Chronology of a mosquito control effort after hurricane Fran.

Introduction

Ideally, mosquito abundance data should be expressed in terms of deviation from historical average values. In past disaster situations, landing counts in excess of 20 mosquitoes per minute have usually been considered necessary for mosquitoes to cause disruption of reconstruction efforts; disruption of community services (fire, police, ambulance, etc.); or intolerable stress in the resident human population (1). In 1996, New Hanover County (Wilmington, North Carolina) experienced its most intense mosquito season in years. As shown in Figure 1, rainfall amounts were 12.81 inches (32.5 centimeters) above average (2). July brought Bertha, a Category II hurricane, which primed mosquito hatch sites by increasing ground water levels and blocking drainage arteries with debris. August conditioned the mosquito eggs laid after Bertha. In September, water deposited by Fran, a Category III hurricane, acted as the catalyst for the increased mosquito populations. The purpose of this article is to describe the county's successful mosquito abatement procedure after Hurricane Fran.

After a hurricane, vector and nuisance mosquitoes assume increased public health importance. Rainfall associated with a storm may significantly increase mosquito populations and augment existing arbovirus cycles in the area (3). An elevated public health risk may necessitate emergency funding to minimize the disease vector potential. Federal assistance for emergency vector surveillance and control is available when a disaster is declared and when at least one of the following conditions is met:

1. Transmission of human or animal disease is in progress or is deemed imminent.

2. Large populations of nuisance species substantially hamper reconstruction efforts.

3. Normal functioning of communities in the disaster area is substantially disrupted.

4. Large nuisance populations place additional stress on the human population (4).

The Mosquito Control Program

New Hanover County (NHC) is a coastal community located in southeastern North Carolina with a population of 146,000 and an area of 185 square miles. Approximately 60 percent of the county has been developed. The county is bordered on the south and west by the Cape Fear River and on the east by the Intracoastal Waterway and the Atlantic Ocean. To the north, it is bordered by Pender County and the flood plain for the Northeast Cape Fear River.

NHC's mosquito program employs seven full-time employees and one temporary summer employee. The division includes a manager, a field biologist, a crew foreman, three vector control operators, and one general-service worker. The facilities equipment consists of nine trucks. Five are equipped with ultra-low-volume (ULV) generators for adulticide application, three are set up for larvicide application, and one is used for surveillance and complaint investigations. Other equipment includes a 16-foot aluminum boat, an all-terrain vehicle with attached granule spreader, two tractors, one dump truck, and a track hoe excavator.

Synthetic pyrethroid products are normally used for adulticide applications. Larvicides include an insect growth regulator, methoprene, in three formulations (liquid, granule, and briquette); Bacillus thuringiensis israelensis (BTI) in three formulations (liquid, granule, and donut); and Bacillus sphaericus (BS) in a granule formulation.

NHC's program has a strong foundation of historical documentation of mosquito breeding sites, a consistent adult mosquito surveillance program, and a public education program.

Larvicide Application

In NHC, mosquitoes are grouped by three distinct habitats based on where the female mosquito lays her eggs. These habitats are salt marsh sites, artificial containers, and freshwater sites that include woodland pools and flood plains. In the weeks following Hurricane Fran, 23 species of mosquitoes were identified from these habitats.

The county's historical information includes daily documentation of all larvicide applications. The methods employed are right-of-way applications, treatment of intermittently flooded sites, treatment of dredge spoil islands, and minnow transplants. Right-of-way larvicide applications treat roadside ditches in residential areas with liquid BTI. Intermittently flooded freshwater sites are monitored after rainfall to determine if treatment is warranted. Sites vary in size from 25 square feet to 4 acres. Dredge spoil islands are considered salt marsh sites; treatment sites vary in size from 5 square feet to 300 acres. Whenever possible, these sites are managed by biological means such as draining, flooding, and establishment of a native minnow population. Because access to the majority of these sites is difficult, an insect growth regulator (IGR) is applied when other biological means are not possible. Finally, in low areas where water can be found year round, the mosquito fish Gambusia affinis is introduced as a biological control. Regardless of habitat, all sites are dipped to determine larval densities and the amount of organic material suspended in the water column. Treatment rates vary depending on larval populations, water turbidity, rainfall amount, temperature, and time of year.

Larval dipping and product applications are documented on a daily worksheet. During the initial survey, a description of the breeding site and accurate directions to the site are recorded. Each site is identified on a master map in the office and added to the appropriate larvicide crew's site book.

Adult Mosquito Surveillance

Adult surveillance incorporates field observations (landing counts) from each of the crew's daily activities and from 12 mosquito light traps, which are baited with C[O.sub.2]. Light traps are deployed at stations around the county once a week. Trap counts for each station are totaled the following morning. When light-trap populations reach elevated levels, ULV generators are routed to reduce the adult numbers. Incorporated into the regular light-trap program are "floating traps." These traps are used to identify species from customer complaints. This species information can lead to the designation of new locations for larvicide application.

Public Education Program

NHC's public education program focuses on the container-breeding mosquito, Aedes albopictus. When breeding sites are located during a complaint investigation, the complainant is contacted and given information regarding Ae. albopictus. If the complainant is not home, a pamphlet is left at the door. During March, April, and May, classroom presentations are given to grades K through 8.

This regimen of larval mapping, weekly adult surveillance, and a proactive public education program prepared NHC's Vector Control program for the emergency presented by Hurricane Fran.

Results

Fran, a Category III hurricane, passed through New Hanover County during the evening hours of September 5, 1996. The hurricane brought sustained winds of 67 miles per hour (with reported gusts up to 115 miles per hour) and a storm surge of 12 to 13 feet. Previous mapping of mosquito breeding sites and an experienced crew were the key to NHC's rapid response to excessive mosquito populations after Hurricane Fran. Vector Control's hurricane response started September 6, 1996, and continued for 26 days, when mosquito populations were found to be well below 1995 light-trap levels. The response consisted of four phases:

* initial assessment,

* surveillance and larval controls,

* full-scale adult controls, and

* follow-up surveillance to evaluate control effectiveness.

Phase I - Initial Assessment (Days 1-3)

The vector control facilities and equipment were surveyed for damage and safety hazards. Ninety-five percent of NHC was without power and water; Vector Control was no exception. Clarke Mosquito Control in Chicago was contacted to initiate a contract for a helicopter with larvicide and adulticide application capabilities. A countywide evaluation of the potential for increased mosquito populations was initiated. In 24 hours, Fran had deposited almost 13.5 inches of rain onto NHC's flood plains. Uprooted trees clogged drainage arteries throughout the county, which increased inland flooding. Storm surge along the coastal edge flooded lower areas at the southern end of the county. The potential for increased mosquito production was enormous.

Phase II - Surveillance and Larval Controls (Days 4-12)

The county was divided into three territories. The northern section consisted of flood plain and woodland pool freshwater sites. The central territory consisted of container mosquito and freshwater hatch sites. In the southern territory, the emphasis was on the salt marsh breeding sites along the coastal edges.

Three crews of two workers each began documenting mosquito-landing counts - that is, the number of mosquitoes landing on their persons over a two-minute period. larval dipping was conducted at previously mapped sites. Vector Control began full-scale larvicide applications of BTI granules at the larger sites and liquid BTI in roadside ditches. All treatment sites met one of two criteria before BTI was applied:

1. The site had a history of mosquito production.

2. The site had not been mapped, and it dipped positive for mosquito larvae.

If an unmapped site did not dip positive, it was mapped and monitored for future reference.

As part of the Phase II response, Vector Control's telephone was forwarded to the main office at the health department, which assisted with hundreds of mosquito complaints and questions. All calls were documented, sorted, and prioritized. Information about Vector Control activities and container breeding mosquitoes was disseminated. These efforts reinforced the public's positive image of mosquito control during the hurricane response.

Initial reports from the central crew suggested that the container mosquito, Ae. albopictus, had survived the hurricane exceptionally well. Container removal became the priority. Public service announcements using television, newspaper, and radio were initiated. The public was asked to "tip and toss" containers to minimize container breeding sites. The public's response was outstanding.

The southern crew reported elevated landing counts during daily larval treatments. Selective light trapping was initiated in southern NHC for the salt marsh mosquitoes, Aedes sollicitans and Aedes taeniorhynchus. Somehow, a larval population of Ae. sollicitans had survived the hurricane. The hurricane-driven hatch was not expected for another five or six days. One ground ULV generator was routed to spot-spray the high-density mosquito areas and to give the National Guard bivouacked in southern NHC some relief. Salt marsh mosquito populations are normally the department's focus in September and October; however, freshwater mosquitoes would be the focus with Hurricane Fran.

On Day 8, the helicopter from Clarke Mosquito Control arrived. The first order of business was to make an aerial survey of the standing water in NHC with a video camera. Northern NHC flood plains were saturated to unprecedented levels. Acres of standing water were recorded in areas not previously known to breed mosquitoes. Along the Cape Fear River and the Intracoastal Waterway, 21 dredge spoil islands were flooded and inaccessible. Two hundred and fifty acres of Eagle Island, a 960-acre dredge spoil island, were flooded. When the aerial surveillance was complete, the next step was aerial larvicide applications.

The helicopter, equipped and calibrated for larvicide application, treated 21 dredge spoil islands and 60 acres of Eagle Island. A review of the historical records and of the aerial video familiarized the pilot with each island and directed placement of the larvicide. The product selected for aerial application was a granule IGR with a 30-day residual at a rate of 2.5 pounds per acre. After a calibration conversion to BTI granules, 175 acres of an inaccessible flood plain were treated next to a large subdivision in northern NHC.

While the helicopter was applying larvicide to inaccessible areas, the central crew was diverted to the Buffalo turbine. The truck-mounted Buffalo turbine is used to treat large sites quickly. It consists of a 100-gallon tank and a turbine engine that produces a high-velocity air stream of approximately 175 miles per hour. The forced air is used to wind-drift liquid BTI across freshwater sites.

Mosquito landing counts, larval development, and increasing mosquito complaints suggested that adult mosquito populations would emerge over the weekend, and they were projected to reach ULV treatment levels on Day 11 or 12. Increased light-trap surveillance confirmed the projection. On Day 11, light trapping was increased from 12 to 25 stations. The traps were deployed across the county between four and nine o'clock p.m. and retrieved early the following morning. Mosquito populations were tallied for each station. Comprehensive light trapping indicated that freshwater flood plain and woodland pool species should be the focus of the adulticide applications [ILLUSTRATION FOR FIGURE 2 OMITTED]. Press releases issued through the Emergency Response Headquarters Media Center informed the public about the spray schedule for helicopter applications. The clerical specialist answering the phones was also provided with schedules of projected aerial and ground applications for the week.

Phase III - Adult Control (Days 13-22)

Aerial adulticide applications started on September 18, 1996, and were completed on September 21, 1996. The helicopter treated 1,700 acres of coastal edge in southern NHC and 8,300 acres of freshwater flood plain in northern NHC. Applications were made near populated areas.

Two truck-mounted ULV generators were dedicated to the populated areas surrounding each aerial application. Ground ULV applications started on September 16, 1996, and continued through September 27, 1996. Fifty-one four-hour spray routes were completed over the 11-day period. Two environmental health specialists, with current public health pest-control licenses, were recruited to supplement an exhausted crew.

Phase IV - Follow-Up Surveillance (Days 23-26)

Light-trap data collected during this week would determine whether continued application of adulticide was required. Telephone complaints dropped to normal levels after Day 23 [ILLUSTRATION FOR FIGURE 3 OMITTED]. The northern and southern larvicide crews both reported adult mosquito populations below normal levels. The 1996 light-trap data were compared with the 1995 data, and the mosquito populations were found to be well below the norm for the last week of September. Vector Control's hurricane response was complete.

Conclusion

Baseline mosquito information is crucial to successful emergency mosquito control. The identification of mosquito breeding sites in a defined control area provides insight into mosquito nuisance populations and their disease potential. The species and number of mosquitoes in a defined control area can be determined by collecting adult mosquitoes using landing counts or light traps at or near the breeding sites. An understanding of the biology of each species is essential in both larvicide and adulticide control measures. Adult mosquito identification helps to determine flight ranges, as well as the time of day when the effects of ULV control efforts can be maximized.

The identification of breeding sites furnishes information on species diversity, seasonality, and distribution of adult mosquitoes. Documentation and mapping of high-density mosquito breeding sites using larval dipping identifies the precise area for control. Employing larval control measures at individually mapped sites greatly reduces the disease potential by minimizing adult dispersion of disease vectors into populated areas. Larvicide applications to a smaller area reduce the amount of adulticide required for control, saving inventory and manpower and reducing the pesticide impact on the environment. Treatment of mosquito larvae with biorationals further reduces the chemical impact on the environment.

Climatological data considered together with baseline mosquito data show population trends throughout the treatment season. Larval site inspections scheduled with an understanding of annual precipitation data provide insight that helps predict the extent of mosquito production. Thus, baseline information directs routine larval and adult control measures and is the foundation of an effective disaster response.

Acknowledgments: The author thanks the New Hanover County Vector Control staff, Rick Hickman, Bruce Harrison, Ph.D., Charles Apperson, Ph.D., and Roger Nasci, Ph.D. for their commitment and expertise.

REFERENCES

1. Centers for Disease Control and Prevention (1996), Eastern Equine Encephalitis Risk in North Carolina After Hurricane Fran, Update 1, Wilmington, N.C.: U.S. Department of Health and Human Services.

2. National Oceanic and Atmospheric Administration, Department of Commerce (1996), Local Climatological Data, Annual Summary - Wilmington, North Carolina, Asheville, N.C.: National Climatic Data Center.

3. Nasci, R.S., and C.G. Moore (1993), "Planning for Emergency Mosquito Surveillance and Control," Wing Beats of the American Mosquito Control Associations, 4(2):4-7.

4. Centers for Disease Control and Prevention (1993), "Emergency Mosquito Control Associated with Hurricane Andrew - Florida and Louisiana, 1992," Morbidity and Mortality Weekly Report, 42(13):240-242.

Corresponding author: Jeffrey S. Brown, New Hanover County Health Department, Vector Control Division, 2029 South 17th Street, Wilmington, NC 28401. E-mail: [e-mail]