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

An Atlas of Historical Gypsy Moth Defoliation & Quarantined Areas in the US.

This is a collection of various maps that describe the history and future of gypsy moth in North America. They fall into three categories:

Gypsy Moth Defoliation Maps

Gypsy moth populations may exist for many years at low densities such that it may be difficult to find any life stages. Then, for reasons that are not completely understood, populations may rise to very high densities and substantial defoliation of the canopy may occur.

Each state in the northeast monitors gypsy moth defoliation annually in all locations using aerial sketch maps. Maps are sketched during a series of low level reconnaissance flights in late July when defoliation is at its peak. Thirty percent (30%) defoliation is considered the lower threshold for detection from the air. In situations where there is doubt as to the cause of the defoliation, ground checks for the presence of gypsy moth life stages are made. Initially the aerial sketch maps are overlaid on standard U.S. Geological Survey (1:24,000) topographical maps. Subsequently a composite mosaic is generated for the entire state at 1:1,000,000 scale. Mapping processes vary among states and years, resulting in a strong likelihood of significant data errors from systematic and non- systematic sources. The likely presence of these errors dictated the course spatial resolution of maps presented.

A geographic information system (GIS), IDRISI, was employed to assemble, collate, and analyze gypsy moth defoliation data(Eastman 1989). IDRISI is a raster-based (grid cell) GIS for capturing, storing, analyzing and displaying geographical data, designed for research applications.A 2 x 2 km grid cell size was selected as standard for all map layers in the GIS. The grid size was selected because it represented the minimum dependable spatial resolution of the defoliation data available from state agencies.

Spatial error is unavoidably generated in the process of recording defoliation on sketch maps from aircraft (Talerico 1981). Error occurs with respect to the exact location, degree, and aerial extent of defoliation, but this locational error is generally less than 1 km in magnitude. One advantage of a raster-based GIS is that the inherent uncertainty of data is maintained and displayed by the 'saw tooth' effect of adjacent cells however the choice of such a coarse scale of resolution raises serious issues concerning data accuracy and the cascading effect of errors as data layers are manipulated (Chrisman 1987). Without corroborating evidence at a fine scale of resolution it is not possible to provide accurate estimates of the errors.

GIS investigation is enabled by the use of multiple layers of geographical data (map layers) each coordinated to the rest by means of geo-reference points. To create a uniform set of geographically-referenced defoliation data, the composite maps for the period 1969 to 1989 were first transferred to mylar stable-base sheets. At least four geo- reference points, on clearly recognizable intersections of county boundaries, were accurately located. The prepared maps were then scanned using a digital scanner set at 150 dots per inch resolution. Binary TIFF files from the scanner were converted to ASCII IDRISI 'image' format and saved as IDRISI images or map layers. The transformation of each map layer to a common base map resolution and projection was achieved through a "rubber-sheeting" procedure (Burrough 1988). In transforming maps of various scales and projections, IDRISI resamples each scanned defoliation image to match the location of the four geo-reference points on the base map (Eastman 1989).

Historical maps of defoliation were not available in some states prior to 1984. It is obvious from these maps that outbreaks occur over large regions, often with considerable synchrony and may persist for many years (Liebhold and Elkinton 1989, Hohn et al. 1994, Williams and Liebhold 1995). Obviously, areas where gypsy moth has only recently invaded will have a lower frequency of defoliation but beyond that pattern, areas with very high defoliation frequencies represent forested areas where composition is highly susceptible to defoliation (Liebhold et al. 1994, Gansner et al. 1993).

1972 1973 1975 1976 1977
1978 1979 1980 1981 1982
1983 1984 1985 1986 1987
1988 1989 1990 1991 1992
1993 1994 1995 1996 1997
1998 1999 2000 2001 2002
2003 2004 2005 2006 2007
Defoliation Frequency 1975-2002

Click here to browse customized maps of gypsy moth defoliation via the online spatial data query system, "Alien Forest Pest Explorer"

Gypsy Moth Quarantine Maps

Beginning with the enactment of the Domestic Plant Quarantine act of 1912, the USDA has regulated the movement of plant material from areas determined to be infested with gypsy moth (Weber 1930). The methods used to designate the infestation status of an area have varied but the designation of infested usually resulted from multiple finds of one or more life stages. Trapping of males in pheromone-baited traps is a powerful tool for detecting incipient gypsy moth populations; these traps have been used to define the infested area since the turn of the century (before the isolation, identification and synthesis of disparlure, agencies often used extracts of live females as trap baits). The official USDA quarantine regulation were used in this study as a method for determining the annual spatial distribution of gypsy moth in the US. Since 1934, the quarantined area has been defined in the annual Code of Federal Regulations under Title 7, chapter 301.45-2a (Administrative instructions designating regulated areas under the gypsy moth and brown-tail moth quarantine and regulation). A county was designated as infested if the regulations listed any part of it as part of the generally infested area, suppressive area, high-risk area or low-risk area. In a few cases (mostly isolated infestations), a county was designated as infested one year, but subsequently was not listed as infested; in these cases we designated a county as infested only if it did not later become "uninfested". The quarantined area was defined in other publications prior to 1934 (Burgess 1915, 1930). Various other sources were used to determine gypsy moth distribution between 1900 and 1912 (Anonymous 1906, 1907; Burgess 1913).

The spatial resolution of the historical descriptions of the infested area varied through time and across different regions. Simple lists of infested counties were the most common method used in these records to describe the area. Therefore, we used US counties as the smallest unit for describing the annual distribution of the generally infested area. Geographical information system software, GRASS (U.S. Army Corps of Engineers 1993), was used to generate maps of the infested area. All maps were drawn using a Lamberts equal area projection (Snyder 1987).

1900 1905 1909 1912 1914 1934 1938 1943
1945 1949 1955 1960 1961 1965 1966 1967
1968 1969 1970 1971 1972 1973 1974 1975
1976 1977 1978 1979 1980 1981 1982 1983
1984 1985 1986 1987 1988 1989 1990 1991
1992 1993 1994 1995 1996 1997 1998 1999
2000 2001 2002 2003 2004 2005 2006 2007
2008 1900-2007 animation

Click here to browse customized maps of historical gypsy moth spread via the online spatial data query system, "Alien Forest Pest Explorer"

Maps of forest susceptibility to gypsy moth

Since the gypsy moth was originally introduced near Boston in 1868 or 1869, it has been slowly expanding its range to include the entire northeastern US and portions of Virginia, North Carolina, Ohio, and Michigan. It is inevitable that the gypsy moth will continue to spread to the south and east over the next century.

Considerable effort has been expended on documenting the spatial extent of gypsy moth defoliation via aerial sketch mapping and other techniques. This information has been used to map the spatial distribution of forests susceptible to the gypsy moth within the generally infested region (Liebhold and Elkinton 1989, Liebhold et al. 1994). In order to plan for the management of the gypsy moth over the next decade and beyond, there is a need to delimit the distribution of susceptible stands in areas that are currently uninfested.

The gypsy moth is a polyphagous insect; North American populations feed on over 300 different shrub and tree species (Leonard 1981). Despite this wide breadth of host preference, there is considerable variation within northeastern North American forests in their susceptibility to defoliation; we use "susceptibility" defined as the probability or frequency of defoliation (for a description of alternative approaches, see Twery et al. [1990]).

Several studies have focused on relating various characteristics of forests to their susceptibility to defoliation by the gypsy moth. These studies have yielded susceptibility models of varying levels of complexity. Probably the most important factor affecting stand susceptibility is the proportion of basal area represented by species highly preferred by the gypsy moth (Herrick and Gansner 1986). Other variables, such as the predominance of chestnut oak, the abundance of tree structural features (e.g. bark flaps), and various site characteristics (e.g. soils), are also known to be correlated with susceptibility (Bess et al. 1947, Valentine and Houston 1979, Herrick and Gansner 1986) but often these correlations are specific for certain regions or the variables are rarely measured in most forest inventories.

Gansner et al. (1993) demonstrated how suscpeptibility models can be applied to forest inventory data in order to map forest susceptibility at the landscape-level. In this section, we applied a similar technique to map forest susceptibility over the conterminous US.

Assessment of forest susceptibility was based upon existing forest inventory data collected throughout the conterminous US. In the eastern United States, all inventory data were obtained from the USDA Forest Service Forest Inventory and Analysis (FIA) (Hansen et al. 1993). In the eastern US, FIA inventories federal land as well as privately held land. In the East, these inventories are usually conducted every 5 to 15 years. Each state typically contains over 1,000 irregularly spaced Forest Inventory and Analysis (FIA) plots. In the western US, FIA does not inventory National Forests. Therefore data on western forests was obtained as a mixture of FIA data as well as inventory data collected by individual national forests.

Sampling methods used for inventorying forest resources varied among regions and organizations conducting inventories (Table 1). All inventory data contained data on individual trees, as well as data about plots. Individual tree records were used to sum total basal area by each species for each plot. These plot records were then expanded (using appropriate expansion factors) to county level estimates of basal area per acre.

Inventory data were available from most portions of the conterminous US (Figure 31). However, inventories were not available from other portions. State and private land in the western 2/3 of Oklahoma and Texas are not inventoried by FIA and therefore these areas were missing. FIA data were available from every state in the country but in the west, FIA does not inventory National Forests and in some cases, National Forest inventory data were not available.

National Forest inventory data were occasionally incomplete. For example, all NFS data from California (Region 5) did not include either districts or counties. Therefore, it was necessary to randomly assign plots within a given National Forest to counties (weighted by the proportion of the national forest in each county). In portions of the southwest (Region 3), county was not included and similar assignments were made to counties within a district.

We adopted proportion of basal area represented by preferred species as the measure of forest susceptibility. While other variables (e.g., proportion chestnut oak) may help to explain more variation in susceptibility, these models are less likely to be successfully applied outside the range of data originally used to calibrate them. Montgomerys (1990) 3-way classification (preferred, susceptible, immune) was used to classify each tree species as preferred by the gypsy moth. This classification was based on a summary of field and laboratory studies, as well as extrapolations based upon taxonomic affinity, and is described in detail elsewhere (Liebhold et al. 1995).

Table 2 lists the top 20 preferred species, ranked on their total basal area over the inventoried area. Out of the highest ranking 10 species, only one species, quaking aspen, occurs in the western US. Some caution should be used in interpreting this ranking because the lack of inventory data in certain western counties (Figure 31) resulted in a bias favoring eastern species. Nevertheless, these data indicate that most of the susceptible basal area (which is closely correlated to foliage area) is concentrated in the eastern US.

White oak was the top-ranking susceptible species (Table 2). Figure 32 shows the distribution of this species. High concentrations of white oak exist throughout the east, but the highest concentrations exist in the Ozarks, Cumberland Plateau, and the southern Appalachians. Most of these areas are currently beyond the expanding range of the gypsy moth. Sweetgum was the 2nd most common susceptible species (Table 2). This species is common throughout the Piedmont from North Carolina to Louisiana, and also exists largely beyond the current range of the gypsy moth (Fig. 33). Quaking aspen was ranked number 3 (Table 2). It is one of only a few tree species whos range extends across the eastern and western portions of the continent though it is most common in the northern portions of the lake states (Figure 34). Most of these areas of high aspen concentration are beyond the current range of the gypsy moth. Northern red oak was ranked no. 4 in terms of its total basal area (Table 2). This species is common throughout the northeast and is common in portions of the lake states (Figure 35). Much of the range of this species occurs in areas already infested by the gypsy moth. The ranges of the other most common preferred tree species are given in Figs. 36-51.

Over-all forest susceptibility was quantified using the total basal area per acre of all preferred species (Fig. 52). The areas with the highest concentration of susceptible forests were in the central and southern Appalachians, the Cumberland Plateau, the Ozark Mountains, and the northwestern lake states. Comparison of these maps with the know distribution of individual susceptible species (Figs. 32-51) indicates that oaks are the major component of susceptible forests in the Appalachian, Cumberland, and Ozark areas but quaking aspen is the major susceptible species in the northwestern lake states. One interesting item is that even though sweetgum is the second most abundant susceptible species (Table 2), it is apparently not abundant enough to cause high levels of stand susceptibility; it is rarely associated with enough other susceptible species to make the stands in the Piedmont highly susceptible (Fig. 52).

There are several caveats that should be attached to the interpretation of these data. As mentioned above, inventories were not available from any urban forests and inventories were missing from several forested areas in the west (Fig. 31). Also, it is important to keep in mind that susceptibility assumptions are based upon several assumptions that have not been completely proven. First, the suitability of many tree species to the gypsy moth is in many cases, based upon incomplete information. For many species, feeding trials have not been performed and for other species for which there is some laboratory data available, data on susceptibility to defoliation in natural forests is unknown (Liebhold et al. 1995).

Despite these limitations, these results should be useful in planning for the future. The finding that the gypsy moth has not yet invaded most of the susceptible forests in the US suggests that there still may be considerable value in limiting the future spread of the gypsy moth. It also indicates that both the impacts of defoliation and costs of gypsy moth management are likely to increase in the future.

Average basal area per acre of preferred species.

Proportion of stand basal area in preferred species.

Proportion of land area covered by susceptible stands. (> 20% of basal area in preferred species)

Proportion of land area covered by highly susceptible stands. (> 50% of basal area in preferred species)

Proportion of land area covered by extremely susceptible stands. (> 80% of basal area in preferred species) for each county.

The 20 most common gypsy moth host trees.


Anonymous. 1906. First annual report of the superintendent for supressing the gypsy and brown-tail moths. Massachusetts Pub. Doc. No. 73. Wright & Potter Printing Co., Boston. 161 p.

Anonymous. 1907. Second annual report of the superintendent for supressing the gypsy and brown-tail moths. Massachusetts Pub. Doc. No. 73. Wright & Potter Printing Co., Boston. 170 p.

Bess, H..A.; Spurr, S.H; Littlefield, E.W. 1947. Forest site conditions and the gypsy moth. Harv. For. Bull. No 22. 56 pp.

Burgess, A.F. 1913. The dispersion of the gipsy [sic] moth. U.S. Dept. Agric. Bull. No. 119. 62 p.

Burgess, A.F. 1915. Report on the gipsy [sic] moth work in New England. U.S. Dept. of Agric. Bull. No. 204. 32 pp.

Burgess, A.F. 1930. The gipsy [sic] moth and the brown-tail moth. US Dept. of Agric. Farmers Bull. 1623. 32 p.

Burrough, P.A. 1988. Principles of Geographical Information Systems for Land Resources Assesment. Clarendon Press, Oxford.

Campbell, R.W.; Sloan, R.J.. 1977. Forest stand responses to defoliation by the gypsy moth. For. Sci. Monog. 19. 34 pp.

Chrisman, N.R. 1987. The accuracy of map overlays: a reassessment. Landscape and Urban Planning 14:427-439.

Eastman, J.R.1989. IDRISI: a Grid-Based Geographical Analysis System. Clark University, Worcester, MA.

Dobson, A.P.; May, R.M. 1986. Patterns of invasions by pathogens and parasites. pp. 58-76 in: Ecology of biological invasions of North America and Hawaii. Mooney, H.A. and Drake, J.A. (eds.). Ecological Studies 58. Springer-Verlag. New York.

Dunlap, T.R. 1980. The gypsy moth: a study in science and public policy. J. For. Hist. 24: 116-126.

Elton, C.S. 1958. The ecology of invasions by animals and plants. Methuen, London. 181 pp.

Forbush, E.H., and C.H. Fernald. 1896. The Gypsy Moth. Wright & Potter, Boston. 495 p.

Gansner, David A.; Drake, David A.; Arner, Stanford L.; Hershey, Rachel R., King, Susan L. 1993. Defoliation potential of gypsy moth. USDA Forest Service Res. Note NE-354

Hansen, M.H.; Frieswyk, T.; Glover, J.F.; Kelly J.F. 1993. The eastwide forest inventory data base: users manual. USDA For. Serv. Gen. Tech. Rep. NC-151.

Herrick, O. W.; Gansner, D.A. 1986. Rating forest stands for gypsy moth defoliation. USDA For. Serv. Res. Pap. NE-583. 4 pp.

Leonard, D.E. 1981. Bioecology of the gypsy moth. p. 9-29 in: The gypsy moth: Research toward integrated pest management. Doane, C.C. and M.L. McManus (eds.). USDA Tech. Bull. 1584.

Liebhold, A.M.; Elkinton, J.S. 1989. Characterizing spatial patterns of gypsy moth regional defoliation. 1989. Forest Science. 35: 557-568.

Liebhold, Andrew; Mastro, Victor; Schaefer, Paul W.. 1989. Learning from the legacy of Leopold Trouvelot. Bull. Entomol. Soc. Am. 35: 20-21.

Liebhold, A.M., J. Halverson and G. Elmes. 1992. Quantitative Analysis of the Invasion of Gypsy Moth in North America. J. Biogeography. 19: 513-520.

Liebhold, A.M.; Elmes, G.A.; Halverson, J.A.; Quimby, J.. 1994. Landscape characterization of forest susceptibility to gypsy moth defoliation. For. Sci. 40: 18-29.

Liebhold, Andrew M.; Gottschalk, Kurt W.; Muzika, Rose-Marie; Montgomery, Michael E.; Young, Regis.; ODay, Kathleen; Kelley, Brooks. 1995. Suitability of North American Tree Species to the Gypsy Moth: A Summary of Field and Laboratory Tests. USDA For. Serv. Gen. Tech. Rep. NE-211.

McManus, Michael L.; McIntyre, Thomas. 1981. Introduction. pp 1-7 in: The Gypsy Moth: Research Toward Integrated Pest Management. (Doane, C.C. and McManus, M.L. (eds.). USDA For. Serv. Tech. Bull. 1584.

Snyder, J. P. 1987. Map projections - a working manual. U.S. Geol. Surv. Prof. Pap. 1395.

Talerico, R.L. 1981. Defoliation as an indirect means of population assessment. p. 38-49 in: The gypsy moth: Research toward integrated pest management. Doane, C.C. and M.L. McManus (eds.). USDA Tech. Bull. 1584. 757 pp.

Twery, M.J.; Elmes, G.A. ; Yuill, C.B.; Millette, T.L.. 1990. Using GIS to assess gypsy moth hazard. in: proc. of the ASPRS/ACSM 1990. 3: 18-23.

Twery, M.J. 1991. Effects of defoliation by gypsy moth. pp. 27-39, in K.W. Gottschalk, M.J. Twery, and S.I. Smith (eds.), Proc. U.S. Dept. Agric. Interagency Gypsy Moth Research Review - 1990. U.S. Dept. Agric. For. Serv. Gen. Tech. Rep. NE-146. 152 pp. U.S. Army Corps of Engineers. 1993. GRASS 4.1 Users Reference Manual. U.S. Army Corps of Engineers. 555 pp.

Valentine, H.T.; Houston, D.R. 1979. A discriminant function for identifying mixed-oak stand susceptibility to gypsy moth defoliation. Forest Sci. 25: 468-474.

Weber, G.A. 1930. The plant quarantine and control administration: its history, activities and organization. Brookings Inst. (Washington, DC), Inst. For Gov. Res. Serv. Mon. of the U.S. Gov. No. 59. 198 p.

Return to Gypsy Moth in NA

Last modified 10-29-03 by Sandy Liebhold .

USDA Forest Service - Northeastern Research Station

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