Importing aerial pictures with scenery SDK

Hi all,I have a set of aerial pictures with which I want to make a scenery. The pictures cover each a 3*3km part of the landscape. They are 256 pixels square.Since I have to do a coordinate transformation (they are not WGS84) I'm wondering, which output format would be best.So my questions are:Should I make images which cover each a LOD13 quadrant or should I make bigger images (eg. LOD 10)?How many rows and columns does an image need, that the resampler doesn't need to a resampling?Is the important thing for the resampler "meters/pixel" or "degrees per pixel"? This is not equivalent for different places on earth, at least for the longitude.Thanks for your helpChristoph