A. von. LOSSAUC
a Consultant to the German Ministry for the Environment, Nature conservation, and Nuclear Safety; formerly Head of Division “soil pollution and soil quality targets”. Private address: Muffendorfer Hauptstr. 46, D-523177 Bonn
b Consultant for tropical animal husbandry and pasture management, Rohnsweg 56 D-37085, Goettingen
c GTZ, Dept. 424 Forest resources management, livestock, fisheries and nature conservation. Postfach 5180, D-65726 Eschborn
1. Dealing with complexity
UNCED in 1992 marked a shift in economic and political thinking. Sustainable use of natural resources has become an increasingly important and challenging issue in planning and governance. In this context, one of the most relevant sectors is agriculture, including animal production. This is characterised by great variability and complexity, because it is highly dependent on natural prerequisites for land-use such as climate, landforms, hydrology, soils, and biodiversity of flora and fauna, as well as social and economic conditions.
The effects of a particular livestock production system on the environment cannot be viewed in isolation. Firstly, they have to be seen within the wider economic and institutional setting. The operational rules for a subsistence economy differ substantially from those for a market-oriented one with a high dependency on external inputs and services. The structure and functioning of institutions to govern economic and social processes are closely linked to the state of economic development. This is especially true for land-use rights.
Secondly, animal production is strongly influenced by socio-cultural factors such as division of labour according to gender, differences in access to resources by individuals and groups, household and local-level decision-making, and food preferences.
Thirdly, animal production depends on agro-ecological conditions. This refers not only to natural conditions, such as climate, soil, topography, or flora and fauna, but also to land improvement such as terracing or drainage. The natural properties and human activities (including animal husbandry) determine land quality. In assessing environmental impact and sustainability, evaluation of land quality is one of the most important tasks.
Box 1: Definition of land quality
Land quality is a complex attribute of land which affects its suitability for specific uses (FAO 1976, 1993; Sombroek & Sims 1995). The concept was formulated in order to avoid having to refer to a large number of individual characteristics (e. g. mean annual rainfall, soil water-holding capacity, slope) in land evaluation. Examples of land quality are availability of water, availability of nutrients, susceptibility of soils to erosion. and quality of natural timber or of grazing land. As different stakeholders use land for different purposes, there is not one land quality, but a range of land qualities (Pieri et al 1996).
Assessing the impact of animal production on the environment and tracking the sustainability of livestock-keeping systems require a methodology which allows us to take natural and economic conditions as well as socio-cultural values into account. We recommend a stepwise approach:
Box 2: Steps and tasks within the framework of assessing and monitoring environmental impact and sustainability of livestock production
Step 1: Pressure-state-response (PSR)analysis
Step 2: Environmental impact assessment (EIA)
Step 3: Sustainability assessment (SA)
Step 4: Monitoring the track towards sustainable development (SD)
2. Identifying the issues and levels of interest
The most striking issue in research and political discussion is the dualism of livestock production for human welfare. On the one hand, it is common understanding that sustainable development in agriculture must include animal husbandry for economic and ecological reasons. Livestock play an important role in rapid recycling of non-food biomass and greatly contribute to the economic stability of farming systems. On the other hand, livestock production can consume large amounts of fossil energy, waste foodstuffs and seriously impair natural resources. Both can be true. An accurate analysis needs to reveal the effects and causes in each specific case.
For evaluating the key processes in livestock-environment interactions, for EIA and SA, and for building a framework for guidance and decision making, we need appropriate sets of indicators. These will differ depending on what level we work on (farm, regional or watershed level, national, government or global level).
This is illustrated by listing some of the different objectives of animal keeping held by farm families as opposed to national governments.
Table 1: Objectives of animal keeping, held by farm families and national governments
|Farm families:||National governments:|
|• to produce food||• to increase the Gross Domestic Product|
|• to generate income||• to reduce imports and save foreign currency|
|• to accumulate savings||• to increase exports and earn foreign currency|
|• to provide draft power||• to create employment|
|• to provide manure as fertiliser or fuel||• to improve income distribution and alleviate rural poverty|
|• to satisfy socio-cultural obligations||• to provide cheap meat and dairy products for urban dwellers|
|• to buffer against food shortages||• to sustain resources, including biodiversity|
|• to sustain natural resources for subsequent use||• to avoid negative impact on ecosystems and to protect wildlife and flora (including transboundary effects).|
|• to be in harmony with the natural surroundings.|
The multiple uses of landscapes must also be taken into account. As an example: rangelands may be used for extensive pasture, game viewing and hunting, collecting medicinal herbs, recreation, water catchment, forest production etc. The interests of the different stakeholders must first be expressed and recognised and then reconciled as part of the land-use planning process.
With respect to the environmental impact of livestock keeping, major issues are:
• unfavourable changes in vegetation composition and structure as a result of overgrazing. In the drylands, vegetation yield may be low, not because of livestock but because of low rainfall, and vegetation may recover quickly when rainfall improves. Where grazing predominates, animals will die before the vegetation is irreversibly damaged, but supplementary feeding can result in overgrazing. This risk is much higher in mixed farming than in pastoral systems;
• nutrient flows within farming systems. This is particularly important in mixed farming systems. Some nutrients are added to cropland via manure, but this can impoverish the rangelands and may not offset the nutrient losses from cropping;
• nutrient accumulation in industrial systems, where concentrate feeds may even come from other continents. This is common in industrial landless systems and can pollute water or overload soils with nutrients. The tremendous regional imbalances jeopardise the ecological equilibrium in a way similar to pollution from heavy industry. Suggestions to focus on industrial animal production units and to spread animal wastes on fields of specialised crop farmers will face very narrow economic and ecological limits;
• transmission of disease among animals and between animals and humans. The widespread use of antibiotics, not only to prevent or cure diseases but also to promote animal growth, leads to the development of resistant bacteria and germs and may jeopardise the possibilities to use antibiotics to cure infections in humans. This is a particular risk in intensive, industrial systems of animal production. Also new diseases, such as BSE, and the increasing salmonella infections of food are mainly linked to industrial systems.
• high inputs of fossil energy and unfavourable input-output ratios. This is especially true for systems with high-yielding animals, requiring concentrates and forages produced with high fertiliser inputs. The current low price of fossil energy promotes its extensive use.
• biodiversity of flora and fauna, including that of domestic animals. Biodiversity is closely linked to eco-systems' resilience, yet understanding of these linkages is still in its infancy. With respect to domestic animals, the difference between nutritional and veterinary requirements between breeds and the subsequent difference in impact on the environment should be taken into account.
Care must be taken not to blame animal production for environmental impacts for which it is not responsible. Soil erosion and loss of biodiversity in animals and plants are often due to road construction, non-sustainable forest exploitation or mono-cropping, rather than to livestock-keeping.
Most of the issues mentioned above must be investigated on a local level. However, important decisions which affect economic and ecological development are made by national governments, e. g. regulations for land tenure and nature conservation, allocation of funds for infrastructure, or subsidies for economic and social activities. Regulations for exports are often made on an international level. These issues should not be neglected in assessing the impact of livestock at a specific site. For international negotiations and consultation (e. g. World Trade Organisation, UNCED), governments also need information about the pressures to which animals are exposed or which they exert on the environment. Finally, we should not forget that sustainable use of natural resources is a prerequisite for sustainable economic and social development.
3. Applying the Pressure-State-Response approach to screening and scoping the Environmental Impact Assessment.
Information of factors which shape economies and the precondition for their improvement must be evaluated in order to structure measures with respect to their impact on the environment and socio-economic security. This has to be done in a way that can be repeated in different settings to ensure comparability of results. The pressure-state-response approach (PSR) offers a method for screening and scoping the main aspect which should be incorporated into the assessment and monitoring of change. At the same time the results of the PSR provide a more comprehensive framework for the environmental impact assessment (EIA). The PSR approach clarifies the most important links between pressures on land quality caused by human activity on one hand and the resultant state of land quality on the other hand, including changes with respect to space and time. It also clarifies reactions of the affected groups of the population and the political and administrative institutions.
The key tasks in evaluating the available data are
A simplified PSR application for range management in the Middle East is shown in Figure 1. Great attention has to be paid to identifying stakeholders and their scope for action.
A modification to the PSR-model as developed by Adriaanse (1993) and Pieri et al (1996) includes time scales into the approach (Fig. 2) and gives an impression of the dynamics and time dimension of pressures, states and responses. Some may change quickly whereas others last over long periods. Structuring and weighing the effects in such an analysis is a prerequisite for monitoring and planning the paths to more sustainability.
4. Assessing the environmental impact of animal production on land quality and ecosystem resilience
The main areas of impacts on the environment by various types of animal production can already be identified with the PSR model. The results of a more in-depth analysis of these impacts can be summarised in an environmental impact study. From an Environmental Impact Assessment (EIA) we expect a systematic identification and evaluation of effects of either present practices or proposed projects, plans, programmes or legislative measures on physico-chemical, biological and - as e. g. wildlife reserves are concerned - socio-economic and cultural components of the human environment.
In the case of planned changes the procedure must:
It is important that causes and effects are clearly identified and also that “well-known facts” are critically examined. The “overgrazing” in the African Sahel mainly proved to be a consequence of climatic variability i.e. low rainfall in some years. There is very little lasting negative effect on the rangeland vegetation. The droughts, however, showed the importance of “social erosion” (breakdown of traditional mechanisms of coping with drought, changes in animal ownership, unfavourable changes in access to land in wetter areas), which left many pastoralists destitute.
According to our approach, EIA should concentrate on impacts of production systems on the ecology.
Figure 1: A simplified example of Pressure-State-Response Analysis (rangelands in Middle East and North Africa).
|Applying the pressure-state-response method|
|Land Use and Management||Pressures|
|State of Land Quality Change|
Initial → Modified
|→||Responses by Land Users, Managers, and Policy Makers|
|Pressures||state and changes||responses|
|-||High profitability of sheep fattening||-||Uneven access to pasture (rich pastoralists favoured because of water transport by truck which allows use of areas far from water points, whereas poorer pastoralists are restricted to areas closer to water points, which are more heavily grazed)||-||Breeding of more drought resistant crops|
|-||Cheap, subsidised concentrates||-||Decreasing plant cover and decline of palatable species||-||Removal of subsidies for concentrates|
|-||Attempts of state to settle nomads||-||Breakdown of traditional land use rights||-||Attempt to introduce group land rights, which restrict pasture users and season of grazing|
|-||Introduction of tractors and trucks||-||Breakdown of natural checks on animal numbers (e. g. seasonal lack of water, increased incidence of disease in areas with high numbers of livestock)||-||Protecting key areas so that indigenous plants can survive|
|-||Increasing human population||-||Many areas under cereal cultivation (with high cropping risks and therefore low inputs, land mining)||-||More law suites because of illegal cropping brought to court|
|-||Higher standard of living and less intensive use of animals (wool and meat only, milk less important) requiring more animals per family||-||Decreasing profitability of sheep fattening|
|-||Many alliances between traders (rich town people) and traditional leaders|
Figure 2: Modified “Pressure-State-Response” Analysis
|t-n||natural catastrophes wars, famines||→||LQ characterised by historical forms of sustainable land use and land degradation||→||differentiation of land tenure, change of traditional land-use practices|
|t||population increase poverty damaging resources||→||LQ affected by intensification (both nutrient mining and nutrient surplus), land degradation||→||technologies with high fossil energy inputs (e. g. import of feed grain from U.S.A. as animal feed), migration; change of land tenure and land-use, strengthening of stakeholder participation, international financial and technical assistance|
|t+m||shortage of capital, unfair distribution of resources and income||→||LQ characterised regional differentiation in advances in sustainability, ecological resilience at risk.||→||advances in education and agricultural extension, development of alternative land-use, information technology, division of labour, more democratic decision making, international co-operation|
key: LQ Land quality, => effects/modification,
t-n = past,
t = presence,
t + m = future.
Source: Adriannse, 1993, Pieri et al 1996, modified
Box 3: Prognosis of impact of water development on rangeland use in Southern Somalia.
Purpose: Assessment of impact of planned government boreholes on future range utilisation, as aid to decide on financial support in 2 southern Somalian regions. The water project was planned to allow better pasture utilisation.
Methods: Use of livestock statistics to estimate grazing pressure; use of rainfall data to estimate potential vegetation yield; use of satellite imagery to determine geophysical units and settlement pattern. Use of maps and observation to estimate distances between existing water points, interviews with pastoralists to get their perception of the environment and information on capacity and organisational aspects of use of existing water points.
Results: Livestock population: camels, goats, sheep and cattle. Livestock numbers indicate that pastures are heavily used in all but very wet years. For dry years feed shortage predicted (also mentioned by pastoralists). Existing network of water points (spring, shallow wells, natural depressions, dams to collect surface water, rivers) is such that pasture can be used (at least by camels) year-round, if there is feed. Heavy use of vegetation evident around water points. Degraded area dependent on capacity of water point. When water point is abandoned, evidence of quick recovery of vegetation. Traditionally access to pasture controlled by access to water. Traditional watering practices (use of wooded troughs and wooded containers to transpo rt water) exerts some pressure on environment (cutting trees to make containers. ) However most areas still rich in trees (and wildlife). Also on some water points contamination of water by faeces and urine from animals. Introduction of additional water points would undermine traditional control of pasture, encourage speculation and potentially lead to pasture degradation.
Recommendations: change of project design, respecting traditional water rights, some minor adjustment on
design of water points to prevent contamination, advice against additional water points.
Source: Salzgitter Consult 1989.
Generally, a difference can be made between indicators, for which a threshold value can be defined and those for which norms have to be defined. An example for threshold values is the nitrate in drinking water, although the threshold value approach can orient itself at international norms e. g. those of the World Health Organisation (WHO). However, as experience shows, even that is a very political question and needs to be negotiated.
An example of “norms” which have to be defined by stakeholders is the vegetation under grazing. This is closely linked to “land quality”. The vegetation we aim for depends on intended use and has to be defined accordingly.
Table 2 lists some issues, indicators and methods of measurements which can be used in EIA. This list is by no means comprehensive and some of the indicators are open to debate. For an EIA the list of issues and the number of indicators has to be more comprehensive than for monitoring. For monitoring purposes only few indicators can be selected, which should be easy to measure and to interpret.
5. Formulating objectives for sustainable livestock production
The results of the PSR analysis and of the EIA have to be combined into an overall assessment, which has to include consideration of sustainability. Sustainable use of natural resources implies a way of resource use, that does not exhaust resources and leaves them for subsequent generations. “The impact (I) of any group of people on the environment can be viewed as the product of population (P) multiplied by the per-capita affluence (A) as measured by consumption, which in turn is multiplied by a measure of the damage done by the technology (T) employed in supplying each unit of that consumption” (Hardin 1992).
Table 2: Some issues, indicators and methods of measuring indicators for EIA of animal production
|Issue||Indicators||Methods of measuring|
|Unfavourable change of vegetation||composition of vegetation, indicator plants, ground cover||evaluation transects/quadrates to measure frequency, estimate of ground cover within quadrates, transects|
|Nutrient flows within mixed farming systems||nutrient budget, water pollution, indicator plants||calculation of nutrient flows, chemical analyses, measuring frequency|
|Nutrient accumulation in industrial systems||nutrient budgets, water pollution, indicator plants||calculation of nutrient flows, chemical analyses, determine frequency of indicator plants|
|Transmission of disease among animals and between animals and humans||frequency of occurrence of disease, number of infected vectors, occurrence of resistant germs||veterinary statistics, epidemiological studies, post mortem of animals, surveys, laboratory analyses|
|Inputs of fossil energy and unfavourable input-output ratios.||energy budgets, input/output ratio-calculations||model calculations, some measurements to complement/verify models|
|Biodiversity of flora and fauna, including of domestic animals||composition of flora and fauna, indicator animals / plants, diversity of land-scape, species/breed composition of domestic animals||observations and systematic enumeration of animal/plant population, mapping of landscape elements, population analysis|
The direction of the most pressing actions for enhancing sustainability can easily be derived from the formula I = P × A × T. In developing countries, with low per capita consumption, but high population growth, the main thrust should go into reducing population growth, since every progress in reducing the damage done to the environment by (T) will be offset by increases in (P). On the other hand, in industrialised countries with low population growth yet increasing affluence (A), the main thrust has to be on reducing consumption of resources or the (T). This is especially true for per capita consumption of fossil energy. Furthermore recycling of raw material and reduction or avoidance of waste is called for. To promote this we may have to change our definitions of “economic growth” and “economic development” to focus on sustainability and sustainable development. The definition of land use should be extended and include settlements, transport routes and other use of land which can affect the ecological functions of an area. Sustainability also raises the issue of the permissible extent and the necessary distribution of various kind of land-use in the light of ecosystems' resilience.
Sustainability is not a state which once reached, will last for ever. Sustainable management of natural resources has to be understood as a process and as a continuos challenge. During this process environmental, economic and social objectives have to be taken into account (Box 4).
Box 4: Five pillars of sustainable land management
Sustainable land management combines technologies, policies and activities aimed at integrating socio-economic principles with environmental concerns in order to simultaneously:
The objectives of productivity, security, protection, viability and acceptability can be called the pillars of sustainable land management and must all be met, if sustainability is to be achieved. Meeting only one or several but not all, will result in only partial or conditional sustainability. Source: Smyth and Dumanski 1993.
Some specific objectives for animal production in relation to sustainability are:
Where grain is used, it should mainly be as a buffer in years of good harvests or when grain stores for human consumption have to be renewed.
Within the overall land-use system some land may be set aside as nature reserves, where natural flora and fauna can find a refuge. It is important that endangered species are not only protected as species but are protected within their habitat. With respect to nature conservation there is a marked difference between a livestock farm, where wildlife may be fed and simply takes the place of domestic animals in damaging the environment, and a true nature reserve, where habitats are protected and animal populations are restricted according to natural carrying capacity. Areas with limited use may also play a vital role in maintaining the water cycle, or provide space, where valuable species can survive, and later recolonize other areas.
There are few situations where all these objectives are achieved. In most situations it is advisable to identify elements of the livestock production system which are non-sustainable, and tackle these to strengthen the process of sustainable natural resource management.
6. Monitoring the track towards more sustainability with appropriate indicators
According to the five “pillars of sustainability”, indicators for monitoring can be divided into the following groups:
Productivity indicators are a measure of efficiency per unit input of labour, production factors, or investment.
Stability indicators serve as measurements for changes and refer to mainly physical issues. The stability indicators should provide information on the normal variation and fluctuations e. g. due to climatic variations but also of trends. Stability has to be seen in connection with resilience. Resilience indicators should show how a farming system can deal with stress (e. g. drought, floods, disease outbreak).
Protection indicators should show how vital ecological functions of a landscape are, such as the water cycle, the soils, or the biodiversity change. Many indicators shown in table 2 can serve as protection indicators.
Economic viability indicators are based on fairly standard economic calculations to show gross margin or profitability. In short-term it is profit, but medium- to long-term it is economic stability which matters.
Indicators for fairness of the distribution or equity depend very much on perceptions of local people. However, the contact with decision-making outsiders can be very selective, and therefore outsiders have the responsibility of knowing with whom they interact. Equity indicators can also refer to how various economic and social groups (crop farmers, pastoralists, large herd- or landowners, women, young people) are favoured or promoted by advice and support in the use of productive technologies (Guiterrez 1994, Müller 1996).
Who benefits from PSR, EIA and sustainability assessment and the use of appropriate indicators? - As can be expected the benefits meet very heterogeneous target groups. Some beneficiaries and potential benefits from information derived from PSR, EIA and SA are:
Farmers/Pastoralists: makes people more aware of the state of environment, facilitates comparison of their own problems and performance with that of neighbours and friends. May give indications of potential solution for local problems.
Local self-help groups and NGOs: gives arguments for supporting people in their struggle to improve economic productivity and ecological security and to oppose inappropriate interference in the local environment.
Extension services: helps creating more awareness of environmental problems and facilitates comparing farmers' perception of problems with that of extension services. May give guidance in revisions of recommendations.
Researchers: pinpoints gaps in information. May help to make research more relevant for local people. Helps evaluating the effects of new technologies on ecological, social, and economic sustainability.
National/International NGOs/Environmental pressure groups: gives them better founded information on relevant environmental issues in order to revise their own position and to put pressure on government and the international community to change rules and regulations and project support in order to encourage sustainable development.
Policy makers, administrators and donors: helps to evaluate effects of own policies and programmes. Gives inputs into reformulating policies; possible impact on legal structure (e. g resource tenure regulations).
7. Need for improving assessment methodology, data generation and dissemination of information.
The assessment and monitoring of the process of sustainable livestock development not only needs carefully selected indicators, it also calls for a reference system of desired values and of thresholds for critical developments. Indicators should not be developed by specialists in isolation, but should be developed with the stakeholders (farmers, other land users, extension agents, researchers etc. ). Outside specialists will not only have to provide their specialist knowledge but may also have to act as facilitators in the process of negotiation for defining appropriate indicators.
Over the last few years a number of approaches for EIA, including for agriculture, have been proposed, but few have been extensively used in practice. The concept of sustainable resource management as a mainstream concept is little more than a decade old. Sustainability proved to be a very important but bulky concept, and it is only now that more progress is made in making it operational. As already shown in the PSR analysis, the time dimension needs much more attention. Furthermore the problems in animal production are so divers that a single blue print for assessment procedures is hardly feasible. We recommend a process approach. The basic concepts, as outlined, should be clear and, depending on the severity of the problems and the data available, a more or less comprehensive analysis should be carried out. In many instances, it is important to start the process of PSR, EIA and sustainability assessment. If monitoring proves to be somewhat inadequate, it can always be improved and expanded. However, like with other data collection exercises, it should be clear why particular data are collected, who is going to analyse and use them. Large data cemeteries are of no help to anybody.
In some areas even very basic data, such as rainfall, are presently not readily available for planning. PSR, EIA and sustainability assessment will to a large extent always have to rely on secondary data (e. g. meaningful rainfall series should include a complete set of data from at least 30 years). The available data should be collected in regional (district/provincial) data bases and be readily accessible for planning purposes. This also includes data on vegetation, animal populations etc.
The results of impact studies should not be shut away in reports or scientific journals, but communicated better to the public. This should not be restricted to “bad news”, but also serve to correct misconception, like the misconception about rangeland degradation in pastoral areas in Africa. This could be done by approaching international journals (e. g. National Geographic), national journals in both industrialised and developing countries, television or international, national or regional radio transmission.
As already stated the problems in livestock x environment interactions and their causes are very divers - depending on eco-region and production systems. The amount of information available (and the level of concern, as shown in the e-mail conference preceding this “face-to-face” conference) is such that a global virtual network would be very difficult to handle, both for organisers and users. We suggest that regional networks are supported, e. g. one for intensive production in industrialised countries, one for range management in Africa etc., Existing structures such as UNSO, the Club du Sahel or the dryland programme of IIED should as far as possible be used for such purposes. If necessary, they should receive additional support from donors, rather than new structures being set up.
Adriaanse, A. (1993) Environmental policy performance indicators - A study on the development of indicators for environmental policy in The Netherlands; The Hague.
Barret D. F. D., Therivel R. (1996) Environmental policies and impact assessment in Japan. Routlegde, Chapman and Hall, New York
Becker B. (1997) Sustainability assessment: a review of values, concepts and methodological approaches. CGIAR/Worldbank, Washington D. C.
Canter, L. W. (1996)) Environmental Impact Assessment; Singapore,
Defoer, T., Kante S., Hilhorst Th., De Groote H. (1996) Towards more sustainable soil fertility management. ODI Agriculture Research & Extension Network, London
Food and Agriculture Organization (1976) A framework for land evaluation. FAO Soil bulletin 32. FAO, Rome
Food and Agriculture Organization (1993) Guidelines for land-use planning; FAO, Rome,
Guiterrez E. (1994) The approximate sustainability index: a tool for evaluation national performance. Paper presented at the Network Seminar on sustainable development indicators. ODI, London.
Hammond, A., Adriaanse, A., Rodenburg, E., Bryant, D., Woodward, R. (1995) Environmental Indicators: a systematic approach to measuring and reporting on environmental policy performance in the context of sustainable development. World Resources Institute, Washington D. C.
Hardin G. (1992) Living within limits - ecology, economics and population taboos. Oxford University Press, New York.
Müller S., (1996) How to measure sustainability: an approach for agriculture and natural resources. Inter-American Institute for Cooperation on Agriculture (IICA)/ Deutsche Gesellschaft für Technische Zusammenarbeit (GTZ).
Pieri, C., Dumanski, J., Hamblin, A., Young, A. (1996) Land Quality Indicators. The World Bank, Washington D. C.
Salzgitter Consult/German Water Engineering (1989) Study on waterpoints in Gedo and Bakool Regions, final report. Salzgitter/Lingen.
Smyth, A. J., Dumanski, J. (1993) FESLM: An International Framework for Sustainable Land Management. Food and Agriculture Organization (FAO), Rome.
Sombroek, W. G., Sims, D. (1995) Planning for sustainable use of land resources: Towards a new approach. Food and Agriculture Organization, Rome.