Biological Station Wijster standard trapping program: Sampling event data for ground beetles (Coleoptera: Carabidae)

Ground beetles (Coleoptera: Carabidae) were sampled year-round in square metal pitfall traps with a perimeter of 1 m. The traps were custom made, based on an earlier design used in the Meijendel coastal dunes (Gerrits and Hemerik 2022). The traps consisted of an outer trap and an inner trap, the latter of which could be lifted out of the outer trap, emptied, and placed back with minimal disturbance to the surrounding soil and vegetation. A small hole, covered with mesh provided drainage of rain water. After 1959, protective covers were placed over the pitfalls to avoid rain, sand and debris building up inside the traps. These were made from opaque metal sheets placed a few centimeters above the trap using pins in the sand.

The traps were placed in groups of three (hereafter referred to as a 'series'), typically set in a straight line with 10m distance between traps, with the middle trap containing a metal funnel equipped with a container filled with 3% formaldehyde solution as killing agent. The two outer traps did not contain killing preservative and the beetles remained alive in these traps until collection. The initial purpose for the live traps was to provide a constant supply of living beetles to be used in ecological experiments. Mostly, however, the beetles trapped in the live traps were killed and collected for identification.

The historical locations of the trap series were obtained from a range of sources, including publications (Den Boer & van Dijk 1994), reports (Van Tol 2000, Verhagen 2005), historical maps and field notes. The coordinates provided in the DWC archive represent the locations to the best of our knowledge, pieced together from the various sources and descriptions. Their accuracy varies between 20 and 10000m.

Each sampling year ran from March 1st until February 28 (29th) of the next year, with some variation around these dates (see Data Range). The traps were typically emptied weekly, on the same week day, but this was variable in winter, with occasionally longer trapping periods around the holidays.

All collected specimens were identified to species level, enumerated, and in most cases separated between males and females by ground-beetle specialists working at the BSW. In the dataset, the sexes are separated as follows:
m = male
f = female
u = unknown.

In the historical period some species pairs were not (always) separated reliably (Den Boer pers.comm. 1999, Van Dijk pers. comm. 1999):
• Calathus melanocephalus / C. cinctus ( = C. erythroderus)
• Bradycellus collaris (= B. caucasicus) / B. harpalinus
• Pterostichus nigrita / P. rhaeticus
• Bembidion lampros / B. properans (always called B. lampros)
• Poecilus versicolor (= Pterostichus coerulescens) / P. cupreus. Poecilus cupreus is presently (since 2002) present in the traps, but it is unclear if it occurred in the historical period. This species pair was not flagged in 1999 by the original data collectors)

In addition, additional species may have been overlooked in the following genera (Den Boer 1999 pers. comm.):
• Agonum fuliginosum (may have been mixed with A. gracile and possibly other Agonum species).
• Asaphidion flavipes group
• Amara
• Harpalus

Samples were taken on a mostly weekly basis in the years 1959, 1961, 1963, 1965 and 1966 using pitfall traps. The traps were placed year-round in remnant forests, heathlands and bogs in the province of Drenthe, The Netherlands. Specifically, the following locations were sampled, which are now mostly under nature protection:
Mantingerbos, National Park Dwingelderveld, Hullenzand (part of Mantingerveld), Bargerveen, Dalerpeel, and locations in the towns of Wijster (on the property of the Biological Station Wijster), Bruntinge and Eursinge.

We used double entry for quality control of our work, and additionally compared it to several sources of data, aiming for a <0.1% error rate: