USDA Forest Service

Gypsy Moth In North America


Res. Work Unit NE-4557
180 Canfield ST.
Morgantown, WV 26505

(304) 285-1503

United States Department of Agriculture Forest Service.

Evaluate Our Service
We welcome your comments on our service and your suggestions for improvement.
Logo of New Century of Service

Gypsy Moth Spread in the Future

The draft gypsy moth environmental impact statement considers six alternative approaches that the USDA can take to gypsy moth management:

  • Option 1: No Action
  • Option 2: Suppression only
  • Option 3: Eradication only
  • Option 4: Suppression and eradication
  • Option 5: Eradication and Slow-the-spread
  • Option 6: Suppression, eradication, and slow the spread

    In this study we attempted to predict the future gypsy moth distribution in the coterminous US in the year 2010 under each of these alternatives. These predictions were based upon the following assumptions:

    • without a slow the spread program gypsy moth will continue to spread in the same manner as it is currently spreading.
    • the transition area consists of a band of 125 km around the generally infested area.
    • a slow the spread program would result in a decrease of 25% - 75% in the annual rate of radial spread
    • If all eradication were abandoned, 1/2 of all isolated infestations scheduled for eradication, would become established.
    • Isolated infestations will not substantially spread until 10 years after discovery.

    Predicting Future Gypsy Moth Spread

    We predicted future gypsy moth spread to follow the same behavior as has been observed over the last 30 years. Historical data on the generally infested area were extracted from the annual Code of Federal Regulations as described in Liebhold et al. (1992). In that paper we also present two spread rates for 1966-1990 for northern (cold = slow spread = 7.6 km/yr) and southern (warm = fast spread = 20.8 km/yr). We applied those spread rates to the existing quarantine area in a GIS to predict area quarantine for 1995-2020. For each year (both past and present), we identified quarantined counties, multiplied each of their areas by the proportion of the infested area that is forested by susceptible stands (stands where the preferred species represent > 20% of the basal area - estimated from FIA data), and totaled these to derive an estimate of the area of the infested area that is susceptible to gypsy moth defoliation. We applied these spread models to generate the expected extent and area of the generally infested area and transition zone in 2010.

    Mapping forests that are susceptible to gypsy moth defoliation

    This process is describe in more detail elsewhere. Unfortunately detailed forest composition maps are not available for all portions of the US. Therefore, we had to map forest susceptibility at the county level. First, we took inventory data collected by several thousand plots by the USDA Forest Service Forest Inventory and Analysis Project and used these data to calculate the proportion of each county in the US that is covered by forests that are susceptible to the gypsy moth. Here we defined a forest as being susceptible to the gypsy moth as one where visible defoliation may occur. Several models are available for relating the probability of defoliation to forest compostion but we used a very simple model developed by Herrick and Gansner (1986). In this model, susceptible forests are defined as those stands where greater than 50% of the basal area is represented by susceptible species. A list of susceptible species is given elsewhere. Estimates of the proportion of each county that is covered by forests susceptible to gypsy moth defoliation were combined in a GIS with maps of past, and future gyspy moth infested areas to calculate past and future areas where defoliation could or will ocurr.

    Gypsy moth defoliation.

    In Liebhold et al. (1994) we were reported that there is on average a 5 year lag from the time of first quarantine to the time of first defoliation in any county. Therefore, we used historical defoliation data to fit a model of yearly defoliation area (from the GM digest) as a function of the susceptible quarantine area lagged by 5 years:
    log(defol) = -9.60 + 2.15 log(quarantine-and-susceptible)

    Isolated gypsy moth infestations.

    The USDA APHIS maintains a database of historical eradication programs. We used this historical data to fit a logistic regression model that predicts the yearly probability of an eradication project (=isolated infestation) as a function of the distance (DIST) from the generally infested front, mean human population density (DENSITY) in the county, mean per capita income (INCOME) for that county and size of the generally infested area:
    B = -8.76 -0.00032 DIST + 0.00018 DENSITY + 0.000323 INCOME + 1.42E-8 SIZE
    We then simulated the yearly generation of isolated infestations from 1995 - 2010. We assumed that without eradication, isolated infestations would remain restricted to individual 1 x 1 km raster cells for 10 years, and then they would spread as predicted above under "Predicting Future Gypsy Moth Spread". Since the the logistic regression model above predicts only a PROBABILITY of an isolated infestation, we use a Monte-Carlo technique to simulate these infestations. We ran simulations from 1995-2010 100 times and then for each of these simulations we calculated the area of the generally infested area and transition areas. The areas given in table 2-2 under options 1 & 2 were thus means of these 100 infestations.

    Current Conditions

    Here's a map of the 1994 quarantine that was used as a starting point for all simulations.


    Here are maps of expected conditions in 2010. Areas in blue are the generally infested area. Areas in red are the transition area.

  • USDA Forest Service - Northeastern Research Station
    Last Modified: Friday, 22 August 2003 at 17:48:24 EDT

    USDA logo which links to the department's national site. Forest Service logo which links to the agency's national site.