Field surveys
Our study was conducted in the Ecological-Botanical Garden (EBG) in Bayreuth, Germany (49°55’45” N, 11°35’10” E, c. 365 m a.s.l.) (Kreyling, Schmid, & Aas, 2015). The EBG, as well as the larger region of Upper Franconia, underlies an oceanic to continental climate, with an annual average temperature of 8.2°C and an average annual precipitation of 724 mm. The months with the highest amount of precipitation and average temperature are June (75 mm, 16.2°C) and July (88 mm, 18.1°C, based on a time series ranging from 1991 to 2020, http://www.bayceer.uni-bayreuth.de/). Being established in 1978 planting began in the early 1980s. Many of the tree species examined in this study were planted in the first years after the garden was established. The EBG can therefore be considered as a large-scale common-garden experiment.
Our investigations include 24 species of Acer, 11 species ofBetula , 13 species of Fraxinus and 29 species ofQuercus . The genera Acer was sampled in 2020, while the genera Betula , Fraxinus and Quercus were sampled in 2021. (see Table S1). Insect sampling was performed only on sunny days with no to minimal wind and was aborted when the conditions got markedly worse. Sampling was conducted at least two full days after the last rainfall event, as bad weather conditions such as heavy rain and strong winds negatively affect insect activity (Weisser et al. 1997) and/or may knock insects off the plants (Kobori & Amano 2003). Tree individuals were sampled at three sampling dates starting at the end of May and ending at the end of August (see Table S2).
Sampling was performed using a beating tray (72 cm diameter, bioform e.K., product number A13a) with an attached sampling flask. Six branches were sampled per tree by hitting each branch three times with a wooden stick (5 cm diameter), and subsequently collecting the insects from the tray. Only larger branches, directly connected to the main stem, were chosen for sampling. Branches between 2 and 9 m above ground were used, starting with the lowest branches on each tree. Only free-hanging branches were used for sampling, meaning branches that did not directly affect other branches during beating. To prevent any subconscious decisions regarding the potential usability of branches, random branches were chosen from four directions, and, when possible, two different heights on the tree. The same branches were used in each sampling run.
Sampling flasks were marked and frozen overnight at -18°C. Insects were separated from debris and plant material and preserved in 96% ethanol for identification. Sampled insects were quantified and determined to order-level in the laboratory, using a stereo microscope (Leica S6E, Leica Microsystems GmbH). Orders Coleoptera and Hemiptera were sub-sampled for further determination on lower taxonomic levels. Coleopterans were identified to species-level using relevant literature (Assing & Schülke 2012; Freude et al. 1965), digital references (Lompe 2022) and by an expert (Boris Büche, Berlin). Individuals of the weevil superfamily (Coleoptera: Curculionoidea, families Anthribidae, Brentidae, Rhynchitidae and Curculionidae) were sent to an expert for determination to species-level (Peter Sprick, Hanover) as were the Heteroptera individuals (Markus Bräu, Munich). Data from all three sampling dates were pooled for statistical analyses.
To assess potential effects of conditions on and around the sampled trees we assessed diameter at breast height, total tree height and number of congeners within a ten-meter radius (see Table S1). We selected these parameters to represent individuals’ fitness, respectively, integrated growth performance as well as the amount of phylogenetic related individuals within the crown area.