Journal of the American Mosquito Control Association, 26(1):57–66, 2010 Copyright E 2010 by The American Mosquito Control Association, Inc. EVALUATION OF EFFICACY AND HUMAN HEALTH RISK OF AERIAL ULTRA-LOW VOLUME APPLICATIONS OF PYRETHRINS AND PIPERONYL BUTOXIDE FOR ADULT MOSQUITO MANAGEMENT IN RESPONSE TO WEST NILE VIRUS ACTIVITY IN SACRAMENTO COUNTY, CALIFORNIA PAULA A. MACEDO,1 JEROME J. SCHLEIER III,2 MARCIA REED,1 KARA KELLEY,1 GARY W. GOODMAN,1 DAVID A. BROWN1 AND ROBERT K. D. PETERSON2 ABSTRACT. The Sacramento and Yolo Mosquito and Vector Control District (SYMVCD, also referred to as ‘‘the District’’) conducts surveillance and management of mosquitoes in Sacramento and Yolo counties in California. Following an increase in numbers and West Nile virus (WNV) infection rates of Culex tarsalis and Culex pipiens, the District decided on July 26, 2007, to conduct aerial applications of EvergreenH EC 60-6 (60% pyrethrins: 6% piperonyl butoxide) over approximately 215 km2 in the north area of Sacramento County on the nights of July 30, July 31, and August 1, 2007. At the same time, the District received notification of the first human WNV case in the area. To evaluate the efficacy of the applications in decreasing mosquito abundance and infection rates, we conducted pre- and post-trapping inside and outside the spray zone and assessed human health risks from exposure to the insecticide applications. Results showed a significant decrease in abundance of both Cx. tarsalis and Cx. pipiens, and in the minimum infection rate of Cx. tarsalis. Human-health risks from exposure to the insecticide were below thresholds set by the US Environmental Protection Agency. KEY WORDS West Nile virus, aerial spraying, insecticide, mosquito control, risk assessment INTRODUCTION intensive larviciding and public education efforts, the District intervened by aerially applying a West Nile virus (WNV, family Flaviridae, formulation of pyrethrins and piperonyl butoxide genus Flavivirus) was first detected in the United (PBO) over an urban/suburban area in Sacra- States in 1999 in New York City, and reached mento County (Elnaiem et al. 2008), which most California in the summer of 2003 (Reisen et al. likely interrupted the WNV transmission cycle 2004). In 2004, WNV amplified to epidemic levels (Carney et al. 2008). and dispersed to all 58 counties in the state, and Although traditionally used in response to was associated with low-level transmission to humans and horses in Sacramento and Yolo epidemics and as part of a sustainable public counties that year (Armijos et al. 2005, Hom et al. health program (Rose 2001), the application of 2005). In 2005, there was a severe outbreak in pesticides often generates public concerns and Sacramento County, with 177 human cases and controversy about the safety of these chemicals to 40 equine cases (Elnaiem et al. 2006). people and the environment as well as the efficacy The Sacramento and Yolo Mosquito and of such practice (Thier 2001, Roche 2002, Hodge Vector Control District (SYMVCD, also referred and O’Connell 2005). A human health risk to as ‘‘the District’’) conducts routine surveillance assessment conducted by Peterson et al. (2006) and management of mosquito populations in for truck-mounted ultra-low volume (ULV) Sacramento and Yolo counties. The District applications of adulticides commonly used in monitors weekly mosquito abundance and West mosquito management programs determined Nile, western equine encephalitis (WEE), and St. risks to be below levels established by the US Louis encephalitis (SLE) viral infection. The Environmental Protection Agency (USEPA), District follows the California Mosquito-Borne which agrees with the current scientific weight Virus Surveillance and Response Plan (Kramer of evidence (NYCDOH 2001, Karpati et al. 2004, 2005) and its own Mosquito and Mosquito-Borne Currier et al. 2005, O’Sullivan et al. 2005). The Disease Management Plan (SYMVCD 2005), and results from Peterson et al. (2006) indicated that applies the principles of integrated pest manage- potential health risks from WNV exceed risks ment (IPM) in its program. When WNV reached from exposure to these pesticides when used at epidemic levels in 2005 despite SYMVCD’s label rates to control adult mosquitoes. Their study used extremely conservative assumptions and estimated exposure after truck-mounted 1 Sacramento-Yolo Mosquito and Vector Control District, 8631 Bond Road, Elk Grove, CA 95757. ULV applications as a worse-case scenario, used 2 Department of Land Resources and Environmental application rates greater than the ones used by Sciences, Montana State University, 334 Leon Johnson SYMVCD, and therefore likely overestimated the Hall, Bozeman, MT 59717. exposure that would be seen for the application 57 58 JOURNAL OF THE AMERICAN MOSQUITO CONTROL ASSOCIATION VOL.26,NO.1 by the District. Schleier et al. (2009b) evaluated Mosquito-Borne Virus Surveillance and Re- probabilistically the deterministic risk estimates sponse Plan (Kramer 2005) and its Mosquito presented by Peterson et al. (2006) and found and Mosquito-Borne Disease Management Plan them to be very conservative. Davis et al. (2007) (SYMVCD 2005), the District made the decision evaluated ecological risks posed by adult mos- on July 26, 2007, to aerially apply pyrethrins and quito management programs and concluded that PBO (Evergreen EC-60-6H) over the area of risks to nontarget organisms from pesticides concern. On the same day that this decision was applied for adult mosquito management are low made, the District received notification of the first and not likely to exceed regulatory levels of human case in the area. The insecticide applica- concern. tions took place on the nights of July 30 and 31, Although vector control strategies and their and August 1, 2007. The objectives of this study effectiveness have generated concern in past were to 1) evaluate the efficacy of the aerial years, there have been few published studies applications in reducing adult mosquito popula- addressing the efficacy of these operations in tions and infection rates of the two main species reducing mosquito populations, infection rates, implicated in WNV transmission in Sacramento and virus transmission. Sacramento County County and 2) assess human health risks for the experienced a WNV epidemic for the first time aerial application of adulticide conducted in in 2005. Although the evaluation of the aerial Sacramento County in 2007 in response to the adulticiding conducted during that year suggested WNV activity. interruption of transmission (Carney et al. 2008) and reduction of vector abundance (Elnaiem et al. 2008), some uncertainties were identified by MATERIALS AND METHODS SYMVCD staff to be addressed in future Aerial applications and study area evaluations, particularly the absence of fixed 2 locations for trapping mosquitoes before and The spray zone was a 215 km area in northern after the applications, the small number of Sacramento County, located in the Central Valley mosquito pools collected from those areas, and of California (Fig. 1). The insecticide Evergreen other confounding factors such as the effect of EC 60-6 (60% PBO and 6% pyrethins; McLaugh- wind shadow at some of the locations. lin Gormley King, Golden Valley, MN) was In 2006, WNV reached epidemic levels in the applied by a fixed-wing Piper Aztec and a Cessna cities of Davis and Woodland in Yolo County, 402 aircraft (VDCI/ADAPCO Vector Control and the SYMVCD, with the collaboration of the Services, Greenville, MS) for three consecutive Center for Vectorborne Diseases at the University nights on July 30 and 31, and August 1, 2007. The of California–Davis, monitored abundance and application rate was 2.8 g/ha (0.0025 lb/ac) of infection rates and conducted aerial applications pyrethins and 28 g/ha (0.025 lb/ac) of PBO. The of pyrethrins and PBO on the nights of August 8 release altitude was 91 m (300 ft), the wind speed and 9, 2006 (Macedo et al. 2007a, Nielsen et al. ranged from 3.7 km/h to 18.5 km/h, and the 2007). Analysis of data from 2005 and 2006 temperature at the time of the applications ranged showed that the aerial applications could have from 34uCto36uC. Application start times ranged been more successful in interrupting virus trans- from 7:34 p.m. to 7:55 p.m. and application end mission to people if they had been conducted times ranged from 9:20 p.m. to 9:51 p.m. 1 wk or 2 wk before, when mosquito abundance and infection rates were higher (Macedo et al. Mosquito abundance 2008). The first indication of active WNV transmis- To evaluate the effect of the aerial applications sion in the District in 2007 was the detection of on mosquito abundance, the District used en- WNV in a dead American crow on May 31. The cephalitis virus surveillance traps (EVS) baited first positive mosquito pool was obtained on July with dry ice, herein referenced to as CO2 traps 4, 2007, in a pool of Culex pipiens L. West Nile (Rohe and Fall 1979), and gravid-female traps virus continued to amplify during the month of (Cummings 1992) to collect mosquitoes inside July, and mosquito abundance and maximum and outside of the aerial spray zone for 3 days likelihood estimates of minimum infection rates before and 3 days after the application events. continued to be monitored. The District identi- Trap collections were brought to the laboratory, fied an area of approximately 215 km2 in the where mosquitoes were anesthetized with trieth- north part of Sacramento County as higher risk ylamine, identified to species and counted, and and intensified all the aspects of its IPM program then frozen at 280uC for later virus testing. Three in an attempt to reduce mosquito populations. CO2 traps and one gravid trap were placed at On the week of July 24, 2007, infection rates had each of the 12 fixed sites in the aerial spray zone reached 10.85 and 7.87 per 1,000 mosquitoes for and six fixed sites in the untreated control zone Culex tarsalis Coquillett and Cx.