Terrestrial vegetation and aquatic chemistry influence larval mosquito abundance in catch basins, Chicago, USA

Abstract

Background: An important determinant of mosquito-borne pathogen transmission is the spatial distribution of vectors. The primary vectors of West Nile virus (WNV) in Illinois are Culex pipiens Linnaeus (Diptera: Culicidae) and Culex restuans Theobald. In urban environments, these mosquitoes commonly oviposit in roadside storm water catch basins. However, use of this habitat is inconsistent, with abundance of larvae varying significantly across catch basins at a fine spatial scale.

Methods: We tested the hypothesis that attributes of the biotic and abiotic environment contribute to spatial and temporal variation in production of mosquito vectors, characterizing the relationship between terrestrial vegetation and aquatic chemistry and Culex abundance in Chicago, Illinois. Larvae were sampled from 60 catch basins from June 14 to October 3, 2009. Density of shrubs and 14 tree genera surrounding the basins were quantified, as well as aquatic chemistry content of each basin.

Results: We demonstrate that the spatial pattern of Culex abundance in catch basins is strongly influenced by environmental characteristics, resulting in significant variation across the urban landscape. Using regression and machine learning techniques, we described landscape features and microhabitat characteristics of four Chicago neighborhoods and examined the implications of these measures for larval abundance in adjacent catch basins. The important positive predictors of high larval abundance were aquatic ammonia, nitrates, and area of shrubs of height <1 m surrounding the catch basins, whereas pH and area of flowering shrub were negatively correlated with larval abundance. Tree density, particularly of arborvitae, maple, and pear, also positively influenced the distribution of Culex during the fruit-bearing periods and early senescent periods in August and September.

Conclusions: This study identifies environmental predictors of mosquito production in urban environments. Because an abundance of adult Culex is integral to efficient WNV transmission and mosquitoes are found in especially high densities near larval habitats, identifying aquatic sites for Culex and landscape features that promote larval production are important in predicting the spatial pattern of cases of human and veterinary illness. Thus, these data enable accurate assessment of regions at risk for exposure to WNV and aid in the prevention of vector-borne disease transmission.

Figure 1

Sixty larval sampling catch basins in four metropolitan Chicago neighborhoods (Alsip, Evergreen Park, Oak Lawn North, and Oak Lawn South). Black-and-white crosses indicate low larval abundance (average <11 larvae per sample), grey crosses indicate medium abundance (average 11–30 larvae per sample), and black crosses indicate high larval abundance (average >30 larvae per sample).

Figure 2

Seasonal abundance of Culex pipiens and Cx. restuans throughout the study period from the week of June 14th to September 27th, 2009.

Figure 3

Larval abundance and aquatic chemistry content (pH, ammonia, and nitrate) for eight catch basins sampled weekly from June 28th to August 16th, 2009.

Figure 4

Regression tree models (A, C) and variable importance scores for random forest models (B, D) of tree genera density as predictors of larval abundance in the early season (A, B) and late season (C, D) time windows.