Regression models for spatial data can be tricky when trying to figure out the effects of different factors like soil, climate, and topography on tree health. The problem is that the way these factors are spread out in space can mess up our estimates. This is called spatial confounding. In the past, we only knew how to deal with this for simple, straight-line relationships between factors and outcomes. But now, we've figured out a way to handle this for more complicated, nonlinear relationships using something called spatial+. This new method helps us see if our estimates are being messed up by spatial confounding and correct them if needed. So, we can now better understand how different factors affect tree health in different areas.