Erfahrungen bei der Erstellung von rechnergestützten Bewertungssystemen für die Ermittlung des Grundwassergefährdungspotentials kontaminierter Altstandorte im Hinblick auf Sanierungsmassnahmen

Balmer-Heynisch, Sabine

Persistent URL: http://resolver.sub.uni-goettingen.de/purl?gldocs-11858/11920
DOI: 10.23689/fidgeo-6224
Balmer-Heynisch, Sabine, 1995: Erfahrungen bei der Erstellung von rechnergestützten Bewertungssystemen für die Ermittlung des Grundwassergefährdungspotentials kontaminierter Altstandorte im Hinblick auf Sanierungsmassnahmen. Berliner geowissenschaftliche Abhandlungen. Reihe D, Geoinformatik; Band 8, Fachbereich Geowiss., FU Berlin, Berlin, 212 S., DOI: 10.23689/fidgeo-6224. 

Abstract

A fast treatment of investigation, evaluation and remediation of contaminated sites and industrial wastelands is necessary to allow a proper recycling of these areas. The ..Ruhrgebiet" is a typical old industrial area in Germany with many closed mining and coke-plants, where the problems of reconditioning of industrial sites have to be solved. It is obvious that computer based systems for an objective and fast analysis of these industrial sites are necessary with respect to subsequent landuse. For this purpose a knowledge based system was developed for the analysis and evaluation of hydrogeological criteria relevant to contaminant transport in underground and groundwater. This project was supportet by DFG (German research association) and by RAG (Ruhrkohle AG). In addition the investigation of the origin, kind and amount of a contamination was included. The evaluation system HYDRISK is based on a layered 3D-model of the underground. The analytic module consists of routines for a spatial analysis of all available data (geological, geophysical and soil-chemistry). The evaluation criteria for each layer are projected on maps, gridded and evaluated for each grid cell using a knowledge based system approach. The thickness of each lay er is taking part of the evaluation in relation to total thickness of all considered layers. The risk assessment system was developed for PC and workstation as well. The evaluation module is interfaced with ARC/Info and relational databases (Dbase, Oracle). In terms of Artifical Intelligence (Al), HYDRISK may be characterized as an heuristic associative diagnostic system.. The evaluation of the spatially distributed data leads to deductive conclusions (forward reasoning) with respect to hydrogeological criteria for groundwater contamination risk. After detailed acquisition and structuring of the knowledge base, domain modules were defined. Three modules were selected for developing prototypes: unsaturated zone, estimation and evaluation of pollutants and their impact on soil and groundwater and the decision-making module for proper remedial actions. These modules differ in technical requirements and knowledge formalisms. „Forward-Reasoning“ and heuristic classifications are used in the„underground-module", database queries and data evaluation in the „pollutant module“. The decision modul for remedial actions represents a typically diagnostic module in the entire system. The concept of HYDRISK represents a first step for the realisation of a system which combines a knowledge based system with evaluation tools for multivariate spatial criteria: hydrogeological and hydrochemical criteria are related to pollutant characteristics as well as subsequent landuse requirements. Due to its structure this program can be enlarged. Most recent standards for environmental protection are taken into account in order to develop a powerful tool for decision making in wasteland remediation.

Subjects
Grundwasser