Integrating geostatistical tools in geographical information systems

Carlos Alberto FelgueirasAntônio Miguel Vieira MonteiroEduardo Celso Gerbi CamargoGilberto CâmaraSuzana Druck Fuks

Geostatistical analysis can be used for spatial modelling in a diversity of geographic applications. In particular, geostatistical methods and tools are of great utility when searching for surface models for fitting a data set sampled as points. By exploring the spatial structure of the data being modeled, the geostatistics yields procedures for surface estimation that contemplates the spatial variability inherent to, for instance, a large set of environmental data. At the same time these developments can generate a spatial estimate of the uncertainty of the information conveyed by those surface representing the attribute information. When developing a geographical application on a GIS platform that has no geostatistical tools, these analysis are performed by exchanging the data to be modeled between the GIS and the geostatistical package where the analysis will take place. These back and forth procedure can be very cumbersome, and in fact , can take a lot of time out that should be used in the task of being thinking on the modelling for that data. These are some reasons for the current trend to incorporate geostatistical facilities in the most popular GIS packages. Initial efforts has been done in the direction to integrate a geostatistical module in the GIS software. Unfortunately these is not enough to guarantee that the geostatistical power is fully incorporated in geographic analysis using GIS packages. Geostatistical procedures must be implemented as operators to be used in spatial modelling languages. Also, the uncertainty related to the modeled data should be considered and propagated to the results of a spatial model that contains non-deterministic parameters and data. This paper addresses the central aspects involved in the incorporation of geostatistical facilities in a GIS modelling environment. The requirements for a geostatistical module to be aggregated into a GIS package and the issues related to geostatistical and error propagation operators in GIS modeling languages are presented and discussed.

Caso o link acima esteja inválido, faça uma busca pelo texto completo na Web: Buscar na Web

Biblioteca Digital Brasileira de Computação - Contato:
     Mantida por: