COS 74-2
Socio-economic, water quality and land-use land-cover gradients interact to create variable West Nile virus risk across metropolitan Atlanta, GA

Wednesday, August 7, 2013: 1:50 PM
L100E, Minneapolis Convention Center
Krisztian Magori, School of Forestry and Wildlife Sciences, Auburn, Auburn, AL
Graeme Lockaby, School of Forestry and Wildlife Sciences, Auburn University, Auburn, AL
Latif Kalin, School of Forestry and Wildlife Sciences, Auburn University, Auburn, AL
Rosmarie Kelly, Vector-Borne & Zoonotic Diseases Team, Acute Disease Epidemiology Section, Georgia Department of Public Health, Atlanta, GA
Wayde Morse, School of Forestry and Wildlife Sciences, Auburn University, Auburn, AL
Navideh Noori, School of Forestry and Wildlife Sciences, Auburn University, Auburn, AL
Rajesh Sawant, School of Forestry and Wildlife Sciences, Auburn University, Auburn, AL
Wayne C. Zipperer, Southern Research Station, USDA Forest Service, Gainesville, FL
Background/Question/Methods

West Nile virus (WNV), the most widespread arboviral pathogen in the US, is associated with urban environments in the South-Eastern US. Potential drivers of this association (land-use land-cover, socio-economic characteristics, and larval habitat availability and quality) vary across the rural-urban gradient, as well as within urban areas. Our objective is to disentangle the effects of these drivers in order to explain the spatial variation of WNV transmission risk across metropolitan Atlanta, GA. First, using mesocosm experiments, we elucidate the effects of water quality on the productivity of Culex quinquefasciatus larval habitats. We then use mosquito surveillance, remote sensing and publicly available datasets at frequently sampled sites to incorporate land-use land-cover and socio-economic conditions into a single geospatial analysis of WNV transmission risk. We extend the list of predictors, through field surveys, with forest structure, composition and configuration data, that we use as an indirect measure of avian community competence. Finally, we incorporate stream water quality predictions obtained through artificial neural network models, based on observed water quality and land-use land-cover characteristics at the watershed level.  

Results/Conclusions

Preliminary results of our mesocosm experiments show that water quality characteristics commonly observed in streams are sufficient for the development and survival of Culex quinquefasciatus larvae, and enable the persistence of mosquito populations. In particular, combinations of nitrogen and phosphorus create optimal conditions in these habitats.  Preliminary results of our geospatial analysis suggest that West Nile virus transmission risk is highly clustered within metropolitan Atlanta, and significantly varies across the city with both land-use land-cover and socio-economic characteristics. In particular, decreased forest cover and low socio-economic status interact to produce the conditions conducive for increased mosquito abundance and West Nile virus transmission risk. Our results have broad implications for other large metropolitan areas across the Southeastern US where Culex quinquefasciatus is the major vector of WNV.