Soil microarthropods, ground-dwelling arthropods and soil properties in mown and grazed grasslands in the Veluwe region

In order to find out which factors limit the restoration of soil life and their ecosystem services under grasslands on sandy soils, we studied 40 grasslands of which 20 had agricultural and 20 nature land use, all after an agricultural history. Site selection Within the Veluwe region (The Netherlands), we selected 40 grasslands: 20 agricultural grasslands and 20 nature grasslands which were managed as new nature reserves since last tillage. Within each of these two land-use types, two types of grassland management were selected: mowing and grazing. Within each of the four combinations of land use and management we selected ten grasslands over a broad age range since last tillage. All grasslands were located on sandy soils (Typic Haploquod and Plaggeptic Haploquod; Soil Survey Staff 1999) with a deep water table to rule out dispersal of soil fauna during waterlogging (Siepel 1996; Jabbour & Barbercheck 2008). Vegetation and insect surveys Within each grassland a 5×5 meter monitoring plot was laid-out for plant cover surveys, insect and soil-microarthropod sampling and soil analyses. The vegetation surveys were carried out in 2019 at the end of May and in early June, using the Braun-Blanquet method (Braun-Blanquet 1932). In June 2019 soil-surface dwelling insects were sampled with a pitfall trap (Wiggers et al. 2015). Three pitfall traps (8 cm diameter, ca. 20 cm deep) were placed in each plot. Traps were half filled with a solution of water and glycol (3:1) and 3 % Extran soap. A plexiglass cover 20 cm above the trap prevented rainfall diluting the liquid. Traps were removed and emptied after seven days. Insects were identified and grouped at the order level, however, predator groups (carabid and staphylinid beetles, ants and spiders) were identified to the species level in order to group those by their feeding guild. Before analyzing the pitfall trap catches we first removed certain groups from the counts because pitfall traps are not well-suited to catch them systematically: Acari, Collembola, Psocoptera, Thysanoptera, Trichoptera, Lepidoptera, Siphonaptera, Diptera, Symphyta, Apocrita, and Parasitica. The remaining 62.0% of the caught individuals were surface-dwelling animals, and their totals (of three pitfall traps per site) were analyzed with negative-binomial generalized linear models. We also analyzed the subset of predators (73.6% of the surface dwellers). Soil chemical and pesticide sampling and analysis On 8, 9 and 16 October 2019, a bulk soil sample of 50 soil cores (0 - 10 cm) was collected from each 5×5 meter monitoring plot. After homogenization a sub-sample was analyzed for soil chemical analysis. Prior to chemical analysis, samples were oven-dried at 40 °C. Soil acidity of the oven-dried samples was measured in 1 M KCl (pH-KCl). Soil Organic Matter (SOM) was determined by loss-on-ignition (Ball 1964). Ammonium-lactate-extractable P (PAL) was determined according to the standard method (Bronswijk et al. 2003). Total potassium (K) in solution was determined using flame photometry after extraction of soil with HCl (0.1 M) and oxalic acid (0.5 M) in a 1:10 M:V ratio and filtration (Bronswijk et al. 2003). Clay (