ok so this has been bugging me for a while and i cant find a straight answer. doing a mapping job next week over a site that has some pretty significant elevation change, maybe 40 meters of relief across the area, not huge but enough that im second guessing my usual settings.

normally i just run 75 front 65 side overlap at a fixed altitude and call it a day for flat stuff. client wants a decent orthomosaic and a dsm, nothing crazy precision-wise but it needs to look professional. the problem is this site has a ridge running through the middle and im worried that if i fly at a fixed altitude above the takeoff point im going to end up with inconsistent GSD across the model and potentially some gaps or artifacts on the steeper parts where overlap effectively drops.

my question is basically whether i should bump overlap up across the board like 85 80 or something, or if i should try to do two separate flight plans at different altitudes and merge the data in post, or if im just overthinking a 40m elevation change and my normal settings are fine. using dronedeploy for planning and processing. the drone is a phantom 4 rtk if that matters.

40m relief honestly you're probably fine just bumping overlap a bit, i wouldnt go crazy with it but like 80 75 would give you a decent cushion without killing your flight time too much. the p4rtk handles terrain reasonably well and dronedeploy has the terrain follow option if you have a decent elevation model loaded which should help more than any overlap tweak anyway. have you looked at enabling that? makes a bigger difference than people realize on sites with actual relief.

ive done sites with way more than 40m of relief and as long as youre using terrain following in your planning software and your GCPs are distributed across the elevation range not just clustered at one level youll be fine. the GCP placement is something people sleep on, ive seen models go weird not because of overlap issues but because all the control points were down at the low end of the site and the high end was just floating a bit. spread them vertically if you can.

the two-flight-plan merge approach is more hassle than its worth for a 40m site in my opinion, save that for like canyon surveys or something where the relief is actually extreme.