Light Trap Collections of Mosquitoes (Diptera: Culicidae) Using Dry Ice and Octenol Attractants in Adjacent Mosquito Control Programs

Abstract

A thorough understanding of the local mosquito fauna is required to develop effective mosquito abatement programs and assess risk of arbovirus transmission in an area. Although many mosquito control districts routinely survey adult populations of mosquitoes, few studies address trap bias, attractant bias, and the sampling effort needed to fully describe mosquito community diversity. Quantifying and visualizing differences in mosquito community composition and abundance, collected in adjacent mosquito control districts that use different attractants, provides a first step toward understanding and communicating cross-district arbovirus risk in continuous geographic areas. We obtained female mosquito collection data from CDC (Centres for Disease Control and Prevention) suction light traps from St. Johns County (baited with octenol) and Duval County (baited with dry ice) in Florida, USA, presenting a unique opportunity to summarize and compare collections across 2 attractants in adjacent Florida mosquito control programs. In the current work, we describe the seasonal distribution of mosquito species, highlight proportions of vector species of importance, quantify and assess diversity, and summarize the variation explained by attractant type using partial redundancy analysis. Numerous actual and potential vector species of importance were abundant throughout the sampling period that included Aedes atlanticus (Dyar & Knab), Anopheles crucians Weidemann, Culex erraticus (Dyar & Knab), and Culex nigripalpus Theobald (all Diptera: Culicidae). Dry ice collections yielded the greatest diversity of species with the least trapping effort. Traps baited with octenol yielded the greatest number of mosquitoes with a greater proportion of vector species. The results of the partial redundancy analysis revealed that attractant explained a significant proportion of the variance in the data set. But a significant linear trend also was present indicating that additional spatially structured variables were responsible for a large proportion of the variance in the data. It is necessary to know the differences between mosquito species and trap numbers when varying collection methods and attractants are used to assess arbovirus transmission risk across mosquito control district administrative boundaries. This information can be used to provide more robust and comprehensive surveillance information capable of identifying new challenges to public health safety.