Author(s)

J. Negreiros, M. Painho, A. Cristina Costa, P. Cabral & F. Aguilar

Abstract

The local confidence uncertainty plume of SAKWeb© J. Negreiros1, M. Painho1, A. Cristina Costa1, P. Cabral1 & F. Aguilar2 1Instituto Superior de Estatística e Gestão de Informação, Universidade Nova de Lisboa, Lisboa, Portugal 2Escuela Politécnica Superior, Universidad de Almería, Almería, Spain Abstract The main goal of this research paper is to introduce a new uncertainty tool based on the Moran I correlogram, rescaled OK variance and local variance in a Web environment. It is hoped that this implementation will be used by users with problems to layout risk analysis environmental maps and plumes assessment. Spatial analysis, Moran I and other uncertainty measures are also reviewed. Keywords: GIS, spatial analysis, Kriging, variogram, Moran I, local variance, local confidence, plumes assessment, SAKWeb©. 1 Spatial analysis: overview The geographical view of spatial analysis is essentially cartographical driven regarding the recognition and description of spatial patterns involving simple statistics and the direct use of visualization. Anselin [1] wrote that spatial data analysis could be defined as the study of statistical phenomena that manifest themselves in space. As a result, location, area, topology, spatial arrangement, distance and interaction become the focus of attention. As confirmed by Longley et al. [9], spatial analysis is a set of methods whose results change when the object location being analyzed changes too. Hence, spatial analysis is more widespread than statistical analysis of non-spatial information because it requires access not only to attributes but also to location and topological knowledge. It could be described as part of the process of transforming spatial data into geographical knowledge. Quoting Longley et al. [9], it can make what is implicit explicit.

Keywords

GIS, spatial analysis, Kriging, variogram, Moran I, local variance, local confidence, plumes assessment, SAKWeb©.