Effects of altered precipitation and wolf spiders on the density and activity of forest-floor Collembola

Abstract

Summary A 2×2 factorial field experiment was conducted in a deciduous forest to determine how changes in moisture and numbers of a common wolf spider, Schizocosa, affect Collembola abundance and activity. Each of four 4-m2 fenced, roofed plots was divided into eight 0.5-m2 fenced subplots, and each subplot was assigned a rainfall treatment (drought or high-rainfall) and a Schizocosa treatment (Schizocosa-removal or Schizocosa-addition). Water was sprayed on the high-rainfall subplots weekly from 7 August to 1 October 2003 at a rate of roughly 2× the long-term mean, while the drought subplots received no water. In order to increase the probability of uncovering possible effects of Schizocosa, predatory arthropods (Schizocosa, other spiders, and centipedes) were first removed from all subplots; Schizocosa-addition subplots were then stocked with five juvenile Schizocosa on 19 August, with additional Schizocosa added throughout the experiment. An index of relative Collembola activity was calculated by comparing the number captured in pitfall traps (activity-density) with absolute density (no./m2) measured by extracting Collembola from litter samples. Decreased moisture reduced overall Collembola density, but led to increased Collembola activity. This response to rainfall was exhibited by the families Entomobryidae, Tomoceridae, and Hypogastruridae. The presence of Schizocosa had no impact on overall Collembola density, nor did spider presence affect Collembola activity, with the possible exception of the Hypogastruridae, for which there was an interaction between rainfall and Schizocosa presence. Hypogastrurid Collembola were more active in the drought subplots and exhibited the same level of activity in the presence or absence of Schizocosa under drought conditions. However, in the high-rainfall treatment, the presence of Schizocosa increased hypogastrurid activity.