出版社：Centro de Estudos Geográficos, University of Lisbon
摘要：GIS APPLICATION FOR INTERACTIVE PROSPECTION OF ENDOKARST. The parameters that affect the development of karst erosion at depth – endokarst - are complex and difficult to understand, due to both their diversity and the difficulty of quantifying the actual contribution of each of them. This work presents a methodology developed in a GIS environment with the aim of identifying the areas in which endokarst is most likely to occur. This case study was conducted within a sample area located to the west of Sesimbra where several of these structures have been identified, with a special emphasis on the caves of Zambujal and Frade. The suggested model is based on the interactive combinatory analysis of the dependent variables according to skillful criteria, taking into account their relative contributions to the development of endokarst. One of the main goals of this study is the development of an interactive software application enabling to test and evaluate the relative importance of the selected parameters. A decision was made to use a simple classification of the supervised or skillful type, in which the variables, their inter-relations and their relations with the surrounding areas are conveyed under the summarised form of a ranking of their relative relevance, as measured by their expected contribution to karstification. The output of the model consists of a set of maps that indicate the likelihood of the occurrence of endokarst at any given point. The hierarchical relevance of each variable, the selection of larger or smaller buffer areas for fractures and fracture intersections, the choice of appropriate variables and their subdivision into a number of categories allow for scrutiny to be applied at a series of different levels, rendering possible the skillful calibration of the results to all the places where the predominant geology is of the carbonate type. At the same time, the comparative analysis of these results provides an excellent opportunity for us to understand the parameters that most contribute to the formation of endokarst. The suggested model was tested interactively and the latter analysis and validation of the results allowed for the positive identification of the appropriate sequence of those variables that most affect the formation of endokarst.
