Research News: Gypsy moth trapping in NE Minnesota

Northeastern Minnesota trapping study: Effects of temperature and trap density on male gypsy moth recapture and flight patterns

By Mark A. White, George E. Host, and David R. Benson Natural Resources Research Institute, University of Minnesota-Duluth 5013 Miller Trunk Highway, Duluth, MN 55811

Gypsy Moth (Lymantria dispar L.) is one of the most destructive insect defoliators in North America. Populations are not yet established in Minnesota, however, the population front has moved into western Wisconsin. Although the Mixed Forest Province of Minnesota (Figure 1) tends to have a relatively low frequency of favorable temperature conditions for gypsy moth establishment, preferred host species (aspen and paper birch) are very abundant. Models based on climate data indicate that gypsy moth establishment is likely in the mixed forest province of Minnesota.

Attempts to control gypsy moth are based on detecting populations using grid networks of pheromone baited traps. Grid trapping densities were developed in regions such as Massachusetts and Lower Michigan where climate conditions differ from northeastern Minnesota. Based on previous work, we hypothesized that temperature conditions will have a significant effect on moth recapture rates.

Our primary objectives in this study were: 1) Assess the utility of three pheromone trap grids (1 trap/mi2, 16 traps/ mi2, 36 traps/mi2) for detecting gypsy moth populations in Northeastern Minnesota 2) Examine the effect of trap density, temperature across the twelve release periods on moth recapture and flight patterns.

map of study region and trapping grid plots

Study Region

The three study plots were located in the Toimi Uplands Subsection of the Minnesota Ecological Classification System (Figure 1). Upland support aspen birch forests with white spruce and balsam fir as associates. Smaller areas of northern hardwood forest occur as well. Lowland areas support, swamp conifers, black spruce, shrub swamps and sedge meadows. General forest conditions here are typical of northeastern Minnesota.

Sample Design

Pheromone baited traps were deployed in three plots at densities of: 1 /mi2, 16/mi2, 36/mi2 (Figure 1). The plot locations were based on the following criteria: similar vegetation and landform, accessibility, three mile minimum distance between nearest boundary. Nine hundred lab reared sterile male pupae were released at each plot center weekly from July 11 to Sept. 27 2002, for a total of 12 releases. Pupae were colored with an external dye so that release date could be identified in recaptured moths; they also had an internal lipid dye to distinguish them from any possible wild moth captures. We monitored emergence and recapture at each plot. Following each release, the core traps at each site were checked for male gypsy moths on a weekly basis. Other traps were checked approximately every 2 weeks. Two weather stations recording temperature and humidity were located at each plot center. Temperature data were summarized as a heat sum statistic for each plot for the 7 day period following each release of male pupae.

Results and Discussion

Emergence, Trap Density and Recapture Rates
Emergence of male moths from the pupal state was related to temperature over the 7 day periods. At very low temperatures (releases 11 and 12 in late September) emergence rates were very low (1-3%), during warmer periods emergence rates were much higher and more variable (39-77%).

The percent recapture by plot shows that there was a strong difference in the probability of recapturing a male in the different trapping grid densities (Figure 2). The lowest density grid recaptured 1% of emerged male moths, while the highest density grid (36 traps/mi2) recaptured 5% of emerged males, indicating that the probability of recapturing a male moth was 5 times greater in the high density versus the low density grid. The probability of recapturing a moth decreases with increasing inter-trap and nearest trap to release point distances.

average % recapture for three trapping grid densities over 10 releases

Temperature and Recapture Rates
Temperature (heat sum) showed no relationship to recapture rates in the 1 trap/mi2 grid, the 16 traps/mi2 grid showed a weak relationship, while the high density grid recapture rates showed a relatively strong association with heat sum. At low temperatures recaptures rates were uniformly low, at higher heat sum values, recapture rates were higher and more variable.

In the low density grid (1trap/mi2) results indicates that other factors led to the mortality of male moths before they could reach the traps. The longer distance to the nearest trap means that other factors such predation and physiological limitation had a greater effect on moth mortality.

In the high density grid (36 traps/mi2) recapture rates increased with higher heat sums because moths have a relatively short distance to travel to the pheromone source and were active for a greater proportion of the release period. At temperatures below approximately 705 F, male moths are inactive. During cooler periods, reaching the pheromone source takes longer and mortality increases due to factors such as predation and low energy reserves. However, during cooler periods, moths may live longer as they have longer inactive periods and thus use less of their available energy. Our data indicate that moths lived as long as 10-14 days during cooler periods.

One gypsy moth phenology model predicts peak male moth flight in NE Minnesota to be in late August to mid-September. In this study, male moths were active and recapture rates were high during this period. Climate records for this period of 2002 shows temperatures significantly above normal for the study region. So, typically during this time period, temperatures would be lower and male moth activity would also decrease.

Release site Predation and Recapture Rates
Beginning with the 5th release in August 2002, we observed significant predation at the release site of newly emerged male moths by black capped chickadees. Chickadees waited for moths to become visible and picked moths off of the release cartons. Once moths were in flight, they were a much more difficult target. We constructed nylon mesh chickadee exclosures, which were very effective in limiting predation. During some release periods, chickadee predation had a significant effect on the number of male moths leaving the release site.

Limiting predation at the release site should give more accurate estimates of the relationship to recapture rates and environmental variables such as temperature.

map of indepencence plot: 36 traps /mi2 percent recapture 10 releases

Recapture Rates and Dispersal Distance
We examined recapture by distance class for all three plots combined. Results show that close to 90% of males recapture occurred in traps within 2,625 ft. (800 m) of the release site while 10% of recaptures occurred at distances greater than 3,609 ft. (1,100 m) (Figure 3). The typical dispersal distance of < 2,625 ft. is similar to distances reported in other studies in Lower Michigan and Massachusetts, indicating that northeastern Minnesota dispersal distances are similar to these other regions.

Population Estimates
We used recapture probabilities from the twelve releases to compare estimated with actual male moth populations and the variability or uncertainty of those estimates for the 3 trapping grid densities.

Average population estimates for all three grid densities were similar to actual average population levels over ten release periods (Figure 4). However the 1 trap/mi2 grid showed substantially higher variability than the two higher density grids. In two instances, 0 moths were recaptured at the 1/mi2 grid when over 500 males emerged at each release.

population estimates for 3 trapping grid densities

Grid center release represents the maximum possible distance a population of males could be from nearest trap or female moth. Recapture probability decreases with increasing distance to nearest trap. For the lowest density grid (1/mi2) the distance to nearest trap is 3733 ft (1138 m), substantially greater than the typical dispersal distance of 2625 ft. Therefore, population estimates at the 1/mi2 density could be viewed as representing maximum population density. Following this logic, zero moths trapped does not equate to zero population but indicates population between 0 and maximum estimated population.

Given that this study represents one field season of data, results should be used with caution. Additional work should be done in similar and different habitats in northern Minnesota to refine habitat and climate-recapture relationships.


  • Recapture rates varied relative to trap density with the probability of recapture decreasing with increased distance between traps and the release point and nearest trap.
  • Recapture rates varied relative to temperature in the higher density plots with higher recapture probability with higher temperatures. Controlling release site predation should yield more accurate temperature-recapture relationships.
  • The typical dispersal distance of < 2,625 ft. is similar to distances reported in other studies in Lower Michigan and Massachusetts, indicating that northeastern Minnesota dispersal distances are similar to these other regions.
  • The 36 traps/mi2 grid produced the most accurate estimates. The 1 trap/mi2 grid showed the greatest uncertainty in population estimates.
  • Grid center release represents maximum possible distance a population of males could be from nearest trap or female moth. Therefore, population density estimates derived from this study should be considered maximum estimates.
  • Zero trap catch values may not indicate zero population levels, particularly in the lowest density trapping grid (1 trap /mi2).


This project was funded through a grant from the USDA APHIS program with support from the USDA Forest Service State and Private Forestry program and the Slow the Spread Foundation. The USDA Otis Plant Protection Laboratory reared the male pupae and provided valuable assistance. Susan Burks (MN DNR), Kevin Connors (USDA APHIS) and Mike Connors (USDA Forest Service) developed the original project concept and design. Field assistance was provided by The Minnesota Department of Agriculture, MN DNR Forest Health Program, USDA Forest Service, and USDA APHIS. The St. Louis County Land Department provided the field sites for this study.