Fall Cankerworm (Lepidoptera: Geometridae), a Native Defoliator of Broadleaved Trees and Shrubs in North America

The fall cankerworm, Alsophila pometaria (Harris), is a species endemic to North America that feeds on broadleaf trees and shrubs. Fall cankerworm is a generalist folivore and larvae feed on a wide range of deciduous trees and smaller woody plants. Common hosts include oak, maple, cherry, ash, apple, beech, and birch. This pest is prone to repeated outbreaks over large areas, can cause extensive defoliation throughout hardwood forest stands, and can be a public nuisance in developed or highly populated areas. Fall cankerworm defoliation can lead to reduced tree health and impact ecosystem function, carbon sequestration, wildlife habitat, and temperature regulation, especially in urban areas. Elevated populations often occur in areas where host trees are stressed or a high density of a preferred host species is present. Fall cankerworm management is often necessary due to their impacts on the local tree canopy and nuisance to the public. Tree banding and the use of the biological insecticide Bacillus thuringiensis (Bt) are the common management tactics used to reduce fall cankerworm populations. Here we review fall cankerworm distribution, life stages, host plants, damage, scouting and sampling procedures, management options, and discuss commonly associated and co-occurring defoliator species.


Description of Life Stages
Eclosion, or the emergence of the adult moth from its pupal case, is triggered when temperatures drop to freezing in late fall or early winter, and adults surface from the soil (Appleby et al. 1975). Males generally surface before females and can be seen in flight on warm days (Harriman and Day 2016). Adult males are gray with a 25-35 mm wingspan. Male forewings are glossy brown and mottled with dark brown and light gray scales, and hind wings are light grayish-brown (Fig. 1a). Some adult males have jagged white lines that form two bands across the forewings (Harris 1841, Hinds 1901, Porter and Alden 1924. Adult females surface without functional wings or mouthparts and are 8-12 mm in length, with dark and pale gray glossy, mottled coloring (Fig. 1b). Females are stout bodied and uniform in diameter, aside from a bluntly tapered abdomen. After emergence, they climb the nearest tree or shrub and release an airborne chemical signal (i.e., sex pheromone) to attract flying males. Males track the pheromone to its source and the pair mates. After mating, the wingless females climb high into the crown of host plants and lay egg masses on the bark of smaller twigs and branches from October through January (Palaniswamy et al. 1986;Fig. 1c).
Adult fall cankerworms are short-lived, usually only surviving for a few weeks. While male fall cankerworms are winged, their sole purpose is to find the female moth and mate, and they die once this task is complete. The mottled, gray-brown coloring of both male and female adults allows for excellent camouflage against tree bark (Harris 1841, Hinds 1901, Porter and Alden 1924Fig. 1d). Because females lack functional mouthparts, they cannot feed and die shortly after ovipositing. Several theories pertain to the origin of female flightlessness, or aptery, all of which suggest this adaptation would aid in increasing the species' fitness. Some suggest that the female fall cankerworm shifts bodily resources from wing and associated flight muscle development and redirects these resources to the production of eggs to increase reproductive potential (Schneider 1980). Because fall cankerworm can feed on a variety of host species, female flight is not as critical to search for feeding and oviposition sites (Barbosa et al. 1989). Alternately, female aptery may be an adaptation to increase synchrony between egg hatch and leaf expansion (Futuyma et al. 1984). Female fall cankerworms can also reproduce parthenogenetically, wherein clones specialize on different tree species as hosts (Futuyma et al. 1981). Clones differ in hatching time to feed on different host trees to limit competition at the time of foliation (Schneider 1979).
Females lay up to 200 eggs in compact masses, arranged in uniform rows in a single layer on smaller twigs and branches (Fig. 1e). The eggs are adhered to each other and to the host tree, creating protective, waterproof clusters (Harris 1841). Eggs are barrel shaped in appearance, with a rounded base and flat top, each about 0.7 mm long. The base is 0.45 mm in diameter and is slightly narrower than the top. Eggs are gray and have a dark brown line around the top rim with a tiny, dark depression on top of each egg (Hinds 1901, Porter andAlden 1924). Fall cankerworm eggs overwinter on host trees for about 3-4 mo. Egg hatch in the spring coincides with bud break and leaf emergence, usually over an interval of 7-10 d from late March to early May. Upon hatching, larvae chew their way out through the flattened top of the egg (Harris 1841, Hinds 1901, Porter and Alden 1924. Within individual egg clusters, most of the hatching is completed within 2 or 3 d after the first egg hatches (Harris 1841, Porter and Alden 1924, Talerico 1971, Futuyma et al. 1984, Drooz 1985. Fall cankerworm larvae are slender and the coloring is highly variable, ranging from light green to dark brown (Harris 1841, Hinds 1901, Porter and Alden 1924Fig. 1f and g). Color morphs are associated with population density, with darker color morphs developing when larvae are exposed to high densities during the early instar phase of development (Schneider 1979). Hinds (1901) described newly hatched fall cankerworm larvae (also known as neonates or first instars) as 1.5 mm long, pale yellow-green to olivegreen, with a cylindrical body that tapers to the tip of the abdomen and has an average diameter of 0.3 mm. Neonates have prolegs on abdominal segments 7 and 10, with a pair of vestigial prolegs (false legs) upon the sixth abdominal segment. Second instars are about 7-8 mm in length, and segments are more distinct. The coloring varies from pale yellow-green to dark green. A dark dorsal stripe begins to appear as second instars mature. On each side of this median stripe is a pair of light, narrow, longitudinal lines, and below the spiracles an irregular light line appears. The pair of vestigial prolegs upon the sixth abdominal segment becomes more prominent as the larvae mature. Third instars are 12-14 mm in length, and both color morphs become more distinct. Light green larvae have two white lines running down the side of the body from the head to the tip of the abdomen while the darker larvae have a black stripe the length of their back accompanying the two distinct light stripes on their sides (Hinds 1901, Porter andAlden 1924). Third and fourth instars have three developed pairs of legs on their thorax, two vestigial prolegs on their abdomen, and a lack of prolegs in the middle of the body. Thus, larvae must draw the hind end up to the thoracic legs to form an arch or loop, and then extend the body to propel them forward. This movement appears as though they 'inch' along as they crawl, thus the common name of inchworms or loopers (Harris 1841). Fourth instars grow from 18 mm to 25 mm in length by the time they reach maturity. Coloring remains highly variable and larvae maintain body markings as described in previous instars (Hinds 1901, Porter andAlden 1924).
To enter the pupal stage, larvae burrow just under the soil surface and spin a cocoon, which is completed in 24-48 h (Hinds 1901, Porter andAlden 1924). Pupae can be found 50-100 mm beneath the surface of the ground and cocoons are tough, ellipsoidal, 10-13 mm in length and made of silk and soil particles. Pupae are yellow to yellow-green when first formed and darken to a reddishbrown with age (Hinds 1901, Porter andAlden 1924;Fig. 1h). The pupation stage is the longest part of the fall cankerworm life cycle, as the pupa remains inside the cocoon for the remainder of the spring, summer, and early fall, usually about 7 mo (Appleby et al. 1975).
Fall cankerworm is univoltine throughout its range (Asaro and Chamberlin 2015). This insect is called fall cankerworm because females emerge from the soil in the fall to oviposit on twigs of host plants. However, larvae are spring defoliators and cause peak defoliation from April to June. Neonates can also produce silken threads and use them to disperse through the wind by 'ballooning' from tree to tree (Harris 1841, Futuyma et al. 1981. Larvae mature through five successive instars over the course of 5-6 wk (Ciesla and Asaro 2013). By mid-to late June, the larvae reach maturity and lower themselves to the ground on silken threads to pupate (Asaro and Chamberlin 2015).

Damage
Fall cankerworm larvae are considered a serious pest of many tree species, in part because they can cause major defoliation of the new spring growth, which is significantly more detrimental to tree health than late-season defoliation (Coulson and Witter 1984). Damage is typically first observed in the spring when neonates begin feeding. Once bud break occurs, cankerworm larvae target the opening buds and feed on the new leaf growth. Young larvae first chew small holes through the leaf surface. As the larvae develop and continue to feed, the holes in the leaf surface become larger and more irregular (Fig. 2a). Eventually, larvae skeletonize new leaves by feeding in the areas in between small veins. Mature larvae consume the entire leaf, leaving only the midrib and major veins (Harris 1841, Hoover 2010Fig. 2b).
Fall cankerworm is an eruptive outbreak pest, and years of defoliation can be followed by many years without any significant sightings or damage (Asaro and Chamberlin 2015). Most outbreaks are relatively localized and short in duration. Typically, fall cankerworm populations crash after 1 or 2 yr of severe defoliation due to natural factors, including natural enemies, disease, limited food, or climate (Sherman 1921, Raushchenberger and Talerico 1967, Kulman 1971, Fedde et al. 1973, Fedde 1980. Light infestations are typically of little consequence, as host plants can re-foliate later in the growing season and continue to photosynthesize. However, if the infestation is severe, complete tree defoliation may occur (Ciesla and Asaro 2013). When host plants experience several consecutive years of severe defoliation, the host plant must continually tap into stored carbohydrate reserves to re-foliate, which become depleted and can lead to branch dieback, reduced growth rate, and potentially death when compounded with other stressors such as drought, poor site conditions, disease, age, and other pests (Kulman 1971, Houston et al. 1981. Young, newly planted, or weakened host plants are especially vulnerable. In these cases, fall cankerworm defoliation can lead to widespread mortality and disruption of forest composition (Coulson andWitter 1984, Asaro andChamberlin 2015).
The fall cankerworm is a highly adaptable species, and populations can occur on an array of hosts in a variety of natural and managed settings. Peak infestation and defoliation levels tend to occur in urban areas (Frank 2014, Noukoun et al. 2014, Walter et al. 2016 or in mature, unmanaged stands where other stressors like drought and high winds are common (Harris 1841, Porter and Alden 1924, Barber and Marquis 2009, Asaro and Chamberlin 2015. In both forested and urban settings, higher levels of cankerworm herbivory are often found in areas lacking host species diversity. When only one or two host species dominate the landscape, fall cankerworm populations can quickly build and spill from preferred plant hosts onto neighboring plants to sustain outbreaks (Dale and Frank 2014). For instance, outbreaks have historically been recorded in oak-dominated forests and managed monocultural settings where host trees are stressed and a high density of a preferred host species is present. In several cases, high fall cankerworm populations occurred on apple trees in neglected orchard settings in Massachusetts, California, and Nova Scotia (Porter andAlden 1924, Patterson 1966). These infestations were likely due to an abundance of host material in the absence of pruning and insecticide treatments. Host trees that sustain prolonged wind and direct sun exposure (i.e., host trees that are stressed) are also susceptible to fall cankerworm outbreaks, as elevated populations have been documented in several counties surrounding Lake Michigan while feeding on American linden (Tilia americana (L.) (Malvales: Malvaceae)), American elm (Ulmus americana (L.) (Rosales: Ulmaceae)), and oak and are often found in shelterbelts and windbreak plantings (Appleby et al. 1975, Hard et al. 1979. Above all, fall cankerworm is known for large, severe, recurring outbreaks in oak-dominated forested landscapes (Noukoun et al. 2014). In fact, fall cankerworm outbreaks are more common than any other forest insect in Virginia and have been responsible for more defoliated acres of oak-dominated forests there than Lymantria dispar dispar L. (Lepidoptera: Erebidae) (Asaro and Chamberlin 2015). Other common fall cankerworm hosts in oakdominated forests include hardwood trees such as elm, beech, linden, and maple (Asaro 2013).
High populations of fall cankerworm are also commonly recorded in urban areas where there is an abundance of a preferred and/or mature host species (Frank 2014, Noukoun et al. 2014, Walter et al. 2016). Recurring outbreaks have caused severe defoliation over vast acreages throughout urban settings in central North Carolina, northern Georgia, and in northern and southern Virginia. These are likely due to the large number of mature preferred host trees, many of which are stressed by age and site characteristics associated with urban landscapes (i.e., compacted soils and paved streets) (Harris 1841, Porter and Alden 1924, Ciesla and Asaro 2013. Heavy infestations can also be a public disturbance if they occur in urban municipalities or highly used public areas. If larvae are disturbed by wind, need to escape an enemy, or find new food, people may encounter larvae spinning to the ground on their silken threads in high numbers (Appleby et al. 1975). As larvae seek places to pupate, large populations can accumulate on fence posts and sides of buildings, and frass commonly rains down on people and vehicles below host trees that have high larval populations (Ciesla and Asaro 2013).
In recent decades, outbreaks of fall cankerworm have increased in urban and suburban areas of the mid-Atlantic. The city of Charlotte, NC, and surrounding counties have experienced fall cankerworm outbreaks almost annually for the last 30 yr. It is not known why Charlotte, specifically, is having such chronic problems with the fall cankerworm, but some have speculated it is due to the large population of mature willow oaks (Quercus phellos (L.) (Fagales: Fagaceae)) in many neighborhoods that otherwise lack vegetative complexity (Ciesla and Asaro 2013). Willow oaks are highly susceptible to fall cankerworm infestation, and the limited species diversity allows populations to flourish, unchecked by natural factors that would exist in a forested environment (Dale and Frank 2014). Fedde et al. (1973) reported other notable outbreaks on over 400 acres of land in urban areas throughout Fauquier County, VA from 1967 to 1972, and recurring populations of fall cankerworm persist in several cities in northern Virginia, where they feed on a variety of trees including maple, hickory, ash, and oak (Asaro andChamberlin 2015, Walter et al. 2016).
Though populations have increased in recent years, fall cankerworm outbreaks are a not a new phenomenon. The first recorded outbreak occurred in Virginia in 1661 (Porter and Alden 1924) and another outbreak was reported in Massachusetts in 1793 (Porter and Alden 1924). Isolated populations were reported from Colorado and California in the early 1900s, and outbreaks were prevalent throughout the state of Pennsylvania from 1932 to 1977 (Hoover 2010). Fall cankerworm was first observed in western North Carolina in 1969, and extensive areas of defoliation occurred in higher elevation mixed broadleaf forests in the mountains of western North Carolina and northwestern Georgia between 1978 and 1980 (Swank et al. 1981, Ciesla andAsaro 2013). Wallner (1971) reported periodic outbreaks on Michigan hardwoods, and Barber and Marquis (2009)

Sampling or Scouting Procedures
Forest and tree managers commonly monitor fall cankerworm populations to determine outbreak trends and hot spots (Waters and Stark 1980). Sticky band traps (commonly referred to as 'tree bands', 'barrier bands', 'bug barrier tree bands', or 'sticky traps') are typically used for monitoring purposes, and to establish 'defoliation thresholds'. Defoliation thresholds relate the number of female cankerworms captured to defoliation severity and are used as a decision-making tool to help landowners determine if management is necessary, and the appropriate management timing to reduce tree mortality (Noukoun et al. 2014, Walter et al. 2016. Defoliation thresholds vary between urban and suburban areas and large forested tracts, and defoliation levels can be influenced by weather, host distribution, natural enemy populations, and other environmental factors. It is important to consider ecological context when choosing a management strategy (Ciesla andAsaro 2013, Walter et al. 2016).
Other predators including spiders, birds, and small rodents attack all fall cankerworm life stages (Sherman 1921, Ciesla andAsaro 2013). Cold winter temperatures and disease may also decrease larval populations, and late spring frosts can kill newly emerging host tree foliage and lead to larval starvation. Higher plant diversity and presence of natural enemies usually keep outbreaks under control within a few years. However, in some areas, outbreak status is consistent and natural factors do not cause populations to decline. If outbreaks occur many years in succession, they can negatively impact forest health, ecosystem services, carbon sequestration, wildlife habitat, and temperature regulation (Gottschalk 1993, Asaro andChamberlin 2015). Additionally, outbreaks in recreation sites or urban areas may require direct control, especially when they take place over several successive years (Walter et al. 2016). As such, several management strategies have been explored to control fall cankerworm populations.

Chemical Control
Cankerworm control can be achieved by treating the affected host trees with a registered insecticide soon after the larvae first become active and after the leaves have sufficiently expanded for coverage by the spray. Materials registered to control cankerworms include carbaryl, acephate, fluvalinate, bifenthrin, naled, methoxyfenozide, acetamiprid, fenpropathrin, and tebufenozide, and the biological insecticide, Bacillus thuringiensis (Berliner) (Bacillales: Bacillaceae) (Bt).. While most effective pesticides are broad spectrum and kill many kinds of arthropods, including beneficial ones, Bt is a naturally occurring bacteria specific to certain orders of insect pests and does not harm other types of insects, wildlife, or humans (Jouzani et al. 2017). The Bt subspecies Bacillus thuringiensis var. kurstaki (Btk) is specific to lepidopteran species. For these reasons, Btk is the most common insecticide used for controlling cankerworm outbreaks (Asaro and Chamberlin 2015). Bacillus thuringiensis formulations can be sprayed aerially or on the ground by a licensed pesticide applicator, but application timing is important. For optimal effectiveness, Bt must be sprayed when larvae are small (less than two centimeters in length), usually around the second week of feeding in early spring; as such, populations must be monitored closely (Porter and Alden 1924). At this time, the damage to the tree will be minor and leaves will have fully expanded-an important point because fall cankerworm larvae must ingest the Bt spores directly for mortality to occur. For this reason, the Bt spores must land directly on the leaf surface (Wallner 1971). Once consumed, the spores disrupt the gut, leading to infection and eventually death (Raymond et al. 2007, Broderick et al. 2009). Read and follow all instructions and safety precautions on labels. Contact local agriculture extension agents, the USDA Forest Service, or state forestry agencies for the most up-to-date information on currently registered and effective insecticides, dosage rates, and method of application.

Mechanical Control
Landscape trees can be banded with sticky material to intercept adult females as they climb, preventing oviposition and therefore reducing larval abundance and defoliation. This method does not involve insecticide use and can be especially effective if used widely and bands are maintained (Noukoun et al. 2014). Two types of sticky band traps are commonly used: sticky traps and barrier bands (LaFrance and Westwood 2006; Fig. 3a and b). Sticky traps typically comprised three components. The first component is a strip of batting, fiberglass insulation, or a similar compressible material, which is installed around the target tree at breast height to fill bark crevices that would otherwise allow moths to crawl under the band. This batting is covered with a sheet of tar paper, roofing felt, or plastic wrap, which is then coated with Tanglefoot, Web-cote, or another similar sticky material. These sticky traps physically trap fall cankerworm females as they attempt to crawl up the host tree after emergence (Noukoun et al. 2014). Barrier bands use a similar form of physical control, but the sticky material is applied on the underside of a plastic barrier, which is installed to cover a layer of cotton batting or fiberglass insulation. The adult female sticks to the sticky material on the inside of the plastic barrier, and the trap is less likely to capture leaves, non-target insects, and other debris (LaFrance and Westwood 2006).
Installation timing is important for both types of tree bands. Bands should be put in place after host trees defoliate, but before adult female cankerworms emerge and begin crawling up host tree trunks to mate. Typically, bands should be installed just after the first hard freeze and removed by April to avoid bycatch of non-target organisms. For best results, bands must be regularly maintained to remain sticky and clear of debris (Otvos andHunt 1986, Asaro andChamberlin 2015). Sticky bands and barrier bands are not practical as a method of control for a widespread outbreak in a forest stand, due to the amount of maintenance they require to remain functional.

Silvicultural Control
Silvicultural management strategies are largely ineffective with such an adaptable generalist herbivore. However, outbreak history indicates a preponderance of mature oaks favors population explosions. When managing for any forest and tree pest, cultivating mixedspecies and mixed-age stands and avoiding overstocking and high grading (removal of only the most valuable timber) are beneficial strategies for maintaining forest health (Gottschalk 1993, Liebhold Downloaded from https://academic.oup.com/jipm/article/12/1/23/6287317 by guest on 01 September 2021et al. 1997, Davidson et al. 1999. Reducing overall tree stress and promoting diverse urban canopies will help reduce potential damage from fall cankerworm (Gottschalk 1993).

Associated Defoliator Species
Eruptive fall cankerworm outbreaks often spatially and temporally coincide with outbreaks of several other defoliator species (Ciesla and Asaro 2013). Between 1981 and 1984, a 'looper complex' defoliated over 0.65 million ha in eastern West Virginia (Butler 1985, Etgen andHicks 1987). This complex was dominated by Phigalia titea (Cramer) (Lepidoptera: Geometridae), but also included the linden looper, Erannis tiliaria (Harris) (Lepidoptera: Geometridae), and Apocheima strigataria (Minot) (Lepidoptera: Geometridae) along with fall cankerworm (Butler 1985;Fig. 4a-c). Another large-scale outbreak of multiple spring defoliators occurred in Virginia from 1981 to 1982. This complex was predominantly comprised of fall cankerworm, but also included P. titea and linden looper (Asaro and Chamberlin 2015; Fig. 4a and b). Similarly, a fall cankerworm outbreak in eastern Virginia in 2012 also featured a complex of other native defoliators that contributed to the defoliation in this area. These species included spring cankerworm, Paleacrita vernata (Peck) (Lepidoptera: Geometridae), elm spanworm, Ennomos subsignaria (Hubner) (Lepidoptera: Geometridae), and forest tent caterpillar, Malacosoma disstria (Hubner) (Lepidoptera: Lasiocampidae) (Ciesla and Asaro 2013;Fig. 4d-f). Winter moth, Operophtera brumata (L.) (Lepidoptera: Geometridae), Bruce spanworm, Operophtera bruceata (Hulst) (Lepidoptera: Geometridae), and the invasive L. dispar dispar can also be Fig. 3. Two common types of sticky band traps are sticky traps (A) and barrier bands (B). On sticky traps, the sticky material is applied to the outside of the apparatus. On barrier bands, the sticky material is applied to the inside of the plastic barrier. In this case, adult female cankerworms get caught between the compressible material and sticky material under the plastic barrier when they climb the host tree. associated with fall cankerworm outbreaks (Roland and Embree 1995;Fig. 4g-i). Outbreaks of these species often occur concurrently on the same host trees in the same geographical location, and it can be difficult to determine which species are responsible for the aggregate defoliation.

Concluding Remarks
Fall cankerworm is an endemic eruptive herbivore that feeds on many deciduous woody species throughout much of North America (Porter and Alden 1924, Ciesla and Asaro 2013, iNaturalist 2021. Larvae are commonly called loopers or inchworms, and despite its common name, defoliation occurs in the spring (Harris 1841). After pupating underground from late spring through fall, flightless females climb the nearest host and use pheromones to attract flying males during late fall and early winter (Hinds 1901, Porter andAlden 1924). Eggs are deposited in rings around small branches, which hatch the following spring (Harris 1841).
Defoliation can be severe and contribute to tree death in both managed and natural settings (Asaro and Chamberlin 2015). Largescale chemical treatments in forested landscapes are often not practical and may cause unintended ecological effects, although Btk has fewer non-target impacts than broad-spectrum insecticides. Further, as fall cankerworm is a generalist defoliator, specific host plant species cannot be targeted for treatment. Damage can be compounded by several lepidopteran species which often co-outbreak with fall cankerworm. Fortunately, natural enemies typically keep populations in check and stifle outbreaks in natural areas (Sherman 1921, Fedde et al. 1973, Fedde 1980, Ciesla and Asaro 2013. Forest health and natural enemy populations can be maintained through proper silvicultural management, by cultivating mixed-species and mixedage stands, and avoiding high grading and overstocking (Gottschalk 1993, Liebhold et al. 1997, Davidson et al. 1999. Fall cankerworm is a common pest of residential areas, and outbreaks typically occur more frequently in urban areas than in natural landscapes (Frank 2014, Noukoun et al. 2014, Walter et al. 2016. Populations can quickly build to injurious levels if monocultures of a preferred host species are maintained in urban areas. In these cases, consistent monitoring is recommended through the use of sticky band traps, which, in addition to establishing defoliation thresholds, can also provide a moderate level of control (Noukoun et al. 2014, Walter et al. 2016. The biological insecticide, Btk provides effective control if applied when early instar fall cankerworm larvae are feeding (Asaro and Chamberlin 2015, Jouzani 2017). Monitoring with sticky band traps can help inform optimal timing for Btk application, and planting diverse urban canopies also reduces potential damage from fall cankerworm (Gottschalk 1993). It is often advantageous to integrate multiple management strategies for the control of fall cankerworm in urban settings.
A consistent and effective monitoring program can gauge fall cankerworm population levels, outbreak potential, and inform management timing. Sticky band traps currently on the market must be frequently maintained to keep the surface free of debris, and bycatch of non-target organisms (including beneficial insects and small birds) is often an unintended consequence of a robust trapping program. Future research should be devoted to developing effective, lower maintenance monitoring strategies. Using synthetic pheromones in sticky band traps may also improve control efficacy, and accuracy in determining seasonal fall cankerworm population levels.