Development of a Novel Approach for Monitoring the Samurai Wasp, Trissolcus japonicus (Ashmead), an Effective Parasitoid of the Brown Marmorated Stink Bug, Halyomorpha halys (Stal)

Objective 1. Examine and compare the effectiveness and efficiency of sentinel BMSB eggs and yellow sticky traps for sampling T. japonicus

Vertical Sampling Transects: This study was in 2017 in Frederick County, VA, where T. japonicus had been detected since 2015. Five mature female tree of heaven growing at the edge of woodlots adjacent to apple and peach orchards were selected for sampling. Trees were of similar height and architecture (i.e. had branches from bottom to top of the trunk). Eyebolts screwed into the top and bottom of the trunk of each tree had a closed loop of rope running through them, enabling sampling devices attached to the rope to be raised and lowered along the trunk as a vertical transect, with sampling locations at the lower, mid, and upper canopy of each tree.

Sentinel eggs and yellow sticky traps. BMSB adults were reared at Virginia Tech’s Alson H. Smith, Jr. Agricultural Research and Extension Center (AREC), Winchester, VA. Egg masses deposited on green bean leaves were collected daily and affixed to double-sided tape on a small piece of cardstock, then held at -80^oC for ≤ 3 weeks before deployment in the field. Egg cards were affixed to the underside of leaves on excised shoot sections of female tree of heaven, which were inserted into 500 mL plastic bottles with water, then attached to the transect rope in each tree and positioned at the three canopy locations mentioned above.

In preliminary studies conducted in 2016, T. japonicus was captured in single-sided yellow sticky traps (Alpha Scents, West Linn, OR) deployed in the canopy of trees at the AREC. These were employed in 2017 for comparison with sentinel eggs. On alternating weeks from mid-May until late August, 2017, sentinel eggs or yellow sticky traps were deployed at lower, mid, and upper canopy locations in each tree. Egg masses were retrieved after 72-hr, since their attractiveness to parasitoids is known to diminish after that time. Egg masses were held in an environmental chamber and assessed daily for several weeks for emergence of parasitoid adults. On alternate weeks, yellow sticky traps were deployed in the same trees at the same canopy locations, and retrieved after 5 days. The parasitoids of interest (i.e. those considered to be potential parasitoids of BMSB) captured in traps and those that emerged from sentinel eggs were tentatively identified, then sent to a specialist for species confirmation.

The low numbers of adult T. japonicus captured or recovered from sentinel eggs in 2017 were not amenable to statistics, so results are presented using descriptive statistics.

Objective 2.  Determine the effect of habitat type on T. japonicus detections

Field Sites: All study sites were in Frederick County, VA.  Tree of heaven was again used as the model host tree. Trees (n = 1 per site) growing in the following common habitat types were selected; 1) spatially isolated patches associated with rock breaks in cultivated fields, 2) windbreaks or hedgerows, and 3) at the edge of woods (n = 5 per habitat type). The same sites and trees were used for T. japonicus trapping in 2018 and 2019.

Trapping Trissolcus japonicus

At each site, a double-sided, backfolding yellow sticky trap mounted atop a 4.8 m bamboo pole was suspended from a mid-canopy branch of female tree of heaven via a hook attached to the pole. Trap deployment at mid-canopy was informed by previous data showing that naturally-laid BMSB egg masses were most abundant in the mid- and upper tree canopy and that T. japonicus was recovered only from eggs collected from those locations. Traps were replaced at 7-day intervals from 3 May until 21 September, 2018 and from 20 April until 30 September, 2019.  All parasitoids of interest captured were tentatively identified in situ, then sent to a specialist for species confirmation. Captures of T. japonicus each year were compared among habitat types using the Kruskal-Wallis test followed by the Bonferroni corrected Dunn’s test (SAS Institute, Cary, NC; SAS Institute Inc. 2018).

Objective 3. Determine the seasonality of T. japonicus detections

Data from Objective 2 were used to examine the onset, progression, and termination of T. japonicus captures in 2018 and 2019. For each year, weekly captures of male and female T. japonicus were pooled across the three habitat types.

 Objective 4. Examine the effect of BMSB host tree species on T. japonicus detections

Field Sites and Tree Species: In 2017 and 2018, trees were sampled at the edge of woodlands in Frederick County, VA. In 2019, following the 2018 results from Objective 2, sampling was conducted in windbreaks. The yellow sticky traps introduced in Objective 2 were deployed in the mid-canopy of trees, as described previously. Selection of the tree species used was based on; 1) census data, 2) known hosts of H. halys, and 3) leaf architecture.  Tree of heaven has been reported to be the most abundant species at the edge of woodlands bordering tree fruit orchards in this region, and as in our previous studies, was considered the standard host. The other species selected were black walnut, black locust, hackberry, and black cherry, and are also among the most common species recorded in a previous census. These trees represent a diversity of plant families and differ in leaf structure; tree of heaven, black walnut, and black locust have compound leaves, while hackberry and black cherry have simple leaves.

Simultaneous trapping in paired tree species at the forest edge: In 2017, tree of heaven was paired with black walnut, black locust, or hackberry (n = 5 per species pairing), with ≥ 10 m and ≤ 25 m between trees in each pair. One trap per tree was replaced at 7-day intervals from 31 July until 29 August, 2017. In 2018, the same trees were used again, and 5 pairs of tree of heaven and black cherry were added. Traps were deployed from 13 June until 20 September, 2018, and replaced at 7-day intervals.

Because trapping for T. japonicus in 2018 showed higher captures in female tree of heaven growing in windbreaks and isolated patches than at the woods edge, in 2019 traps were deployed in trees in windbreaks (n = 5 sites), each containing tree of heaven, black walnut, and black locust. Adjacent trap trees in each windbreak were 23.7 ± 8.6 m apart and the distance between the traps at the ends of the sampling area was 47.4 ± 15.9 m. Traps were replaced at 7-day intervals from 17 June until 11 August, 2019. As in the previous studies, all parasitoids of interest captured were tentatively identified in situ and subsequently sent to a specialist for species confirmation.

In 2017 and 2018, captures of T. japonicus were compared by species pair using a Wilcoxon signed-rank test, while in 2019 captures were compared among three host species using the Kruskal-Wallis test followed by the Bonferroni corrected Dunn’s test. All statistical comparisons used SAS (SAS Institute, Cary, NC; SAS Institute Inc. 2018) and were considered significant at P < 0.05.