Soil Conservation

GIS Models to Measure Soil Erosion Risk

By: Atiqur Rahman & Asif, The authors are Professor and Assistant Professor, Department of Geography, Faculty of Natural Sciences, Jamia Millia Islamia, New Delhi, respectively
Soil erosion is the removal of the top layer by natural agents. However, deforestation, overgrazing and infrastructural development accelerate soil erosion. As soil is a critical natural resource, the risk of its erosion must be addressed urgently. Use of modern technologies like remote sensing and GIS can help build an understanding towards risk mitigation.
Magazine Articles

Soils are a complex mixture of minerals, water, air, organic matter and countless organisms that lie on the surface and as such is the ‘skin of the earth’ providing a natural medium for the growth of plants (Soil Science Society of America, 2005). The properties of soil depend on the parent material, topography, climate, time and the organisms present. There can thus be hundreds of different kinds of soil. Since soil is one of the most important natural resources and is vital for the functioning of ecosystem, the problems associated with it, particularly erosion should be addressed with utmost care and on top priority. Water, wind, moving ice and more are the primary agents responsible for soil erosion and depletion of soil fertility. Heavy rainfall, nature of the soil (loose, lighter and sandy soil is more susceptible for soil erosion) and steep slope are some of the natural factors responsible for soil erosion—broadly classified into:

  • Sheet erosion-removal of top soil in the form of sheet (thin layers), widespread particularly in areas having smooth and gentle slope;
  • Rill erosion-removal of top soil in finger-like narrow grooves on uneven slopes; and
  • Gully erosion-where rills join to form wider and deeper channels called gullies.

Although soil erosion is a natural phenomenon, its pace has increased manifold in the latter half of the 20th century and first two decades of the 21st century due to unsustainable and unplanned developmental activities around the world, especially in the hilly areas. Deforestation, overgrazing, infrastructure development, faulty agricultural practices and forest fires aggravate the problem of soil erosion. Apart from loss of soil fertility, erosion exacerbates the spread of desert, famine, flash flood, siltation, destruction of wild life, climate change and more. Worldwide, soil erosion losses are highest in agro-ecosystems of Asia, Africa and South America, averaging 30 to 40 tonnes/ha/year (Pimentel, 2006). An estimate shows that the rate of soil erosion in India is 16.35 tonnes/hectare/annum, which is higher than the naturally determined limits of 4.5 to 11.2 tonnes/hectare/annum. Also, the average annual loss of soil nutrients due to soil erosion has been estimated as 5.4-8.4 million tonnes (Fertiliser Association of India, 2008). About 29 per cent eroded soil is washed out into oceans and seas, 10 per cent is deposited in reservoirs and ponds while the remaining 61 per cent is relocated from one place to another (Dhruva and Babu, 1983).

World agriculture accounts for about three-quarters of the soil erosion worldwide (Food and Agriculture Organisation, undated). Although methods of soil conservation such as contour ploughing, terrace cultivation, keyline design, windbreak and crop rotation were in practice for centuries, the estimation of soil erosion was barely undertaken as it was a tough task until the recent past. With the advent of modern technologies—remote sensing and geographical information system (GIS), soil erosion risk could be estimated with better precision and ease. GIS is complimentary to remote sensing as it enables data from varied sources to be incorporated, analysed and even modelled by its powerful analytical functionality. Both these tools are very important in identification and solving complex geospatial problems including the risk of soil erosion.

Soil Conservation | Soil Erosion Risk Assessment

From 1972 onwards satellite data from Landsat, IRS and MODIS and various sensors such as IRS (LISS-III, LISS-IV, WiFS, aWiFS etc., Landsat (MSS, TM, ETM, ETM+, ASTER etc., started throwing up good resolution data making the accurate prediction of soil risk assessment easy. There are standard models which have been developed for this. The Universal Soil Loss Equation (USLE) was the first standard method for assessing soil erosion using remote sensing and GIS techniques which was later enhanced to the Modified Universal Soil Loss Equation (MUSLE) and further to Revised Universal Soil Loss Equation (RUSLE).

The USLE was developed even before the beginning of the Landsat programme by Wischmeier and Smith (1960) for United States Department of Agriculture (USDA) as a multiple soil loss model to estimate annual soil erosion. This model was used extensively in South East Asia too, apart from USA and consequently in several other countries around the world. The USLE model can be expressed as a compound of five factors including rainfall erosivity factor (R), soil erodibility factor (K), spatial distribution of crop management factor (C), spatial distribution of conservation and preservation factor (P) and slope length and slope inclination factor or topographical factor (LS).

In this model, rainfall type determines the erosive (R factor) capabilities—a function of volume, duration and intensity which can be calculated from a single function or a series to identify the cumulative erosivity of a particular time. The erodability of soil (K factor) is related to the resistance of the soil to detachment and transport, with the K values of different soil types matched with Soil Erodability Nomograph (USDA, 1978).The C factor reveals the consequence of soil related activities that depend on plant cover, crop combination sequence and productivity. It can be defined as the ratio of the losses of soil from the cropped land under some specific conditions to the corresponding fallow land. Since a lot of variations are found in spatial and temporal land cover, it is good to use remote sensing data for the measurement of the C factor. The P factor identifies the result of slope on the loss of soil—its value ranging from 0 to 1. The maximum values are assigned to areas without any management practices and the minimum values are given to built-up area or area under plantation or contour cropping. The length and steepness of the slope (LS factor)is also significant and to generate the LS factor, distribution and consolidation of soil is important. The computation of LS requires steepness of slope and flow accumulation which can be derived from shuttle radar topography mission (SRTM) data available in 1-degree tiles, digital elevation model (DEM) provided by USGS server ( since 2003. And it can be analysed in ArcGIS Spatial analysis ( and Arc-hydro tool (

This USLE model requires digital data for rainfall, soil property, land use, DEM and soil distribution. Soil loss simulation model using environmental GIS is simple, reasonable and useful for assessment of soil erosion (Bleecker et al. 1995; Savabi,1995). In order to get the desired result, specific soil loss risk models need to be formulated through a vegetation index calculated from remote sensing data. A risk map of soil erosion can be developed using the soil loss risk model following which soil loss and the factors of soil erosion can be assessed. On the basis of these estimations some measures for soil conservation may be suggested.

A second model—the MUSLE was proposed by Williams and Berndt in 1974 and later modified in 1977. As outlined earlier, the USLE model tries to identify average annual erosion based on rainfall energy. However, in the MUSLE model the rainfall energy is substituted by surface runoff. This plays a vital role in improving the prediction of sediment yield and makes it possible for the model to be applied to a single storm event (Neitsch et al., 2005). The runoff aspect is computed with the use of calculated values of runoff and peak rate of runoff at the channel of the watershed for MUSLE. The measurement of topographic factor (LS) and crop management factor (C) are done with the help of GIS and field survey for land use/land cover. The use of MUSLE at the watershed level is a very elegant data processing procedure which needs expert application of GIS. If the model is applied correctly, accurate soil erosion and risk assessment can be undertaken at the watershed level (Pandey and Chowdary, 2009).

Based on the USLE and MUSLE models, the RUSLE model was developed in early 1990s in the National Soil Erosion Research Laboratory (NSRL) USA. RUSLE can calculate long-term and mean-annual erosion by water for a wide variety of farming, management, mining, construction, and forestry uses. This model gives more accurate estimation of R, K, C, P factors and soil erosion (Van Remortel et al., 2004). Analysis of the database which was not included in USLE model has been included in the RUSLE model in a theoretical framework describing the fundamental hydrologic and erosion process. As the distribution of rainfall is not uniform throughout the year, if rainfall occurs when the land is fallow, the soil is most vulnerable to erosion. Thus in the assessment of erosion the degree of R factor and its seasonal distribution need to be taken into consideration along with the cropping system. One of the major improvements in the RUSLE is the erosivity map. Previously the map was based on very few point data calculations, but in the revised version data from more than 1,000 locations is collected and analysed.

Another change which has been included in RUSLE is related to rock fragments in the soil—which is considered as mulch in the C factor. Since the development of the USLE, a large number of questions have been raised about the L factor. The reason being that the choice of slope length depends on the judgment of the user. RUSLE provides important guidelines to choose slope length values to provide consistent results amongst different users. An assessment of the magnitude of soil erosion and its impact on the land and environment are major challenges in the present day scenario.


Models like USLE, MUSLE and RUSLE facilitate accurate and timely pattern and risk analysis of soil erosion in terms of cost, time and accuracy. USLE is used by soil scientists, water resource experts and preservationists. A few revisions and changes adopted in RUSLE have enhanced its value as an important tool to estimate soil erosion. The mitigation of the problem of soil erosion can begin by proper estimation through remote sensing data and GIS techniques at the local, regional and national level.


Bleecker M., S.D. De Gloria, J. L. Hutson, R. B. Bryant and R. J. Wagenet, 1995. Mapping atrazine leaching potential with integrated environmental databases and simulation models, Journal of Soil and Water Conservation, 50 (4):388-394.

Dhruva V. V. N. and R. Babu, 1983. Estimation of soil erosion in India, Journal of Irrigation and Drainage Engineering 109 (4): 419-434.

Food and Agriculture Organization (FAO), Undated. Restoring the Land, Available at:,.

Fertiliser Association of India, 2008. Fertilizer Statistics of India, Available at:

Neitsch S. L., J. G. Arnold, J. R. Kiniry, J. R. Williams, 2005. Soil and Water Assessment Tool theoretical documentation. Grassland, Soil, and Water Research Laboratory, Blackland Research Center.

Pandey A.V. M. and Chowdary, B. C. Mal, 2009. Sediment yield modelling of an agricultural watershed using MUSLE, remote sensing and GIS, Paddy and Water Environment (Springer) 7(2): 105–113.

Pimentel D., 2006. Soil erosion: A food and environmental threat. Environment, Development and Sustainability 8: 119–137.

Savabi M. R., D. C. Flanagan, B. Heble, B. A. Engel, 1995. Application of WEPP and GIS-GRASS to a small watershed in Indiana, Journal of Soil and Water Conservation 50(5): 477-483.

Soil Science Society of America, 2005. Glossary of Soil Science Terms,Available at:

Van Remortel R. D., R. W. Maichle and R. J. Hickey, 2004. Computing the LS Factor for the Revised Universal Soil Loss Equation through array based slope processing of digital elevation data using C++ executable, Computers and Geosciences, 30: 1043- 1053.

Wischmeier H. W.  and D. D. Smith, 1960. A universal soil-loss equation to guide conservation farm planning, Transactions of 7th International Congress Soil Science, 1 : 418-425.

Leave a Reply

Your email address will not be published. Required fields are marked *