Anoplophora glabripennis (Asian longhorned beetle)

Widespread planting of susceptible poplars within the native zone led quite rapidly to the build-up and spread of A. glabripennis, which was previously of no particular importance there. In China, the species has only been a problem since the 1980s, when the first significant outbreak was discovered in Ningxia province (Pan HongYang, 2005). The Asian longhorned beetle A. glabripennis was first intercepted outside its native range in 1992 in North America, where an established population was first observed in 1996 (Haack et al., 1996) in the Brooklyn area of New York (USA). The first European invasive population was detected in Austria in 2001. Since then populations have been detected in Canada as well as in several European countries (Javal et al., 2019a; EPPO, 2020).

The genus Anoplophora includes 36 species, all native to Asia, the taxonomy of which has been stabilized with the revision of Lingafelter and Hoebeke (2002). This revision designates several synonyms for Anoplophora glabripennis (Motschulsky, 1854): Cerosterna glabripennis Motschulsky, 1853; Cerosterna laevigator Thomson, 1857; Melanauster nobilis Ganglbauer, 1890; Melanauster luteonotatus Pic, 1925; Melanauster angustatus Pic, 1925; Melanauster nankineus Pic, 1926; and Melanauster glabripennis var. laglaisei Pic, 1953. A. glabripennis used to be considered as part of the glabripennis complex, comprising A. glabripennis, A. freyi Breuning, 1947, A. flavomaculata Gressitt, 1935 and A. coeruleoantennata Breuning, 1947 (Wu and Jiang, 1998) but is now described as one single taxonomic entity.

Anoplophora nobilis (Ganglbauer, 1890) is also a synonym of A. glabripennis on the basis of cross-mating experiments (Gao et al., 2000), comparison of isoenzymes (Tang and Zheng, 2002) and random amplification of polymorphic DNA (RAPD; An et al., 2004), but nobilis is still considered as a form of A. glabripennis. In A. glabripennis, elytra are marked with white spots; in the nobilis form, these spots are yellow (Lingafelter and Hoebeke, 2002).

Eggs

About 5-7 mm, off-white, oblong (Haack et al., 2010). The ends of the eggs are slightly concave (Peng and Liu, 1992). Just before hatching, the eggs turn yellowish-brown.

Larva

The larva is a legless segmented grub up to 50 mm long when fully grown. It is creamy white, with a chitinized brown mark on the prothorax. The head has strong dark mandibles, and can be up to 5 mm wide (Cavey et al., 1998; Pennacchio et al., 2012).

Adult

Typically cerambycid in shape, adults can be 17 to 39 mm long. Females are generally larger than males, but important variation in size can be observed both within and between sexes. Males can be distinguished from females by their smaller size and longer antennae: when folded over the body, five antennae segments protrude from the apex of the elytra in males, compared to only one or two in females (Lingafelter and Hoebeke, 2002; Meng et al., 2015).

Adults of both sexes are characterized by shiny black elytra with white to yellowish spots, which is why they are called ‘starry sky beetle’ in some regions. The cuticle presents a bluish pubescence at the joints of the legs and antennae.

A. glabripennis is indigenous to East Asia. In China, the species is considered as native throughout the country. However, its prevalence and range has increased as a result of widespread planting of susceptible poplar hybrids (Pan HongYang, 2005). The pest has therefore a remarkably broad distribution in China, and can be found in most of the country at varying abundances (Yan, 1985; Li and Wu, 1993). The species has also been recorded throughout Korea (Lingafelter and Hoebeke, 2002) but is present at a moderately low density in riparian areas (Williams et al., 2004a). Despite some museum records which suggested that the species could have been native to Japan, it was established that A. glabripennis was not part of the Japanese endemic fauna (Lingafelter and Hoebeke, 2002). A. glabripennis has never been recorded in Taiwan (EPPO, 2020) despite extensive surveillance.

A. glabripennis has been detected in North America. In the USA, the first established population was discovered in New York City in 1996 (Haack et al., 1996). Several other populations have then been detected in the north-eastern part of the country as well as in Canada (Carter et al., 2010, EPPO, 2020). In Europe, established populations have been detected since 2001 (Hérard et al., 2006) and the species is now present in multiple other European countries (Javal et al., 2019a; EPPO, 2020).

Records of A. glabripennis in Maryland, Pennsylvania and Georgia (Nowak et al., 2001) published in previous versions of the Compendium are invalid as Nowak et al. (2001) does not mention detection of the pest in these states. For information on the states where ALB is considered present and under quarantine, see: https://www.aphis.usda.gov/aphis/resources/pests-diseases/asian-longhorned-beetle/Quarantines

The invasion process of the Asian longhorned beetle may have already started in its native zone, since its distribution has been altered by anthropic activities in Asia, as suggested in China (Javal et al., 2019b) and in the Korean peninsula (Lee et al., 2020). In Asia, niche modelling shows that the species is likely to settle in eastern and central China, North and South Korea, and Japan, and has a lower probability to establish in south-east China and east India (Peterson et al., 2004).

The steady increase in international trade from Asia since the early 1980s (Normile, 2004) has certainly increased the likelihood of unintentional introductions of the species. Once introduced, transportation corridors and abundance of preferred hosts are predictors of the ability of the species to spread (Shatz et al., 2013). In addition, the comparison between the climatic niche occupied by the species in its native zone and in its zone of introduction reveals only a negligible shift and expansion of niche (Hill et al., 2017), indicating an absence of changes physical and biological demands of the species during the invasion process.

Attention was drawn to A. glabripennis by its introduction into the USA in 1996 (Haack et al., 1996), where a major eradication programme is underway, and strong measures have been taken to reduce the risk of further introduction with wood packing from China. Niche modelling shows that most of the eastern USA has a high risk of Asian longhorned beetle establishment (Peterson and Vieglais, 2001; Peterson et al., 2004).

As forecasted by the detailed pest risk analysis done by MacLeod et al. (2002), the species was then introduced in Europe in 2001 (EPPO, 2020). Indeed, using the climate-matching system CLIMEX (Skarratt et al., 1995) and the EPPO Standard for pest risk assessment (EPPO, 1997), areas in southern Europe had previously been identified as those where the pest was most likely to establish and cause economic damage. More recent studies of climatic and environmental suitability show that the species might be able to spread in most of Europe, except for the most northern areas (EFSA, 2019). The invasion in Europe is most likely due to several introductions, both from the native zone, but also from already invaded areas such as the USA, highlighting the fact that international trade from all areas where the species is recorded should be the focus of management measures (Javal et al., 2019b).

Several assessments of A. glabripennis dispersal were made in its native area and were based on harmonic radar or CMR methods (Smith et al., 2001, 2004; Williams et al., 2004b). Adults tend to stay very close to the tree in which they have grown to feed and reproduce, as long as the resource is sufficient to support their offspring (Hu et al., 2009; Sawyer, 2009). Natural dispersal is nonetheless possible, and preliminary studies on the dispersal abilities of the species in its native range have shown that adults could fly over distances greater than 2 km over the total duration of their life (Smith et al., 2001; 2004; Williams et al., 2004b). However, some authors suggest that rare events of natural dispersal over long distances may partly explain the structure of Asian longhorned beetle populations (Trotter and Hull-Sanders, 2015; Hull-Sanders et al., 2017). This is indeed supported by flight mills experiments that showed that in a standardized environment, some specimens could fly more than 14 km in their lifetime (Lopez et al., 2017; Javal et al., 2018) This dispersal capacity could also explain part of the genetic patterns observed both in the original and invaded areas (Carter et al., 2009, 2010; Javal et al., 2019a, b).

Anthropogenic transport occurs mainly during very long-distance dispersal. Individuals can be transported as larvae in untreated wood packaging and, in very rare cases, live plants (Bartell and Nair, 2003; Hu et al., 2009; Haack et al., 2010). Such packaging is typically of relatively low grade, used for building materials (e.g. tiles). Individual larvae and adults have been detected in several European countries in wood packaging accompanying consignments from China (EPPO, 2003).

Human-mediated transport can also act on a finer scale like in Switzerland, where beetles from an infested area were transported to another zone in infested firewood (Eidg. Forschungsanstalt WSL, 2014; Tsykun et al., 2019; EPPO, 2020).

The species is very polyphagous and is found on a wide range of tree species (MacLeod et al., 2002; Haack et al., 2006; Hu et al., 2009; Sjöman et al., 2014; Meng et al., 2015). The genera most often described as preferential hosts for A. glabripennis include poplar (Populus spp.), willow (Salix spp.), maple (Acer spp.), birch (Betula spp.), elm (Ulmus spp.) and plane tree (Platanus spp.) (Sjöman et al., 2014). In China, A. glabripennis has mainly been recorded on poplar, willow, elm and maple trees (Pan HongYang, 2005) and the major hosts are species and hybrids of section Aegeiros of the genus Populus: P. nigra, P. deltoides, P. x canadensis and the Chinese hybrid P. dakhuanensis. Some poplars of the other sections of the genus (Alba and Tacamahaca) are also attacked, but are only slightly susceptible (Li and Wu, 1993). The insect has never been found on conifers, nor apparently on such important forest genera as Fagus and Quercus. This great diversity of hosts is explained by the peculiarities of the microbiome of its digestive system which allow it both to increase the rate of degradation of certain components of the wood, but also to access essential nutrients that the insect cannot synthesize on its own or metabolize from its diet (Scully et al., 2014). The great polyphagia of A. glabripennis is also linked to regulation mechanisms of genes involved in particular in the digestion and assimilation of nutrients to compensate for development in wood that is poor in nutrients (Mason et al., 2016). The amplification and functional divergence of genes associated with specialization of food on plants, including genes inherited from fungi or bacteria by horizontal transfer, have also contributed to the expansion of A. glabripennis’ metabolic repertoire (McKenna et al., 2016).

Research has been evaluating which tree species are most at risk from larval feeding in the invaded range. Bancroft et al. (2002), rearing larvae in freshly cut logs, ranked eight tree species in the following order for larval weight gain (from largest to smallest): Ulmus chinensis [U. parvifolia], Acer platanoides, Ulmus americana, Gleditsia triacanthos, Acer saccharum, Quercus rubra, Fraxinus americana and Fraxinus pennsylvanica. Smith et al. (2002) found that adult survival and reproductive capacity were higher on A. platanoides than on Acer rubrum and Salix nigra. Ludwig et al. (2002) investigated oviposition under caged conditions, and insertion of first-instar larvae into potted trees as experimental methods for determining host potential; they showed that eggs are laid on, and larvae develop in, species which are not yet known to be hosts (e.g. Q. rubra). Morewood et al. (2003) showed in laboratory experiments that the number of oviposition sites and living larvae was significantly higher in sugar maple (A. saccharum) than on red maple (A. rubrum), green ash (F. pennsylvanica) or red oak (Q. rubra).

Asian longhorned beetle infestation can be detected via oviposition and exit holes. Sap might be oozing from these holes and larval activity forms frass that accumulates either against the trunk or at the base of the tree. Masses of wood shavings and frass extruding from round exit holes are also signs that adults have emerged from infested wood.

Larval activity is recognized by the presence of galleries under the bark and, later, tunnels in the wood (Haack et al., 2010; Meng et al., 2015).

Based on random amplified polymorphic DNA (RAPD) fragments, Kethidi et al. (2003) describe DNA markers for the molecular identification of all development stages, and frass, of A. glabripennis, as distinct from related species. Other RAPD markers have also been used for intraspecific differentiation of populations (An et al., 2004). Molecular barcoding, based on mitochondrial DNA, provides a powerful tool to identify specimens, including immatures, to the species level (Javal et al., 2019a). Microsatellite markers have also been developed and used to visualize population structure and to decipher invasion routes (Carter et al., 2008, 2009, 2010; Javal et al., 2019b). A rapid diagnostic protocol based on loop-mediated isothermal amplification (LAMP) has been developed (Rizzo et al., 2020).

A. glabripennis is very similar to A. freyi from south-western China (Lingafelter and Hoebeke, 2002; Wu WeiWen and Chen Bin, 2003) and to A. coeruleoantennata (Lingafelter and Hoebeke, 2002).

In its invaded zone, A. glabripennis can be confused with pine sawyer species (Monochamus spp.). However, Monochamus species are usually smaller, usually do not show the same elytral patterns, and have a different phenology. The cotton borer is another similar species; however, it is characterized by black spots on light yellow elytra. In addition, its North-American distribution does not currently overlap with that of A. glabripennis (Meng et al., 2015). The greatest risk of confusion is with another invasive species, Anoplophora chinensis (citrus longhorned beetle, synonymous with A. malasiaca) (Haack et al., 2010; Pennacchio et al., 2012). Adults of A. chinensis differ from A. glabripennis in the presence of tubers at the base of their elytra, and the larvae of the two species show patterns of different size and shape on the disc of their pronotum (Lingafelter and Hoebeke, 2002; Pennacchio et al., 2012). A. chinensis has also been categorized as a quarantine pest by European countries (EPPO, 1997).

Adult specimens can be found between April and December (Haack et al., 2010), but the most active period for adult activity is late June to early July (Li and Wu, 1993) and oviposition most often occurs in early summer. Females form an oviposition well with their mandibles in tree bark, usually at the base of the lowest branch crown (Li and Wu, 1993; Haack et al., 2006). About 45-62 eggs are laid one by one by females (Wong and Mong, 1986; Keena, 2002) between the bark and the cambium. Eggs develop faster at higher temperatures, and take between 54.4 and 13.3 days to hatch at temperatures ranging from 15 to 30°C (Keena, 2006). While developing, the larva first digs galleries under the bark before digging deeper inside the trunk, where it eventually forms a pupal compartment to spend the 12 to 50 days necessary for its metamorphosis (Keena and Moore, 2010). Another 4-7 days are needed for sclerotization to be complete, and 4-5 days for the adult insect to burrow the trunk to the surface and extricate itself from the trunk (Sánchez and Keena, 2013). The emergence of adults lasts throughout the summer, with peak emergence most often between May and June, varying according to region and climatic conditions (Haack et al., 2010). Short-range pheromones allow partner recognition (Zhang et al., 2002; 2003a). The longevity and fecundity of insects are conditioned by the larval host and temperature conditions, but adults usually live for about a month (Li and Wu, 1993; Keena 2002, 2006; Morewood et al., 2003). The life-cycle takes place in 1 to 2 years, depending on the environmental conditions: the individuals spend the winter in larval form, and it is necessary that the larvae have reached a minimum mass at the beginning of winter in order to be able to induce pupation the following spring (Keena and Moore, 2010). In China, the number of annual generations varies with climate and latitude. The further north A. glabripennis is found, the longer it takes for a generation to develop. In eastern China, a generation may take 1 or 2 years to develop, whereas in northern China (Neimenggu), a single generation takes 2 years to develop. Thus, there can be one or two overlapping generations per year, depending upon the climate and feeding conditions (Pan HongYang, 2005). The adults usually remain on the tree from which they emerged, or fly short distances to nearby trees, and feed there on leaves, petioles and young bark before mating and laying eggs (Smith et al., 2001, 2004; Pan HongYang, 2005).

In the native range, most populations are found in urban environments, mainly in range trees and in parks, as well as in monocultures (Smith et al., 2009) but in some cases they can also be observed on the edge of the forest (Williams et al., 2004a). In Europe, no outbreaks have been detected outside urban areas (EPPO, 2020). The situation is similar in North America, with the exception of a population detected in 2008 in a peri-urban forest in Massachusetts (Dodds and Orwig, 2011). Given the current distribution of the species, its biology, and its regular interceptions around the world, it is possible that A. glabripennis could become established in much of Europe (MacLeod et al., 2002), but also in North America and Asia (Hu et al., 2009). On the basis of this information, the species has been classified as a quarantine pest in both Europe (EPPO, 2020) and North America (USDA, 1998).

Luo et al. (2003) briefly review possibilities for biological control of the pest. Multiple control strategies are still under study, but do not seem conclusive at the moment (Pan HongYang, 2005; Dubois et al., 2008; Haack et al., 2010; Ugine et al., 2013; Brabbs et al., 2015) often due to their toxicity or their low specificity of action. Indeed, A. glabripennis has few natural enemies because the majority of its development cycle takes place hidden inside trees, making its management by biological control methods difficult (Pan HongYang, 2005). For example, entomopathogenic bacteria have been identified as a potential means of controlling insects (Pan HongYang, 2005; Brabbs et al., 2015) but the tests carried out to date in the laboratory have not revealed any significant effect on the different stages of development of individuals (D'Amico et al., 2004). Some entomopathogenic fungi of the genera Beauveria and Metarhizium are also known for their virulence, and have the advantage of being able to form epidemics within populations (Shimazu et al., 2002; Zhang et al., 2003b; Dubois et al., 2008; Hu et al., 2009; Ugine et al., 2014; Brabbs et al., 2015) and some strains are even registered as biocontrol agents in several regions of the world (Dubois et al., 2008). Several nematode species also attack A. glabripennis, mainly belonging to the genus Steinernema (Fallon et al., 2004) and can trigger high level of mortality (Liu et al., 1992). However, their use as a biological control agent is made tricky by the difficulty of maintaining certain species under laboratory conditions, but also by their sensitivity to the method of release into the environment which largely conditions their survival rate (Brabbs et al., 2015). The option of parasitoids was also explored. Among the known species of Anoplophora parasitoids, larvae of the beetle Dastarcus helophoroides are ectoparasites of A. glabripennis larvae (Pan HongYang, 2005; Hu et al., 2009; Haack et al., 2010; Brabbs et al., 2015). In Europe, native parasitoids attacking the larvae of A. glabripennis have been identified, but their dissemination in urban areas is made impossible for public health reasons (Brabbs et al., 2015). Finally, A. glabripennis has few predators. No study shows any predation on A. glabripennis by any other insect species (Haack et al., 2010). Some woodpecker species, on the other hand, can feed on the larvae. The grey-headed woodpecker (Picus canus) and the great spotted woodpecker (Dendrocopos major) for instance, are both found in both Asia and Europe, and are effective predators of A. glabripennis. Their nesting is therefore encouraged as much as possible (Pan HongYang, 2005).

A. glabripennis is now considered one of the 100 most problematic invasive species in the world (Lowe et al., 2000). It has a high destructive potential because unlike many cerambycid species it attacks a priori healthy trees, mainly urban or peri-urban. By digging its galleries, the larva physically alters the vascular system of the tree, eventually leading to its death. The boring larvae damage the phloem and xylem vessels, resulting in heavy sap flow from wounds which are then liable to attack by secondary pests and infection. Infested trees lose turgor pressure, and leaves become yellow and droop. Structural weakening of trees by the larvae in urban regions poses a danger to pedestrians and vehicles from falling branches. The adults can also cause damage by feeding on leaves, petioles and bark. It is therefore a harmful species, both in its native area and in its introduction zones.

In China, it has been known since the Qing Dynasty (1644-1911), but has only been a problem since the 1980s, when the first significant outbreak was discovered in Ningxia Province (Pang HongYang, 2005). It was also during this period that China undertook major reforestation work in the north-east of its territory in order to protect the capital and agricultural land from the wind, limit soil erosion, and create new resources for timber and paper production. Initially, the Chinese government largely favoured monocultures of fast-growing species with low resistance to A. glabripennis (i.e. Populus spp., Salix spp. and Ulmus spp.), thus leading to a dramatic increase in its populations. Since then, silvicultural practices have been adapted, and emphasis has been placed on multi-species stands and the use of poplar clones that are less susceptible to insect attacks (Yin WeiLun and Lu Wen, 2005). Many factors have contributed to the extremely rapid explosion of the A. glabripennis population in China, including the lack of natural enemies in newly planted forests, as well as the lack of control and monitoring which has led to the transport of cut timber from infested areas, allowing the insects to be transported over long distances. In the most affected areas, the infestation rate could reach 80-100% of the trees (Pan HongYang, 2005). The species is estimated to cause about 10 billion yuan of economic loss per year (1.5 billion dollars), that is to say 12% of the economic losses due to forest pests in China (Hu et al., 2009). Poplar wood damaged by A. glabripennis larvae can be downgraded and lose value by up to 46% (Gao et al., 1993). Severe damage is caused between 21° and 43°N and 100° and 127°E in China (Yan, 1985). In contrast, A. chinensis has so far shown only a low propensity to form outbreaks, with few exceptions in pecan orchards in southern China.

In North America and Europe, A. glabripennis represents a considerable threat to urban trees. In the USA, for example, its maximum potential impact is estimated to be an economic loss in the order of $669 billion, with a loss of nearly 35% of the canopy and 30% tree mortality (Nowak et al., 2001). For instance, suppressing a 1996 infestation in New York State cost more than 4 million USD (USDA, 1998). Between 1998 and 2006, the USA invested nearly $249 million in eradication programmes for the species (Smith et al., 2009). In Canada, projections show that potential costs for management of Asian longhorned beetle in the urban environment (i.e., removing and replacing damaged trees) ranged from CDN $8.6 to $12.2 billion. Damage on merchantable maple (Acer spp.) timber was evaluated between CDN $1.6 billion and CDN $431. The impact on edible maple products was estimated at CDN $358 million annually (Pedlar et al., 2020).

The actual cost of an A. glabripennis infestation depends on the method used to estimate the overall cost of tree loss in urban areas (Faccoli and Gatto, 2016; Pedlar et al., 2020). It includes the cost of the eradication programme which involves scientific expertise to confirm the identification of the species, the control of trees in the infested area, the felling of trees in which the insect grows and their removal, and finally the replacement of these trees (Faccoli and Gatto, 2016). In north-eastern Italy, the rapid implementation of an eradication programme led to a 52% reduction in the number of infested trees expected the following year, representing a saving of nearly €250,000 compared to inaction. The ornamental value of the preserved trees is about six times higher than the cost of their protection (Faccoli and Gatto, 2016). However, in urban areas, the implementation of eradication programmes that include the felling of trees from private gardens can have repercussions on the feelings of populations who, often due to a lack of information, do not see how local control and prevention can stem a global phenomenon (Porth et al., 2015). In addition to the impact on the ornamental value of attacked trees, the presence of A. glabripennis reduces forest value and production (Pan HongYang, 2005; Smith et al., 2009; Pedlar et al., 2020) but could also, in some regions, influence other aspects of the economy such as forest-related tourism or maple syrup production (Smith et al., 2009; Pedlar et al., 2020).

Existing problems with A. glabripennis mainly concern plantations in China and urban trees in North America and Europe. Larvae dig galleries in the sapwood and heartwood of their hosts. One single larva can eat up to 1000 cm³ of wood (Yan and Qin, 1992). Poplar (Populus euramericana [P. canadensis]) can lose up to 22-49% of diameter of the trunk at breast height (DBH) and up to 5-25% of vertical height after being infested with Asian longhorned beetles for 3 years (Gao et al., 1993).

It is not clear how damaging this pest could be to its host trees in natural or managed forests, nor to the native fauna in its invaded range.

In China, control measures include the direct application of insecticides (Chen et al., 1990; Liang et al., 1997), trap trees combined with insecticide treatments (Sun et al., 1990) or the use of insect-pathogenic nematodes (providing up to 94% mortality; Liu et al., 1992). As certain poplar hybrids are relatively resistant (Qin et al., 1996), the planting of such hybrids is now preferred, and the use of very susceptible hybrids is avoided. Control strategies in China have been reviewed by Luo et al. (2003) and Pan HongYang (2005).

The case of A. glabripennis has been the main stimulus for the development by the FAO Interim Commission on Phytosanitary Measures of an international standard 'Guidelines for regulating wood packaging material in international trade' (ICPM, 2002). In the USA and in Europe, strong measures have been taken for wood packing materials (packing cases and dunnage) from China. Such packaging should be treated by methods recognized to have adequate efficacy against all wood pests. These currently include heat treatment (to an internal temperature of 56°C for 30 min). Once treated, packing wood is unlikely to be re-infested, so such wood (especially crates and pallets) can continue to be used in trade. An internationally recognized mark is stamped on the treated wood. The international standard was approved in 2002 and is now progressively being implemented worldwide, in order to prevent introductions.

In the invaded zones, control measures aim to contain and eradicate the outbreaks in urban areas (Lance, 2003; Haack et al., 2010). Unger (2003) gives an example of the measures needed to exclude the pest from Germany. Detection of outbreaks is difficult and is often incidental. Indeed, the cryptic lifestyle and tendency of the beetle to lay small numbers of eggs on several trees combine to make it difficult to define the limits of the outbreak and thus to eradicate the beetle without destroying large numbers of trees. Once the presence of the species is confirmed, a security perimeter is defined around the infested trees, and these trees are immediately felled and then crushed or incinerated on site. Host species in the vicinity of the outbreak are usually controlled visually. Felling of infested trees is currently the only means of control implemented systematically in Europe, Canada and the USA (Hérard et al., 2006; Meng et al., 2015; Turgeon et al., 2015). This method has proven to be effective in relatively recent, small-scale outbreaks. Inspection of susceptible trees is, however, often not sufficient to detect the entire infestation. Indeed, as the most obvious sign of the presence of the species is the emergence hole, the infestation is very often detected when the adults have already emerged from the tree and have potentially dispersed further. In addition to the visual inspection, dogs have been trained to detect the presence of A. glabripennis (Hoyer-Tomiczek et al., 2016). These dogs are capable of smelling the insect in trees, and are also used to inspect suspicious shipments at potential points of entry.

The difficulty in detecting infestations explains, for example, why new infested trees were detected in 2013 in the Toronto area (10 years after the first discovery of the species in the region) a few months after the species was considered as eradicated from the monitoring area. Growth rings from felled infested trees showed that the individuals had been present for several years, suggesting that this second finding was probably a satellite population of the original population observed (Turgeon et al., 2015).