Biodiversity drives ecological functioning, ultimately providing ecosystem services. Ecosystem processes are favored by greater functional diversity, particularly when groups of functionally different species interact synergistically. Many of such functions are performed by insects, among which dung beetles stand out for their important role in dung decomposition. However, anthropogenic disturbances are negatively affecting their ecological dynamics and ecosystem services. We conducted a manipulative field study, to evaluate the effect of human disturbance on dung beetle diversity (abundance, species richness, and functional group richness) and dung removal rates, comparing perturbed and conserved forests in three regions of Colombia (Caribbean, Andes, and Amazon). We also assess the relationship between dung beetle diversity and dung removal rates. Dung beetle diversity was assessed using pitfall traps, and specimens were measured and assigned to functional groups according to body size and dung relocation strategy. We used exclusion control units and experimental units to assess dung degradation with and without dung beetle activity and evaluate differences in removal rates between two dung removal strategies: paracoprids and telecoprids. Dung removal rates, abundance, and functional group richness were lower in perturbed forests compared to conserved forests. Dung removal increased with abundance, species richness, and functional group richness. Moreover, dung removal performed by telecoprids increased with species richness of telecoprids and paracoprids. Our results evidence a negative effect of human perturbation on dung beetle richness, abundance, and dung removal rates, and also that dung beetle diversity and functional group richness enhance dung removal rates.

Sampling design consisted of a 1,000 m linear transect per site, placed at the center of each forest (i.e., at least 100 m from the border of the forest to minimize any possible edge effect). Each transect included 20 baited pitfall traps (hereafter referred as ‘sampling units’; Figure S1B) placed 50 m from each other (Larsen & Forsyth, 2005; Figure S1B). The distance between our pitfall traps and experimental units is a potential caveat, taking into consideration that some dung beetle individuals could fly up to 1.5-2 km per day (Paik 1976). Each pitfall trap consisted of a 1 L plastic bucket buried at ground level and filled with 500 ml of water mixed with kitchen salt and scentless soap (following Noriega & Fagua, 2009). On top of the bucket, a metallic wire sustained a surgical mesh filled with 300 g of a mixture of fresh pig and human dung (homogenized and mixed in 4:1 proportion; Figure S1B). The amount of dung was chosen to be similar to that produced by large terrestrial mammals in Colombia, and we used the mixture because previous sampling campaigns showed that it is highly attractive to dung beetles in this region (J.A. Noriega pers. obs.). Traps were left in the field for 48 hours. After this period, dung beetle specimens were collected and labeled in individual containers filled with ethanol (70%), to be later identified to species level using different taxonomic identification keys (Edmonds, 1994; Génier, 1996; González et al., 2009; Camero, 2010; Edmonds & Zidek, 2010; Vaz-de-Mello et al., 2011). Voucher specimens were deposited in the Entomological Collection of the Museum of Natural History of Los Andes University (EANDES) and the reference collection of the first author (CJAN).