Research question is an econometric Monte Carlo approach to checking misspecification when geospatial factors matter. Someone must have done this. If not, you are very lucky to be doing this. But I would be surprised if someone has not done this. What does Luc say?
I wanted to understand your results better. They looked very interesting, but I didn't quite understand how the pictures and table supported your conclusion.