High scatter-hoarding intensity, phylogenetic relatedness, and body-size similarity increase species co-occurrence of sympatric rodents in subtropical forest fragments

Species co-occurrence has been widely investigated in various species, but the roles of trait similarity and environmental factors are not fully investigated. Here, we examined the effects of species traits (i.e., phylogenetic relatedness, body size, diet similarity, and seed-hoarding behaviors) and environmental factors (i.e., patch size, seed richness, seed abundance, and stand age) on species co-occurrence probability among sympatric rodent species with seven-year field data using live and camera traps in a subtropical forest. We found species co-occurrence probability of rodents was significantly and positively associated with phylogenetic relatedness, body size similarity, diet similarity, and scatter-hoarding intensity of pairwise rodent species using live traps at the species-pair level (such correlations were significant only for phylogenetic relatedness and diet similarity using camera traps). At the community level, species co-occurrence probability of rodents was significantly and positively correlated with patch size using camera traps, and significantly and negatively correlated with stand age, seed richness, and seed abundance using live traps. Our results suggest that species traits and environmental factors are important in shaping species co-occurrence patterns of rodents in forest ecosystems, but the underlying mechanisms may be multiple and, thus, not necessarily equivalent to species coexistence. The effects of phylogenetic relatedness and diet similarity can be well explained by the environmental filtering hypothesis, while the effects of body size similarity, patch size, seed richness, and seed abundance can be well explained by the niche or competition hypothesis, and the effects of scatter-hoarding intensity can be explained by the reciprocal pilferage hypothesis. Besides, the estimation of species co-occurrence probability is highly dependent upon the methodologies because animals may have different responses or sensitivity to different survey approaches.