Environmental influences on West Nile virus transmission in the United States---Space and time analyses.

West Nile virus (WNV) was introduced in New York in 1999 and become one of the most important vector-borne diseases in the U.S. This dissertation analyzed WNV national case data from CDC and mosquito and bird infection from Saginaw County Mosquito Abatement Commission (SCMAC) to investigate various environmental factors and vector ecologies influencing risk. Spatial analysis on national-wide human cases during 2002 through 2006 indicated the Central region of the U.S. remained a major transmission focus beginning in 2003. Midwestern and southern states also experienced intense transmission during these years. WNV transmission predominated within mixed rural/non-agriculture and rural/agriculture counties. Agricultural land use was a risk factor regardless urbanization level. Investigations in Saginaw County showed that both Culex pipiens and Cx. restuans were important vectors with high infection rates. Infected dead Crows appeared earlier than Blue Jays, but WNV incidence for both species peaked around mid-August. Spatial clusters were found in different locations each year and included urbanized areas. The high WNV infection among mosquitoes and wild birds demonstrates significant transmission; however, the transmission to people may be limited by spatial and/or mosquito abatement. A long-term mosquito database (1989--2005) was analyzed to explore the association between mosquito abundance and climate variation. Preceding precipitation and relative humidity were significantly associated with abundance of Ae. vexans during the whole period. For Cx. pipiens , precipitation associated abundance was negative in summer and the prior minimum temperature was more important in July. Abundance of Cx. restuans was associated with preceding relative humidity in May and June. Maximum temperature had negative effects for all species during summer. This study demonstrated that climate has different impacts on mosquito abundance depending on the species, suggesting implications for WNV transmission. These results demonstrated WNV hotspots in the U.S. which should be considered as disease prevention foci. The spatial and temporal patterns of WNV transmission and climate-mosquito interactions at local areas may provide useful information for disease prevention and vector control.