Modelització dels marges entre camps a partir de l’anàlisi i tractament paisatgístic pel municipi de Vilademuls

Abstract

Margins are a land-use that can have different conformations depending on the neighbouring patches, and because the ecosystem services that bring, they have ecological importance. Nevertheless, cause the intensification process, the patches that were margins between crops, now are disappearing in pro of the crops unification. This process is currently happening in Vilademuls and was identified on a cartographic and statistic land use study. The study focuses on the cartographic treatment and analysis for the margins between crops category of land use using the data of the Vilademuls study. In the process, a theoretical model that categorize the margins is extracted, and it’s checked using the same procedure analyzing herbaceous crops land use. To obtain the model, the first step is to revise the cartographic data and then do the individualization treatment for the patches and posterior transformation. For the landscape analysis comes the index selection to extract the values with it. Then comes the statistical analysis of the values for every plot using a principal component analysis in the first place, and for a hierarchical aggrupation in the second. The results for the statistical analysis PCA for the margins model show that the more explanatory indexes are the area, that was calculated for margin land use, and the area, fractal index and Largest Patch Index calculated for the landscape plots. In comparison with the crop model, the variables are the cohesion obtained for the herbaceous crop category, and the Shannon’s Evenness Index, LPI, perimeter and cohesion extracted for landscape patches. The classification results are the same, with two optimal groups, and the difference for the aggrupation it’s only for one plot. One explanation might be that because of the use of Euclidean distance at the similarity matrix, it can bias the results, causing the aggrupation in two groups. The results of the study show that the theoretical model is useful but it is inconsistent because it relays a lot on the resolution used in the cartographic digitization. Plus, the land use used in the model is the margins between crops category only, and it could be interesting if using all the types of margins that can be found in the plots