I am facing some difficulties in processing MODIS data set.
To have MODIS data over my study area, I need to mosaic two tiles of MODIS (h29 v12 and h30 v12). I have used the MRT tool by NASA (the website based) to reproject the MODIS data from sinusoidal to UTM (zone 55s). And used VISAT (software by ESA) to mosaic the data.
It is not clear for me why the final subset over the study area comes with some coordinate shifts (13') when it is compared against google earth and land use map derived from Landsat imagery. I guess the mosaicing process is the main reason for this coordinate shift, but I heard that it could be because of reprojection process.
Can you give me some clue about what could be the reason for this coordinate discrepancy or how to solve this problem?