In my previous life, I was a maker of maps, a land surveyor, and as a hobby I delved into remote sensing and geographic information systems. Translated, I enjoy converting data into colored maps to see what unexpected patterns might show up.
A couple of years ago, I posted a map of the US which I had color coded by the incidence of PD in each state based on data in one of Dr. Lieberman's books. This map is interesting but very limited because the source of Dr. Lieberman's data is not clear; it is given as an average for each state so the result is "coarse" with no detail within each state; my map further coarsened the data by subjectively assigning the states to a simplified high risk - low risk color coding system. Still, it is a pretty neat map and I haven't seen a better one.
Yesterday, I dusted that one off and went in search of another that might show UV exposure over the same area. UV should translate into Vit D production overall. I found two, from 1992 and 2006. They were comparable soI chose the latter because it was simpler to understand. In particular, it shows several dozen monitoring sites as little football shaped icons with relative index numbers for each so you can compare with a little finer resolution than the "whole state" approach.
So I merged the two maps and looked for features that appeared on both. That is, I looked for conditions that tied UV-Vit D to PD at the same geographical location. Given the poor quality of data, I would have been pleased to have found even one good one. Surprisingly, I found three and have marked them on the attached merged map. <PDoverlay.jpg>
On this crude merger, the colors represent my PD incidence data and the numbered footballs the UV - Vit D figures (the higher the number the more you have). The three anomalies that show up in both are-
1) On the east coast, Virginia and the Carolinas show both a drop in PD (thus the blue color) and a rise in the UV compared to the neighboring states (6 & 7 vs 3 - 5)
2) In the Dakotas, the highest PD rate (the orange color) coincides with the lowest UV ( a "2")
3) In the Southwest, the contour between blue and green PD values is a good match with the similar division between the 5 & 6 UV (the Rocky Mtns skew things a little).
Admittedly crude, but rather convincing to me.