from well-planned projects intended to have long-term development benefits.
Together, the first and the second selection give an indication of the range of the share of climate-affected
development activities.
In addition, the share of emergency-related activities was calculated. This category includes
emergency response and disaster mitigation projects, as well as flood control. The size of this selection
gives an indication of the development efforts that are spent on dealing with natural hazards, including,
often prominently, climate and weather related disasters.
The implications of this classification should not be overstated. If an activity falls in the “climateaffected”
basket, which does not mean that it would always need to be redesigned in the light of climate
change or even that one would be able to quantify the extent of current and future climate risks. Instead,
the only implication is that climate risks could well be a factor to consider among many other factors to be
taken into account in the design of development activities. In some cases, this factor could be marginal. In
others, it may well be substantial. In any case, these activities would benefit from a consideration of these
risks in their design phase. Hence, one would expect to see some attention being paid to them in project
documents, and related sector strategies or parts of national development plans. Figures 6 and 7 show the
results of these selections, for the three years 1998, 1999, and 200015.
13 Traditional power plants are not included. Despite their long lifetime, these facilities are so localized
(contrary to, e.g., roads and other transport infrastructure) that climate risks will generally be more limited.
Due to the generally large investments involved in such plants, they could have a relatively large influence
on the sample, not in proportion with the level of risk involved.
14 Note however, that they often lack attention to trends in climate records, and do not take into account
indirect risks of infrastructure projects on the vulnerability of natural and human systems.
15 The three-year sample is intended to even out year-to-year variability in donor commitments. At the time
of writing, 2000 was the most recent year for which final CRS data were available. Note that coverage of
the CRS is not yet complete. Overall coverage ratios were 83% in 1998, 90% in 1999, and 95% in 2000.
Coverage ratios of less than 100% mean that not all ODA/OA activities have been reported in the CRS. For
example, data on technical co-operation are missing for Germany and Portugal (except since 1999), and
partly missing for France and Japan. Some aid extending agencies of the United States prior to 1999 do
not report their activities to the CRS. Greece, Luxembourg and New Zealand do not report to the CRS.
Ireland has started to report in 2000. Data are complete on loans by the World Bank, the regional banks
(the Inter-American Development Bank, the Asian Development Bank, the African Development Bank)
and the International Fund for Agricultural Development. For the Commission of the European
Communities, the data cover grant commitments by the European Development Fund, but are missing for
grants financed from the Commission budget and loans by the European Investment Bank (EIB). For the
United Nations, the data cover the United Nations Children's Fund (UNICEF) since 2000, and a significant
proportion of aid activities of the United Nations Development Programme (UNDP) for 1999. No data are
yet available on aid extended through other United Nations agencies. Note also that total aid commitments
in the CRS are not directly comparable to the total ODA figures in Figure 5, which exclude most loans.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
23
Figure 6. Share of aid amounts committed to activities affected by climate risk and to emergency in Tanzania
(1998-2000
d a rk : a ffe c te d b y
c lim a te r is k s
(h ig h e s tim a te )
26%
74%
dark: affected by
climate risks
(low estimate)
12%
88%
dark: emergency
activities
97
%
3%
Figure 7. Share (by number) committed to activities affected by climate risk and to emergency activities in
Tanzania (1998-2000
dark: affected by
climate risks
(high estimate)
31
%
69
%
dark: affected by
climate risks
(high estimate)
21
%
79
%
dark: emergency
activities
96
%
4%
Emergency projects make up 3 to 4% of all activities. In monetary terms, between one-eighth and
a quarter of all development activities in Tanzania could be affected by climate change. By number, the
shares are higher, between about 20 and 30 percent16. In addition to providing insight in the sensitivity of
16 Note that the number of activities gives a less straightforward indication than the dollar amounts. First of
all, activities are listed in the CRS in each year when a transfer of aid has occurred. Hence, when a donor
disburses a particular project in three tranches, that project counts three times in our three-year sample. If
the financing for a similar three-year project is transferred entirely in the first year, it only counts once.
Secondly, the CRS contains a lot of non-activities, including items like “administrative costs of donors”.
Moreover, some bilateral donors list individual consultant assignments as separate development activities.
In most cases, such transactions will fall outside of the “climate-affected” category. Hence, the share of
climate-affected activities relative to the total number of activities (which is diluted by these non-items) is
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
24
development activities in Tanzania as a whole, the classification also gives a sense of the relative exposure
of various donors. These results are listed in Table 2 and 3 (again in the years 1998, 1999, and 2000).
Table 2. Relative shares of CRS activities, by total disbursed amounts, for the top-five donors in Tanzania
(1998-2000)
Amounts of activities
(millions US$)
Activities affected by
climate risks
(high estimate)
Activities affected by
climate risks
(low estimate)
Emergency activities
Donor
Amoun
t % Donor
Amoun
t % Donor Amount % Donor
Amoun
t %
Total 2916
100
% Total 761 100% Total 356 100% Total 81
100
%
UK 524 18%
CEC/ED
F 134 18%
German
y 50 14% USA 31 39%
IDA 453 16% Denmark 81 11% UK 41 12% AfDF 13 16%
Japan 326 11% Germany 72 9% Japan 38 11% UK 11 13%
CEC/EDF 264 9% Japan 71 9% IFAD 33 9% CEC/EC 8 10%
Denmark 215 7% UK 58 8% Norway 32 9% Sweden 7 8%
Table 3. Relative shares (by number) of CRS activities for the top-five donors in Tanzania (1998-2000)
Numbers of activities
Activities affected by
climate risks
(high estimate)
Activities affected by
climate risks
(low estimate)
Emergency activities
Donor Number % Donor Number % Donor
Numbe
r % Donor Number %
Total 1745 100% Total 536 100% Total 369 100% Total 76 100%
Sweden 232 13% Ireland 72 13% Ireland 64 17% Switzerl. 16 21%
UK 222 13% UK 67 13% UK 42 11% Sweden 15 20%
Norway 210 12% Norway 53 10% Norway 37 10% UK 14 18%
Ireland 191 11% Sweden 46 9% Sweden 31 8% Norway 6 8%
Germany 124 7% Germany 36 7%
Germany 30 8% Finland 5 7%
Given the extensive share of development activities in Tanzania that could be affected by climate
risks, one would assume that these risks are reflected in development plans and a large share of
development projects. The following sections examine to which extent this is the case.
5.2 Attention to climate risks in donor strategies
Tanzania regularly suffers from various climate-related hazards, including droughts that have
substantial effects on economic performance and poverty. Many development plans and projects recognize
this influence, and Tanzania’s climate even turns up in the context of economic analyses. However, few of
the development plans and projects that were reviewed take these risks into account. Given that current
climate risks are already being neglected, it comes as no surprise that climate change is often ignored
lower. On the other hand, the shares by total amount tend to be dominated by structural investments (which
tend to be more costly than projects in sectors such as health, education, or environmental management).
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
25
altogether. In the few cases where climate change does receive attention, the focus is on mitigation, rather
than adaptation.
Several donor strategies recognize Tanzania’s dependence on favourable weather, and the
linkages between poverty, drought, and food security. For instance, the AfDB Country Strategy Paper
highlights the impact of weather on economic performance: “growth rates have been fluctuating from year
to year reflecting the vulnerability of the economy to external shocks. Although strong growth was
registered in FY 1996/97 (4.2 percent), it declined to 3.3 percent in FY 1997/98 due to the adverse impact
of the drought on agricultural output. The drought was followed by the El-Nino floods late 1997 and early
1998, which destroyed some of the crops and damaged roads, thereby, disrupting internal movement of
agricultural commodities as well as export shipments.” IFAD’s Country Strategic Opportunities Paper
estimates that the country has a structural food deficit of about 700 tons, with imports rising to up to 1.5
million tons in times of flood or drought. This vulnerability cannot be attributed to weather conditions
alone. For instance, the AfDB paper notes that less than 20% of the irrigation potential is utilized,
unnecessarily exposing agricultural production to droughts. “While droughts have contributed to water
supply problems, the underlying factors include weak institutional capacity in the sector, poor water
resource management, and the dilapidated condition of the water schemes and distribution networks in the
rural and urban areas resulting from the under-funding of maintenance and rehabilitation.” The UN
Development Assistance Framework (UNDAF) also highlights Tanzania’s vulnerability to climate-related
disasters, due to natural and human factors: “Natural and man-made disasters erode the coping capacity of
the vulnerable population especially in drought-prone areas. There have been poor rains in Central
Tanzania for the last three years, and traditional coping strategies are breaking down as land pressure
increases. These types of shocks have become a frequent phenomenon in Tanzania in recent years. Floods
and droughts, epidemics and crop pests, environmental damage and economic instabilities, have all had
their effects on people’s capacity to meet their basic needs and subsequently their ability to survive and
pursue their development ambitions and potential”.
Despite these strong linkages between climate and economic performance, as well as the
relationships between droughts, environmental degradation and poverty, none of the donor strategies even
mentions climate change. Attention to current hazards, particularly droughts, varies from donor to donor.
Some of them, including SIDA, Ireland Aid and the EU, do not explicitly recognize the impact of current
climate-related risks on the success of development investments. Others, such as DFID and IFAD, have
components that aim to address Tanzania’s vulnerability to such risks.
In 2001, a joint “Emergency Consolidated Appeal for the Drought in Tanzania” was launched by
a number of UN agencies. Instead of just addressing short-term relief, the appeal intended to address the
underlying causes of the chronic droughts, including early warning systems, and drought mitigation
measures in Rural and Agricultural Development Strategies. Despite the longer-term focus of the appeal,
climate change was not considered.
5.3 Climate risks in selected development programs and projects
None of the (relatively few) development projects that were reviewed paid attention to the risks
associated with climate change. For instance, a World Bank forest conservation and management project,
which explicitly addresses climate change through carbon uptake, does not address current or future
climate-related risks. A regional GEF-funded World Bank project to improve the long-term environmental
management of Lake Victoria, does not consider the potential impacts of climate change on the water
resources and ecosystems at stake, and a USAID-sponsored coastal management partnership neglects sea
level rise in its analysis of integrated coastal zone management options.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
26
6. Attention to climate concerns in national planning
Since attaining its independence in 1961, Tanzania has been addressing its development process
through long, medium and short-term development plans and programs, which are developed by the
Planning Commission in the Ministry of Planning and Privatization. See Table 4 for an overview on
Tanzania’s planning history. The latest medium-term program is the so-called Three Year Rolling Plan and
Forward Budget, which rolls on an annual basis and has been in place since 1993/94 up to the present. The
major macroeconomic and sectoral policy objectives and cross-sectoral issues included in Tanzania’s plan
are poverty alleviation, population, science and technology as well as environmental protection.
Besides, Tanzania also embarks on long term planning, the latest being the National
Development Vision 2025, which aims for economic prosperity, equity, self-reliance, the transformation
from a rural based agricultural economy to a more diversified and industrialized one, as well as
sustainability by the year 2025. Despite the Vision’s long time horizon, climate change is not mentioned. It
neither discusses climate-related risks, nor strategies to mitigate or to adapt to them (such as irrigation,
reforestation, and crop diversification). Similarly, the shorter-term (5-year) Tanzania Assistance Strategy
(“a medium-term framework for promoting local ownership and development partnerships”) does not
discuss climate change either. However, climate-related risks, mainly floods and droughts, feature
prominently. Besides specific attention to disaster preparedness activities, the plan also advocates the
integration of disaster mitigation in Tanzania’s development plans.
Table 4. Tanzania’s main planning documents
National Plans Period
National Development Plans
Three Year Plan
First Five Year Plan
Second Five Year Plan
Third Five Year Plan
First Union Five Year Plan
Second Union Five Year Plan
Three Year Rolling Plan and Forward Budget (rolls annually)
1961-1963/64
1964/65-1968/69
1969/70-1973/74
1976/77-1980/81
1981/82-1985/86
1988/89-1992/93
1993/94 to date
Emergency Plans
National Economic Survival Programme
Structural Adjustment Programme
Economic Recovery Programme (ERP-I)
Economic Recovery Programme (ERP-II)
1982
1983-85
1986/87-1988/89
1989/90-1991/92
Long Term Perspective Plans
15 Year Development Plan
20 Year Development Plan
National Development Vision 2025
1964-1980
1981-2000
1998-2025
In 1997 Tanzania developed a first National Action Plan on Climate Change, which contained an
inventory of emissions by source and removal by sinks of greenhouse gases based on 1990 data. Besides
the Action Plan, various studies focusing on technological and other options for mitigating greenhouse
gases in Tanzania as well as on the assessment of vulnerabilities and possible adaptation measures have
been completed. Tanzania has also signed or ratified a number of multilateral environmental agreements,
and has a number of national level environmental and sectoral plans that intersect with responses that may
be required to manage climate variability and long term climate change.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
27
6.1 National Action Plan on Climate Change
The National Action Plan on Climate Change was developed in 1997 and has different objectives
for various timeframes:
6.1.1 Short term program (Year 1 - 2)
In the beginning efforts should be undertaken to raise awareness of possible impacts stemming
from climate change on various social and economic activities. The overall aim of these meetings would be
to explore possibilities of how current activities or sectoral plans could complement climate change
mitigation options. Besides, there is a need to analyze the effects of governmental macroeconomic policies
in relation to climate change.
6.1.2 Medium term program (Year 2 - 5)
In the medium term, projects already internalizing climate change aspects, especially those
reducing GHG emissions, should be supported. Support will either be sought from internal such as the
Government budget or from external sources. In addition, climate change aspects should be included into
the educational curriculum, preferably starting at secondary school level. Also, the Government should
start introducing environmental economic instruments such as fiscal measures (pollution taxes, input taxes,
product taxes, import tariffs, royalties , land user taxes, tax differentiation etc), property rights (ownership
right, user right, and development rights], and performance bonds (land reclamation bond, waste delivery
bond, environmental performance bond, etc.) as incentives to increase environmental conservation.
6.1.3 Long term program (Year 10 - 20)
In the long-term, large projects in the energy and transport sector should be undertaken. In
addition, adaptation measures to cope with a rising sea level and its adverse effects on coastal
infrastructures should be implemented.
6.2 National communications to international environmental agreements
Tanzania is a party to various international environmental agreements, including the UNFCCC,
UNCCD, and UNCBD. Tanzania recently submitted its Initial National Communication to the UN FCCC
(in July 2003), and is currently preparing a National Adaptation Programme of Action (NAPA).
While Tanzania’s National Report to the UN Convention on Biodiversity does not mention
climate change at all, its First National Report to the UN Convention to Combat Desertification refers only
to climate change mitigation mainly through the diversification of Tanzania’s energy resources. The
Second National Report to the UNCCD, however, does highlight the linkages between climate change and
desertification. It also notes that desertification programs have been quite successful, not only in terms of
awareness raising, but also by mainstreaming desertification concerns in national and sectoral plans and
policies.
Tanzania’s National Report to World Summit on Sustainable Development (2002) refers to the
national vulnerability and adaptation assessment, and explicitly lists agriculture, water resources, forestry,
grasslands, livestock, coastal resources and wildlife and biodiversity as vulnerable to climate change.
Nevertheless, adaptation receives very little attention (except in agriculture, where further work is
planned), in sharp contrast to mitigation, which is discussed extensively. Several components of potential
climate change adaptation strategies are included in efforts to address current-day vulnerability to climaterelated
risks, including better water management, for instance in the context of irrigation development, and
research on drought-resistant, high-yield crops.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
28
6.3 Poverty Reduction Strategy Paper (PRSP)
Although Tanzania’s PRSP recognizes the grave impact of weather and climate hazards on
development, and particularly on the poor, it neglects climate change. The important impact of climaterelated
risks, however, is clearly recognized. For instance, stakeholder groups that were interviewed in
preparation for the poverty strategy voiced their worries: “A major concern of the poor is their
vulnerability to unpredictable events. In Tanzania, famine often results from either floods or drought. Since
the mid-1990s, Tanzania has in fact experienced a series of adverse weather conditions, which undermined
food security. […] There is, therefore, a growing need for safety-nets.” In response, the PRSP lists a
number of activities to reduce this vulnerability, including early warning systems, irrigation, better food
supply systems, development of drought resistant crops, facilitation of the provision of adequate, safe and
clean water to the rural areas from 48.5% population coverage in 2000 to 85% by 2010, promotion of the
use of rainwater harvesting and sustained efforts in reforestation as well as sustained efforts in adaptation.
The PRSP progress report, which was published a year later, notes that agricultural growth has
been lagging behind expectation “owing to adverse weather and the collapse of export prices”. Despite this
observation, the report’s response to this lagging growth includes no direct measures to reduce
vulnerability to climate risks, not even the ones mentioned in the original PRSP a year earlier. Similarly,
these options are also neglected in the World Bank/IMF Joint Staff Assessments of the PRSP and the PRSP
progress report, suggesting that climate-related risks do not get much attention in the PRSP oversight
process.
6.4 Other national policies of relevance to climate change
Tanzania has put in place a number of environmental and sectoral policies and plans especially
during the 1990s, which are intended to increase its ability to cope with current environmental problems as
well as with additional risks posed by climate change. The following paragraphs discuss some of the most
relevant policies.
The National Environmental Action Plan (NEAP) of 1994 was a first step towards incorporating
environmental concerns into national planning and development. NEAP identified six priority
environmental concerns, namely land degradation; lack of accessible, good quality water for both urban
and rural inhabitants; pollution; loss of wildlife habitats; deterioration of marine and freshwater systems;
and deforestation. In order to address these issues the National Environmental Policy (NEP) was
promulgated in December 1997 to provide a framework for mainstreaming environmental considerations
into the decision-making processes in Tanzania. Though NEP does not pay explicit attention to climate
change, the primary environmental issues brought forward include many of the concerns that would be
addressed by no-regrets climate change adaptation measures. In particular, the NEP highlights the
importance of integrating environmental management in several sectoral programs and policies.
A particularly strong example of such integration is found in the agriculture sector, which is
crucial for food security and the eradication of rural poverty. The NEP, for example, proposes “the
improvement of land husbandry through soil erosion control and soil fertility improvement; the
minimization of encroachment in public lands including forests, woodlands, wetlands, and pastures; the
strengthening of environmentally sound use, monitoring, registration and management of agrochemicals;
as well as the improvement in water use efficiency in irrigation, including control of water logging and
salinization.” In addition, the forestry section of NEP is most explicit in giving attention to cross-sectoral
environmental issues: “the main objective is the development of sustainable regimes for soil conservation
and forest protection, taking into account the close linkages between desertification, deforestation,
freshwater availability, climate change, and biological diversity.” The only other paragraph in the NEP
that relates to climate change reads as follows “The need to undertake climate studies in order to come up
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
29
with mitigation options is stressed. In view of Tanzania’s vulnerability to climate variations, an assessment
of impacts of climate change and climate variations will be undertaken. In this regard strategies will be
evolved to ensure that options which are pursued do not unduly sacrifice national development endeavors.”
Similarly, the 1998 National Forestry Policy (NFP), which is a review of the 1953 one, gives no
direct references to climate change despite the vulnerability of Tanzanian forests to changed climatic
conditions. One of the main objectives of the NFP is to ensure ecosystem stability through conservation of
forest biodiversity, water catchments, and soil fertility. The policy states that new forest reserves for
conservation will be established in areas of high biodiversity value and that biodiversity conservation and
management will be included in the management plans for all protected forests. This policy is a great
departure from the traditional forestry approach of command-and-control by involving communities and
other stakeholders through joint management agreements.
Likewise, despite the criticality of climate change impacts on water resources the new National
Water Policy (NAWAPO), which has been approved by the Tanzanian cabinet in July 2002, does not
explicitly mention the issue. Nonetheless, the NAWAPO is participatory, multi-sectoral, river-basin based
and tries to integrate land use with water use and water quality as well as quantity. The four key issues in
the revised policy are 1) the demand respond approach, which leads to community ownership and
management of water and sanitation facilities; 2) private sector participation; 3) integration of water supply
and sanitation and finally 4) decentralization of service delivery from central government to district
councils.
7. Climate change and Mount Kilimanjaro
Mount Kilimanjaro derives its name from the Swahili words Kilima Njaro meaning “shining
mountain”, a reference to its legendary ice cap. It is the retreat of this ice cap, arguably linked to rising
temperatures, that has made the Kilimanjaro a prominent symbol of the impacts of global climate change.
Beyond the symbolism of the ice cap Kilimanjaro is also a hot spot of biodiversity with nearly 3000 plant
species and providing a range of critical ecosystem services to over one million local inhabitants who
depend on it for their livelihoods, as well as to the broader region that depends on water resources that
originate at the Kilimanjaro. The Kilimanjaro ecosystem is also subject to wide ranging impacts that may
be more directly attributable to changes in temperature and precipitation patterns, and which may have far
greater significance than the melting of the ice cap itself. This in-depth analysis has two objectives: (i) to
provide an overview on the impact of climatic changes on Mt. Kilimanjaro and on the resulting impacts on
the environment, ecosystems and on the human population; and (ii) to describe adaptation responses to
reduce or manage some of the most critical impacts and their synergy or conflict with other environmental
and development priorities.
Mt. Kilimanjaro is located 300 km south of the equator in Tanzania, on the border with Kenya. It
is the highest mountain in Africa, a huge strato-volcano (ca. 90 by 60 km), composed of three single peaks,
Kibo, Mawenzi and Shira that reach respectively an altitude of 5,895, 5,149 and 3,962 meters (Figure 8).
Kilimanjaro is also the world’s highest free standing mountain, looming 5,000 meters above an open
undulating plain that averages around 1,000 meters above sea level. The morphology of the upper areas of
Mt. Kilimanjaro is formed by glaciers which reached down to an altitude of 3000 m above sea level (asl)
during the ice age (Downie & Wilkinson 1972, Hastenrath 1984).
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
30
Figure 8. Mount Kilimanjaro
7.1 Climate, glaciers, and hydrology
Mt. Kilimanjaro is characterized by a typical equatorial day-time climate. Due to its near-equator
location, it experiences two distinct rainy seasons: the long rains from March to May forming the main
rainy season; and the short, but heavy rains centered on the month of November of the small rainy season.
The driest period falls into the months from July to October, while April and May are the wettest months.
However, rainfall and temperature vary with altitude and exposure due to the dominant wind blowing from
the Indian Ocean. Annual rainfall reaches a maximum of around 3,000 mm at 2,100 meters on the central
southern slope in the lower part of the forest belt, clearly exceeding precipitation on other East African
high mountains (HEMP 2001a). Higher up at 2,400, 2,700 and 3,000 meters, approximately 90, 70 and 50%
respectively of this maximum precipitation has been observed. The northern slopes, on the leeward side of
the mountain, receive much less annual rainfall (Figure 9).
The mean annual temperature in Moshi township (813 m) is 23.4°C (Walter et al. 1975). It
decreases to 9.2°C at an altitude of 3100 m, 5.0°C at 4000 m (HEMP, unpub. data) and –7.1°C on top of the
Kibo peak at about 5800 m (Thompson et al. 2002), with a lapse rate of about 0.6°C per 100 m increase in
altitude. The climate in the alpine belt above 3500-4000 m is characterized by extremes, with nightly frosts
and intense sunshine during daytime all year round (HEDBERG 1964).
Kilimanjaro represents a rare instance of the occurrence of glaciers in equatorial regions and like
the glaciers of Rwenzori and Mt. Kenya these are a relic of the colder and wetter climatic conditions of the
region during the Pleistocene (Downie & Wilkinson 1972). At present permanent ice exists only on Kibo -
covering an area of 2.6 km2 (Thompson et al. 2002). Yet, the distribution of moraines reaching down to an
altitude of 3000 m indicates that a much greater area of the mountain was formerly covered by ice (Downie
& Wilkinson 1972, Hastenrath 1984).
Mt. Kilimanjaro is a critical water catchment for both Tanzania and Kenya. High rainfall and
extensive forests give Mt. Kilimanjaro its high catchment value. The southern and the south-eastern forest
slopes form the main upper catchments of the Pangani River, one of Tanzania’s largest rivers, which drains
into the Indian Ocean near Tanga. Although the greater aridity of the northern slopes is reflected in a
sparser network of valleys on this side, in their shallower cross-section and in the general absence of
running water above 3000 m (Downie & Wilkinson 1972), the north-western slopes form the catchment of
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
31
the Tsavo River, a tributary of the Galana River, one of Kenya’s major rivers. The Amboseli National Park
in Kenya also depends on the hydrology of Mt. Kilimanjaro17.
Figure 9. Annual precipitation on Mt. Kilimanjaro
Source: Hemp 2001a
Mt. Kilimanjaro’s ice cap is relatively small in comparison to its height and surface area and its
contribution in developing water sources must be assumed to be equally slight (Ramsay 1965)18. Very few
streams originate in this zone and most of these have small flows. In contrast, the montane forest belt
between 1600 and 3100 m provides most of the water (96%) coming from the mountain (Ramsay 1965). In
17 Further afield in Kenya, it is likely that the mountain has an effect on Ol Turesh swamp and possibly
Mzima Springs, whose primary catchment is the Chyulu Hills.
18 Only Weru-Weru and Kikafu River, important branches in the headwaters of the Pangani River, are linked
by permanent streams to glaciers on the south-west edge of Kibo. The relatively few springs between the
ice cap and the forest belt indicate that the percolation of melt water downwards through the permeable
surface volcanic ash is small as well. Therefore - except below the glaciers of south-west Kibo - the valleys
above 3600 m are dry for a large part of the year (Downie & Wilkinson 1972).
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
32
this zone the rainfall is very high while evaporation losses are low due to an almost permanent cloud cover.
A great amount of water from this zone flows underground, directly to the savanna plains.
7.2 Ecosystems, biodiversity and land tenure on Mount Kilimanjaro
The Kilimanjaro Region consists of six administrative districts: Moshi Rural (1,713 km2), Moshi
Urban (58 km2), Hai (2,111 km2), Rombo (1,442 km2), Same (5,186 km2) and Mwanga (2,698 km2) of
which only the four former are immediately adjacent to Mt. Kilimanjaro. Regional headquarter is Moshi.
Most of the forest is part of the Mt. Kilimanjaro Forest Reserve (107,828 ha). The upper areas of Mt.
Kilimanjaro that lie above the 2,700 meters contour fall within Kilimanjaro National Park with 75,575 ha.
Mt. Kilimanjaro has a rich diversity of ecosystems, particularly of vegetation types that result mainly from
a large range in altitude and rainfall (summarized in Table 5). Due to the high diversity of its ecosystems,
Mt. Kilimanjaro is also very rich in fauna and flora, including about 2200 vascular plant species and 140
mammals. Details on the fauna and flora on Mt. Kilimanjaro are summarized in Box 3 and 4 respectively.
Kilimanjaro is one of the main agricultural regions of Tanzania contributing approximately 30%
of the country’s high quality Arabica coffee in 1985/1986 (O’KTING´ATI & KESSY 1991). In addition to
coffee the other cash crops are sugar cane, sisal, pyrethrum and cotton. Mt. Kilimanjaro is also important in
terms of food crops such as bananas, beans, rice and millet. Most of this activity on the southern and
(north) eastern slope of Mt. Kilimanjaro is performed by smallholders of the Chagga tribe, who use the
vegetation zones in various ways (see Table 6), depending on the climatic conditions (cp. HEMP et al.
1999). On the southern slopes of the mountain, the area below the montane forest was traditionally divided
into two zones. The upper part, the highland area of the irrigated banana belt in the submontane zone
(“kihamba” land), was permanently cultivated and inhabited by the Chagga for reasons of suitable climate
and defense against the Masai. The lower part, the “shamba” land of the colline savanna zone was
cultivated seasonally and provided annual crops like maize, beans and finger millet as well as fodder for
cattle.
Box 3. Flora of Mount Kilimanjaro
About 2,200 vascular plant species occur on Mt. Kilimanjaro (HEMP, unpub. data). These are 22% of the
approximately 10,000 vascular plant species of Tanzania (BRENAN 1978). Dissecting diversity into different types of
habitats or formations, the forest belt is the most important habitat in terms of species diversity on Mt. Kilimanjaro.
Nearly 900 species occur in the forests of Kilimanjaro, representing roughly 45% of the whole vascular flora (HEMP,
unpub. data). Besides the richness in epiphytes another prominent feature of the forests of Mt. Kilimanjaro is the
wealth of ferns, especially on the southern slope, due to the high humidity. 145 taxa of pteridophytes, constituting
roughly 35% of the pteridophyte flora of Tanzania, occur on the mountain, most of them (over 90%) in the forests
(HEMP 2001 a, b, 2002). The number of vascular plants capable of enduring the harsh climate conditions in the alpine
zone is rather small – together with the ericaceous subalpine zone Kilimanjaro harbours in its alpine belt only 350
species of vascular plants (HEMP, unpub. data), 13 of which are endemic to Kilimanjaro (HEDBERG 1961). About 600
species of bryophytes (of which 415 are mosses and 185 are liverworts) and approximately 120 lichens occur on Mt.
Kilimanjaro. 12 bryophytes are strict endemics. The richest belt for bryophytes is between 2100-4100 m (PÓCS 1991).
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
33
Table 5. Altitudinal zones and main vegetation units at Mount Kilimanjaro
Altitude
(meters) Main vegetation type
Altitudinal zone
according to Hemp (2001
a)
4400
Cushion vegetation (Helichrysum) 11
lower
Alpine
3800
Erica shrubland, Helichrysum cushion vegetation 10
upper
Erica shrubland, Erica excelsa forest, Hagenia-
Rapanea forest 9
middle
2800
Erica excelsa forest, Podocarpus forest, moorland 8
lower
Subalpine
2700
Podocarpus-Ocotea forest, Erica excelsa forest 7
upper
Ocotea-Podocarpus forest 6
middle
Agauria-Ocotea forest, Cassipourea forest 5
1600
Agauria-Ocotea forest, coffee-banana plantations,
Bulbostylis meadows 4
lower
Montane
1500
Coffee-banana plantations, Croton-Olea forest,
Hyparrhenia meadows 3
upper
900
Coffee-banana plantations, savanna bushland and
grassland, agriculture, pasture 2
lower
Submontane
800
700
Savanna bushland and grassland, agriculture,
pasture 1 Colline
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
34
Box 4. Fauna of Mount Kilimanjaro
GRIMSHAW et al. (1995) recorded about 140 species of mammals for Mt. Kilimanjaro, a number far exceeding the
diversity known for Mt. Kenya (GATHAARA 1999). Among them, 87 species are regarded as being pure forest species.
Black Rhinoceros is now extinct in the area, as possibly are reedbuck and klipspringer. Twenty four antelope species
are recorded in the area, as well as 25 species of carnivores and 7 species of primates. The forest is home to the
largest known population of Abbot’s duiker, which is globally threatened. There are also 25 species of bats
(Chiroptera).
SJÖSTEDT (1909) listed 405 bird species in his expedition report for Mt. Meru and Mt. Kilimanjaro, while GRIMSHAW
(1996) gives a number of 179 highland bird species inhabiting Mt. Kilimanjaro. In an ethno-zoological study, 82 bird
species were recorded on the southern slopes in the area of the Chagga home gardens, mostly from an altitude of
1400 m (Hemp et al. 1999) reflecting the high diversity of bird habitats. 4 bird species which are globally threatened
occur on Kilimanjaro. These are Lesser Kestrel, the Taita Falcon, the Corncrake and Abbot´s Starling. The
Madagascar Pond-Heron and the Pallid Harrier are near threatened species
418 reptile species are recorded for East Africa of which 302 are listed for Tanzania. The habitat range of 88
reptile species lies within Mt. Kilimanjaro (SPAWL et al. 2002). Thus Kilimanjaro harbours about 21% of the reptile fauna
of East Africa and 29% of Tanzania. The side-spotted dwarf gecko (Lygodactylus laterimaculatus) known only from Mt.
Kilimanjaro and the Taita Hills, and the Mt. Kilimanjaro two-horned Chameleon (Chameleo tavetanus) occurring on Mt.
Kilimanjaro and Mt. Meru, the adjacent North and South Pare Mts., and the Chyulu and Taita Hills in Kenya are locally
restricted species.
SJÖSTEDT recorded 1,310 species of beetles (Coleoptera), 594 Hymenoptera, 447 bugs and allies (Hemiptera),
and 537 butterflies and moths (Lepidoptera) species for the area including Mt. Meru, but with a main focus on Mt.
Kilimanjaro. The insect materials collected highlight the diversity of Mt. Kilimanjaro and the large number of endemic
species: 47 of the 107 known Homoptera species were endemic to the mountain, as well as 27 of the 57 recorded
Darkling beetles (Tenebrionidae). A high rate of endemism was also recorded for the Rove beetles (Staphylinidae,
39% endemism), the Scarab beetles (Scarabaeidae, 25% endemism) and the long-horned beetles (Cerambycidae,
36% endemism in the mountain among all species known in East Africa) (FORCHHAMMER & BREUNING 1986; HEMP &
WINTER 1999; HEMP, C., 2001). Grasshoppers and locusts (Saltatoria) have been well studied on Mt. Kilimanjaro; 140
species of Acridoidea have been collected around the mountain in the past 10 years (HEMP & HEMP, in press), which
represent 33% of the species found in entire Tanzania according to a list published by JOHNSEN & FORCHHAMMER
(1975). Together with the Ensifera, about 190 species of Saltatoria are recorded on the mountain, of which 12 species
are only known from Mt. Kilimanjaro localities (HEMP, C., in press), and three species are still un-described,
representing 8% endemism in this insect group.
439 species of Odonata are reported for East Africa of which 171 occur in Tanzania (CLAUSNITZER 2001).
Nevertheless, the number of dragonflies recorded for Tanzania is constantly growing with every field survey due to the
very poor original data base. There are 16 species restricted to Tanzania which means a share of 9% endemism of
dragonflies for Tanzania. Mt. Kilimanjaro alone harbours 85 species (20%), among them are 14 species typical for
montane areas (17%) (CLAUSNITZER, pers. comm.). In comparison to other montane habitats of volcanic origin in East
Africa, Kilimanjaro, though being the youngest, shows an unusual high diversity due to Eastern Arc species, which
reached Kilimanjaro via the adjacent North Pare Mountains. Thus, this particular insect group exemplary reflects the
high diversity of habitats on Kilimanjaro.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
35
Table 6. Land-use in the different vegetation zones of Mount Kilimanjaro
Altitude
(meters) Land-use Altitudinal zone
3200
2700
• South eastern slopes: forest border with tussock grasses and giant lobelias
are fringed by so-called moorland zones into Erica bushland at steeper
slopes
• Tussock grassland, although already situated in the Kilimanjaro National
Park (KINAPA), is in some areas cut by fodder collectors
• Bee hives were seen up to over 3000 m with bees sucking on Erica. Except
for the tourist climbing routes the afro-alpine zone of the National Park is
mostly undisturbed by direct human impact
Subalpine zone
1700
• Southern and eastern slope: half-mile forest strip ranges between the
plantation belt and the forest reserve; provides timber and firewood (mostly
pines, cypress and eucalyptus)
• Meadows reach far into the montane forest, especially along the rivers
• Forest strip grades into natural montane forest, which should be excluded as
“forest reserve” from any usage. Nevertheless, since the 1950s about 12%
of the forest was changed into cypress and pine plantations
• Northern, north eastern and western slope: large forest plantations
• Honey collectors also frequent the montane forest zone
• Special type of land use: Shamba (Taungay) system practices (allowing
local farmers to inter-crop annual agricultural crops – mainly potatoes,
carrots and cabbage – with tree seedlings in forest plantation areas until the
third year of tree growth. By the third year, the young tree canopy casts too
much shade for the normal growth of agricultural crops. At this point farmers
move out and are allocated another plot, if available)
Montane forest
1000
• Most intensively cultivated by the Chagga (population density 500 person
per km2)
• Tree layer provides firewood, fodder and shadow, banana trees (in about 25
varieties)
• Network of irrigation canals
• The Chagga live among their home gardens in single dwellings, villages as
such do not exist
• Livestock like cattle, goats, sheep and pigs, sometimes even chicken, are
kept in stalls
• Bee-keeping plays an important role (Two bee species are kept: the bigger,
stinging honey-bee Apis mellifera ssp. monticola resembling the European
honey-bee, and a small stingless bee of the genus Meliponula)
Submontane coffeebanana
zone
700
• Southern foothills: most areas planted with maize and beans
• North-eastern foothills: maize, finger millet (important ingredient of local
beer), pigeon peas, groundnuts and sunflowers
• Western and north western foothill: large farms owned by big companies or
the government growing mainly wheat
• East of Moshi: rice
• South of Moshi: sugar plantations
Colline savanna
zone
7.3 Climatic trends on Mount Kilimanjaro
Over the past millennium, equatorial East Africa has witnessed a series of contrasting climate
conditions19. A drastic climatic dislocation took place during the last two decades of the 19th century,
19 A significantly drier climate than today occurred during the “Medieval Warm Period” (~AD 1000-1270)
and a relatively wet climate during the so-called “Little Ice Age” (~AD 1270-1850), which was interrupted
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
36
manifested in a drop of lake levels and in the onset of glacier recession (Hastenrath 1984, 2001,
Verschuren et al. 2000, Nicholson 2000, Nicholson & Yin 2001)20. A decrease in the annual precipitation
of the order of 150 mm with attendant albedo and change of cloudiness during the last quarter of the 19th
century constitutes also the most likely cause of the retreat of the Lewis Glacier on Mt. Kenya. In contrast,
the continuation of ice retreat beyond the early decades of the 20th century – as is the case for Kilimanjaro
– has been favored by a warming trend (Kruss 1983). Further, weather patterns on the Kilimanjaro are
intricately linked to landscape characteristics (e. g. Altmann et al. 2002)21. During the past few decades
vast savanna woodlands have increasingly been turned to agricultural use and thousands of hectares of
forest cover on the mountain have been destroyed by logging and burning. Whether such reciprocal effects
caused by (mostly man-made) landscape changes or whether climatic changes are of higher influence on
the Kilimanjaro remains an open question.
The most striking and most easily recognizable evidence for a steady change in regional climatic
conditions on Mt. Kilimanjaro, directly influencing landscape characteristics, are the vanishing glaciers. As
there are no signs of an increasing volcanic activity on a major scale this phenomenon has to be linked to
climatic conditions. Also, the fact that such glacier retreat is coincident with similar patterns elsewhere
around the globe leads to the assumption that their causes are also of a global character (Kaser 1999). In
contrast to this direct climatic impact, there are other even larger landscape changes, which are linked
indirectly to changing weather conditions. During the last century not only were the glaciers melting
rapidly, but there was also a significant increase in number and intensity of wild fires on Mt. Kilimanjaro,
which are most likely caused by the same climatic changes and which are simultaneously enhanced by
human influence. Changing weather patterns influence not only landscape characteristics but also animal
distributions (cp. Altmann et al. 2002). A changing migration behavior and population dynamic of big
game has been observed in the forests of Mt. Kilimanjaro.
Analysis of proxy data reveals that annual precipitation decreased by 150 mm, this means a lapse
rate of 7.5 mm/year between about 1880 and 1900. Since 1935 there are actual daily rainfall records from
the Lyamungu Coffee Research Institute which is located at an altitude of 1200 meters in the submontane
cultivated zone on the southern slope of Mt. Kilimanjaro. The annualized values are shown in Figure 10.
by three episodes of several decades of persistent aridity more severe than any recorded drought of the
twentieth century (Verschuren et al. 2000).
20 This glacier recession was caused by enhanced solar radiation due to diminished cloud cover which
accompanied the reduced precipitation. The drastic drop of the water level of Lake Victoria from around
1880 to the turn of the century was caused by a reduction in annual precipitation of about 150-200 mm
(Hastenrath 1984). These data are apparently indicative of an important precipitation reduction throughout
an area exceeding East Africa (cp. the variations of the water level of Lake Chad (Street-Perrott & Perrott
1990), where severe droughts started from the year 1900), followed by little secular precipitation variation.
21 The role of temperature and rainfall in shaping the landscape has long been recognized. More recently both
empirical evidence and mathematical models have highlighted the reciprocal impact of landscape changes
on weather patterns.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
37
Figure 10. Annual precipitation at the Lyamungu Coffee Research Institute
Source: Lyamungu Coffee Research Institute
It appears that there is a decrease in precipitation since 1935 of about 11% or 177 mm (equivalent
to 2.6 mm/year) at Lyamungo or a lapse rate of 2.6 mm per year. If this rate is extrapolated back to the year
1900 this would mean an annual loss of over 400 mm compared with the situation before 1880.
This records from Lyamungu are consistent with a general reduction in rainfall throughout most
of Africa since 1950 (Nicholson 2000) and in the area of Kilimanjaro according to the maps presented by
Hay et al. (2002) for the time interval 1941-1995 between 1941-1960 and 1971-1995. In addition to the
decline in annual precipitation, the Lyamungo data also reveal that the number of dry months with less
than 30 mm increased, whereas wet months with more than 125 mm were stable. With regard to
temperature, the maps presented by Hay et al. (2002) indicate that spatially averaged temperatures in the
area of Kilimanjaro rose between 1951-1960, were stable or decreased slightly between 1960-1981, and
increased again between 1981-1995. While no time-series exists for a particular location on the mountain,
there is however a 25-year temperature record (from 1976) from the Amboseli region just to the north
(Altmann et al. 2002). This record shows daily temperatures increased dramatically throughout the same
25-year period. Mean daily maximum temperatures increased with a rate of 0.275 °C per annum, with
increases being greatest during the hottest months of February and March.
To summarize, available climate records reveal a declining trend in precipitation on the
Kilimanjaro at least since 1880. Although available data is not sufficient to infer temperature trends at
different altitudes on the mountain, a distinct overall warming trend has been observed for most of the
period since 1950 to present. Observations from neighboring Amboseli in fact indicate a local warming
rate of 0.275 °C per decade between 1976-2000, significantly higher than globally averaged warming.
Either of these trends – declining precipitation or increasing temperature – contribute to enhanced
glacier melting, as well as to enhanced fire risk22. Consistent with the pronounced decrease of precipitation
22 Decreased precipitation reduces cloud cover and therefore enhances the sunlight reaching the glacier,
causing it to melt faster. The effect of increased temperatures on glacier melting is self evident. With
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
38
at the end of the 19th century fires in the areas of the subalpine forests were documented by the first
Europeans on the mountain (Meyer 1890, Volkens 1897, Jaeger 1909). At the same time the glaciers
started to recede (Kaser et al. under review) – driven by such changes in precipitation23. In the following
decades the climatic situation was more stable while the glaciers changed more slowly (Kaser 1999).
Enhanced glacier melt and fire risk have both been empirically observed in recent decades. These effects
are consistent with the simultaneity of precipitation decline and temperature increase24 which has been
observed during the same time period.
7.4 Potential impacts of climatic changes: glacier retreat
The ice cap on the Kilimanjaro has been in a general state of retreat since the end of the Little Ice
Age around 1850. This retreat was driven by natural climatic shifts (particularly a decline in regional
precipitation), but appears to have accelerated due to the warming observed in the second half of the 20th
century. Later in 1976 the glaciers covered 4.2 km2 (Hastenrath & Greischar 1997) compared with only 2.6
km2 in 2000 (Thompson et al. 2002). Measurements taken in 2000-2001 on Kilimanjaro show that its
glaciers are not only retreating but also rapidly thinning (Thompson et al. 2002). Figure 11 shows the
diminishing extent of the glaciers on Kibo between 1962 and 2000. Over these 38 years, Kilimanjaro has
lost approximately 55 % of its glaciers. There is general consensus that the ice cap of Kilimanjaro will
have disappeared by the year 2020 for the first time in the surveyed period of over 11,000 years.
Figure 11. Development of the Kilimanjaro (Kibo) ice fields from 1912 to 2000
Source: Thompson et al. 2002
The symbolism of this loss notwithstanding, it is important to note that the impact of the
disappearance of the ice cap on the natural and human systems would be very limited. The present glaciers
of Kibo cover an area equivalent to 0.2% of the area covered by the forest belt on Mount Kilimanjaro.
regard to forests, drier and hotter conditions both contribute to enhanced inflammability of the forest,
thereby enhancing fire risk.
23 The causes of such changes in precipitation are likely natural and not linked to climate change.
24 The warming in recent decades is consistent with climate change.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
39
Only two rivers are directly linked by very small streams to the glaciers, while 90% of the precipitation is
tapped by the forest belt. Even if the glaciers have melted till 2020 there will still be precipitation on Kibo
feeding springs and rivers, although not so continuously and to a much lesser degree.
Therefore, contrary to the opinion expressed by Thompson et al. (2002) it is very unlikely that the
loss of the glaciers will have a major impact on the hydrology of the mountain. Kaser et al. (under review)
come to the same conclusion. Further, observations of dry river beds are not necessarily an indicator of
long term climatic changes or the impact of shrinking glaciers. Dried out rivers in some areas are much
more likely the result of forest destruction or of increasing water demands of the rapidly growing
population. Water diversion has in fact quadrupled in certain areas during the last 40 years (Sarmett &
Faraji 1991).
Today Kilimanjaro National Park (KINAPA) is a major tourist attraction in Tanzania and gains
the most foreign exchange of any National Park in Tanzania (Newmark & Nguye 1991). Most visitors are
mainly interested in reaching the summit of Kibo, known as Uhuru Peak, the highest point in Africa. Since
the establishment of the Park in 1972, the number of visitors of KINAPA has multiplied by five. Without
any doubt Mt. Kilimanjaro will lose part of its beauty with the inevitable loss of its glaciers. However, it
will still remain the highest mountain in Africa – and incentive enough to climb. Therefore, a decline in
tourist numbers is unlikely.
7.5 Potential impacts of climatic changes: enhancement of fire risk
A less publicized and possibly far more significant impact of climate change on Mount
Kilimanjaro is the intensification of fire risk and its attendant impacts on biodiversity as well as ecosystem
services. In theory rising temperatures should result in the upward migration of vegetation zones, as
observed in the Alps by GRABHERR & PAULI (1994). This effect however has been offset by the
intensification of fire risk as a result of warmer temperatures and declining precipitation. This risk is
particularly acute in the vast ericaceous belt in the upper reaches of the vegetation. Consequently, climatic
changes have actually pushed the upper forest line downward on the Kilimanjaro.
On Mt. Kilimanjaro fires are common in the colline savanna zone, in the (sub-) alpine zone and –
to a lesser degree – in the submontane and lower montane forest zone, whereas in the middle montane
forest zone – at least on the southern slope – fires are rare. Most of these fires are lit by man (often as a
maintenance tool), especially in the cultivated areas on the lower reaches of the mountain. The situation
however is different in the upper regions of the mountain where no grazing or agriculture exists above the
forest belt and logging in the upper forest zone is also rare25. Since climate change is the objective of this
analysis, man lit fires are of minor interest. Therefore the destructive role of fires in the forests and in the
alpine zone where climatic conditions play a more critical role are the focus of this discussion.
7.5.1 Elevation distribution of species richness and its relationship to fire
Fire variously influences species diversity, composition and vegetation structure in the different
altitudinal zones on Kilimanjaro (Hemp, in press). Figure 12 shows the species numbers of vascular plants
25 Although even these remote areas are not free from human influence, as population on the foothills has
increased enormously. Since 1895 population has multiplied by 20. As a result an increasing human
activity can be seen in all altitudinal zones and areas, promoted in particular by tourism. Since the
establishment of the park in 1972, the number of visitors of KINAPA has multiplied by five. Together with
porters, guides and tourists about 100.000 people visit the upper regions of Kilimanjaro per year. Such
increasing numbers of visitors have certainly effects on the environment. Thus, the (natural) impacts of the
changing climatic conditions are additionally enhanced by human influence.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
40
at 100 m elevation intervals between 700 and 4500 m. Vascular plants have their maximum (745 species)
in the 1300 - 1400 m interval in the area of the banana-coffee plantations of the submontane zone.
Figure 12. Absolute species numbers of ferns and of all occurring vascular plants on the southern slope
of Mount Kilimanjaro
Source: HEMP 2001a
As shown in Figure 12 species numbers are highest in moderately cultivated or disturbed areas
and not in natural, completely untouched areas. In this context the second peak at 2600 m at the lower
border of the subalpine zone is of interest. In this altitude fire starts to be an important ecological factor on
Mt. Kilimanjaro, creating a mosaic of different fire induced stages of forest, shrub and tussock grassland
communities. This high diversity in habitats - compared with the closed forest at lower altitudes and the
monotonous heath lands at higher altitudes – leads to a high diversity in species numbers. This trend is
enhanced by the occurrence of fire-tolerating species, which show the same bimodal altitudinal distribution
with a gap in the wettest central forest parts where fires are uncommon. They can therefore be regarded as
fire indicators.
7.5.2 Influence of fire on regeneration, composition and structure of forests
Forest fires are frequent in the subalpine zone and also, less frequent in the submontane and
lower montane zone between 1300 and 2000 m above sea level (asl). Fires in the submontane and lower
montane forests are mostly set by people. In these forests, fire changes species composition and structure
of the tree as well as the herb layer (HEMP, A. in press). This is of major importance for forest
regeneration, as the dense cover of bracken impedes the sprouting of trees. In the subalpine forests
between 2800-3000m asl fire causes sharp discontinuities in the floristic composition and structure26. Once
Erica excelsa has established, regeneration of a broad-leaved forest becomes more and more improbable
(HEMP & BECK 2001). If the frequency of fire becomes too high it degrades Erica excelsa forests into bush
lands in which E. excelsa is substituted by E. trimera and E. arborea. This Erica bush extends between
26 Giant heather (Erica excelsa) becomes dominant at this altitude forming dense mono-specific stands,
which border the Podocarpus and Juniperus forests without any transition (HEMP & BECK 2001).
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
41
3200 and 3900 m asl. Continuously high frequency of fires destroys this bush vegetation, ultimately
resulting in Helichrysum cushion vegetation.
7.5.3 Major impacts of fire on Mount Kilimanjaro’s ecosystem
On Mt. Kilimanjaro structure and composition of the subalpine vegetation is strongly influenced
by recurrent fires. Above 3200 m asl Erica excelsa forest is replaced today by Erica trimera and E.
arborea bush in most areas. But from field observations and historical descriptions (JAEGER 1909, KLUTE
1920) it can be assumed that the forest extended up to 3600 m in some areas of Kilimanjaro at the
beginning of the twentieth century while an open Erica forest was reported at altitudes of over 3900 m; this
is 800 m higher than today (HEMP & BECK 2001). On the south-eastern slopes at an altitude of 2800 m
Erica excelsa stands and the “moorland” tussock vegetation produce very abrupt boundaries. Tree-islands
consisting of a core of Podocarpus forest are surrounded by a fringe of Erica trees and various shrubs. In
this area, a mosaic of Podocarpus forest, Erica forest and subalpine grassland occur at the same altitude.
Substantial microclimatic differences can thus be ruled out as an explanation for this pattern. Rather,
recurrent fires may be the crucial factor pushing the forest back from the subalpine to lower and moister
regions.
The comparison of two classified Landsat images from 1976 and 2000 reveals enormous changes
in the upper vegetation zones of Mt. Kilimanjaro during the last 24 years (Figures 13 and 14)27.
Figure 13. Vegetation cover in the montane and alpine zone on Mount Kilimanjaro (1976)
27 It should also be mentioned that on Fig. 13 and 14 differences in glacier size are apparent. While in 1976
the glaciers covered 4.2 km2 (HASTENRATH & GREISCHAR 1997) in 2000 they have been shrunk to 2.6 km2
(THOMPSON et al. 2002).
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
42
Figure 14. Vegetation cover in the montane and alpine zone on Mount Kilimanjaro (2000)
In 1976, the Erica trimera bush, which today is depressed in the western and northern parts of the
mountain below 3400 m, reached up as a continuous belt to over 4100 m into an area which today is
covered by Helichrysum cushion vegetation. Erica forests covered nearly 5 times the current area (166 and
36 km2 respectively), extending in many places up to 3700 m. This equals a loss of 130 km2 or over 10%
of Kilimanjaro’s forest cover due to fire since 1976.
As discussed earlier, Erica vegetation is largely influenced and controlled by fire. The growing
influence of fire pushed down the forest line replacing Erica forests with Erica bush. Fire has also shifted
the upper border of Erica trimera bush by replacing it with Helichrysum cushion vegetation. The
Helichrysum cushion vegetation is not threatened by fire because its little biomass provides little fuel. In
addition, distances between vegetation patches and cushions are too high to allow fire to spread. When fire
reaches this vegetation zone it stops. Therefore, the upper line of this vegetation formation has been stable
for the examined 24 years. On the lower edge, however, fire was able to spread into the Erica bush zone. It
can be assumed that most of the Erica bush of the year 1976 as shown in Figure 13 has still been Erica
forest at the end of the 19th century while most of the Erica forest of 1976 was still broadleaved forest at
that respective time. This constitutes a loss of over 300 km2 of upper montane forest (or a third of the
present forest size) during the last 120 years. As a consequence, the ericaceous belt on Kilimanjaro with
the easily inflammable heathlands became larger, giving rise to more and bigger fires.
7.5.4 Socioeconomic impact of increasing fire intensity
The increase in fire intensity on the slopes of Mt. Kilimanjaro has very significant impacts on
both the natural and human systems that it sustains. The most direct impact is a significant decline in water
resources; other impacts include effects on farming and other activity, as well as a loss of biodiversity.
7.5.4.1 Water resources
The devastation of 13,000 ha of forests, mostly of Erica forest, in the upper reaches of the
Kilimanjaro since 1976 by fire has caused a serious disturbance in the water balance of the entire
mountain, given that the forest belt functions as the main water catchment area. Montane and subalpine
mossy or cloud forests are of great importance for watersheds in East Africa. They play an active and
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
43
important role in the protection of slopes against erosion by controlling the damaging effects of torrential
rainfall and regulating the outflow patterns of watercourses. In cloud forests about one third of the total
rainfall is absorbed by the dense epiphytic layer (PÓCS 1976). Destruction of these forests reduces the
function of the forest belt as a water filter and reservoir. Instead of remaining in the thick epiphytic
biomass, humus and upper soil of the forest, percolating slowly to the groundwater, rainwater flows off
quickly on the surface to the rivers eroding the soil and increasing the danger of floods on the foothills.
Another consequence of the quicker rate of rain flow is water shortage during periods without rain.
In addition to the function of filtering and storing water the upper montane and subalpine cloud
forests have a high potential of collecting cloud water (fog interception). Fog interception or fog deposition
refers in this case to the small cloud droplets that do not settle on horizontal surfaces and, thus, are not
collected in a rain gauge. Cloud water droplets are blown by the wind against the vegetation where they
coalesce to form large drops that run off and fall to the ground. Fog droplets have to be intercepted by the
vegetation and do not precipitate spontaneously (cp. CAVELIER et al. 1996, GLASOW & BOTT 1999).
Above 2000 m asl fog and mist occur nearly every day, above 2600 m asl every day. Thus, fog
interception increases with altitude, especially its relative share of water input. The amount depends on the
height and leaf area index of the vegetation providing wetting capacity for interception, the frequency of
fog, and exposure to the prevailing wind (CAVELIER & GOLDSTEIN 1989, CAVELIER et al. 1996, GLASOW
& BOTT 1999, ZIMMERMANN et al. 1999). Several studies suggest that fog can supply different amounts of
liquid water to tropical montane cloud forests. In some areas fog interception represents 99% of the water
input while in others only 3.5%. In general, fog interception is an important additional water source at sites
with regular and frequent occurrence of fog, contributing far more than one third of the bulk precipitation
in tropical montane forests (cp. e. g. CAVELIER & GOLDSTEIN 1989, JUVIK & NULLET 1993, CAVELIER et
al. 1996). In lower montane tropical rain forests an average of about 16% was measured (CAVELIER et al.
1996).
The following calculations are based on a comprehensive ecological and meteorological database
collected by a consultant to this report. A vegetation map was produced by analyzing over 1200 vegetation
plots. In addition, 16 meteorological stations along 4 transects inside the forest belt were established,
producing the first reliable weather data (rainfall, temperature, air humidity, radiation, wind speed etc.)
from this vegetation zone of Mt. Kilimanjaro. Using these data a map of mean annual rainfall and mean
annual temperature was created. According to the distribution of the different forest types, the annual
rainfall, the estimated amounts of cloud water collection and evapotranspiration (based on measured
vegetation density, altitude, climatic parameters and numbers given in literature e.g. LARCHER 1984,
CAVELLIER et al. 1997) the forest belt was divided in 11 eco-climatic zones. For the first time this
approach allows to estimate the water output of the 939 km2 of indigenous forest (excluding forest
plantations) of Mt. Kilimanjaro (Table 7).
Table 7. Hydrometrical data of the forest belt on Mount Kilimanjaro
Water Input Water Output
Rain (million m3) fog (million m3) Evapotranspiration (million
m3)
groundwater and streams
(million m3)
1,533.5 560.0 797.4 1296.1
73.3% 26.7% 38.1% 61.9%
The indigenous forests of Mt. Kilimanjaro receive 2093.5 million m3 water annually of which
73% is by rainfall and 27% by fog interception. The intercepted moisture has to be considered to be a net
gain, since the energy used in its evaporation from the leaf surfaces during fog-free periods would have
been used in transpiration of an equal amount of water from the soil (KERFOOT 1968). In contrast, half of
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
44
the amount of rainwater re-evaporates back to the air by evapo-transpiration. The circa 1,300 million m3 of
remaining water percolate into the groundwater or run off as surface flow into streams. The approximately
800 million m3 of water evapo-transpired are not lost for the ecosystem. The forest dampens the air,
leading to permanent high air humidity over the forest belt. This results in cloudiness and rain showers
even during the dry seasons. Therefore the forest stores water not only in its biomass and the forest soil,
but even in its surrounding air. This mechanism enhances the forest’s function as a water reservoir
regulating the outflow patterns of watercourses. Without such a permanent cloud cover over the forest
evapo-transpiration would be much higher (due to higher temperatures) and rain showers during the dry
season would be absent.
In his analysis of the value of East African forests in influencing climate and water supply
NICHOLSON (1936) estimates the condensing capacity of montane forests add up to at least 25% of the total
annual rainfall. This amount (in the case of the forests on Kilimanjaro equivalent to 383.4 million m3) has
to be added to the 560 million m3 water of fog interception, to get a more reliable impression about the
influence of the forest on the water balance. This gives 943.4 million m3 or a surplus of 146 million m3
water (nearly 10% of the rain water input) which forests on Kilimanjaro contribute more to the water
balance every year than comparable open areas. Table 7 further shows that fog interception is an important
factor in the hydrological balance of the mountain. About one quarter of the atmospheric water input in the
forests derives from this source. Without the cloud water collecting forests this water would be lost for the
mountain. If the surface and groundwater run-off is compared with only the “ordinary” precipitation, i. e.
rainfall, the role of forest for the water production becomes evident.28
Consequently, the loss of 13,000 ha of Erica forest since 1976 results in a water yield reduction
of about 58 million m3 of fog water annually. This number represents over 10% of the annual fog water
input of the entire forest belt or the equivalent of the annual drinking water demand of nearly three million
inhabitants on the mountain (this calculation is based on numbers given by UNITED REPUBLIC OF
TANZANIA & CES 2002). In this calculation, however, are neither the several 10,000 ha of destroyed
ericaceous bush land nor the montane forests, which have been lost due to logging activities included.
Since the Chagga with their irrigation system are highly dependent on a steady river discharge
changes to the water balance present a serious threat to their existence. During the dry seasons water
shortages especially on the lower foothills become increasingly common. Women and children have to
spend a big part of the day fetching water. Yet, the water demand grows rapidly. The hydrometric report of
the Hai district water supply Phase IV (UNITED REPUBLIC OF TANZANIA & CES 2002) referring to an
selected area on the south western, western and northern parts of the mountain presents the following
numbers: Currently, population in this area totals 132,258 inhabitants with a daily demand of 7,200 m³
water and it is expected to rise until 2015 to 162,570 inhabitants demanding daily about 8,900 m³ water.
Besides, the situation on Mt. Kilimanjaro affects the entire region. The Pangani River, one of
Tanzania’s largest rivers, provides water to the hydropower plants of Nyumba ya Mungu (8 MW), Hale
(17 MW) and Pangani Falls (66 MW), which generate some 20% of Tanzania’s total electricity output. A
water shortage during the dry periods would increase the number of power cuts which have already inhibit
economic prosperity. Fishing in Nyumba ya Mungu dam yields a maximum catch of approximately 4,000
tonnes annually. The river also supplies the large scale South-East Moshi rice scheme. Furthermore, the
28 Although in general water output by run off is lower than the input caused by rains, this relation varies
within the 11 distinguished eco-climatic zones. In the relatively dry submontane Croton-Calodendrum
forests the run-off counts only to 40% of the rainwater input. In higher altitudes with lower evapotranspiration
but higher fog interception this ratio turns: In the Juniperus forests above 2600 m the total
water run-off is 110% of the rainfall input. This is due to the additional water available from fog
deposition. In the Podocarpus, Hagenia and Erica forests this ratio lies even higher at 120%.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
45
southern slopes provide water to Arusha Chini sugarcane plantation. In Kenya, the Amboseli ecosystem
including the wetlands of Ol Tukai and Kimana, which support Masai pastoralists and an abundance of
wildlife, depend on the Kilimanjaro water supplies.
7.5.4.2 Other ecosystem services diminished by fire
Forest fires do not only reduce the water budget of the mountain, but they also directly and
indirectly destroy other goods and benefits. Forest fires burn huge amounts of precious wood including fire
wood, which people are allowed to collect, and timber, which people cut illegally. Besides, fires reduce the
beauty of the heathlands that attract tourists and destroy the flower trees for bees. Bee-keeping is important
on Mt. Kilimanjaro. An ethnobotanical study (HEMP 1999) showed that the Chagga make use of their plant
environment in a variety of ways. The plants serve as forage for households and agricultural purposes, and
many are used in medicinal applications either as drugs or for “magic” purposes. The montane forest is
home to many of such plants. In addition, repeated burning also modifies the nutrient balance of soils
(CRUTZEN & ANDREAE 1990).
7.6 Other threats to the Mount Kilimanjaro ecosystem
The climate related threats to the Kilimanjaro ecosystem need to be viewed in conjunction with
other stresses stemming from human activities as well as changed migration behavior and population
dynamics of big game. The results of a 2001 aerial survey (LAMBRECHTS et al. 2002) and the examination
ground data revealed that the forests of Mt. Kilimanjaro are heavily impacted by illegal logging of
indigenous trees in most areas below 2,500 metres on the western, southern and eastern slopes, and by the
establishment of forest villages in the western and northern slopes. Logging activities affect the entire
broadleaved mixed forests below an altitude of 2,500 metres on the southern slopes of Mt. Kilimanjaro.
The moist Ocotea forests which cover most of the southern slopes are subject to serious destruction due to
intensive illegal logging of camphor trees.29
In addition, large tracts of indigenous forests on the north-western and northern slopes have been
converted into forest plantation, using fast growing exotic tree species, such as pine and cypress. On the
north western slopes, the expansion of the forest plantations has reduced the indigenous forest belt to a
width of less than one kilometer. The majority of the clear felled compartments within the forest
plantations have not been replanted as required by the normal rotation management. To summarize, the
aerial survey revealed that the forest belt is threatened on its upper and lower border, thus shrinking on
both sides. This further exacerbates the adverse impacts on the water balance of the mountain.
Changing climate patterns not only influence landscape characteristics but also animal
distributions. The Kenyan Amboseli National Park is situated on the northern foothills of Mt. Kilimanjaro.
This area has experienced extensive habitat changes since the early 1960`s (ALTMANN 2002). These
include dramatic loss of tree and shrub cover which was partly caused by an increasing elephant population
and temperature changes. The “natural” landscape alterations are further enhanced by a steadily growing
Masai population on the whole northern foothill of Mt. Kilimanjaro. According to rangers of Kilimanjaro
29 During the survey, over 2,100 recently-logged camphor trees were counted. On the lower slopes bordering
the half-mile forest strip, there was no recent logging of camphor trees since these areas have already been
depleted. However, other indigenous tree species were targeted; some 4,300 recently-logged indigenous
trees were recorded. As a result, evidence of 57 landslides in the heavily impacted Ocotea forests was
recorded. To the east, above Marangu, 19 cleared fields have been opened up in the forest, and a large
number of livestock was seen up to 8 kilometers deep into the forest. There were fewer observations
recorded in the half-mile forest strip because this zone is virtually denuded of indigenous trees. Some areas
have been completely cleared. Logging activities also impact heavily the east and west sides of the
northern slopes; 574 recently-logged cedar trees were counted, as well as over 800 other indigenous trees.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
46
National Park elephant migration from the Amboseli National Park through the so-called “Kitendeni
corridor” into the forests of Kilimanjaro has increased. In addition, more elephant herds stay permanently
inside Kilimanjaro’s forests given the better conditions compared with the Amboseli basin.
A ground survey of the forests on the western and northern slopes of Kilimanjaro reveals that in
most places elephants and buffaloes are abundant. Besides former logging activities (the last sawmills
inside the indigenous forest were closed in the 1970´s) grazing patterns of big game cause a change in the
dense forest cover towards a mosaic of openings and patches of closed canopies. If the openings become
larger, forest regeneration is impeded. In the long term, this development will destroy the forest and change
it into a bush land with scattered trees with all the known disadvantages.
7.7 Scenarios for 2020 with respect to fire impact
Assuming that the observed trends in fire frequency continue in a linear mode the following
scenarios are probable. Regarding the upper forest line, most of the remaining subalpine Erica forests will
have disappeared within five years. As a result, Mt. Kilimanjaro will have lost its most effective water
catchment area. Compared with the situation of 2000, this means an annual loss of 16.2 million m3 fog
water. Subsequently, the upper forest line will retreat more slowly because on the one side mostly broadleaved
forests remain, which are to a much lesser degree inflammable and because on the other the lower
areas receive an increasing amount of precipitation. Nevertheless, an average retreat of the upper forest line
of about 100 m in altitude seems to be probable by 2020, when the glaciers will have melted. Forest
regeneration will completely be inhibited and regressive succession will prevail, as illustrated in Figure 15,
substituting increasing areas of Erica heathland with low layered Helichrysum cushion vegetation.
Figure 15. Forest succession after continued fires
Linear increasing temperature and decreasing precipitation in combination with increasing
logging activity will also result in more forest fires, which will heavily destroy the lower forest zone up to
an average altitude of 2000 m (for example closed forests will be replaced with an open bush that cannot
carry out the necessary ecological functions). These trends will cause a further shrinking and fragmentation
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
47
of the forest belt. Especially on the western and eastern but also on the south eastern slopes the forest zone
will be interrupted by large gaps with all the known disadvantages for wildlife and the ecological balance.
7.8 Climate risks in perspective: shrinking glaciers versus enhanced fire risk
With an average thickness of 30 m as indicated by ice core drilling (THOMPSON 2000,
THOMPSON et al. 2002) and likewise observations from KASER et al. (under review) the existing 2.6 km2 of
glaciers constitute a water volume of about 72 million m3. However, most of this water is not available for
the lowlands since most glacier ablation occurs as sublimation and the remaining melting water evaporates
immediately into the atmosphere (KASER et al. under review). If one quarter (or 18 million m3) of glacier
water would percolate into the rivers, an average annual water output of about 0.9 million m3 would result
until 2020, when the glaciers are predicted to have being melted. But even then, one can still expect
precipitation on Kibo to feed springs and rivers although not so continuously and to a lesser degree.
In contrast, Mt. Kilimanjaro receives 58.5 million m3 less water each year due to forest depletion
and vegetation changes incurred as a result of forest fires since 1976. The number is likely an
underestimate since the calculation assumes the timberline to remain stable for the next 20 years, which is
very unlikely. Moreover, the calculation did not include the several 10,000 ha of ericaceous bush land
which has been substituted by low Helichrysum cushion vegetation.
Summarizing, compared with roughly 1.3 billion m3 of water, contributed every year by the 1000
km2 of indigenous forest, the consequences of losing 2.6 km2 of glaciers providing an annual water output
of about 0.9 million m3, the loss of Mt. Kilimanjaro’s ice cap is negligible. Still, the melting glaciers are
certainly an alarming indicator of severe environmental changes on Mt. Kilimanjaro.
8. Policy responses for Mount Kilimanjaro
The preceding section has laid out the complex interaction between climatic and other stresses
that are causing significant changes in the Kilimanjaro ecosystem and adversely impacting the ecosystem
services it provides. While the most visible impact – glacier retreat – may only have limited consequences,
enhancement of fire risk that has resulted from climatic trends and human interference poses significant
threats not only to the viability of the ecosystem, but also neighboring regions through its critical influence
on regional water resources. Some of these changes (such as glacier retreat) may be inevitable, but others
can be managed to make the ecosystem more sustainable. However, this requires a comprehensive set of
policy responses that take into account the underlying demographic, environmental and climatic stresses.
This section starts with a brief discussion of policy responses to the shrinking ice cap, to the general
environmental threats facing the Kilimanjaro ecosystem, as well as to the enhancement of fire risk. Finally,
given that human livelihood choices might provide the trigger for forest fires, the section reviews alternate
livelihood strategies that might alleviate some of these stresses.
8.1 Policy responses to the shrinking ice cap
The melting of Kilimanjaro’s ice cap receives much attention. Articles in local, regional and
international newspapers described the results of ice core drilling by American scientists during the years
2000-2001. A study on climatic changes in the context of the receding ice level on Mt. Kilimanjaro was
chosen to be a topic among the proposed research priorities for the next years of TANAPA (Tanzanian
National Parks). However, there is obviously nothing that could be done by way of policy responses to
avoid or even delay its eventual loss.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
48
8.2 Policy responses to general environmental threats
The vulnerability of the Kilimanjaro to climate change can be alleviated at least partially by
reducing other environmental stresses on it. Since Mt. Kilimanjaro is a UNESCO World Natural Heritage
Site, the environmental problems of this unique volcano have attracted international attention and a number
of conservation projects are already under way. The United Nations Development Programme (UNDP) and
the United Nations Foundation (UNF) have jointly disbursed 264,000 US$ to the Tanzanian government
for running different environmental conservation projects and promoting eco-tourism on Mt. Kilimanjaro.
A first comprehensive inventory of threats - including wild fires – to Mt. Kilimanjaro was taken during an
aerial survey in September 2001.30 As a result of this survey it was decided by the Ministry of Natural
Resources and Tourism that the forest belt of Mt. Kilimanjaro and Mt. Meru will be taken away from the
Forest Department and included into Kilimanjaro National Park and Meru National park respectively. A
similar shift in management from Forest Department to Kenya Wildlife Service in 2000 on Mt. Kenya
resulting from an aerial survey had dramatic consequences. The illegal cutting dropped drastically.
Comparing the situation in 2002 with 1999 logging of camphor was reduced by 96%, logging of cedar by
73% and logging of other indigenous trees by 92% (LAMBRECHTS, pers. com.). It is therefore expected that
such a shift in management in the Kilimanjaro will have similar effects.
While these efforts are underway, several important challenges remain. One major threat is the
cross-border migration of big game from the Amboseli National Park. There is a need for the formulation
of a cross-border response between Tanzania and Kenya. An initial step would be to survey and count the
numbers of elephants and buffalos. Based on these figures further steps, including control or reduction
measures have to be taken into account. Also, the animals should be provided with adequate areas in the
Amboseli basin by restricting permanent settlements of the Masai.
8.3 Policy responses to enhanced fire risk
There are two general ways to cope with wild fires: first, reduction of fire risk and second,
fighting of fires. The main area of interest in this respect is the upper montane and subalpine zone on Mt.
Kilimanjaro, where fires are most common. The first aim is to protect the still existing upper montane and
subalpine forests from further destruction by fire. Second, since the potential climatic and historic tree line
is much higher than the actual fire-induced one it should be tried to increase the forest area and to push up
the actual forest line to areas that were formerly covered by forests.
8.3.1 Responses to forest destruction
While the natural montane forest on the Kilimanjaro has had protected status since the early
twentieth century, the cutting of indigenous trees continued to increase until 1984 when the severe forest
destruction led to the banning of all harvesting from the catchment forests on Kilimanjaro by a Presidential
Order. Prior to the ban local people were used to entering the reserve without restriction to utilize its
resources. Therefore, the new restrictions were not effective and encroachment activities have continued
illegally. Nevertheless, general awareness for the protection of Mt. Kilimanjaro’s natural resources
especially of its forests is high among local people, governmental and non-governmental institutions.
Everywhere Panda miti! (Plant trees!) stickers can be seen in offices, governmental cars and schools. The
government awards prizes to those villages that have planted the most trees. Unfortunately, no indigenous
30 The request for the aerial survey of the forests of Mt. Kilimanjaro was originally presented by UNDP/GEF
Small Grants Programme, New York. The objective was to identify the type, extent and location of the
threats to the forests and provide a baseline assessment for the newly developed Community Management
of Protected Areas Conservation Project (COMPACT). The aerial survey on Mt. Kilimanjaro was
supported by the Ministry of Natural Resources and Tourism and the Tanzania National Parks.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
49
trees are used for such competitions. The churches, which have great influence on people, also support
afforestation measures.
Many NGO’s like the Tanzania Association of Foresters (TAF) run reforestation projects on the
mountain. Some villages like Mbokomu and private institutions like the Maua Seminary do the same,
mostly without any support from the forest department in Moshi and sometimes even against the
authorities31. Another significant development was the initiation of the catchment forestry project in 1988.
The first phase of the project (1988-1992) focused on improving catchment forest management by
establishing an inventory of the forest by resurveying, replanting boundaries, mapping and through
reviving the management and protection activity. The second phase (1992-1996) tried to improve
management of the forest through boundary marking, mapping, policing and people’s participation. Efforts
are currently being made to involve the local communities in the management of the forest reserve.
Villages adjacent to the forest have now the responsibility to watch that there is no encroachment into the
forest. Village conservation committees are responsible for establishing tree nurseries, to organize patrols
into the forest, to mobilize the people for fire fighting and to control the entrance into the forest by issuing
permits. Timber that has been confiscated during the patrols becomes (partly) property of the village
(MISANA 1999). However, these activities of the forest department were not very successful, which
became apparent during the aerial survey of LAMBRECHTS et al. (2002).
In 2000, the GEF Small Grants Program implemented by UNDP, in collaboration with the United
Nations Foundation (UNF), launched the Community Management of Protected Areas Conservation
Project (COMPACT). The main objective of COMPACT is to demonstrate, by complementing and adding
value to existing conservation programs, how community-based initiatives can significantly increase the
effectiveness of biodiversity conservation in and around World Natural Heritage Sites (WNHS). The
project also aims at (i) enhancing the capacities of local organizations and NGOs whose existence and
future prospects are closely linked to these protected areas; (ii) increasing local awareness of, and concern
for, the protection of WNHS, (iii) promoting and supporting communication and cooperation among park
management personnel and other concerned groups, particularly local communities, (iv) increasing general
understanding of the synergies between community development and the role of globally significant
protected areas in contributing to sustainable development, and (v) drawing lessons from project
experience that can be shared widely at local, national and international levels.
Mount Kilimanjaro is one of six World Natural Heritage Sites on three continents participating in
COMPACT. A common methodology to prioritize COMPACT interventions at the six sites has been
developed. It involves a participatory approach to identify the main threats to the protected area, and to
assess the types of activities that may be carried out by local communities to address those threats while
improving their quality of life and livelihoods. This planning process involves a wide range of stakeholders
of Mt. Kilimanjaro: community-based organizations, local and national NGOs, local and national
authorities with management responsibilities of the mountain, and other programs and projects present in
the area. It is too early to assess the effect of COMPACT, although the expectation of the funding agencies
31 The Maua Seminary, a Franziscanian monastery leading a vocational school, is a good example for
possibilities and problems of private engagement in environmental projects. The Padres of this monastery
are very active replanting the whole valley of the Mue river inside the half-mile forest strip. The trees,
although paid and planted by private effort are the property of the government. To cut down expenses and
to get local people involved and interested in the project, they tried to use the Shamba (Taungya) system
practices as it is done in other forest plantations on the mountain. Many forest plantations in West and
North Kilimanjaro have usually been established by allowing local farmers to inter-crop annual agricultural
crops with tree seedlings in forest plantation areas until the third year of tree growth. But the Padres have
been unsuccessful fighting for eight years to get the permission from the forest office to use the Shamba
system. Since they cannot pay a lot, incentives for local people to cooperate are not very high and the
afforestation of the valley takes long.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
50
and the host government is that the project empowers local communities to participate effectively in
reversing extractive pressures that have adverse impacts on the mountain’s resources.
8.3.2 Responses to forest fires
Since most areas heavily affected by fire - Erica forests and bush lands, are located inside the
Kilimanjaro National Park (KINAPA), effective management of this park is one of the keystones to reduce
the fire risk on Mt. Kilimanjaro. During the 1997 fire outbreak, a contingent of 700 fire fighters including
the Tanzanian army was needed to extinguish the fire. A special fund of US$5,000 was subsequently set up
to fight fires on the mountain. Fundraising to collect the money has targeted various donors including the
business community, environmental institutions and other interested parties. However, since large amounts
are necessary, the Tanzanian government is seeking new donor funding to conserve forests on Mt.
Kilimanjaro (THE EAST AFRICAN, 9.10.2002).
One step towards fire prevention was already taken by the national park authorities by banning
camp fires. However as most fires are caused by pit-sawyers, poachers or honey gatherers more effort has
to be undertaken to cut down these illegal activities. A paramilitary ranger troop patrol the forests could
serve as an effective deterrent, as proven successful on Mt. Kenya.
The construction of open strips as fire breaks seems generally not suitable for the Kilimanjaro
due to the very difficult, inaccessible and steep slopes. One suitable area however exists on the south
eastern slopes where there is a plateau around 2700-2800 m with moorland vegetation, formed by tussock
grasses, occurring at the fringe of the forest. In this area grassland fires affecting the bordering forests are
very common and the construction of open strips to prevent fire from spreading into the forest seems to be
possible and effective. At the lower forest boundary, fire lines could be reactivated and cleared before the
dry seasons.
There is also a need for better forest fire early warning system on Mt. Kilimanjaro. One
possibility is the establishment of fire observation points such as fire towers on higher hills or ridges near
existing ranger posts or tourist camps and huts. The fire-fighting capabilities could be significantly boosted
with the provision of one or two small airplanes, as is the case for Mt. Kenya. A national park of the size,
topography and importance like Kilimanjaro cannot be managed properly in many respects (e. g.
observation of poachers) without such modern equipment. The fire fighting equipment also needs to be
supplemented by a suitably equipped task force.
As mentioned earlier, many NGO’s, some villages and private institutions run re- and
afforestation projects on the mountain. However, a large scale effort is missing. The forest department in
Moshi, which acts as the official initiator of such projects, has failed completely to protect the indigenous
forests. Nurseries are used as maize fields and illegal cutting of timber is not prohibited effectively. No
government supported tree nursery exists on the southern slopes. A first step might be the recent decision
to incorporate the forest belt into the National Park, since this institution should be able to employ well
paid rangers to take care for its resources. In addition, bearing in mind the financial resources – the annual
park income is US$ 6.5 million (20,000 park visitors stay on average 5 days paying a daily entrance fee of
US$ 65) park authorities should ensure that a certain fixed share of the income flows directly back to
Kilimanjaro.
The local people should also be involved and get benefits out of the forest to be more interested
in its protection. As shown on Mt. Kenya and some Kilimanjaro villages adjacent to the forest, people
should be given the responsibility to ensure that there is no encroachment into the forest, to organize
patrols and to mobilize people for fire fighting. These efforts should be compensated by rewarding the
participating villages with timber which has been confiscated during patrols.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
51
Likewise, the half-mile forest strip should be managed similarly by involving local people. This
strip of 8,769 ha on the southern and eastern slope, ranges between the plantation belt and the forest
reserve. It is meant to provide timber and firewood, and in some areas even pines, cypress and eucalyptus
have been planted. However, people use this strip mainly in an uncontrolled way to collect fodder for their
livestock. Sometimes the area even serves as pasture land. Since nobody feels responsible, this area is
highly degraded and not managed properly. This is quiet different from the situation in the beginning. The
half-mile forest strip was established resulting from a request of the Chagga Council in 1941 brought to the
colonial government in response to the need of unrestricted availability of forest products, which could not
be gained in the forest reserve. During the 20 years of managing the forest, the local people contributed
substantially in planting trees, demarcating the boundary and fighting fires. In return, they were allowed to
obtain forest products freely or at minimal costs (MISANA 1999). In 1962, however, management of the
strip was transferred to the District Council and later (1972) the central government took control of the
strip, which was then managed by the Forest Division, which restricted local people from collecting forest
products freely. This situation has not changed up to now, although the management of the strip has been
referred to the district councils again in 1987, together with the Forest Division.
Currently, a discussion has started to give the area back to the villages, which has been already
done in some areas. This may offer the possibility that the local population will take more care of the land
in the future than today. Since many fires originate in the agricultural land surrounding the forest and in the
half-mile forest strip, it would reduce the fire risk for the forest as well. However, it cannot be excluded
that this land would be abused as agricultural land and for settlement. Therefore, governmental control
would be necessary. In any case, this huge area - when properly used and planted with timber trees - has a
high potential to reduce the pressure from the indigenous forests.
Regarding replanting it appears to be a necessity to employ well-trained foresters, perhaps from
outside the country, to start forestation projects and to educate local foresters. Especially the choice of
suitable tree species is of importance. A variety of different species, not only the widely used exotic ones,
but also indigenous could be used. Riverine areas could exclusively be replanted with indigenous trees, and
existing natural forest patches in the half-mile forest strip situated mostly near rivers should not be
replaced with forest plantations.
Right now, the forest plantations in West and North Kilimanjaro are not managed properly.
During the aerial survey (LAMBRECHTS et al. 2002) it became evident that over 50% of the Shamba system
areas are not under tree growing, either because replanting was not successful or because it was not
undertaken at all. Since the production of timber is the primary goal and growing vegetables only a
secondary one, the share of tree-planted areas in the forest plantations has to be much higher, even though
this implies the removal of illegally erected villages inside the forest reserve.
The Chagga home gardens (vihamba) are an old and very sustainable way of land use that meets
several different demands. Besides crop production, the sparse tree layer provides people with fire wood,
fodder and timber. However, the high demand for wood and the introduction of coffee varieties that are
sun-tolerant endangers this effective system. In some areas of the mountain (e. g. on the eastern slopes) the
trees in the banana fields are very scattered or already missing. Therefore it seems to be necessary, in order
to reduce the pressure on the forest, to support the tree planting in the Chagga home gardens with their
unique agro forestry system. There could also be a program that rewards farmers to have a certain share of
their land covered with trees. As the banana belt is nearly as extensive as the forest reserve, this will
certainly have major effects in terms of forest protection and water balance. In combination with new
marketing and farming strategies for growing organic coffee using traditional methods an advertising
campaign should be started. The campaign should point out that the consumer buys high quality
ecologically grown coffee supporting not only sustainable land use and an old African cultural heritage but
he is also protecting the rain forest. A certain share of the coffee prize should be used to run this
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
52
environmental Chagga home garden program. Government programs and donor agencies should cover any
additional program costs instead of using financial resources for other expensive and in the long-term less
effective projects such as dairying, which are not suitable for the Chagga home gardens.
Finally, a comprehensive, holistic environmental development plan focusing on fire risk and
forest destruction while defining conservation strategies to ensure the long term sustainability of the
mountain should combine all the different requirements, constraints and aims under one leading guideline.
The high complexity of Mt. Kilimanjaro’s ecosystem requires the expertise of scientists familiar with the
biodiversity patterns and ecological conditions of the mountain for the preparation of such a report.
8.4 Promotion of ecosystem friendly livelihood opportunities
A key adaptation response to the threats facing the Kilimanjaro ecosystem is also to reduce
human and livelihood pressures that make it vulnerable to other stresses such as climatic change. Humans
have continuously occupied the slopes of Mt. Kilimanjaro for the last 2000 years (SCHMIDT 1989).
However, the population has multiplied by 20 during the 100 years since 1895. In general, the growth rate
is exponential, albeit it is slowly decreasing since 1978. The annual average population growth has been
2.1% between 1978 and 1988 and decreased to 1.6% between 1988 and 2002. In 1991 GAMASSA estimated
the doubling of the population within 39 years. The population increase was much higher in urban than in
rural areas. While the population has doubled between 1967 and 2002 on Kilimanjaro, the population in
Moshi town multiplied 5 times during that same period.
The overall population density for the four districts that comprise Mt Kilimanjaro region was 198
people per km2 in 2002. If population density would be based upon actual land availability, this number
would be approximately 331 people per km2. Most of the population is concentrated at an altitude between
1100 and 1800 m. Here, densities varying from 500 to 1000 people per km2 have been recorded in certain
places (TIMBERLAKE 1986, FAO 1986). From these data it is evident that every effort in environmental
protection, which ignores the demands of a still fast growing population, will fail. Therefore, it is necessary
to boost livelihood prospects in sectors that do not pose threat to the Kilimanjaro ecosystem.
Today, the bulk of development processes is departing from the mountain, although most of the
population still remains there. Manufacturing in the region has collapsed following the closure of most
leading factories. Even in the tourism market neighboring Arusha out-competes Moshi in the Kilimanjaro
region.
Agriculture, the livelihood for most residents, accounts for over 85% of the total regional income
with coffee being the main cash crop. Lately, however, coffee has become less profitable due to traditional
farming techniques and very low coffee prices. Today (March 2003) a farmer gets only 400 TSH
(equivalent to 0.4 US Dollar) for one kg of coffee and many farmers think about replacing their coffee
trees with other crops such as passion fruits. In the 1970s 35,000 tonnes of coffee were harvested annually
in the region whereas today only 12,000 to 15,000 tonnes are produced. Still, coffee accounts for over 60%
of the region’s income and authorities plan to raise coffee production to over 45,000 tonnes. The Coffee
Revival Programme, which was launched in 1998 aimed at producing 100,000 tonnes. The strategy
involves the formation of coffee revival committees and replacement of old coffee trees (THE EAST
AFRICAN, August 26-September 1, 2002). However, the question whether this could be the solution
remains open considering the over supply on the world market. A better way seems to be to raise the
quality. According to Tanzania Coffee Board officials (THE EAST AFRICAN, August 26-September 1, 2002)
new crop marketing and farming strategies aim at growing organic coffee through traditional methods
without any use of pesticides and artificial fertilizers.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
53
Despite its large cattle herds and successive government efforts to promote dairying, Tanzania is
a net importer of dairy products (MDOE & WIGGINS 1997). Since independence in 1961 the government of
Tanzania has tried to encourage more domestic milk production to achieve self-sufficiency. Since most
cattle are stall-fed, and fodder has to be collected and brought from remote areas, large scale dairying in the
Chagga home-gardens offers no alternative. Expensive dairy development projects therefore may not be
the right way to improve livelihood in the submontane banana zone in the long-term.
In terms of tourism Kilimanjaro National Park (KINAPA) is a major tourist attraction in
Tanzania and earns the most foreign exchange of any National Park in Tanzania (NEWMARK & NGUYE
1991). Most visitors are mainly interested in reaching the summit of Kibo, known as Uhuru Peak, the
highest point in Africa. In the year 1976 5,000 people tried to reach the summit (CARLÉ 1977). In the year
2002 this number had increased to 20,000 (Chief Park Warden KINAPA, pers. comm.). Since its
establishment in 1972, the number of visitors of KINAPA has multiplied by five. Today, about 100,000
people – porters and tourists combined - frequent the alpine areas of Kilimanjaro every year. Such
increasing numbers of visitors have certainly effects on the environment. Especially the alpine zone with
its highly specialized flora and fauna is a very sensitive ecosystem. Since a national park is meant for
nature protection, it appears that tourism has reached a level which should not be exceeded. Therefore,
alternatives to the mountain climbing tourism have to be explored.
During the last years a strong development of ecotourism could be observed world wide.
Ecotourism can not only help in protecting the environment especially on Mt. Kilimanjaro put also it
allows local population to participate in its economic potential. In the long run this type of tourism could
be an alternative to the usual climbing tourism on Mt. Kilimanjaro. Another possibility is one or two day
guided nature trips to the forest or to the lower alpine zone organized by the Kilimanjaro National Park.
Many tourists would prefer to visit only the lower vegetation zones of the park instead of climbing. A
special training program for guides should provide them with sufficient, basic knowledge about main
vegetation types, flora and fauna to explain the mountain ecosystem to interested tourists. For the
promotion of tourism it is of fundamental importance to generally improve tourist facilities and in
particular to raise the quality of Tanzanian tourist hotels.
To summarize, due to a rapidly growing population, the decline in coffee production, and the
collapse of manufacturing industry, the Kilimanjaro Region, which once has been one of Tanzania’s
leading economical areas, is now among the most poverty stricken. The region’s annual per capita income
is less than TSH 96,390 (US$ 96) (THE EAST AFRICAN, August 26-September 1, 2002). The most
promising economic alternatives for the region currently appear to be the promotion of high quality organic
coffee rather than necessarily increasing the quantity of production, the production of new cash crops such
as passion fruits and flowers, as well as the improvement of eco-tourism and improvement of tourism
infrastructure.
9. Concluding remarks
Climate change poses significant risks for Tanzania. While projected trends in precipitation are
uncertain (and may differ for various areas of the country), there is a high likelihood of year-round
temperature increase, as well as sea level rise. The sectors potentially impacted by climate change include
agriculture, forests, water resources, coastal resources, human health, and energy, industry and transport.
Given the low level of human development, extreme poverty, and high dependence on agriculture and
natural resources, Tanzania may be quite vulnerable to projected climatic changes.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
54
9.1 Differentiated adaptation strategy
While uncertainties in climate change and impacts projections pose a challenge for anticipatory
adaptation for any country, Tanzania’s case has several specific characteristics that may argue for a
differentiated adaptation strategy.
First, the climate change projections on which all national impact and vulnerability assessments
(all the way to the Initial National Communication of 2003) rely on an older generation of climate models
and scenarios (circa early 1990s). A preliminary analysis based on more recent climate models conducted
as part of this study concludes that temperature increases might be somewhat lower than (although broadly
consistent with) the estimates used in the National Communication and the National Climate Change
Action Plan. Updating of climate scenarios and impact projections through the use of multiple and more
recent models might therefore be advisable prior to the formulation of aggressive (and potentially
expensive) adaptation responses. This should not however affect “no regrets” adaptation measures such as
leakage prevention and water conservation.
A second characteristic feature of Tanzania is that certain sectors are projected to experience both
negative and positive impacts under climate change – for example, while production of maize is projected
to decline, the production of two key cash crops (coffee and cotton) which contribute significantly to the
GNI is projected to increase. Similarly, while stream-flow declines are projected to decline in two of three
key river basins (Ruvu and Pangani), they are projected to increase in the third (Rufiji). The implication for
adaptation therefore may be to not only cushion adverse impacts, but also to harness positive opportunities.
This suggests consideration of an enhanced portfolio of linked-adaptation responses – for example a
strategic shift from maize to cash crops over the medium term, and inter-basin transfers in the case of water
resources. Such strategic shifts however may entail economic and dislocation costs – and therefore require
careful screening, particularly with regard to their effects on equity and rural livelihoods. More rigorous
testing of particular crop and stream-flow projections may also be advisable prior to undertaking such
adaptation responses.
A third key characteristic is that unlike most other countries where the need for adaptation relies
on projections of future impacts, some discernible trends in climate and attendant impacts are already
underway in Tanzania. Such impacts – as is the case of the Kilimanjaro ecosystem - argue for more
immediate adaptation responses as opposed to a “wait and see” strategy.
9.2 Climate change and donor portfolios
Tanzania receives close to a billion dollars of development assistance annually. An analysis of
donor projects using the OECD/World Bank Creditor Reporting System (CRS) database reveals that
roughly 12 – 25% (in terms of investment dollars) and 20-30% (in terms of number of projects) of donor
portfolios in Tanzania may be potentially affected by climate change. This includes both activities in
sectors which may themselves be impacted by climate change, as well as those projects and other activities
which may influence the vulnerability of natural or human systems to climate change. These numbers are
only indicative at best, given that any classification based on sectors suffers from over-simplification.
Nevertheless, such measures can serve as a crude barometer to assess the degree to which particular
projects or development strategies may need to take climate change concerns into account. Several donor
strategies in fact already do make frequent references to the impacts of climate variability (such as El
Nino) and linkages between such events and economic performance. There is however as yet no explicit
reference to climate change. The (relatively few) development projects that were reviewed for this report
did not pay attention to the risks associated with climate change.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
55
9.3 Attention to climate change concerns in national planning
At the national level meanwhile Tanzania has a draft National Action Plan on Climate Change
since 1997 that highlights priorities on three time-scales (Short term 1-2 years; Medium term 2-5 years;
and Long term 10-20 years). The short term primarily focuses on capacity building through conferences;
the medium term flags “projects internalizing climate change aspects… especially those reducing GHG
emissions” and recommends the introduction of economic instruments to accomplish such goals; and the
long term identifies major infrastructure projects in energy, transportation, and coastal zones as priority
areas. While the sequencing appears reasonable, the plan remains short on specific details on how it may
be implemented. Tanzania’s recent National Communication to the UN Convention on Biodiversity, and
its report to the World Summit on Sustainable Development only make tangential references to climate
change. Its Poverty Reduction Strategy paper (PRSP) does explicitly recognize the significance of current
climatic impacts on the poor, although the potential links between climatic factors and performance of key
sectors (such as agriculture) are generally not discussed.
There is however considerable synergy between priorities of at least some national plans and the
measures that may be required for adaptation. Specifically, the National Environmental Policy which
emphasizes measures to improve the resilience of the agricultural sector, the National Water Policy that
highlights efficient water use and water conservation, and the National Forest Policy which highlights
forest conservation and biodiversity preservation. However, some of these goals (such as Water
Conservation) have been articulated in previous plans, but have not been successfully implemented.
Therefore, despite the obvious synergies between such policies and climate change adaptation, a key
obstacle facing successful “mainstreaming” is successful implementation.
9.4 Climate risks in perspective on Mount Kilimanjaro
The second half of this report discusses in-depth climate change impacts and policy responses on
the Mount Kilimanjaro ecosystem – Africa’s highest mountain and largest glacier, a biodiversity hotspot,
and a UNESCO World Heritage Site. Glaciers on Mount Kilimanjaro have been in a general state of retreat
on account of natural causes for over a hundred and fifty years. A decline in precipitation coupled with a
local warming trend that has been recorded in the second half of the twentieth century accelerated their
retreat, and the ice cap is projected to vanish entirely by as early as 2020. While the symbolism of this loss
is indeed significant, this analysis concludes that the impact of such a loss on the physical and socioeconomic
system is likely to be very limited. The present glaciers are already very small, and cover an area
which is only 0.2% of the forest belt on Mount Kilimanjaro. Glaciers do not feed any major rivers, and
even when they would have melted the mountain will still receive precipitation. Further, even without
glaciers Mount Kilimanjaro will remain the world’s highest free standing mountain and with Africa’s
highest peak. Therefore, it is unlikely that the loss of glaciers would have a significant long-term impact on
tourism. It must however be emphasised that ice-cores on the Kilimanjaro are a repository of paleoclimatic
records, and valuable climatic records would be irreplaceably lost with the loss of the ice cap.
The increase in temperatures and a concomitant decline in precipitation have also significantly
enhanced the intensity and risk of forest fires on the Kilimanjaro. Climatic changes have not shifted
vegetation zones upwards as in the case of other mountains, but on Mt. Kilimanjaro they have pushed the
upper forest line downward as a result of increase in forest fire risk and intensity on the upper fringes of
the forest. A whole vegetation zone, the ericaceous belt, has moved downwards since 1976 by several
hundred meters, substituting 13000 ha of forest. The replacement of the fog intercepting forest belt by low
lying shrub has already seriously impacted the hydrological balance of the mountain as fog intercepting
cloud forests play a key role in the water budgets of high altitude drainage basins. The decline of 13000 ha
of cloud forest since 1976 has already resulted in a reduction of the water yields of about 58 million m3 of
water every year. This constitutes about 10% of the annual fog water input of the whole forest belt. Not
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
56
included in this calculation are the several thousands of hectares of destroyed ericaceous bush land and the
loss of montane forests due to human activities such as logging. These impacts have implications that
extend beyond the region as it feeds the Pangani river, one of Tanzania’s largest, which is responsible for
20% of Tanzania’s electricity output.
Looking into the future, a continuation of current trends in climatic changes, fire frequency and
destructive human influence most of the remaining subalpine Erica forests could disappear within five
years. With this, Mt. Kilimanjaro will have lost its most effective water catchment area as fog interception
is of highest importance in the Erica forests. A further retreat of the upper forest line by about 100 m
altitude seems to be probable until 2020. Increasing logging activity in combination with a higher number
of forest fires is also expected to destroy the lower forest zone up to an average altitude of 2000 m. This
will result in a further shrinking and fragmentation of the forest belt.
9.5 Policy responses for Mount Kilimanjaro
While glacier retreat is inevitable and cannot even be delayed, forest fire risk can indeed be
reduced. Climate change only adds to the urgency of fire prevention and control, as well as forest
conservation activities on Mount Kilimanjaro. Among the measures identified by this report are
institutional measures such as the inclusion of the forest belt into the Kilimanjaro National Park and
creation of a paramilitary ranger group (as in Mount Kenya) to deter logging, as well as better investments
in early warning systems, particularly the purchase of one or two aircraft for aerial surveillance. There is
also a need to limit cross-border migration of big game from neighboring Amboseli, which is adding to the
stress on the Kilimanjaro ecosystem.
In addition to such piecemeal solutions there is an urgent need to better understand the livelihood
needs of the local population to engage them more successfully in conservation and fire-prevention efforts.
For example, an earlier policy response – the banning of camp-fires – did not have the desired effect
because most of the fires were actually being lit not by mountaineers, but by honey collectors. A more
sustainable solution therefore needs to identify viable livelihood opportunities that take some of the human
pressures away from the forest. Creative solutions to boost local incomes, such as provision of incentives
to switch to more lucrative specialty coffee production, may therefore be part of a package of responses
that may help reduce the pressures on activities like logging and honey collection. Finally, there is a critical
need to develop a comprehensive and holistic development plan focusing on fire-risk and forest destruction
as well as to identify conservation strategies to ensure the long term sustainability of the valuable resources
of Mount Kilimanjaro.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
57
APPENDIX A: PREDICTIVE ERRORS FOR SCENGEN ANALYSIS FOR TANZANIA
The table below shows the predictive error for annual precipitation levels for each SCENGEN
model for each country. Each model is ranked by its error score, which was computed using the formula
100*[(MODEL MEAN BASELINE / OBSERVED) - 1.0]. Error scores closest to zero are optimal. The six
models with the highest error scores from the estimation were dropped from the analysis.
Predictive errors for each SCENGEN model for Tanzania
Average error32 Minimum error Maximum error
Models to be kept for estimation
ECH3TR95 7% 3% 12%
ECH4TR98 13% 1% 27%
CCSRTR96 14% 5% 26%
HAD3TR00 18% 7% 30%
CERFTR98 22% 18% 25%
BMRCTR98 23% 2% 45%
HAD2TR95 24% 10% 42%
GFDLTR90 25% 1% 37%
CSI2TR96 32% 24% 40%
PCM_TR00 34% 7% 45%
CSM_TR98 35% 19% 57%
Models to be dropped from estimation
IAP_TR97 40% 7% 93%
GISSTR95 48% 4% 125%
LMD_TR98 63% 29% 100%
CCC1TR99 73% 53% 98%
W&M_TR95 94% 32% 136%
MRI_TR96 132% 94% 154%
32 SCENGEN outputs data for 5×5 degree grids. To estimate for an entire country, a 10×10 degree area was
used and the data output from the resulting four 5×5 grids were averaged. The maximum and minimum of
these four 5×5 grids are also reported.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
58
APPENDIX B: LIST OF PURPOSE CODES INCLUDED IN THE SELECTION OF CLIMATEAFFECTED
PROJECTS, ORGANIZED BY THE DAC SECTOR CODE
DAC
code
General sector name Purpose codes that are included in the selection
110 Education -
120 Health 12250 (infectious disease control)
130 Population -
140 Water supply and Sanitation
14000
14010
14015
14020 (water supply and sanitation – large systems)
14030 (water supply and sanitation – small systems)
14040 (river development)
14050 (waste management/disposal)
14081 (education/training: water supply and sanitation)
150 Government & civil society 15010 (economic & development policy/planning)
160 Other social infrastructure and
services
16330 (settlement) and
16340 (reconstruction relief)
210* Transport and storage All purpose codes
220 Communications -
230 Energy 23030 (renewable energy)
23065 (hydro-electric power plants)
[23067 (solar energy)]
23068 (wind power)
23069 (ocean power)
240 Banking and financial services -
250 Business and other services -
310 Agriculture, forestry, fishing All purpose codes
320 Industry, mining, construction -
330 Trade and tourism 33200 (tourism, general)
33210 (tourism policy and admin. management)
410 General environment protection 41000 (general environmental protection)
41010 (environmental policy and management)
41020 (biosphere protection)
41030 (biodiversity)
41040 (site preservation)
41050 (flood prevention/control)#
41081 (environmental education/training)
41082 (environmental research)
420 Women in development -
430 Other multi-sector 43030 (urban development)
43040 (rural development)
510 Structural adjustment -
520* Food aid excluding relief aid 52000 (dev. food aid/food security assist.)
52010 (food security programmes/food aid)
530 Other general programme and
commodity assistance
-
600 Action relating to debt -
700* Emergency relief 70000 (emergency assistance, general) #
710* Relief food aid 71000 (emergency food aid, general) #
71010 (emergency food aid) #
720* Non-food emergency and
distress relief
72000 (other emergency and distress relief) #
72010 (emergency/distress relief) #
910 Administrative costs of donors -
920 Support to NGOs -
930 Unallocated/unspecified -
* sector codes that are excluded in the second selection (low estimate).
# purpose codes that are included in the emergency selection
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
59
APPENDIX C: REVIEW OF SELECTED DONOR STRATEGIES FOR TANZANIA
C. 1 United National Development Program (UNDP)/United Nations Population Fund (UNPF)
Second country cooperation framework for the United Republic of Tanzania 2002-2006 (2001)
This cooperation framework focuses on governance and institutional aspects of poverty reduction, as well
as government services. Little attention is being paid to natural resources dimensions. Climate change,
current climate-related risks, or even food security in general, are not discussed.
C. 2 United Nations Development Assistance Framework (UNDAF) 2002-2006 (2001)
The UNDAF does not mention climate change. However, it recognizes the linkages between poverty and
degradation of natural resources. In particular, it mentions the increasing risk of desertification (with 60%
of Tanzania being composed of dry lands), partly caused by extensive deforestation. A fairly
comprehensive section on Tanzania’s vulnerability to natural hazards also highlights climate-related
concerns: “Natural and man-made disasters erode the coping capacity of the vulnerable population
especially in drought-prone areas. There have been poor rains in Central Tanzania for the last three years,
and traditional coping strategies are breaking down as land pressure increases. These types of shocks
have become a frequent phenomenon in Tanzania in recent years. Floods and droughts, epidemics and
crop pests, environmental damage and economic instabilities, have all had their effects on people’s
capacity to meet their basic needs and subsequently their ability to survive and pursue their development
ambitions and potential” In addition, the UNDAF observes a worrisome trend: “Some claim that during
recent years emergency preparedness has actually decreased and dependency on external support in these
kinds of situations has increased. Long term disaster management strategies to deal with predictable,
poverty related emergencies are needed to use available resources most effectively.” This general concern
is not yet translated into concrete activities in the UNDAF.
C. 3 United Nations Emergency Consolidated Appeal for the Drought in Tanzania 2001
This appeal illustrates Tanzania’s high vulnerability to climate variability: “The 1999/2000 rains were very
poor in many parts of northern and central Tanzania. This has resulted in abnormally low levels of food
production, particularly of the staple crop, maize grain, and has also caused a very poor cash crop harvest,
thereby further reducing the cash income of the drought-affected households. This has been highly
damaging to the household food security of many farming families in the semi-arid areas, who have
suffered a fourth consecutive year of poor harvests and low-income levels. This cumulative effect has
greatly undermined their purchasing power, forcing many of the poorest families to sell productive assets
in order to survive. The recurrent nature of these food crises exposes the underlying layer of core poverty”.
The appeal is intended to address Tanzania’s consecutive and chronic droughts, which affect the lives of
over 9 million people, almost 30% of the total population of Tanzania. About 1.3 million of these live in a
situation of total food insecurity. Instead of applying emergency measures year after year, the appeal
proposes a more fundamental approach, part of longer-term integrated development strategies, including
the Rural Development Strategy and the Agricultural Development Strategy which are currently under
development, as well as improved early warning systems. Despite this longer-term focus, shifts in risks, for
instance due to climate change, are not discussed. The implementation of the appeal, and particularly the
longer-term components, could involve up to eight members of the UN system: FAO, WFP, UNICEF,
ILO, UNIDO, IFAD, UNDP and the World Bank.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
60
C. 4 African Development Bank
Country Strategy Paper 1999-2001 (2000)
Country Economic Profile
In its macroeconomic analysis, the strategy notes that overall macroeconomic performance has been
satisfactory, but “growth rates have been fluctuating from year to year reflecting the vulnerability of the
economy to external shocks. Although strong growth was registered in FY 1996/97 (4.2 percent), it
declined to 3.3 percent in FY 1997/98 due to the adverse impact of the drought on agricultural output. The
drought was followed by the El-Nino floods late 1997 and early 1998, which destroyed some of the crops
and damaged roads, thereby, disrupting internal movement of agricultural commodities as well as export
shipments.” In response, the strategy underlines the need for “an aggressive export promotion drive and
continued diversification of the export base.” The direct causes of Tanzania’s vulnerability to natural
hazards are not analyzed in the macroeconomic analysis or in the sectoral sections.
In a section on poverty, the strategy highlights the links between poverty, drought, and food insecurity:
“Since the poor are entirely dependent on agriculture (mainly crops) for their livelihood, their incomes and
food consumption are vulnerable to droughts. Food insecurity is therefore a major feature of poverty,
especially in drought-prone areas of the country”. The strategy also notes that less than 20 percent of the
irrigation potential is utilized, unnecessarily exposing agricultural production to droughts. At the same time
however, the agriculture section of the strategy attributes poor agricultural performance mainly to limited
access to agricultural credits and weak extension services, poor transport infrastructure, limited use of
modern inputs (mainly due to high costs), weak agriculture planning and program implementation, and low
budgetary allocation to the sector. Dealing with droughts and irrigation are not mentioned here. Donor
support has in the past included, among others, small-scale irrigation and soil conservation. However, the
results have been mixed, mainly due to lack of counterpart funding, weak institutional capacity in the
Ministry of Agriculture, and the lack of a coherent sector framework.
Similarly, the strategy point to the underlying patterns causing water supply problems: “While droughts
have contributed to water supply problems, the underlying factors include weak institutional capacity in the
sector, poor water resource management, and the dilapidated condition of the water schemes and
distribution networks in the rural and urban areas resulting from the under-funding of maintenance and
rehabilitation.” Hence, the AfDB also focuses mainly on sector reform, operation efficiency, and
rehabilitation and expansion of existing facilities. In addition, it has adopted a river basin approach for
water management improvements. Finally, the strategy also notes the widespread environmental problems,
including land degradation, desertification, loss of biodiversity and wildlife, and the depletion of marine
and coastal resources. The environmental degradation is attributed to widespread poverty, high population
growth, and poor natural resource management practices. In that context, climatic factors (including
wildfires and flood and drought risks related to climate variability) are not discussed. Climate change is not
mentioned anywhere in the strategy.
The AfDB Country Economic Profile (from 1995) highlights the interrelationships between natural hazards
and natural resources management: “There is widespread consensus that one of the major problems facing
the nation is land degradation. This takes many forms: soil erosion, deforestation, bush fires and
overgrazing. The root cause often lies in the actions of the agricultural producers themselves. Land
degradation results in a loss of productivity in agriculture, land use conflicts, loss of biodiversity and
changes in water catchment areas which have led to both drought and floods”. These issues are highlighted
throughout the study, and illustrated by several examples. Deforestation in particular is highlighted as a
pressing problem, for biodiversity, but also floods and droughts. Bush fires are both a cause and
consequence. The profile also describes the environmental problems in coastal areas, including pollution,
but also clearance of mangrove forests and destruction of corals (particularly by dynamite fishing). Among
the consequences is a depletion of fishery resources.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
61
While the document reviews policy options to respond to the challenges, it mainly emphasizes the
implementation and further development of government policies that were already in place or under
consideration. Climate change is not discussed33.
C. 5 World Bank
Country Assistance Strategy (2000)
The effect of climate on the country’s performance is recognized by the fact that climatic conditions are
mentioned as part of the inputs to a (low-case) macro-economic scenario for Tanzania’s development. In
addition, the strategy states that “Tanzania is vulnerable to external shocks, commodity price changes and
droughts”, putting climatic conditions on a par with major economic issues that are discussed at length. In
addition, it describes that an infrastructure project has been restructured to address “El Nino damages”.
However, vulnerability to floods and drought is not mentioned, not as a risk to the Bank’s own projects,
nor as a development opportunity that could have been addressed by concrete activities. Climate change is
not mentioned.
C. 6 IFAD
Country Strategic Opportunities Paper (1998)
According to IFAD’s strategy paper “agriculture remains exposed to the vagaries of nature”. For instance,
the high growth in maize production (the main staple crop) is highly susceptible to weather conditions.
While the country has a structural food deficit of about 700 tons, imports rise to up to 1.5 million tons in
times of flood or drought. The main constraints to agricultural production are lack of irrigation,
unavailability of credit for the poorest segment of the population, and absence of an appropriate
institutional framework to support agricultural development activities. Donor support has been ineffective
due to poor counterpart funding from the government, cumbersome and centralized procedures, lack of
beneficiary participation and ownership, and lack of appropriate targeting criteria for women. IFAD aims
to address these sector-wide issues in order to make the agricultural sector more productive; at the same
time, this should contribute to a decrease in vulnerability to adverse weather conditions, particularly for the
smallholders who account for about 85% of the cultivable land. Climate-related risks to IFAD projects are
not discussed explicitly, although the report mentions the flood damage to irrigation schemes during the
1997/98 El Nino. Climate change is not mentioned.
C. 7 DFID
Country Strategy Paper (1999)
This country paper recognizes that Tanzania’s agriculture, accounting for half the GDP and 75% of
exports, is “highly vulnerable to climatic shocks”. In the past, DFID has provided substantial support in
response to natural disasters. The strategy for the coming years contains assistance to help protect poor
people’s livelihoods and strengthen the government’s capacity to prepare for and manage disasters.
However, climate risks to development investments and their outcomes are not recognized as a concern,
and climate change is not even mentioned.
C. 8 EU
Tanzania Strategy Paper for the Period 2001-2007 (2002)
This country strategy paper defines the priorities for EU assistance to Tanzania in the period 2001-2007.
The main sectors to be targeted are transport infrastructure and basic education. Further assistance will go
to governance and macro-economic support in line with the PRSP objectives. Ongoing programs in
agriculture, water & sewerage, and environment, will be continued. Despite the vulnerability of some of
these sectors, even current climate risks are mentioned only once, in the margins of an agriculture section.
33 It is noted that Tanzania had not yet ratified the UNFCCC, but was in the process of doing so.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
62
C. 9 Ireland Aid
Country Strategy Paper for Bilateral Aid Programme 2000 – 2002 (1999)
Ireland’s country strategy focuses on poverty reduction. While increasing food and livelihoods security are
among the key goals, weather and climate-related risks are not mentioned at all. In the coastal zone, Ireland
Aid is financing the Tanga Coastal Zone Conservation and Development Programme, managed by IUCN,
which aims to address issues like the destruction of coral reefs and mangrove swamps. Again, no reference
is made to climate change.
C. 10 JICA
Country Study for Japan’s Official Development Assistance to the United Republic of Tanzania
(1997)
Country Profile on Environment (1999)
The JICA country study recognizes the severe economic implications of climate risks in Tanzania:
“Several factors are considered directly responsible for the weakened economy. They include certain
political and economic policy choices made following independence, a rapidly growing population, climate
anomalies, and a deteriorating conditions for trade in the international market.” Recognizing the severe
pressures on Tanzania’s natural resource base, JICA aims to provide assistance to alleviate those pressures,
particularly in the forestry and water resources sectors. However, no attention is paid to interactions of
climate-related risks, poverty, and land degradation and water scarcity, or to ways to reduce that
vulnerability.
The JICA Country Profile on Environment gives a complete overview of environmental problems facing
Tanzania. It includes issues like desertification, deforestation and forest fires, but does not mention new
risks due to climate change. Even climate variability is largely ignored, for instance when discussing water
resources: “Tanzania is a well-watered country with moderate to good rainfall and with many rivers and
lakes. However, rainfall is seasonal and water is not readily available in the dry season.” The most
pressing problems, however, occur not in the average dry season, but in a dryer than normal period, due to
climate variability.
C. 11 SIDA
Tanzania Country Strategy 2001-2005 (2000)
The country strategy mentions that desertification, deforestation, and problems with coastal and marine
environments and urban settlements threaten Tanzania’s sustainable development. However, neither
climate change, nor even current climate risks that interact with these issues, are discussed.
C. 12 USAID Tanzania
Summary Strategic Plan for Environment and Natural Resources (1999)
Annual Report (2002)
This strategic plan provides an update of previous work of USAID in the area of environmentally
sustainable natural resources management. Its new focus will be on improved conservation of coastal
resources and wildlife in targeted areas. Climate change and sea level rise are not among the listed causes
of coastal degradation. Similarly, several possibly vulnerable biodiversity projects neglect climate risks.
Similarly, climate risks do not appear in USAID’s Annual Report for Tanzania.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
63
APPENDIX D: REVIEW OF SELECTED DEVELOPMENT PROJECTS/PROGRAMS
D.1 Projects dealing explicitly with climate related risks
D.1.1 US Country Studies Program
The US Country Studies Program supported several studies in Tanzania, on both mitigation and
vulnerability & adaptation. Results from these studies, which were performed by the Centre for Energy,
Environment, Science and Technology (CEEST) in Dar es Salaam, are reviewed in the Tier-1 component
of this project.
D.1.2 Draft National Action Plan on Climate Change in Tanzania (CEEST, 1998)
This Plan, developed with underlying materials from the US Country Study, gives a comprehensive
overview of Tanzania’s vulnerability to climate change in various sectors, and discusses both mitigation
and adaptation options. For adaptation, the focus is on no-regrets measures integrated in sectoral
development. Some adaptation options are proposed in Agriculture, Livestock, Forestry, Water Resources,
and Coastal Zones. No attention is paid to possible overlaps of adaptation and mitigation options, such as
in the forestry sector.
While this 1998 Draft Plan is comprehensive and detailed, it is unclear what its impact has been. It appears
that it has not been formally adopted by the Government. Moreover, its recommendations are not well
reflected in subsequent sectoral and national development plans.
D.1.3 GTZ (Energy and Transport Division) : Measures to Implement the UN FCCC: Technological
and other Options for the Mitigation of Greenhouse Gases in Tanzania (1995)
This somewhat older report discusses GHG mitigation options in Tanzania, including in the forestry sector.
This sector might allow for projects that integrate adaptation and mitigation goals, but these are not
discussed.
D.2 Other Development Programs and Projects
D.2.1 World Bank Forest Conservation and Management Project
Project Appraisal Document (2001)
Social and Environmental Considerations (2002)
This project focuses on the development of the forestry sector, and on biodiversity conservation in
Tanzania’s forests. The latter component, which is supported by the GEF and implemented jointly with
UNDP, focuses on the Eastern Arc forests, which are recognized as “biodiversity hotspots”. Besides their
biodiversity value, these forests provide local livelihoods, and are crucial as water catchment areas for
Tanzania’s water supply and hydroelectric power generation.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
64
The project also aims to contribute to carbon sequestration, partly by limiting forest fires: “The main
techniques for increasing carbon uptake in miombo is the reduction in fire frequency. Experiments in many
parts of Africa have shown that woody biomass and soil carbon both increase if fires are excluded.
Permanent fire exclusion is virtually impossible in the strongly seasonal miombo climate, but a reduction
in frequency is probably achievable at reasonable cost. This would simultaneously increase carbon dioxide
uptake and decrease the emission of methane and ozone precursors.”
While the project thus explicitly addresses climate change, current climate-related risks to the project itself
are not discussed, and possible risks due to climate change, including more frequent forest and direct
threats to biodiversity, are entirely ignored. It is unclear whether such considerations would have changed
the project design, which in its current form already contributes to a reduction in the vulnerability of these
valuable forests.
D.2.2 GEF/World Bank Lake Victoria Environmental Management Project (supplemental credit)
Project Information Document (2001)
Integrated Safeguards Sheet (2001)
This project addresses the management of Lake Victoria, and affects the three countries around the lake
(Uganda, Kenya, and Tanzania), with Tanzania acting as the regional coordinator. It had many
components, varying from community-level management, to watershed improvement, to hydrometeorological
monitoring. Given the far-reaching environmental issues at stake, the project aims to put
the region on a long-term path of better management of the Lake and its surrounding natural resources.
Despite this long-term focus, climatic changes, which might have strong effects on water resources and
ecosystems, are not considered.
D.2.3 GEF/UNDP
Aerial Survey of the Threats to Mt Kilimanjaro Forests (2002) [www.tz.undp.org]
This aerial survey is part of UNDP’s Community Management of Protected Areas Conservation Project
(COMPACT), which promotes community-based biodiversity conservation in and around World Heritage
Sites (such as Kilimanjaro). The main threats identified for the Kilimanjaro region were: logging of
indigenous trees, forest fires, and establishment of settlements. No specific attention was paid to issues
related to changing climatic circumstances.
D.2.4 USAID
Tanzania Coastal Management Partnership: Options for a national integrated coastal management
policy (undated.)
This report (prepared by Tanzania’s National Environment Management Council and the University of
Rhode Island/Coastal Resources Center and supported by USAID) analyzes options for integrated coastal
zone management in Tanzania. Climate change and sea level rise are not discussed. While an ICZM
approach would certainly contribute to the sustainable development of Tanzania’s coastal areas,
opportunities may be missed.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
65
APPENDIX E: SOURCES FOR DOCUMENTATION
Statistics
CRS database, OECD/World Bank http://www.oecd.org/htm/M00005000/M00005347.htm
Government Documents
PRSP related documents www.worldbank.org/prsp
Poverty Reduction Strategy Paper (PRSP) (2000)
PRSP progress report (2001)
PRSP Joint Staff Assessment (by IDA and IMF) (2001)
PRSP Progress Report Joint Staff Assessment (by IDA and IMF) (2001)
Other national strategies www.tzonline.org
Tanzania Assistance Strategy (A Medium Term Framework for Promoting Local Ownership and
Development Partnerships) Consultation draft, Ministry of Finance (2001)
Tanzania Development Vision 2025
National Environmental Policy (1997)
UN Conventions
UN Convention on Climate Change (UNFCCC) www.unfccc.int
UN Convention to Combat Desertification (UNCCD) www.unccd.int
Proposed National Action Programme (1999)
Second National Report (2002)
UN Convention on Biodiversity (UNCBD) www.biodiv.org
National Report (2001) www.biodiv.org
World Summit on Sustainable Development www.johannesburgsummit.org
National Report to The Earth Summit on Sustainable Development (2002)
Country Profile (2002)
Donor Agencies
AfDB www.afdb.org
Country Environmental Profile, Environmental and Social Policy Working Paper Series, no. 26
(1995); Country Strategy Paper 1999-2001 (2000)
DFID www.dfid.gov.uk
Country Strategy Paper (1999)
GEF/UNDP
Aerial Survey of the Threats to Mt Kilimanjaro Forests (2002)www.tz.undp.org
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
66
EU
Tanzania Strategy Paper for the Period 2001-2007 (2002)
IFAD www.ifad.org
Country Strategic Opportunities Paper (1998)
JICA www.jica.go.jp
Country Study for Japan’s Official Development Assistance to the United Republic of Tanzania
(1997)
Country Profile on Environment (1999)
SIDA
Tanzania Country Strategy 2001-2005 (2000)
UN
United Nations Emergency Consolidated Appeal for the Drought in Tanzania 2001 Development
Assistance Framework (UNDAF) 2002-2006 (2001)
UNDP www.undp.org.np
United National Development Programme (UNDP)/Population Fund (UNPF) Second country
cooperation framework for the United Republic of Tanzania (2002-2006) (2001)
UNEP www.unep.org
USAID www.usaid.gov
Summary Strategic Plan for Environment and Natural Resources (1999)
Annual Report (2002)
Tanzania Coastal Management Partnership: Options for a national integrated coastal management
policy (n.d.)
World Bank www.worldbank.org
Country Assistance Strategy (2000)
World Bank Forest Conservation and Management Project. Project Appraisal Document (2001),
Social and Environmental Considerations (2002)
GEF/World Bank Lake Victoria Environmental Management Project (supplemental credit), Project
Information Document (2001), Integrated Safeguards Sheet (2001)
US Country Studies Program
GTZ (Energy and Transport Division)
Measures to Implement the UN FCCC: Technological and other Options for the Mitigation of
Greenhouse Gases in Tanzania (1995)
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
67
REFERENCES
Agrawala, S. and M. Berg, 2002, “Development and Climate Change Project: Concept Paper on Scope and
Criteria for Case Study selection. COM/ENV/EPOC/DCD/DAC(2002)1/Final, OECD, Paris.
Altmann, J.; Alberts, S. C.; Altmann, S. A. and Roy, S. B. (2002): Dramatic change in local climate
patterns in the Amboseli basin, Kenya. In: Afr. J. Ecol. 40: 248-251.
Beck, E., Scheibe, R. & Senser, M. (1986): The vegetation of the Shira plateau and the western slopes of
Kibo (Mount Kilimanjaro, Tanzania). In: Phytocoenologia 11: 1-30.
Brenan, J. P. M. (1978): Some aspects of the phytogeography of tropical Africa. In: Annals of the Missouri
Botanical Garden 65: 437-478.
Brewin, D. R. (1965): Kilimanjaro Agriculture. In: Tanganyika Notes and Records 64: 115-117.
Carlé, W. (1977): Der Kilimandjaro. In: Jh. Ges. Naturkde. Württ. 132: 50-109.
Cavelier, J. & Goldstein, G. (1989): Mist and fog interception in elfin cloud forests in Columbia and
Venezuela. In: Journal of Tropical Ecology 5: 309-322.
Cavelier, J., Jaramillo, M., Solis, D. and León, D. d. (1997): Water balance and nutrient inputs in bulk
precipitation in tropical montane cloud forest in Panama. In: Journal of Hydrology 193: 83-96.
Cavelier, J., Solis, D. and Jaramillo, M. A. (1996): Fog interception in montane forests across the Central
Cordillera of Panamá. In: Journal of Tropical Ecology 12: 357-369.
CIA (2002): World Fact Book. Central Intelligence Agency, available on-line
http://www.cia.gov/cia/publications/factbook/ Accessed January 16, 2003.
Clausnitzer, V. (2001): Notes on the species diversity of East African Odontata, with a checklist of species.
In: Odonatologica 30: 49-66.
Crutzen, P. J. and Andreae, M. O. (1990): Biomass burning in the tropics: impact on atmospheric chemical
and biogeochemical cycles. In: Science 250: 1669-1678.
Diaz, H. F. and Graham, N. E. (1996): Recent changes in tropical freezing heights and the role of sea
surface temperature. In: Nature 383: 152-155.
Downie, C. and P. Wilkinson (1972): The Geology of Kilimanjaro, Sheffield.
Eva, H. and Lambin, E. F. (2000): Fires and land-cover change in the tropics: a remote sensing analysis at
the landscape scale. In: Journal of Biogeography 27: 765-776.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
68
FAO (1986): Rehabilitation of traditional schemes (Phase I), Tanzania. AG: TCP/URT/4522, Mission
Report, Rome.
Forchhammer, P., Breuning, S. (1986): Annotated list of Cerambycidae, subfamily Lamiinae collected in
Tanzania, Zambia and Botswana 1969-80 with description of 86 new taxa. In: Natura Jutlandica 22:
1-72.
Gaffen, D.J., Santer, B. D., Boyle, J. S., Christy, J. R., Graham, N. E. & Ross, R. J. (2000): Multidecadal
changes in the vertical temperature structure of the tropical troposphere. In: Science 287: 1242-1245.
Gamassa, D. M. (1991): Historical change in human population on Mount Kilimanjaro and its
implications. In: Newmark, W. D. (ed.): The Conservation of Mount Kilimanjaro. The IUCN
Tropical Forest Programme: 1-8.
Gathaara, G. N., Woodley, B and Lambrechts, C (1999): Aerial survey of the destruction of Mt. Kenya,
Imenti and Ngare Ndare Forest Reserve. Kenya Wildlife Service.
Glasow, R. v. and A. Bott (1999): Interaction of radiation fog with tall vegetation. In: Atmospheric
Environment 33: 1333-1346.
Grabherr, G. and Pauli, H. (1994): Climate effects on mountain plants. In: Nature 369: 448.
Grimshaw, J. M. (1996): Aspects of the ecology and biogeography of the forest of the northern slope of
Mt. Kilimanjaro, Tanzania. Thesis, University of Oxford.
Grimshaw, J. M.; Cordeiro, N. J. and Foley, C. A. H. (1995): The mammals of Kilimanjaro. In: Journal of
East African Natural History 84: 105-139.
Hastenrath, S. and Kruss, P. D. (1992): The dramatic retreat of Mount Kenya’s glaciers between 1963 and
1987: greenhouse forcing. In: Annals of Glaciology 16: 127-133.
Hastenrath, S. & L. Greischar (1997): Glacier recession on Kilimanjaro, East Africa, 1912-89. In: Journal
of Glaciology 43: 455-459.
Hastenrath, S. (1984): The glaciers of Equatorial East Africa, Solid Earth Sciences Library, Dordrecht,
Boston, Lancaster.
Hastenrath, S. (2001): Variations of East African climate during the past two centuries. In: Climatic
Change 50: 209-217.
Hay, S. I., Cox, J., Rogers, D. J., Randolphs, S. E., Stern, D. I., Shanks, G. D., Myers, M. F. and Snow, R.
W. (2002): Climate change and the resurgence of malaria in the East African highlands. In: Nature
415: 905-909.
Hedberg, O. (1961): The phytogeographical position of the afroalpine flora. In: Recent Advances in
Botany 1: 914-919.
Hedberg, O. (1964): Features of afroalpine plant ecology. In: Acta Phytogeographica Suecica 49: 1-144.
Hemp, A. & Beck, E. (2001): Erica excelsa as a fire-tolerating component of Mt. Kilimanjaro’s forests. In:
Phytocoenologia 31: 449-475.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
69
Hemp, A. (1999): An ethnobotanical study on Mt. Kilimanjaro. In: Ecotropica 5: 147-165.
Hemp, A. (2001a): Ecology of the pteridophytes on the southern slopes of Mt. Kilimanjaro. Part II: Habitat
selection. In: Plant Biology 3: 493-523.
Hemp, A. (2001b): Life form and strategies of forest ferns on Mt. Kilimanjaro. In: Gottsberger, G. &
Liede, S. (eds.): Life forms and dynamics in tropical forests. Disserationes Botanicae 346: 95-130.
Hemp, A. (2002): Ecology of the pteridophytes on the southern slopes of Mt. Kilimanjaro. Part I:
Altitudinal distribution. In: Plant Ecology 159: 211-239.
Hemp, A. (in press): The impact of fire on diversity, structure and composition of Mt. Kilimanjaro’s
vegetation. Proceedings of the GMBA Symposium 19-24 August in Moshi, Tanzania.
Hemp, A., Hemp, C. and Winter, J. C. (1999): Der Kilimanjaro – Lebensräume zwischen tropischer Hitze
und Gletschereis. Natur und Mensch 1998, Jahresmitteilungen der Nürnberger Naturhistorischen
Gesellschaft: 5-28.
Hemp, C. & Hemp, A. (in press): Saltatoria cenoses of high altitude grasslands on Mt. Kilimanjaro,
Tanzania (Orthoptera: Saltatoria). In: Ecotropica.
Hemp, C. (2001): Ethnozoological Research on invertebrates on Mt. Kilimanjaro, Tanzania. In: Ecotropica
7: 139-149.
Hemp, C. (in press): Comparison of the endemic Saltatoria fauna (Orthoptera) of Mt. Kilimanjaro, Mt.
Meru and the Monduli Range. In: Journal of East African Natural History.
Hemp, C., Winter, J. C. (1999): Ethnozoologische Feldforschung am Kilimanjaro. Arthropoda. In:
Bayreuther Forum Ökologie 64: 167-199.
IPCC (2001): Climate Change 2001. Impacts, Adaptation, and Vulnerability, Intergovernmental Panel on
Climate Change, Geneva.
Irion, R. (2001): The melting snows of Kilimanjaro. In: Science 291: 1690-1691.
Jaeger, F. (1909): Forschungen in den Hochregionen des Kilimandscharo. In: Mitteilungen aus den
Deutschen Schutzgebieten 22: 113-146; 161-196.
Johnsen, P. and Forchhammer, P. (1975): Checklist of the Acridomorpha of Tanzania. In: Natura
Jutlandica 18: 26-52.
Jones, C.A. and J.R. Kiniry (eds.) (1986): CERES-Maize. A Simulation Model of Maize Growth and
Development, Texas A&M Press, College Station.
Juvik, J. O. and Nullet, D. (1993): Relationships between rainfall, cloud-water interception, and canopy
throughfall in a Hawaian montane forest. In: Hamilton, L. S., Juvik, J. O. & Scatena, F. N. (eds.):
Tropical montane cloud forests. Proceedings of an international symposium at San Juan, Puerto
Rico, 31 May-5 June 1993: 102-125.
Kaser, G. (1999): A review of the modern fluctuations of tropical glaciers. In: Global and Planetary
Change 22: 93-103.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
70
Kaser, G. (2001): Glacier-climate interaction at low latitudes. In: Journal of Glaciology 47: 195-204.
Kaser, G., Hardy, D. R., Mölg, T. & Bradley, R. S. (under review): Modern glacier retreat on Kilimanjaro
as evidence of climate change: Observations and facts.
Kaser, G., Hastenrath, S. and Ames, A. (1996): Mass balance profiles on tropical glaciers. In: Zeitschrift
für Gletscherkunde und Glazialgeologie 32: 75-81.
Kerfoot, O. (1968): Mist precipitation on vegetation. In: Forestry Abstracts 29: 8-20.
Klute, F. (1920): Ergebnisse der Forschungen am Kilimandscharo 1912, Berlin.
Körner, C. (1999): Alpine Plant Life. Functional Plant Ecology of High Mountain Ecosystems. – Springer,
Berlin-Heidelberg.
Kruss, P.D. (1983): Climate Change in East Africa. A numerical simulation from the 100 years of terminus
record at Lewis Glacier, Mount Kenya. In: Zeitschrift für Gletscherkunde und Glazialgeologie 19:
43-60.
Lambrechts, C., Woodley, B., Hemp, A., Hemp, C. and Nnyiti, P. (2002): Aerial survey of the threats to
Mt. Kilimanjaro forests, UNDP, Dar es Salaam.
Larcher, W. (1984): Ökologie der Pflanzen, Ulmer, Stuttgart.
Martens, P.; R.S. Kovats; S. Nijhof; P. de Vries; M.T.J. Livermore; D.J. Bradley; J. Cox; and
A.J. McMichael (1999): Climate change and future populations at risk of malaria. In: Global
Environmental Change 9: 89-107.
Mbonile, M. J. (1999b): Population mobility and migration in Mount Kilimanjaro. In: Mount Kilimanjaro:
Land Use and Environmental Management. French Institute for Research in Africa. IFRA Les
Cahiers 16: 24-33.
Mdoe, N. and Wiggins, S. (1997): Returns to smallholder dairying in the Kilimanjaro region, Tanzania. In:
Agricultural Economics 17: 75-87.
Meyer, H. (1890): Ostafrikanische Gletscherfahrten. Forschungsreisen im Kilimandscharo-Gebiet, Leipzig.
Miehe, S. and G. Miehe (1994): Ericaceous Forests and Heathlands in the Bale Mountains of South
Ethiopia. Ecology and Man’s Impact. Hamburg.
Misana, S. B. (1999): Community utilization and management of forest resources on Mount Kilimanjaro.
In: Mount Kilimanjaro: Land Use and Environmental Management. French Institute for Research in
Africa. IFRA Les Cahiers 16: 48-58.
Mitchell, J.F.B.; C.A. Senior; and W.J. Ingram (1989): CO2 and Climate. A missing feedback? In: Nature
330: 238-240.
Mwaipopo, O.U. (2001): Implications of accelerated sea-level rise (ASLR) and climate change for
Tanzania. In: Proceedings of Third SURVAS Expert Workshop on: “African Vulnerability &
Adaptation to Impacts of Accelerated Sea-Level Rise (ASLR),” de la Vega-Leinert, A.C., Nicholls,
R.J., Hassan, A. and El-Raey, M. (eds.) Second workshop, Cairo, Egypt — 5th-8th November 2000,
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
71
Flood Hazard Research Centre, Middlesex University, London, available on-line at
www.survas.mdx.ac.uk
Mwandosya, M., B. Nyenzi, and M. Luhanga. (1998): The Assessment of Vulnerability and Adaptation to
Climate Change Impacts in Tanzania. Centre for Energy, Environment, Science and Technology,
Dar es Salaam, Tanzania.
Newmark W. D. and Nguye, P. A. (1991): Recreational impacts of tourism along the Mrangu route in
Kilimanjaro National Park. In: Newmark, W.D. (ed.): The Conservation of Mount Kilimanjaro. The
IUCN Tropical Forest Programme: 47-52.
Nicholson, J. W. (1936): The influence of forests on climate and water supply in Kenya. In: E. Afr. agric.
J. 2: 48-53, 164-170, 226-240.
Nicholson, S. E. (2000): The nature of rainfall variability over Africa on time scales of decades to
millennia. In: Global and Planetary Change 26: 137-158.
Nicholson, S.E. and X. Yin (2001): Rainfall conditions in Equatorial East Africa during the nineteenth
century as inferred from the record of Lake Victoria. In: Climatic Change 48: 387-398.
O'Kting´ati, A. and Kessy, J. F. (1991): The farming system on Mount Kilimanjaro. In: Newmark, W. D.
(ed): The Conservation of Mount Kilimanjaro. The IUCN Tropical Forest Programme: 71-80.
Pócs, T. (1976): The role of the epiphytic vegetation in the water balance and humus production of the rain
forests of the Uluguru Mountains, East Africa. In: Boissiera 24: 499-503.
Pócs, T. (1991): The significance of lower plants in the conservation of Mount Kilimanjaro. In: Newmark,
W. D. (ed.): The Conservation of Mount Kilimanjaro. The IUCN Tropical Forest Programme: 21-34.
Ramsay, J. C. (1965): Kilimanjaro - sources of water supplies. In: Tanganyika Notes and Records 64: 92-
94.
Sarmett, J. D. and Faraji, S. A. (1991): The hydrology of Mount Kilimanjaro: an eximination of dry season
runoff and possible factors leading to its decrease. In: Newmark, W. D. (ed.): The Conservation of
Mount Kilimanjaro. The IUCN Tropical Forest Programme: 53-70.
Schmidt, P. R. (1989): 8. Early exploitation and settlement in the Usambara Mountains. In: Hamilton, A.
C., Bensted-Smith, R. (1989): Forest conservation in the East Usambara Mountains Tanzania. The
IUCN Tropical Forest Programme: 75-78.
Sjöstedt, Y. (ed) (1909): Wissenschaftliche Ergebnisse der Schwedischen Zoologischen Expedition nach
dem Kilimanjaro, dem Meru und den umgebenden Massaisteppen Deutsch-Ostafrikas 1905-1906.
Spawls, S., Howell, K., Drewes, R. and Ashe, J. (2002): A field guide to the reptiles of East Africa.
Academic Press.
Street-Perrott, F. A. and Perrott, R. A. (1990): Abrupt climate fluctuations in the tropics: the influence of
Atlantic Ocean circulation. In: Nature 343: 607-612.
Thompson, L. G. (2000a): Ice core evidence for climate change in the tropics: implications for our future.
In: Quaternary Science Reviews 19: 19-35.
COM/ENV/EPOC/DCD/DAC(2003)5/FINAL
72
Thompson. L. G. (2000b): Ice core paleoclimate study of east African Monsoon and ENSO variability
from the ice fields of Kilimanjaro. Unpubl. report.
Thompson, L. G., Mosley-Thompson, E., Davis, M. E., Henderson, K. A., Brecher, H. H., Zagorodnov, V.
S., Mashiotta, T. A., Lin, P.-N., Mikhalenko, V. N., Hardy, D. R. & Beer, J. (2002): Kilimanjaro ice
core records. Evidence of holocene climate change in tropical Africa. In: Science 298: 589-593.
Thompson, L. G., Mosley-Thompson, E., Davis, M. E., Lin, P.-N., Henderson, K. A., Cole-Dai, J., Bolzan,
J. F. and Liu, K.-B. (1995): Late glacial stage and Holocene tropical ice core records from Huacarán,
Peru. In: Science 269: 46-50.
Timberlake, L. (1986): Krisenkontinent Afrika. Hammer, Wuppertal.
Tol, R.S.J. and H. Dowlatabadi (2002): Vector-borne diseases, development, and climate change. In:
Integrated Environmental Assessment 2:173-181.
United Republic of Tanzania, Ministry of Water & CES, Consulting Engineers Salgitter GmbH (2002):
Hydrometic report of the Hai district water supply Phase IV (2002).
United Republic of Tanzania (2003): Initial Communication under the United Nations Framework
Convention on Climate Change (UNFCCC), Dar es Salaam, available on-line:
http://unfccc.int/resource/docs/natc/tannc1.pdf.
Verschuren, D., Laird, K. R. & Cumming, B. F. (2000): Rainfall and drought in equatorial east Africa
during the past 1,100 years. In: Nature 403: 410-414.
Volkens, G. (1897): Der Kilimandscharo. Darstellung der allgemeineren Ergebnisse eines
fünfzehnmonatigen Aufenthalts im Dschaggalande, Berlin.
Walther, H.; E. Harnickell and D. Mueller-Dombois (1975): Climate-diagram maps of the individual
continents and the ecological climatic regions of the earth.
World Bank. 2002. World Bank Development Indicators. On CD Rom. World Bank, Washington, DC.
Zimmermann L., Frühauf, C. and Bernhofer, Ch. (1999): The role of interception in the water budges of
spruce stands in the eastern Ore Mountains/Germany. In: Phy. Chem. Earth 24: 809-812.