J. Dumanski, Agriculture and Agri-Food, Ottawa, Canada,
C. Pieri, World Bank, Washington D.C., USA
The LQI programme is being developed to harmonize the combined objectives of food production and environmental protection, and is one of several responses to the challenges put forward by UNCED in Agenda 21. Because of its complexity, the programme is being initiated through a coalition of international agencies (World Bank, FAO, UNDP and UNEP). Additional partners are actively being solicited for the programme which will initially focus on LQIs for developing countries.
LQIs are needed to address major land-related issues of national and global significance, such as land-use pressures, land degradation, and soil and water conservation, as well as policy related questions on sustainable land management. Once developed and standardized through international scientific protocols, LQIs will be used for policy and programme formulation for district, national and global assessment, environmental impact monitoring, and to promote technologies, policies and programmes to ensure better use of natural resources and sustainable land management.
The LQI initiative is similar in concept to previous programmes sponsored by national and international agencies, such as indicators of economic and social performance and state-of-the-environment reporting. These programmes were initiated by groups of interested parties working together, using an iterative process of indicator development, testing, refinement and standardization. This ultimately resulted in the standard economic, social and environmental indicators that are now used routinely for monitoring national economic performance, and air and water quality. Something similar is planned in the LQI programme for development of indicators for land quality. In this context, land refers not just to soil, but to the combined resources of soil, water, vegetation and terrain that provide the basis for land use. Land quality is the condition or health of the land relative to its capacity for sustainable land use and environmental management.
Although the programme is still new (preliminary activities started in 1994), much has already been accomplished:
1. Two regional workshops (Cali and Nairobi), and one coordination workshop (Washington) have been held. The regional workshops developed and tested some of the pertinent concepts and preliminary indicators for specific agro-environments; the Washington workshop (June, 1995) was the first formal meeting between the World Bank, FAO, UNDP and UNEP to establish the basis for the global coalition necessary to launch the programme.2. The Pressure-State-Response (PSR) Framework, used by OECD, SCOPE and other national and international organizations for environmental performance monitoring, has been adopted as the common approach for the programme.
3. A discussion paper, describing the concept of LQIs, some examples of possible LQIs for specific regions, and some recommendations for their development, has been prepared (Pieri et al., 1995).
4. An institutional structure for the programme has been developed consisting of a small secretariat to develop the programme and coordinate activities, a Core Advisory Committee responsible for scientific and methodological standards, and a Donor Support Group to ensure liaison among the partners, identify priorities and ensure coordination and funding.
The objectives of the LQI programme are:
1. To develop a set of standardized LQIs for managed ecosystems (agriculture and forestry) in the major agro-ecological zones (AEZs) of tropical, sub-tropical and temperate environments.2. To identify sources of data and develop standardized methods for analyses, aggregation, and application of the results.
3. To validate and disseminate the findings among the major institutions responsible for collection of LQI data, and to identify the institutional capacity needed for setting and implementing land and natural resources priorities, policies and technologies at district, regional, national and global levels.
The outputs of the programme will be:
1. A core set of LQIs related to major policy-related questions on land management in tropical, sub-tropical and temperate regions.2. A set of targets and thresholds for the state LQIs to provide guidance towards more sustainable land management for the different eco-regions.
3. A meta-database on land-related information set up on Internet and the World Wide Web, with documentation on what data are stored with which agencies, the quality and reliability of the data, and how the data can be accessed. Emphasis will be on land suitable for cultivation and forestry, biological production potentials, current land management technologies, and other related information necessary to monitor changes in land quality.
4. Assessment of trends in land quality for various AEZs for use at district, national and global levels.
5. Case studies, using data from representative countries, will be an integral part of the programme. Development of regional and district level LQIs through case studies in representative AEZs. The case studies will focus on developing countries selected on the basis of agro-environments, intensity of land use, and amounts of data available (AEZ x land use intensity x data availability). The case studies will utilize data available from the national census bureaux, but will also include results from on-ground studies being conducted by many of the national and international agencies, including NGOs. Tentatively about six AEZ case studies are being planned.
6. National capacity building: This activity will be closely integrated with capacity building activities being conducted by agencies such as UNEP, FAO, the International Agricultural Research centres and others. Major activities will be training on geo-referenced data acquisition, training on PC-based data management, and training on the use of local farmer knowledge and the development of quantitative indicators from farmer knowledge.
The objective is to complement existing and new initiatives on land resource management. This requires that the work be done in a coordinated, cooperative fashion, using a common framework.
The results of this programme will be used to support planners, decision-makers in national governments, international institutions and donor agencies, NGOs involved in community-level development, and others. Initial activities will focus on decision-makers at district and national levels.
THE WHAT AND WHY OF INDICATORS
Indicators are statistics or measures that relate to a condition, change of quality, or change in state of something valued. They provide information and describe the state of the phenomena of interest, but with a significance beyond that directly associated with an individual parameter (OECD, 1993). Indicators need to be developed according to their perceived applications, and this requires reliable statistics and raw data. Because of regional requirements and priorities, global agreement on a single set of indicators would be difficult and unnecessary for many issues. However, a common aggregate of key indicators could be used as a basis for international comparison (Dumanski, 1994; Bakkes et al., 1994).
A distinction must be made between indicators and other types of statistics (Bakkes, et al., 1994; Eswaran et al., 1994). Measurements of some event or phenomena produce raw data, which after processing are often published as statistics. These statistics can provide underlying information, or they could be indicators if they have some added significance and are tied to a specific problem. If the number of indicators is reduced by aggregating them according to some formula, then these are called indices. Examples of useful indices are the Human Development Index, the Air Pollution Index, the UV Radiation Hazard Index, and the Water Quality Index. Indicators of land quality (LQIs) are statistics which report on the condition and quality of the land resource, but also on the cause-effect relationships which may result in changes in quality and the responses to these changes by society.
LQIs are also needed for the World Bank's new project management called Portfolio Management (OPD, 1993), and for development of Country Assistance Strategies, and National Environmental Action Plans. LQIs need to be embedded in the objectives of a project to ensure that they serve as monitoring instruments during and after the life of the project. Broad level statistics may provide some of the data needed to develop these indicators, but very often some new data will have to be collected, particularly information on how farmers manage their land and the rate of adoption of conservation practices. Criteria for the selection and development of indicators have been developed by many agencies (OECD, 1993; EPA, 1994; Bakkes et al. 1994; WRI/UNEP/UNDP, 1994).
AVAILABLE FRAMEWORKS FOR RESOURCE ASSESSMENT
LQIs are part of the family of indicators being developed for environmental reporting. However, along with other types of indicators they are also useful for reporting on the performance of the agricultural sector, because they report on land as part of the production system. A framework of some sort is required to guide the development of LQIs, and to ensure they are relevant to those land management issues in a country.
Frameworks serve to organize the (large quantities of) data used for developing an indicator, to improve the accessibility of the indicator, and to increase its value-added (EPA, 1994). Frameworks can also serve to link individual monitoring programmes, identify duplication and gaps, facilitate development of new indicators and increase the use of this information for development of policies and programmes.
There are currently many frameworks available for reporting on environmental issues. These can be classified as accounting frameworks, reporting frameworks, and sustainability frameworks.
Frameworks for environmental accounting can be:
a. Social (welfare) accounting frameworks - these are true accounting frameworks which reflect resource flows throughout a system, but which have been adjusted to include environmental costs and the distribution of these among economic activities. The social accounting frameworks require balancing inputs with outputs of production by incorporating with economic and environmental costs and to make explicit statements of gains and losses in the system.b. Environmental (natural resource) accounting or "green" accounting - this is a new accounting approach which requires that GDP or NDP be adjusted for the depreciation (cost) of using natural resources in economic activities (Costanza, 1991; Lutz, 1993). This may involve defensive environmental costs, such as those used for prevention of degradation, as well as user costs for degraded natural resources. Natural resource accounting is linked with the System of National Accounts, and has become an important tool for environmental policies and strategies. A recent extension of this concept, with application to land resources, is to consider costs of degradation and resource consumption as depreciation from nature's endowment of national wealth (Serageldin, 1995).
Frameworks for environmental reporting can be:
a. Models of decision-making processes - these are organizational frameworks which mimic environmental decision making in government and other agencies. They are early forms of environmental reporting frameworks, but they are still used for activities such as the Environmental Monitoring and Assessment Program (EMAP), and in international activities such as the International Joint Commission (IJC) (EPA, 1994).b. Pressure-state-response (PSR) framework - This framework links pressures on the environment as a result of human activities, with changes in the state (condition) of the environment (land, air, water, etc.). Society then responds to these changes by instituting environmental and economic programmes and policies, which feed back to reduce or mitigate the pressures or repair the natural resource (OECD, 1993). This framework has been adopted by many OECD countries and by the World Bank for environmental reporting. It is described in greater detail in subsequent parts of this paper.
c. Spatial frameworks - this is a general grouping of frameworks consisting of those used solely for monitoring air, water, land, flora and fauna, as well as the various ecological land classification systems which are used for environmental assessment, planning and management. Although readily understandable, they do not adequately reflect ecosystem patterns or processes, and they do not highlight environmental issues of importance to governments. In essence they deal only with the "state" dimension of the PSR approach, without linking this to pressures of human activity or responses of society.
Frameworks for assessing sustainability and sustainable development can be:
a. Agro-ecosystem approach - this approach is based on assessing the performance of agro-ecosystems according to ecological, economic and social dimensions, using four criteria for sustainability, namely productivity, resilience, stability and equity. Crosslinking the flows of resources and materials with the dimensions and criteria for sustainability enables one to assess the performance and the sustainability of agro-ecosystems.b. Total (factor) productivity - this is the ratio of the value of all outputs divided by the value of all economic and environmental inputs, normalized to remove changes in prices (Lynam and Herdt, 1988; Harrington et al., 1994; Whitaker, 1994). Total productivity is the inverse of the unit cost of agricultural production when the costs of environmental degradation are included. Agricultural systems are deemed to be sustainable when total productivity shows a non-declining trend. This framework has been adapted by several of the CGIAR centres for research purposes.
c. International Framework for Evaluation of Sustainable Land Management - sustainable land management (SLM) describes the complementary goals of maintaining and enhancing the quality of the land, while providing economic, social and environmental opportunities for the benefit of present and future generations. Sustainable land management can be assessed by evaluating the performance of the five pillars of SLM, namely maintenance or enhancement of productivity, reduction of risk, enhancement of environmental (land) quality, economic; viability, and social acceptability (Smyth and Dumanski, 1994). A logical analysis framework is used to accommodate the imprecise concept of sustainability, i.e., assessments of sustainability are inferred in degrees of probability. This framework relates closely to the PSR framework for environmental reporting, and it is being evaluated in a number of case studies throughout the world.
Each framework has a different set of objectives, and each is used for a different purpose. Although sets of indicators are required for each framework, these are somewhat different in each case. The following sections describe some recommended procedures for application of the PSR approach for development of LQIs.
APPLICATION OF THE PRESSURE-STATE-RESPONSE FRAMEWORK FOR DEVELOPMENT OF LAND QUALITY INDICATORS
LQIs report on the biophysical condition of land, but also on how it is being managed and the policy and social environment for instituting improvements in land management or which foster deterioration. This is achieved using the PSR framework.
The PSR framework has been modified slightly and adopted by the World Bank for development of environmental indicators and LQIs (O'Connor, 1994b). For the LQIs, the PSR framework is used to structure and classify information, and to assist in the identification of the key set of indicators that best describe how farmers are managing their lands and the impacts of this management.
The PSR (Figure 1) is a convenient representation of the linkages among the pressures exerted on the land by human activities (pressure box), the change in quality of the resource (state box), and the response to these changes as society attempts to release the pressure or to rehabilitate land which has been degraded (response box). The interchanges among these form a continuous feed-back mechanism that can be monitored and used for assessment of land quality.
FIGURE 1. Pressure-state-response framework
A single index of land quality is neither feasible nor desirable at this time, but groups of indicators can be developed to reflect the PSR dimensions of land quality.
Three groups of LQIs have been developed to reflect the PSR structure:
Group 1. Pressure on the land resource
Indicators in this group include those activities that relate to the degree of intensification and diversification of agricultural land uses, and result in increased pressure on land quality. This may include the number of crops in a cropping system per year or per hectare, type and intensity of tillage, degree of removal of biomass, integration with livestock systems, number of food and fibre products produced annually, etc. These indicators must be seen within the context of major socio-demographic factors such as population pressures, land tenure, etc., but the latter do not qualify for inclusion as LQIs. This is because these major forces do not influence land quality directly, but rather through the land practices that are adopted by farmers as a consequence. It is these management systems and their impacts that we wish to capture as LQIs, although changes in the major driving forces may provide some "early warning" signals.
Group 2. State of land quality
State indicators reflect the conditions of the land as well as its resilience to withstand change as a consequence of sector pressures. This may include indicators which express changes in biological productivity (actual and potential), extent and impacts of soil degradation, including erosion, salinization, etc., annual and long-term balance of nutrients (exported and imported by the cropping systems), degree and type of contamination or pollution (by direct application, atmospheric transport, etc.), changes in organic matter content, water holding capacity, etc. The changes in state may be negative with poor management, or positive with good management.
Group 3. Societal response(s)
The response mechanisms are normally achieved through direct actions by the farmers themselves in evolving or adopting improved land management systems, or through complementary activities whereby adoption of conservation technologies is stimulated by general economic, agricultural and conservation policies and programmes. In rare instances, environmental regulations may be necessary to effect proper control of land resource degradation. Response indicators may include number and types of farmer organizations for soil conservation, extent of change in farm technologies, risk management strategies, incentive programmes for adoption of conservation technologies, etc. Response indicators should be distinguished into those categories promoted by governments, those undertaken by individual farmers and those supported by agri-business.
AVAILABLE DATA AND PROCEDURES FOR DEVELOPMENT OF LQIs
Good baseline information on soils, climate, land use and land management are necessary for developing reliable LQIs. Although this information is often incomplete or lacking in many developing countries, some data are almost always available. Also, several international agencies have collated and organized some of these data in computer compatible formats. These stores of available data provide a useful point of departure for development of LQIs, although it must be emphasized that no single agency has yet put these data together into an organized database directly suitable for development of LQIs.
Some of the more useful data already collated are described briefly below. These are arranged according to the PSR framework, i.e., the usefulness of these data to develop Pressure indicators, State indicators, or Response indicators.
Pressure LQIs - Available databases and procedures for development
Pressure LQIs can be developed in most cases using available statistical (census) databases. For projects financed by the World Bank, the most useful databases are:
a. The World Bank Economic and Social (BESD) Database 1: The BESD database contains about 2 million time series in 40 data files from the World Bank, IMF, UN, UNIDO, UNESCO, FAO, OECD, and ILO (World Bank, 1993). It is maintained by the International Economics Division, in collaboration with data users and compilers throughout the Bank. For LQIs, the most useful data files are FAOPROD, FAOFERT, and some of the "Social Indicators for Development" (SOCIND) related to agricultural labour. These are typical statistical (commodity data obtained from agricultural census) data files, obtained on a regular basis from the FAO statistical office. They consist primarily of national production of agricultural commodities (crops, animals, area, yield), inputs (use of fertilizers and other inputs) and cultivation (land use, area harvested, irrigation). Continual time series data are available from 1961. These data can be used to develop broad-level, national indicators such as ratios of:
1The Economic Research Service (ERS)-USDA has prepared a global database called "World Trade: Trends and Indicators, 1970-91", which they update periodically. They have aggregated data from USDA, FAO, World Bank, IMF, UN and country publications, and processed them as time trends. However, most of the database is on economic and trade indicators, and the only indicators related to land are crop land, area harvested and yield. There is no comparative advantage to using this database compared to the BESD.¤ cultivated area/arable land
¤ production/arable land (yield)
¤ soil conserving/soil degrading crops
¤ nutrient inputs/nutrient exports.
Indicators such as these identify potential pressures on the land due to agricultural activities (Pressure LQIs). Although targets and thresholds for these indicators are not yet available, trend lines of performance provide very useful information for estimating changes of impact over time.
b. World Resources Institute 1994-95 Database: This is a PC-compatible database consisting of 503 variables for 198 countries. It is the source database for WRI's publication on global conditions and trends (WRI/UNEP/UNDP, 1994). In some aspects the database is similar to the BESD database in that it contains some of the same variables from the same sources, e.g., FAOPROD, but the data are processed further and often summarized as ratios, and this generally makes the data more useful as indicators. The database provides economic, population, natural resource and environmental statistics as one-time data and 5-, 20-, 30-, and 40-year time series for many of the variables. The most useful data files for LQIs are Land Cover and Settlements, Food and Agriculture, Forests and Rangelands, Biodiversity, Water, and Atmosphere and Climate.
c. UNEP/GEMS/GRID Database: The Global Environmental Monitoring System (GEMS) is operated by UNEP in collaboration with UN organizations, national governments, environmental groups and scientific bodies. GEMS does not generate new data, but rather accepts data from other national and international agencies, such as FAO, and then harmonizes and pre-processes the data to make them compatible and more accessible and useful to users. Through the Global Resource Information Data Base (GRID), GEMS maintains a variety of global and regional environmental databases, including data on soil, vegetation, climate and cultivation that are useful for LQIs. GEMS receives and stores environmental indicators based on remote sensing data, such as the weekly and seasonal vegetation indices from NOAA and NASA, and the cultivation intensity index from the Goddard Institute for Space Studies. These indices provide useful information on the extent of land cover and the hazards of soil erosion.
Conclusions on Pressure LQIs:
¤
The World Bank BESD database provides the most useful set of data for development of Pressure LQIs, but this database is not sufficiently comprehensive and must be supplemented by data from the WRI and GEMS databases.
¤ Although these databases are global, the origin of the data is national and sometimes regional, and therefore can be disaggregated and applied for national and regional projects. Some Pressure LQIs have national meaning, but many have regional significance and these should be geo-referenced using GIS.
¤ A standard set of these indicators could be selected for reporting in the World Development Reports produced by the World Bank. Over time, these would evolve as international standards, similar to the economic and social indicators routinely reported by the World Bank and others.
State LQIs - Available databases and procedures for development
Whereas Pressure LQIs can be developed using available databases, the situation is not as straightforward for State LQIs. Some national and global computerized databases are being developed, and the most useful of these for State LQIs are described briefly below. In most cases, however, the data remain dispersed among individual, specialized data banks. Also, these data will have to be supplemented with data from other sources, such as long-term experiments and by estimates obtained from physical process models, remote sensing and related techniques. The most useful of these for State LQIs are also described below.
Global and Regional Databases for State LQIs
i. FAO Databases: In addition to the statistical data described above, FAO also maintains a collection of soil and agro-ecological databases, some of which originate solely from FAO, and some jointly between FAO and other international institutions, such as IIASA, UNEP and ISRIC. The most useful databases for LQIs include the following:
Agro-ecological Zones Data Bank:
This is a global database with data on soils, climate, landforms and some land use. It consists of 500 000 records with a time-span from 1969-1990. Also, the database is linked to a generalized crop growth model which is used to estimate "constraint-free yields" (crop yield based on genetic coefficients of the crop, photosynthetically active radiation, temperature, length of the growing season), and "anticipated yield" (constraint-free yield adjusted by soil limitations such as soil water, salinity, acidification, erosion) for economically important crops. In practice, "constraint-free yield" is not a useful indicator (LQI), except as a theoretical maximum for experimental purposes. However, "anticipated yields" provide useful targets of agronomically and economically feasible yields (providing that these estimates are interpreted as estimates only), and can be compared with current farm yields to develop an LQI such as actual (farm) yield.
Anticipated (potential) yield: As this LQI approaches (or exceeds) 1.0, this indicates increasingly intensive land management, signalling the need to check for possible water quality problems and excessive fertilization. On the other hand, values < 0.2 indicate marginal or submarginal performance of the crop in that area. This could be due to low levels of fertilization (possible nutrient mining), a potentially serious degradation problem, or an unsuitable crop for the region.
CDROM Soil Map of the World: This is a soil database consisting of the digitized version of the Soil Map of the World (1:5M), and various soil interpretations. Georeferenced soil profiles (descriptions and analytical data) are stored in the FAO/ISRIC Soil Database. It is estimated that FAO currently maintains records for about 175 soil profiles.
ii. ISRIC Databases:
SOTER:
This database consists of attribute files for soil-terrain maps for selected countries mostly in Latin America, the Near East and East Africa, produced in ISRIC-FAO-UNEP collaboration. The information is available at scales of 1:1M or smaller, although larger scales are available in some countries. The SOTER databases are being expanded as the technology is used in different regions.
GLASOD: This global database consists of the digitized map of human-induced land degradation (1:10M), produced by ISRIC with the support of UNEP. It provides estimates of the kind, degree and extent of degradation in all countries of the world. The map and data are based on estimated evaluations of degradation (rather than measured values) provided by local scientists in each country and have been criticized because of this by the soil science community. Nonetheless, it provides the only global estimates of land degradation available, and it will continue to be used until better information becomes available.
WISE: This combined database consists of soil data available from FAO, NRCS and ISRIC. This combined store contains 4 353 soil profiles (Africa - 1 799; South, West, North Asia - 522; China, India, Philippines - 553; Australia, Pacific Islands - 122; Europe - 492; N. America - 226; S. America and Caribbean - 599). These profile data are complemented with a simplified grid cell (half degree) database of the World Soil Map.
iii. World Soil Resources Database: This database is maintained by the Natural Resource Conservation Service-USDA (formerly the Soil Conservation Service), and it consists of a series of global, national and regional, digitized soil maps (ARC/Info and GRASS) at various scales, as well as special files on soil pedons (profiles), soil carbon and soil climate. The soil pedon database consists of about 17 000 complete records, of which 2 437 are geo-referenced (345 from Central and South America, 179 from Asia and South Pacific, 422 from Europe, 43 from North America (excluding the USA), and 1 386 from other regions). The soil carbon database is built on pedons selected from the soil pedon file, and consists of 2 120 pedons of which 743 are outside the USA. This file is useful for studies on carbon sequestration and organic matter management in relation to climate change. The soil climate file is developed through a procedure which accesses a global climate database of over 27 000 stations, the FAO soil database of the world, and the soil pedon file. Through this interaction, the following information can be generated:
¤
soil temperature and moisture regimes;
¤ length and dates of growing season;
¤ moisture and temperature stress;
¤ moisture and temperature calendars;
¤ the FAO soil classification.
Currently, the spatial data files (digitized maps) are minimal, consisting only of 26 maps, all outside the USA, although there is a continental soil map of Africa (1:5M) and a digitized world soil map (1:30M), both derived from an earlier version of the Soil Map of the World. The store of digitized maps, however, is continually increasing.
iv. Databases maintained by CGIAR Centres: These are research-oriented databases, maintained by several of the Centres in support of their regional and global research mandates. The data holdings in some of the Centres are quite extensive, particularly in CIAT and ICRAF, but continental or global coverage of any theme is rare (this is normal in research-oriented organizations, since most of the data files originate from project activities). These holdings are summarized as follows:
CIAT:
The most complete coverage is climate information for all the tropics (19 000 stations), and topographic elevations. Soil data are minimal, including only the FAO Soil Map of the World for Latin America and Africa. Additional coverages include roads and legally protected areas (Latin America), vegetation (South America), land systems (lowland tropics, Latin America), administrative boundaries (Latin America and Africa) and additional coverages such as population densities (Africa), tribal boundaries (Africa), etc., which originated from the Africa cassava study. Many of the coverages are stored in 10 arc-minute format, and results are produced as digital elevation models (DEMs). CIAT was a pioneer in GIS systems in the CGIAR, and has a large, well functioning installation, operating with ARC/Info and IDRISI software.
ICRAF: This is a new and smaller installation, but similar to that of CIAT in terms of capability and capacity. Most data holdings are for continental Africa. The most complete coverage so far is for climate data, consisting of actual and estimated mean monthly data, as well as daily data (8 000 stations, with 17 year time series). Climate coefficients were obtained from CIAT, Centre for Resources and Environmental Studies (CRES) (Australia) and several national climatic agencies. Soils data are sketchy, but digitized soil maps are available for seven countries (1:1M). Other data, such as vegetation, land cover, etc. are incomplete. DEMs, obtained from the Earth Resource Observation System (EROS) Data Centre, are available for all of Africa. The ICRAF system is still in a rapid stage of development, and data holdings are expected to increase rapidly over the next few years.
ICRISAT: The ICRISAT data holdings include long-term daily climate data for seven countries in Africa, two in Asia and one in South America. Some detailed soil pedon data are available for six countries in Africa and one in Asia. Data from detailed village studies are available for three countries in Africa and for India, and field productivity data are available for India and six countries in Africa. It is not clear if these data are maintained in a GIS format.
Long-term Agronomic Experiments
The Rockefeller Foundation has recently completed a global inventory of continual long-term agronomic experiments as a source of research information for issues related to agricultural sustainability (Steiner and Herdt, 1993). The records include ten experiments from Africa (from 1912), 24 experiments from Asia (from 1909), and nine from Central and South America (from 1941). The types of data, and often the quality, are varied as would be expected, but often include agronomic data, soil characteristics, physiology, weather and economic data. These data would be a valuable source of supporting information for development of LQIs.
Use of Physical Process Models
The overall scarcity of information for development of State LQIs requires the use of indirect measures, and some of the best of these are physical process models. There are many models available in soil science, but most have been developed primarily for research and cannot be easily applied for operational programmes such as LQIs. However, there is a growing family of models which have been verified in many environments, including the tropics to some extent, and are beginning to have a good track record. The most relevant of these for LQIs are described below.
i. Erosion Productivity Impact Calculator (EPIC): EPIC was developed by the United States Department of Agriculture (USDA) and Agricultural Research Service (ARS) originally as a tool to analyse the impacts of soil management and erosion on crop yields, but more recently it has been expanded to include assessments of water quality, pesticides, etc. EPIC consists of ten major subroutines, namely, weather, hydrology, wind and water erosion, nitrogen and phosphorus transformations, soil temperature, crop growth, tillage, plant environment control (irrigation, lime, etc.), pesticide routines and economic crop budgets. Interim and final output is available from each subroutine, either in daily, monthly or annual increments. Although the model inputs are flexible through the use of many data defaults (for missing data), the model requires reliable data on soil properties, crop inputs and tillage management (weather is generated through a weather generator). EPIC generates several potentially useful outputs for LQIs, namely:
¤yield, for several economically important crops;
¤ erosion, wind and water, rate (t/ha) and impacts on yield;
¤ change in nitrogen and phosphorus (crude estimate).
Rates of change are calculated by running EPIC using various land management scenarios over many years (usually 30 years). Increasingly, EPIC is being adapted to many temperate as well as tropical regions as a tool to evaluate land management practices, particularly tillage and residue management. It also has been integrated with large economic optimizing models to provide analytical systems for evaluation of environmental impact prior to implementation of agricultural policies and programmes.
ii. CENTURY: The CENTURY model simulates the effects of erosion on long-term storage of soil organic carbon under field conditions. Briefly, soil organic matter is divided into pools with active (1.5y), slow (25y) and passive (1 000y) turnover rates. A plant production subroutine simulates the allocation of carbon into shoots and roots, dividing plant residue into a metabolic (0.1-1y) and a structural (1-5y) pool based on the lignin: nitrogen ratio. The model then transfers the carbon to the soil, and simulates carbon stability through interactions with clay and organic molecules. Estimates of soil carbon change are obtained by running CENTURY under initial (usually current) conditions, then again for future scenarios under new management technologies. Output useful for LQIs include:
¤
total soil carbon, used to estimate carbon sequestration;
¤ rapid turn-over fraction, a surrogate for microbial biomass
In terms of land quality, rapid turn-over of carbon is a better LQI than total carbon.
iii. NUTMON: This is a recently developed model for estimating regional losses or gains of nutrients as a consequence of nutrient inputs (mineral fertilizers, organic manures, wet and dry deposition, nitrogen fixation, sedimentation), compared to nutrient losses (harvested product, crop residue removal, leaching, erosion, denitrification) (Smaling, 1993). Data for nutrient inputs and nutrients removed by harvest are gathered for various land use systems, and estimates for the other variables are calculated using various available models. NUTBAL then calculates whether the systems are gaining or losing for each macronutrient. Results can be extrapolated to wider areas using GIS techniques. NUTBAL is still experimental, but it has been used for studies in Kenya with good success.
Development of Proxy Indicators
Proxy indicators of State LQIs need to be developed where more reliable information from other sources is not available. Some proxies, such as the Normalized Difference Vegetation Index, can be obtained from remote sensing, or through analyses of aerial photographs. Also, various proxy indicators are available from known cause-effect relationships, such as:
¤
completeness of vegetative ground cover, proxy for erosion risk hazard;
¤ sediment load in water bodies, proxy for water erosion;
¤ removal of crop residues, proxy for nutrient removal and for soil carbon sequestration;
¤ price of fuelwood, proxy for rate of harvesting of woody materials;
¤ presence of indicator plants, usually weeds;
¤ increase in salinity or acidification.
Conclusions on State LQIs:
¤
There are currently few global and regional databases that can be used for State LQIs, because historically little effort has been expended nationally and internationally to document how land is being used (except for North America and parts of Europe).
¤ Considerable knowledge is available, however, on cause-effect relationships between land use and change in land quality (pressure-state relationships), and this knowledge has been captured in physical process models of various kinds. Some of the better tested models can be used to develop estimates of State LQIs, but application of the models requires development of large, reliable input files (normally soil data, land management information including tillage, and long-term daily weather data), as well as technological expertise to run the models, verify the output, and interpret the results.
¤ Proxy State LQIs can be developed using techniques such as remote sensing, aerial photo interpretation and field identification of indicator plants, etc.
¤ In most cases, State LQIs are location specific and should be geo-referenced at appropriate scales using GIS techniques.
Available Data and Procedures for Development of Response LQIs
Response LQIs primarily involve adoption of soil conservation. This concerns both awareness of the conservation problem and knowledge of what to do about it, as well as adoption of specific conservation technologies by farmers. Response LQIs are distinguished by those promoted by governments, those supported by agri-business, and those undertaken by individual farmers. Response LQIs reflecting government actions are normally national (sometimes regional) in scale, whereas adoption of conservation technologies by farmers is local or regional.
Information Sources for National Response LQIs
Response LQIs can be developed by collating and evaluating the activities undertaken by governments, the private sector, NGOs and farmers in response to problems of soil degradation. Governments normally respond with programmes to increase public awareness, incentive programmes for adoption of conservation practices, improved advisory services, and, in rare instances, legislation. Many of these activities may be undertaken through partnerships among governments, the private sector and NGOs.
Information on these initiatives is readily available from government departments and from the records of private NGOs, chemical and machinery companies, etc. Also, much useful information on these activities can be gleaned from the numerous Participatory Rural Appraisal studies completed through Bank-financed studies, NGOs and others. This information can be used to develop Response LQIs such as:
¤
kind, duration, funding of awareness and incentive programmes;
¤ kind, duration, funding of incentives programme;
¤ legislation for conservation;
¤ activities, size, membership of conservation associations;
¤ etc.
The presence and impact of farmer self-help groups for soil conservation, such as conservation clubs and "Club de Terra", can be powerful Response LQIs. If these associations are strong and active, this indicates that farmers are aware of the degradation problems and are prepared to make investments to overcome them (normally an incentive programme is necessary to get them started, and to assist with the investment costs). Farmer conservation clubs and associations should not be confused with farmer marketing cooperatives, however. In some cases the marketing coops may also promote activities in soil conservation, but often they focus only on marketing and may actually be counterproductive.
Information on Adoption of Soil Conservation Technologies
The adoption of improved technologies in soil conservation by farmers is an important statistic for development of LQIs, yet this is never available and must be gathered. Through programmes such as the agricultural census (the full census or special surveys), or it can be instituted as part of the project activities (rapid project surveys). Whatever procedure is chosen, a special land management questionnaire has to be developed to gather the relevant data. The questionnaire must be strategically designed, short and cost effective to implement if it is to be used (an example is given in Appendix 1). With such data, Response LQIs could be developed such as:
¤
use of conservation structures, % farmers, extent;
¤ use of conservation tillage, % farmers, extent;
¤ use of special inputs (manures, lime, etc.);
¤ integration with livestock, agroforestry;
¤ participation in soil conservation associations;
¤ etc.
These could be summarized as national Response LQIs, or they could be summarized for specific regions covered by a project. Response LQIs based on average adoption rates are useful, but they have to be correlated with the specific biophysical or production constraints prevailing in an area, e.g. absence of techniques for control of wind erosion in a high wind erosion risk area may indicate lack of awareness of the problem. Caution must be exercised to ensure that average national conditions do not mask important local variability, e.g., an individual farm may be sustainable because of superior management, in an area that is judged overall to be experiencing major problems, and vice-versa.
Major shifts in land use, sometimes called indicators of unsustainability (Jodha, 1990), are also useful Response LQIs. These shifts occur when the performance of one land use has deteriorated beyond an acceptable threshold and survival depends on adopting some alternative, often less intensive system. Such response LQIs may be:
¤
retirement of marginal lands;
¤ shift from cultivation to pasture;
¤ shift from cattle to goats;
¤ abandoned terraces;
¤ increased seasonal migration;
At the other end of the scale, major shifts in land use can also occur due to new market opportunities or some major change in resource availability. In all cases, Response LQIs must be interpreted on the basis of knowledge on the motivation(s) for the response.
GUIDELINES FOR APPLYING THE PSR FRAMEWORK
Some general guidelines have been developed for using the PSR framework with LQIs. The pressure and response indicators are generally considered at the level of the agricultural sector (agriculture in this case, but may also be forestry or other biological land uses), whereas the state indicators relate directly to change in condition and in some cases the quantity of the land resource. Changes in state, even if they are of small magnitude such as oxidation of organic matter or nutrient balance, can be of considerable impact if they occur over large areas.
Sectoral pressures and response are useful for expressing the impacts of the sector on the condition of the land, and therefore often relate directly to the policy arena. Indicators that can show relationships between pressures and change in state generally have the most meaning for environmental decision makers.
The application of the PSR approach to LQIs requires that key land issues be identified for each cluster of indicators, i.e., what are the key policy-related questions on land that must be answered? These should be developed carefully, since they are crucial for identifying the short list of strategic indicators and sub-indicators to be associated with each issue and each cluster. Normally, these issues are associated with specific geographic regions, and reflect the priorities of these regions. Local farmer knowledge and the advice of experienced agronomists can be very useful in this portion of the exercise.
Targets or goals for each indicator should then be developed if possible (Adriaanse, 1993), as well as thresholds where the systems may become unsustainable (Smyth and Dumanski, 1994). Performance of the sector can then be monitored in relation to these targets, goals and thresholds, i.e., the contribution of the sector to resource maintenance or resource degradation can be assessed. If realistic, reliable and scientifically sound targets and thresholds cannot be developed, however, then trends in performance can still provide very useful information.
The pressure-state-response framework has been used for state-of-the-environment reporting and for national environmental performance reviews. For the LQIs, it is used to structure and classify information, and to assist in the identification of the key set of indicators that best describe how farmers are managing their lands and the impacts of this management.
The PSR framework remains in a continuous state of evolution. The EPA (1994) is proposing to extend the framework to include the effects of changes in state on the environment (pressure-state-response/effects). UNEP (1994) is discussing the development of pressure-state-impact-response (PSIR) framework. O'Connor (1994a) has extended the PSR framework towards development of a "sustainability matrix". In a recent development, EPA (1994) is proposing to re-orient environmental reporting more towards "place-driven" approaches, using ecosystem stratification, spatially referenced data, and geographic information systems.
MAJOR PROGRAMME ACTIVITIES AND METHODOLOGY FOR THE LQI PROGRAMME
The LQI programme is targeted to specific objectives, outputs and beneficiaries. The intent of the programme is cost-effective development and validation of harmonized indicators that reflect a broad consensus. Activities and programmes already under way will be included to avoid duplication of effort and resources. Activities related to the LQIs, such as those currently under way in FAO, UNEP, UNDP, OECD, the CGIAR, and several of the international scientific unions such as SCOPE and International Council of Scientific Unions (ICSU), will be used.
REFERENCES
Adriaanse, A. 1993. Environmental Policy Performance Indicators. A Study on the Development of Indicators for Environmental Policy in the Netherlands. Sdu Uitgeverij Koninginnegracht, The Netherlands. 175 p.
Bakkes, J.A., Van den Born, G.J., Swart, R.J., Hope, C.W. and Parker, J.D.E. 1994. An Overview of Environmental Indicators: State of the Art and Perspectives. UNEP/EATR.04-01; Environmental Assessment Sub-Programme, UNEP, Nairobi. 72 p.
Costanza, R. 1991. The Ecological Economics of Sustainability: Investing in Natural Capital. In: Environmentally Sustainable Economic Development. Building on Bruntland, Goodland, R., Daly, H. and El Serafy, S. (eds.). 1991. Environment Working Paper No. 46. World Bank, Washington D.C. 85 p.
Dumanski, J. 1994. Proceedings of the International Workshop on Sustainable Land Management for the 21st Century. Vol. 1: Workshop Summary. The Organizing Committee. International Workshop on Sustainable Land Management. Agricultural Institute of Canada, Ottawa.
EPA [Environmental Protection Agency]. 1994. A Conceptual Framework to Support the Development and Use of Environmental Information. Environmental Statistics and Information Division. Office of Policy, Planning and Evaluation. EPA 230-R-94-012, USEPA, Washington D.C.
Eswaran, H., Pushparajah E. and Ofori, C. 1994. Indicators and their Utilization in a Framework for Evaluation of Sustainable Land Management. In: Proceedings of the International Workshop on Sustainable Land Management for the 21st Century, Wood, R.C. and Dumanski, J. (eds.). Vol. 2: Plenary Papers. The Organizing Committee. International Workshop on Sustainable Land Management. Agricultural Institute of Canada, Ottawa. pp. 205-225.
Harrington, L., Jones P. and Winograd, M. 1994. Operationalizing Sustainability: A Total Factor Productivity Approach. Unpub. paper given at Cali LQI workshop, June, 1994. World Bank, Washington, D.C.
Jodha, N.S. 1990. Sustainability of mountain agriculture: some imperatives. Entwicklung und Landlicher Raum 3 (90): 16-19.
Lutz, E. (Ed.). 1993. Towards improved accounting for the environment. Proceedings of an UBSTAT-World Bank Symposium. World Bank, Washington, D.C. 329 p.
Lyman, J. and Herdt, R. 1988. Sense and Sensibility: Sustainability as an Objective in International Agricultural Research. CIP-Rockefeller Conference on Farmers and Food Systems. CIP, Lima, Peru.
O'Connor, J.C. 1994a. Environmental performance monitoring indicators. In: Monitoring Progress on Sustainable Development. A User-Oriented Workshop. 22-23 Sept., World Bank, Washington D.C.
O'Connor, J.C. 1994b. Towards Environmentally Sustainable Development. Measuring Progress. Paper given at IUCN 19th Session of the General Assembly, Buenos Aires, 18-26 Jan. 1994.
OECD. 1993. OECD Core Set of Indicators for Environmental Performance Reviews. A Synthesis Report by the Group on the State of the Environment. OECD, Paris. 35 p.
OPD. [Operations Policy Department]. 1993. Portfolio Management: Next steps. World Bank, Washington D.C. 23 p. with appendixes.
Pieri, C., Dumanski, J., Hamblin, A. and Young, A. 1995. Land Quality Indicators. World Bank Discussion Paper 315. World Bank, Washington D.C. 63 p.
Serageldin, I. 1995. Sustainability and the Wealth of Nations: First Steps in an on-going Journey. Third Annual World Bank Conference on Environmentally Sustainable Development. World Bank, Washington D.C. 19 p.
Smaling, E. 1993. An Agro-Ecological Framework for Integrated Nutrient Management, with Special Reference to Kenya. Doctoral Thesis. Agricultural University, Wageningen, The Netherlands.
Smyth, A.J. and Dumanski, J. 1994. FESLM: An international framework for evaluating Sustainable land management. World Soil Resources Report 73. FAO, Rome. 74 p.
Steiner, R.A. and Herdt, R.W. (eds.). 1993. A Global Directory of Long-Term Agronomic Experiments. Vol. 1: Non-European Experiments. The Rockefeller Foundation, New York.
UNEP. 1994. World Environment Outlook: Brainstorming Session. ENEP/EAMR. 94-5. Env. Assessm. Prog., Nairobi. 18 p.
Whitaker, M. 1994. Using long-term experimental data in evaluating productivity and Sustainability of alternative crop technologies. In: Proceedings of the International Workshop on Sustainable Land Management for the 21st Century. Vol. 2. Plenary Papers, R.C. Wood and J. Dumanski (eds.). The Organizing Committee. International Workshop on Sustainable Land Management. Agricultural Institute of Canada, Ottawa. pp. 297-311.
World Bank. 1993. BESD (Bank Economic and Social Database). Intern. Econ. Dept., World Bank, Washington D.C.
WRI [World Resources Institute]. UNEP and UNDP. 1994. A Guide to the Global Environment. World Resources 1994-5. Oxford University Press, New York.
APPENDIX 1
DEVELOPMENT OF A SOIL CONSERVATION MODULE FOR ADOPTION OF CONSERVATION TECHNOLOGIES
The adoption of soil conservation technologies is an important source of information for developing LQIs, but these data are almost never available. Procedures must therefore, be developed to collate these data either through national census, special project studies, or research. Adoption of soil conservation technologies provides information on how farmers are managing their land, and this is necessary to develop good Response LQIs and to enable proper interpretation of Pressure and State LQIs.
Soil conservation technologies are the foundation for attaining sustainable land management, i.e., maintenance of soil quality must first be assured if other interventions and investments are to be effective. Also, many of these technologies have been tested in many parts of the world, and much is known about changes in soil quality with adoption of conservation.
Adoption of soil conservation technologies by a farmer is never done in isolation, but as part of a larger farm management strategy. Therefore, a variety of data is required in addition to information on adoption. The complete information required includes the following:
¤
size, distribution and tenure of farm;
¤ crops grown, area and yield;
¤ integration with livestock and/or agroforestry;
¤ crop and livestock management and inputs;
¤ soil conservation and tillage practices.
These data are essential for development of Response LQIs, but also for input to models and other procedures for State LQIs.
Part of these data is already being gathered in many standard census, questionnaires, but some are additional to the normal census. The following section outlines how part of the information requirements can be integrated with existing census data forms, and then proposes a new module to be added for soil conservation.
The example shown is generic, and it is more complete than would be necessary for many countries. Also, it would have to be carefully tailored to the land-use systems prevailing in the country of application to be useful. Although the questionnaire appears to be quite long, most responses are multiple choice, thereby making the questionnaire easy to implement in the field and effective for computerized data analyses.
TABLE 1. Example of a questionnaire for adoption of conservation practices
The first set of questions (or something similar) usually standard on most census questionnaires | ||||||||
1. Size, distribution and tenure of farm |
| |||||||
|
What is the total area of your farm |
_____u.a 1 | ||||||
|
What is the total area owned |
_____u.a | ||||||
|
What is the total area leased, rented or sharecropped: |
| ||||||
|
|
from governments or tribal authorities |
_____u.a | |||||
|
|
under Islamic law |
_____u.a | |||||
|
|
from others |
_____u.a | |||||
Has the number of fields on your farm changed over the last five years |
| |||||||
Increased_______ |
Decreased_______ |
| ||||||
2. Crops grown, area and yields (?) |
| |||||||
On your farm, do you usually grow: |
| |||||||
|
one crop per growing season |
_____ | ||||||
|
more than one crop per growing season |
_____ | ||||||
|
both, depending on the field |
_____ | ||||||
If you grow more than one crop per growing season, which of the following do you use: |
| |||||||
|
relay cropping |
_____ | ||||||
|
intercropping |
_____ | ||||||
|
strip cropping |
_____ | ||||||
|
other (please specify) |
_____ | ||||||
Report the total area of crops grown (seeded or to be seeded) on your farm: |
| |||||||
Maize: |
for grain area |
_____u.a |
yield |
_____u.a |
| |||
|
for silage area |
_____u.a |
yield |
_____u.a |
| |||
Wheat: |
spring area |
_____u.a |
yield |
_____u.a |
| |||
|
winter area |
_____u.a |
yield |
_____u.a |
| |||
3. What draught power do you usually use for cultivating your fields: |
| |||||||
hand labour |
|
|
_____ | |||||
animal power |
owned |
|
_____ | |||||
|
hired |
|
_____ | |||||
Do you use a tractor to cultivate your fields |
| |||||||
no_____ |
yes_____ |
| ||||||
small tractor (< 30 hp) |
owned |
_____ | ||||||
|
hired |
_____ | ||||||
large tractor (> 30 hp) |
owned |
_____ | ||||||
|
hired |
_____ | ||||||
4. Do you irrigate any portion of your farm |
| |||||||
no_____ |
yes_____ |
area_____ |
| |||||
|
gravity or flood |
|
_____ | |||||
|
centre pivot |
|
_____ | |||||
|
portable |
|
_____ | |||||
|
linear |
|
_____ | |||||
|
trickle |
|
_____ | |||||
|
drip |
|
_____ | |||||
|
microjet |
|
_____ |
1
Please specify the unit area used (e.g., ha).
The second set of questions would be added to the census questionnaire | |||||
1. Integration with livestock and/or agroforestry |
| ||||
|
Do you keep animals on your farm: |
| |||
|
no_____ |
yes: _____ |
| ||
|
|
cattle, sheep or goats |
| ||
|
|
chickens, turkeys, etc |
| ||
|
|
horses, mules, etc |
| ||
|
|
camels |
| ||
|
|
other (specify) |
| ||
Are trees important for your farm: |
| ||||
|
no_____ |
yes: _____ |
| ||
|
|
fruit, nuts |
| ||
|
|
fuelwood |
| ||
|
|
building materials |
| ||
|
|
poles, stakes |
| ||
|
|
windbreaks |
| ||
|
|
shade |
| ||
|
|
other (specify) |
| ||
2. Crop and livestock management and inputs |
| ||||
What was the area of land on which each of the following was used: |
| ||||
|
commercial fertilizer |
_____u.a | |||
|
manure |
_____u.a | |||
|
mulches or compost |
_____u.a | |||
|
herbicides |
_____u.a | |||
|
insecticides or fungicides |
_____u.a | |||
|
field drainage |
_____u.a | |||
On your farm, do you use any: |
| ||||
local fertilizers such as rock phosphate, etc. |
| ||||
|
no_____ |
yes_____ |
| ||
lime |
| ||||
|
no_____ |
yes_____ |
| ||
crop varieties tolerant to acid soils |
| ||||
|
no_____ |
yes_____ |
| ||
special procedures to control soil salinity (alkali) |
| ||||
|
no_____ |
yes_____ |
| ||
3. Soil conservation and tillage practices |
| ||||
Is soil erosion a problem on your farm |
| ||||
|
no_____ |
yes_____ |
| ||
If yes, which of the following practices do you use to control soil erosion: |
| ||||
|
none |
_____ | |||
|
crop rotations using cultivated grasses, legumes, etc. |
_____ | |||
|
cover crops |
_____ | |||
|
cultivation along the contour |
_____ | |||
|
strip-cropping along the contour |
_____ | |||
|
mechanized terraces |
_____ | |||
|
biological terraces (with barrier crops) |
_____ | |||
|
grassed waterways |
_____ | |||
|
windbreaks or shelterbelts |
_____ | |||
|
stonelines |
| |||
|
other (please specify) |
_____ | |||
Which of the following implements do you use to prepare your land for seeding: |
| ||||
|
traditional farm plough |
_____ | |||
|
hand hoe/rake |
_____ | |||
|
land leveller |
_____ | |||
|
mouldboard plough |
_____ | |||
|
disk harrow |
_____ | |||
|
spike harrow |
_____ | |||
|
spike cultivator (plough) |
_____ | |||
|
rotary tiller |
_____ | |||
|
other (please specify) |
_____ | |||
What area of land is prepared for seeding using: |
| ||||
|
convention tillage (most of the residues are incorporated into the soil) |
_____u.a. | |||
|
conservation tillage (most of the crop residue (trash) is retained on the soil surface) |
_____u.a. | |||
|
no tillage (seed placed directly into stubble or sod with a special planter) |
_____u.a. | |||
4. Soil conservation associations/programmes |
| ||||
Do you know of any soil management and conservation associations in your area: |
| ||||
|
no_____ |
yes_____ |
| ||
Are you a member of any of these associations: |
| ||||
|
no_____ |
yes_____ |
| ||
Are any incentive programmes for soil conservation available in your area: |
| ||||
|
no_____ |
yes_____ |
| ||
|
government |
_____ | |||
|
NGO |
_____ | |||
|
private companies |
_____ | |||
Do you participate in any of these incentive programmes: |
| ||||
|
no_____ |
yes_____ |
|