Environmental modeling with uncertainty treatment in geographical information systems: the indicator geostatistical paradigm / Modelagem ambiental com tratamento de incertezas em sistemas de informação geográfica: o paradigma geoestatístico por indicação

AUTOR(ES)
DATA DE PUBLICAÇÃO

1999

RESUMO

The use of Geographic Information Systems (GIS)to model environmental phenomena limits decision-making processes because the products of GIS modeling, such as agriculture suitability and vulnerability to erosion maps or economic-ecological zoning, do not encompass quantitative information on data quality. Current GIS technology does not associate the notion of data quality to the various scenarios resulting from integrating and combining different spatial data. This situation requires attention and jeopardizes decision-making processes based on GIS. In most cases there is no information on data quality aggregated to the final products. As a consequence, what should be a scenario of possible answers to a certain modeling turns out to be the expression of truth. The present research is concerned with issues that come from this context. This thesis is a proposal towards a systematic computational modeling in GIS based on the paradigm of indicator geostatistic modeling. The proposed techniques and methods are implemented and demonstrated through case studies. This work is especially concerned with the formulation of estimation techniques and uncertainty metrics that can treat both thematic (categorical)and numerical (continuous)geographic data. It also discusses and tests the hypothesis of using soft data to improve the estimate accuracy when only a few samples are available. The integration of these procedures in a GIS environment, as well as the basic requirements for such integration, is discussed and presented. This research contributes to guide GIS designers and developers to incorporate spatial modeling tools required for an appropriate handling of geographic data, taking their uncertainties into consideration when the data set is directly involved in environmental modeling.

ASSUNTO(S)

geoprocessamento simulação estocástica uncertainty modeling geostatística qualidade de dados geographic information systems stochastic simulation modelagem de incerteza data quality krigeagem geostatistic sig kriging sistemas de informacao geográfica

ACESSO AO ARTIGO

Documentos Relacionados