Um Contributo para a Delimitação da Área Metropolitana do Noroeste de Portugal

Abstract

The objective of this paper is to present an approach for the definition of a metropolitan region in the Northwest part of Portugal, standing on the current Oporto Metropolitan Region. The proposed approach, which constitutes an option to avoid the endless confrontations that may derive from the essentially subjective political criteria, analyses the socio-economical evolution observed from 1991 and 2001 Census Data and explores two branches of Spatial Analyses: Spatial Statistics and Spatial Modeling. Exploratory Spatial Data Analyses (ESDA) tools are used to identify the characteristics of local associations in order to build dynamic models of the studied region. Moran´s scatterplots can be used as means to classify the behavior of each particular zone regarding the value of the attribute under consideration and the mean value of the same attribute for neighbour zones. In such a classification scheme, four possibilities are considered, which correspond to quadrants in the scatterplot. The analyses of both the points in the scatterplots and their associated location in a map give a clear indication of clusters of zones with similar characteristics, which can be seen as uniform regions. These homogeneous regions, however, do not necessarily represent the future metropolitan regions, unless there is a previous knowledge of what is a positive aspect for the considered variable. Another potential problem of this simple view could come from a unique observation of the data in time. However, the regularity of census surveys offers a reliable alternative for long-term analyses, if carried out in the way abovementioned, because the analyses are not directly based on attribute values but on the association of zones according to their attribute relationships, as shown here. The proposed method constitutes a promising option for the integrated and holistic approaches required for the delimitation of metropolitan areas. This is particularly true if it is combined with other strategies that allow the identification of forms of not contiguous spatial arrangements that can characterize metropolitan areas or polynucleated cities (such as constellations or clusters).