Linking regional and global functional trait data: insights from mammal communities in a fragmented Atlantic Forest landscape

Accurate functional trait data are essential for understanding ecosystem services and processes in fragmented landscapes. We evaluated whether the global EltonTraits 1.0 database adequately represents the functional structure of mammal communities in forest fragments and restoration sites in a highly fragmented Atlantic Forest landscape. We used local frequency of occurrence data obtained at the landscape level via camera traps and sand plots for 30 mammal species and compared regionally compiled trait values (from 82 studies) with corresponding values from the global database. We focused on three key traits (diet, foraging stratum, and activity cycle), and assessed the associations between regional and global trait sets using species‐level functional uniqueness (K¯i), community‐weighted means (CWM), functional diversity using Rao's Q, community‐level functional uniqueness (U), and functional richness (FRic). Comparisons between regional and global trait datasets demonstrated that: 1) 85% of species in fragments and 77% in restoration forests showed higher K¯i with regional traits; 2) CWM values differed significantly, with regional data capturing a broader range of ecological strategies across habitats; 3) global data consistently underestimated Rao's Q and U in both habitats, suggesting that trait convergence in global datasets masks regional‐scale variation; and 4) FRic differed significantly between datasets, suggesting greater niche occupancy when using regional trait data. These discrepancies likely arise because EltonTraits 1.0 averages many trait values across largely pristine ecosystems, reducing niche space variation, while regional communities face multiple effects of environmental filtering. Our findings indicate that reliance on global trait databases can lead to underestimation of the functional roles of specialized species in fragmented ecosystems, highlighting the need to incorporate regional trait information into conservation planning in tropical landscapes. Our results also highlight the importance of integrating regional trait information into functional ecology metrics, in order to avoid biased assessments of community structure.