ABSTRACT. Indiana Department of Natural Resources and USDA Forest
Service applied pheromone flakes, disparlure, to a very low-level gypsy
moth (Lymantria dispar) population in south-central Indiana in June 2008
to eradicate the infestation. One aerial application of Disrupt II, a
disparlure-based mating disruptor, was applied over the forest canopy.
The fall of the flakes through the forest canopy to the forest floor was
monitored using 50.8 x 76.2 cm white foam core boards. Percent canopy
cover and distance from the plane's flight lines had no effect on
canopy through-fall. Only 0.73 flakes per [m.sup.2] were recorded,
compared to 22 flakes per [m.sup.2] on average that would have been
deposited on areas without tree canopy. These figures indicated that in
this study 96.7% of the applied flakes were retained in the canopy and
only 3.3% constituted canopy through-fall. There was only a minimal
chance that pheromone flakes would land in streams or cave openings
within the area studied.
Keywords: Gypsy moth, pheromone flakes, mating disruption
European gypsy moth (Lymantr& dispar) is not native to the
United States and lacks effective natural controls. The caterpillars
feed on the foliage of many host plants. Although oaks are the preferred
host species, the caterpillars defoliate many species of trees and
shrubs when oaks are not available (Liebhold et al. 1995). When high
numbers of gypsy moth caterpillars are present, forests may suffer
severe and repeated defoliation that can result in reduced tree growth,
branch dieback, and tree mortality. High numbers of caterpillars also
create a public nuisance and can affect human health (Sharov et al.
2002). The national strategy for managing gypsy moth includes
eradication in areas not yet generally infested and strategically
applied suppression in generally infested areas (Sharov et al. 2002).
Indiana Department of Natural Resources detected gypsy moth on the
Hoosier National Forest in 2006. Additional trapping in 2007 indicated
that there was a very low level incipient population in the treatment
area. Although individual gypsy moths were present in the area, they had
not yet become established. The optimum time to treat this potential
infestation was before gypsy moths became established and spread.
Without intervention, this population was expected to grow and
contribute to a faster rate of spread into other non-infested areas.
Mating disruption relies on the use of the gypsy moth sex pheromone,
disparlure. Female European gypsy moths are flightless and naturally
release this pheromone to attract males. The objective of the
application was to disrupt mating by saturating the treatment area with
enough pheromone sources to confuse male moths and thus prevent them
from finding and mating with female moths. Mating disruption using the
pheromone disparlure is considered specific to gypsy moth and is not
known to cause impacts on non-target organism populations, water
quality, microclimate, or soil productivity and fertility (USDA 2008,
Vol. III, Appendix H). The product Disrupt II, which incorporates
disparlure, has proven effective at eliminating gypsy moths at very low
The purpose of this cooperative project between the USDA Forest
Service and Indiana Department of Natural Resources was to eradicate the
gypsy moth by preventing moths in the treatment area from reproducing.
Although final flake location, whether in the forest canopy or
forest floor, does not affect the effectiveness of Disrupt II, some
persons had expressed interest in the possible deposition of flakes into
streams or cave openings. They were concerned the flakes would be eaten
by cave fauna causing undo stress in an already nutrient-poor
environment. The objective of this study was to estimate the amount of
pheromone flakes reaching the forest floor or potentially entering
streams or cave openings.
[FIGURE 1 OMITTED]
Study area.--The treatment area (748 ha.) covered both private land
(17%) and Hoosier National Forest System land (83%) in southeastern
Monroe County, Indiana (Fig. 1) and was in the Brown County Hills
Ecological Subsection (Zhalnin & Parker 2007). A 91 m buffer
surrounded the treatment area. Fifty-seven percent (425 ha) of the
treatment area was in the Charles C. Deam Wilderness on the Hoosier
National Forest. The Ecological Land Types were: Transect A, a broad
flat ridge; Transect B, a north-facing slope; and Transect C, the south
side of a narrow ridge (Zhalnin 2004).
The area around Transect A, an old field in 1939 (Jenkins &
Parker 2000), had an over story of poletimber-small sawtimber trees with
yellow-poplar (Liriodendron tulipifera L.), white oak (Quercus alba L.),
American beech (Fagus grandifolia Ehrh.), and bigtooth aspen (Populus
grandidentata Michx.). The understory consisted of pawpaw (Asimina
triloba L.), sugar maple (Acer saccharum Marsh.), and American beech.
The area around both Transects B and C was forested in 1939.
Sawtimber-size white oak, black oak, (Quercus velutina Lam.), sugar
maple, and yellow-poplar comprised the overstory at Transect B, and the
understory consisted of sugar maple and American beech. At Transect C
the overstory was dominated by large sawtimber white oak with a few
large sawtimber black oak. The understory was dominated by American
beech, flowering dogwood (Cornus florida L.) and sugar maple.
Data sampling.--Before treatment application, three transect
locations were generated randomly using ArcGIS. All of the transects
were in the Charles C. Deam Wilderness. The western point of each
east-west transect was located using a Garmin GPS unit and the generated
UTM coordinates. Twenty points, 3.048 m apart, were located along each
transect due east of the generated coordinates. Transects were
perpendicular to the flight lines.
[FIGURES 2-4 OMITTED]
To estimate overstory canopy above each point, digital images were
taken at approximately 66 cm above ground level (66 cm was as low as the
camera could be placed using a tripod). A tripod-mounted Sony Cybershot
DSC-S85 digital camera with the lens at wide angle was used. To position
the camera lens axis horizontally, we used bubble levels on the tripod
and one placed over the lens. The camera was oriented so that the long
dimension of the image ran north and south. Images were taken the
mornings of June 6 and 9, 2008. ERDAS Imagine software was used to
classify the images by percent as either sky or canopy (Fig. 2). Some
images were edited, using Microsoft Photo Editor, by changing colors to
those matching leaves when portions of boles or leaves were initially
classified as sky due to bright sunshine.
To monitor pheromone flake through-fall, we placed 50.8 x 76.2 cm
white foam core boards at the 20 points along each of the three
transects. The points marked in the initial layout were used as the
southwest corner of the boards. The long dimension was oriented north
and south. The foam boards were placed 5-13 cm above the ground the day
before application (Fig. 3) as level as possible. Hours after aerial
application of the flakes, the boards were inspected and each flake
location was marked on the board with a permanent marker. Flakes were
counted after the boards were returned to the office.
Treatment application.--The treatment consisted of one aerial
application of a mating disruptor called Disrupt II (Hercon
Environmental, Emigsville, Pennsylvania) on 23 June 2008--prior to the
emergence of male moths. Disrupt II is typically referred to as
pheromone flakes because it consists of 1 mm x 3 mm plastic-laminated
flakes with a layer of pheromone sandwiched between the outer layers of
plastic (Fig. 4). The flakes are coated with glue so they will adhere to
tree parts in the canopy.
GPS equipment recorded the plane's flight paths while applying
the pheromone flakes. The data was then exported to ArcGIS. GPS
locations were recorded for one end of each transect. A Trimble XT GPS
unit was used to locate more accurately the eastern point at Transect A
and the western point at Transects B and C. The UTM easting coordinate
for each point along the transect was developed by adding or subtracting
3.048 m to or from the previous point. We then determined the distance
from each point to the nearest flight line.
Statistical analysis.--Data were analyzed with analysis of variance
(ANOVA) and linear regression using SAS (SAS 2004). Duncan's
multiple-range test was used for means separation in the ANOVA.
RESULTS AND DISCUSSION
Transect B had a significantly higher percent canopy cover (90.7%)
(P < 0.05) than Transect A (89.3%) and Transect C (89.2%). Though
Transect B had significantly higher percent canopy cover, this
difference may not be important. Transect B percent canopy cover ranged
from 87.6-93.5%, Transect A ranged from 8Y8-92.6%, and Transect C ranged
from 87.2-94.4%. There was only one point on Transect A with 83.8%
canopy cover. The next lowest percent canopy cover on that transect was
86.4, very similar to Transects B and C. The fact that Transect A was an
old reverting field, compared to the continuously forested Transects B
and C, would explain the lower canopy coverage on some points for
Transect B recorded eight flakes on seven boards, the highest count
of the three transects. Transect C had six flakes on five boards and
Transect A had three flakes on three boards. Although percent canopy
cover had no significant influence on flakes falling on the boards
([R.sup.2] = 0.003), Transect A, a reverting old-field, may have had a
more compact and compressed canopy structure and that may have prevented
the flakes from falling to the forest floor. Also, because Transect B
was on a slope, there might have been gaps in the canopy caused by a
stair-stepping of the overstory canopy. This could have permitted more
flakes to filter through the canopy.
According to a study in Virginia, flakes tend to be distributed in
a non-uniform pattern (Thorpe et al. 2006). They tend to peak beneath
the pods or distribution points of the plane and bottom out under the
fuselage and wing tips of the plane. However, in our study on the
Hoosier National Forest, the distance from a flight line had no
significant influence on flakes falling on the boards ([R.sup.2] =
0.01). There were also insufficient flakes to determine a distribution
pattern comparable to the Virginia study.
At the dose used in this treatment (37.5 grams active ingredient
per hectare), an average of 22 flakes per [m.sup.2] would have been
deposited on areas without tree canopy (Thorpe et al. 2006). This
corresponds to an expected density just over 8 flakes per board for a
total of approximately 500 flakes for all 60 boards. However, only 17
flakes were actually recorded on the 60 boards. This corresponds to 0.73
flakes per [m.sup.2] or one flake for every 1.4 [m.sup.2]. These figures
indicate that in this study 96.7% of the applied flakes were retained in
the canopy and only 3.3% constituted canopy through-fall. Thus the
canopy intercepts the vast majority of flakes, keeping them from
reaching the forest floor.
The glue remained tacky; therefore, the chances were high that
pheromone flakes would stick to the first object encountered on the
forest floor. This greatly reduced the probability that a pheromone
flake would wash into a stream or enter a cave opening.
Though there was only a 3.3% through-fall of flakes at the time of
application, nearly 100% of the flakes would fall to the forest floor in
autumn with leaf fall. These flakes would become part of the forest
litter layer and not move oft site into streams or caves. During
application it was possible to halt the application of flakes when
flying over streams or other non-target areas, thus preventing flakes
from falling into streams not covered by tree canopy.
Therefore, with the small number of flakes falling through the
canopy to the forest floor, the glue remaining tacky, and the ability to
halt application of pheromone flakes over streams, there was only a
minimal chance that pheromone flakes would land in streams or cave
openings within the area studied.
Manuscript received 23 September 2008, revised 19 December 2008.
Jenkins, M.A. & G.R. Parker. 2000. Changes in the forest
landscape of the Charles C. Deam Wilderness, southern Indiana,
1939-1990. Natural Areas Journal 20(1):46-55.
Liebhold, A.M., K.W. Gottschalk, R. Muzika, M.E. Montgomery, R.
Young, K. O'Day & B. Kelley. 1995. Suitability of North
American tree species to the gypsy moth: A summary of field and
laboratory tests. USDA Forest Service, North eastern Forest Experiment
Station, General Technical Report, NE-211. 34 pp.
SAS Institute Inc. 2004. SAS OnlineDoc[R] 9.1.2. Cary, North
Carolina: SAS Institute Inc.
Sharov, A., D. Leonard, A.M. Liebhold, E.A. Roberts & W.
Dickerson. 2002. "Slow the Spread", A national program to
contain the gypsy moth. Journal of Forestry 100(5):30-35.
Thorpe, K., R. Reardon, K. Tcheslavskaia, D. Leonard & V.
Mastro. 2006. A review of the use of mating disruption to manage gypsy
moth, Lymantria dispar (L.). USDA, Forest Service, FHTET-2006-13.
U.S. Department of Agriculture. 2008. Gypsy moth management in the
United States: A cooperative approach. Draft Supplemental Environmental
Impact Statement, Vols. I-IV. USDA Forest Service and USDA APHIS.
Zhalnin, A.V. 2004. Delineation and spatial analysis of ecological
classification units for the Hoosier National Forest. Ph.D.
dissertation. Purdue University, West Lafayette, Indiana. 268 pp.
Zhalnin, A.V. & G.R. Parker. 2007. Land type association
delineation and spatial analysis for the Hoosier National Forest in
southern Indiana. Proceedings of the Indiana Academy of Science
Dale R. Weigel: Hoosier National Forest, 811 Constitution Avenue,
Bedford, Indiana 47421 USA
Todd Dempsey: Hoosier National Forest, 248 15th Street, Tell City,
Indiana 47586 USA
Correspondence: Dale R. Weigel, Hoosier National Forest, 811
Constitution Avenue, Bedford, IN 47421, Phone: 812/276-4774, FAX:
812/279-3423, Email: firstname.lastname@example.org