Table 1.
List of parameters recorded for each sample.
Table 2.
Abundance and pest status of mosquito species found in SCMs.
Table 3.
Mosquito abundance per sample (APS) found in SCMs.
Fig 1.
Prevalence and APS of mosquitoes found in SCM sites during (A) 2021 and (B) 2022. The prevalence of mosquitoes indicates the percentage of sites with at least one mosquito larva collected (positive count data). The APS (mosquito abundance per sample) was calculated by dividing the total number of mosquitoes collected in a site during a single sampling period by the number of samples taken. Sites with zero APS were removed. The box plots show the median, quartiles, minimum, maximum, and outliers among the sites belonging to each SCM type and period.
Table 4.
Results of the ZA-GLMM estimating mosquito presence (binomial model) and APS (conditional model) in 2021.
Table 5.
Results of the pairwise comparisons between SCM types in ZA-GLMMs estimating mosquito APS in 2021.
Table 6.
Results of the ZA-GLMM estimating mosquito presence (binomial model) and APS (conditional model) in 2022.
Fig 2.
Partial Least Squares Canonical Analysis (PLSCA) plots representing the correlations between predictor and response variables.
Plots represent data from the (A, C) 2021 season and (B, D) 2022 season. (A, B) Vectors emanating from the origin of the circle to each pair of variables form angles that represent their respective relationships. Blue vectors represent mosquito species or total mosquitoes (response variables), purple vectors represent SCM types (predictor variables), and orange vectors represent environmental parameters (predictor variables). Acute angles represent positively correlated variables with more acute angles indicating a stronger correlation, whereas obtuse angles represent variables that are negatively correlated. The circle in each correlation plot has a radius of 1.0. The length of the vector represents the magnitude of how well the variable can be directly interpreted from the axes chosen for the plot. (C, D) The clustered image maps (CIMs) show the strength and sign of the approximate Pearson correlations between every pair of predictor and response variables. The values used to create these CIMs can be found in S1 Table. The dendrogram associated with each CIM is a representation of the hierarchical clustering used to organize the variables in each CIM. Abbreviations are as follows: Aedes vexans (Ae.ve), Anopheles spp. (An.), Culiseta inornata (Cs.in), Culex erraticus (Cx.er), Cx. pipiens (Cx.pi), Cx. restuans (Cx.re), Cx. salinarius (Cx.sa), Cx. territans (Cx.te), Psorophora ferox (Ps.fe), Uranotaenia sapphirina (Ur.sa), total mosquitoes (Mosq), constructed wetlands (CW), detention ponds (DP), retention ponds (RP), conductivity (Cond), pH (pH), shade (Shad), temperature (Temp), turbidity (Turb), vegetation on land (VegL), and vegetation within water (VegW).
Table 7.
Mean and standard deviation of alpha diversity indices for each SCM type in 2021.
Table 8.
Mean and standard deviation of alpha diversity indices for each SCM type in 2022.
Fig 3.
Non-metric multidimensional scaling analyses using Bray-Curtis dissimilarities of mosquito communities.
Plots represent data from the (A) 2021 season and (B) 2022 season. Each individual point represents a single site, each color represents a different SCM type, and each colored ellipse represents the 95% confidence interval encompassing the population of sites. The NMDS plots were determined to be fair representations of the data (2021: stress = 0.010; 2022: stress = 0.11).