As pointed out above, any hypothesized etiology of Parkinson's must be consistent with the Willis [1] paper's message. Here we note that this must be true both in general terms and in local terms. It is perhaps too easy to put local variations down to "noise".
Winds, both at surface and higher levels, can spread material through the atmosphere. Ultimately, this paints the general picture. But it takes meteorological deposition, the process whereby particles (dust, metals, spores, viruses etc.) fall from the atmosphere when there is insufficient updraught to support their weight or when gases are washed out of the atmosphere by precipitation, to paint in the fine detail.
I've not been able to find any material of this type for Parkinson's. However, Firestone et al. [2] do a meteorological analysis of the spread of equine flu. They take into account factors such as wind velocity, humidity, precipitation and temperature. Additional factors, such as temperature inversions, could also be considered.
Finally, let's address how we might narrow down the number of possible causes of PD. We could run the model in reverse, taking backward trajectories, leading from the destination to the source. The problem is that we don't know the size of particle we are dealing with. But we might be able to argue that the lumpiness of the prevalence distribution precludes certain particle sizes.
References
[1] "Geographic and Ethnic Variation in Parkinson Disease: A Population-Based Study of US Medicare Beneficiaries"
Allison Wright Willis,a,* Bradley A. Evanoff,b Min Lian,b Susan R. Criswell,a and Brad A. Racettea
Neuroepidemiology. 2010 April; 34(3): 143–151.
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2865395/
[2] "The Influence of Meteorology on the Spread of Influenza: Survival Analysis of an Equine Influenza (A/H3N8) Outbreak"
Simon M. Firestone, Naomi Cogger, Michael P. Ward, Jenny-Ann L. M. L. Toribio, Barbara J. Moloney, Navneet K. Dhand
PLoS ONE 7(4): e35284. doi:10.1371/journal.pone.0035284, 2012
http://www.plosone.org/article/info%...l.pone.0035284
John