Assessing mosquito breeding sites and abundance using an unmanned aircraft

Abstract

An unmanned aircraft system (UAS; i.e., drone) with an attached multispectral camera was used to quantify accumulated surface water on a 0.54-km2 tidal marsh that abuts San Francisco Bay, CA, USA. The results of the survey showed unequal accumulation of surface water and provided information for focused inspections of potential mosquito breeding areas and identified areas where existing ditches needed improvement for increasing water circulation in the marsh to reduce mosquito breeding. The UAS was also outfitted with a high-magnification zoom video camera and piloted at varying heights to measure the video camera's ability to visualize immature mosquitoes in 2 small containers of contrasting colors during simulation tests in a marsh habitat. Immature mosquitoes could be seen clearly in white or black containers at heights up to 14 and 8 m, respectively. An artificial intelligence algorithm identified mosquito larvae and pupae in videos of the white tray with 94.1% and 52.8% accuracy, respectively. Together, our studies show that an UAS equipped with multispectral and zoom cameras provides a means for vector control agencies to rapidly and quantitatively assess the landscape for the presence of surface water and mosquito larvae.