Using weather to predict mosquito abundance

Short-term Forecasting of Daily Abundance of West Nile Virus Vectors Culex pipiens-restuans (Diptera: Culicidae) and Aedes vexans Based on Weather Conditions in Southern Québec (Canada). Ripoche M, Campagna C, Ludwig A*, Ogden NH*, Leighton PA. J Med Entomol 2019 Apr 16;56(3):859-72 doi:

This science story describes the use of a statistical model to forecast disease vector abundance to help understand and reduce public health risks. Risk models could incorporate other factors linked to disease vectors and may help improve understanding of the effects of climate change on the potential expanding range of West Nile virus vectors.

What was known about this area prior to your work, and why was the research done?

West Nile virus (WNV) is a virus that is usually spread by infected mosquitos. Canada’s first human case of WNV was detected in 2002. Infections have been reported annually with a peak of over 2200 reported cases in 2007. As there are no vaccines against WNV nor medications to treat infections, the best method to prevent a WNV infection is to prevent mosquito bites. Despite ongoing monitoring and control efforts, outbreaks of WNV in Canada continue to occur sporadically and remain challenging to predict. The risk of human WNV infection is strongly associated with the abundance of infected mosquitoes in the environment. Predicting seasonal trends in mosquito abundance has the potential to provide timely and relevant information for public health, making a useful indicator of human exposure risk. A previous study in Ontario developed a statistical model to predict daily abundance of the Culex pipiens-restuans (Cx. pipiens-restuans) mosquito in the Toronto area, a region with a similar epidemiological and ecological context to that of southern Québec. This research built on the previous model to include an additional WNV vector (Aedes vexans) and other factors to predict the abundance of mosquitos in southern Québec using short-term weather conditions.

What are your most significant findings from this work?

Mosquito abundance in the study region was determined by using insect traps in a variety of habitat locations (e.g. urban and forested areas) over an extended period of time. Temperature and precipitation were found to be significantly associated with mosquito abundance. However, the lag time for these factors to affect insect counts was different, a similar distinction was also observed between the mosquito species. Once configured properly, the model was able to predict the mosquito abundance data of 2013 and 2014 for Cx. pipiens-restuans, the main vectors for West Nile virus in southern Québec. The model was less precise for the second vector, Aedes vexans. The consideration of other factors (e.g. insect control measure, presence of hosts, habitat type) significantly improved the model accuracy.

What are the implications or impact of the research?

The fight against vector borne diseases is multi-pronged and further strengthened through innovative work like the development of accurate statistical models. Being able to forecast WNV vector mosquito abundance during summer using short-term weather conditions will help anticipate periods of risk for WNV transmission to humans. The predicted trend in mosquito abundance would provide timely and relevant information for appropriate public health action. Proactive steps could be taken to mitigate potential exposure and alert Canadians to WNV risk ahead of seasonal outbreaks.

Additional References of Significance:

  • Wang J, Ogden NH*, Zhu H. The impact of weather conditions on Culex pipiens and Culex restuans (Diptera: Culicidae) abundance: a case study in Peel Region. J Med Entomol 2011 Mar; 48(2):468-75. doi:
  • Ogden NH*, Lindsay LR*, Ludwig A*, Morse AP, Zheng H, Zhu H. Weather-based forecasting of mosquito-borne disease outbreaks in Canada. Can Commun Dis Rep 2019 May 2;45(5):127–32. doi:

Date modified: