Answers are grounded in this paper and cite their sources.
Artificial light at night (ALAN) is rapidly expanding worldwide and disrupting the behaviour of organisms, including cross-species interactions that are essential for ecosystem services. Biocontrol by invertebrate predators relies on predator-prey interactions occurring in the correct time and place, and altering these dynamics may be ecologically and economically costly. We used a Before-After Control-Impact (BACI) design to assess the abundance and functional composition of ground-dwelling invertebrates in a temperate evergreen forest of Aotearoa New Zealand. Pitfall traps at lit plots were sampled during baseline (i.e. the period prior to exposure), acute (three-day exposure), and short term (four-week exposure) periods of treatment with ALAN and compared to control (dark) pitfall traps sampled at the same time points. We expected that the beneficial hunting conditions provided by ALAN would attract predators. However, in contrast to studies in open habitats, we found that predatory and omnivorous invertebrates in our closed-canopy ecosystem declined in abundance (3-fold and 2-fold reduction respectively) relative to controls. ALAN did not shift overall composition but increased variability among lit plots, with changes driven primarily by shifts in taxon-abundance rather than species turnover. These shifts may have cascading effects on prey populations, with potential consequences for ecosystem functioning. Our findings highlight that ALAN's ecological impacts are not universal: predator attraction in open ecosystems may increase top-down control, whereas avoidance in evergreen forests may reduce biocontrol. Protecting dark refugia remains essential for maintaining diverse trophic guilds of invertebrates and ensuring ecosystems continue to function despite the global spread of artificial lighting.
No takes yet. Share an insight, caveat, or question.
Farnworth et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: