Urban Wetlands and West Nile Virus: The effects of edge and degree of isolation on mosquito species richness, vector presence, and disease risk

Background/Question/Methods

The effects of urbanization on local vector communities, such as population isolation and homogenization, can have dramatic consequences on the prevalence of zoonotic diseases residing within fragmented urban natural areas, particularly within urban wetlands.  The aim of this study was to analyze the influence of habitat fragmentation and degree of isolation of urban wetland patches on the prevalence of West Nile virus (WNV) within the northeastern United States.  To accomplish this, the impact of wetland fragmentation and isolation on mosquito species richness, the community presence of WNV-competent enzootic and bridge vectors, and the prevalence of WNV in the state of New Jersey (USA) was analyzed over two transmission seasons (2011-2012).  Surveys were conducted within wetland patches representing three size classifications; large (>30 ha), medium (15-30 ha) and small (<15 ha).  The degree of wetland isolation was calculated as each wetlands core to edge area ratio.  Core habitat was defined as all wetland area located at least 100 m from the edge of the wetland. These results were then compared to results originating from a highly developed urban control site and urban residential sites to better understand the influence of urbanization on the transmission of WNV in the Northeast.

Results/Conclusions

It was discovered that a decrease in wetland isolation and an increase in urbanization were positively correlated (R²=0.66, P=0.014) with increases in WNV infection rates in Culex spp. populations.  Overall, no WNV-positive mosquito pools were collected from the most isolated wetland site in our study (1.15:1 core/edge ratio) over the two year study period.  Oppositely, the least isolated wetland site in our study (0.25:1 core/edge ratio) recorded the highest WNV infection rates for all of the wetland patches surveyed (MIR=10.14/1000).  The least isolated wetland patch had an infection rate most similar to that of urban residential (MIR=9.8/1000) areas and the urban control site (MIR=11.5/1000).  Furthermore, a decrease in wetland isolation and an increase in urbanization were positively correlated (R²=0.74, P=0.038) with increases in the community presence of WNV-competent vector species.  Overall, WNV-competent vector species (i.e. Culex pipiens, Aedes albopictus, Ochlerotatus japonicus) comprised 76.7±8.3% of the mosquito community residing in the urban control site and only 21.4±5.9% of the mosquito community within the largest, most isolated, wetland patch.  Given these results, we recommend the continued conservation of large and medium sized urban wetland patches as a potential method of disease control.