engleski

Influence of different thematic and spatial data resolution on the detection of landcover changes

In order to detect and quantify landcover changes, we need historical data from various time periods. However, often these data are of different spatial and/or thematic resolution. Therefore, we have to adjust our data to those with the coarsest resolution. Comparison of habitat diversity in four Croatian protected areas in terms of different spatial (1: 25 000 and 1: 100 000 scale) and thematic resolution (three category levels of Croatian National habitat classification scheme) has shown that spatial resolution will influence both “presence” (Shannon-Wiener) and “spatial” (number of patches, edge density, mean perimeter – area ratio, mean shape index) diversity with statistical significance. Thematic resolutions of habitats had a greater impact on “presence” than on “spatial” diversity. For temporal variability analyses, we usually have to use any available data in terms of data resolution. A coarser thematic resolution will yield clearer trends in changes, but less understandable dynamic of those changes, which is more helpful in prediction scenarios. Nevertheless, spatial and thematic resolution are not independent, and coarser spatial resolution usually implies coarser thematic resolution, e.g. as in Corine Land Cover (CLC) map. Use of CLC maps for change analyses can result in complete absence of detected changes, while on the same thematic resolution usage of finer spatial resolution map will yield detection of significant changes. This was demonstrated on a sample of two protected areas in Meditteranean part of Croatia, where is in general habitat heterogeneity pronounced. It is therefore essential to match the size of an area of interest with the spatial scale of available data, prior to making final conclusions on effective landcover changes.

vegetation / habitat map; vegetation dynamics; diversity

nije evidentirano