Abstract

Trial applications of liquid Bti were used in an attempt to control Aedes vexans in a river flood plain area. Mortality ranged from 0-100%, and was most likely a result of varying vegetative covers in the control areas preventing the Bti from reaching the water surface.

Introduction

The Sudbury River flood plain is one of several major sources of pest mosquitoes in the East Middlesex Mosquito Control district. The Sudbury River begins at the Cedar Swamp in Westborough. Initially the river is narrow and flows rapidly. However, beyond the Saxonville Dam in Framingham the gradient changes to barely a one inch drop per mile and the river broadens and meanders through marsh meadows in Sudbury, Lincoln, Concord, and Wayland. These meadows usually flood annually in the spring. Periodically there are also summer and fall floods. These typically occur when more than 3" of rain falls along the Sudbury River drainage area during a short time period (Henley personal communication). Since much of this area consists of grass, reed, and shrub wet meadows, and because there is a delay in the cresting of water along the flood plain following a flooding rain, the area provides ideal habitat for Aedes vexans during periods of summer flood. Aedes vexans eggs typically hatch within 24-48 hrs after re-flooding (Horsfall 1958). The larvae require 7-10 days to complete their aquatic stages, depending on temperature (Gerberg et al. 1994). The most effective control using Bti in the past occurred with larvae at the lst and 2nd instars.

In the late summer of 1994, between Aug 13 and Aug 22, three storms dropped 5.81 " of rain along the Sudbury River flood plain. Larval mosquito surveys done on Aug 22nd and 23rd showed that larval populations were excessive, indicating that a massive emergence of Aedes vexans was likely. In the past, summer emergences of Aedes vexans have been controlled using helicopter applications of both granular Bti, with good results, and Bti pellets, with less success. The decision to use a liquid formulation of Bti this summer was based primarily on cost considerations. Liquid Bti has been used successfully to control spring Aedes spp. in freshwater wetlands and to control Aedes sollicitans at salt marshes. Use of this product for summer control of flood water mosquitoes could save as much as 50% over the cost of granular applications. The products used for this application included 125 gallons of Acrobe and 15 gallons of Vectobac 12AS. American Cyanamid, through Northeast Vector Management, contributed 85 gallons of Acrobe for this field trial. The remaining 55 gallons was purchased in 1994. It should be noted that 80 gallons of the Acrobe contributed by American Cyanamid was formulated in 1992 and 1993. The field trials had originally been planned for 1993 but were postponed a year due to dry conditions.

Material and Methods

We used both Acrobe and Vectobac 12AS applied at a per acre rate of 16 oz of Bti plus either 16 oz of water for a 1:1 mix or 48 oz of water for a 3:1 mix to Sudbury River flood plain areas in Sudbury and Wayland. Sites chosen for control were open canopy sites with varying amounts of button bush, sedge, reeds, grasses, cattail, purple loosestrife, and other wetland vegetation. A helicopter-mounted Simplex liquid spray system, holding fourteen raindrop nozzles, was used to apply droplets with a volume median diameter (VMD) of 200-40OAt over the targeted areas. This system produces droplets that come out in a swirl. This, combined with rotor back wash reduces drift, which remains a potential problem with aerial applications of liquid Bti. However, because the Bti mist does settle slowly compared with the heavier Bti granules and pellets, the helicopter pilot can more accurately line up each swath by following the mist that is settling from the previous swath. The Bti application was done on Aug 24 by Joe Brigham, Inc. using a Bell Jet Ranger helicopter carrying 80 gallons of Bti/water mix per trip and laying down a 100' swath.

In addition to pre and post-spray larval surveys, adult mosquito surveys were also done using dry ice baited CDC light traps placed at two locations in Wayland and one location in Sudbury adjacent to the flood plain. These sites, along with 46 others in 16 towns within the East Middlesex Mosquito Control district, provide data used in determining when control measures are needed in each area. Data for many of these traps have been collected since 1984. These data give an approximation of adult mosquito populations in the trap site area and allow an estimation of nuisance levels. They are not a good indication of level of control in areas targeted for aerial applications of Bti as not all vexans producing sites are included in these applications. However, they do give an indication of relative differences in mosquito populations from year to year. These differences are affected by many factors, including control measures.

Results and Discussion

Ten Wayland and 3 Sudbury sites were chosen for larval survey within the 3:1 Bti application rate area (Table 1). Twenty-five to fifty samples were taken per site, for a total of 450 pre-spray dips and 480 postspray dips. Estimated acreage is given for each wetland; some wetlands were surveyed at more than one site. Although sites chosen for control were open canopy sites, there was some variation in vegetation. Pod Meadow was primarily grasses and shrubs. Meadowview was predominantly cattail, as was the Griscom/Erwin road site and north of Rt. 20 on the west side of the river. The wetland behind Wayland High School was predominantly grasses and purple loosestrife with areas of cattail and shrubs. The Pelham Island Road site was heavy cattail and grasses. North of Route 20 to the east side of the river and the Community Gardens sites consisted primarily of shrubs, as did the Longfellow site. The Grove Street site was predominantly reed canary grass. Pantry Brook was predominantly grasses and purple loosestrife with areas of button bush. Great Meadows Headquarters was predominantly purple loosestrife, cattail and buttonbush.

Water depth varied considerably throughout these areas, ranging from several inches to several feet. In most areas it was possible to see water through the cover of vegetation. Percent reduction in number of larvae ranged from 0 - 100%, probably due to type and amount of vegetation at each site and water depth. It is likely that thick vegetation prevented the Bti droplets from reaching the water. Overall larval reduction in the 3:1 application area was 55 %.

Two Wayland and 4 Sudbury sites were chosen for larval survey within the 1:1 Bti application area (Table 2). Again, 25-50 samples were taken per site, for a total of 250 pre-spray dips and 250 post-spray dips. All the sites surveyed were purple loosestrife, cattail and button bush wetlands. Water depth varied, with areas of water visible through the vegetation at most sites. In these areas percent reduction of larvae ranged from 0 - 67 %, with an overall reduction of 36%. For comparison, 1989 spring application of Bti granules in Bedford, Sudbury and Weston resulted in larval reductions ranging from 31 - 99% with an overall reduction of 83 %. It was possible to find Bti granules at these sites, indicating that the Bti was reaching the targeted areas. Similar results were observed in the spring of 1990 and in summer applications of Bti granules.

CDC light trap data show some reduction in Aedes vexans in some areas compared with previous years (Fig 1). In 1990, some aerial application of granular Bti was done in Sudbury. In 1991, Bti pellets were used. In other years, no summer Bti applications were done in Wayland or Sudbury. In 1990, there were heavy rains from late July through early August. In 1991, Hurricane Bob deposited 5.2" of rain in the Wayland/Sudbury area on August 19th. In 1992, 6.23" of rain fell in the area between August 9-18. In 1993, there was a drought that started in May and ran through October. Therefore, variations in adult population data reflect differences in temperature and rainfall as well as efficacy of control (Table 3).

Although the cost of liquid Bti is $4-5 less per acre compared to the lowest successful application rate of granular Bti (Table 4), overall efficacy of control, which is less by as much as 40%, negates any cost benefit. Drift was not a factor contributing to poor control as low wind speeds prevailed throughout the application. Given that the sites were fully inundated on the day of application, it is likely that heavy growths of purple loosestrife, button bush, cattail, and other vegetation prevented the Bti from reaching the target sites. Prior applications of Bti granules have penetrated through shrub and emergent vegetation to the water surface and then slowly sank to the bottom releasing Bti from the granules exterior coating. Although Bti pellets also penetrated through the shrub and emergent vegetation, they most likely fell to the bottom of the wetland before the pellet could break apart and release it's Bti. In depths greater than several inches, the Bti from the pellets may not rise high enough in the water column to be effective. An attempt was made using liquid sensitive paper to see if Bti droplets were reaching the control areas. However, early morning mist coming off the flood plain was so heavy that the papers became saturated and unreadable. Because the more open areas generally showed the best control, it is likely that aerial application of liquid Bti would show better results in open canopy areas in the Spring before the vegetation emerges.

Literature Cited

Gerberg, E. J., D. R. Barnard & R. A. Ward, 1994. Manual for Mosquito Rearing and Experimental Techniques.

American Mosquito Control Association, Bulletin #5. Allen Press, Inc., Kansas. 98 pp.

Horsfall, W. R., 1958. Eggs of Floodwater Mosquitoes. V. Effect of oxygen on hatching of intact eggs. Ann Entomol. Soc. Amer. 51:209-213.