Spatio-temporal patterns of arthropod leaf-litter communities in protected Mexican tropical deciduous forests revealed by metabarcoding

Resumen

Leaf-litter arthropods represent an essential component of most terrestrial ecosystems, yet comprehensive studies of their species richness and composition remain considerably scarce due to the Linnean shortfall, i.e. the vast number of undescribed species that hampers accurate biodiversity assessments. Here, we explored the spatial and temporal diversity and structure of arthropod leaf-litter communities inhabiting two protected areas of tropical deciduous forest in Mexico using metabarcoding: the Chamela–Cuixmala Biosphere Reserve and Huatulco National Park. As this biome is characterized by marked seasonality, leaf-litter sampling was conducted during both the rainy and dry seasons. A total of 18,885,578 high-quality reads of 418 bp were obtained for the animal DNA barcoding region. Clustering at 97% similarity and subsequent taxonomic assignment yielded 1238 MOTUs assigned to Arthropoda, with diversity estimators indicating that ~66% of the total expected biodiversity was sampled. Insecta was the most species-rich class (~69% of sampled MOTUs), followed by Arachnida (~23%). Formicidae (Hymenoptera), Cecidomyiidae (Diptera) and Chrysomelidae, Curculionidae and Staphylinidae (Coleoptera) were the most species-rich insect taxa, whereas Araneae and Acari dominated Arachnida, highlighting their key functional roles in decomposition, nutrient cycling, and predation in tropical deciduous forest leaf-litter communities. Significant differences in MOTU richness were found between the two regions, with higher diversity during the rainy season in both. Community composition varied more strongly than species richness, indicating that intra-annual variation primarily reflects species turnover rather than net gains or losses. Around 36.5% of the MOTUs detected were also recovered in a previous metagenetic study of these forests using Malaise trap sampling. Our results thus underscore the complementary nature of ground and aerial sampling and the influence of seasonality and spatial heterogeneity in shaping arthropod assemblages. These findings also emphasize the urgent need to expand metabarcoding reference libraries and baseline datasets for leaf-litter arthropod communities, both to overcome the Linnean shortfall and to strengthen long-term biodiversity monitoring and conservation in tropical deciduous forests.