PART II. EXTRACTS AND SUMMARIES OF CONTRIBUTIONS

At the start of each week, a brief introduction to the group of themes and a set of discussion questions were sent to all participants. These background materials have been reproduced in Annex 4. The full interventions and contributions of the participants are available on the LADA conference Web site and on the CD-ROM in the back pocket of this volume. Extracts and summaries of the contributions have been systematically arranged by theme below.

Paragraphs in italics are extracts from contributions. Different extracts from one contribution may appear under more than one theme. Paragraphs in normal typeface are summaries by the papers directly relevant to the issue being discussed. The original contributions or papers are cited at the end of each summary. They can be read by clicking on the link in the electronic version of this document or by copying the web address from the paper version of this volume into a browser. Note: apparent spaces in a web addressed should be typed in as low dashes (shift-hyphen).

Theme 1. Land degradation assessment methods, indicators and a conceptual framework

Methods for land degradation assessment

Methods that are currently being used

Finally, there are a host of modelling techniques that could be mentioned as relevant to the LADA methodology, (e.g. simulation models, GIS spatial modelling, etc.)

Raul Ponce-Hernandez (see contribution under Theme 4).

The following approach is currently being tested in the Kgalagadi District, SW Botswana and builds upon work done by the Botswana Rangeland Inventory and Monitoring Programme:

Our approach focuses on participatory degradation appraisal. ... Given the desire of LADA to further facilitate the CCD, an emphasis on local participation may be particularly relevant - the CCD heavily emphasizes the importance of grass-roots desertification monitoring and response. The approach borrows from the field of Participatory Monitoring and Evaluation (see Estrella and Gaventa, 2000 for a recent review)...Our approach integrates three of the five degradation assessment methods identified by van Lynden and Kuhlmann (2002 - LADA draft): land user opinion and farm-level field criteria; field monitoring; and productivity changes. I attach a draft paper outlining the full methodological framework. Field data are currently being added by our partners at the University of Botswana. The paper is couched in terms of sustainability to widen its appeal, but essentially addresses degradation monitoring.

Mark Reed and Andy Dougill

The paper is available as an E-mail conference contribution at:

ftp://ftp.fao.org/agl/emailconf/lada/lada1_reed_mark_and_dougill_andy.doc

In Morocco an integrated assessment approach is being used. This approach emphasizes gaining a better understanding from local populations (e.g., on their experiences in coping with and implications of land degradation). Field evaluations proved to be very useful, to among other things, gain a better understanding of the impacts of land management improvement: field discussions helped indicate a key indicator of programme success - various people voiced they no longer got "sand in their couscous".

Abdelkader Allali

In Argentina the National Patagonic Center, CONICET is using aerial photography and remote sensing to assess the degree of dryland deterioration.

Jorge Ares

More information in Jorge Ares' E-mail conference contribution, at:
http://www.fao.org/landandwater/agll/lada/emailconf.stm.

Use of various methods

This means that one cannot [prioritize any one of] these methods [expert opinion, remote sensing, etc.], but rather [one needs to] consider a palette of mixtures...

Jorge Ares

Graduated and broadly based methods

It requires ... a very simple, broad-brush system. Andrew Warren

...[it requires] a graduated system in which national system is extremely broad-brushed (units distinguished and data aggregated only at the District or province level), with increasing levels of detail, and devolved responsibility with smaller and smaller [units of analysis].

Andrew Warren

[it requires] people who understand the limitations of the assessment system (well-trained people).

Andrew Warren

Need for purpose-driven methods and tools

Land degradation assessment is an exercise for a purpose. The purpose must lead and the methods are then chosen to achieve that purpose. Yes, integration is important if we are dealing with complex inter-linked issues related to livelihoods, food security or even the implementation of soil conservation projects. But specify what that purpose is first.

Michael Stocking

Whereas models provide insight and information that can be used to improve management, they do not change things by themselves: something still has to be done. A relevant question, therefore, is whether Land Degradation Models can provide information that can improve decision-making .... The answer is yes: Land Degradation Models can support farming indirectly by being a source from which guidelines, diagrams, and extension service advice can be derived, and by enabling explicit alternatives for agricultural development to be drawn up.

Vicente Espinosa Hernandez (see contribution under Theme 4).

Need for cost-effectiveness

Consider for instance that [the] cost of Remote Sensing [is] probably one (or more) order(s) of magnitude lower than Field Monitoring or Productivity Changes.... Since Cost is so critical a criterion, this would explain why attempts on biophysical assessment of dryland degradation at a regional-national-global scale that are based on Field Monitoring-Productivity Changes have been... unsuccessful [so far].

Jorge Ares

Need for models

When used correctly and of high quality, models allow us, based on relatively few data points and for relatively low costs, to provide a wealth of information. They allow us to take into account things that by their very nature are not directly measurable, for instance, because of the time or spatial scales involved. They also allow us to carry out experiments and what-if scenarios that would be unthinkable in the real world. Process-based models have one other vital benefit, which is that they are based on, and may thus reveal a degree of insight in feedback mechanisms that the current indicator frameworks do not reveal.

David Niemeijer

In our experience, simulation models have been needed to help us understand several important aspects of dryland degradation. (See Jorge Ares under E-mail conference contributions, http://www.fao.org/landandwater/agll/lada/emailconf.stm). While recognizing that models might not constitute an efficient communication tool in some circles of stakeholders, our experience is that once we understand things better by using models, we can better communicate our views to other parties.

Jorge Ares

Reluctance to use models (in general)

These underlying models are rarely articulated, and few have great certainty. If nothing else, assessors should spell out their models (their assumptions).

Andrew Warren

[a national level framework is] in danger of coming under the control of poorly informed, defensive and rigid cadres.

Andrew Warren

Proposed or possible models and methods

You might take a look at quite a comprehensive model that formally integrates economics, biophysical and agronomic data: Adams, D., R.J. Alig, J.M. Callaway, and B.A. McCarl (1994) 'Forest and Agricultural Sector Optimisation Model: model description'. Washington, DC: Environmental Protection Agency, Climate Change Division. FASOM is a multi-period nonlinear programming model whose objective criterion maximizes the discounted sum of consumer and producer surpluses, net of transport and management costs, of U.S. agriculture and forest sectors over a finite time horizon. It accounts for changes in quantities of carbon in the major carbon pool in the private timberland and cropland, and over the life-cycle of wood products. It is also interesting because it illustrates one of the problems that we often encounter in these kind of exercises, namely that the model uses regional aggregates, hence, the results will be less relevant for land use planners if they would like to know precisely where to undertake a specific activity.

Ben Sonneveld

Regarding a method to assess degradation I believe we should apply an adaptable method (principles of which can be given in a framework) [using an] expert system and fuzzy logic ....

Abbas Farshad (see E-mail conference Web site for reference)

Using model results cautiously; need for testing

I agree with Andrew Warren's reluctance concerning the use of models. Generally, the problem with models is that they will always produce output whatever you feed into them. So even a perfect model (and there are few in the field of land degradation) will deliver poor output if the input data and/or parameters are of poor quality or inappropriate. So, while a statistical or process-based model can help us fill in data gaps it can equally well conceal data gaps (by providing data that only on closer inspection will turn out to be too poor to use). This implies that when models are used for land degradation assessment we need more than an evaluation of the scientific merits of the model or a sensitivity analysis - we need, for each area in which the model is used:

1. a serious validation of the model output;

2. a proper uncertainty analysis that also takes the uncertainties of the input data and parameters into account;

3. locally relevant documentation and usage protocols; and

4. well-trained staff aware of the limitations of the model.

Validation should ideally be done with measured data of the actual field situation and local land management practices (i.e. not just validation with experimental plots or on-station trials).

David Niemeijer

In sum, as I see it, both measurement based indicators and computer models have their pros and cons and both need to be used with care and perhaps even more importantly interpreted with care.

David Niemeijer

Yet, it may be emphasized that careful testing and evaluation remain essential when models are applied to new situations.

Vicente Espinosa Hernandez (see contribution under Theme 4).

Indicators for land degradation assessment

Need for indicators

...if properly selected and used, indicators can provide a lot of important and relevant information at relatively low costs, so yes, despite the risks involved indicators are useful to evaluate land degradation.

David Niemeijer

We agree that indicators are a key tool for monitoring land degradation, as non-specialists can use them rapidly, cheaply and accurately.

Mark Reed and Andy Dougill

Indicators are important. They are our means of grasping aspects of complicated issues.

Michael Stocking

My opinion [regarding] indicators is that indicators are very important and necessary for land degradation assessment.

Vicente Espinosa Hernandez

Need to use indicators cautiously (e.g., accounting for local conditions, not assuming single universal causality, and being aware of misleading assumptions

Land degradation is very difficult to evaluate. Because of this indicators need to be treated with extreme care. They should be seen merely as guidelines, within which trained individuals are allowed very considerable latitude. This is because the same amount of physical change has different impacts depending on...: [varying] household circumstances..., risk-avoidance strategies... [etc.].

Andrew Warren

Indicators are generally selected because we think they reveal an underlying pattern that ties a cause with a problem or a problem with an effect. However, in some cases local circumstances may be so different in one respect or another that the same indicator is a reflection of very different underlying relationships than the one we think it reveals. To give an example from the village- or watershed-level indicators suggested for LADA, the biophysical indicator "number of water harvesting structures" which is mentioned in relation to erosion may be a very appropriate indicator in an area where rainfall is very low and water harvesting an effective tool to harvest water and nutrients and at the same time reduce erosion. A high number could thus be interpreted as an indication of proper response to erosion. In wetter areas the number is likely to be lower, not because people do not respond adequately to erosion, but because this is not an appropriate solution to tackle erosion in these areas. This is a fairly simple and obvious example, but I think it nicely illustrates the need to be cautious when interpreting indicators and not to assume a single universal causality relation for a particular indicator. It also shows how difficult it can be to compare indicators between regions. Some indicators are great for one region but would completely miss the point in another one.

David Niemeijer

Indicators could work in providing a realistic picture of land degradation in drylands; however, care must be taken in their selection. In doing so we must not lose sight of simplicity and relevance to the aim of the LADA methodology.

Raul Ponce-Hernandez (see contribution under Theme 4).

[An] important point to keep in mind when using indicators for land degradation is that natural endowments differ greatly from region to region, country to country and locality to locality. Many potentially appropriate bio-physical land degradation indicators have a hard time distinguishing between a naturally poor state and a human-induced poor state. This implies that ideally indicators should be considered in a historical perspective (which obviously in many cases is not possible for lack of data) and used in combination with other indicators to be able to better interpret the information we think an indicator conveys.

David Niemeijer

In the case of land degradation indicators we need to be very careful not to make misjudgements concerning the underlying relations we think an indicator reveals and not to be misled by false hits that actually point at differences in the natural state. To give an example, absolute population density by itself is meaningless as an indicator for land degradation because natural and climatic endowments vary (and thus carrying capacity) and because societies can respond at different times in different ways to increases in population density following more Boserupian or more Malthusian scenarios.

David Niemeijer

Danger of using 'simple' or a restricted number of indicators

The danger of trying to capture all aspects of land degradation through a necessarily restricted number of simple indicators is that one arrives at yet another system that can only very approximately assess degradation. Indicators often focus on single aspects of (land) degradation and neglect the various interactions (either cumulative or opposite) that take place. So "simple" and "restricted number" [are] already out of the question.

Godert van Lynden

(Alternatively) Developing too long a list of indicators

[A long] list or table of indicators is typically the outcome of a consultation process whereby people contribute what they think are essential indicators for a particular process, compartment or issue. This usually leads to too many indicators, as is also the case for the current LADA draft (which is also acknowledged in the draft itself). So in a next step, people typically try to reduce such a list by asking experts for what they consider the key indicators, or by assigning attributes to each of the indicators in terms of how cost-effective they are, how reproducible, and so forth, and then determining criteria each of the indicators should meet. While this will lead to a much reduced list of indicators it is not very satisfactory in theoretical terms, because it will always involve a certain degree of arbitrariness and generally will not really take into account the DPSIR model. Instead, I would propose that the causal relationships between the indicators be used as primary selection criteria.

David Niemeijer (see also under "On criteria to help select indicators" for more detail)

But what frustrates me most is that, when asked for indicators, soil scientists (especially) come up with long lists of standard soil variables.

Michael Stocking

Need for indicators to capture complicated issues

Our indicators must integrate. ... a good indicator does not need to know what all these processes are; it just has to capture them. Land degradation assessment must not follow the road of reductionist science in its search for indicators.

Michael Stocking

Local community input

Given the difficulty of developing relevant and feasible indicators at national and global levels (Snel and Bot), it may be relevant to prioritize local and district scale monitoring, where it can be more easily linked to grass-roots action, feeding the results of this into national institutions and policy. While the information policy makers may receive from such decentralized, grass-roots monitoring systems may be more complicated and messy, it may better capture the complexities that so often characterize land degradation debates. It may also enable policy-makers to better target policy at district levels.

Mark Reed and Andy Dougill

The Reed et al paper (Mark Reed and Andy Dougill, under E-mail conference contributions at http://www.fao.org/landandwater/agll/lada/emailconf.stm) outlines some of the advantages of identifying and selecting indicators with local communities as the first stage in degradation monitoring. With subsequent empirical validation, this approach can increase the number and quality of indicators, and improve the relevance of short-listed indicators to land-users. ... [Furthermore] we suggest that final indicator selection should be carried out firstly by local communities, who are able to evaluate their ease of use. The approach could be made more flexible by providing land users with lists of key indicators approved both by the community and field researchers, and asking them to select a minimum number of indicators from a number of categories (e.g. vegetation, soil etc.) to monitor regularly. In this way, indicator use can be adapted to the skills and assets of individual land users.

Mark Reed and Andy Dougill

Basing indicators on accurate and agreed-upon data

Indicators are often composed data, the usefulness of which has been agreed upon by a large group of scientists and decision makers. These indicators might give a good idea on land degradation when they are able to address [the dynamics and interactions with regards to agroecological, socio-economic, and cultural processes, as well as to human decision making processes, knowledge differentiation among stakeholders, and issues of perception]. Key, however, is that they are based on accurate data that are either observed or the result of reliable modelling exercises. For analytical purposes you will need the observed data that later can be aggregated to indicators. Nonetheless, it is a good idea to develop indicators that appeals to large groups of stakeholders and truly reflects the status of land degradation.

Ben Sonneveld

Criteria to help select indicators (e.g., ease of use, focusing on indicators that show causal relationships, recognizing purpose, data availability, etc.)

User-friendliness

The list of key [draft LADA biophysical] indicators... tends to focus on soils. This contrasts with indicators generated by land users in the Kalahari, which tended to focus on vegetation and livestock (Reed and Dougill, 2002). One of the reasons for this is that soil indicators tend not to be harder for non-specialists to measure. 'Ease of use' was a key evaluation criterion for indicators cited by local communities in the Kalahari, and something that is under-represented in the SMART criteria.

Mark Reed and Andy Dougill

Explaining causal relationships

Indicators should be selected because they take up key positions in the causal processes of land degradation. Indicators taking up similar positions can be evaluated on the basis of the SMART concept (highlighted in the draft) to make a proper selection when multiple indicators are available. But, the first criterion, in my view should be the position of indicators in relation to the land degradation processes leading from driving forces to pressures to changes in the state to impact and to responses. In other words, we need to borrow from the process-based models and apply process knowledge to our selection of indicators.

David Niemeijer

Data availability and purpose-driven

Which of the key indicators we should actually use also depends on the kind of study to be undertaken. If we want to do field work we preferably need end-of-chain indicators (typically state and impact indicators) that reveal the effect of multiple processes on land degradation. If we want to do a desk study we typically do not have access to data for end-of-chain indicators but instead need to look for the pressure indicators that are predictive of a wide range of processes that exert pressure on the land.

David Niemeijer

Misuse of indicators for political purposes (e.g., to obtain additional donor funds)

... we have to [be aware] that some countries prefer to use [specific] indicators because these indicators may [highlight] degradation ... and ... permit ... claim [to] financial support from international organizations.

Abdelkader Allali

DPSIR and other conceptual frameworks

Usefulness of the DPSIR

[The DPSIR framework] seems to offer opportunities for conceptually linking diverse approaches regarding dryland degradation.

Jorge Ares

I think that the...Driving forces-State-Impact-Response framework to evaluate land degradation is useful and crucial to make [a] diagnostic on...land degradation around the world.

Abdelkader Allali

... the [DPSIR] model is useful to sort out cause and effect. The advantage is that it categorizes different aspects and allows us to focus on process links.

Michael Stocking

Shortcomings of the DPSIR framework (e.g., complicated framework and linear processes)

I do (and should) know what a DPSIR framework is, but I think it is a fairly complicated approach (to an admittedly very complicated issue). Rather than making things clearer, it only complicate things, at least to me....It is often difficult to separate, for instance, driving forces from pressures or state from impact (I would say the state is a result of the impact, but apparently that is not how it is seen)....This leads indeed to confusion and ambiguities.

Godert van Lynden

The DPSIR framework represents...an often complicated, recursive dynamic process. The big challenge, however, is to formalize these relationships in a quantitative manner that can be used for analytical purposes, clearly indicating the marginal contribution of the causative factors to the land degradation process.

Ben Sonneveld

Despite the fact that indicators are nowadays often used within a PSR or DPSIR framework cantered on a causality chain, true insight in the complexity of inter-relations and feedbacks is seldom provided.

David Niemeijer

At present most indicator-based studies present the selected indicators in the form of lists or tables that conveniently categorize the indicators as D, P, S, I, or R indicators. This hints at the causal position of each of the indicators in the causality chain, as it is often called, but fails to draw attention to the actual linkages and causative patterns that relate one indicator to another.

David Niemeijer

The disadvantage [of the DPSIR] is that it [may be] promoted as an all-singing, all-dancing framework to be used at all scales and all purposes.

Michael Stocking

Jargon and the need for clarity

I don't know what a Driving Force-Pressure-State-Impact-Response framework is. Does this kind of jargon confuse anyone else?...I believe we should have quickly understood terms and concepts throughout.

Andrew Warren

Frameworks other than the DPSIR

A paper by Rapport and Singh (2002) notes that current frameworks (including the DPSIR) insufficiently recognize complex interactions (e.g., natural system's resilience to stress, time delays, and synergetic effects on ecosystems) and inappropriately establishes causation by suggesting linear causal chains. They propose various alternative frameworks for environmental assessments and reporting including the Human Ecosystem Model (HEM), ecosystem health framework, and human vulnerability to environmental change framework. According to the authors these frameworks better account for the interaction between human and ecological systems and provide information on its consequences on human well-being (since people want to know how environmental changes is affecting them).

For a full copy of this paper refer to:

ftp://ftp.fao.org/agl/emailconf/lada/lada1_singh_ashbindu.doc.

Another method for land degradation assessment that I have found useful is environmental risk assessment. It is compatible with modelling and the DPSIR indicator framework. It is doubly useful because it also forces scientists to think about risk communication and risk management.

Elisabeth Bui.

Theme 2. National land degradation assessment

Some proposed indicators and methods

Soil moisture index

An article by Smith et al. discusses the use of radar to acquire uninterrupted drought-relevant information such as on soil moisture and vegetation stress. The ability of radar to penetrate clouds and obtain data under most weather conditions and to obtain daily and nightly images (radar is not dependent on the sun), makes the use of such an active sensor system especially interesting for routine monitoring. Two types of radar sensors may be used: synthetic aperture radar (SAR) - which typically provides high spatial (e.g., for the ESR-1 and -2 SAR at 25 m for a 100x100 km area) but poor temporal resolution (e.g., for the ERS-1 and -2 SAR a 35 day repeat cycle) - and scatterometer - that conversely provides poor spatial (e.g., at 50 km for a 500x500 km area) but good temporal resolution (e.g., at a repeat cycle of 3 to 4 days). The Radar SAR can be configured for different spatial and temporal resolutions. The full article is available under

Theme 2 at ftp://ftp.fao.org/agl/agll/ladadocs/activesensor.doc.

Andries Rosema indicates that data on actual evaporation from the Energy and Water Balance Monitoring System may be used to estimate soil moisture, among other indicators.

Rosema's full Email conference contribution is available at:

ftp://ftp.fao.org/agl/emailconf/lada/lada2_rosema_andries.doc).

Soil depth and productivity

...a good indicator does not need to know what all ... processes are; it just has to capture them. ...Soil depth is [an example of an excellent indicator].

Michael Stocking (see contribution under Theme 1).

To monitor soil degradation it is important to include pH (acidification), organic carbon (fertility), bulk density (compaction), and texture. One may complement these observations depending on the degradation problems with micro-elements or heavy metals.

Michel Robert

Besides ... soil fertility ..., indicators of [soil] productivity may include the yields, their stability, the level of management and on-farm income, the herd sizes per household, and the degree of intensification, diversification and on-farm processing achieved by the land user. ... Indicators of the depletion of soil fertility ... can best be given by past and present records of the soil analyses (nutrients and soil organic matter) when available and comparable (this is rare). Otherwise the depletion of soil nutrients can be grossly assessed from the amounts exported by the crops, the amounts of nutrients brought back to the soil by crop residues, animal wastes, and chemical fertilizers (sediments from floods and sand winds as well as precipitations also play a role locally). When such balances cannot be worked out, a rough indication of under use of chemical fertilizer can be given by the average chemical fertilizer use per cultivated area or somewhat more precisely by fertilizer use per crop and per hectare. Philippe Mahler (see contribution under Theme 4).

As noted by one of the participants, other soil properties (other than soil depth) will need to be evaluated: A good description of the soil structure is vital in soil degradation studies. Often the conventional description of soil structure is not enough and must be combined with the description of other soil properties, such as porosity, surface feature and consistence.

Abbas Farshad

Organic carbon

An article by Anware et al. discusses the use of hot-water carbon (HWC) to evaluate soil quality. It is known that pools of carbon in soil (that includes microbial biomass, HWC, etc.) are sensitive indicators to capture soil stability, stress, and restoration. The usefulness of HWC, which is strongly correlated with microbial biomass, is tested as an indicator of soil quality: more specifically in detecting subtle changes of pastoral systems and the long-term impacts of cropping, market gardening, pastoral agriculture, and native vegetation. Results show that HWC is one of the most sensitive measures to grazing intensity and fertilization (of N or P) in pastoral soils and that it can be used as an integrated proxy variable for soil quality.

The full article is available under Theme 2 at ftp://ftp.fao.org/agl/agll/ladadocs/1650.pdf.

Nutrient balance

An article by Syers at al. discusses a nutrient audit model that provides national-level nutrient balance information. The model is based on coefficients for estimating nutrient outputs and inputs and FAO crop and animal production data - data that are readily and freely available. The model was used to evaluate annual nutrient (N, P and K) balances for Republic of Korea, Vietnam and China for 1961 to 1998. The article emphasizes the importance of collecting time series data: yearly fluctuations make results of evaluations that rely on "snapshot-in-time" data misleading.

The full article is available under Theme 2 at: ftp://ftp.fao.org/agl/agll/ladadocs/1641.pdf.

Erosion and sedimentation

Use of Caesium-137

Caesium-137 (Cs-137) is an artificial radionuclide with a half-life of approximately 30 years. It was released into the stratosphere by the testing of above ground thermonuclear weapons in the late 1950s and early 1960s and deposited as fallout. Due to the high affinity of Cs-137 to fine soil particles and its world-wide distribution it became a practical tracer for studying upland soil erosion and downstream sedimentation. The technique permits a quantitative assessment and detection of spatial patters of the medium-term erosion rates. Whereas existing measurement techniques like run-off plots posses important limitations in terms of temporal and spatial sampling and long term tedious efforts, the Cs-137 technique provides in a single field a comprehensive distributed pattern for parameterization and testing of erosion and sediment yield models. Moreover, it facilitates geostatistical calculations of error distribution which makes it amenable to an accurate representation through a GIS interface.

An extensive literature overview on Cs-137 applications is available at: http://www.ex.ac.uk/~yszhang/caesium/welcome.htm and a full summary under Theme 2 at

ftp://ftp.fao.org/agl/agll/ladadocs/erosedmeth.doc.

Wind erosion and sand deposition

The frequency and duration of sand winds, the amount of suspended particles in the air and the presence and extent of sand deposits (sparse or contiguous sand sheets or dunes) are the most common [wind erosion and sand deposition] indicators.

Philippe Mahler (see contribution under Theme 4).

Water erosion

... useful [indicators of water erosion] is ... information on the landform, the degree and pattern of its slopes and position in the landscape, on vegetation cover and its recent changes, erosivity of soil (texture, structure in particular the occurrence of compacted layers, sealing of pores and crusting of topsoil), land use system and practices and their recent changes and the impact of recent remedial action, if any. ... At national level, however, complex patterns of water erosion will usually limit the possibility of collecting meaningful data at this level.

Philippe Mahler (see contribution under Theme 4).

Biodiversity

While a set of biodiversity indicators is potentially very large, a smaller subset are of relevance to land degradation and required to represent genetic, species and ecosystem diversity in relation to land use and land management practices. There are two ongoing processes under the CBD of relevance to LADA which consider indicators and assessment: the Programme of Work on Dry and Sub-Humid Lands, which includes dryland, Mediterranean, arid, semi-arid, grassland and savannah ecosystems, and the Programme of work on the conservation and sustainable use of agricultural biodiversity which addresses the worldwide range of agro-ecosystems.

FAO is working with its Members to develop relevant biodiversity indicators, for example, through the State of the World's Plant Genetic Resources (1996):

(http://www.fao.org/WAICENT/FAOINFO/AGRICULT/AGP/AGPS/Pgrfa/wrlmape.htm),

the State of the World's Domesticated Animal Genetic Resources (http://dad.fao.org/), and the State of the World's Forest Resources (http://www.fao.org/forestry/fra).

In addition, OECD is in the process of developing agri-environmental indicators in OECD countries through a series of expert meetings. In assessing environmental impact of agriculture, the OECD biodiversity indicators include: crop and livestock genetic resources, wild species and non-native species as well as habitat quantity, habitat quality and their linkages, in addition to indicators of soil quality, water quality, land conservation, greenhouse gases and landscape. Other OECD agri-environmental indicators that are also of interest to LADA address the following categories: agriculture in the broader economic, social and environmental context; farm management and the environment; use of farm inputs and natural resources. Various references on this can be found under"biodiversity indicators of relevance to land degradation" at:

http://www.fao.org/landandwater/agll/lada/emailconf.stm.

Biodiversity losses may be indicated by reductions in the number of species of fauna and flora, loss of key habitats for migratory species and other vulnerable animal species, depletion of wildlands, occurrence of plant associations which indicate overgrazing...etc. While biodiversity losses also occur as a result of other forms of land degradation ..., they are usually more significant, more easily measurable and more critical when associated with the depletion of natural vegetation cover. Similarly losses of carbon in soil occur as a result of erosion but unless past records of soil carbon content of cultivated soils are available, assessments of carbon losses will be attempted only where the depletion of natural vegetation cover can be documented.

Philippe Mahler (see contribution under Theme 4).

Accumulation of salts

Indicators [on the accumulation of salts] are the presence of salts above a certain level/concentration in the soil; in the parent material; in the irrigation water; in the groundwater; in incoming runoff waters and winds from the sea and salt flats. The minimum levels above which the problem should be signalled should be discussed and agreed by specialists (they usually fluctuate during the year). Additional information on hydrology and water management as well as on the soil profiles may help to determine whether the problem is due to a natural process (e.g. salt marshes) or caused by man (e.g. saline irrigation water) or a combination of both.

Philippe Mahler (see contribution under Theme 4).

Waterlogging

Useful indicators [on waterlogging] are the regime of precipitation and that of water bodies, the importance, duration and frequency of floods as well as information on the origin of the floods .... Other important indicators are the position of the land area in the landscape ..., the presence of gley or pseudogley in the soil profile, recent changes in soil and water management (e.g. irrigation) and ... the presence of plants indicating an excess of water on or ... presence of a surface or subsoil drainage systems .... For a national assessment [on waterlogging], the focus will be placed on the identification of ... watercourses along which flooding hazards are important ...; on the sections of ... watersheds which play or may play a critical role in generating floods downstream; and on areas which have become ... waterlogged as a result of changes in land and water use ... or new infrastructure developments....

Philippe Mahler (see contribution under Theme 4).

Land use and cover

Information [to monitor soil degradation] should also include land cover and land use (management practices, inputs, outputs) ....

Michel Robert

Beside the identification of the type of natural vegetation, the main indicators are the percentage of barren areas or with limited vegetation cover as compared with areas covered with primary natural vegetation....

Philippe Mahler (see contribution under Theme 4).

FAO is currently developing sub-national data on land use that may be used in land degradation assessments to identify heterogeneous areas (see http://www.fao.org/ag/agl/agll/lclu.stm).

FAO TERRASTAT has available sub-national data (on CD) on, among other items, the ratio of land used for arable agriculture to the extent of suitable land potentially available - which may be useful to LADA (see http://www.fao.org/ag/agl/agll/terrastat/wsr.asp).

Land-Use Compatibility Index (LUCI)

[The Land-Use Compatibility Index]... explains the degree to which a land utilization type agrees with the land suitability class. Abbas Farshad Full contribution at:

ftp://ftp.fao.org/agl/emailconf/lada/lada2_farshad_abbas.doc.

Economic productivity

An article by Byerlee provides an overview of the use and constraints of Total Factor Productivity (TFP) measures to help evaluate sustainability. TFP accounts for changes in agricultural production (e.g., crops, livestock, etc.) in relation to changes in inputs (e.g., land, labour, water, fertilizers, etc.). The TFP was modified into the Total Social Factor Productivity (TSFP) which incorporates non-market inputs and outputs and externalities. While TFP measures have been widely accepted among economists as the measure of sustainable agricultural systems, there is need for modifications to these approaches and to evaluate these measures in relation to trends in resource quality.

The full article is available under Theme 2 at:

ftp://ftp.fao.org/agl/agll/ladadocs/econoprod_byerlee.doc

An article by Murgai et al. discusses the use of TFP to evaluate productivity growth and sustainability in the Punjab of India and Pakistan. Changes in agricultural production (e.g., crop and livestock product costs) and changes in inputs (e.g., land, labour, water, machinery, draught animals, fertilizers, and pesticide costs) were evaluated. TFP was estimated separately for different agro-ecological zones (defined in terms of cropping systems).

The full article is available under Theme 2 at:

ftp://ftp.fao.org/agl/agll/ladadocs/econoprod_murgai.doc.

Poverty and food insecurity

The poverty of the land user is often a major cause and effect of land degradation. It is a reflection of a series of factors which include: the adverse conditions of soils and climate for land use, a low level of land management (in particular limited capacities to undertake land protection and rehabilitation ...), limited access to land (small size of the farms and/or no secure land tenure)...etc.. Food insecurity and population density are also relevant factors to take into account in the assessment of land degradation in connection with poverty as both often force the land user to overexploit the land resources.

Philippe Mahler (see contribution under Theme 4).

Numerous countries are developing sub-national poverty maps (e.g., based on income consumption/expenditure data). The development of several of these poverty maps relies on small area estimation - a sophisticated technique in which a set of identical variables (e.g., household characteristics) are identified in both a census and survey. Upon combining census and survey data, this approach takes advantage of the wide coverage of the census and detailed evaluation of the household survey (more information on poverty mapping and initiatives at: http://www.povertymap.net).

As noted by one contributor, an evaluation of poverty should only be evaluated where a clear link between poverty and land degradation is evident (e.g., in many developing countries):

Land degradation in Australia underscores that extreme climatic variation and variability combined with inappropriate land uses and/or land management practices are really at the heart of land degradation-much more so than poverty or food insecurity. Poverty and food insecurity reflect lack of control over land use/land management; they are not causes of land degradation but are probably good Socio-economic indicators (D, P, I level) of potential land degradation in developing countries.

Elisabeth Bui

Crop yield forecasting and crop cover curve

... I think that the crop cover curve is one example of an excellent indicator. ... Historical crop yield is an indicator of productivity, which in turn is an indicator of LD status.

Michael Stocking (see contribution under Theme 1).

Andries Rosema indicates that data on actual evaporation from the Energy and Water Balance Monitoring System (EWBMS) may be used to forecast crop yields; and that data on actual evaporation from the EWBMS may also be used to estimate...a climatic moisture index.

Rosema's full Email conference contribution is available under Theme 2 at:

ftp://ftp.fao.org/agl/emailconf/lada/lada2_rosema_andries.doc.

Further information at: http://www.earlywarning.nl/.

Land pollution

Depending on the pollutant, its source, its mode of disposal, propagation, degradation or persistence and the amount discharged, [land pollution] indicators may be searched in various media (soil, water, air), or/and at different levels in the food chains or/and in the records of toxic effects on the animals or human population. At national level, the main problem areas may be identified in relation with the sources (agro-chemicals, urban solid wastes and urban effluents, ...) and mode of disposal (on the soil, in the air, in surface water).

Philippe Mahler (see contribution under Theme 4).

Other indicators

The following national and district- or watershed-level indicators were suggested by Raul Ponce-Hernandez:

National-level land degradation

• Land cover (extent, type and duration - this incorporates the temporal dimension through satellite image interpretation and band-ratio indices such as NDVI and GVI).

• Land aridity indices (computed from meteorological data)

• Land productivity changes (productivity records and statistics form national agencies can be useful here)

• Spatial distribution of land use activities

• Economic parameters, such as per capita income distribution, land tenure and value

• Demographic variables associated to changes in land productivity.

District- or watershed-level

• Soil loss (erosion) by water and wind

• Yield variations (spatial and temporal) across the watershed.

• Changes in soil cover.

• Evidence of soil chemical imbalance, contamination or toxicity etc.

Raul Ponce-Hernandez (see contribution under Theme 4).

... after going through the list of indicators proposed, I would suggest to add as indicators both at country and global levels: wild fauna (wildlife) diversity/quantity; level of technologies adoption; existence/implementation of National Plan of land management or Soil Fertility Action Plan; [and] the presence of refugee camps to be linked with the occurrence of war within a region.

Lamourdia Thiombiano (see contribution under Theme/Week 4).

Some suggested methods and approaches

Adapted GLASOD approach

I believe that GLASOD principles can be adapted for studies at [the] national level. Obviously, the larger the scale the more important... data collection [becomes] [(e.g.,]... use of remote sensing data and fieldwork). (see full contribution for more detail on GLASOD) Abbas Farshad

The methodology for country-wide assessments and above will... have to be based on a mix of local studies, generalizations and interpretations based on expert and stakeholder judgement: in other words, something not basically different from the GLASOD methodology but with a wider range and interaction of contributors and multi-level consultations, combined with a broader use of the data bases now available.

Philippe Mahler

Indicators and methods currently being used

Australia's National Land and Water Resources Audit (NLWRA) and CSIRO

Australia was recently engaged in a National Land and Water Resources Audit that established baseline data for soils and vegetation, assessed the nation-wide level of water-borne erosion, Carbon and nutrient balances, soil acidification, salinity...I think that the NLWRA could serve as a template for many other countries' land degradation assessments in terms of data required, methods for making the most from existing data, modelling, and various indicators.

Results of the audit are available at www.nlwra.gov.au.

Elisabeth Bui

The CMIS Remote Sensing and Image Integration group at CSIRO (Mathematical and Information Sciences) has been developing methods to map and monitor rangelands, forests, and agricultural areas. More specifically this includes monitoring of the spread of salinity, wind erosion occurrence, seasonal waterlogging effects, and remnant vegetation conditions.

More information at: http://www.cmis.csiro.au/rsm/casestudies/index.htm.

OECD work on agri-environmental indicators

While OECD is developing its AEIs [(Agri-Environmental Indicators)] from the broader perspective of tracking environmental trends in agriculture at national/regional scales ..., [these] approaches and indicators ... could be of use in the context of dryland agriculture. ... A number of key areas where the OECD AEIs could be of value to a wider group of countries ... with dryland conditions concern the nutrient balance, soil quality, water use and biodiversity indicators.

Kevin Parris

More detail is available in the OECD papers posted under Theme 2 at:

http://www.fao.org/landandwater/agll/lada/emailconf.stm and at the OECD Web site on agri-biodiversity indicators: http://www1.oecd.org/agr/biodiversity/index.htm.

WOCAT (World Overview of Conservation Approaches and Technologies) indicator on technology and approaches

...WOCAT [identified]... a number of key indicators ... for SWC technologies and approaches respectively. These indicators were based on grouped answers in very comprehensive WOCAT questionnaires and consisted of:

• Indicators for Technologies

  - Economic benefits/disadvantages (outputs vs. inputs or benefits vs. costs)
  - Acceptance or adoption
  - Required or available knowledge
  - On-site ecological benefits or disadvantages
  - Off-site ecological benefits or disadvantages

• Indicators for Approaches

  - Focus
  - Involvement of stakeholders
  - Effectiveness of training
  - Effectiveness of extension
  - Monitoring and Evaluation
  - Use of incentives
  - Impact on land management
  - Replicability and expansion
  - Land ownership. Godert van Lynden

World Resources Institution's goods and services indicators

The Pilot Analysis of Global Ecosystems (PAGE) is the first attempt to synthesize information from national, regional and global assessment. The PAGE studies report on the condition of the world's ecosystems in terms of their ability to deliver a number of key goods and services: food, feed and fibre, water services, biodiversity, and carbon storage. A similar goods and services approach/inventory could be very useful to LADA. The study reports on five major categories of ecosystems:

• Agro-ecosystems: http://wri.igc.org/wri/wr2000/agroecosystems_page.html

• Grassland ecosystems: http://wri.igc.org/wri/wr2000/grasslands_page.html

• Forest ecosystems: http://wri.igc.org/wri/wr2000/forests_page.html

• Freshwater systems: http://wri.igc.org/wri/wr2000/freshwater_page.htm

• Coastal and marine ecosystems: http://wri.igc.org/wri/wr2000/coast_page.html

Soil degradation assessment in Senegal

The following extract is from Mamadou Khouma on soil degradation assessment in Senegal.

In Senegal we have [mapped the] main types of soil degradation based on the following factors: erosion (wind and water), acidification and salinization. Sodication [and biodiversity] [are] recognized but not mapped... (the [latter being] most difficult to evaluate)... We have evidence of wind erosion by qualitative appraisal (particle removal and deposit). Quantitative figures are available for water erosion in term of tons of soil loss per hectare and year. Acidification assessment is based on acidity survey and delineation of acid sulphate soils, [while] salinization assessment is based on bare surface and specific vegetated area affected by soluble salt excess. Remote sensing [has proven to be] the cheapest and reliable way to assess salt-affected area.

Mamadou Khouma

More detail on land degradation assessment in "Land degradation in Senegal" at:

ftp://ftp.fao.org/agl/emailconf/lada/lada2_khouma_mamadou.doc).

Monitoring land degradation in Mexico

Vicente Espinosa Hernandez notes that the following indicators are being considered in Mexico to monitor land degradation in respectively the arid and semiarid zone and the humid and subhumid zone.

Arid and semiarid zone:

Wind erosion: fragility of loose surface (composition), wind-eroded land, surface roughness, formation of sandstone, reactivation of fixed sand dunes, instability of annual precipitation

Population pressure and intensive land use: over-reclamation of sandy surface and over-reclamation on slope land (due to overgrazing, collection of fuelwood, intensive human activity)

Humid and subhumid zone:

Water erosion: concentrated rainfall, loose weathered crust, easily erodible rock, surface erosion, and gully erosion

Population pressure and intensive land use: same as above.

More detail at: ftp://ftp.fao.org/agl/emailconf/lada/lada2_hernandez_vicente.doc.

Desertification indicators for Italy

Maurizio Sciortino describes an initiative to develop desertification indicators for Italy and the Mediterranean region. In Italy, desertification maps have been developed - in consultation with local populations - for Italy's four southern regions (Sicily, Puglia, Basilicata and Sardegna). The Medalus model was used to and map identify areas prone or sensitive to desertification. This model is based on four indices: soil quality, climate quality, vegetation, and management quality. More information on this model, indices, and other steps is available at:

ftp://ftp.fao.org/agl/emailconf/lada/lada4sciortinomaurizioattachment.doc. Information on other desertification initiatives in the European Mediterranean region can be found at a Web site that provides documents on desertification in the Mediterranean: www.desertification.it.

Diagnosis of early desertification in Argentina

In the Patagonian Monte, high resolution panchromatic images were used to detect early desertification:

... signs of early desertification... can be detected through the analysis of Fourier signatures in nadir-oriented panchromatic images digitized to resolutions in the range 0.09m< r < 0.70m (Ares et al. 2002). The procedure allows the periodic monitoring of extended geographical areas at cost-efficient rates with equipment usually available in routine range laboratories and agencies.

Jorge Ares

The full contribution is available at:

ftp://ftp.fao.org/agl/emailconf/lada/lada2_Ares_Jorge.doc.

Stabilization of gullies

In Nazareno, Minas Gerais (Brazil) a project is underway to stabilize gullies. An emphasis of the project has been the development of environmental education to help control and stabilize gullies - this has included environmental training of teachers, adoption of alternative technologies viable for small farmers, and the promotion of seed collection and seedling production.

Rogério Ferreira

Key issues in national land degradation assessment

Development of land monitoring systems

Urgent need for national land degradation assessment

The crux of the matter remains that Donors, the UN-CCD (and other conventions) and national governments are all urgently in need of comprehensive environmental monitoring systems. This implies the establishment of a measured baseline environmental status in every country. To move quickly to establish such national monitoring systems one could start off by measuring a small number of (land degradation) indicators (e.g., carbon, nutrient balances, salinity for irrigated sites, heavy metals in industrialized countries), complemented by remotely sensed data (e.g. moisture status, land cover change) and ground observations on land use (input/output analysis) and its socio-economic implications (e.g., population pressure, poverty, gender). Note that the indicators between brackets are only examples. ... [Such environmental monitoring systems] would enable to take the pulse of the environmental health of a country and in the long term will confirm or establish causes and relationships among the complex factors involved. On the other hand the results could immediately be used as diagnostics in WOCAT or other case study-based programmes to suggest possible solutions.

Freddy Nachtergaele

We should keep in mind that we will have to come up with a concrete plan of action for land degradation assessment.

Godert van Lynden

Monitoring of interest to drylands and other issues

This system would be of interest not only to drylands and land degradation but obviously to other conventions and assessments as well. The GTOS Network made also a start of this and could serve as an initial base for such an initiative.

Freddy Nachtergaele

Acknowledging constraints and limitations/retaining modest ambitions

At national level and above, there will always be a lack of sufficient factual data, (particularly time series), to cover all the aspects which have been idealistically identified in developing the proposed lists of indicators and the comprehensive methodological framework for LADA. While these are useful references, particularly for local assessments, the existing constraints and limitations in many countries -and those of time and money in the LADA project- will impose the use of simpler working procedures for national, regional and global assessments. What counts most at these levels is the production of comparative/relative assessments over large geographical zones which should enable countries to set priorities among diverse land resource areas, compare and transfer their experiences and establish appropriate policies, plans and programmes and projects.

Philippe Mahler

While our approach to, and understanding of land degradation issues get wider in scope, many countries still do not have the information required and the means to undertake systematic, comprehensive, scientifically conducted assessments with such scope.

Philippe Mahler

Routine and timely monitoring

Ideally, one would like to make land degradation assessments that are based on time series, derived from a dense network of relevant observations. Unfortunately, these data are seldom available. Yet, an effort to collect important land quality indicators at regular time intervals and at representative locations would largely benefit the accuracy of land degradation assessments.

Ben Sonneveld

Indicator Development

The need to identify key issues and questions about which indicators are identified and framed

Going through the contributions to date (abstracts) it appears to me that we may end up with an endless list of factors that each individually (or in combination) may be quite relevant for land degradation assessment. I purposely call them factors because as indicators I do not believe they will help us much further. I do not believe that the list should be reduced by deleting some or many of them until a reasonable number remains, but that these factors should be grouped into a manageable number of key issues or in other words: indicators (to me the term key indicators is a tautology). Whether we should first have a long list with all the individual and sometimes very detailed factors - as is currently already being built up - and then group these, or whether we should (have) start(ed) by identifying what the key issues for land degradation assessment are, does not really matter as long as the key issues are clear and the individual factors can be grouped according to these issues (not very original, but I refer to WOCAT again).

Godert van Lynden

... it is important to see indicators within the context of the land degradation processes we are trying to capture. Indicators should not be selected on their own individual merits alone, but primarily in relation to other indicators and the specific questions we are trying to answer. I would even want to go as far as suggesting that we first must think of a specific question we want to answer and which processes play a role in that particular issue from driving forces down to impacts. Once we have defined which processes are relevant for a particular question we can try to narrow down on particular indicators appropriate for getting a handle on each of those processes, taking into account the kind of information and data typically available at the particular scale we are dealing with.

David Niemeijer

At national, regional, global levels, a selection and some kind of hierarchy/stratification/prioritization of indicators are required. What kind of assessment results may be of interest to the users at national level and above? For what kind of decision-making and action? Can we provide the information wanted and can we convey this information in a form which will be understood, convincing and useable? How far do we need indicators to produce this information at national level?

Philippe Mahler

Selecting indicators based on national capacity

To sum-up, the use of indicators for LADA at national level and above should be placed in the perspective of both what will be required by the users at these levels and what is feasible taking into account the actual availability, reliability and comparability of data within and across the countries and the constraints of the LADA project.

Philippe Mahler

... an active approach is needed to involve national decision makers to clearly define the goals for the land degradation assessment. [Upon defining goals of the land degradation assessment], we can... see if specific data sets are available at the national scale that are relevant for the defined objective of the land degradation exercise.

Ben Sonneveld

From here on, suggested indicators and attributes must be matched against the reality of existing statistics and other data sets...

José Benites (see contribution under Theme 4).

Synergies with existing monitoring initiatives

Where monitoring systems have already been established (as in many irrigation schemes for example) an existing system of gathering data exists - of water volumes for example. This could be harnessed to additional monitoring of say, crop or soil samples. ... information on nutrient budgets, water use, finance and production can [furthermore] be reviewed ....

José Benites (see contribution under Theme 4).

Need to evaluate root causes of land degradation (e.g., policies)

The root causes ... may be discovered through analysis of the flows of the different capital (financial, social, natural) and related policies (that have created disparities of access to opportunities among different social groups and ethnicities to local and external markets, financing, technical support, policy incentives, etc.). It is important to attempt to relate data ... with the situation on the ground....

Raul Ponce-Hernandez (see contribution under Theme 4).

Indicator criteria

Adequately acknowledging local conditions

For a national assessment, a given indicator will not have the same significance in different natural or socio-economic environments of the country. In a certain sense, the blind use of indicators independently from each others would be a negation of the diversity of the ecosystems and of the multiplicity of interactions within ecosystems. Therefore one may conclude that indicators should be used with extreme caution and certainly not as a panacea for LADA at national level and higher. Trying to reduce the number of indicators to be used at this level will not help and may be misleading. One should rather use other tools in parallel.

Philippe Mahler

[Indicator selection] will partly be dictated by the specific conditions and priorities of each country with regard to land degradation and how it affects the goods and services produced in its drylands (the effect of the loss of land productivity on food security, on land and water pollution, on the socio-economic conditions of the population). Therefore by searching for a precise and unique set of indicators we may be fooling ourselves.

Freddy Nachtergaele

[In WOCAT]...for none of these indicators absolute threshold values can be given: what is high costs in one place may mean peanuts in another (compare highly mechanised and computerized farming in W. Europe and US with smallholder farmers in Africa or Asia).

Godert van Lynden

In WOCAT the indicators are only meant as guidance for users to make their own (relative) judgement of a technology or approach for a given set of conditions. The same technology may be highly successful in place A, but a complete failure in location B, for many different reasons.

Godert van Lynden

The experience of local people and the expertise of specialists of diverse disciplines and the views of diverse stakeholders should take precedence over the indicators and over the use of the DPSIR or other models in arriving at an assessment of the land degradation.

Philippe Mahler

It's important to have a broad list of indicators but we must keep in mind that land suitability... is location specific [and] cannot be dissociated from history, cultural, social and economic factors. [For example,] If we consider sandy soils and low activity clay (0.2 % organic carbon, 0.01 % nitrogen in the first 20 cm)... [most extensive] in Senegal... [but] ranked [globally] as [one of] the worst soils.... in the Senegalese context one finds that [this soil] is suitable for millet, groundnut, cowpea provided [that] they are properly managed.

Mamadou Khouma

Need to evaluate indicators in context (in combination with other indicators)

I do not think it is a good idea to try to evaluate the utility of indicators without treating them as part of a consistent set. For example, the indicator on infant mortality needs to be used in conjunction with several other indicators to be able to say that it really provides information on the Impact of land degradation and not just on the lack of a good medical infrastructure or political turmoil (read war). So, in my view indicators are as good as the way in which they are used (and in particular how they are combined with other indicators). To paraphrase Andrew Warren's paper about land degradation being contextual I would say that the utility of individual indicators is also very much contextual.

David Niemeijer

Need to integrate indicators for comprehensive assessment

The main problem for the implementation of the national LADAs resides in the fact that most countries do not have integrated, comprehensive inventories of their lands as defined for the LADA project. In most cases, separate inventories of individual components/facets of lands are available such as vegetation (e.g. forest resources assessments), soils (or soils and terrain such as SOTER), climate or AEZ, land use (such as the FAO survey farming systems of developing countries). Land degradation should not be assessed merely by using separate assessments of the degradation of these individual land components. How then can the degradation of land be assessed comprehensively at national level ...?

Philippe Mahler (see contribution in Theme 4).

... an integrated inventory of land resources is an implicit prerequisite to the assessment of land degradation.

Philippe Mahler (see contribution in Theme 4).

Cost-effectiveness

...Organic carbon content, soil moisture, nutrient balance,...[indicators,] while process-relevant from an ecological point of view, must be tested [to see if they are cost-effective]...

Jorge Ares.

Financing monitoring and indicator development

Need for long-term investment

[The development of environmental systems requires]...an initial relatively large investment and a long term monitoring effort, partly independent of immediate results or rehabilitation of degraded sites.

Freddy Nachtergaele

Alternatively developing sustainable (self-sufficient) monitoring

Unless we subsidize the monitoring of degradation...with the revenues from other activities in the national economy (and...present mainstream economic theories do not seem to accommodate this view), we must plan for these lands to "pay" for their own conservation. This means that monitoring and assessment systems should be adequate to cover extensive areas, with regular periodicity, at costs commensurate with the rent obtainable from the land.

Jorge Ares

Estimating the costs of monitoring

In the Patagonian Monte we adopted [a] standard practice...[to estimate] costs of land degradation assessment ... per km2/per year .... We estimated that our monitoring system to detect early desertification...[would require] US$3.30/km2/year..., an affordable amount considering usual profits of land utilization in [this] area. This...includes...costs such as...equipment repositioning, building rental,... image processing, flight rental, etc.

Jorge Ares

Theme 3. Local land degradation assessment

Types of possible local indicators

Some suggested farm- or village-level indicators

It is at this level that we have to rely on the farmers' knowledge base in order to select the appropriate and relevant indicators. A preliminary attempt should include:

Field indicators of soil loss:

• sheet erosion,

• rills and gullies

• tree-root and rock exposure

• soil build-up against structures

• formation of pedestals

• changes in colour

• exposure of fragipan or armour layer, etc.

Field indicators of productivity changes:

• yield declines...

• increase of spacing between plants in a farming plot

• changes in soil colour (darker to lighter)

• evidence of increases in toxicity

• increasing amounts of nutrients exported from the farming systems (nutrient mining)...

Field indicators of toxicity in soil:

• decline in soil productivity associated to presence of salts

• evidence of salty spots

• irrigation with sewage water

• patchy or very localized crop decline in the field

• proximity to heavy industry, mining, smelting or major roads

Field indicators of biological degradation of the soil:

• evidence of lack of organic residues added to the soil after harvest

• changes in soil colour (to lighter)

• increased exposure to direct solar radiation

• lighter structure and weak aggregation

Raul Ponce-Hernandez (see contribution under Theme 4).

Indicators for success or failure and sustainability

In Uganda, indicators for success or failure and sustainability were developed for different farming systems using a sustainable rural livelihoods approach (see under Methods). This approach first discussed assets of importance to the community (i.e., natural, physical, financial, human, and social) through participatory work with stakeholders. These local indicators were then compared with external indicators identified by researchers using a sustainable livelihoods framework. While local indicators tended to focus on human assets, the external or expert indicators highlighted financial and natural assets (note that few indicators of social assets were easily measurable). The approach highlights the wide range and different types of indicators selected among and between different communities and researchers. A combined set of sustainability indictors were compiled based on the local and external or expert indicators.

Sally Bunning

A table of the indicators and more detail on the methodology used can be found at:

ftp://ftp.fao.org/agl/emailconf/lada/bunning_sally_attachment.doc

Sustainability indicators

Various sustainability indicators were identified and selected by land users in the Kalahari. These indicators were then compared and integrated with scientific sustainability indicators. The indicators were categorized with respect to soil, vegetation, livestock, wild animals, and socio-economic aspects.

Mark Reed and Andy Dougill

More information in Table 2 in Reed et al. (2002):

ftp://ftp.fao.org/agl/emailconf/lada/reed_mark_attachment.doc

Soil health

A farmer should be able to evaluate the health of his soils and be able to recognize changes in soil health that occur in response to changes in land management or land use. ...Guidelines are presented on how to evaluate land quality in the field using simple techniques that require only domestic tools without needing to resort to sophisticated equipment or laboratory analyses. The list of indicators was developed on the basis of an assessment of soil health done by farmers in several field projects, especially in Central America.)

More information at:

ftp://ftp.fao.org/agl/emailconf/lada/lada3_attachment_benites_jose.htm

José Benites

Soil morphology

One important local indicator of land degradation is soil morphology, mainly in the superficial horizons (colour, texture, aggregation, porosity, biological features, etc.) .... Soil morphology indicates physical, chemical, biological soil degradation and its modifications influence ... water, air, and life .... [It] is not difficult to establish this inventory [and it]...is a very easy tool for evaluation of land health: everybody can learn to observe and to interpret soil morphology.

Alain Ruellan

More information at ftp://ftp.fao.org/agl/emailconf/lada/lada3_ruellan_alain.htm

Suggested methods for local assessment

PRA, rapid rural appraisals, and community focus groups

I believe PRA processes can be successfully used for a structured and guided assessment of land degradation and impact on livelihoods, including local perceptions and adoption of NRM practices (extent and reasoning) at local level.

Sally Bunning

When data are scarce, detailed conventional surveys and systematic censuses may be replaced by a combination of diverse expert opinions, judgements and advice, questionnaires and on-site observations and consultations of stakeholders and study of existing documents. Rapid rural appraisals can be particularly effective when combining all these methods. Data can then be largely replaced by qualitative descriptions and analyses.

Philippe Mahler

...we are initially generating indicators from local knowledge and evaluating them in community focus groups. To ensure that they are accurate, reliable and linked to degradation processes, we are testing these indicators using field-based methods. Final selection will take place on the basis of these results in consultation with local communities. If successful, there are plans to extend this approach to other districts in Botswana, in collaboration with the University of Botswana....

Mark Reed and Andy Dougill

Adapted "Sustainable livelihood analytical process

A sustainable livelihood analytical process may be adapted and used for land degradation assessment. The following is a summary of steps from David Howlett et al., 2000 (Stakeholder analysis and local identification of the success and sustainability of farming based livelihood systems) adapted for LADA.

The paper by Howlett, D., Bond, R., Woodhouse, P., Rigby, D. 2000. Working paper 5: Stakeholder Analysis and Local Identification of Indicators of the Success and Sustainability of Farming Based livelihood systems is available at:

http://les.man.ac.uk/ses/research/CAFRE/indicators/wpaper5.htm.

Sally Bunning

Manual development

We are developing indicator-based degradation assessment manuals with the Ministry of Agriculture in Botswana, for distribution to farmers in Kgalagadi District....Different approaches to the development of user-friendly manuals will be evaluated during a trial with land users prior to publication and distribution. Manuals will link assessment results to a range of management strategies to slow or reverse degradation. Strategies will be developed from existing literature and local knowledge to provide a range of options to suit land users with differing capital assets.

Mark Reed and Andy Dougill

Details of the methodology and examples of indicators elicited from Kalahari communities are given in Reed et al. (in prep. - available under Theme 1 contributions) and at:

ftp://ftp.fao.org/agl/emailconf/lada/reed_mark_attachment.doc.

USLE

Very often, a modified version of the Universal Soil Loss Equation (MUSLE, RUSLE, etc) is used for local studies where water erosion is involved. Unfortunately, this method has been misused ...(Wischmeier and Smith, 1978). A similar method (parametric) has also been used to assess wind erosion. Remote sensing data have been employed to calculate biomass, which can be used in the assessment of nutrient depletion. Examples of using USLE and Morgan models (used in a GIS environment) are given in the applications guides of ILWIS (ITC, 2001) under ILWIS at: www.itc.nl.

Abbas Farshad

Feasibility of a local focus for national assessment

Monitoring by land users themselves

We are developing an approach to local-scale assessment...[that] involves mobilizing communities to monitor degradation themselves.... Devolving degradation assessment tasks to land users sets the stage for equipping individual land users with the knowledge and tools to slow or reverse land degradation. Appropriately selected degradation indicators can enable non-specialists to monitor land degradation themselves.

Mark Reed and Andy Dougill

Use of hot spots

But local scale assessments are notoriously time-consuming and costly. ...We are developing an approach to local scale assessment in...degradation hot spots...

Mark Reed and Andy Dougill

As [PRA] would require substantial resources (human, time, funds) it would require prior determination of hot spots, bright spots and representative farming systems and degradation situations for such in-depth local analysis (i.e. a good sampling basis).

Sally Bunning

Advantages of a local focus for national assessment

Linking with land degradation intervention

We would suggest that local scale assessments link better to land degradation interventions, which are increasingly focusing on communities and land user groups. This is also consistent with calls for grass-roots monitoring and response to desertification in the UNCCD.

Mark Reed and Andy Dougill

By giving local government, land user groups or individuals a role in developing indicators and a limited choice over which indicators they use, indicators are more likely to be relevant to conditions in the local agro-ecosystem...[and] can better link to farm-level interventions to combat land degradation.

Mark Reed and Andy Dougill

If these indicators are linked to a range of potential management responses, land users should be able to respond in a timely and effective manner to degradation. ...By empowering the majority of land users to carry out these tasks themselves, ...it may be possible to facilitate informed land management responses to degradation at a grass-roots level.

Mark Reed and Andy Dougill

It encourages a targeted response to degradation, rather than broad-brush national solutions that may not be relevant to all Districts or sectors of society.

Mark Reed and Andy Dougill

Few examples link indicators to policy and institutional development at farm, community and district levels (see work by Howlett in Rigby, 2000) but [indicators] can [help in]:

• develop[ing] capacity and commitment towards more sustainable land use and allow farmers to evaluate their own practices;

• ... diagnosis of problems and improvements to farming practices and identification of appropriate research and extension activities;

• assessment and monitoring spatial and temporal sustainability of different farming systems for evaluation, planning and management

• enhancing relationships between farmer, researcher and extension agent and encouraging farmer participation incorporation of indigenous knowledge and increase adoption of improved technologies.

Sally Bunning

Cost-effective

[Local indicators] can be managed by local groups of ranchers with moderate technical guidance, employs local hand-labour and is affordable within the local productive systems, without need of external subsidies (... see... contribution to Week2).... If we compare [this to national and global assessment]..., we conclude that these techniques might not be affordable without international subsidies because of the needed technological platforms, expertise, etc. These techniques cannot detect EARLY desertification at a cost-efficient scale.

Jorge Ares

... the approach builds on and rationalizes the use of existing institutional infrastructures and services, rather than requiring investment in long-term field-based monitoring capacity. It may be a more cost-effective way to monitor wide tracts of land, building degradation monitoring capacity among local communities and linking to grass-roots strategies for tackling degradation.

Mark Reed and Andy Dougill

Improved extension services

In many countries, local scale degradation assessment is still the domain of (often external) experts, usually extension workers. Such services are often overstretched. By empowering the majority of land users to carry out these tasks themselves, extension services may be able to target their assistance to the minority of land users who are still unable to carry out these tasks (most likely the poorest in society).

Mark Reed and Andy Dougill

Early detection

At a local scale in the Monte, we propose looking at structural changes of the vegetation cover (...also see our contribution to Week 1, Ares et al 2002). This allows monitoring EARLY desertification. This technique can be multiplied at many local places and can constitute a national system of local detection of early desertification. At a national scale (a low-resolution spatial scale), only LATE (advanced or extensive) desertification processes can be monitored.

Jorge Ares

Recognizing local conditions

By giving local government, land user groups or individuals a role in developing indicators and a limited choice over which indicators they use, indicators are more likely to be relevant to conditions in the local agro-ecosystem.

Mark Reed and Andy Dougill

Local indicators of success or failure and sustainability for selected farming systems were identified in Uganda based on assets of importance to communities (natural, physical, financial, human, and social). The use of this approach highlights the wide range and different types of indicators selected by different communities.

Sally Bunning

More reliable results (field-based results)

Given the problems associated with national scale indicators, farm-level collection of data may lead to more reliable results when aggregated across a District or agro-ecological zone, despite being collected by non-specialists. Integration of data across districts could be carried out during routine visits by extension services, or through a representative sample of volunteer land users who could be visited annually to collect data. At this scale, results could be fed into local government, and from there into national government.

Mark Reed and Andy Dougill

(Conversely, less reliable results)

The suggested approach [monitoring of local indicators by land users themselves] is unlikely to provide the quality and comparability of data collected in purely field-based approaches.

Mark Reed and Andy Dougill

Some key issues regarding local indicator development

Identifying key issues and questions

... rather than trying at the outset to identify ex ante indicators for local assessments and wondering which minimum data sets are required in general and then whether such sets are available, it seems preferable to formulate the set of basic questions to be answered for each main category of hot spots....

Philippe Mahler

...indicators and methods for detailed studies should be developed problem-wise (erosion, salinity, pollution, ...) by formulating sets of specific questions to be addressed for each category of problem. According to the question raised, indicators may be selected and used in connection with the type of question raised.

Philippe Mahler

Land degradation assessment is an exercise for a purpose. The purpose must lead and the methods are then chosen to achieve that purpose. Yes, integration is important if we are dealing with complex interlinked issues related to livelihoods, food security or even the implementation of soil conservation projects. But specify what that purpose is first.

Michael Stocking (see contribution under Theme 1).

Some key issues about which indicators may be structured

Abbas Farshad provides an overview of some key land quality issues. See p. 2 in:

ftp://ftp.fao.org/agl/emailconf/lada/lada3_farshad_abbas_attachment.doc.

So I think that the crop cover curve is one example of an excellent indicator. Soil depth is another. Historical crop yield is an indicator of productivity, which in turn is an indicator of land development status.

Michael Stocking (see contribution under Theme 1).

Need for indicators to capture issues and integrate the different causal processes

... what frustrates me most is that, when asked for indicators, soil scientists (especially) come up with long lists of standard soil variables. These so-called 'indicators' are nothing of the kind. Our indicators must integrate. They are likely 'black-box' variables, such as the relationship between plant cover and erosion - why is it curvilinear, generally exponential? The answer is that there are numerous drivers, such as plant roots, or the effect of raindrop interception on infiltration .... But a good indicator does not need to know what all these processes are; it just has to capture them. Land degradation assessment must not follow the road of reductionist science in its search for indicators.

Michael Stocking (see contribution under Theme 1).

Different indicators for different levels

...the degree of expectation ...varies from one level to the other. The indicators applicable at [the local] level should be different from those applicable to other levels.

Abbas Farshad

Importance of using indicators to raise awareness (in environmental education)

I think ... the use of indicators [in] environmental education [is very important] ... to [help] achieve the restoration of degraded areas ....

Vicente Espinosa Hernandez

Issues regarding methods and models for local assessment

Need for a dialogue among and between researchers and policy makers

To me it seems more important to invest in a better and common understanding of land degradation among researchers from biophysical and socio-economic disciplines and also to promote a better understanding of the problem between researchers and policy makers.

Charles Lilin

Modelling for local assessment

Before asking how to [integrate biophysical and socio-economic data] one should ask the question if this is really a necessary step.... I doubt very much that there is a demand from the user community for such models or that it is a very useful exercise in practical terms.

... This does not mean that user interest cannot be generated, for instance policy makers would be provided with a tool that apparently tackles a situation over which they seem to have little control.

Charles Lilin

Importance of monitoring in hot and bright spots

Need for evaluation in hot spots

It should be kept in mind, however that some hot spots are critical because they combine several types of land degradation problems and therefore several sets of questions should be associated, adapted and used concurrently.

Philippe Mahler

Emphasis on assessing hot spots and bright spots

...in the framework of the LADA project, priority should be given to the monitoring of the hot spots in view of the risk involved and to that of the bright spots in order to confirm their success and get further experience in their maintenance.

The detailed studies for monitoring of the hot spots should first aim at verifying the diagnosis of the causes and impacts .... The second objective of this monitoring should be to assess the validity of the option taken for the urgent remedial and preventative action undertaken. ...

The detailed studies for monitoring the bright spots should also be established on an ad hoc basis and focusing on their specific vulnerabilities. In addition, more data should be collected in general to ascertain the costs and benefits of their establishment, the costs of their maintenance and the effectiveness of the technologies used.

Philippe Mahler

Criteria to identify hot spots

What are then the indicators or rather the criteria which characterize a hot spot? These include

• the nature, severity and speed of the land degradation process, including the frequency and dimensions of related natural disasters or of economic and/or social crises;...

• the value of the land being degraded and that of the other land areas at risk (economic, social, ecological, or cultural); this criterion is often overlooked whereas it is important for decision making;...

• the size and conditions of living of populations affected and at risk; precedence may be given to hot spots with high population density and prevalence of poverty.

Philippe Mahler

Types of classifications that may be used for hot spots

Philippe Mahler suggests classifications that could be used to categorize hot spots:

• A. lands with highly critical problems (or risk) of erosion requiring urgent attention (hot spots)

• lands with high productive value

• lands of low productive value but of high ecological, or cultural importance

• degraded lands threatening other areas of economic, social, ecological or cultural importance

• degraded lands with widespread poverty, food security problems or other major social problems

  B. Lands with highly critical problems (or risk) of salinity requiring urgent attention.

  C. D. E. F. ditto... with other major problem or risk.

Philippe Mahler

Need to assess hot spots and reverse land degradation

... by definition, a hot spot requires urgent action which cannot be overly delayed by the needs for detailed studies. Some compromise has to be found between the need to fully understand the land degradation process and the need to take immediate remedial action against existing those damages and risks which are both critical in a hot spot.

Philippe Mahler

Identification criteria and indicators for hot spots should not be limited to their problems, however, they must also include those related to their potentialities and limitations for prevention and remedial action. As part of the detailed assessment of hot spots, additional information is required to determine the feasibility of diverse options (such as those compiled by WOCAT). Of particular importance are the local capacities to implement the option, the requirements for technical, social or financial support services, the needs for human resource capacity building, and those for equipment, investment and land use controls.

Philippe Mahler

As action on hot spots is by essence urgent, decisions will have to be taken on the basis of the answers to a minimum set of basic questions, leaving other questions to be answered by monitoring activities and adjusting the decision/action taken accordingly.

Philippe Mahler

Use of bright spots to help identify successful management

The main issue is to determine how the experience gained in a bright spot can be extended to wider areas and to identify the other factors which, beside the technology (see Godert van Lynden's contribution 15/10/02), have played a role in their successful management. For example, in addition to local factors, broader policy aspects of trade, land use, tenure..etc may have had a positive determining influence on the occurrence of a bright spot and the possibility of transfer of its success to other areas.

Philippe Mahler

Data collection in hot spots

Then it is necessary to examine whether, for the specific hot spot concerned, these questions are relevant and whether they should be amended and can be answered with the use of available data or/and by collecting new data or/and by other means (consultations of documents, experts, stakeholders, available studies of similar cases, WOCAT data bases..etc.).

Philippe Mahler

Importance of local assessment for LADA

From the guidance given by the LADA project document, it is clear that local assessments may be conducted for diverse objectives and at different stages of the overall LADA project. Some of these objectives partly overlap and the same local assessment may serve several of them, while others are quite distinct and their results are meant for different kinds of users. A recapitulation of the objectives of local assessments can be presented as follows

i. to experiment locally the validity and feasibility of the methods (including indicators) to be used for national and sub-regional assessments (pilot testing in the preparatory phase)

ii. as part of the methods for national or sub-regional assessments, to analyse in more detail sample areas representative of specific problems identified in larger areas at national level or subregional level

iii. to study in detail specific sites identified as hot spots and assess the associated problems and risks

iv. to study in detail specific sites identified as bright spots and assess the effectiveness and transferability of the experience and technologies used with a view to developing further a compendium of best practices of control and prevention

v. to establish baseline studies for the selection and establishment of monitoring sites

vi. to prepare pilot action projects in priority areas.

Philippe Mahler

Theme 4. Global land degradation indicators, a drylands network, next steps

Global assessment

Importance of national assessments

It is recognized that reliable data on dry and subhumid lands biodiversity should come from national assessments. Therefore it is recommended that support to national assessments and monitoring programmes should be a primary focus, especially in dry and sub-humid lands. Systematic data collection, processing thereof and effective communication of the assessment results ought to take place on a national level.

Juliane Zeidler

Linking national assessment to global assessment

It will be important for ...national level work to be linked to existing regional and global assessments, and mechanisms and methods should be developed to foster this.

Juliane Zeidler

Sub-national global data sets of importance to global assessments

Lists of core sub-national FAO global data sets (FAOcoredatasetslist2.doc) and maps developed for the Poverty and Food Insecurity Mapping Project that may be useful to LADA (TERRASTAT I Beta Maps list.doc), as well as the FAO map of global irrigated areas can be found on the LADA conference Web site and on the CD-ROM; the irrigated areas map also at http://www.fao.org/ag/agl/aglw/aquastat/irrigationmap/index.stm. Subnational spatial global data not distributed by FAO that may also be useful to LADA include:

• protected areas by WCMC (available at http://www.wcmc.org.uk/protected_areas/);

• land cover characterization by USGS EDC (http://edcdaac.usgs.gov/glcc/glcc.html);

• population density 2000 by Landscan (available at http://www.ornl.gov/gist/landscan/LandScan_2000_Release.htm); and

• percent arable land by IFPRI (http://www.ifpri.org/pubs/books/page/maps.htm).

Need for conformity: data standards and guidelines

While recognizing the specificities of the needs of participating countries and diverse regions... a degree of congruence should be sought among the national methodologies and their products if a global assessment is to be obtained. Otherwise exchange of experience among countries will remain limited, opportunities for international cooperation in research, monitoring, rehabilitation activities and capacity building will be seriously reduced and, of course, the feasibility of making a global assessment by compiling the diverse sub-national and national assessments will be almost impossible.

Philippe Mahler

[The development of] core guidelines should include provisions for a minimum of common elements for all the national assessments as regards [to] terminology, basic definitions, classifications, presentation of results...etc while leaving sufficient flexibility for elaboration and diversification of the norms at national level. In this way, LADA would secure a minimum common denominator to all the national assessments which would allow for reviews of progress at regional and global levels and facilitate the compilation of national products into subregional, regional and global assessments, as anticipated by the LADA project.

Philippe Mahler

A wider set of criteria are required by the national and international agricultural agencies that have the responsibility of monitoring the course of development into the future.

José Benites

... I have been thinking of the utility of LCCS for LADA as a standard for assessing the land cover and land use. I find quite several 'selling points' for LCCS worth considering.

... LCCS ... is suited to assessing biological and physical aspect of the landscape ... LCCS is scale insensitive and therefore suitable for LADA's multi-level assessment needs. ... LCCS can provide easily comparable information from different ... climatic or agro-ecoregions.

Timo Maukonen

Need for flexible methodologies

If a rigid methodology is adapted, the risk is that as conditions change (as they will when going from one country to the next, or from one scale to another) the methodology can no longer deal with the factors that were not considered when developing it. ...If we take erosion risk as an example, we note that during the erosion crisis of the 1930s in the USA, the work of the American Soil Conservation Service (SCS) was well adapted to the context ... and [the USLE] model.. performed well. ... When in the 1980s an erosion crisis erupted in France, ... under pressure [the]...USLE [was adapted] to French conditions. But as historical and social conditions were very different, ... this approach could not work. ...In developing countries the land degradation situation is again very different ..., it is not the result of a pioneering agriculture (US) or of overhasty modernization (France) but rather a result of misguided (dualistic) developing policies. Here again, as in France, only negotiations and participatory approaches will work, but probably with lower expectations of immediate results given the more difficult context.

Charles Lilin

... I would like to underline ... the necessity to take into account five main considerations at global level:

1. Diversity of climatic conditions: land degradation occurs across a quite range of climatic conditions (from Sahel to sub-humid zones);

2. Diversity of causes: from structural fragility of ecosystems to human mismanagement of lands;

3. Heterogeneity of contexts in term of land degradation severity levels: ...;

4. Heterogeneity of available data and even institutional capacities and national policies relating to the awareness on land degradation, its ecological and socio-economical impacts;

5. And finally the challenge to propose a minimum set of indicators which is flexible enough to take into account the considerations above: accurate enough and relevant to catch the state of degradation and also the process (modelling) of land degradation.

Lamourdia Thiombiano

Need for synergy

Synergy with the CBD

It is recommended that further links between LADA and the dry and subhumid lands programme of the...[Convention on Biological Diversity] or a cross-cutting theme on indicators and assessments, or both, should be established.

Juliane Zeidler

More information at: http://www.biodiv.org/decisions/default.asp?lg=0&dec=V/23.

The draft framework for the development of a national set of biodiversity indicators developed by a Liaison group under the cross-cutting theme of indicators and assessments of the CBD could provide some interesting guidelines also for the development of a national core set of land degradation indicators, ....

Juliane Zeidler

The draft can be read at: http://www.biodiv.org/doc/meetings/sbstta/sbstta-03/information/sbstta-03-inf-13-en.pdf

Developing synergy with MA, Forest Resource assessment, GEO, and other initiatives

Amongst others it is recommended to closely interlink with LADA and other [Global Environmental Assessments] GEA's. The Millennium Ecosystem Assessment (MA), the FAO Forest Resource Assessment process and the Global Biodiversity Outlook (GEO) initiative are amongst those GEA's that are potentially relevant and can contribute directly to the dry and subhumid lands programme.

Juliane Zeidler

It is recognized that LADA is focusing on land degradation assessment. There are clear linkages between biodiversity, ecosystem function and land degradation. Therefore we recommend further exploring such linkages and reflecting them in the development of indicator development, subsequent assessments and monitoring programmes. The framework of the MA makes some interesting connections. These though should be brought forward in the scope of national assessments, which should preferably include much more direct biodiversity indicators, and eventually, over a longer term, include inventories and studies as described in the Global Biodiversity Assessment (UNEP, 1995).

Juliane Zeidler

The draft MA framework currently under review is available at:
http://www.millenniumassessment.org/en/about/cfreview.htm.

Linkages with desertification initiatives

We are aware [that] current work of the International Geosphere and Biosphere Programme (IGBP) and the related Land Use and Cover Change (LUCC) are currently developing desertification indicators, which will be discussed during an internal workshop in November of this year. The outcomes and recommendations from such work should be included in the LADA development process.

Juliane Zeidler

IGBP at: http://www.igbp.kva.se/cgi-bin/php/frameset.php
LUCC at: http://www.indiana.edu/~act/focus1/index.html.

Strengthening the implementation of existing initiatives

The overview of ...different kinds of land degradation [data and information] at country and regional levels generated from LADA could provide basic support to boost the implementation of the International Scheme for the Conservation and Rehabilitation of African Lands (ISCRAL).

Lamourdia Thiombiano

Indicator development

Possible use of a dusty events frequency indicator for desertification

Over the last decades, it has been recognized that dusty events frequency analysis was one of the major indicators to assess the evolution of the desertification process. ...This indicator is available at the global scale for long-term periods of observations. Its application at the national and regional level should help decision-makers estimate the trend of desertification which is, in many countries, one of the most severe impediments to poverty reduction, improvement in the quality of life of local populations, and sustainable development.

Pierre Ozer

Possible sampling units

...at global level it could be more appropriate to define [a] specific set [of indicators] for a specific global agro-ecological zone (AEZ). This will give the possibility to capture the bio-physical, socio-economical and cultural factors within a given global AEZ.

Lamourdia Thiombiano

Upscaling issues

The merit of including real household, community and local level indicators into assessment and monitoring programmes cannot be stressed enough. However, difficulties have been encountered in the upscaling of such data, as well as in the systematic collection and communication thereof. Case experiences will be discussed during a workshop of the MA, as well as mechanism and methods how to deal with linking local knowledge and global assessments. The outcomes of this meeting could be useful for the further development of LADA, as would potentially be country experiences.

Juliane Zeidler

Meeting Web site:
http://www.millenniumassessment.org/en/meetings/meeting.subglobal.2.htm

Network for drylands

Need for a network, although premature

While ... a logical follow-up, ... establishing a network already now may be premature .... Nevertheless, ... while a monitoring network is premature, a cooperative network of stakeholders for a comprehensive and balanced assessment is a must. This cooperative network may then be used subsequently to discuss and select the problems, the areas and the indicators to be monitored and to establish the monitoring sites and their linkages. Philippe Mahler

Next steps

Some proposed steps

Proposal 1: A proposed set of steps (and figure) on how LADA proceed to conduct land degradation assessment was sent to participants along with the agenda for Week 4 (Annex 5). Extracts follow.

Stage 1: User needs assessment and National stakeholder meeting

• identify key issues in political and economic terms

Stage 2: Inventory of data and capacity

• identify and collect existing data sets (review biophysical data and socio-economic data)

• identify important missing data sets (e.g., socio-economic)

• identify indicators (based on user needs)

Stage 3: Implementation

• select appropriate stratification and sampling

• develop models (e.g., that link biophysical and socio-economic data)

Stage 4: Analyse results

• determine cause, state, and effect (frame data within the DPSIR framework)

• conduct participatory local surveys in hot and bright spots

• discuss results with stakeholders

Stage 5: Disseminate results and policy recommendations

• link results with solutions and policy recommendations

• disseminate results (e.g., meetings with policy makers, brochures, etc.)

Stage 6: Monitoring and evaluation

• repeat steps above

Proposal 2: Dominique Lantieri (FAO) provided a proposal on how LADA could proceed with its land degradation assessment. Extracts follow:

1. Undertaking of a user needs assessment: ... there is a need to undertake such a survey to specify better what types of different products LADA is supposed to provide for each main category of users: e.g. decision makers and technicians working for development, scientists, civil society and NGOs, environmentalists, wide public and mass media ....

2. Definition of main common methodological steps...:

...details on the methods [are not critical] but rather a broad consensual and flexible framework [is needed]. ...:

• step 1: The collection of the existing data (biophysical, socio eco, remote sensing etc...)

• step 2: The stratification of the territory into homogeneous land systems at around 1: 500 000 ...

• step 3: The further stratification or analysis within these land systems of main land use or farming systems with identification of farming practices (essential), cropping strategies, inputs levels (human, technical and financial), land tenure and market issues, cultural behaviours, main environmental problems ...

• step 4: The pre-identification of hot spots

• step 5: Sampling strategy to perform the field survey ...

• step 6: Field survey using a range of methods such as farmers enquiries, measurements on plots, rapid visual assessment by experts ...

• step 7: Integration of results, analysis and synthesis ...

• step 8: Preparation and dissemination of LADA information products ...

• step 9: Definition of the specifications of the desertification monitoring system ...

• step 10: Consultation process on actions to be done to support the effective use of LADA products for land degradation.

3. Implementation of Pilot studies ... to test the above approach and assess some possible practical methods for various steps ....

Dominique Lantieri

The complete set of steps can be viewed at:

ftp://ftp.fao.org/agl/emailconf/lada/lada4_lantieri_dominique.doc.

Proposal 3: Philippe Mahler (consultant) provided the following "next steps" on how LADA may proceed. Extracts follow:

[Stocktaking exercise on past experiences/review of national capacity:] Rather than starting with a user survey, right away at national level, a national lead institution should initiate a preliminary stocktaking exercise with the main sources of information and experience available in the country and from international data bases. ...

[Presentation of national capacity review and stocktaking results to potential users:] The results of this preliminary stocktaking should then (and only then) be presented to and discussed with a range of potential users. Out of these discussions, a clearer picture would emerge on, not only what are the diverse gaps to be filled in the assessment of land degradation, but also the requirements in national capacity building, in overcoming communication gaps and institutional barriers, and other lessons learnt.

[Developing a consensus:] This national process of stocktaking and review of past experiences should be undertaken with a view to arriving at a consensus among national stakeholders on: a) the present state of knowledge and state of the art as regards land degradation assessment ...; b) the available knowledge of the interactions among the driving forces, pressures, impacts and responses related to the degradation problems ...; c) the major conclusions and strategic implications of this stocktaking exercise .... The output of this national review should take the form of an executive summary ... and include a descriptive account of its method of preparation....

[Discussion on methodology/needs assessment based on national review/capacity:] On this basis, a wider discussion on the methodology should take place based on the nature and range of the LADA products which are ... needed and feasible/attainable in the conditions of the country concerned. In fact, the user needs - and the LADA products which could meet these needs - cannot be examined without reference to the means available in the country, the constraints and the methodology. ...

[Development of guidelines/procedures for the national assessment:] The next steps should ...include the development of core of common principles and procedures for the national assessments by translating the methodological framework ... into guidelines. The drafts of these guidelines should be widely circulated prior to regional workshops which would refine them and develop a consensus on the ways to proceed. ....

Philippe Mahler

The full contribution can be read at:

ftp://ftp.fao.org/agl/emailconf/lada/lada4_mahler_philippe.htm.

Clearly stating objectives and not losing sight of LADA's modest ambitions

... the ambitions of LADA should remain modest: in most dryland countries, a comprehensive baseline assessment at national scale ([e.g.,] 1: 1 million) should be the main objective with an identification of the diverse hot spots and bright spots.

Philippe Mahler

... we must not lose sight of simplicity and relevance to the aim of the LADA methodology.

Raul Ponce-Hernandez

... the main objectives of the assessment of land degradation could be formulated as follows:

i. to identify the type, degree and relative extent of the main land degradation problems encountered in the country(ies);

ii. to provide basic contextual information on the main causes and impacts of the land degradation problems ... as well as on the main features of the natural and socio-economic and institutional environments which influence land use and land degradation processes.

iii. on the basis of the above, to delineate the major land degradation problem areas (LDAs) of the country(ies) and indicate the relative ... urgency of their requirements for ... protection or rehabilitation. Within the land degradation problem areas, the location and nature of the most critical sites of degradation (hot spots) should be indicated.

iv. ...to identify and characterize the land areas with no significant land degradation problem in their present state ...[and] to assess their degree of stability or vulnerability and the requirements for their maintenance.

Philippe Mahler

It was recommended that reference to the full GEF project be made. Extracts follow:

Expected results and outcomes of the full GEF project

a. Standardized methods and guidelines for dryland degradation assessment and monitoring....... It should be a multi-level system that enables the inclusion of detailed field studies into national and regional statistical frameworks

b. A baseline map of dryland land degradation at sub-regional scale. The baseline map should be based on the collection and collation of existing maps and databases...

c. Global assessment of actual dryland degradation and degradation hazards.... It will enable an initial identification of areas at greatest risk (hot spots)

d. Detailed assessment of land degradation at national level, focusing on areas at greatest risk (hot spots) and areas where degradation has been successfully reversed (bright spots): This detailed assessment will use all available methods, with emphasis on field studies and indigenous knowledge. It will cover not only the state of degradation but its causes and impacts, together with identification of remedial measures. A field database will be developed to allow monitoring of changes...

e. Analysis of the effects of land degradation areas at risk. The effects of land degradation areas at risk will be analysed in terms of the types, extent and severity of degradation, their driving factors or causes, their impacts on the environment (ecosystems, carbon emissions, international waters, etc.) and on human livelihoods, food security, poverty, migrations, etc.

f. Best practices for the control and prevention of land degradation in drylands. Best practices for the control and prevention of land degradation will be developed considering indigenous and traditional knowledge, existing information on indigenous practices (e.g. the World Overview of Conservation Approaches and Technologies, WOCAT), the findings of UNEPs program on Success Stories in Desertification Control, and results of the LADA project. The WOCAT project has established a methodology for collecting soil and water conservation experience that could be employed by the future LADA project

g. Communication and exchange of land degradation information, and promotion of its use in decision making. The information generated on land degradation in drylands will be packaged, communicated and exchanged especially for use by policy and decision-makers in various ways, such as for Policy Guidance; GEF Interventions; Priority Actions; Lessons Learned and Best Practices; Monitoring Tools; ....

Acknowledging national capacity

Rather than starting with a user survey, right away at national level, a national lead institution should initiate a preliminary stocktaking exercise with the main sources of information and experience available in the country and from international data bases. A series of ad-hoc working sessions with the national experts concerned should enable them to determine jointly what is known, what are the gaps, what, in their opinion, were the past successes, failures and shortcomings of past assessments in mobilizing political will and action and what might be undertaken in the framework of LADA. ... The results of this preliminary stocktaking should then (and only then) be presented to and discussed with a range of potential users.

Philippe Mahler

Users' needs assessment

Need for a users' needs assessment

I agree ... with [regards] to [the need for] a user assessment ... - it will [ensure that] ... information that is needed [to answer key] ... questions [is obtained].

Vicente Espinosa Hernandez

Ensure user needs assessment does not overextend LADA's scope

The survey of user needs as proposed is certainly desirable but this should not lead to a cumbersome process by which the same ground may be covered again and again without much progress. ...Of course, more potential users may be consulted, but as we may broaden the range of users to be consulted, we may also overextend the scope of LADA and end up with such multi-purpose exercise that it would be even more difficult to develop and agree on a methodology (and indicators!) and complete the assessments in a reasonable time (while land degradation is ongoing at a rapid pace in many areas).

Philippe Mahler

Use of the LADA PDF-B document to help assess users' needs

To some extent, a fairly good idea of what is needed can be obtained from reviewing the preparations that enabled the launching of the LADA project, the LADA project document and consultations held so far before and during the PDF-B phase, including visits to some countries. A number of stakeholders were involved, supported the project, advised on its preparations and were convinced on its usefulness.

Philippe Mahler

Need to be more reserved in asking user needs: potential users may expect answers from LADA

[Stakeholders involved in the PDF-B phase of LADA] now assume that LADA is on its way to be implemented. They might be surprised and perhaps disappointed if they would receive questionnaires asking them to indicate or otherwise develop proposals on what they need.

Philippe Mahler

Many potential users do not even know what can be provided by LADA and may turn back the questions and ask what LADA may have to offer and how its results could be used.

Philippe Mahler

Selecting an appropriate sampling unit

When covering the whole territory of a country, ... the difficulty is to delineate the areas where these problems occur and to show them on a map at a relatively small scale (e.g. 1:1 million).

Philippe Mahler

Some [indicators] ... are not easily aggregated .... Major watersheds and drainage basins are the most appropriate boundaries for delineating surface waters, land-terrain features and major soil categories, but to fit into the present System of National Accounts, natural resources must be aggregated along administrative boundaries to the level of countries.

José Benites

The scale problem ... cannot be answered simply. ...The following aspects need to be coherently considered to clarify the problem of scale: objective of the Land Degradation Assessment of Drylands; level of organization or scale of process; possibility of clearly defining subsystems..; possibility of validating the model ...; possibility of returning to the field ...; [and the] action plan for the rehabilitation .... ... It should be kept in mind that in order to generate corrective or preventive measures, homogeneous areas are required....

Vicente Espinosa Hernandez

Data collection

Collecting new data vs. using existing data

... it will be important to determine whether a particular question should be answered on the basis of existing (usually less detailed data) or on the basis of detailed field level data (which will often have to be collected) ...

David Niemeijer

... we ... should resist the temptation of creating ever larger data bases that in the end do not significantly contribute to the solution of the problem.

Ben Sonneveld

... there are numerous data inventories (AEZ, SOTER, WOCAT, census, household surveys) available that allow or support the analysis of land degradation. There is also a large array of accumulated scientific process knowledge that has been built up in the last decades and that can not be neglected. The challenge is to formalize relationships by combining process knowledge with statistical techniques, so that we understand and quantify the land degradation process and find the answers for the relevant policy questions. I am also sure that once the country studies get started a lot of indicators, data sets, methods and techniques will fall in place like a jigsaw puzzle.

Ben Sonneveld

The emphasis in LADA should be on design of a policy tool and not on collection of new data. There are three (or four) important reasons for this. ... To warrant land degradation a place on political agendas we need hard evidence that people will benefit from investments in land degradation and this should be substantiated with decision making tools.

Ben Sonneveld

Collecting data for selected indicators

... [it should] be made more explicit, that ... data [are collected] only for those indicators that [have been] selected.

David Niemeijer

Need to collect data over time

Land degradation acts in space and time and it is therefore [important] to establish a georeferenced monitoring system that [can provide better information] about the actual state of land degradation [and] ... the evolution over time, and [enables] evaluation of the effect of rehabilitation interventions.

Michel Robert (see contribution under Theme 2).

There exists ... a network in Europe ... France's network has 2000 observation sites. Observations take place...[at intervals] depending on the indicator concerned (1 to 10 years).

Michel Robert (see contribution under Theme 2).

Identifying hot and bright spots

...[more details should be given to a] description of the hot spots in the [Land Degradation Problem Areas] ... with indication of their occurrence and location and an explanation of the problems which make their degradation particularly critical (risks and impacts).

Philippe Mahler

Philippe Mahler provided a number of possible categories to help delineate hot spots or land degradation problem areas. These include categories to distinguish between those areas that are eroded due to water or wind. He also describes characteristics of bright spots or stable land areas. More detail on land degradation problem areas and stable land areas respectively in points 4 and 6 in Philippe Mahler's first contribution to Theme 4 at:

ftp://ftp.fao.org/agl/emailconf/lada/lada4_mahler_philippe.htm.

Indicator development

Framing indicator development about important questions or needs

..., it will be important ...[that] we not just get into listing good indicators, but try to frame them within the context of the specific questions that need to be answered. ... the structuring should ... be in terms of ... what are the best indicators to use for each of the questions we would like to see answered.

David Niemeijer

Recognizing land users perspective in developing indicators

From the perspective of the farmer (who ultimately experiences in her/his own livelihood the impacts of land degradation) land degradation has a meaning, which LADA implementing technical staff on the ground need to be able to capture, ...

Raul Ponce-Hernandez

Selecting indicators carefully

Indicators could work in providing a realistic picture of land degradation in drylands - however, care most be taken in their selection.

Raul Ponce-Hernandez

Broad issues and categories for indicator development

Eight broad categories for indicators are proposed by Philippe Mahler: accumulation of salts, wind erosion and sand deposition, water erosion, waterlogging, depletion of natural vegetation cover, biodiversity and soil carbon, degradation of land or soil productivity, land pollution, and poverty. The full contribution can be read at:

ftp://ftp.fao.org/agl/emailconf/lada/lada4_mahler_philippe.htm under his first contribution for Theme 4. Also see summary notes in Theme 2 under "Some proposed indicators and methods".

José Benites provided a set of key issues and clusters about which indicators specifically for drylands may be developed. The following three clusters, followed by issues, were suggested:

1. Diversity and intensity of land use: resource availability, availability of fuel wood, and decreased water supply;

2. Soil and surface water quality: inappropriate soil management strategies

3. Societal commitment and response: policy environment unsympathetic to conservation and lack of awareness.

Various land degradation indicators for drylands are recommended for each of these clusters. They can be read at: ftp://ftp.fao.org/agl/emailconf/lada/lada4benitesjoseattachment.htm.

Need for standard definitions

... it is necessary to identify relevant set of indicators for each land degradation problem category and, for each indicator, to adopt standard definitions and, where feasible and needed, classes to qualify and quantify the indicator.

Philippe Mahler

Need to build upon existing national to local capacity

From here on, suggested indicators and attributes must be matched against the reality of existing statistics and other data sets, ....

José Benites

José Benites (FAO) provides a number of priorities to help ensure that indicator development builds upon (and strengthens) existing national to local capacity. Extracts follow:

• Develop methods to increase usefulness of available data ([e.g., census, ...])

• Increase reliability of national census, timeliness and data quality ([e.g.,] special census...)

• Increase ... modules for on-farm land management in the national census to collect necessary data...

José Benites

Routine monitoring may be premature in the LADA context; need for a baseline policy support tool first

Discussing ... how to start a routine multi-level monitoring of land degradation would open a kind of Pandora box which would unduly prolong the discussions on indicators.

Philippe Mahler

While monitoring should be kept in mind as a logical follow-up, elaborating a monitoring scheme ... already now may be premature in a number of countries: most users will be convinced on the need for monitoring when and only when the results of a thorough baseline assessment are available.

Philippe Mahler

Information dissemination

Danger of distributing preliminary products

I find the idea of providing some information or intermediate products ... pretty dangerous. ... It is my experience ... [that intermediate products] in the hands of policymakers often become actual degradation assessments on which policy will be based. If such material is made public there will also be a lot of confusion once final products tell a different story.

David Niemeijer