Remotely-sensed vegetation indices identify mosquito clusters of West Nile virus vectors in an urban landscape in the northeastern United States

Heterogeneity in urban landscapes can influence the effectiveness of mosquito-borne disease control. We used remotely sensed vegetation indices to discriminate among mosquito habitats within a densely populated urban environment in New Haven, CT. ASTER derived vegetation indices were identified for 16 sites where adult mosquitoes were trapped over the summer of 2004. Canonical correlation analysis showed a significant relationship between the environmental variables (normalized difference vegetation index, disease/water stress index and distance to water) and four local West Nile virus competent vectors (Cx. pipiens, Cx. restuans, Cx. salinarius, and Ae. vexans) (0.93, P = 0.03) explaining 86% of the variance in the environmental and mosquito measures. Sites were clustered based on these remotely sensed environmental variables. Three clusters were identified which provide insight into the distribution of West Nile virus vectors in an urban area. Identification of habitat differences of mosquitoes within the urban landscape has important implications for understanding West Nile virus transmission and for control of vector-competent mosquito species.