2026/04/27 by Robert Künast, Sebastian Jeschke, Patrick Mäder +1 · 1 voice
Environmental Science · Earth and Planetary Sciences · #Wildlife-Road Interactions and Conservation #Physiological and biochemical adaptations #Subterranean biodiversity and taxonomy
paper · doi:10.1016/j.baae.2026.04.007
Mowing is essential for maintaining species-rich grasslands but causes short-term mortality and habitat disruption for arthropod communities. However, mid-term recovery dynamics after mowing within the same season remain poorly understood, mainly due to limitations of traditional sampling methods. Camera traps equipped with artificial intelligence are promising tools for providing large-scale, high-temporal-resolution insight into such ecological dynamics. To deepen our understanding of the mid-term effects of mowing, we deployed AI-based camera traps across 27 grasslands. We monitored arthropod activity from April to November, focusing on 2 weeks before to 10 weeks after mowing for 30 mowing events. Over 20,727 hours, we recorded 145,226 arthropod observations and trained and evaluated an identification algorithm to differentiate among higher taxonomic levels. We assessed short- and mid-term effects, recovery, and the impact of long-term management intensity on arthropod activity. Mowing caused significant short-term declines in overall arthropods, Hymenoptera, and Diptera activity. Comparing post-mowing time intervals with pre-mowing levels showed significant differences up to 8 weeks after mowing for overall arthropods and 4 weeks for Diptera. This indicates that recovery within the same season is possible but takes longer than previously thought and may depend on the taxa, highlighting the importance of choosing mowing dates that align with conservation aims. Regarding long-term effects, we observed a positive and significant relationship between the five-year average land-use intensity and the activity of overall arthropods, Diptera, Hymenoptera, Hemiptera, and Orthoptera. This may reflect a shift in communities in highly disturbed habitats toward species with high mobility or enhanced activity as a stress response. Our findings demonstrate that prolonged arthropod suppression following mowing—lasting 8 weeks for most taxa—likely depends on recolonization from unmown landscape refuges, underscoring the need for spatially and temporally staggered mowing regimes that maintain continuous recolonization sources across conservation landscapes.