Integrated management of the mosquito vectors in West Nile virus in Wisconsin

In order to effect efficient, economical and environmentally sound mosquito abatement it is necessary to be forearmed with information about the local species spatial and temporal population dynamics as well as their potential to transmit West Nile virus (WNv). To this end we conducted several research projects in the Madison, Wisconsin area.;The first project involved generating a wetlands inventory of all possible mosquito larvae habitats in the area and sampling for larvae. The majority of sites did not produce any mosquito larvae and a subset of 12 sites was continuously productive for Culex larvae.;It was determined in the laboratory that fathead minnows (Pimephales promelas) would consume mosquito larvae. We introduced of fathead minnows in three of the 12 productive sites. No further abatement was needed for the sites where fish were introduced, but the nine control sites required abatement.;Other potential oviposition sites for Culex include the subterranean storm water catch basins. I looked at a randomly selected subset of these catch basins for three years (2004--2007). During this period I sampled for the presence of Culex larvae, took water quality measurements, and recorded physical characteristics of the catch basins. When 2.54 cm of rainfall occurs in the preceding seven days few of the catch basins have mosquito larvae. Water quality is not correlated to mosquito populations in catch basins. One physical parameter is negatively associated with mosquito larvae, the more inlets which feed into a catch basin the lower is the probability of finding larvae.;The final research project focused on adult mosquito populations. Utilizing a previous published risk model which incorporated a mosquito species' relative abundance, vector competence, predilection for feeding on mammals, and its prevalence of infection it was possible to predict which species are most likely to transmit WNv to humans. In the Madison area, Aede. vexans and Ochlerotatus trivittatus pose the greatest risk, while the risk from Culex pipiens/restuans is low. The model is capable of predicting the short term number of human WNv cases and was accurate for the Madison area as well as two of three other Midwestern cities.