Contents - Previous - Next

Chapter 4. Methods and sources

A wide variety of methods are used in land-use planning They are taken from the natural sciences (climatology, soil science, ecology), from technology (agriculture, forestry, irrigation engineering) and from the social sciences (economics, sociology). Some of the methods, notably land evaluation, are interdisciplinary.

It is impossible to give detailed accounts of these methods in the guidelines. Many require handbooks of substantial length. The following notes indicate some of the principal sources in which such accounts can be found, details of which are given in the Bibliography, p. 93.

Land-use planning: general accounts

• Beatty, Peterson and Swindale (1978): the focus is on planning in developed countries.
• Davidson (1980): soils and land-use planning.
• FAO (1991b): includes accounts of 13 land-use planning applications, mainly in developing countries.
• ILRI (1980): includes detailed checklists of possible activities in regional planning.
• Laconte and Haimes (1985): water resources and land-use planning.

National land-use planning handbooks

National handbooks vary in content. Some are strongly focused on soil conservation, others include elements of land evaluation. Some examples are:

• Bangladesh: Brammer (1983).
• Brazil: Ramalho Filho et al. (1978).
• Canada: Lang and Armour (1980).
• Colombia: Vargas (1992).
• Ethiopia: FAO (1984b).
• Lesotho: Greenhow (1991).
• Sri Lanka: Dent and Ridgway (1986).
• United Republic of Tanzania: Corker (1983).
• Zimbabwe: Zimbabwe Federal Department of Conservation and Extension (1989).

Land-use planning applications

• OAS (1984): regional development planning.
• FAO (1991b): includes accounts of 13 land-use planning applications.

Information management

Decision-making depends on timely information on the present land-use situation, on possible ways of improving this situation and on the consequences of implementing each alternative solution. The gathering and storing of data requires much time in planning, but is not an end in itself. It is important to reserve time to interpret and apply these data to the task in hand. To manage information effectively, it is essential to know the place of each operation in the information system as a whole (Fig. 15). Otherwise, it is easy to concentrate on one part of the system, or one specialist task, without recognizing the implications it has for the whole operation. For example, if a lot of information is required, then a big investment in data collection, storage and reporting is necessary.

Good information management involves keeping a balance in the system. Information needs should drive data collection. Therefore, if there is a problem with grazing or drainage, focus attention there. Only collect data if it is known what they will be used for.

There must often be a trade-off between the excellence of the data and the time and cost of collecting them. Ways of making surveys cost effective include:

• Establish what is already available but check the reliability of the data.

• Collect data incrementally. Begin with a rapid overview of the whole area. Use this to identify those areas from which more detail is needed.

• Stratify the planning area. Divide the planning area according to the kind of information needed. This may reflect the potential land-use pattern or a physical characteristic that may be a limiting factor; for example in hill country, where land use is limited by steep slopes, detailed soil information may not be needed. Usually, the land mapping units delineated in Step 3 will provide a basis for stratification.

• Carry out a pilot study. Work through the whole planning process, or at least Steps 3 to 6, for a small representative area. This will identify more clearly the relationships between the collaborating workers and agencies. Discussions between all parties will be more fruitful when the output of several steps is available. For example, financial analysts may not be very clear about what natural resource information they require for a specific project until they have had experience in using it.

• Know the method of analysis. Design each survey with the method of analysis to be used in mind.

• Organize and store the data systematically, paying attention to:

- Quality control. Always list where and how data were obtained, in the field or from printed sources;

- Protection. If stored on paper, protect from fire, damp and insects; if stored on computer diskettes, keep backups, both on diskette and as printouts; prevent unauthorized people altering data;

- Updating. Record when data were last revised or updated.

Figure 15: Elements of a land-use planning information system

Box 10
Questions on the use of computerized methods of information management

Any organization contemplating the introduction of computer assistance for natural resources surveys, land evaluation and land-use planning should seriously consider at least the following ten questions (Burrough, 1986):

1. Identify your problems carefully. What do your customers or users expect of you and will a computer help provide the service they need?

2. How much money have you for investment and operation?

3. Have you trained staff available to operate the computer or to ensure the strict organization that efficient use of a computer requires?

4. Are you aiming at countrywide systems or a single-project system?

5. How much data and what kinds of data will you have to process at any one time?

6. What is the structure of your data? Will you have to interface with other types of data that may possibly have a different organization?

7. What is the quality of graphic output you can afford and what quality will your staff and clients accept?

8. Have you the necessary physical support facilities such as stable electricity supplies, air-conditioning and low-humidity rooms?

9. Can you collect data that are of a reasonable quality and worth the investment of a computer system?

10. What existing systems can you make use of? Can you acquire them in a modular way, allowing the gradual buildup of a comprehensive computer system? How permanent are those systems and how dependent will you be on a particular company or supplier for support when things go wrong or become obsolete?

A computer may not be necessary or cost effective. For small projects, it certainly will not. The most critical problem is likely to be the availability of trained staff.

Systems analysis

Data are often processed by means of systems analysis; that is, the analysis and modelling of interrelated processes. A system to be modelled must have defined boundaries. Within the system there are often stores of flows (of materials, energy, money, etc.). External flows cross the system boundaries as inputs and outputs. For example, in modelling nutrient dynamics in a plant-soil system, stores of nutrients occur in the plants, the organic soil fraction and the mineral soil fraction; internal flows refer, for example, to litter fall, humification and plant uptake from soil; and external flows to inputs such as atmospheric nitrogen fixation and outputs such as harvest of crops, or to the loss of nutrients in eroded soil. In modelling a farming system, some of the flows will refer to materials, such as seed and fertilizers, and others to energy or money.

Accounts of systems analysis, its potential and problems are given in:

• Arnold and Bennett (1975).
• Bennett and Thomas (1982).
• Biswas (1982).
• IIASA (1980).
• Morris (1977).
• National Research Council (1976).
• Quade and Miser (1982).
• Romero and Rehman (1989).
• Rossmiller (1978).

Geographic information systems

A geographic information system (GIS) is a computer-based system of storage and manipulation of data which is organized by area or location. Areas can be identified by a grid of cells (cell-based or raster systems), or information can be stored by means of the boundaries of mapped areas, e.g. land units or administrative units (polygon-based systems). A GIS enables different kinds of information to be recalled and combined; for example, areas that are both suitable for export crops and within a specified distance of an all-weather road could be overlain and mapped.

Most kinds of data processing undertaken on a GIS can also be done manually, by overlay of transparent maps, comparison and calculation. For small areas and few mapping units, this is the quickest way to do it. A GIS becomes efficient where there are numerous mapping units and many combinations of data are needed.

A GIS can offer valuable facilities to land-use planners. First, disaggregated data can be stored and retrieved by location. For example, crop yields may have been collected in order to calculate a financial measure of performance like the gross margin; these data can be stored and subsequently retrieved and used again for other purposes. Point data can be stored as such, rather than being lost by incorporation into mapping units. Thus, in a soil survey, data such as soil depth and texture, gathered for individual locations in the field, can be stored and retrieved for use in land evaluation. A further facility is to undertake complex and manually tedious calculations using any combination of the data in store. In this way, tables and maps of interpreted information can be produced very quickly. More important, the data can be updated or corrected and the methods of calculation revised by changing the computer program so that new maps and tables can be produced rapidly.

The cost of a GIS is now low and quite powerful systems can be run on personal computers. Systems have been developed for land-use planning, ranging from those that are relatively simple and easy to use (e.g. Ridgway and Jayasinghe, 1986) to complex ones (e.g. Wood and Dent, 1983; Schultink, 1987).

Accounts of the nature and potential of GISs are given in Burrough (1986) and Maguire, Goodchild and Rhind (1991). The IDRISI system is relatively easy to use and its capacity is substantial. The CRIES system (Schultink, 1987) and the ILWIS system (Valenzuela, 1988) are specifically designed for land resource evaluation. A powerful but very complex system is ARC / INFO.

Natural resource surveys

There is a large number of publications on the survey of natural or environmental resources: following is a selection.

• Bunting (1987): a collection of methods that have been used for agro-ecological characterization and classification; see especially the chapters by Young (1987) and Brinkman (1987).

• Dent and Young (1981): an account of soil survey methods, suited to different scales and purposes, and of land evaluation, including a comparison of the FAO framework with other methods.

• FAO (1979b): soil survey methods for irrigation planning.

• FAO (1984c, 1987): agroclimatological data for Africa and Asia, respectively.

• Landon (1991): a useful reference for many aspects of soil evaluation and classification

• Carver (1981): air photography in land-use planning.

• Lindgren (1985): applications of remote sensing methods in land-use planning.

Box 11
Climatic data for land-use planning



• Sufficiency of energy

• Temperature regime, sunshine hours, day length

• Frost hazard

• Probability of frost (local occurrence - not adequately recorded in standard data)

• Sufficiency of water

• Reference evaporation Eo
Crop water requirement = Eo x crop coefficient
Rainfall probability
Effective rainfall

• Irrigation need/drought hazard

• Rainfall probability - crop water requirement

• Length of growing season

• Period of energy and water sufficiency

• Hazard of high winds, high temperature, hail, low humidity

• Probability of occurrence in the growing season

• Erosion hazard

• Rainfall intensity

The Agricultural Studies Unit of CIAT has created land system and agroclimatic databases to support agricultural research management. These and complementary agronomic techniques help CIAT in the selection of high-yielding crop varieties with farm-effective organic and mineral fertilizer recommendations for a given ecosystem, while contributing to the successful conservation and use of soil resources in tropical South America (Cochrane et al., 1984).

Rural land-use analysis

Three methods have been described for the analysis of problems of rural land use: farming systems analysis, diagnosis and design and rapid rural appraisal. These have much in common: all are centred on interviews with a sample of rural land users, preferably stratified according to identified classes of farming system. The methods are not confined to problem diagnosis but include elements of later steps in land-use planning, particularly the design of improved land-use types and social analysis.

Farming systems analysis. (Fresco et al., 1992; FAO, 1991b, p. 147-152.) This is centred on the identification of farm-level constraints and aims to develop adapted technologies for specific farming systems. The publications cited outline how it can be combined with land evaluation as an integrated sequence, i.e. land evaluation and farming systems analysis (LEFSA).

Diagnosis and design (D&D). (Raintree, 1987a; Young, 1986.) This approach was developed specifically for the design of agroforestry systems but can be applied to other types of land use. Diagnosis means the identification of problems with land-use systems and the analysis of their causes; design is the formulation of promising land-use types that might help solve these problems. The analogy is with the medical profession where a doctor must diagnose an illness before it can be treated. One way in which diagnosis and design can be integrated with land evaluation procedures is given in Young (1986).

Rapid rural appraisal. (Abel et al., 1989; FAO, 1989a; McCracken, Pretty and Conway, 1988). This approach is intended as a relatively rapid way of acquiring (in a matter of weeks) essential information on existing rural land-use systems, including the problems they entail.


There is a large and increasing number of computer models relevant to different aspects of land-use planning. Most models consist essentially of quantitative predictions based on input data, for example the prediction of plant evapotranspiration from weather data or the prediction of net present return from data on inputs, production, costs and prices. Note that:

• models are only as reliable as the data which are entered into them;

• wherever possible, models should be calibrated for the planning area, its climate, soil types, etc.; data should be entered and the results compared with an independent measure, for instance crop yield.

Box 12
Water resource data for land-use planning

• Present water use

- River abstraction, tanks, groundwater
-Location of abstraction points, sluices, dams, wells and boreholes, with yields

• Present storage capacity of tanks and reservoirs

• Reliable yield of water for each river catchment - 75% and 90% probability low flow (from hydrograph records) or 75% and 90% probability rainfall - Eo over seven- or ten-day day periods x area of catchment

• Safe yield of groundwater (from test pump data or well records)

• Depth below surface of useful groundwater

• Location of aquifers

• Water quality

• Location of irrigable land

• Legal and customary rights

Figure 16: Example of the relationship between a land characteristic and plant growth: the effect of soil pH on the growth of storage roots of sweet potato

Source: Hackett (1988).

Results from modelling can be combined with a GIS to show the spatial extent of the effects modelled (e.g. crop yield, tree growth).

Examples of the purposes for which modelling has been applied in land-use planning are:

• Agriculture (Heady and Srivastara, 1975).

• Crop growth, for example the CERES/DSSAT set of models (Jones and Kiniry, 1986); WOFOST (van Diepen et al., 1988).

• Crop water requirements (FAO, 1977; 1979a).

• Decision-making in land-use planning (Cocks et al., 1983; Ive, 1984; Ive and Cocks, 1987).

• Forestry and agroforestry (Davey, Prinsley and White, 1991).

• Land evaluation, e.g. automated physical land evaluation (APLE); automated land evaluation system (ALES) (Beck, Burrough and McCormack, 1987; Higgins et al., 1987; van Keulen et al., 1987).

• Soil erosion, e.g. the universal soil loss equation (USLE) (Wischmeier and Smith, 1978); soil loss estimator for southern Africa (SLEMSA) (Elwell and Stocking, 1982).

• Soil response to land use, e.g. soil changes under agroforestry (SCUAF) (Young and Muraya, 1990); CENTURY (Parson et al., 1989).

• Descriptions of a range of models applicable, with references, are found in Bunting (1987) and Davey, Prinsley and White (1991).

Land evaluation

Land evaluation in its broad sense covers Steps 1 to 6 in Chapter 3, from the setting of goals to land suitability evaluation, including environmental, economic and social analysis. It has been most widely applied as qualitative (physical) land evaluation, as in Step 5. Among the information to be found in land evaluation handbooks are checklists of descriptors for land-use types, land qualities and land characteristics as relevant to different kinds of land use.

The basis of the approach is described in FAO (1976). Other accounts are given in Dent and Young (1981) and McCrae and Burnham (1981). Recent developments in quantitative land evaluation, including computer programs and modelling, are described in Beek, Burrough and McCormack (1987). Detailed guidelines are available on land evaluation for:

• rain-fed agriculture (FAO, 1983);
• irrigated agriculture (FAO, 1985a);
• forestry (FAO, 1984a);
• extensive grazing (FAO, 1991b).

Requirements for plant growth

The FAO Soil Resources, Management and Conservation Service is establishing a two-level database (ECOCROP 1 and 2) covering the ecological requirements and responses of plants, with emphasis on economic crops. ECOCROP 1 which, by July 1993, contained data for 1200 species, identifies arable crop, pasture and tree species for defined environments.

In ECOCROP 2, designed to support a wide range of existing and future models, information is held in the form of pairs of coordinates representing the response of a whole plant or plant process at a given level of an environmental factor, with specified values for the other factors. For example, growth rate at a given temperature or rate of photosynthesis at a given light intensity. The aim will be to have at least three or four of such pairs in order to define a response curve.

At an intermediate level, empirical relationships for plant/ environment response have been collected for a large number of plants by Hackett (1988).

Some countries have begun to collect data on plant growth requirements at a national level under the direction of national soil survey or land-use planning organizations. Other local systems will be found in previous land evaluation surveys. Criteria should not be taken uncritically from previous surveys but rather examined and, if possible, tested.

Financial and economic analysis

Financial and economic analysis for the purpose of land-use planning uses essentially the same basic methods as do other kinds of project analysis. The foundation of this method is set out in Gittinger (1982) and Bridger and Wipenny (1983). The application of economic analysis to natural resources is discussed in Pearce and Turner (1990), Wipenny (1991) and Whitby and Willis (1978).

A specific problem encountered is that of choosing the discount rate for investments of which the returns will not be received for many years, e.g. most kinds of forestry. This is discussed by Leslie (1987). The application of economics to land-use planning is discussed in Harrison (1977).


• Romero and Rehman (1989).

People's participation

• Huizer (1983).

Land tenure

• FAO (1989b): appraisal of tree and land tenure.
• Raintree (1987b): land and tree tenure in agroforestry.
• Dale and McLaughlin (1988): cadastral aspects of land-use planning.


• Mollett (1984).

Legislation for land use

• FAO (1971).
• Roberts (1977).
• FAO (1985b).

Contents - Previous - Next