<back>

Final Report
UNEP/FAO/UNCCD
Workshop on changes in the Sahel
Nairobi
14-16 October 2003

SUMMARY

For the last four decades there has been sustained scientific interest in contemporary environmental change in the Sahel. It suffered several devastating droughts and famines between the late 1960s and early 1990s. Speculation about the climatology of these droughts is unresolved, as is speculation about the effects of land clearance on rainfall and about land degradation in this zone. However, recent findings suggest a consistent trend of increasing vegetation “greenness” in much of the region.

Time series data from satellite have been analyzed for the African Sahel to study recent trends in vegetation greenness. A strong increase in seasonal greenness was observed over large areas of the Sahel during the period 1982-1999. Preliminary studies indicate a continuation of the trend through 2003. Analyses of rainfall data indicate increasing rainfall during the same period. However, the greening trend cannot be explained solely by rainfall. While extensive, the greening is not uniform, suggesting that factors other than rainfall may be contributing to greening of some areas and not others. In addition, the resolution of the satellite data set is coarse (8 km). So, the pattern of greening that might help explain its causes may be partly obscured by the resolution of the data.

“Recovery” is not a helpful term. It means a return to conditions that existed at some assumed equilibrium point in the past. When an equilibrium system is disturbed, its tendency is to seek equilibrium. In the case of the ecology of the Sahel, disturbances might be drought or, in the case of livestock, it might be pressure from grazing animals. When that pressure is removed (i.e., the rains return or the animals move elsewhere), the system returns to equilibrium. This implies that during the period of disturbance, vegetation cover is reduced and some species may decline, others increase, and still others may invade. Recovery, in this sense, would be a return to initial species composition and cover. However, there are other models of non-equilibrium that argue that, rather than a single equilibrium point, there may be multiple equilibrium points. Thus, in our simple example, the state at which the original array of species and cover has been reduced and invaders increased may represent a new equilibrium point because the relationship among species or nature of the soil has changed because of erosion. Yet, it may have a “greenness” value that is similar to the original state because overall plant cover may be the same although species composition has changed.

Human livelihood systems might be viewed to operate in a similar manner. Using a similar equilibrium model, a given household or community might enjoy some level of income (agricultural production) and command a set of resources (e.g., livestock; farm implements) in their livelihood system at some point in the past. When this livelihood system is disturbed (e.g., drought) there appear to be two types of trajectories that households might follow. In the common degradation model, households must draw down their resources (i.e., sell their livestock and farm implements), seek employment, or migrate. In the equilibrium model, it is assumed that when rains return, opportunities for agricultural production improve and households then may “recover.” However, because they have a much-diminished productive capacity at this new equilibrium point, it is extremely difficult for them to move to a new, higher equilibrium point, even if rains remain good.

An alternative adaptation model has emerged to counter the degradation model. Limited ground studies have shown that some farmers and communities have adapted to the changes during the droughts experienced in the Sahel and improved management of water and soil fertility that increase production. As these change increase returns, farmers invest in more inputs, livestock, and crop diversification. This suggests that they are moving toward a more positive equilibrium point than in the degradation trajectory. While promising in that it shows that success can be achieved with and without outside intervention, these success stories serve a broader purpose by suggesting model strategies that might be pursued.

Altogether, these findings suggest continued caution in interpreting the greening phenomenon, particularly with respect to how it might influence policy and any actions that might be taken in the near future. While it may be true that climatic conditions have improved, it is not possible to predict how long this may continue. It is certain, though, that drought will return and that policy should be flexible enough to accommodate that certainty and the non-equilibrium conditions that accompany it.

As suggested above, there are other positive developments that come from long term environmental and agricultural studies in the region that can provide a new narrative to guide efforts to stabilize and improve agriculture and natural resource management in the region.

Several studies on long-term environmental and agricultural change in the Sahel (in Niger, Nigeria, Burkina Faso and Senegal). These studies have found evidence of significant transitions from degradational land use trajectories to more sustainable and productive production systems. These include increases in cereal yields, higher densities of trees, improved soil fertility management, locally higher groundwater tables, reductions in rural poverty, and decreased outmigration. These changes coincided with growth in rural populations and introduction of structural adjustment policies.

While limited in their extent, it is likely that more such ‘success stories’ will be found in the Sahel, which collectively challenge the narrative of Sahelian degradation which has informed policy for the last few decades. These success stories could provide the basis for a new narrative that is based on the ability of farmers and livestock producers to manage their livelihoods under non-equilibrium conditions of variable rainfall, finite land resources and low bioproductivity. The biophysical and management changes that have been identified suggest that they are less the helpless victims of environmental change than agents who try to make the best use of productive and investment opportunities.

 

RECOMMENDATIONS

1. Confirmation and exploration of greening

Initiate further interrogation of the data used to detect the greening trend. This would include:

• updating the time series,
• quality assessment and correction,
• further analysis of patterns at varying spatial and temporal scales,
• derivation of biophysical information

2. Assessment of recovery

Initiate validation campaign comparable to those associated with other remote sensing observations (e.g., TRMM) to establish systematically (a) what greenness means on the ground, (b) how changes relate to human well-being and environment.

• Identification of areas of change,
• Select sample study sites from larger area,
• Characterize samples,
• Establish degrees of change,
• Establish causative factors for observed changes.

3. Assessment of apparent anomalies

While there is evidence that the conditions for farmers and livestock producers have declined in terms of production and soil fertility, other ground studies have documented success stories of farmers and livestock producers who have adapted successfully to conditions in the Sahel. These should serve as lessons that could guide future policy and intervention strategies that might lead to a more positive perspective on options for renewing agriculture in the Sahel. This would be achieved through a studied that was focused on:

• The policy lessons drawn from the existing studies should be applied to other areas and scaled up to macro-policy level.
• More policy-relevant studies should be undertaken on ‘success stories’ of agricultural and natural resource management in order to draw lessons from a wider range of environments, and to identify constraints to upscaling.
• These studies could be linked to a review of land rehabilitation activities across the Sahel (approaches, technologies, impact, upscaling potential).

Ultimately, this should lead to policies which:

• Through appropriate policy instruments, incentives should be provided for innovation and investment in natural resource management (for example, ensuring security in land and tree tenure, information diffusion).
• Provide the widest possible range of technical and livelihood options for rural families.
• Promote the use of institutional channels (existing or adapted) for locally based management of natural resources, especially common pool resources.

4. Data continuity and exploitation

Data will continue to be a critical issue, in terms of availability and access (i.e., price), but is essential for monitoring and assessment. To continue and enhance the flow of data:

• Encourage national and regional climate data sharing with partners
• Invest in data mining of historical and systematic gathering of socio-economic data to provide multi-scale picture of interactions between climate and changes
• Strengthen climate observation network

Final Report
UNEP/FAO/UNCCD
Workshop on changes in the Sahel
Nairobi
14-16 October 2003

1.0 Background

The Sahel region of Africa is a dynamic ecosystem that responds to climatic variability and human exploitation of biospheric resources. Over the long-term, changes in rainfall may have resulted in changes in land use patterns. While there has been a tendency to refer to the desertification of the Sahel, results from analysis of different types of satellite and ground data have not resulted in consensus on the direction of changes.

Since the early 1980s satellite mapping of the global land biosphere has generated long time-series measurement of vegetation that can be used as a proxy for understating the dynamics of variability of the Sahel. A number of studies using these and other data have shown the close coupling between rainfall and primary production in the Sahel.

Therefore, United Nations Environment Programme (UNEP), the United Nations Food and Agriculture Organization (FAO) and the Convention to Combat Desertification (CCD) jointly organised a Changes in the Sahel workshop, from 14 to 16 October 2003 in Nairobi, Kenya, to bring together scientists (Attachment 1) who have conducted significant studies in this region to synthesize results over the last 20 years.

 

2.0 Objectives

The workshop is intended to produce a statement of the current state of scientific knowledge of changes in the Sahel.

Specifically, participants will:

• Present results of their recent work addressing the current state of processes in the Sahel;

• Produce conclusions drawn from these results;

• Based on these conclusions, provide recommendations of future actions to the sponsoring organizations;

• Prepare and present a summary of the workshop findings at the conclusion of the workshop.

3.0 Approach

As reflected in the Workshop Agenda (Attachment 2), individual papers were presented by the participants, followed by group discussions of the presentations. From these presentations and ensuing discussions conclusions were drawn and agreed upon by the group. Actions to address these conclusions were then agreed upon and are presented in the recommendations section of this report. Workshop results were presented at the conclusion of the workshop.

 

4.0 Results and Discussion

4.1 Evolving contexts of the desertification debate

A great many debates have grown up around the notion of desertification as a process of degradation that affects the arid, semiarid and subhumid zones of the globe. A fundamental and continuing debate has been over whether desertification actually exists and, if so, how it might be defined, measured and assessed. Rather than simply review the evolution of these debates we examine the contexts in which they take place and how those contexts have contributed to the evolution of our understanding of the intertwined processes that contribute to desertification. The fact that these “contexts” have changed over time, and that some of these contexts are often ignored have helped sustain debate. We consider four “contexts” that frame much of the debate and consider what impact each has had: (1) changes in our understanding of climate variability; (2) changes in our understanding of vegetation responses to perturbation; (3) changes in our understanding of social processes, including household responses to economic perturbation; and (4) changes in our understanding of desertification as a political process or artifact.

4.2 Policy implications of the Sahelian recovery

The Sahelian greening is one phase in a continuous sequence of wet and dry periods. Here, we examine policies in the light of these changes. We consider policies involved both in other greenings and in the Sahel itself. Others include Sinai, southern Russia, Arizona and the Loess Plateau in northern China. In Sinai, the policies that brought about greening include exclusion, emigration and nature conservation. In Russia, the policy that had impact was the collapse of the Soviet collective farm system. In Arizona, it came after the Taylor Grazing Act of 1934 which allowed the US Forest Service to manage grazing. On the Loess Plateau, it has been the result of irrigation, tree-planting, the extension of responsibility of land users and perhaps aerial seeding and fencing. Most of these policies are inapplicable to the Sahel, with the possible exception of some of the Chinese measures.

In the Sahel, some policies in Burkina Faso, and elsewhere did enable soil and water conservation projects, but the success of these projects is more due to the harnessing of local skills. More generally, policy has had little to do with the greening (as with the earlier browning), even though the Sahel has been a seething cauldron of policy debate for three decades.

Sahelian environmental policies face two main challenges: fast and occasionally extreme fluctuations in rainfall; and changes in the scientific dialectic. There should be less emphasis on the negative policy of controlling land degradation, and more on positive policies that encourage adaptation and allow for rapid deployment to make use of good years and insurance policies against bad ones.

4.3 Greening of the Sahel

For the last four decades there has been sustained scientific interest in contemporary environmental change in the Sahel. It suffered several devastating droughts and famines between the late 1960s and early 1990s. Speculation about the climatology of these droughts is unresolved, as is speculation about the effects of land clearance on rainfall and about land degradation in this zone. However, recent findings suggest a consistent trend of increasing vegetation greenness in much of the region. It is not possible to explain the vegetation trend by rainfall only. There are other possible causes of this trend such as land use change, migration and armed conflicts. There are also policy implications of a positive trend in biophysical conditions. One conclusion is that more site-specific information on the interaction of biophysical dynamics (climate, soils, vegetation) and farming systems (farming practices and risk management strategies) is needed in order to better understand and ultimately support the development challenge in the Sahel.

4.4 The impact of soil and water conservation on agriculture and environment on the northern part of the Central Plateau of Burkina Faso between 1980 and 2001

In the beginning of the 1980s the situation on the northern part of the densely populated Central Plateau was dramatic: drought years succeeded each other, food shortages at household level were endemic and the environment had become degraded. The commonly held view is that the process of environmental degradation continued in the second half of the 1980s as well as in the 1990s.

A study was recently undertaken by a multidisciplinary team of national researchers, who looked at long-term changes in agriculture and environment in this region. Most of their research findings show positive trends. Cereal yields have increased by about 50% since 1984-88; in two of the three provinces studied the cultivated area remained stable during the last 15 years; tree density and species diversity are higher on fields treated with soil and water conservation than on untreated fields; livestock numbers have increased and livestock management is evolving from extensive to semi-intensive and a survey in 59 villages shows that, according to the villagers, local groundwater levels have improved substantially since the start of soil and water conservation. Based on criteria used by villagers, which are mainly related to levels of household food security, rural poverty seems to have decreased significantly (up to 50%) in villages with soil and water conservation.

Much has been achieved, but much more remains to be done. It is urgent to improve soil fertility on cultivated land and to fight against land degradation on uncultivated land, which continues unabated.

4.5 Twenty-three years of Sahelian vegetation dynamics from NOAA-AVHRR

Satellite measurements of the global biosphere in the form of the normalized difference vegetation index (NDVI) have generated a 23 year time series appropriate for the studies seasonal to interannual vegetation dynamics of the Sahel region. The close coupling between Sahelian rainfall and the green-up of vegetation has made it possible to utilize this vegetation index data set as a proxy for the land surface response to climate variability. Examination of the this time series principally reveals two major periods: (a) 1982-1993 marked by below average vegetation and persistence of drought with a signature large scale drought during the 1983-1985 period; and (2) 1994-2003, marked by a trend towards “greener” conditions with region-wide above normal vegetation conditions in 1994. Spatial patterns enable us to conclude that is not a footprint of desertification, rather they indicated the variability of green vegetation biomass over the region in response to interannual variations in rainfall. Systematic studies of changes on the landscape at local scales using high spatial resolution satellite data sets such as those from LANDSAT, SPOT and MODIS will allow for an improved documentation of degradation of Sahelian land resources that could lead to desertification.

4.6 Do farmers’ long-term responses to economic and environmental change support an hypothesis of Sahelian desertification?

The term ‘desertification’ can be considered in terms relevant to small farmers’ practices and strategies, and the expectations generated by an hypothesis of Sahelian desertification reviewed. Long-term data sets (1960-2000) collected at the district level in three Sahelian countries (Diourbel Region, Senegal; Maradi Department, Niger; and the Kano region, northern Nigeria), together with village-level field enquiries, conducted in 1999-2000, these data were used to construct profiles of change in a range of relevant variables. These profiles were then synthesized to obtain lessons for policy, which was one of the principal drivers of farmers’ responses in each of the districts. A simple theory of ‘desertification’ is found inadequate for understanding the complexity, diversity and flexibility of farmers’ responses to change. While questions of sustainability arise in relation some variables (e.g., fertility management under cultivation), some sub-hypotheses of desertification are found to be in need of re-thinking in relation to others (e.g., tree management). Natural resource management as a whole cannot be understood except in terms of livelihood strategies based on diverse portfolios of on- and off-farm income sources, and interaction between Sahelian and sub-humid or humid agro-ecological zones and cities.

4.7 UNCCD perspectives on the changes in the Sahel

The plight of the Sahel was brought into the international arena by the reports of the debilitating drought of the early seventies and its attendant massive loss of life and property. The international community made attempts to address the issues of drought and desertification by the establishment by UNCOD in 1977, but the resulting Plan of Action fell short of expectation.

The UNCED process and the resultant Chapter 12 of Agenda 21 led to further consideration, but there was still a need for a legally binding instrument. The birth of the UNCCD, a Convention to address the fight against drought and desertification the world over was the vehicle. This was a call for paradigm shift, a different way of dealing with the twin problems of drought and desertification.

The “United Nations Convention to Combat Desertification in Those Countries Experiencing Drought and/or Desertification, Particularly in Africa,” with its entry into force in December of 1996, has 190 signatories to date. The thrust of the UNCCD is for concerted local level action, but with international level support and partnership. Its objective is:

“to combat desertification and mitigate the effects of drought…..through effective action at all levels, supported by international co-operation and partnership arrangements, in the framework of an integrated approach which is consistent with Agenda 21, with a view to contributing to the achievement of sustainable development in affected areas.”

The UNCCD recognizes that achieving this objective will involve long-term integrated strategies that focus simultaneously, on improved productivity of the land and the rehabilitation, conservation, and sustainable management of land and water resources, leading to improved living conditions, in particular at the community level.
The Sahel is a major entry point in understanding and eventually addressing the complex problems of recurrent droughts and the ever-present desertification and land degradation. The particular attention given to Africa by the UNCCD includes the Sahel, as testimony to the need to take concrete action in that region.

Various interacting forces come into the picture, in influencing not only natural processes, but also possible remedial measures. Traditional knowledge and coping strategies of the affected peoples, early warning systems, provide some of the tools required to address the challenges that imposed by the often adverse climatic factors, ecological factors, anthropogenic factors encountered in the region. These are all important components to be taken into consideration when looking at the “Changes in the Sahel”. The combined action of these at local level, national level as well as appropriate international interventions in one way or another hold the key.

4.8 Monitoring vegetation growth and mapping changes in landscape: Senegal case study

A lack of early warning system in the early 1970s prevented governments and even researchers from having a good appreciation of the droughts that occurred, in semi-arid zones, particularly in 1972.

Since this period, developing tools to monitor vegetation growth and to estimate available forage for cattle has been a major concern for decision makers. The Centre de Suivi Ecologique (CSE) has developed various products to address these needs. Decadal NDVI information and annual biomass maps using NOAA/AVHRR and SPOT/Vegetation data are some of the products successfully elaborated and regularly published since 1987. More recently, some other indicators like VCI and SPI have been successfully applied to drought monitoring in Senegal.

The process has allowed CSE to set up a NOAA database of well-calibrated images. These data have been used to look at the tendency of vegetation growth at a national level. At a more localized level, several studies based on satellite images and aerial photos from the 1960’s until 2000 have been conducted to map changes in land use and land cover. CSE is still continuing to work on these kinds of studies with FAO, UNEP and GEF in the framework of the Land Degradation Assessment (LADA) project. It is aiming to identify hot and bright spots and to document their causes using mapping and census data.
Early warning systems are now used in some Sahelian countries on a more operational basis and can be expanded to all Sahelian countries. Detection of changes in the Sahel landscape can be done at large and localized scales (access to high resolution data still needs to be improved). Generalization on very large areas should be done carefully.

4.9 Long-term precipitation variability in the Sahel

The Sahel has undergone tremendous fluctuations of rainfall throughout historical times. Extreme and prolonged droughts are an inherent feature of the environment. The fluctuations have been particularly extreme during the last half of the 20th century. The mean rainfall for 30-year periods, the traditional time period for a climatic “normal”, decreased 25% to 40% in the Sahel between 1931 – 1960 and 1968 – 1997. the contrast is even greater when the wettest and driest decades, the 1950s and 1980s, are compared. During those decades, the entire continent was affected. That clearly demonstrates that the main causes of rainfall variability are to be found in the large-scale general atmospheric circulation. These in turn are at least partially driven by sea-surface temperature variability, however El Nino/La Nina does not play a large role.
Most of the change takes place in August. Wet/dry conditions in the Sahel tend to be associated with a northward/southward displacement of the rainbelt over West Africa. In some cases, however, the reduction in rainfall is associated with an overall weakening of the tropical rainbelt. There is little relationship between the amount of rainfall during the season and length of the season or its onset date. These facts have strong implications for both adaptive strategies and predictability.
The persistent dry conditions prevailed from the late 1960s through the mid-1990s. TRMM satellite data that was validated with a dense gauge network showed that the region became markedly wetter in 1998. Relatively good rainfall continued through the next year. As a whole rainfall of the last 6 years has been better than average, but the “wet” conditions of the 1950s were not matched.

4.10 Regional variability, local relative degradation: How to manage scale

Desertification has to be clearly distinguished from desert encroachment, which is linked with the romantic idea of the desert encroaching irreversibly upon green areas. Different thresholds of land and soils degradation have been used to assess the loss of natural resources, and differ considerably according to the idea of irreversibility as the ultimate stage of degradation (which is desertification on a 25 years basis). We have no evidence to say that the drought period has ended, nor can we say that there is a global trend towards a drier climate since the last century. So that the greening Sahel, following the yellowing Sahel of the 70’s, is just an expression of climate variability: the desert did not encroach upon the arid Sub-Saharan regions.

Whether land use changes (towards fewer range pastures, more fields and fallow fields, less fallow period, etc) are responsible or not for soil and land degradation cannot be assessed just through a global survey at regional scale: degradation (loss of soil material, decreasing in production) depends upon the way populations are using their space and resources (human density, technology, etc.).

Examples from north Saharan countries, where animal pressure increased during the drought period due to national actions taken to mitigate the effect of the drought, show that we must differentiate between land degradation and/or desertification (which correspond to the loss of production capability) and resources degradation, which expresses a decreasing value for human use.

Combining bio-physical and socio-economical assessment and monitoring, functional models at a local scale and structural monitoring at the national or regional scale, could help understanding the trends in desertification. These are the objectives of the ROSELT/OSS program.

4.11 Environmental and land cover changes in the Sahel region: Lessons learned, challenges and priority actions

During the last three decades, the Sahel region has been confronted with various forms of environmental and ecological degradation due to climate change and anthropogenic pressures. These changes have been described as the most important that the region has ever faced. Recent studies and analysis show spatial and temporal patterns changes and variability in landscape features, tree-crop patterns, and forest cover, with severe degradation of soils and fragile ecosystems.
However, their impacts could have been mitigated or even reversed through constructive policies and concerted and collaborative efforts with a focus on priority areas where the conservation and rehabilitation of fragile lands could be the most cost effective.

Since 2002, the LADA team (a partnership among FAO, UNEP and GEF) is carrying out studies focused on the assessment of the status and trends of these changes, including their impacts on livelihoods and identification of hotspots. LADA aims to generate up-to-date information related to ecological and environmental changes, including economical, social and technical aspects, traditional knowledge and practices on land management which have occurred in drylands during these decades.

The principle objective of the LADA project is to develop methods and tools to assess and quantify the nature, extent, severity and impacts of land degradation on ecosystems, watersheds and river basins, carbon storage and biological diversity in drylands at a range of spatial and temporal scales.

The project will also build national, regional and global assessment capacities to enable the design and planning of interventions to mitigate land degradation and establish sustainable land use and management practices.

4.12 The utilization of geoinformation technology for agroenvironmental applications in Egypt

Remote sensing can provide valuable and timely information about natural resources and environmental conditions which are important for sustainable development. However, in developing countries, the utilization of such advanced technologies differs from one country to another. Egypt, as a developing country, has experience in the utilization of Earth observation satellites and aircraft remote sensing data in soil mapping as well as assessment of land degradation.

Agriculture in Egypt depends mainly on irrigation, which has been practiced since pre- historic times by Egyptians, and is still used today in similar ways. The main source of irrigation water in Egypt is the Nile River. Despite implementing a number of projects to regulate the Nile, much of its water is still not used properly with an efficiency of less than 50% and salinization as a consequence. By adopting irrigation practices that seek to match irrigation amount with different crop requirements will save a great deal of water while conserving soils and rendering them saline.

5.0 Conclusions

After presentation were made, the group identified seven issues that are associated with the changes that have been observed in the Sahel. They were framed as questions.

1. IS THE SAHEL “GREENING”?

This question and its policy implications are the primary purpose of the meeting.

Conclusion:

For the last four decades there has been sustained scientific interest in contemporary environmental change in the Sahel. It suffered several devastating droughts and famines between the late 1960s and early 1990s. Speculation about the climatology of these droughts is unresolved, as is speculation about the effects of land clearance on rainfall and about land degradation in this zone. However, recent findings suggest a consistent trend of increasing vegetation greenness in much of the region.

AVHRR NDVI data have been analyzed for the African Sahel to study recent trends in vegetation greenness. A strong increase in seasonal NDVI was observed over large areas in the Sahel during the period 1982-1999. Preliminary studies indicate a continuation of the trend through 2003.

Although strong shifts in satellite overpass times have led to shifting solar zenith angles (SZA) over the time period, only minimal influence of SZA’s on the Pathfinder NDVI has been found in the data. Meticulous quality assessment of other parameters related to the data, strongly supports the conclusion that the observed trend is a real change on the land surface.

Analyses of rainfall data indicate increasing rainfall during the same period. However, the greening trend cannot be explained solely by rainfall.

Implications:

“Getting it wrong” will have adverse impacts on the development future of local and national populations in the Sahel, as well as the institutions designed to deal with problems of dryland management everywhere.

2. IS THIS RECOVERY?

“Recovery” is not a helpful term. It means a return to conditions that existed at some assumed equilibrium point in the past. When an equilibrium system is disturbed, its tendency is to seek equilibrium. In the case of the ecology of the Sahel, disturbances might be drought or, in the case of livestock, it might be pressure from grazing animals. When that pressure is removed (i.e., the rains return or the animals move elsewhere), the system returns to equilibrium. This implies that during the period of disturbance, vegetation cover is reduced and some species may decline, others increase, and still others may invade. Recovery, in this sense, would be a return to initial species composition and cover. However, there are other models of non-equilibrium that argue that, rather than a single equilibrium point, there may be multiple equilibrium points. Thus, in our simple example, the state at which the original array of species and cover has been reduced and invaders increased may represent a new equilibrium point because the relationship among species or nature of the soil has changed because of erosion. Yet, it may have a “greenness” value that is similar to the original state because overall plant cover may be the same although species composition has changed.

Human livelihood systems might be viewed to operate in a similar manner. Using a similar equilibrium model, a given household or community might enjoy some level of income (agricultural production) and command a set of resources (e.g., livestock; farm implements) in their livelihood system at some equilibrium point in the past. When this livelihood system is disturbed (e.g., drought) the income stream is disrupted. There appear to be two types of trajectories that households might follow. In the common degradation model, confronted with reduced income, households must draw down their resources (i.e., sell their livestock and farm implements), seek employment, or migrate. In the equilibrium model, it is assumed that when rains return, opportunities for agricultural production improve and households then may “recover.” However, because their labor force is reduced, and productive assets are gone, they have reached a new equilibrium point. Because they have a much-diminished productive capacity, it is extremely difficult for them to move to a new, higher equilibrium point, even if rains remain good.

An alternative adaptation model has emerged to counter the degradation model. Limited ground studies have shown that some farmers and communities have adapted to the changes experienced in the Sahel over the past 30 years (see 4.4. and 4.6). Adaptations begin with improved management of water and soil fertility. As these change increase returns, farmers invest in more inputs, livestock, and crop diversification that are all intended improve economic their position. This suggests that they are moving towared a much different – and positive – equilibrium point than in the degradation trajectory. While promising in that it shows that success can be be achieved with and without outside intervention, these success stories provide model of strategies that might be pursued.

Conclusion:

From an agricultural, ecological, and socioeconomic perspective, it is not only risky but very probably wrong to assume that increased greenness signals “recovery.” More importantly, given the range of potential outcomes that might result for any community from a general improvement of conditions in the Sahel, the concept of recovery is does not seem to be particularly useful. “Change” is a less loaded and more useful concept.

Implications:

If it is assumed that increased greenness signals recovery, policies that are intended to benefit affected populations might be scaled back or eliminated. However, rather than “recovered,” conditions on the ground may be quite different and, in fact, less favorable than they were “before the emergency.” Moreover, given the range of outcomes of the past 20 years, policies based on an equilibrium or “recovery” model may do damage to one or several groups.

3. HAS THIS RECOVERY OCCURRED EVERYWHERE?

Adding to the difficulty of determining what the greening of the Sahel might mean is its pattern and scale.

Conclusion:

While extensive, the greening is not uniform, suggesting that factors other than rainfall may be contributing to greening of some areas and not others. In addition, the resolution of the satellite data set is coarse (8 km). So, the pattern of greening that might help explain its causes may be partly obscured by the resolution of the data.

Implications:

Sorting out the patterns of green and nongreen areas is essential for making sense of the greening phenomenon.

4. WHAT OTHER UNEXPECTED CHANGES HAVE OCCURRED?

Several studies on long-term environmental and agricultural change in the Sahel (in Niger, Nigeria, Burkina Faso and Senegal) have found evidence of significant transitions from degradational land use trajectories to more sustainable and productive production systems. These include increases in cereal yields, higher densities of trees, improved soil fertility management, locally higher groundwater tables, reductions in rural poverty, and decreased out-migration. These changes coincided with growth in rural populations and introduction of structural adjustment policies.

Conclusion:

It is likely that more such ‘success stories’ will be found in the Sahel, which collectively challenge the narrative of Sahelian degradation which has informed policy for the last few decades.

Implications:

Known and currently unknown success stories offer a wide range of potential lessons that could be widely applied. In essence, they represent a wide array of successful experiments that are as yet unknown but with great potential value.

5. HOW DO THESE CHANGES REDEFINE THE CHALLENGES TO POLICY?

The changes observed in the Sahel over the past 30 years show it to be a region of extreme variability. Part of the variability is climatic, but as seen above, it is also a region in which farmers are capable of responding creatively to that variability. Policy must be adept at responding directly to the variations in climate (both positive and negative) in ways that release the creativity of farmers and livestock growers to respond to them.

Conclusion:

In some places, as in the central Plateau of Burkina Faso, central government policies have enabled greening through soil and water conservation (although the results generally exceeded expectation), but most of the post 1984 greening was independent of policy or intervention. This is partly because many policies have arisen from an unfortunate narrative about degradation, for which the empirical evidence is inconclusive. Further, they have not addressed the most important influences on crop production, or the construction of fertility (as opposed to the inherent fertility) of soils.

Implications:

Until now, policy had been informed by a narrative of degradation in the Sahel. Policy was largely ineffective, but farmers found ways to improve their conditions in managing, among other things, soil fertility. If policy can be informed by a narrative that arises from the creativity of farmers and livestock producers, they may be better positioned to deal with the inevitable return of drought.

6. HOW DO THESE CHANGES REDEFINE THE CHALLENGES FOR LAND USERS?

Farmers and livestock producers will still have to manage their livelihoods under conditions of variable rainfall, finite land resources and low bioproductivity. However, the biophysical and management changes that have been identified suggest that they are less the helpless victims of environmental change than agents who try to make the best use of productive and investment opportunities. Externally funded interventions and macro-policies have played a part in this evolution, together with local initiatives on a sustained and significant scale.

Conclusion:

Farmers have a capacity to experiment and to innovate, especially in areas with high pressure on the available resources. This capacity should be supported and enhanced, for example by means of participatory research and extension, and strengthened local institutions.

Implications:

Continued innovation in land use practices and investment in natural resources is a condition for sustainable livelihoods in the longer term.

7. ARE THE DATA WE GATHER AND ANALYZE APPROPRIATE FOR DETERMINING AND UNDERSTANDING THIS TREND?

Evidence of the greening pattern over the region has been drawn from analysis of a limited set of biophysical (greenness) and climatic data (rainfall). However, part of the greening phenomenon cannot be attributed to rainfall alone. Thus, other data may be required to understand greening and what it means. Second, rainfall data is problematic for the region in terms of availability. This may be partially overcome by relying on satellite estimates of precipitation. However, in addition to its intrinsic value for studying climate and its use in driving climate models, rainfall data are used to validate satellite estimates.

Conclusion:

Efforts must be made to continue to make rainfall data available. An over- reliance unvalidated satellite data is problematic. Other types of data (i.e., socioeconomic) should be explored to help explain changes that are observed within the Sahel.

Implications:

To have confidence in observed trends in greenness requires a convergence of evidence. It is risky to rely on a single source of information. Data from rainfall gauges is sometimes problematic, but so too is unvalidated satellite data. Similarly, while these two measures might suggest trends in biophysical conditions, those may be unrelated to conditions of human well-being.

6.0 Recommendations

1. Confirmation and exploration of greening

The data used to detect the greening trend should be interrogated further. Four main aspects can be distinguished:

• Continuation of the time series: The present time series generally available covers the period from July 1982 through 2000. The year 2000, however, is of inferior quality due to the gradual shift of SZA. It is important to make the continuation of the time series also generally available.
• Further quality assessment and correction: The time series available has not gone through state-of-the art data processing. By reprocessing the entire database using the latest processing schemes we can anticipate further removal of noise and atmospheric contamination.
• Further data analysis: There are a number of intriguing patterns that deserves further studies. The analyses so far have concentrated to the time integrated NDVI. Other characteristics may reveal patterns that can contribute to the disentangling of the biophysical meaning of the trend, such as i) the seasonal timing and amplitude of NDVI, ii) the onset and end of seasons and iii) the rate of increase and decrease of the seasons.
• Derivation of biophysical information: In order to derive biophysical information, the NDVI time series should be used as a driving force of a biophysical model. This will result in a much more meaningful, and for land management more useful, information, expressed as Gross/Net Primary Production (kg/m2)

2. Assessment of change

Initiate validation campaign comparable to those associated with other remote sensing observations (e.g., TRMM) to establish systematically (a) what greenness means on the ground, (b) how changes relate to human well-being and environment. This major effort would include:

• Identification of areas of change. All positive changes would be inventoried to establish population of targets from which to draw sample study areas. They would be identified from the (a) satellite record, and (b) areas of documented change on the ground (irrespective of corroborative satellite data). These would be compared with rainfall data to determine the degree to which changes could be attributed to climate.
• Select sample study sites from pool. Samples of change with controls (no change) would be drawn from the population for systematic study. These would form a base from which the impacts of change could be established now, but also monitored at a future time to develop a more profound understanding of change (both positive and negative).
• Characterize samples. First, samples would be studied using remote sensing (satellite data and aerial photography) and field sampling to characterize them biophysically (soil, vegetation, land use, land cover, infrastructure). In addition, historical remote sensing data would be used to establish conditions (to the degree possible) at some reference point in the past. Second, household surveys would be conducted within subsampled areas within each sample to characterize (a) their livelihood systems and (b) socioeconomic conditions and how they had changed between now and some point in the past. These, too, would serve as long-term reference points for longitudinal studies of human response to change.
• Establish degrees of change. Changes (positive and negative) will be manifest in a number of different ways that will vary from place to place. Likely biophysical measures would include changes in vegetation and land cover, infrastructure development, conservation interventions, and soil erosion. Likely socioeconomic measures might include changes in population, agricultural production, wealth, and measures of well-being (e.g., child weight-to-height ratios, morbidity and mortality, malnutrition), and general food security.
• Establish causative factors for observed change. The data collected will be analyzed to understand the reasons change occurred and the degree to which different factors influenced outcomes. Improved rainfall may be an important factor, but others such as changes in soil and water management practices are expected to be important.

3. Assessment of apparent anomalies

While there is evidence that the conditions for farmers and livestock producers have declined in terms of production and soil fertility, other ground studies have documented success stories of farmers and livestock producers who have adapted successfully to changing conditions in the Sahel. These should serve as lessons that could guide future policy and intervention strategies that might lead to a more positive perspective on options for renewing agriculture in the Sahel.. This would be achieved through a study that was focused on:

• The policy lessons drawn from the existing studies should be applied to other areas and scaled up to macro-policy level.
• More policy-relevant studies should be undertaken on ‘success stories’ of agricultural and natural resource management in order to draw lessons from a wider range of environments, and to identify constraints to upscaling.
• These studies could be linked to a review of land rehabilitation activities across the Sahel (approaches, technologies, impact, upscaling potential).

Ultimately, this should lead to policies which:

• Through appropriate policy instruments, incentives should be provided for innovation and investment in natural resource management (for example, ensuring security in land and tree tenure, information diffusion).
• Provide the widest possible range of technical and livelihood options for rural families.
• Promote the use of institutional channels (existing or adapted) for locally based management of natural resources, especially common pool resources.

4. Data continuity and exploitation

Data will continue to be a critical issue, in terms of availability and access (i.e., price), but is essential for monitoring and assessment. To continue and enhance the flow of data:

• Encourage national and regional climate data sharing with partners
• Invest in data mining of historical and systematic gathering of socio-economic data to provide multi-scale picture of interactions between climate and changes
• Strengthen climate observation network

NOTE: Relevant results from the workshop will be refereed and printed in a special issue of the Journal of Arid Environments.

ATTACHMENT 1: LIST OF PARTICIPANTS

1. Dr. Assaf Anyamba
NASA/Goddard Space Flight Center
E-mail: assaf@ltpmail.gsfc.nasa.gov

2. Dr. Charles Hutchinson
College of Agriculture and Life Sciences
University of Arizona
E-mail: chuck@Ag.arizona.edu

3. Dr. Sharon E. Nicholson
Dept. of Meteorology
Florida State University
E-mail: sen@met.fsu.edu

4. Dr. Lennart Olsson
Centre for Environmental Studies
University of Lund
lennart.olsson@miclu.lu.se

5. Dr. Andrew Warren
University College London
E-mail: a.warren@geog.ucl.ac.uk

6. Dr. Assize Toure
Centre de Suivi Ecologique (CSE)
Senegal
Email: assize@cse.sn

7. Dr. Wilbur K. Ottichilo
Regional Centre for Mapping of Resources for Development (RCMRD)
Nairobi, Kenya
E-mail wilber.ottichilo@rcmrd.org

8. Mr. Antonio di Gregorio
Africover / FAO
Nairobi, Kenya
E-mail: antonio.digregorio@africover.org

9. Mr. Jean-Marc d'Herbes
ROSELT/OSS
Montpellier CEDEX 05, France
Email: Jean-marc.Dherbes@mpl.ird.fr

10. Mr. Chris Reij
International Cooperation center
Vrije Universiteit
E-mail: cp.reij@dienst.vu.nl

11. Dr. Michael Mortimore
Drylands Research
United Kingdom
Email: mikemortimore@compuserve.com

12. Dr. Mahmoud H. Ahmed
National Authority for Remote Sensing and Space Science,
Cairo, Egypt
E-mail: mahahmed_narss@yahoo.com

13. Mr. Jim Weber
International Center for Remote Sensing of Environment (ICRSE)
E-Mail: isrse@symposia.org

14. Dr. Abdoulaye Kignaman-Soro
ACMAD
Niamey, Niger
e-mail: sd@acmad.ne

15. Mr. Mounkaila Goumandakoye
Drylands Development Center /UNDP
Nairobi, Kenya
Email: mounkaila.goumandakoye@undp.org

16. Dr. Syaka Sadio
Forest Resources Division, FAO
Rome, Italy
E-mail: syaka.sadio@fao.org

17. Mr. Ndegwa Ndiang'ui
UNCCD
Bonn, Germany
E-mail: nndiangui@unccd.int

18. Mr. Timo Maukonen,
DEWA / UNEP,
Nairobi, Kenya
E-mail: timo.maukonen@unep.org

19. Dr. Ashbindu Singh,
DEWA/RONA/UNEP
Washington, D.C. USA
E-mail: as@rona.unep.org

20. Dr. Steve Lonergan,
DEWA / UNEP
steve.lonergan@unep.org

21. Mr. Ivar Baste
DEWA / UNEP
ivar.baste@unep.org

22. Dr. Anna Tengberg,
DGEF / UNEP
anna.tengberg@unep.org

23. Dr. Mohamed Sessay
DGEF / UNEP
mohamed.sessay@unep.org

24. Mr. Jinhua Zhang
DEWA / UNEP
jinhua.zhang@unep.org

Observers:

Mr. Michael Norton-Griffiths
Nairobi, Kenya
Email: "M.Norton-Griffiths" <MNG5@compuserve.com>

Mr. Dave MacDevette
Cape Town , South Africa
Email: dmacdev@icon.co.za

ATTACHMENT 2: WORKSHOP AGENDA

UNEP / FAO / UNCCD Workshop on Changes in the Sahel

Nairobi

14-16 October 2003

Tuesday, 14 October 2003

0900 Welcome and opening statement S. Lonergan, UNEP, Kenya
0930 Introductions and Administrative Announcements T. Maukonen, UNEP, Kenya
1000 Review and Adoption of Agenda J. Weber, ICRSE, USA
1030 Coffee
1100 UNEP perspectives T. Maukonen, UNEP, Kenya
1130 FAO Perspectives S. Sadio, FAO, Italy
1200 UNCCD perspectives N. Ndiangui, UNCCD, Germany
1230 Lunch
1400 Evolving Contexts of the desertification debate C. Hutchinson, U of Arizona, USA
1430 Sahelian Rainfall S. Nicholson, Florida State U, USA
1500 23 years of Sahelian vegetation dynamics A. Anyamba, NASA/GSFC, USA
1530 Greening of the Sahel L. Olsson, U of Lund, Sweden
1600 Coffee
1630 Discussion Participants
1730 Adjourn

Wednesday, 15 October 2003

0900 Regional Variability J. d’Herbes, ROSELT, France
0930 Egypt’s perspective M.H. Ahmed, NARSSS, Egypt
1000 A new land cover classification system W. Ottichilo, A. DiGregorio, Kenya
1030 Coffee
1100 Discussion Participants
1200 Lunch
1400 Changes and vegetation growth in the Sahel A. Toure, CSE, Senegal
1430 Impacts of change in the Sahel A. Kignaman-Soro, ACMAD, Niger
1500 Impact of Soil/Water conservation on Agriculture C. Reij, ICC, Netherlands
1530 Coffee
1600 Discussion Participants
1700 Adjourn

Thursday, 16 October 2003

0900 Policy Implications of the Sahelian Recovery A. Warren, UCL, UK
0930 Farmers responses to economics and environment M. Mortimore, Drylands Research, UK
1000 Coffee
1030 Discussion Participants
1200 Lunch
1400 Drafting of preliminary report Participants
1700 Presentation of Preliminary Report Participants
1800 Adjourn

<back>