A possibility theory approach for estimating available water capacity fromimprecise information contained in soil databases

Soil survey data are required for different kinds of environmental and agro nomic studies, especially for estimating general soil quality indicators su ch as Available Water Capacity (AWC) over large areas. Generally, AWC estim ates take the form of a unique value per soil class, and the uncertainty pe rvading these estimates is not assessed, although it is of crucial importan ce for the users. To overcome this problem without requiring expensive-to-m easure data, we propose an alternative approach which estimates AWC from tw o different types of information currently available in soil databases: (i) imprecise descriptions of soil classes, i.e. synthesis profiles or qualita tive descriptions of soil classes, (ii) pedotransfer functions and other st atistical relations between input soil properties. The imprecise descriptions of soil classes were considered as sets of fuzzy intervals of values of soil properties and represented by possibility dist ributions. Statistical relations were translated into fuzzy relations of so il properties by applying a "probability-to-possibility" transformation. AW C was then estimated by means of a fuzzy interval calculus that was impleme nted as a Constraint Satisfaction Problem (CSP), allowing the representatio n of relations between the variables. AWC estimates were expressed by possi bility distributions that were translated into a set of prediction interval s corresponding to different possibility levels. The approach was applied t o the estimation of the AWC from soil class descriptions from three detaile d soil maps in the Languedoc plain (southern France). The validation study demonstrated that the proposed approach provided relia ble but fairly imprecise estimates, which allows the soil classes to be ran ked with respect to the AWC. The best reliability-precision compromise was obtained at possibility level of 0.8 by using qualitative descriptions as a data source. Because it only uses currently available information of soil databases, the proposed approach can be extensively used in the future for regional applications which allow for a certain amount of (known) imprecisi on, or in studies aimed at defining improvements to existing soil databases . (C) 2001 Elsevier Science BY. All rights reserved.