engleski

Detecting habitat changes in cross-border areas using MODIS EVI (MOD13Q1)images: a case study of spoonbill (Platalea leucorodia)

The breeding population of spoonbill (Platalea leucorodia) in Croatia, currently estimated at 111-220 breeding pairs, is classified as endangered (listed in Red Data Book of Croatian Birds and it’s breeding population). Most important reason for the endangered status are changes in breeding and feeding habitats. The purpose of the study was to detect changes in the habitat types previously assessed as important for spoonbills feeding. The area of interest is the central part of the Sava River (Croatia – Bosnia and Herzegovina). The research area was chosen for three reason: 1) the alluvial wetlands of the Sava River holds the biggest colony of spoonbills in the Croatia (until 1989 it was the only one) ; 2) there are research on feeding ecology of spoonbills along alluvial wetlands of Sava River and 3) we wanted to test usefulness of the method since the specific shape of the Croatia, having important breeding grounds for numerous species placed just along country borders. We used the MODIS Enhanced Vegetation Index (EVI) as the main indicator for the changes in the habitat structure. 20 16-days composites for years 2000 and 2008 with the spatial resolution of 250 m were obtained from the NASA server. Downloading, re-projection and analysis of images was implemented using the open-source software MRTool, R, SAGA. This software combo enables data processing automation and thus can be used to process large volume of data. MODIS images are especially attractive for analysing habitat in cross-border areas where the availability of GIS layers is limited or the layers are not harmonized. The results of the analysis revealed places with significant changes in EVI index along important habitat types for feeding of the spoonbills. The probability value of for t-test (difference in EVI distributions for two periods per pixel) has been produced. The resulting grid was further exported to kml format for visualization in Google Earth i.e. free exchange of results. For the Croatian part of the research area, the resulting grid was overlaid with layers presenting important habitats for spoonbills (pastures, rivers, river edges, fishponds and pastures along highest wetness potential) and analyzed to detect which land cover types are most viable to habitat changes.

MODIS; EVI; monitoring; habitat

nije evidentirano