We mapped a couple of fields this spring with Green Aerotech
. I'm not totally sure what the X/Y resolution is, but they measured some pretty small features (ditches) that a previous mapping by a ground-based rig didn't even register. They have a fixed-wing UAV that they fly in a grid pattern, and then use multiple images to create parallax which they can determine ground elevation with. It's pretty remarkable how well it works with a single camera.
They have an online tool for identifying drainage channels and showing you what your cuts would be like. They claim their elevations are accurate to within a few cm. I think they can export data to a shapefile, so drainage software packages should be able to work with it.
One of the problems with trying to get elevation data from your sprayer, tractor, or combine--even with RTK--is that you really only get data every so often depending on your width. But could give you some baseline measurements.
Another challenge is combine data from different sources. Even with RTK, it's best to think of your data as relative. Unless all the RTK systems are calibrated with known survey markers, they may disagree with each other (but in consistent ways). Just something to consider if you ever import data from different sources into a GIS editing program.
I want to set up a mobile RTK unit using a ZED-F9P that can talk to my phone over bluetooth. Battery to run the ZED and the radio receiver, plus a bluetooth serial adapter to get the NMEA feed to my phone. Then I could use a logging program (something like Mobile Topographer Pro) to collect data. In many cases I know where the drainage needs to go, but I don't know exactly how much needs to be cut in different places. With the mobile gear i could drive a quad through the bottom of the area and get some very specific data to play with that way.