Climatic and landscape correlates for potential West Nile virus mosquito vectors in the Seattle region

Transcription

1 22 Journal of Vector Ecology June 2007 Climatic and landscape correlates for potential West Nile virus mosquito vectors in the Seattle region Heidi L. Pecoraro 1*, Heather L. Day 2, Robert Reineke 1, Nathan Stevens 2, John C. Withey 1, John M. Marzluff 1, and J. Scott Meschke 2 1 Division of Environmental Science and Resource Management, College of Forest Resources, University of Washington, Seattle, WA 98104, U.S.A. 2 Department of Environmental and Occupational Health Sciences, School of Public Health and Community Medicine 2, University of Washington, Seattle, WA 98104, U.S.A. Received 7 June 2006; Accepted 2 February 2007 ABSTRACT: Climatic and landscape patterns have been associated with both relative mosquito abundance and transmission of mosquito-borne illnesses in many parts of the world, especially warm and tropical climes. To determine if temperature, precipitation, or degree of urbanization were similarly important in the number of potential mosquito vectors for West Nile virus in the moderately temperate climate of western Washington, mosquitoes were collected using CDC carbondioxide/light traps set throughout the Seattle region during the summers of 2003 and The type and abundance of recovered species were compared to ecological correlates. Temperature and mosquito abundance were positively correlated, while precipitation was not strongly correlated with numbers of mosquitoes. Potential WNV mosquito vectors were most abundant in urban and suburban sites, including sites near communal roosts of American crows (Corvus brachyrhynchos). Exurban sites had the greatest vector species diversity, and Culex pipiens was the most abundant species throughout the region. Journal of Vector Ecology 32 (1): Keyword Index: West Nile virus, Culex, mosquito abundance, climatic correlates. INTRODUCTION The emergence of West Nile virus (WNV) in 1999 on the east coast of the United States and its subsequent westward dispersal throughout North America has renewed interest in mosquito vectors. The definitive hosts of WNV are birds, especially passeriforms (Komar et al. 2003), while human and other mammals are considered incidental hosts. Several species of mosquitoes have been reported as competent vectors of WNV (Sardelis et al. 2001, Turell et al. 2001, Godard et al. 2002, Turell et al. 2005), and much has recently been reported regarding the ecology of WNV transmission in the northeastern United States (Gingrich et al. 2006, Anderson et al. 2006, Diuk-Wasser et al. 2006). Despite these reports, however, the relationships between avian host, mosquito vectors, and environmental factors, which have resulted in very limited WNV transmission in western Washington, remain unclear. Even though thousands of human, mosquito, bird, and veterinary cases of WNV have been reported throughout the contiguous United States (USGS 2002, 2005), western Washington has been nearly exempt from WNV activity. In samples submitted to the Washington State Department of Health during 2005, only one Black-billed magpie (Pica hudsonia) of 660 birds, one of 32 horses (Equus caballus), and two Culex pipiens of 915 mosquito pools, all from south-central Washington, tested positive for WNV (USGS 2002, 2005, WSDOH 2005, 2006). Not until the summer of 2006 was the first human case of WNV reported in western Washington and were three dead American crows (Corvus brachyrhynchos) found infected with WNV near Seattle (USGS 2006). The continued low incidence of WNV has surprised health officials because of previously detected WNV amidst potential avian hosts and the presence of potential WNV vector species throughout the state (Duffy 2003). Previous field studies conducted in warm tropical climes, as found in Africa, New Zealand, and Australia, have determined associations between climatic and landscape patterns with both relative mosquito abundance and transmission of mosquito-borne illnesses (Reisen et al. 1993, Dhileepan 1996, Koenraadt et al. 2004, Afrane et al. 2005, Reisen et al. 2005, Hu et al. 2006, Leisnham et al. 2006). Results of several recent studies investigating similar possible correlations for WNV in the United States have varied depending on geographical region. A countyby-county assessment of WNV in Colorado, Nebraska, Louisiana, and Pennsylvania showed high temperature and agricultural land use associated with WNV infection (Miramontes et al. 2006). Urbanization was positively correlated to WNV risk in Georgia (Gibbs et al. 2006). In Texas, the abundance of WNV vector species was correlated with temperature, precipitation, and canopy cover (Bolling et al. 2005). Precipitation, rather than temperature, was linked with general arborvirus activity in a Rhode Island study (Takeda et al. 2003), while in Florida, spring drought followed by rain was specifically associated with incidence of WNV (Shaman et al. 2005).

2 Vol. 32, no. 1 Journal of Vector Ecology 23 Two recent studies have reported occurrence and distribution of specific mosquito species in Washington (Sames et al. 2004, Sames and Pehling 2005). However, neither of these studies, nor a recent Washington State Department of Health report detailing mosquito pool surveillance for WNV infection (Duffy 2003), has examined the ecological relationship between avian hosts of WNV, potential mosquito vectors, and ecological factors in western Washington. To determine if abundance of previously identified WNV vectors is related to climate or landscape in Seattle, we compared mosquito abundance to temperature and precipitation and to urban, suburban, and exurban landscapes during the summers of 2003 and MATERIALS AND METHODS Study area The city of Seattle contains 217 km 2 mixed urban, suburban, and exurban landscapes and 152 km 2 water, and is part of a large metropolitan region consisting of King, Snohomish, Pierce, and Kitsap counties. Temperatures peak late July to early August, with a maximum average of 22.5 C in July through September since Since 2002, when WNV was first reported on the west coast, monthly average temperatures in Seattle from June through September have not climbed above 20 C. Precipitation is low throughout the summer months, averaging 11.7 cm total for June through September during the last decade. Active surveillance for Figure 1. Trapping sites ( ), 1 km buffers (shaded circles around trapping sites), and American crow roosts ( ) in the Seattle region. Landscape class is represented by the light to dark gradient: white (water), light gray (urban), gray (mixed urban), and dark gray (exurban). WNV has been underway throughout the entire area since 2002, including testing of humans and horses presenting with symptoms consistent with WNV, dead birds, and mosquito pools. The study area is home to several confirmed WNV hosts, including an abundant population of American crows, numbering over 12,000 with four known large roosts in the Seattle area (Audubon Society 2006). Other confirmed WNV hosts abundant in Seattle include European starlings (Sturnus vulgaris), American robins (Turdus migratorius), house finches (Carpodacus mexicanus), and rock pigeons (Columba livia) (Audubon Society 2006, CDC 2006, Molaei et al. 2006). Mosquito collection Adult mosquitoes were collected from late June through early October in 2003 and 2004 using CDC carbon-dioxide/ light-baited traps as outlined in the CDC surveillance guidelines (CDC 2003). Traps were set for a total of 187 trap nights in 2003 and 203 trap nights in 2004 among 26 sites within 600 km 2, encompassing the city of Seattle and its borders (Figure 1). One trap night consisted of h to account for the various temporal feeding patterns of female mosquitoes. One or two traps were set at each site one to five times weekly during the season in which mosquitoes are most active (June through October) and were hung 1-3 m off the ground in trees to accommodate for both human and bird-biting mosquitoes. Site selection criteria generally included presence of water (in ponds, gardens, and other stagnant pools) where mosquito larvae were observed, and a place to hang the trap within a few meters of the water source. Five additional sites were selected without apparent standing water within a 0.8 km radius. Trapped mosquitoes were immobilized in the field using remaining dry ice from the CDC light traps in a cooler or by placing trap nets in a -20 C cooler upon return to the laboratory. Mosquitoes were then microscopically identified to species using scale patterns and other anatomical features. Larvae were also collected using the simple scoop method (O Malley 1995) from most sites and matured in rearing chambers before being submitted for identification. Landscape analysis Using ArcView software, land cover was determined from a 30 m resolution LANDSAT Thematic Mapper image that was classified at a resolution of 90 m, and generally ground-truthed for accuracy (Alberti et al. 2004). Landscape categories were defined according to the guidelines developed by Marzluff et al. (2001) and were calculated at a scale of 1 km 2. Urban landscapes contained >70% urban and <30% mixed urban-forest land cover and contained more than ten single family, multi-family, or commercial/ industrial dwellings per hectare. Suburban landscapes were covered by <50% urban, >35% mixed urban-forest, and <30% forest, and contained primarily single family dwellings per hectare. Exurban landscapes contained more than 30% forest land cover, and less than 2.5 single family dwellings per hectare. Of the 26 trapping sites, four were exurban, four urban, and 18 suburban. For the four

3 24 Journal of Vector Ecology June 2007 Table 1. Average catch per trap night for potential West Nile vector species in 2003 and Species Ur (38 trap 2003 Catch/trap night* Su (114 trap Ex (35 trap 2003 Total Catch/trap night SE Ae. vexans 0.01 (1) (1) An. punctipennis 0.71 (25) 0.08 (25) 0.03 Cq. perturbans 0.07 (8) 0.71 (25) 0.13 (33) 0.05 Cx. pipiens (681) (2,214) 1.11 (39) 8.96 (2,934) 1.82 Cx. tarsalis 0.31 (35) 2.00 (70) 0.33 (105) 0.10 Cs. inornata 0.04 (5) 0.11 (4) 0.03 (9) 0.01 Total (681) (2,263) 4.65 (163) (3,107) 1.93 * Ur Urban, Su Suburban, Ex Exurban. Ur (32 trap 2004 Catch/trap night* Su (171 trap Ex (0 trap 0.02 (3) 0.36 (62) (2,601) (9,118) 0.09 (3) 1.71 (293) 0.06 (2) 0.26 (44) (2,606) (9,520) 2004 Total Catch/trap night 0.01 (3) 0.24 (62) 42.7 (11,719) 1.10 (296) 0.16 (46) 44.3 (12,126) SE 2003/2004 Catch/trap night t (14)

4 Vol. 32, no. 1 Journal of Vector Ecology 25 Table 2: Partial correlations between temperature and vector abundance holding precipitation constant and between precipitation and vector abundance holding temperature constant. Species Temperature and vector abundance American crow roosts, which were located in suburban areas, trapping sites were examined within a 1 km radius of each roost to accommodate Cx. pipiens potential nightly dispersal. Climate data Temperature and precipitation data collected at the Seattle-Tacoma International Airport National Climate Data Center s weather station (22 km south of downtown Seattle; 8 km outside the sampling area) were used for the study (NCDC 2005). Weekly minima and maxima temperatures were averaged from June through October for a total of 15 weeks each year. Total precipitation for each of the 15 weeks was considered. Monthly temperature averages and total precipitation for the summer months were compared to decade trends. Data analyses Paired t-tests were used to evaluate significant differences between 2003 and 2004 temperatures, precipitation, and vector abundance, as well as between temperatures and precipitation averages and decade averages. A linear regression method similar to that described in Takeda et al. (2003) was used to examine independent relationships between average mosquito catch per trap night and weekly average precipitation or temperature. Multiple regression analysis was also performed to examine correlations between mosquito abundance and both temperature and precipitation. All statistical tests were performed using SPSS for Macintosh (SPSS Inc., Chicago, IL 2004). RESULTS Precipitation and vector abundance Ae. vexans (P = 0.19) (P = 0.70) An. punctipennis (P = 0.14) (P = 0.84) Cq. perturbans (P = 0.04) (P = 0.95) Cx. pipiens (P < 0.01) (P = 0.36) Cx. tarsalis (P = 0.04) (P = 0.81) Cs. inornata (P = 0.01) (P = 0.04) Vector abundance Of the nine WNV potential vector species previously documented to be present in the Seattle region - Aedes cinereus, Aedes vexans, Anopheles punctipennis, Coquillettidia perturbans, Culex pipiens, Culex tarsalis, Culiseta inornata, Ochlerotatus canadensis, and Ochlerotatus japonicus japonicus (Goddard et al. 2002, Duffy 2003, Andreadis et al. 2004, Turell et al. 2005, Anderson et al. 2006) - Cx. pipiens was by far the most abundant species in both years, followed by Cx. tarsalis and Cq. perturbans (Figure 2). Identification of reared larvae suggests that Cx. pipens and Cx. tarsalis were also the most prevalent species found in ponds and stagnant pools in the study area (data not shown). Only a few adult An. punctipennis were caught in 2003 and Ae. vexans was least abundant both years. Ae. cinereus, Oc. canadensis, and Oc. japonicus japonicus were not caught either year. All trapped species peaked in late July-early August, except Cq. perturbans, which tended to peak in early July. More total mosquitoes were caught per trap night in 2004 than in 2003 (Table 1). Relative abundance was higher for all species in 2004 with significant increases in numbers of Cx. pipiens and Cs. inornata trapped per week. Only An. punctipennis decreased in 2004 relative to Landscape trends Mosquito community composition varied among the three different landscape classes (Table 1). Exurban sites contained the potential vector species Cs. inornata, Cx. tarsalis, Cx. pipiens, Cq. perturbans, and An. punctipennis. Suburban sites had the potential vector species Cx. tarsalis and Cx. pipiens and non-vector species Cs. incidens. Traps in urban sites contained only Cx. pipiens and Cs. incidens. The proportion of Cx. pipiens at urban sites was nearly five times that seen at exurban sites, resulting in urban areas having the highest total number of mosquitoes caught per trap night (Table 1). Species trapped in very low numbers (<35 total) included Ae. vexans, An. punctipennis, Cx. territans, Cs. impatiens, Cs. minnesotae, Cs. particeps, Oc. aborginis, Oc. fitchii, Oc. incepitus, and Oc. sierrensis. Crow roost trends Species compositions in traps set near known American crow roosts were similar to other suburban sites. Brier, Foster Island, and Tukwila roosts overlapped with one or more 1 km buffers set around trapping sites (Figure 1). Cx. pipiens was the primary species (~90%) present at all roosts, followed by Cx. tarsalis (3%). The three Ae. vexans mosquitoes trapped in 2004 were near the Foster Island roost. Temperature trends Weekly temperature maxima for the 15 weeks from July through September were not significantly different between 2003 and 2004 (2003 mean 23.9 C, 2004 mean 22.9 C, t (14) = 1.21, P = 0.25). However, compared to the decade trend of 22.5 C, the monthly maxima temperatures July through September were significantly warmer in 2003 (t (3) = 5.67, P = 0.01). Temperature averages per week did not reach above 30 C, although daily temperatures occasionally peaked above 30 C. Daily temperature minima fluctuated between 10 and 20 C. Total mosquito abundance increased as temperatures rose in late July and generally decreased at the end of August as temperatures declined (Figure 2a-c). Cx. pipiens, Cx. tarsalis, and Cs. inornata abundance was highest during temperature peaks. Mosquito abundance was significantly correlated with temperature during both years of trapping (r 2 (Adj) in 2003 = 0.32, P = 0.01; r2 (Adj) in 2004 = 0.67, P < 1).

5 26 Journal of Vector Ecology June 2007 a b c Figure 2. Relative abundance of West Nile vector mosquito species : (a) Cx. pipiens; (b) Cx. tarsalis; (c) Ae. vexans, An. punctipennis, Cq. perturbans, and Cs. inornata. Note differences in range of each y-axis. Temperature ( C) is represented in 2a and 2b as shaded regions. Precipitation trends There was significantly more weekly precipitation from July through September in 2004 compared to 2003 (2003 mean 0.05 cm, 2004 mean 0.20 cm, t (14) = -2.35, P = 0.03), especially in August with a total monthly rainfall of 7.66 cm in 2004 versus 0.84 cm in Total summer (June-Sept) precipitation in 2003 was significantly lower than the decade trend of 11.7 cm (2003 total 4.65 cm, t (3) = -6.11, P = 0.01). Average precipitation was near zero during temperature peaks and was highest when mosquito catch was lowest. Precipitation was not strongly correlated for mosquitoes in either year (r 2 in 2003 < 0.01, P = 0.39; (Adj) r2 in 2004 = (Adj) 0.03, P = 0.26). Even when the lag time (~ten days) between a rainfall event and its subsequent hatch was adjusted for, there was no significant correlation between mosquito abundance and precipitation (r 2 in 2003 = 0.11, P = 0.12; (Adj) in 2004 = -0.01, P = 0.36). r 2 (Adj) Multiple regression analysis In multiple regression analyses for vector species Cq. perturbans, Cx. pipiens, and Cx. tarsalis, the partial correlations between temperature and vector abundance (r 0.36; P 0.04) were much stronger than those between precipitation and vector abundance (r 0.17; P 0.36). While partial correlations between temperature and vector abundance were still stronger than those for precipitation and abundance of the potential vectors Ae. vexans, An. punctipennis, and Cs. inornata, temperature correlations were not significant for Ae. vexans and An. punctipennis. However, the precipitation correlation was significant for Cs. inornata (Table 2). The sample sizes for Ae. vexans, An. punctipennis, and Cs. inornata were smaller than for Cq. perturbans, Cx. pipiens, and Cx. tarsalis (Figure 2c). DISCUSSION This study has demonstrated that potential WNV vectors are present in all landscape classes throughout the Seattle region, and their numbers increase with temperature. This is the first study showing definitively that a correlation exists between the abundance of several potential mosquito vector species and temperature in western Washington s mild climate. Though admittedly the CDC light traps used in this study are known to demonstrate collection biases for certain species and may be placement biased (Mboera et al. 2000, Burkett et al. 2001), results of larval sampling support the identification of Cx. pipiens and Cx. tarsalis as the dominant species trapped in the study area. Cx. pipiens, a potential bridge vector (Fonesca et al. 2004, Kilpatrick et al. 2005), was found at high abundances and increased with rising temperatures in an urban area with high densities of both humans and several of the native bird species listed in the CDC s WNV bird mortality database (CDC 2006). The American crow, in particular, has suffered WNV mortality rates as high as 68% in a tracked cohort and 100% in controlled experiments (McLean et al. 2001, Komar et al. 2003, Yaremych et al. 2004). In Champaign and Urbana, IL, where large numbers of American crows have died from West Nile virus, Culex spp. accounted for most of the mosquitoes caught at night as well as near diurnal crow roosts (Yaremych et al. 2004; Sarah Yaremych, personal communication). Similarly, WNV-infected Cx. pipiens mosquitoes were found in Los Angeles county the same year 38 birds there died from the disease (Reisen et al. 2004). American robins, house finches, and house sparrows (Passer domesticus) are also abundant in Seattle (Audubon Society 2006), and may play an even greater role in WNV prevalence as the preferred feeding host of Cx. pipiens (Molaei et al. 2006, Reisen et al. 2006). Monthly summer temperatures during both 2003 and 2004 were nearly the same; however, precipitation was ~7 cm below the decade mean in 2003 and ~6 cm above the decade mean in High populations of Cx. pipiens are typically associated with a early season heavy rainfall and high summer temperatures, followed by only light rainfall, which provides more breeding habitat in stagnant pools of water in tires, tree holes, other containers, ponds, etc. (Robert McLean, personal communication). Thus, despite having no overall correlation with mosquito abundance in our study, precipitation may affect vector species abundance on a more local scale (e.g., at individual sites) and for individual species such as Cs. inornata, as shown by the multiple regression analysis. The effects of site-specific microclimate should be evaluated in future studies. Additional studies to examine

6 Vol. 32, no. 1 Journal of Vector Ecology 27 temporal and other environmental condition biases (e.g., wind dispersal) also warrant further investigation for their potential impact on mosquito vectors and the possibility of WNV transmission. Overall, our study shows the abundance of potential WNV vectors was positively correlated with temperature in the Seattle metropolitan region, and that there was no association between mosquitoes and precipitation patterns during the two years of trapping. Potential WNV vector species abundance also increased with degree of urbanization, although species diversity was greatest in areas with the least land development. If Seattle temperatures warm in response to global climate change, further understanding of the ecological determinants of WNV transmission, or lack thereof, will be needed to better inform management strategies and develop predictive models of disease emergence. Acknowledgments Mosquito identification was conducted in collaboration with the Washington State Department of Health Zoonotic Disease Program (CDC WNV grant), which also provided the CDC traps. Special thanks to Dr. Robert McLean from the National Wildlife Research Center for information on mosquito emergence-weather patterns in Colorado and Sarah Yaremych-Hamer of the University of Michigan for mosquito abundance data at crow death sites in east-central Illinois. This work was partially supported by NSF grants DEB and (JMM, RR, JCW). REFERENCES CITED Afrane, Y.A., B.W. Lawson, A.K. Githeko, and G.Y. Yan Effects of microclimatic changes caused by land use and land cover on duration of gonotrophic cycles of Anopheles gambiae (Diptera: Culicidae) in western Kenya highlands. J. Med. Entomol. 42: Alberti, M., R. Weeks, and S. Coe Urban land cover change analysis for the central Puget Sound: J. Photogramm. Engineer. Remote Sens. 70: Anderson, J.F., T.G. Andreadis, A.J. Main, F.J. Ferrandino, and C.R. Vossbrinck West Nile virus from female and male mosquitoes (Diptera: Culicidae) in subterranean, ground, and canopy habitats in Connecticut. J. Med. Entomol. 43: Andreadis, T.G., J.F. Anderson, C.R. Vossbrinck, and A.J. Main Epidemiology of West Nile virus in Connecticut: a five-year analysis of mosquito data Vector Borne Zoon. Dis. 4: Audubon Society th Annual Christmas Bird Count. table.html. Bolling, B.G., J.H. Kennedy, and E.G. Zimmerman Seasonal dynamics of four potential West Nile species in north-central Texas. J. Vector Ecol. 30: Burkett D.A., W.J. Lee, K.W. Lee, H.C. Kim, H.I. Lee, J.S. Lee, E.H. Shin, R.A. Wirtz, H.W. Cho, D.M. Claborn, R.E. Coleman, and R.A. Klein Light, carbon dioxide, and octenol-baited mosquito trap and host-seeking activity evaluations for mosquitoes in a malaria area of the Republic of Korea. J. Am. Mosq. Contr. Assoc. 17: Centers for Disease Control and Prevention Epidemic/epizootic West Nile virus in the United States: guidelines for surveillance, prevention, and control. U.S. Department of Health and Human Services, Fort Collins, CO. pp Centers for Disease Control and Prevention Avian mortality database 1999-present. ncidod/dvbid/westnile/birdspecies.htm. Dhileepan, K Mosquito seasonality and arboviral disease incidence in Murray Valley, southeast Australia. Med. Vet. Entomol. 10: Diuk-Wasser, M.A., H.E. Brown, T.G. Andreadis, and D. Fish Modeling the spatial distribution of mosquito vectors for West Nile virus in Connecticut, USA. Vector Borne Zoon. Dis. 6: Duffy, A West Nile virus nonhuman surveillance in Washington State. Washington State Department of Health: Olympia, WA. Fonseca, D.M., N. Keyghobadi, C.A. Malcolm, C. Mehmet, F. Schaffner, M. Mogi, R.C. Fleischer, and R.C. Wilkerson Emerging vectors in the Culex pipiens complex. Science 303: Gibbs, S.E., M.C. Wimberly, M. Madden, J. Masour, M.J. Yabsley, and D.E. Stallknecht Factors affecting the geographic distribution of West Nile virus in Georgia, USA: Vector Borne Zoon. Dis. 6: Gingrich, J.B., R.D. Anderson, G.M. Williams, L. O Connor, and K. Harkins Stormwater ponds, construction wetlands, and other best management practices as potential breeding sites for West Nile virus vectors in Delaware during J. Am. Mosq. Contr. Assoc. 22: Goddard, L.B., A.E. Roth, W.K. Reisen, and T.W. Scott Vector competence of California mosquitoes for West Nile virus. Emerg. Infect. Dis. 8: Hu, W.B., S.L. Tong, K. Mengersen, and B. Oldenburg Rainfall, mosquito density and the transmission of Ross River virus: A time-series forecasting model. Ecol. Modeling 196: Kilpatrick, A.M., L.D. Kramer, S.R. Campbell, E.O. Alleyne, A.P. Dobson, and P. Daszak West Nile virus risk assessment and the bridge vector paradigm. Emerg. Infect. Dis. 11: Koenraadt, C.J.M., A.K. Githeko, and W. Takken The effects of rainfall and evapotranspiration on the temporal dynamics of Anopheles gambiae s.s. and Anopheles arabiensis in a Kenyan village. Acta Trop. 90: Komar, N., S. Langevin, S. Hinten, N. Nemeth, E. Edwards, D. Hettler, B. Davis, R. Bowen, and M. Bunning Experimental infection of North American birds with the New York 1999 strain of West Nile virus. Emerg.

7 28 Journal of Vector Ecology June 2007 Infect. Dis. 9: Leisnham, P.T., P.J. Lester, D.P. Slaney, and P. Weinstein Relationships between mosquito densities in artificial container habitats, land use and temperature in the Kapiti-Horowhenua region, New Zealand. New Zealand J. Marine Freshwater Res. 40: Marzluff, J.M., R. Bowman, and R. Donnelly A historical perspective on urban bird research: trends, terms, and approaches. In: J.M. Marzluff, R. Bowman, and R. Donnelly, (eds). Avian ecology and conservation in an urbanizing world. Kluwer Academic Publishers: Norwell, MA, pp Mboera, L.E.G., B.G.J. Knols, M.A.H. Braks, and W. Takken Comparison of carbon dioxide-baited trapping systems for sampling outdoor mosquito populations in Tanzania. Med. Vet. Entomol. 14: McLean, R.G., S.R. Ubico, D.E. Docherty, W.R. Hansen, L. Sileo, and T.S. McNamara West Nile virus transmission and ecology in birds. Ann. N.Y. Acad. Sci. 951: Miramontes, R., W.E. Lafferty, B.K. Lind, and M.W. Oberie Is agricultural activity linked to the incidence of human West Nile virus? Am. J. Prev. Med. 30: Molaei, G., T.G. Andreadis, P.M. Armstrong, J.F. Anderson, and C.R. Vossbrinck Host feeding patterns of Culex mosquitoes and West Nile virus transmission, northeastern United States. Emerg. Infect. Dis. 12: National Climatic Data Center Annual climatological summary. O Malley, C Seven ways to a successful dipping career. Wing Beats 6: Reisen, W.K., R.P. Meyer, S.B. Presser, and J.L. Hardy Effect of temperature on the transmission of Western Equine Encephalomyelitis and St-Louis Encephalitis viruses by Culex-tarsal (Diptera, Culicidae). J. Med. Entomol. 30: Reisen, W., H. Lothrop, R. Chiles, M. Madon, C. Cossen, L. Woods, S. Husted, V. Kramer, and J. Edman West Nile virus in California. Emerg. Infect. Dis. 10: Reisen, W.K., Y. Fang, and V.M. Martinez Avian host and mosquito (Diptera: Culicidae) vector competence determine the efficiency of West Nile and St. Louis encephalitis virus transmission. J. Med. Entomol. 42: Reisen, W.K., C.M. Barker, R. Carney, H.D. Lothrop, S.S. Wheeler, J.L. Wilson, M.B. Madon, R. Takahashi, B. Carroll, S. Garcia, Y. Fang, M. Shafii, N. Kahl, S. Ashtari, V. Kramer, C. Glaser, and C. Jean Role of corvids in epidemiology of West Nile virus in southern California. J. Med. Entomol. 43: Sames, W.J., W.E. Herman, D.A. Florin, and F.A. Maloney Distribution of Ochlerotatus togoi along the Pacific Coast of Washington. J. Am. Mosq. Contr. Assoc. 20: Sames, W.J. and D. Pehling Update on Ochlerotatus japonicus in the state of Washington. J. Am. Mosq. Contr. Assoc. 21: Sardelis, M.R., M.J. Turell, D.J. Dohm, and M.L. O Guinn Vector competence of selected North American Culex and Coquillettidia mosquitoes for West Nile virus. Emerg. Infect. Dis. 7: Shaman, J., J.F. Day, and M. Stieglitz Drought-induced amplification and epidemic transmission of West Nile virus in southern Florida. J. Med. Entomol. 42: Takeda, T., C.A. Whitehouse, M. Brewer, A.D. Gettman, and T.N. Mather Arbovirus surveillance in Rhode Island: assessing potential ecologic and climatic correlates. J. Am. Mosq. Contr. Assoc. 19: Turell, M.J., M.L. O Guinn, D.J. Dohm, and J.W. Jones Vector competence of North American mosquitoes (Diptera: Culicidae) for West Nile virus. J. Med. Entomol. 38: Turell, M.J., D.J. Dohm, M.R. Sardelis, M.L. Oguinn, T.G. Andreadis, and J.A. Blow An update on the potential of North American mosquitoes (Diptera: Culicidae) to transmit West Nile Virus. J. Med. Entomol. 42: United States Geological Society West Nile Virus - Washington cumulative bird map. usgs.gov/2002/. United States Geological Society West Nile virus maps. United States Geological Society West Nile virus maps. Washington State Department of Health Dead bird from Yakima County tests positive for West Nile virus. In: WSDOH News Release September 16, Washington State Department of Health: Olympia, WA. Washington State Department of Health West Nile Surveillance 2005 Testing Summary. Washington State Department of Health: Olympia WA, Appendix 6. Yaremych, S.A., R.E. Warner, P.C. Mankin, J.D. Brawn, A. Raim, and R. Novak West Nile virus and high death rate in American crows. Emerg. Infect. Dis. 10: