Supporting data for: Ecological predictors of mosquito population and arbovirus transmission synchrony estimates

Description

We examined seasonal mosquito and arbovirus surveillance data collected in Connecticut, United States from 2001 – 2020 to quantify spatial relationships in 19 mosquito species and 7 arboviruses time series while accounting for environmental factors such as climate and land cover characteristics as well as Euclidean distance. We determined that mosquito collections, on average, were significantly correlated up to 10 km though highly variable among the examined species. Few arboviruses displayed any synchrony and significant maximum correlated distances never exceeded 5 km. After accounting for distance, mixed effects models showed that mosquito or arbovirus identity explained more variance in synchrony estimates than climate or land cover factors. Correlated mosquito collections up to 10 – 20 km suggest that mosquito control operations for nuisance and disease vectors alike must expand treatment zones to regional scales for operations to have population-level impacts. Species identity matters as well, and some mosquito species will require much larger treatment zones than others. The much shorter correlated detection distances for arboviruses reinforce the notion that focal-level processes drive vector-borne pathogen transmission dynamics and risk of spillover into human populations. Specific metadata files and underlying data are supplied in this archive and methods needed to reproduce the results are detailed in the manuscript. In brief, the .csv files beginning with "Yr..." contain the annual, effort-corrected collections for either mosquito species or arboviruses collected in either ground-level, CO2-baited CDC light traps or ground-level CDC gravid traps at 87 surveillance sites in Connecticut, U.S. The .csv file beginning with "Coordinates.." contains the latitude and longitude coordinates (in decimal degrees) for each of the 87 surveillance sites. The file named "Mmetadata_Upload) contains brief explanations of each .csv file. Users will need a working knowledge of programming in R in order to replicate any and all methods and results presented in the manuscript.

Steps to reproduce

Please see the manuscript for specific details on how the data was collected, analyzed, and displayed.

Institutions

Categories

Funding

Licence