Developing a non‐sticky trap design for monitoring jewel beetles

There is an urgent need in Europe to prepare resources for the arrival of the emerald ash borer, Agrilus planipennis (Buprestidae, Coleoptera) from European Russia, and possibly other invasive jewel beetles. A lightweight, easy to handle, non‐sticky trap could facilitate monitoring and detection to derive information about emerald ash borer and other jewel beetle populations. In two experiments carried out over two consecutive years in an oak forest, a new non‐sticky multi‐funnel trap design with a light‐green (sometimes described as fluorescent yellow) visual cue was developed. Altogether, there were 238 (2018) and 194 (2019) specimens captured often (2018) and eight (2019) Agrilus species, eight of which are oak‐related and one (A. convexicollis) was linked to ash. The new light‐green multi‐funnel trap performed similarly to the sticky design with a similar coloured surface. Our results suggest that the new trap design may be suitable for catching a wide range of buprestid species. It may also have the potential to be further optimized with respect to visual and olfactory cues, which would provide an even more useful tool for monitoring both invasive and indigenous buprestids.


IMREI Et al.
Efficient tools for the recognition of jewel beetle presence or changes in their population dynamics (including establishment at newly invaded sites) are currently missing. This is due to jewel beetles being small, swift flyers and because the larvae develop hidden under the bark of trees, resulting in their presence being invisible to unexperienced eyes (Chamorro et al., 2015;Muskovits & Hegyessy, 2002). Similar problems are faced by phytosanitary inspectors during visual inspections of wood packaging materials at ports of entry and other high-risk sites. Due to such difficulties, EAB infestations often go undetected during initial colonization resulting in insidious damage (Haack et al., 2002). By the time the characteristic D-shaped exit holes are recognized and the emerging adults are noticeable at ground level, the tree is already significantly damaged. Furthermore, the EAB mating system does not typically involve long-range pheromones, which can often be useful for detecting some individuals in sparse populations.
Monitoring methods currently in use for EAB include purple and green sticky prism traps (Francese, Crook, et al., 2010a;Francese, Fraser, et al., 2010b), and sometimes green or purple Lindgren multi-funnel traps (Francese, Fraser, Lance, & Mastro, 2011). The US federal government EAB monitoring system is based on purple prism traps with an estimated use of approximately 12 000 traps yearly (APHIS, 2018). These sticky traps have overwhelming operational costs, so a more user-friendly trap would be helpful to program managers, surveyors and researchers (Francese et al., 2011). Non-sticky trap designs can make handling and identification of specimens far easier (Domingue et al., 2013), and they are more suitable for quantitative comparisons than sticky designs (Wall, 1989). The efficacy of such traps would not rely on the residual strength of the adhesive, and there is less likelihood that trap saturation would affect future catches.
Furthermore, sticky traps also catch a significant number of non-target insects. These incidental catches result in reduction of the sticky surface and require additional labour if the surface is to remain clear (Lelito et al., 2007). In addition, because the area of sticky surface is reduced by each trapped insect, this is an inherent disadvantage compared with non-sticky traps, which have a constant catching capacity over time (Wall, 1989).
The objective of this study was to initiate development of a nonsticky trap type with a similar efficacy to sticky traps. The intention was also for the trap to be easy to handle and relatively inexpensive, F I G U R E 1 Trap designs tested a: lightgreen sticky PALz; b: transparent sticky PAL; c: light-green multi-funnel non-sticky MULTz; d: purple non-sticky MULTp; and e: transparent non-sticky MULT [Colour figure can be viewed at wileyonlinelibrary. com] thus being more economical for monitoring and detecting jewel beetle populations on a regular basis.
As attractant cues, light-green (sometimes called fluorescent yellow) and purple colours were trialled. The tested light-green was a similar hue to that previously found to be the strongest general attractant for a wide range of Agrilus spp. in Europe (Domingue et al., 2013). The purple colour was the same as that previously tested by Brown et al. (2017). In 2018, sticky and non-sticky traps, with or without light-green colouration, were compared; in 2019, the light-green, transparent and purple versions of the non-sticky design were compared. The Tanglefoot Company). When deploying the traps, the sticky sheet was folded with the sticky side facing outward and the rear edges fastened together with clips, forming a "cloak-like" structure ( Figure 1a,b).

| Statistics
Catch data were analysed by the non-parametric Kruskal-Wallis test (Kruskal & Wallis, 1987) because the data did not fulfil requirements for a parametric analysis. When the Kruskal-Wallis test indicated significant differences, pairwise comparisons were performed with Mann-Whitney U tests (Zar, 1999)

| D ISCUSS I ON
Results suggest that the newly developed non-sticky MULTz trap design is suitable for catching Agrilus jewel beetles. Efficacy of the MULTz was similar (even higher in absolute numbers caught) to that of sticky traps (PALz). Jewel beetle catches free from sticky residue are highly advantageous as there is no need to clean them with chemicals before identification. The maintenance and operation of a non-sticky trap in the field is also much easier and more straightforward than sticky traps.  (Francese, Rietz, & Mastro, 2013). Because Agrilus catches significantly increase with the placement of the trap at heights from 1.5 m to 13 m for EAB in the United States (Francese, Fraser, et al., 2010b;Francese et al., 2008;Ryall et al., 2012), from 3 m to 10 m for A. laticornis in the UK (Brown et al., 2017)   inexpensive flexibility depending on the trapping purpose. The importance of colour choice is supported by earlier findings. Numbers of male EAB beetles caught were much higher in green traps than in purple traps, regardless of height or trap type (Francese, Fraser, et al., 2010b;Francese et al., 2013). On the contrary, purple traps were found to be more attractive to females, showing female-specific sensitivity in the red regions of the spectrum at 640-670 nm (Francese, Crook, et al., 2010a;Francese et al., 2013). In Europe, purple traps attracted 2-3 times more adults of the beech-infesting A. viridis, with high (>95%) female ratios, when compared to green traps in the same study.
Furthermore, purple prism traps caught A. biguttatus and A. sulcicollis in oak forests in the UK (Brown et al., 2017) and C. undatus in a cork oak forest in Spain (Fürstenau, Quero, Riba, Rosell, & Guerrero, 2015). In our study, no Agrilus specimens were caught using purple traps. Eight out of eleven Agrilus species trapped in this study were linked to oaks (Muskovits & Hegyessy, 2002 Muskovits and Hegyessy (2002).

F I G U R E 3 Mean numbers (+SE) of
A. convexicollis was caught in significant numbers in light-green traps of both types, confirming a report from Slovakia where it was almost exclusively caught in traps with a green visual cue (Rhainds et al., 2017), and a report from Italy where catches in green multi-funnel Lindgren traps were higher than in purple traps of the same design (Rassati et al., 2019). A. angustulus and A. graminis are also recorded from the Russian Far East, although both are described as Western Palearctic species (Jendek & Nakladal, 2019 (Muskovits & Hegyessy, 2002).
A. viridis, a common European and North African species (Muskovits & Hegyessy, 2002), which was caught in small numbers in this study, has a wide range of hosts. At high population densities, it is described as a pest of common beech (Fagus sylvatica L.) (Molnár, Bruck-Dyckhoff, Petercord, & Lakatos, 2010), although no beech trees are known to exist in the vicinity of the experimental site.
Finally, the larvae of A. derasofasciatus commonly develop in the woody parts of grape species (Vitis vinifera L., V. sylvestris Gmel.) all over Europe according to Muskovits and Hegyessy (2002). It is important to note that all of these were caught in the PALz traps.
The MULTz traps form a good basis for further trap optimization.
Efficacy may be improved by further improving the green colouration and by adding combinations of other attractive visual (Crook et al., 2009;Gwynne & Rentz, 1983;Lelito et al., 2007) or olfactory cues. The latter may be plant-related (de Groot et al., 2008;Grant, Poland, Ciaramitaro, Lyons, & Jones, 2011;Grant, Ryall, Lyons, & Abou-Zaid, 2010) or beetle-related (Bartelt, Cosse, Zilkowski, & Fraser, 2007;Silk et al., 2011). Selectivity may also be improved as our understanding increases with respect to both the generalities of buprestid behaviour and the complexities of the communication systems of each jewel beetle species. Council's Industrial Strategy Challenge Fund.

CO N FLI C T O F I NTE R E S T
None declared.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are openly avail-