Integrating parametric uncertainty and modeling results into an advisory system for watershed management

Abstract This paper describes an approach to integrate complex modeling experience into a decision support framework for non-point source pollution modeling of a watershed. The approach employs probabilistic reasoning techniques and derives probability distributions from previous model simulations. Thus, the sensitivity of a given model to its inputs is captured in such a way that the system can be used to find solutions to management problems through the application of probabilistic inference. A graphical probability model is a visual formalism encoding random variables and relationships between random variables as nodes and directed links in a graph. In our model, the nodes represent individual parameters for each of the fields in a watershed. Directed links connect correlated nodes, such as nodes representing the management practice in a field to nodes representing soil loss rates. The directed links between the nodes in the probability model follow the drainage network of the watershed. The relationships (or links) between the nodes are quantified via two methods. The first method integrates data (cases) derived from Monte Carlo simulation of a non-point source (NPS) pollution model. In the second method, deterministic functions defined in the NPS model are used to specify the relationships. The Monte Carlo simulations are performed to include the influence of parametric uncertainty on model results. The network for an entire watershed is complex with a large number of nodes, therefore, a spatial analysis/visualization tool was developed for interacting with the large probability model.

[1]  Judea Pearl,et al.  Evidential Reasoning Using Stochastic Simulation of Causal Models , 1987, Artif. Intell..

[2]  Christopher Joseph Pal,et al.  Case Libraries and Information Theoretic Case Matching for Soil and Water Resources Management , 1999, ISESS.

[3]  Ramesh P. Rudra,et al.  Targeting remedial measures to control nonpoint source pollution. , 1990 .

[4]  W. H. Wischmeier,et al.  Predicting rainfall erosion losses : a guide to conservation planning , 1978 .

[5]  Judea Pearl,et al.  Fusion, Propagation, and Structuring in Belief Networks , 1986, Artif. Intell..

[6]  Olli Varis,et al.  A Belief Network Approach to Optimization and Parameter Estimation: Application to Resource and Environmental Management , 1998, Artif. Intell..

[7]  C. T. Haan,et al.  Effect of Parameter Distributions on Uncertainty Analysis of Hydrologic Models , 1998 .

[8]  Sakari Kuikka,et al.  Joint use of multiple environmental assessment models by a Bayesian meta-model: the Baltic salmon case , 1997 .

[9]  Ramesh P. Rudra,et al.  Identification of soil erosion and fluvial sediment problems , 1986 .

[10]  David J. Spiegelhalter,et al.  Local computations with probabilities on graphical structures and their application to expert systems , 1990 .

[11]  J. Cain,et al.  Application of belief networks to water management studies , 1999 .

[12]  G. J. Wall,et al.  GAMES—A Screening Model of Soil Erosion and Fluvial Sedimentation on Agricultural Watershed , 1986 .

[13]  John W. Crawford,et al.  An application of belief networks to future crop production , 1994, Proceedings of the Tenth Conference on Artificial Intelligence for Applications.

[14]  J. J. Warwick,et al.  USE OF FIRST‐ORDER UNCERTAINTY ANALYSIS TO OPTIMIZE SUCCESSFUL STREAM WATER QUALITY SIMULATION 1 , 1997 .

[15]  Ramesh P. Rudra,et al.  A delivery ratio approach for seasonal transport of sediment , 1986 .

[16]  James M. Hamlett,et al.  DETERMINING THE DECISION-MAKING RISK FROM AGNPS SIMULATIONS , 1998 .

[17]  D. Heckerman,et al.  Toward Normative Expert Systems: Part I The Pathfinder Project , 1992, Methods of Information in Medicine.

[18]  R. Denzer.  Environmental Software Systems: Environmental Information and Decision Support, IFIP TC5 WG5.11 3rd International Symposium on Environmental Software Systems (ISESS'99), August 30 - September 2, 1999, Dunedin, New Zealand , 2000, ISESS.