Estimating land cover type and proportion of vegetation using remote sensing

Estimation of the land cover type and proportion of vegetation using NASA´s EOS/MODIS data is verified as a case study for Turkey where 7 locations were selected. The correlation between the NDVI/EVI data is analyzed. EVI is shown to be slightly more accurate than NDVI. Land cover types are determined for these 7 locations, and similar to other locations, İstanbul and Samsun are found to perfectly match the `forest type´. İzmir is found to be different from the currently available land cover types which are assumed to be due to its mixed canopy structure. Annual and monthly average measurements in various bands are analyzed and expected correlation with the number of days of precipitation is observed.