Surveys in northern Utah for egg parasitoids of Halyomorpha halys (Stål) (Hemiptera: Pentatomidae) detect Trissolcus japonicus (Ashmead) (Hymenoptera: Scelionidae)

The highly polyphagous and invasive brown marmorated stink bug, Halyomorpha halys (Stål) (Hemiptera: Pentatomidae), has become a significant insect pest in North America since its detection in 1996. It was first documented in northern Utah in 2012 and reports of urban nuisance problems and plant damage have since increased. Biological control is the preferred solution to managing H. halys in North America and other invaded regions due to its alignment with integrated pest management and sustainable practices. Native and nonnative biological control agents, namely parasitoid wasps, have been assessed for efficacy. Trissolcus japonicus (Ashmead) (Hymenoptera: Scelionidae) is an effective egg parasitoid of H. halys in its native range of southeast Asia and has recently been documented parasitising H. halys eggs in North America and Europe. Field surveys for native and exotic egg parasitoids using wild (in situ) and lab-reared H. halys egg masses were conducted in suburban and agricultural sites in northern Utah from June to September 2017–2019. Seven native wasp species in the families Eupelmidae and Scelionidae were discovered guarding H. halys eggs and adult wasps from five of these species completed emergence. ‡ ‡ §,| ‡ ‡ This is an open access article distributed under the terms of the CC0 Public Domain Dedication. Native species had low mean rates of adult emergence from wild (0.5–3.7%) and labreared (0–0.4%) egg masses. In 2019, an adventive population of T. japonicus was discovered for the first time in Utah, emerging from 21 of the 106 wild H. halys egg masses found that year, and none from lab-reared eggs. All T. japonicus emerged from egg masses collected on Catalpa speciosa (Warder). Our results support other studies that have observed biological control of H. halys from T. japonicus and improved parasitoid wasp detection with wild as compared to lab-reared H. halys egg masses.


Introduction
The brown marmorated stink bug, Halyomorpha halys (Stål) (Hemiptera: Pentatomidae), is a severe agricultural and urban nuisance pest that originates from southeast Asia (Hoebeke and Carter 2003) and has invaded numerous countries worldwide (Cesari et al. 2015, Gariepy et al. 2014, Haye et al. 2015b, Macavei et al. 2015, Milonas and Partsinevelos 2014, Vétek et al. 2014. As of 2020, it has been detected in 46 U.S. States and four Canadian Provinces, with 11 States reporting severe agricultural damage (StopB MSB.org). Halyomorpha halys was first detected in Utah in 2012 and has been considered a pest to fruit and vegetable crops since 2017. With the threat of increasing economic agricultural damage, development of proactive management tactics is imperative. In the U.S. Mid-Atlantic region, where H. halys has been a severe pest, effective control has relied on broad spectrum insecticides, leading to increased application frequency and disruption of integrated pest managment, including secondary pest outbreak , Leskey et al. 2012a, Leskey et al. 2012b. Physical or cultural control (e.g. trap cropping and mass trapping) can offer some mitigation of plant damage, but may not be economically viable . The most effective management tactics have paired cultural and chemical tactics (e.g. orchard perimeter insecticide applications and treatment of trap trees) .
Biological control by egg parasitoids has proven effective in suppressing H. halys populations in its native range (Yang et al. 2009, Zhang et al. 2017. Halyomorpha halys sentinel egg mass surveys in North America have identified parasitism by native parasitoids in the families Scelionidae, Encyrtidae and Eupelmidae (Abram et al. 2017, Cornelius et al. 2016a, Cornelius et al. 2016b, Balusu et al. 2019, Talamas et al. 2015a, Talamas et al. 2015b. However, parasitism rates are low, likely due to inability of native species to overcome healthy H. halys egg defences (Abram et al. 2017, Abram et al. 2014. Measuring native parasitoid effectiveness against H. halys eggs solely by wasp emergence may underestimate their impact, as partial development of a native wasp inside H. halys eggs can cause egg mortality (Abram et al. 2019a, Abram et al. 2014, Cornelius et al. 2016a, Schumm 2020. Therefore, evaluating native wasp parasitism rates, especially in novel landscapes where new behaviour or species may be observed, deserves critical analysis. Trissolcus japonicus (Ashmead) (Hymenoptera: Scelionidae) is an egg parasitoid native to the home range of H. halys (Yang et al. 2009). Adventive T. japonicus have been discovered emerging from H. halys egg masses in North America (Milnes et al. 2016, Talamas et al. 2015b and Europe (Stahl et al. 2019, Sabbatini Peverieri et al. 2018. Research has assessed the effectiveness of these adventive populations against H. halys and a recent study in Washington State revealed parasitoid emergence rates reaching 77% (Milnes and Beers 2019). Conversely, initial parasitism of H. halys eggs by T. japonicus in Europe has been as low as 2%. (Stahl et al. 2019 The primary objective of this study was to utilise H. halys egg mass surveys to identify potential parasitoid species for suppression of this invasive insect pest in northern Utah. Northern Utah provides novel geographic and environmental conditions for detection of H. halys parasitoids, most notably high elevation (>1200 m) and arid sites with a hot summer and cold winter climate (Utah Climate Center 2020), as compared to other regions where H. halys occurs , Jarrett et al. 2019, Milnes et al. 2016, Stahl et al. 2019. Secondly, we compared parasitism rates of wild (in situ) versus lab-reared egg masses to better understand effective survey approaches and projection of natural parasitism rates in the field (Abram et al. 2017).

Survey Sites
Surveys for native and exotic parasitoid wasps of H. halys eggs in northern Utah were conducted from June through September in each of 2017, 2018 and 2019. The surveys included a total of 17 field sites. Sites 1, 4, 5, 8 and 10-17 were located in suburban landscapes containing mixed woody ornamental trees and shrubs. Sites 2, 3, 6, 7 and 9 were in conventionally-managed agricultural row crops and orchards (Fig. 1). Survey sites were chosen, based on areas of established H. halys populations and preferred host plant availability (Tables 1, 2).

Stink Bug Colony
Halyomoprha halys egg masses were reared in the Department of Biology at Utah State University, Logan, Utah. The colony was initiated and continuously supplemented from wild H. halys collections in northern Utah beginning in 2016 and further supplemented in 2019 by egg masses from a colony at the New Jersey Department of Agriculture in Trenton, New Jersey. The lab colony was maintained at 25-28°C, 40-60% RH, with a 16:8 hr photoperiod.

Survey Methods
Fresh lab-reared egg masses were deployed at field sites within 24-48 hr post-oviposition. All lab-reared egg masses were oviposited on to paper towels, assessed for the number of eggs they contained and attached to wax-covered cardstock (4 cm x 4 cm), using doublesided sticky tape with sand to cover excess adhesive before field deployment. Lab-reared egg masses mounted on cardstock were attached to the underside of plant leaves (Table  1)  Upon collection, all egg masses were inspected for the presence of guarding parasitoid wasps. If present, wasps were collected with an aspirator (Carolina Scientific Supply Co. Burlington, NC) and placed into a 47 mm plastic Petri dish (Fisher Scientific Co. L.L.C. Pittsburgh, PA) with the associated egg mass to allow for further oviposition during transport to the lab in a cooler at ambient temperature, 15.5-24°C.
In the lab, egg masses were stored under the same conditions as the H. halys colony described above. Guarding female wasps were removed upon arrival at the lab, preserved in ethanol and later pinned for identification. Collected egg masses were inspected for the number of hatched (H. halys emergence), parasitised (parasitoid wasp emergence), missing (number of lab-reared eggs not present after field collection), unhatched or predated eggs (e.g. chewing or sucking damage) present approximately one week after collection, following procedures established by . Egg masses were observed again six weeks after collection to identify late-emerging wasps or those with partially-developed wasps within eggs (Stahl et al. 2019). Wasp species were identified using the keys to Nearctic Trissolcus (Talamas et al. 2015a), Nearctic Telenomus (Johnson 1984) and Nearctic Anastatus (Burks 1967).

Statistical Methods
Parasitism (defined as the proportion of egg masses in which one or more eggs produced adult wasps) was compared amongst years and egg types (wild and lab-reared) using a generalised linear model with a quasi-binomial distribution to account for over-dispersion due to small sample sizes in some years and zero-inflation. We report means and intervals that have been inverse-linked from the logit scale of the statistical model to the original proportion scale. Computations used the glm function in the stats package and various functions in the car (Fox and Weisberg 2019) and emmeans (Lenth 2019) packages in R software (R version 3.6.1; R Core Team 2019).

Voucher Specimens
Three voucher specimens of Trissolcus japonicus from this study have been deposited in the Florida State Collection of Arthropods, Gainesville, Florida (FSCA 00090589, FSCA 00090661, FSCA 00090662). A Darwin Core Archive of the data associated with these specimens is provided in Suppl. material 1.

Results
Over the three year survey period, a total of 39 parasitoids from five native wasp species emerged from six wild and five lab-reared H. halys egg masses. Anastatus mirabilis (Walsh & Riley), A. pearsalli Ashmead, A. reduvii Ashmead, Trissolcus euschisti (Ashmead) and T. hullensis (Harrington) were documented from both guarding females and successful emergence from H. halys egg masses (Fig. 2). Trissolcus utahensis (Ashmead) and Telenomus podisi Ashmead were observed guarding H. halys eggs, but did not successfully emerge as adults. Catalpa speciosa (Warder), Malus domestica Borkh and Prunus persica (L.) Batsch were the only plant species on which lab-reared egg masses were parasitised and this parasitism was by native wasp species exclusively (Table 1). In June 2019, the Asian parasitoid T. japonicus was first discovered in Utah emerging from two wild H. halys egg masses at Site 1 in Salt Lake City (Figs 1, 2, 3) and was detected consistently from June through September. Further, a single egg mass with emergent T. japonicus was found at Site 17 in August 2019. Trissolcus japonicus was only detected at these two suburban landscape sites. A total of 452 T. japonicus emerged from 21 of the 106 wild egg masses found in 2019 (Fig. 2). Parasitised wild egg masses were collected on two tree species, C. speciosa and Acer grandidentatum Nutt, with attack by T. japonicus occurring only on C. speciosa (Table 2).
When native wasp species successfully emerged from H. halys eggs, the mean number of parasitised eggs per affected egg mass was low, 4-25%. When considering only those egg masses giving rise to adult T. japonicus in 2019, the mean egg parasitism rate per mass was 78.5%. Additionally in 2019, a group of 19 wild egg masses experienced a similarly high mean parasitism rate of 67.3%, though these egg masses did not have adult wasps present at the time of collection, only signs of chewing and emergence (Fig. 2).
Mean parasitism of lab-reared egg masses was 0.42% and 0.05% in 2017 and 2018, with no wasps emerging in 2019. Mean parasitism rates of wild-collected egg masses in 2017, 2018 and 2019 were 2.9%, 3.7% and 28.2%, respectively (Table 3). The generalised linear model did not reveal a significant two-way interaction between year and egg type (P = 0.196, F = 1.63). Significantly more wasps emerged from wild than lab-reared egg masses (P = 0.002, F = 9.50). There were no significant differences in mean wasp emergence amongst years (P > 0.797, F = 0.23). Percent parasitism (± SE) of eggs in wild and lab-reared egg masses with adult wasp emergence in northern Utah, 2017-2019. Sample size (n) represents the number of egg masses parasitised by the indicated wasp species in each year. Bars without standard error lines represent single egg masses. The Unknown category represents egg masses in which parasitoid wasp emergence was confirmed, but no wasp specimens remained to confirm species identification. Table 3. Table 3. Mean parasitism of lab-reared and wild egg masses collected in northern Utah, 2017-2019. LCL and UCL refer to the lower and upper limit of a 68% confidence interval, respectively and approximately depict the mean +/-1 SE. See Suppl. materials 2, 3.

Discussion
Surveys of wild and lab-reared H. halys eggs in northern Utah demonstrated relatively high diversity of native parasitoid wasps, but these native species all exhibited low rates of parasitism. These findings are congruent with other North American surveys of H. halys egg parasitoids , Abram et al. 2019b. Low native parasitism rates could be caused by deterrence from natural chemical defences on and within H. halys eggs or a lack of effective venom at the time of female oviposition needed for successful wasp development in the exotic host egg , Tognon et al. 2016. Other research suggests that the use of parasitism emergence as a metric of parasitoid effectiveness underestimates native wasp effects on H. halys eggs since partiallydeveloped wasps that do not complete emergence often kill the stink bug host (Abram et al. 2019a, Cornelius et al. 2016b, Schumm 2020. Although our egg dissections revealed many unhatched H. halys eggs with undifferentiated contents (Milnes and Beers 2019), the ultimate cause of egg death could not be ascertained.
Our results support those of Jones et al. (2014) and Abram et al. (2017) who found that wild (in situ) egg masses more accurately detect the presence and ability of parasitoid wasps to emerge from H. halys eggs than do field-deployed lab-reared egg masses. This difference may be due to a variety of factors, including the age of the egg mass upon  2019) has also shown that kairomones, left by H. halys on host plant leaves, are detectable by T. japonicus and the wasp resides on these leaves longer than those lacking such kairomones. Therefore, lower parasitism rates of lab-reared egg masses could be due to reduced chemical cues.
Although current parasitism by the exotic T. japonicus in northern Utah is modest, relative to those in its native range (Yang et al. 2009, Zhang et al. 2017, our results indicate that T. japonicus has the potential to provide biological control of H. halys in the Intermountain West. Parasitism rates were not shown to be different amongst years, but our data clearly show higher parasitism in 2019, when T. japonicus was discovered, as compared to previous years. The dissonance of biological and statistical conclusions in our results is likely due to the variable and low sample size of egg masses. Trissolcus japonicus may have killed more H. halys eggs than we were able to document, based on identification of the causal wasp. Indeed, many egg masses were attacked by parasitoids that had already emerged from eggs before collection in 2019, with higher mean parasitism rates in affected egg masses than those observed for native wasp species, suggesting that at least some of the unidentified parasitoids were T. japonicus. It is also of interest to point out that T. japonicus was detected in two suburban landscape sites in Salt Lake County in 2019 and not in agricultural sites. The northern Utah region differs in its climate and topography from most locations in which T. japonicus has been documented or predicted to become established in North America (Avila and Charles 2018). Given the arid, high elevation conditions of northern Utah that include cold winters and hot summers, detection of an adventive T. japonicus population implies potential for range expansion into other locations within the greater Intermountain West region. These results support the possibility of an eventual intersection of eastern and western T. japonicus populations in North America (Jarrett et al. 2019, Talamas et al. 2015b, Milnes et al. 2016. Further research should focus on the capacity of T. japonicus to persist in the Intermountain West, specifically focusing on overwintering behaviour where heavy snowfall accumulation and consistent sub-zero temperatures occur (Lowenstein et al. 2019, Nystrom Santacruz et al. 2017. In fact, follow-up surveys in 2020 have documented continued detection of T. japonicus in Salt Lake and expansion into Utah counties (K. Richardson, personal communication). Laboratory rearing and releases, in conjunction with conservation efforts, are critical next steps in supporting the future establishment of T. japonicus populations in Utah.

Conclusions
Our findings show that an adventive population of T. japonicus in northern Utah is causing higher levels of reproductive parasitism of H. halys eggs compared to native wasp species and wild (in situ) egg masses provide a more accurate measure of parasitoid activity compared to those deployed from lab colonies. This study reports the first detection of T. japonicus in the Intermountain West, a novel geographic location for this parasitoid in North America.