C J Ritsema, Alterra, Wageningen, The Netherlands G W J van Lynden, ISRIC, Wageningen, The Netherlands V G Jetten and S M de Jong, Utrecht University, Utrecht, The Netherlands
ß 2005, Elsevier Ltd. All Rights Reserved.
are the result of human disturbance of either a natural or anthropogenic state of equilibrium. Some of these are: Agricultural causes: Defined as the improper management of cultivated arable land. It includes a wide variety of practices, such as insufficient or excessive use of fertilizers, shortening of the fallow period in shifting cultivation, use of poor quality irrigation water, absence or bad maintenance of erosion-control measures, improper use of heavy machinery, etc. Degradation types commonly linked to this causative factor are erosion (water or wind), compaction, loss of nutrients, salinization, and pollution (by pesticides or fertilizers). Deforestation or removal of natural vegetation: Defined as the near complete removal of natural vegetation (usually primary or secondary forest) from large stretches of land, for example by converting forest into agricultural land (hence sometimes followed by agricultural mismanagement), large-scale commercial forestry, road construction, urban development, etc. Deforestation often causes erosion and loss of nutrients. Overexploitation of vegetation for domestic use: Contrary to ‘deforestation or removal of natural vegetation,’ this causative factor does not necessarily involve the (near) complete removal of the ‘natural’ vegetation, but rather a degeneration of the remaining vegetation, thus offering insufficient protection against erosion. It includes activities such as excessive gathering of fuel wood, fodder, (local) timber, etc. Overexploitation of natural water resources: This leads to water shortages for the natural ecosystem and in the long term to the removal of the natural vegetation cover. The result is an increased vulnerability of the land for surface runoff, soil erosion, and soil surface crusting. As soon as the process of vegetation deterioration starts, it normally has a self-enhancing effect which is difficult to stop or to reverse. Overgrazing: Besides actual overgrazing of the vegetation by livestock, other phenomena of excessive livestock amounts are also considered here, such as trampling. The effect of overgrazing usually is soil compaction and/or a decrease in plant cover, both of which may in turn give rise to water or wind erosion. Industrial activities: All human activities of an industrial or bioindustrial nature are included: industries, power generation, infrastructure and urbanization, waste handling, traffic, etc. It is most often linked to pollution of different kinds (either point source or diffuse) and loss of productive function.
Soil is under increasing threat from a wide range of human activities that are undermining its long-term availability and viability. One third of the world’s agricultural soils, or approximately 2 billion hectares of land are affected by soil degradation. Water and wind erosion account for most of the observed damage, while other forms such as physical and chemical degradation are responsible for the rest. Appropriate soil and water conservation strategies are needed to prevent and combat the effects of soil degradation in the field and at the planning level. Soil degradation is ‘‘a process that describes humaninduced phenomena which lower the current and/or future capacity of the soil to support human life.’’ In a general sense, soil degradation could be described as the deterioration of soil quality, or in other words: the partial or entire loss of one or more functions of the soil. Quality should be assessed in terms of the different potential functions of the soil. Land degradation is the reduction in the capability of the land to produce benefits from a particular land use under a specified form of land management. Seven main groups of land-degradation processes are normally distinguished: (1) mass movement (such as debris flows and avalanches), (2) water erosion (sheet, rill, gully erosion), (3) wind erosion, (4) excess of salts (salinization, sodification), (5) chemical degradation (acidification, contamination, toxicity), (6) physical degradation (crusting, compaction, oxidation), and (7) biological degradation (loss of soil biodiversity). An important aspect of many soil and land degradation processes are the so called off-site effects; for example, dust storms or eroded sediment cause problems such as damage by mudflows, siltation of dams, or pollution of drinking water in downwind or downstream areas.
Factors and Processes Affecting Degradation of Soils
Various types of human activities may lead to soil degradation. Although some degradation processes may also occur naturally, many degradation types
Types of Soil Degradation
The type of soil degradation refers to the nature of the degradation process. Soil particles may be displaced by the action of water or wind (erosion and sedimentation), which may cause damage to crops, infrastructure, buildings, and the environment in general. Erosion can be linear, i.e., concentrated along certain channels (rill or gully erosion and mass wasting such as landslides), sometimes creating very deep scars in the landscape (Figure 1). Less conspicuous, but often even more detrimental to crops is the gradual removal of the topsoil layer (sheet erosion). Off-site effects of erosion may consist of siltation of reservoirs and river beds and/or flooding, or dune formation and ‘overblowing’ in the case of wind erosion. Degradation in situ, i.e., without movement of soil particles, can be chemical (soil pollution by chemical wastes or excessive fertilization; fertility decline due to nutrients being removed by harvesting, erosion and leaching; salinization due to irrigation with saline groundwater and/or without proper drainage in semiarid and arid areas, acidification due to pH-lowering additions to the soil from fertilizers or from the atmosphere), or physical (compaction due to the use of heavy machinery; deteriorating soil structure such as crusting of the soil surface; waterlogging due to increased water table or its opposite, aridification).
erosion status for a watershed or the pollution status for a province. Qualitative assessments are based on expert judgement and hence more liable to subjectivity than quantitative methods. A method does not have to be fully qualitative or quantitative, mixtures may occur. Some frequently used methods or tools are: 1. Expert opinion: Qualitative assessment on a controlled mapping base and semiquantitative definitions, as employed for instance in the Global Assessment of Human-induced Soil Degradation (GLASOD) survey. GLASOD and related methods are based on an assessment of land suitability by national experts that use defined, semiquantitative class limits on a given mapping base. Its major disadvantage is the inevitable degree of subjectivity. Its major advantage is its capacity to produce results, such as achieving complete world coverage (Figure 2), in a short time and on a small budget. Costs per unit area are relatively low. In Figure 3 an integrated global soil degradation severity map is shown, indicating areas with different degradation rates; 2. Remote sensing: Analysis of low- and highresolution satellite data and airborne imagery (e.g., analysis of composite indices such as the Normalized Difference Vegetation Index (NDVI)). Remote sensing always includes linkages with ground observations. The basis of this method is comparison of remotely sensed imagery of different dates, for regional coverage, mainly low-resolution imagery; and, specifically, comparison of the NDVI, derived from imagery collected by the sensor aboard the National Oceanographic and Atmospheric Administration (NOAA) satellite, and more detailed imagery. This method was tested amongst others in Saudi Arabia and shows areas where vegetation response to rainfall is decreasing (degradation of resources) or increasing (rehabilitation of resources). It has been applied particularly to early warning systems. For longer-term comparisons, some form of calibration for preceding rainfall is needed. Costs are relatively low. It is recognized that remote sensing cannot be used alone. Spectral mixture analysis (SMA): Since 1985 hyperspectral remote sensing has been developed, opening new methods to survey and assess degradational state of the soil surface. Hyperspectral remote sensing refers to the collection of images in the solar spectrum, with many narrow spectral bands allowing the collection of very accurate spectra of objects and the earth surface and identification of absorption features of plants and of soil minerals in these spectra. SMA is a technique to unravel the spectral information in the remote-sensing images by assuming that the spectral variation is caused by a limited number of surface material (green vegetation, senescent vegetation, a
Assessment of Degradation
The status of soil degradation can be assessed in a qualitatively broad manner or in a more detailed quantitative manner. The former generic approach is better suited for small-scale assessments, such as for entire countries, continents, or global overviews. A quantitative approach is required for more specific and detailed assessments, e.g., to determine the
Figure 1 Severely degraded soils on the Loess Plateau of China.
Degradation severity (Extent + Degree)
Low Medium High Very high
Low Medium High Very high
Chemical deterioration Physical deterioration
Low Medium High Very high Low Medium High Very high
Nonused wasteland Ocean, inland water
Figure 2 Global assessment (in 1990) of the status of human-induced soil degradation. (Reproduced with permission from Oldeman LR, Sombroek WG, and Hakkeling R (1991) World Map on the Current Status of Human-Induced Soil Degradation. An Explanatory Note, 2nd edn. Wageningen, the Netherlands: ISRIC/Nairobi, Kenya: UNEP.)
Figure 3 Global soil degradation severity map as produced by the GLASOD initiative.
number of soil types, and water). A reference library of these surface materials collected in the field or in the laboratory yields the basis for SMA of the remotesensing images. This approach has been applied successfully in a number of case studies to survey soil conditions and to identify classes of degradation. The SMA approach normally improves on results using the NDVI but requires more spectral bands: SMA is successfully applied to separate, in images, bare soil surfaces from senescent vegetation and yellow vegetation
from green vegetation. These three factors are important inputs in soil-erosion models because they act differently with respect to raindrop interception. 3. Field monitoring: Stratified soil sampling and analysis, and field observation of vegetation and biodiversity under certain land-use or management practices and climate variability. To date, soil monitoring has been applied mainly in developed countries, and tests are needed of its cost-effectiveness in developing countries. In areas where baseline studies have
been established, monitoring of changes will be undertaken; in other areas, establishment of a baseline will be a priority. Stratified soil-sampling with analysis, and/or benchmark sites, repeated over 5- to 10-year intervals, has been advocated as a basic activity for national soil survey organizations. Examples of application to date (2003) are few, but successful: the method has been applied to 20 000 sites over a 25-year period in Japan; is currently being used for a national 16-km-grid in France; and has been started in Denmark and Switzerland. The same approach has been applied to field observations of vegetation, along transects or in sampling plots, and to biodiversity. Costs per unit area are relatively high, but could be reduced by application to priority areas only, on a stratified sampling basis. 4. Productivity changes: Observation of changes in crop yields, biomass production, and livestock output, which directly apply to the definition of land degradation in terms of lowered productivity, although they are influenced by many other factors. There is a range of possibilities: At national level, use might be made of national yield statistics (of which the reliability is still under debate), adjusted for fertilizer use and climate. At local level, yield monitoring is possible by comparisons with a standard crop, either without fertilizer or with standard fertilizer and management. Substantial problems arise in that productivity decline could be due to factors other than land degradation, e.g., removal of fertilizer subsidy or civil strife. The same cost constraints apply as for soil monitoring. 5. Sample studies at farm level, based on field criteria and the expert opinion of land users. Even at national level, such detailed studies are essential on a sample basis, to obtain grass roots views both of the severity of degradation and its causes, together with practicable remedies (Stocking and Murnaghan, 2001). Field indicators of soil degradation were developed about 20 years ago, and could be extended to condition of vegetation. Talking with farmers means getting the views of farmers, and other land users, on whether things have got worse – which are of course, subjective and perhaps systematically biased, but still essential to get grass roots view at local level. The method is clearly applicable only at a local scale, and thus on a selective sampling basis. Observations of the state of the land can be combined with assessment of driving factors and impacts. 6. Modeling: Based on data obtained by other methods, modeling can be used in many ways, such as: (1) prediction of degradation hazard; (2) operational definition of degradation in terms of unfavorable changes in plant productivity, soil properties, and hydrology; (3) design of conservation
measures using climatic data with a specific return period (worst-case scenario modeling); (4) extending the range of applicability of results; (5) integrating biophysical with socioeconomic factors. Much research has been put into devising models for the prediction of soil-erosion hazard. There are established methods for the modeling of both water and wind erosion, which have been widely applied, in part because it is vastly cheaper than any form of field observation. The modeling approach is mainly relevant to degradation hazard, but can be applied to actual degradation first, as a means of calibration of the model to the specific requirements of an area, optimizing sampling design, or to extrapolate the applicability of results obtained on a sampling basis. Risk reflects a potential development in the future, while status reflects the development to date. Models vary widely in complexity and data requirements, depending on the type of degradation they are addressing and the size of the area under investigation. Models are useful to learn and understand degradation processes, but both the model and input data are a simplification of reality, hence extrapolation of models should be done with care. Very often models are developed for experimental plots or pilot zones of a restricted size and under more or less controlled conditions, which should be taken into account when applying the model elsewhere. The data requirements and structure of a model, and the type of processes included in it, depend on many things: (1) the temporal scale of the research objectives: Is an annual, daily, or event-based result required? (2) the spatial scale: Are predictions needed for a single plot, a field, complex spatial catchment, or an entire region? (3) Is the emphasis on the on-site effects of land degradation (e.g., soil erosion or crop yield changes) or on the off-site effects such as water sediment levels and pollution? Spatial and temporal scales are often linked, as, for example, is the case for physically based spatial erosion models (Figure 4) that simulate single events for first-order catchment with a high level of detail. They can be used to answer subtle questions about the effects of specific land-use changes or soil and water conservation measures in the catchment upon reducing runoff and erosion (Figure 5). On the other hand there are lesscomplex empirical models that can simulate continuous periods mostly for fields or hillslopes, but they can only show the change in annual erosion or soil loss.
Potentials and Limitations
It is useful to emphasize some potentials and limitations of land degradation assessments. It is obvious that an assessment at a small scale (e.g., 1:1 M) does not have a direct value for activities at the field level,
but can be highly useful (if well done) to planners, government agencies, legislative bodies, educational institutions, nongovernment organizations (NGOs), and the general public in highlighting (potential) problem areas and decision-making for further action. Besides geographic coverage and scale, another factor that determines the usefulness of an assessment methodology is the range of degradation issues the assessment tries to cover. Land degradation is a very broad issue, covering a wide range of degradation
issues, which makes its assessment a rather generic or alternatively highly unwieldy exercise. It is already quite complicated to assess one specific type of soil pollution, not to mention the various other types of soil degradation, which in itself is just one aspect of land degradation. This also means that the frequently observed desire to have ‘simple’ assessment methods is not entirely realistic, if this is supposed to be anything more than just a general awareness-raising tool. Two types of assessments have been identified one is ‘backward-looking’ and determines the result of degradation over a recent past period. The other approach is forward-looking in the sense that it makes predictions for the future based on models and scenarios. Although the backward-looking approach considers the current status, it does not necessarily reflect the result of the degradation process over that period, but the net result of a number of acting and counteracting factors. Degradation is one of these, but remedial activities compensating the degradation effect to some extent is another one. Though the wish to have ‘simple’ degradation assessment methods is often expressed, it should be realised that soil degradation is a complex process, determined by a range of factors of a natural and socioeconomic character. Hence a simple method will tend to correspond less with reality than a more complex and comprehensive one.
Figure 4 Model structure of a physically based spatial hydrologic and soil-erosion model, in which water and sediment are routed to the outlet of a catchment and produced as discharge. Input var, input variable; LAI, leaf area index; Cov, soil cover; Ksat, hydraulic conductivity; theta, moisture content; RR, surface roughness; ldd, runoff network; n, flow resistance; slope, terrain slope; As, aggregate stability; COH, Cohesion; D50, median grain size of suspended sediment.
Degree and Impact of Degradation
Degree of Degradation
Degree is defined as the intensity of the soil degradation process, e.g., in the case of erosion, the amount
Changed land use with external support
Present land use
Negligible erosion Slight erosion Moderate erosion Serious erosion Severe erosion
(0−2.5 t ha−1) (2.5−10 t ha−1) (10−25 t ha−1) (25−100 t ha−1) (100−2000 t ha−1)
Figure 5 Computed soil losses in a first-order watershed for the current land use and management conditions and for an alternatively defined land-use distribution.
of soil washed or blown away. The FAO has proposed values for maximum acceptable limits of soil loss by erosion with respect to decreased agricultural productivity. Four classes are distinguished, ranging from no loss of productivity to severe loss of productivity. The classes are <12, 12–25, 25–50, and >50 t haÀ1 yearÀ1. Relative changes of the soil properties are other good indicators of soil degradation: the percentage of the total topsoil lost, the percentage of total nutrients and organic matter lost, the relative decrease in soil moisture-holding capacity, changes in buffering capacity, etc. However, although such data may exist for experimental plots and pilot areas, precise and actual information is often lacking at a regional scale.
Rate of Soil Degradation
The recent rate of degradation relates to the rapidity of degradation over the past 5–10 years or, in other words, the trend of degradation. A severely degraded area may be quite stable at present (i.e., low rate, hence no trend toward further degradation), while other areas that are now only slightly degraded may show a high rate, hence a trend toward rapid further deterioration. From a purely physical point of view, the latter area would have a higher conservation priority than the former. Areas where the situation is improving (through soil conservation measures, for example), can also be identified. A comparison of the actual situation with that of the preceding decade may suffice, but often it is preferable to examine the average development over the last 5–10 years to level out irregularities. Whereas the degree of degradation only indicates the current, static situation (measured by decreased or increased productivity compared with some 10–15 years ago) the rate indicates the dynamic situation of soil degradation, namely the change in degree over time.
Impact of Degradation
Impact refers to the effects of soil degradation on the various soil functions. Changes in soil and terrain properties (e.g., loss of topsoil, development of rills and gullies, exposure of hardpans in the case of erosion) may reflect the occurrence and intensity of soil degradation but not necessarily the seriousness of its impact. Removal of a 5-cm layer of soil may have a greater impact on a poor shallow soil than on a deep fertile soil. The impact depends on the function and/ or use of the soil: a heavily compacted soil is unsuitable for agriculture, but may be an appropriate basis for road construction. Whereas the degree of degradation mainly refers to the degradation process, the impact of degradation can be manifold, depending on the current function
(or use) of the soil. In many cases, the impact of degradation types will be on its biotic functions, or more specifically on its productivity. A significant complication in indicating productivity losses caused by soil degradation is the variety of reasons that may contribute to yield decline. Falling productivity may be caused by a wide range of factors such as erosion, fertility decline, improper management, drought, or waterlogging, quality of inputs (seeds, fertilizer), pests, and plagues, often in combination with each other. However, if one considers a medium- to longterm period (e.g., 25 years), large aberrations resulting from fluctuations in the weather pattern or pests should be leveled out. The effects of soil degradation can be partially hidden by various management measures such as soil conservation, use of improved varieties, fertilizers, and pesticides. Some of these inputs are used to compensate for the productivity loss caused by soil degradation, for example application of fertilizers to compensate for lost nutrients. In other words, yields could have been much higher in the absence of soil degradation (and/or costs could have been reduced). Therefore, productivity changes should be seen in relation to the degree of input or level of management. The latter may include use of fertilizers, biocides, improved varieties, mechanization, various soil conservation measures, and other important changes in the farming system. Changes in productivity should be expressed in relative terms, i.e., the current average productivity compared with the average productivity in the nondegraded situation and in relation to inputs. For instance, if previously an average yield of 2 t of wheat haÀ1 was attained while at present only 1.5 t is realized in spite of high(er) inputs – and all other factors being equal – this would be an indication of strong soil degradation. Sometimes the impact may be ranked as negligible, even when degradation occurs, because of the capacity of the soil to resist a certain amount of degradation. Although for most degradation types the dominant impact is on productivity, some types (pollution in particular) may have additional or different impacts, e.g., on human or animal health or on entire ecosystems.
Preventing and Combating Degradation
There are a wide variety of measures to prevent or combat land degradation. These measures are generally known as soil conservation or soil and water conservation (SWC), especially when related to aspects like erosion, soil-moisture problems and soil fertility. More broadly applicable are names such as land husbandry or sustainable land management.
The WOCAT (World Overview of Conservation Approaches and Technologies) network, which constitutes an international consortium of institutions and individuals from all over the world, provides an evaluation tool for SWC activities, an informationmanagement system designed to collect, analyze, present, and disseminate SWC knowledge and a decision-support system designed to assist in the search for SWC options appropriate to the prevailing biophysical and socioeconomic settings. WOCAT was initated in 1992 and has developed a common framework and methodology, consisting of three comprehensive questionnaires (in English, French, and Spanish) for the documentation and evaluation of SWC. The WOCAT methodology consists of three major modules: 1. Questionnaire and database on SWC technologies; 2. Questionnaire and database on SWC approaches; 3. Questionnaire and database on the geographic distribution of SWC (mapping). The first two modules aim at a comprehensive and detailed description of specific technologies, i.e., agronomic, vegetative, structural, and/or management measures used in the field (Figure 6), and the ways and means used to implement an SWC technology on the ground. The mapping module is more or less similar to the qualitative methodology for degradation assessment described earlier. In this approach, information is collected for individual units of a (physiographic or other) base map on the following items: . Land use: type, extent, trend in area, trend in intensity; . (Per land use type) degradation, as above, but only for water and wind erosion and fertility decline in erosion-prone areas;
Figure 6 Categorization of soil and water conservation measures according to the World Overview of Conservation Approaches and Technologies (WOCAT) initiative: (a) agronomic measures such as mixed cropping, contour cultivation, mulching, etc. which: (1) are usually associated with annual crops, (2) are repeated routinely each season or in a rotational sequence, (3) are of short duration and not permanent, (4) do not lead to changes in slope profile, (5) are normally not zoned, (6) are normally independent of slope; (b) vegetative measures such as grass strips, hedge barriers, windbreaks, etc. which: (1) involve the use of perennial grasses, shrubs, or trees, (2) are of long duration, (3) often lead to a change in slope profile, (4) are often zoned on the contour or at right angles to wind, (5) direction,
(6) are often spaced according to slope; (c) structural measures such as terraces, banks, bunds, constructions, palisades, etc. which: (1) often lead to a change in slope profile, (2) are of long duration or permanent, (3) are carried out primarily to control runoff, wind velocity, and erosion, (4) require substantial inputs of labour or money when first installed, (5) are often zoned on the contour/against wind direction, (6) are often spaced according to slope, (7) involve major earth movements and/or construction with wood, stone, concrete, etc.; (d) management measures such as land use change, area closure, rotational grazing, etc. which: (1) involve a fundamental change in land use, (2) involve no agronomic and structural measures, (3) often result in improved vegetative cover, (4) often reduce the intensity of use; (e) combinations in conditions where they are complementary and thus enhancing each other. Any combinations of the above measures are possible, e.g.: structural: terrace, with vegetative: grass and trees, with agronomic: ridges.
WOCAT data available Preliminary WOCAT data Agronomic SWC measure: 1 − conservation tillage Vegetative and management SWC measure: 2 − agroforestry 3 − afforestation, forest protection 4 − rotational system 5 − grazing land management: control stocking rates, enclosure, reseeding Combination of structural and vegetative SWC measure: 6 − contour bunds, grass strips, forward-sloping terraces Structural SWC measure: 7 − irrigation terraces 8 − stone terraces 9 − water-harvesting: microbasin, small ponds, ditches
Figure 7 Reported soil conservation measures as recently compiled by the WOCAT initiative.
. (Per land use type) conservation: type, extent, period of implementation, effectiveness, trend in effectiveness, and reference to a corresponding questionnaire in the technology database for more detailed information; . (Per land use type) productivity: trend, contribution of SWC or degradation to this trend, average production value, average input value. An overview of soil and water conservation measures applied in different geographic regions of the world is shown in Figure 7.
See also: Desertification; Erosion: Water-Induced; Wind-Induced; Salination Processes
Blaikie PM and Brookfield HC (1987) Land Degradation and Society. London, UK: Methuen. Blum WEH (1988) Problems of Soil Conservation. Nature and Environment Series 39. Strasbourg, France: Council of Europe. Crosson P (1997) The on-farm economic costs of erosion. In: Lal R, Blum WEH, Valentin C, and Stewart BA (eds) Methods for Assessment of Soil Degradation. Boca Raton, FL: CRC Press. Dregne HE (1997) Desertification assessment. In: Lal R, Blum WH, Valentin C, and Stewart BA (eds) Methods
for Assessment of Soil Degradation. Boca Raton, FL: CRC Press. Lal R, Blum WH, Valentine C, and Stewart BA (eds) (1997) Methods for Assessment of Soil Degradation. Boca Raton, FL: CRC Press. Liniger HP, van Lynden GWJ, and Schwilch G (2002) Documenting field knowledge for better land management decisions – Experiences with WOCAT tools in local, national and global programs. Proceedings of ISCO Conference 2002, vol. I, pp. 259–267, Beijing, China: Tsinghua University Press. Oldeman LR, Sombroek WG, and Hakkeling R (1991) World Map on the Current Status of Human Induced Soil Degradation. An Explanatory Note, 2nd edn. Wageningen, the Netherlands: ISRIC/Nairobi, Kenya: UNEP. Ritsema CJ (ed.) (2003) Soil erosion and participatory land use planning on the Loess Plateau in China. (Special issue.) Catena 754. Stocking M and Murnaghan N (2001) Handbook for the Field Assessment of Land Degradation. Sterling, VA: Earthscan Publications. Van der Meer FD and de Jong SM (eds) (2001) Imaging Spectrometry: Basic Principles and Prospective Applications. Bookseries Remote Sensing and Digital Image Processing, vol. 4. Dordrecht, the Netherlands: Kluwer Academic Publishers. Van Lynden GWJ (1995) European soil resources. Current status of soil degradation, causes, impacts and need for action. Nature and Environment Series 71. Strasbourg, France: Council of Europe.