Erosion database interface (EDI): a computer program for georeferenced application of erosion prediction models

The multidisciplinary approach of soil erosion research often requires erosion to be treated as spatial georeferenced information. This condition is essential so as to be compatible with information analyzed via Geographic Information Systems (GIS). The original versions of important soil erosion prediction models such as the Universal Soil Loss Equation (USLE) and Water Erosion Prediction Project (WEPP) do not operate on a georeferenced basis. The Erosion Database Interface (EDI) is a computer program for georeferenced application of USLE and WEPP. EDI uses, as input, a text format database with points defined by coordinates (x, y, and z) representing hillslopes, each point associated to soil type and land use. Such input data can be provided by different methods. Exclusive field work with ordinary topographic equipment and GIS procedures are examples of methods that can be used for this purpose. Flexibility in the methods adopted for providing input data is an important prerequisite for erosion prediction in tropical and developing regions. where soil erosion is a major concern and the availability of digital data is usually restricted, Hillslopes for EDI were defined as straight line segments beginning at the upper slope and ending down at runoff output, This restricts EDI as a complete erosion-prediction method for areas where runoff deflecting features predominate or where channel or gully erosion is to be considered. As output, EDI provides georeferenced soil erosion values in another text format database. This database can be used directly for statistical or geostatistical analysis or imported into a GIS for further processing. A practical example representative of a sugarcane-growing area located at the southeastern part of Brazil is used to illustrate EDI's performance. In this example, soil erosion maps were produced from GIS data using EDI as interface for erosion calculations for WEPP and USLE. (C) 2002 Elsevier Science Ltd. All rights reserved.