2026-10-14T01:55:05Z https://www.scidb.cn/oai

10.57760/sciencedb.45642 2026-08-07T10:38:58Z

2026-08-07 Topographic heterogeneity decouples the diversity–productivity relationship in mammalian assemblages across mountain landscapes doi:10.57760/sciencedb.45642 en Mountains sustain exceptional biodiversity, but the mechanisms behind variable diversity–productivity relationships constitute the “Humboldt enigma”. Topographic heterogeneity may modulate energy-driven diversity patterns, yet its divergent effects on mammalian taxonomic/functional richness and uniqueness remain poorly quantified. Here, we show that topographic heterogeneity fundamentally reshapes productivity–diversity relationships by regulating how energy availability influences mammalian community assembly. Using camera-trapping surveys of medium- and large-sized mammals across the eastern Himalaya, we integrated taxonomic and functional diversity metrics with environmental gradients to test how elevation and terrain complexity modify energy–diversity relationships. We found that terrain roughness amplified the positive effect of net primary productivity (NPP) on species richness, whereas elevation reversed the influence of NPP on functional richness through stronger environmental filtering. Elevation and roughness also strengthened productivity-driven increases in community uniqueness, revealing contrasting controls of energy and topography on biodiversity dimensions. Consequently, richness and uniqueness exhibited pronounced spatial mismatch: species-rich communities were concentrated in productive mid-elevation forests, whereas unique assemblages occurred predominantly in high-elevation and topographically complex habitats. Environmental threshold analyses identified distinct NPP and elevation conditions defining integrated biodiversity hotspots. Our findings reveal a hierarchical topographic regulation of diversity–energy relationships, demonstrating that landscape configuration is essential for explaining mountain biodiversity patterns and for identifying conservation priorities under global environmental change.Data and code used in the manuscript of "Topographic heterogeneity decouples the diversity–productivity relationship in mammalian assemblages across mountain landscapes". Here, we compiled an extensive camera-trapping dataset of medium and large-sized mammals from 830 sites within 30 camera trap grids across the eastern Himalayas. We investigated how topographic heterogeneity affects the diversity–productivity relationship in mammalian assemblages across mountain landscapes.'community_data.RData' includes the records of species in each camera trap grid for each sampling replicate: (1) 'v.1', the non-replicated dataset; (2) 'v.6', the dataset with 6 sampling replicates; (3) 'v.11', the dataset with 11 sampling replicates.'env.RData' includes 'env' and 'envrast'. (1) 'env': the coordinates and the environmental variables of each camera trap grid, including 'lon' (lontitude), 'lat' (latitude), npp (net primary productivity), ele (elevation), rough (terrain roughness), bio4 (temperature seasonality) and pd (patch density). (2) 'envrast': a stacked raster map data of environmental variables across the eastern Himalayas.'env_data.RData' includes all the standardized environment variables used for the ICOM: npp (net primary productivity), ele (elevation), rough (terrain roughness), bio4 (temperature seasonality), pd (patch density) and camera (the number of camera sites). Take NPP as an example. 'npp': The NPP data for all grids; 'npp.1': Non-replicated grid's NPP data; 'npp.6': NPP data of the grids with 6 sampling replicates; 'npp.11': NPP data of the grids with 11 sampling replicates.'traits.rds': Functional traits of each species. 'Scientific': Latin names of species; 'Body.Mass': body mass; 'Diet': consumption proportion of food resources; 'ForStrat': foraging stratum; 'Activity': circadian rhythms; 'Litter.Size': litter size; 'Habitat.Breadth': number of habitat types; 'Dispersal': daily movement distance.R version 4.4.3 were used to run the Bayesian Integrated Community Occupancy Model (ICOM), the Generalized Dissimilarity Modelling (GDM), the Generalized Linear Models (GLMs), as well as the Permutational Multivariate Analysis of Variance (PERMANOVA, 999 permutations), and to produce Figure 2-5 in the manuscript. biodiversity hotspots; camera trapping; community uniqueness; conservation prioritization; energy–diversity theory; functional diversity Yang Liu Xueyou Li RESTRICTED dataset Science Data Bank