SEARCH

SEARCH BY CITATION

Keywords:

  • Mosquito community;
  • weather;
  • land use/cover

ABSTRACT:

  1. Top of page
  2. ABSTRACT:
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments
  8. REFERENCES CITED

Mosquito community composition and population dynamics were compared to weather variables and land use/cover data during 2008 to determine which variables affected population dynamics at the J.W. Jones Ecological Research Center in southwestern Georgia. Models relating adult mosquito distributions to weather variables and time of year were compared using Akaike's Information Criterion (AIC) model selection. Precipitation, temperature, humidity, and Keetch-Byram Drought Index were important factors correlated with mosquito abundance or presence/absence for the species considered. A cluster analysis, which grouped eight sites based on the percentages of land use/cover and hydric soils located in a 1-km radius surrounding collection sites, and an indicator species analysis were used to investigate the associations among 11 mosquito species and sites with similar land use/cover. Aedes albopictus (Skuse), Culex coronator Dyar & Knab, Culex quinquefasciatus Say, and Culex salinarius Coquillett were associated with sites that had the most anthropogenic influence, while Coquillettidia perturbans (Walker) and Psorophora ferox (von Humboldt) were associated with natural land cover such as wetlands and forested land. This study demonstrates that regional climate and land use/cover data can be predictive of the population dynamics of certain mosquito populations and is the first to examine how the distribution of Cx. coronator adults relate to land use/cover in the southeastern United States.


INTRODUCTION

  1. Top of page
  2. ABSTRACT:
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments
  8. REFERENCES CITED

Understanding the spatial and temporal dynamics of mosquito communities is important for implementing control measures as well as assessing vector-borne disease prevalence within an area (Godsey et al. 2005, DeGroote et al. 2007). Several mosquito-borne viruses that cause disease in humans and other animals circulate in Georgia each year, including St. Louis encephalitis virus (SLEV), LaCrosse encephalitis virus (LACV), Eastern Equine encephalitis virus (EEEV), and West Nile virus (WNV) (Lance-Parker et al. 2002). Epizootics of these viruses occur sporadically but within relatively well-defined seasonal periods (GDCH 2010).

The occurrence and distribution of specific mosquito species in Georgia have been documented in a number of studies (eg., King et al. 1960, Smith and Floore 2001, Gray et al. 2005, Kelly et al. 2008). Other investigations have examined the relationship between land use/land cover data, human demographic data, climate ecological correlates, and WNV infection in birds (Gibbs et al. 2006, Bradley et al. 2008). However, none of these studies examined the relationship between potential disease vectors and ecological factors affecting mosquito populations in the Gulf Coastal Plain of southwestern Georgia. We compared mosquito community composition data to selected weather variables and land use/land cover data during the 2008 mosquito season to determine whether mosquito population dynamics are related to weather patterns or landscape variation.

MATERIALS AND METHODS

  1. Top of page
  2. ABSTRACT:
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments
  8. REFERENCES CITED

Study area

The study was conducted at Ichauway, the property of the J.W. Jones Ecological Research Center in southwestern Georgia. Ichauway is a remnant 11,736 hectare longleaf pine-wiregrass ecosystem embedded within an agricultural landscape dominated by center-pivot irrigated row crops (Michener et al. 1998, Smith et al. 2006). Relative humidity, precipitation, temperature, and Keetch-Byram Drought Index data for the study were obtained from the Georgia Automated Environmental Monitoring Network weather station on Ichauway property (listed as the Newton station at http://www.georgiaweather.net).

Sampling methods

Seven locations for collection sites were selected by creating two intersecting transects across Ichauway property and choosing points along those transects. To sample the mosquito population at a location that receives agricultural runoff, we chose an additional site that was not on the transects (Hall Pond) (Figure 1). Adult female mosquitoes were collected through weekly trapping at each site from mid-May through early November, 2008, using Centers for Disease Control and Prevention (CDC) miniature light traps and gravid traps (John W. Hock, Gainesville, FL) (Silver 2008). Light traps were baited with dry ice, powered with 6-V batteries, and suspended 1.5 m above the ground. Gravid traps were baited with hay-infused water and powered with 6-V batteries. Traps were operated from approximately 17:00 until 07:00 the next morning. Collected mosquitoes were transported alive to the laboratory and anaesthetized using triethylamine (Kramer et al. 1990). Adults were identified to species using standard keys (Darsie and Morris (2003) and Darsie and Ward (2005)). For collections that resulted in large quantities of mosquitoes after Tropical Storm Fay in August, the total number of individuals collected and species composition were estimated using the mean of three sets of 100 randomly chosen individuals of mixed species that were weighed and identified (Schäfer et al. 2008).

image

Figure 1. The eight study sites at Ichauway in southwestern Georgia, USA, where adult mosquitoes were collected from May to November 2008. The 1-km radius surrounding each site used in spatial analysis is outlined in black.

Download figure to PowerPoint

Spatial analysis

To quantify land use/cover around sites, a map was created from 1:12,000 m scale 1992 color infrared aerial photography that is updated frequently to reflect changes in land cover from more recent photographs. Using ArcGIS (ESRI, Redlands, CA, v. 9.1), the percent area of nine land use/cover categories (Table 1) were determined within a 1-km radius of each collection site (total area = 3.14 km2). In order to quantify the percentage of hydric and non-hydric soil within the 1-km radius of each site, a data layer was created from the Natural Resources Conservation Service Baker County Soil Survey.

Table 1.  The percent area of nine land use/cover categories and hydric soils within a 1-km radius of eight study sites (total area = 3.14 km2) at Ichauway assessed using ArcGIS (ESRI, Redlands, CA, v. 9.1).
Land use/cover category and soilPercent area within 1-km radius of sites
Big SloughCollins PondHorse HammockMim's DrainRichardson FlatFlint RiparianParmalee PondHall Pond
Coniferous forest76.979.270.886.664.443.059.854.4
Deciduous forest1.65.12.43.611.83.91.81.8
Bottomland deciduous3.60.012.92.60.013.10.00.5
Forested wetland0.65.11.50.00.10.05.60.2
Non-forested wetland0.00.20.00.022.61.32.33.6
Scrub4.81.20.90.80.05.91.00.0
Agriculture8.88.33.45.91.025.826.339.5
Residential0.10.95.20.20.00.13.20.1
Open water3.50.02.90.50.06.90.00.0
Hydric soils10.519.39.50.024.810.313.18.8

Data analysis

Eleven mosquito species collected from both light and gravid traps were selected for use in statistical analyses based upon their abundance in sampling collection and/or their status as a potential arbovirus vector (Table 2) (Turell et al. 2005, Cupp et al. 2007, Kelly et al. 2008). Given that the weather data fluctuated over the study, while the spatial data remained constant, the relationships between these factors and total mosquito abundance were analyzed independently. As part of the weather data analysis, 52 a priori models were developed to explain daily abundance or presence/absence of selected mosquito species using combinations of 11 predictor variables (Table 2) based on the importance of these variables in the literature and field observation. Though the possible combinations of the 11 predictor variables are endless, the number of a priori models used in this analysis was limited to 50 along with the null and global models (Conner et al. 2010).

Table 2.  Best fitting models explaining the effects of weather variables and time on mosquito abundance or presence/absence for all species considered. Models with lowest second order Akaike's information criterion difference (Δ AICc) and greatest Akaike weight (wi) were more supported by data and are presented here. The number of predictor variables (K) in each model included the intercept term.
ModelKΔ AICcwi
Ae. albopictus   
wk_2_rh wk_1_rh wk_2_temp40.000.44
Ae. triseriatus   
wk_4_precip wk_6_precip30.000.23
Ae. vexans   
wk_2_precip wk_4_precip wk_6_precip d_1_rh d1_temp wk_1_temp wk_2_temp time90.000.46
Cq. perturbans   
wk_1_temp d1_temp wk_1_rh40.000.13
Cx. coronator   
wk_2_precip wk_4_precip wk_6_precip d_1_rh d1_temp wk_1_temp wk_2_temp time90.000.30
Cx. erraticus   
wk_2_precip wk_4_precip wk_6_precip d_1_rh wk_1_rh time70.000.48
Cx. nigripalpus   
wk_4_precip wk_6_precip wk_2_rh d1_temp wk_2_temp DI70.000.19
Cx. quinquefasciatus   
wk_6_precip d1_temp wk_1_temp wk_1_rh50.000.21
Cx. restuans   
wk_2_precip wk_6_precip wk_2_rh d1_temp wk_1_temp60.000.28
Cx. salinarius   
wk_4_precip20.000.08
Ps. ferox   
wk_6_precip wk_1_temp DI time50.000.17

The second-order Akaike's Information Criteria (AICc) analysis for small sample sizes was used to identify the best model for each mosquito species and to predict variable importance (Burnham and Anderson 2002). As part of the AIC analysis, daily log-transformed abundance of Aedes vexans (Meigen) was analyzed as a function of all 52 a priori models using PROC GENMOD (SAS Institute Inc., Cary, NC). Due to the presence of multiple zero values in abundance data for all other mosquito species considered, the daily presence/absence of those species was analyzed as a function of all 52 a priori models using PROC LOGISTIC (SAS Institute Inc., Cary, NC). For each mosquito species, the model with the lowest AICc value was considered to be the best model (Burnham and Anderson 2002, Conner et al. 2010). Model averaging was used to calculate parameter estimates and unconditional standard errors for all predictor variables (Burnham and Anderson 2002, Conner et al. 2010). Predictor variables were considered useful in explaining mosquito abundance or presence/absence if the 95% confidence interval surrounding the parameter estimate did not contain zero and the variable weight was high (i.e., > 0.80) (Miles et al. 2006, Miller and Conner 2007, Conner et al. 2010).

To study landscape effects, we used a cluster analysis to group sites based on percentages of land use/cover and hydric soils in the 1-km radius surrounding collection sites using PC-ORD (MjM Software Design, v.5.10). Associations between groups of sites and the subset of mosquito species selected for analysis were assessed using the indicator species analysis (ISA) in PC-ORD (Dufrene and Legendre 1997). ISA was performed on log-transformed weekly mosquito abundance data collected from each site grouped according to the cluster analysis results. The indicator species analysis produces indicator values (IV) for each species based on the combination of species abundance and faithfulness of occurrence for a particular group (McCune et al. 2002). If the P-value for the indicator value is statistically significant, then the species is considered to be an indicator of that group (McCune et al. 2002). An IV closer to 100 indicates a stronger species association with that group (McCune et al. 2002).

RESULTS

  1. Top of page
  2. ABSTRACT:
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments
  8. REFERENCES CITED

Mosquito species

Approximately 58,000 mosquitoes were collected over 355 trap nights on 78 dates. The mosquitoes belonged to 30 species within the genera Aedes, Anopheles, Coquillettidia, Culiseta, Culex, Orthopodomyia, Psorophora, and Uranotaenia (Table 3). Four species (Ae. vexans (50 %), Ps. ferox (25%), Aedes sticticus (Meigen) (4%), and Anopheles crucians Weidemann (3%)) comprised over 80% of the total mosquitoes collected.

Table 3.  Number of mosquitoes per species collected during May-November, 2008 at eight sites in Ichauway, the property of the J. W. Jones Ecological Research Center in southwestern Georgia. Locations of sites can be found in Figure 1.
SpeciesBig SloughCollins PondFlint RiparianHall PondHorse HammockMim's DrainParmalee PondRichardson Flat
Ae. albopictus (Skuse)21117329305
Ae. atlanticus (Dyar & Knab)03000012
Ae. canadensis canadensis (Theobald)712090086153134
Ae. fulvus pallens Ross669631376734
Ae. infirmatus (Dyar & Knab)18829722120279865
Ae. mitchellae (Dyar)3500828541815
Ae. sticticus (Meigen)1,3641366106412394021
Ae. thibaulti Dyar & Knab1090060634
Ae. triseriatus (Say)232713536104550
Ae. vexans (Meigen)2,7305,8873,6793,3843,0073,0805,0084
An. crucians Weidemann1058701465439141022,428
An. punctipennis (Say)75252595651182
An. quadrimaculatus Say1451389478185
Cq. perturbans (Walker)02113278
Cs. melanura (Coquillett)010001012
Cx. coronator Dyar & Knab331085722580
Cx. erraticus (Dyar & Knab)71624882225569
Cx. nigripalpus Theobald8345421502839948
Cx. quinquefasciatus Say592816935951146169112
Cx. restuans Theobald1318806831113279
Cx. salinarius Coquillett405211,54493031
Cx. territans Walker3537127409109
Or. signifera (Coquillett)42221534
Ps. ciliata (Fabricius)129223073401350
Ps. columbiae (Dyar & Knab)1323775214527
Ps. cyanescens (Coquillett)9101703841516097
Ps. ferox (von Humboldt)7,20342617766,09427134120
Ps. howardii Coquillett17361341340645
Ps. mathesoni Belkin & Heinemann48000000
Ur. sapphirina (Osten Sacken)11180712221
Total12,2058,5434,3947,38910,5394,2187,3143,357

Weather patterns

Over the sampling period, temperatures ranged from –0.4° C to 37.4° C. Average daily humidity ranged from 45.1% to 98.2%. The total rainfall during the study was 50.62 cm. Weather patterns had a strong influence on the mosquito species modeled, but predictive variables varied by species. The model for predicting daily abundance of Ae. vexans that yielded the lowest second order Akaike's information criterion difference (Δ AICc) and greatest Akaike weight (wi= 0.46) contained all precipitation variables considered, one day average relative humidity and temperature, one and two week average temperature, and time (Table 2). The best model for predicting the presence/absence of all other mosquitoes varied by species (Table 2).

Focusing on the importance of individual variables through the use of model averaged parameter estimates suggested that abundance of Ae. vexans was best predicted by the following variables: total precipitation for two and six weeks, average temperature for one day and two weeks, and average relative humidity for one day (Table 4). Of the variables examined, the daily presence/absence of Ae. albopictus was best predicted by average relative humidity for one and two weeks. Six-week total precipitation best described daily presence/absence of Aedes triseriatus (Say). Cx. coronator daily presence/absence was best described by total precipitation for four weeks and time of year, while Culex erraticus (Dyar & Knab) daily presence/absence was best described by these variables as well as average relative humidity for one day and one week. Six-week total precipitation, one-day average temperature, and drought index were most useful in describing Culex nigripalpus Theobald daily presence/absence. The pattern of Culex restuans Theobald daily presence/absence was best described by the one-day average temperature.

Table 4.  Model-averaged parameter estimates, their unconditional standard errors (SE), 95% confidence intervals (CI), and variable weights for parameters used to predict daily abundance or presence/absence of selected mosquito species during the 2008 mosquito season in southwestern Georgia, U.S.A. Data are only presented for those parameters with 95% confidence intervals that did not contain zero and with a high variable weight.
VariableCoefficientSE95% CIVariable Weight
Ae. albopictus    
wk_1_rh0.240.090.06–0.420.9698
wk_2_rh–0.350.11–0.58––0.130.9862
Ae. triseriatus    
week6precip–0.010.003–0.02––0.0010.9999
Ae. vexans    
week2precip–0.010.003–0.015––0.0020.9999
week6precip0.0040.002–0.17––0.030.9999
d1_temp0.180.080.03–0.340.9887
wk_2_temp–0.650.27–1.17––0.130.9987
d_1_rh–0.060.02–0.11––0.020.9900
Cx. coronator    
week4precip0.090.040.01–0.160.9774
time2.971.510.01–5.930.9996
Cx. erraticus    
week4precip0.070.030.01–0.140.9968
d_1_rh–0.110.05–0.20––0.010.9994
wk_1_rh–0.310.13–0.57––0.060.9489
time1.590.750.1162–3.05970.8666
Cx. nigripalpus    
week6precip–0.080.04–0.15––0.0040.9999
d1_temp–0.970.38–1.72––0.230.9900
DI–0.030.01–0.0434––0.00770.9724
Cx. restuans    
d1_temp–0.450.23–0.89––0.0020.9929

Spatial distribution

The cluster analysis grouped sites based on percentages of land use/cover and hydric soils into the following three assemblages: Group I: Big Slough, Collins Pond, and Horse Hammock, Group II: Mim's Drain and Richardson Flat, Group III: Flint Riparian, Hall Pond, and Parmalee Pond (Figure 2). Of the 11 species considered, ISA identified seven with significant indicator values (P < 0.05) (Table 5). Ps. ferox (IV = 27.5, P= 0.002) was the only species associated with the sites in Group I. Ae. albopictus (IV = 17.7, P= 0.027) and Cq. perturbans (IV = 12.8, P= 0.006) were both associated with the sites in Group II. Cx. coronator (IV = 17.1, P= 0.048), Cx. quinquefasciatus (IV = 34.7, P= 0.006), and Cx. salinarius (IV = 28.8, P= 0.002) were all associated with the sites in Group III.

image

Figure 2. A cluster analysis dendrogram of the percent area of the nine land use/ cover categories listed in Table 1 and hydric soils within a 1-km radius of eight study sites created using PC-ORD. Site groupings interpreted from the dendrogram are outlined in black.

Download figure to PowerPoint

Table 5.  Indicator species analysis after Dufrene and Legendre (1997) computing indicator value (IV) coefficient of selected mosquito species across sites grouped by landscape features in southwestern Georgia, U.S.A.
SpeciesGroupIVMeanS. DevP*
  1. Group = identifier for group with maximum observed IV.

  2. * Indicates P is significant at 95% level.

Ae. albopictusII17.712.02.360.027*
Ae. triseriatusII15.713.62.540.176
Ae. vexansI35.834.21.440.127
Cx. coronatorIII17.112.22.450.048*
Cx. erraticusIII20.116.92.670.121
Cx. nigripalpusII12.512.42.430.396
Cx. quinquefasciatusIII34.726.02.510.006*
Cx. restuansII8.810.92.470.822
Cx. salinariusIII28.811.42.460.002*
Cq. perturbansII12.85.91.880.006*
Ps. feroxI27.512.52.590.002*

DISCUSSION

  1. Top of page
  2. ABSTRACT:
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments
  8. REFERENCES CITED

AIC analysis suggested that total precipitation over two and six weeks, average temperature over one day and two weeks, and daily average relative humidity were the most influential weather variables affecting the daily abundance of Ae. vexans. The correlation with abundance and precipitation is likely because this mosquito is a floodwater species, and immature stages are often found in temporary rain-filled pools (Carpenter and LaCasse 1955). The positive correlation between Ae. vexans daily abundance and average temperature on collection day is potentially a reflection of warm temperatures encouraging host-seeking (Shone et al. 2006). The Asian tiger mosquito, Ae. albopictus, is susceptible to body fluid loss at low relative humidity (Hylton 1967), which may explain the importance of average relative humidity in modeling the daily presence/absence of this species.

In the southeastern United States, Cx. coronator and Cx. erraticus lay their eggs on the surface of standing water and usually exhibit a peak in population density in mid- to late summer (Bolling et al. 2005, Goddard et al. 2006). These life-history characteristics potentially explain the importance of four-week total precipitation and time of year in modeling their presence/absence. Negative correlations between daily presence/absence of Cx. nigripalpus, six-week total precipitation, and Keetch-Byram Drought Index we observed are consistent with those of Shaman et al. (2003), who found that antecedent drought, coincident wetting, and the emergence of Cx. nigripalpus were all associated with increased prevalence of SLEV in wild birds in southern Florida. Ae. triseriatus is a tree hole/container breeder and its negative correlation with six-week total precipitation may reflect the effect of rainfall on the volume and persistence of water within tree holes and other potential breeding sites. In a survey of tree holes in Pennsylvania, Paradise (2004) reported that higher densities of Ae. triseriatus larvae were always found in tree holes that maintained high water volumes.

Cq. perturbans, Cx. quinquefasciatus, Cx. salinarius, and Ps. ferox presence/absence were not successfully predicted using weather variables in our study. However, the results of the ISA suggested that land use/cover could be useful in predicting the presence/absence of these species. The sites delegated into Group I by the cluster analysis are all forested wetlands with surrounding land use/cover comprised of primarily coniferous forest with little cultivated land (Table 1). Ps. ferox's ISA-based association with these sites reflects its preference for breeding in woodland pools (King 1960).

The ISA-based association between Cq. perturbans and Group II sites is likely because it was collected in greatest abundance from Richardson Flat, the site with the largest percentage of wetland within its 1-km radius (Table 1). Rochlin et al. (2008b) found that abundance of Cq. perturbans in Suffolk County, NY increased with wetland area surrounding a site. Ae. albopictus is associated with Group II due to its abundance at Mim's Drain (Table 3). Although natural breeding sites for Ae. albopictus (i.e., tree holes) could be found at all sites, the abundance of this species collected from Mim's Drain may be because it is located near a residential area where artificial containers are present, which could allow for increased breeding. In a recent north-central Florida study, Obenauer et al. (2009) found the greatest abundance of Ae. albopictus near residential areas.

Cx. coronator's association with Group III sites (approximately 25–40% of their 1-km radius in cultivated land) is potentially important, because this mosquito species, common to the American tropics, has recently expanded its range in the southeastern United States (Gray et al. 2008). Although observations on the occurrence and distribution of Cx. coronator larvae and adults in the southeastern United States exist (Goddard et al. 2006, Gray et al. 2008, Kelly et al. 2008, Moulis et al. 2008), our study is the first that examines Cx. coronator adult distribution in relation to land use/cover. WNV and SLEV have been detected in Cx. coronator mosquitoes (Hammon and Reaves 1943, Mackay et al. 2008), but the effect of this species on arbovirus transmission in areas where it has been recently found is unknown (Kelly et al. 2008). Studying the link between the distribution of this species, its life history, and land use/cover in the southeastern United States is important to understanding its potential as a disease vector.

Cx. salinarius’ relatively high IV coefficient of 28.8 with Group III is most likely due to 95% of its total abundance trapped at Hall Pond (Table 3). Cx. salinarius breeds in fresh and saltwater marshes, lakes, and ponds but can also be found in water polluted with organic matter (Rochlin et al. 2008a). Hall Pond receives runoff from an adjacent agricultural area and has higher nutrient and suspended solids concentrations compared to undisturbed wetlands (S. Golladay, J.W. Jones Research Center, unpublished data), which makes it a suitable breeding site for Cx. salinarius. Cx. quinquefasciatus, the primary vector of WNV in Georgia, also breeds in water rich in organic matter (Calhoun et al. 2007). Its IV coefficient of 34.7 was the highest of all species examined for the group of sites with the least natural land cover. Our observations are similar to those reported on the spatial distribution and abundance of Cx. quinquefasciatus in Hawaii by Reiter and Lapointe (2007), who found agricultural lands and forest fragmentation significantly increased the abundance of this species.

Although our study represents an initial, exploratory analysis, it demonstrated that regional climate and land use/cover data can be predictive of the population dynamics of certain mosquito populations. It is important to note that the seasonal distribution and abundance of mosquitoes documented here reflect a year that had extreme drought conditions followed by a tropical storm. Therefore, our findings may not apply to years when different climatic conditions prevail. Sampling of larval as well as adult mosquitoes at replicated sites encompassing a wider range of conditions would enable us to improve predictive models and better understand this component of mosquito ecology and arbovirus transmission in southwestern Georgia. In combination with similar analyses of vertebrate host population ecology, these methods could provide a much clearer picture of mosquito interactions that affect human and veterinary health in this region.

Acknowledgments

  1. Top of page
  2. ABSTRACT:
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments
  8. REFERENCES CITED

We thank Greg Knothe for assistance in the field and in the lab. The Joseph W. Jones Ecological Research Center and University of Georgia provided funding for this project. We thank Jean Brock for her assistance with ArcGIS. Mike Conner's help with AIC was invaluable. We appreciate the comments of two anonymous reviewers whose suggestions greatly improved this manuscript.

REFERENCES CITED

  1. Top of page
  2. ABSTRACT:
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments
  8. REFERENCES CITED
  • Bolling, B.G., J.H. Kennedy, and E.G. Zimmerman. 2005. Seasonal dynamics of four potential West Nile vector species in north-central Texas. J. Vector Ecol. 30: 186194.
  • Bradley, C.A., S.E.J. Gibbs, and S. Altizer. 2008. Urban land use predicts West Nile virus exposure in songbirds. Ecol. Appl. 18: 10831092.
  • Burnham, K.P. and D.R. Anderson. 2002. Model Selection and Multimodal Inference: A Practical Information-Theoretic Approach, 2nd ed. Springer-Verlag, New York .
  • Calhoun, L.M., M. Avery, L. Jones, K. Gunarto, R. King, J. Roberts, and T.R. Burkot. 2007. Combined sewage overflows (CSO) are major urban breeding sites for Culex quinquefasciatus in Atlanta, Georgia. Am. J. Trop. Med. Hyg. 77: 478484.
  • Carpenter, S.J. and W.J. LaCasse. 1955. Mosquitoes of North America (North of Mexico). University of California Press, Berkeley , CA .
  • Conner, L.M., J.C. Rutledge, and L.L. Smith. 2010. Effects of mesopredators on nest survival of shrub-nesting songbirds. J. Wildl. Manage. 74: 7380.
  • Cupp, E.W., H.K. Hassan, X. Yue, W.K. Oldland, B.M. Lilley, and T.R. Unnasch. 2007. West Nile virus infection in mosquitoes in the Mid-South USA, 2002–2005. J. Med. Entomol. 44: 117125.
  • Darsie Jr., R.F. and C.D. Morris. 2003. Keys to the Adult Females and Fourth Instar Larvae of the Mosquitoes of Florida (Diptera, Culicidae). E.O. Painter Printing Company, DeLeon Springs .
  • Darsie Jr., R.F. and R.A. Ward. 2005. Identification and Geographical Distribution of the Mosquitoes of North America, North of Mexico, 2nd ed. University Press of Florida, Gainesville .
  • DeGroote, J., D.R. Mercer, J. Fisher, and R. Sugumaran. 2007. Spatiotemporal investigation of adult mosquito (Diptera: Culicidae) populations in an eastern Iowa county, USA. J. Med. Entomol. 44: 11391150.
  • Dufrene, M. and P. Legendre. 1997. Species assemblages and indicator species: the need for a flexible asymmetrical approach. Ecol. Monogr. 67: 345366.
  • Georgia Department of Community Health. 2010. Mosquito-borne Viral Diseases Web Page. http://health.state.ga.us/epi/vbd/mosquito.asp.
  • Gibbs, S.E.J., M.C. Wimberly, M. Madden, J. Masour, M.J. Yabsley, and D.E. Stallknecht. 2006. Factors affecting the geographic distribution of West Nile virus in Georgia, USA: 2002–2004. Vector-borne Zoonot. Dis. 6: 7382.
  • Goddard, J., W.C. Varnado, and B.A. Harrison. 2006. Notes on the ecology of Culex coronator Dyar and Knab, in Mississippi. J. Am. Mosq. Contr. Assoc. 22: 622625.
  • Godsey, M.S., M.S. Blackmore, N.A. Panella, K. Burkhalter, K. Gottfried, L.A. Halsey, R. Rutledge, S.A. Langevin, R. Gates, K.M. Lamonte et al. 2005. West Nile epizootiology in the southeastern United States, 2001. Vector-Borne Zoonot. Dis. 5: 8289.
  • Gray, E.W., B.A. Harrison, M.L. Womack, J. Kerce, C.J. Neely, and R. Noblet. 2005. Ochlerotatus japonicus japonicus (Theobald) in Georgia and North Carolina. J. Am. Mosq. Contr. Assoc. 21: 144146.
  • Gray, K.M., N. Burkett-Cadena, and M.D. Eubanks. 2008. Distribution expansion of Culex coronator in Alabama. J. Am. Mosq. Contr. Assoc. 24: 585587.
  • Hammon, M.D. and W.C. Reeves. 1943. Laboratory transmission of St. Louis encephalitis virus by three genera of mosquitoes. J. Exp. Med. 78: 241253.
  • Hylton, A.R. 1967. Low humidity water-retention ability in Eretmapodites chrysogaster and Aedes albopictus. J. Insect Physiol. 13: 153157.
  • Kelly, R., D. Mead, and B.A. Harrison. 2008. Discovery of Culex coronator Dyar and Knab (Diptera: Culicidae) in Georgia. Proc. Entomol. Soc. Wash. 110: 258260.
  • King, W.V., G.H. Bradley, C.N. Smith, and W.C. McDuffie. 1960. A Handbook of the Mosquitoes of the Southeastern United States. Agriculture Handbook No. 173. United States Department of Agriculture, Washington D.C .
  • Kramer, L.D., S.B. Presser, E.J. Houk, and J.L Hardy. 1990. Effect of the anesthetizing agent triethylamine on Western Equine Encephalomyelitis and St. Louis Encephalitis viral titers in mosquitoes (Diptera: Culicidae). J. Med. Entomol. 27: 10081010.
  • Lance-Parker, S., C. Rebmann, S. Kramer, and R. Kelly. 2002. Arboviruses in Georgia. Emergence of West Nile virus. J. Med. Assoc. Ga. 91: 3234.
  • Mackay, A.J., A. Roy, M.M. Yates, and L.D. Foil. 2008. West Nile virus detection in mosquitoes in East Baton Rouge Parish, Louisiana, from November, 2002, to October, 2004. J. Am. Mosq. Contr. Assoc. 24: 2835.
  • McCune, B., J.B. Grace, and D.L. Urban. 2002. Analysis of Ecological Communities. MjM Software Design, Gleneden Beach , Oregon .
  • Michener, W.K., E.R. Blood, J. Brim Box, C.A. Couch, S.W. Golladay, D.J. Hippe, R.J. Mitchell, and B.J. Paulk. 1998. Tropical storm flooding of a Coastal Plain landscape: Extensive floodplains ameliorated potential adverse effects on water quality, fishes, and molluskan communities. BioScience 48: 696705.
  • Miles, A.C., S.B. Castleberry, D.A. Miller, and L.M. Conner. 2006. Multi-scale roost-site selection by evening bats on pine-dominated landscapes in southwest Georgia. J. Wildl. Manag. 70: 11911199.
  • Miller, D.A. and L.M. Conner. 2007. Habitat selection of female turkeys in a managed pine landscape in Mississippi. J. Wildl. Manage. 71: 744751.
  • Moulis, R.A., J.D. Russell, H.B. Lewandowski, P.S. Thompson, and J.L. Heusel. 2008. Culex coronator in coastal Georgia and South Carolina. J. Am. Mosq. Contr. Assoc. 24: 588590.
  • Obenauer, P.J., P.E. Kaufman, S.A. Allan, and D.L. Kline. 2009. Host-seeking height preferences of Aedes albopictus (Diptera: Culicidae) in north central Florida suburban and sylvatic locales. J. Med. Entomol. 46: 900908.
  • Paradise, C.J. 2004. Relationship of water and leaf litter variability to insects inhabiting treeholes. J. N. Am. Benthol. Soc 23: 793805.
  • Reiter, M.E. and D.A. Lapointe. 2007. Landscape factors influencing the spatial distribution and abundance of mosquito vector Culex quinquefasciatus (Diptera: Culicidae) in a mixed residential-agricultural community in Hawai'i. J. Med. Entomol. 44: 861868.
  • Rochlin, I., M.E. Dempsey, S.R. Campbell, and D.V. Ninivaggi. 2008a. Salt marsh as Culex salinarius larval habitat in coastal New York. J. Am. Mosq. Contr. Assoc. 24: 359367.
  • Rochlin, I., K. Harding, H. S. Ginsberg, and S. R. Campbell. 2008b. Comparative analysis of distribution and abundance of West Nile and eastern equine encephalomyelitis virus vectors in Suffolk County, New York, using human population density and land use/cover data. J. Med. Entomol. 45: 563571.
  • Schäfer, M.L., J.O. Lundström, and E. Petersson. 2008. Comparison of mosquito (Diptera: Culicidae) populations by wetland type and year in the lower River Dalälven region, Central Sweden. J. Vector Ecol. 33: 150157.
  • Shaman, J., J.F. Day, and M. Stieglitz. 2003. St. Louis Encephalitis virus in wild birds during the 1990 south Florida epidemic: The importance of drought, wetting conditions, and the emergence of Culex nigripalpus (Diptera: Culicidae) to arboviral amplification and transmission. J. Med. Entomol. 40: 547554.
  • Shone, S.M., F.C. Curriero, C.R. Lesser, and G.E. Glass. 2006. Characterizing population dynamics of Aedes sollicitans (Diptera: Culicidae) using meteorological data. J. Med. Entomol. 43: 393402.
  • Silver, J.B. 2008. Mosquito Ecology: Field Sampling Methods 3rd edition. Springer, Dordrecht , The Netherlands .
  • Smith, J.P. and T.G. Floore. 2001. New mosquito collection records for southern Georgia. J. Entomol. Sci. 36: 114121.
  • Smith, L.L., D.A. Steen, J.M. Stober, M.C. Freeman, S.W. Golladay, L.M. Conner, and J. Cochrane. 2006. The vertebrate fauna of Ichauway, Baker County, GA. Southeast. Nat. 5: 599620.
  • Turell, M.J., D.J. Dohm, M.R. Sardelis, M.L. O'Guinn, T.G. Andreadis, and J.A. Blow. 2005. An update on the potential of North American mosquitoes (Diptera: Culicidae) to transmit West Nile virus. J. Med. Entomol. 42: 5762.