Alternative Approaches to Soil Erosion Prediction and Conservation Using Expert Systems and Neural Networks

In response to on-site and off-site water-borne soil erosion impacts, considerable effort has been expended in recent decades in conceptualising and developing quantitative models for soil erosion prediction. Such models have focused on deductive process-based modelling, emphasising replication of the physical laws operational in soil erosion and seeking to encapsulate the conditions, processes, and practices deemed responsible for erosion events. This paper proposes an alternative, inductive, use of two variants of empirical models: those of expert systems and artificial neural networks. Though empirically based, these models offer significant advantages and potential for both soil erosion prediction and conservation. In these models the theoretical understanding of the soil erosion process is expressed in the sample erosion sites selected for rule generation and in the related site attributes. Importantly, the approach is empirical not one of empiricism.

[1]  R. Evans.  Water erosion in British farmers' fields- some causes, impacts, predictions , 1990 .

[2]  J. Boardman,et al.  Soil erosion, climatic vagary and agricultural change on the Downs around Lewes and Brighton, autumn 1982 , 1985 .

[3]  John A. Dearing,et al.  Soil erosion on the South Downs: a review. , 1990 .

[4]  Jean Poesen,et al.  A long flume study of the dynamic factors affecting the resistance of a loamy soil to concentrated flow erosion , 1990 .

[5]  Peter Jackson,et al.  Introduction to expert systems , 1986 .

[6]  G. R. Foster.  Advances in wind and water erosion prediction , 1991 .

[7]  J. Boardman.  Severe erosion on agricultural land in East Sussex , 1988 .

[8]  J. Boardman,et al.  Climate change and soil erosion on agricultural land in england and wales , 1990 .

[9]  John W. Tukey,et al.  Exploratory Data Analysis. , 1979 .

[10]  James L. McClelland,et al.  Parallel distributed processing: explorations in the microstructure of cognition, vol. 1: foundations , 1986 .

[11]  John R. Williams,et al.  EPIC-erosion/productivity impact calculator: 1. Model documentation. , 1990 .

[12]  Geoffrey E. Hinton,et al.  Learning internal representations by error propagation , 1986 .

[13]  J. Boardman,et al.  A rule-based expert system approach to predicting waterborne soil erosion. , 1990 .

[14]  Richard P. Lippmann,et al.  An introduction to computing with neural nets , 1987 .

[15]  William J. Clancey,et al.  The Epistemology of a Rule-Based Expert System - A Framework for Explanation , 1981, Artif. Intell..

[16]  J. Boardman,et al.  Simple methods of characterizing erosive rainfall with reference to the South Downs, southern England , 1993 .

[17]  J. Boardman,et al.  Soil erosion: its impact on the English and Welsh landscape since woodland clearance. , 1990 .

[18]  D. Favis-Mortlock.  Evaluation of Field-Scale Erosion Models on the UK South Downs , 1998 .

[19]  D. Robinson,et al.  Soil Erosion and Conservation , 1988 .

[20]  Donald Gabriëls,et al.  Characteristics of water-eroded fields in lowland England. , 1980 .

[21]  Anne Morris,et al.  Prototyping expert solutions: an evaluation of Crystal, Leonardo, GURU and ART-IM , 1992 .

[22]  Judith E. Dayhoff,et al.  Neural Network Architectures: An Introduction , 1989 .

[23]  James P. Ignizio.  An introduction to expert systems : the development and implementation of rule-based expert systems , 1991 .