Biology and Behavior
Emerald Ash Borer: Research and Technology Development Meeting
From: V. Mastro and R. Reardon (compilers), Emerald Ash Borer Research and Technology Development Meeting, Romulus, Michigan, Oct 5-6, 2004. USDA Forest Service publication FHTET-2004-15.
Diane E. Brown-Rytlewski, IPM Program, Michigan State University, B 18 NFSTC, East Lansing, MI 48824 and Mary A. Wilson, Michigan State University Extension, G-4215 W. Pasadena Ave., Flint, MI 48504
Emerald ash borer (Agrilus planipennis) is a serious exotic pest of ash trees. Determining when emerald ash borer (EAB) adults first emerge, reach their emergence peak, and reach the end of emergence has important implications for its management. Determining when adult emergence begins and ends is critical for cutting infested ash trees and deciding when transport of the wood can safely be accomplished. It is also important to predict when adult emergence will begin to set monitoring traps out on a timely basis. Predicting when peak emergence will occur may be critical for timing certain pesticide applications. In this study, weekly observations of new emergence holes were recorded to track emergence from the trunk by direction (NW, NE, SW, and SE), track key stages of emergence, record corresponding degree day accumulations, and look at concurrent bloom stages of common landscape plants to develop easily observable benchmarks for key periods of adult EAB activity.
In 2003, adult EAB emergence was tracked weekly at a single site in Ann Arbor, Michigan, from early May until the end of August (sampling discontinued two weeks after the last detected adult emergence). Eleven trees were planted in parking lots and mulched with stone; nine were planted in lawns. In 2004, adult EAB emergence was tracked weekly at locations in Troy and Novi, Michigan, from early May until late September (sampling discontinued two weeks after the last detected adult emergence). All trees monitored in 2004 were planted in grassed parkways between the sidewalk and road. Degree-day accumulations and bloom stage of common landscape plants were recorded both years at the adult monitoring sites along with three other locations (gardens with large collections of ornamental plant materials) in Flint, East Lansing, and Novi, Michigan.Adult emergence was monitored by counting and marking all emergence holes found within two-foot sections of trunk (4-6 ft. and 10-12 ft) on each of twenty ash trees of approximately the same age at each location. Dataloggers (Watchdog model 400, Spectrum Technologies, Plainfield, Illinois) were used to record hourly temperature readings. The modified sine-wave method (Baskerville-Emmons) was used to calculate degree-days, using a base temperature of 50º F.
The 2003 data were analyzed as a split-split plot; for each tree a whole plot and high/low locations on the trunk and direction of emergence holes (NE, NW, SE, or SW) as within tree measurements. There were significant differences in emergence by direction on the trunk. Greater emergence of adults occurred on the sunniest exposures of the trees (SW and SE), and earlier emergence on trees mulched with stone and planted in parking lots. There was greater emergence of EAB adults high on the trunk than low on the trunk for trees planted in grass. Adult emergence began sometime between June 5-13, 2003, within the degree-day range of 471-584 GDD base 50ºF; peak was between June 13-19, 2003 (584-705 GDD base 50ºF); date of last recorded emergence was August 16, 2003 (2083 GDD base 50ºF). Adult emergence continued for a ten-week period.
The data from 2004 have not yet been analyzed, but it appears that direction of emergence may not be as significant this year. First emergence at the Novi site began sometime between May 11 and May 18, 2004 (348- 463 GDD base 50ºF), and emergence peaked between June 1- June 8, 2004 (572-759 GDD base 50ºF). A second emergence period began sometime between August 17-24 and continued until sometime between September 2-7, 2004 (2089-2230 GDD base 50ºF). Emergence from the SW and SE quadrants occurred several weeks earlier than emergence from the NW and NE quadrants. At the Troy site, first emergence occurred sometime between May 25-June1, 2004 (444-518 GDD base 50ºF), and peak emergence between June 22-29 (894-1027 GDD base 50ºF). A second emergence period began sometime between August 17-24 and continued until sometime between September 2-7, 2004 (1948-2052 GDD base 50ºF). Adult emergence continued for a seventeen-week period. The significance of the second emergence period is not yet known.
It appears that black locust (Robinia pseudoacacia) from early to late bloom and doublefile viburnum (Viburnum plicatum tomentosum) from full to late bloom may be good indicators for first emergence of adults. Japanese tree lilac (Syringa reticulata) from first to full bloom may be a good indicator for peak emergence, and purple coneflower (Echinacea purpurea) at late bloom and Joe-Pye weed (Eupatorium purpureum) at full to late bloom may be good indicators for the end of adult emergence. Plans are to continue monitoring emergence of adults, degree day accumulations and bloom times of selected landscape plants in 2005.
Robin A. J. Taylor1, Leah S. Bauer2, Deborah L. Miller2 and Robert A. Haack2
1Department of Entomology, OARDC/OSU, 1680 Madison Ave., Wooster, OH 44691
The emerald ash borer (EAB), Agrilus planipennis Fairmaire (Coleoptera: Buprestidae), is an invasive pest of ash trees (Fraxinus spp.) that is rapidly spreading from the probable introduction site in Detroit, Michigan. The rapid spread to areas outside Michigan is undoubtedly due to phoretic transport on nursery stock, logs, and firewood. However, not all the range expansion can be attributed to human agency. Despite attempts to contain the core infestation to the counties surrounding Detroit and Essex County, Ontario, EAB range has continued to expand. This is due in part to the natural dispersal of EAB. Failure to understand the natural dispersal will impede attempts contain and control EAB; knowledge of flight behavior and physiology is needed to estimate dispersal capabilities in order to develop effective containment strategies.
A cooperative research venture between The Ohio State University and USDA-Forest Service is using computer-monitored flight mills with tethered EAB adults to measure flight speed, duration, and periodicity. Preliminary results from 28 adults, flying without rest, food, or water, showed that about half of the tethered beetles flew >50 m, while one 3-day old male flew a total of 5.2 km in 40 hrs. Subsequent data have confirmed the maximum flight speed as 1.5 m/sec (3.5 mph) which occurs in bouts of about 1 min each. The individual that flew thefurthest in 24 hrs started with 70 sec flight bouts followed by an idle periods of about 130 sec. After about 2 hr, the idle time increased, rising to about 20 min at 24 hr. Although the detailed bout patterns differ between individuals, this overall pattern appears to be the norm. Bigger differences are observed in the length of time spent flying. In particular, females flew twice as far as males in 24 hr (P < 0.002) and mated females flew twice as far as unmated females (P < 0.0001). The average distance flown in 24 hrs by mated females was 1.7 km. The frequency distribution of distance flown by all females in 24 hrs is skewed to the right (mode = 800 m, median = 1 km, mean = 1.7 km, 20 percent flew >2km, 1 percent flew > 4km).
The discovery that mated females fly longer, farther, and faster than either males or unmated females is rather alarming as it suggests females are programmed to make a dispersal flight. The absence of a correlation (R2 = 0.007) between distance flown and size (mg) of mated females suggests there are no other distinct classes of migrants.
A simple random walk model suggests that ~20 percent of mated females are displaced >250 m while flying 2 km; ~1 percent are displaced ~500m while flying 4 km. The random walk assumption is probably optimistic; the flight is probably less random, which means that these are underestimates of the actual displacement of gravid females in their dispersal flight. In order to determine how significant this is for control and containment efforts, we need to know how directional the flights actually are and how receptive gravid females are to cues from ash trees for stopping their dispersal flight to settle.
Alicia M. Bray 1, Leah S. Bauer 1,2, Robert A. Haack 1,2, and James J. Smith 1,3
1 Department of Entomology, Michigan State University, 243 Natural Science Building, East Lansing, MI 48824
Emerald ash borer (EAB) was first detected in Michigan and Canada in 2002. Efforts to eradicate this destructive pest by federal and state regulatory agencies continue. Knowledge of EAB genetics will be useful in understanding the invasion dynamics of the beetle and to help identify geographic localities of potential biocontrol agents. Genetic techniques, such as mtDNA gene sequencing and amplified fragment length polymorphisms (AFLP) will help determine the geographic origin of EAB in its native range throughout eastern Asia.
In an initial analysis, we collected EAB individuals from several localities in Michigan and three populations in China. Analysis of mtDNA cytochrome oxidase subunit I (COI) sequences from 20 individuals from Michigan, three individuals from Dagong (Tianjin City), one individual from Hangu (Tianjin City), and three individuals from Harbin (Heilongjiang Province) indicated that all COI sequences (~500 nucleotides) were identical. However, differences between individuals were observed using AFLPs. AFLP analysis using three primer pairs yielded fingerprints from EAB individuals from Michigan (19), Dagong (2) and Hangu (1) (Tianjin City), and Harbin (4) (Heilongjiang Province). Eighty-two scoreable bands, coded as binary characters (presence/absence), were analyzed in a neighbor-joining (NJ) analysis. The NJ tree showed that individuals from MI cluster with individuals from Dagong and Hangu (Tianjin City), while EAB individuals from Harbin (Heilongjiang Province) fell into a separate, more distantly related group. Therefore, with this limited sample, AFLP appears to reveal population-level differences between EAB populations, and the Michigan populations appear more closely related to EAB from Tianjin Province than to EAB from Heilongjiang Province. Nonetheless, more thorough sampling in China is necessary to better characterize the relationships of the Michigan and Chinese EAB populations. Due to the rarity of EAB in Korea, Japan, Mongolia, Taiwan, and Russia, no samples have yet been found for genetics. We now plan more intensive sampling of EAB populations within Michigan to provide information on invasion genetics and possible age of infestation. We also plan more extensive sampling in North America including EAB from Ohio, Maryland, Virginia, Indiana, and Ontario, Canada, to determine if there was a single or multiple introductions of EAB into North America. Overall, mtDNA sequences appear to be a good positive control that ensures all individuals in our analyses are indeed EAB. AFLP fingerprints detected differences between EAB populations, important when locating potential biological control agents.
David Cappaert1, Deborah McCullough1, and Therese Poland2
1 Department of Entomology, Michigan State University, 243 Natural Science Building, East Lansing, MI 48824
To establish the life cycle of EAB was one of the first objectives of EAB research. Our expectation was that Agrilus planipennis phenology would coincide roughly with that of well-charac-terized Agrilus species such as A. anxius and A. bilineatus: 1) mid-summer adult flight and oviposition; 2) complete four stages of larval development by fall; 3) non-feeding prepupal stage overwinter; and 4) pupation in late spring. Evidence in support of this assumption includes observations demonstrating synchronous pupation and adult flight and summer/fall dissection series showing steady progression of larval size. However, several anomalies raised questions about the universality of a synchronous, one-year cycle. At some locations, a majority of larvae failed to complete feeding in the fall. Most conspicuously, winter/spring dissections of very lightly infested trees at outlier sites revealed mostly 2nd and 3rd stage larvae. A dissection series in spring 2004 confirmed that small winter/spring larvae did not complete development before mid-summer, and failed to form pupae during the annual “window” for that life stage.
Several lines of evidence now demonstrate that some fraction of EAB requires two years for development:
1. In a series of dissections of lightly-to-moderately infested trees during spring/summer 2004, we found that 2nd and 3rd stage larvae (oviposited the previous summer) present in April did not complete larval development before the summer pupation window. Prepupae formed by these larvae in late summer had not resulted in adult emergence by October.
2. A 2004 experiment compared larval density between unsprayed trees or trees treated with a Tempo cover spray. Despite bioassay evidence indicating excellent coverage and persistence of toxin, there was no significant effect of the treatment. Subsequent re-examination revealed that the poor performance of Tempo was attributable to protected prepupae present in the trees before the May treatments began.
3. Examination of larval galleries clearly demonstrates two-year development. In the Tempo study and other dissections conducted in fall of 2004, galleries of mature larvae were of two types: a continuous track contained entirely within 2004 growth tissue (1- or 2-year larvae), or a two-stage track beginning in (now) dead wood overgrown by 2003 tissue and concluding in a final tunnel through 2004 growth (2-year larvae).
We are not yet clear on the mechanisms that determine the proportions of one and two-year EAB. Clearly, the seasonal temperature profile may be important: we know that many insects including other Agrilus species have prolonged development where temperatures are lower. However, our data suggest host condition may be key: proportions of two-year larvae tend to be higher in lightly infested trees.
The occurrence of two-year larvae has many implications for research, management, and the containment/eradication effort. The likelihood that trees have some degree of resistance (inhibiting larval development) suggests that there may be opportunities for enhancing resistance via breeding or chemical treatment. The presence of mature larvae in the spring/ summer creates a new (and more difficult) target for pesticide applications. Two-year larvae at outlier sites may mean a delay in detection of initial outbreak or resurgence from outliers; on the plus side, dissemination from an outlier will be slowed. Further research will focus on determining the prevalence and underlying mechanism of the two-year phenomenon.
Reconstructing the Temporal and Spatial Dynamics of Emerald Ash Borer in Black Ash: A Case Study of an Outer Site in Roscommon County, Michigan
Nathan W. Siegert1, Deborah G. McCullough1, Andrew M. Liebhold2, and Frank W. Telewski3
1 Departments of Entomology &: Forestry, Michigan State University, 243 Natural Science Bldg., East Lansing, MI 48824-1115
The temporal and spatial dynamics of emerald ash borer (EAB), Agrilus planipennis Fairmaire (Coleoptera: Buprestidae), in an outlier site in Roscommon County, Michigan, were reconstructed using dendrochronological analyses. The site was characterized by pockets of black ash, Fraxinus nigra Marsh., located in swampy areas surrounded by ash-free, higher terrain consisting mainly of oaks (Quercus spp.) and pines (Pinus spp.). Ash eradication operations were underway during the summer of 2004 within an elongated eradication zone delimited by Michigan Department of Agriculture personnel. Within the eradication zone were two main swampy areas that were separated by a distance of more than a half mile. Thirty black ash trees, ranging in vitality from ‘apparently healthy’ to ‘declining’ (i.e., reduced leaf size and canopy dieback) to ‘dead’, in the two main swampy areas were selected, partially debarked, and cored in late June to early July 2004. Signs of EAB presence were evident on all dead and declining ash trees sampled. Increment cores were prepared using standard dendrochronological techniques and crossdated using skeleton-plots and verified using COFECHA software. Crossdating analyses indicated that trees began to die in 2001 in one of the swampy areas. Preliminary results suggest that three trees in close proximity were initially infested with EAB and the infestation radiated out from that point in subsequent years. In the other swampy area, EAB infestations did not begin causing tree mortality until 2003. Phenological development of EAB in the lower portions of the trees was considerably less advanced than in the upper portions of the trees. Dendrochronological examination of wood growth during gallery formation indicated that, under certain conditions, successful EAB development from egg to adult can be extended over multiple years (e.g., early instars present in 2002 and adult emergence in 2004). Additional dendrochronological analyses are in progress to determine when EAB initially infested the sample trees. Implications of this research were discussed in relation to future management guidelines.
John A. Witter1 and Andrew J. Storer2
1 School of Natural Resources and Environment, University of Michigan, 430 E. University, Ann Arbor, MI 48109-1115
Two ash monitoring systems are being established throughout rural forests and recreational areas of Michigan to address three objectives:
The Rural Ash Monitoring Plot System (RAMPS) was initiated during Summer 2004 with the establishment of 160 plots along five gradients running through the state of Michigan. As of October 2004, we have established approximately 65 percent of the plots along the Upper Peninsula (U.P.) gradient, 50 percent of the plots along the southern Lower Peninsula (L.P.) gradient, and portions of both the northern and eastern L.P. gradients. A western L.P. gradient will also be established. To date, approximately 30 percent of the plots fall within the quarantine area and contain EAB. We expect to establish 240 additional plots by the end of Summer 2005, for a total of 400 rural forest plots in this system. At these plots, variables measured include stand age, soil texture, live and dead basal area by species, crown variables (live crown ratio, light exposure, transparency, density, and dieback), tree vigor, presence of EAB, and other types of tree damage. During 2004 in the L.P., the percent of dead ash varied from 0–100 percent by plot. When plot data were pooled by county, examples of the average percent of dead ash in 2004 are: Arenac County—1 percent, Midland County—3 percent, Gladwin County—5 percent, Lapeer County—6 percent, Cheboygan County—7 percent, Oakland County—15 percent, Washtenaw County—31 percent, and Wayne County—61 percent.
The second ash monitoring system examines ash at over 250 sites in Lower Michigan that are in or near recreational areas (parks, picnic areas, rest areas, boat landings, campsites, etc.). A variety of tree health variables are measured at these sites, along with recording presence of EAB and other types of tree damage.
The major results from Summer 2003 are:
Data for recreational plots from Summer 2004 are currently being analyzed.