Thresholds are in the eye of the beholder: plants and wildlife respond differently to short-term cattle corrals

Getting better with age: Lessons from the Kenya Long-term Exclosure Experiment (KLEE)

The Kenya long-term exclosure experiment (KLEE) was established in 1995 in semi-arid savanna rangeland to examine the separate and combined effects of livestock, wildlife and megaherbivores on their shared environment. The long-term nature of this experiment has allowed us to measure these effects and address questions of stability and resilience in the context of multiple drought-rainy cycles. Here we outline lessons learned over the last 29 years, and how these inform a fundamental tension in long-term studies: how to balance the need for question-driven research with the intangible conviction that long-term data will yield valuable findings. We highlight the value of (1) identifying experimental effects that take many years to manifest, (2) quantifying the effects of different years (including droughts) and (3) capturing the signatures of anthropogenic change. We also highlight the potential for long-term studies to create a collaborative community of scientists that brings new questions and motivates continued long-term study.

Weak negative responses of spider diversity to short-term ‘kraaling’

The influence of short-duration, concentrated kraaling (enclosure) has been documented for plants, wildlife, and macro-invertebrates. However, limited information is available on its impact on ground-dwelling spiders. The purpose of this study was to assess the effect of short-duration kraaling, time since cattle removal, and microhabitat variables on spider assemblages in Matabeleland North Province, Zimbabwe. We used a matched-pair and space for time design (inside vs outside previously kraaled inclusions) across 11 sites, using four cattle herds (H1, H6, H7 and HNguni). Spiders were sampled in the early and late rainy season with pitfall traps left open for 14-day sampling periods and emptied twice in each period. We captured 634 spiders, comprising 63 species in 44 genera and 18 families. The most abundant family was Lycosidae (37%; 16 spp.), followed by Gnaphosidae (15%; 10 spp.) and Salticidae (14.5%; 7 spp.). Generalised linear mixed models showed that generic richness was greater in sites with more bare ground. However, this effect was reversed in previously kraaled sites, and was particularly evident for spider abundance that responded negatively relative to unkraaled sites. Furthermore, with a U-shaped recovery, generic richness increased with time since kraaling. Model-based multivariate models showed that short-duration kraaling had a significant impact on spider assemblage structure, but this impact was relatively small compared with the effect of seasonality. Most of the species that made significant contributions to this multivariate response were less abundant in kraaled sites. Spider diversity, therefore, had a weak negative response to short-term kraaling. However, these impacts should also be assessed at broader scales, including areas where cattle go to graze during the day.