Vincent de Leijster

40 Chapter 3 3.2.3.4 Arthropod community composition Canopy arthropods In May 2017 canopy arthropods were collected using the beating technique (Benhadi-Marin et al., 2011). One sample consisted of the arthropods collected from four branches of the same tree. Each branch was hit six times and all the arthropods that fell off were caught in a round beating tray with a diameter of 1 m. This was done for four trees in each plot and repeated within two weeks. Collection took place between 12 and 3 pm. The beating tray was emptied with a pooter and the arthropods were stored in 70% ethanol. Canopy arthropods in the sample were identified based on morphotype to the level of order and when possible to a more detailed taxonomical level, especially when more detailed taxonomic identification was necessary to attribute an ecological role and ecosystem service. Then, we classified the arthropod taxonomic groups into ‘pests’ and their ‘natural enemies’ based on existing literature on almond pests (Appendix Table A3-3). In the pest group we included, from the order Hemiptera, the suborder Sternorrhyncha, which comprises almond aphids, and the family Tingidae (lace bugs, for example Monosteira unicostata -‘tigre de almendro’-))(Almatni and Khalil, 2008; Benhadi-Marin et al., 2011; Santos et al., 2012). Further, we included the order Thysanoptera and the larvae of Lepidoptera, which have both been identified as pests in almonds (Minaei, 2014; Santos et al., 2012). In the natural enemy group we included the family Coccinellidae (Coleoptera) and the order Araneae (Benhadi-Marin et al., 2011). Syrphid flies, lacewings and parasitic wasps are also considered to be important natural enemies (Almatni and Khalil, 2008), but the beating method is not sensitive to these arthropods so these could not be considered (Müther and Vogt, 2003). The data is expressed as pest abundance (see 2.3.5. for ‘pest absence’), natural enemy abundance, and natural enemy- pest ratio (NE:P), which was calculated by dividing the number of individuals categorized as natural enemies by the number of individuals categorized as pests (Denys and Tscharntke, 2002). Moreover, we calculated ‘pest absence’ from the pest abundance to maintain a more- is-better scale. This approach is explained in section 3.2.3.5 ‘Ecosystem service index’. Aerial arthropods In February and March 2017 aerial arthropods were collected using the pan trap method, which has been demonstrated to be the most efficient method for sampling aerial arthropods in European agricultural areas (Westphal et al., 2008). We followed the method as proposed by Westphal et al. (2008), using five clusters of pan traps per plot, each with three colored plates (blue, yellow and white). Pan traps were set up between 9 and 10 am in all plots and removed between 4 and 5 pm. The method was not repeated. The arthropods caught were