Sado Estuary Management Areas: Hard versus Soft Classification Maps Comparison

S. Caeiro (1), S. Sousa (2), R.G. Pontius Jr (3), M. Painho (2)

(1) ISEGI/CEGI, Institute for Statistics and Information Management of the New University of Lisbon (PT)
(2) IMAR, Depart. of Exact and Technological Sciences of the Portuguese Distant Learning University, Lisbon (PT)
(3) Department of International Development, Community and Environment, Graduate School of Geography, Worcester (US)

In the different GIS applications, Coastal zone management in particular, compare or detect different maps is an essential issue. The accuracy of a comparison procedure based on a more reliable and robust approach could have a marked improvement in the ability to detect a map change. Costal hydrodynamics makes it difficult to define sampling grids in exact positions and therefore a single cell-by-cell analysis comparison is less representative. Therefore, in this case, a neighbourhood cells comparison is more appropriate. Using the neighbourhood to compare categorical maps could be computed using a hard or fuzzy classification. Hard classification has the disadvantage of modifying the maps before the comparison. After hardening, there could be a substantial change in the quantity of each category, leading to errors and misleading results. By applying fuzzy set theory for the comparison of categorical maps it is possible to obtain a special and gradual analysis of the similarity of two maps. Fuzzy set theory implements classes or groupings of data with boundaries that are not sharply defined. Two sources of fuzziness are considered, the first is fuzziness due to vague distinctions between categories the second is fuzziness due to the gliding scale of similarity.

In order to divide and define the Sado Estuary in homogenous areas for future environmental management of this ecosystem, three different geostatistical multivariate techniques were used.
The aim of this work is to assess the difference between the three maps, and discuss the more appropriate method classification for that comparison. A great agree of similarities will further support the choice of any of the methods as appropriate for environmental management, and hence the less significance of choosing one of the methods. Visual map overlays were used either for single cell, or neighborhood using hard and fuzzy data map comparison. Neighborhood operations summarize the attributes occurring in the vicinity of each location. It creates a map where the value assigned to a location is computed as a function of independent values surrounding that location. This group of operations can be conceptualized as "moving windows" sliding throughout the mapped area. Each location is a function of the input cells of different neighborhood sizes (3, 5 and 7), instead of one input cell-by-cell comparison, each cell corresponding to 100 meters. Then, map algebra was used again to obtain the difference between each of the two maps and create a classification of their differences. For quantification of map comparison approaches, Kappa statistics (kstandard, klocation and kquantity) and agreement space were used.

Although the three methods of homogenous areas were computed with different statistical techniques, their results are similar. These approaches demonstrate that using either single cell or a neighborhood number of cells the estuarine management areas maps are still moderately correlated. Nevertheless, differences between hard and fuzzy map comparison were discussed.