Data for: The Biotic Floor of Old-Growth Forests: Rove Beetles (Staphylinidae) as Sentinels of Soil–Litter Tipping Points After Wildfire

This dataset contains the raw ecological and environmental data collected one year post-fire (2022) in the Acatti old-growth forest (Aspromonte National Park, Italy). The study investigates the impact of wildfire severity, quantified via the differenced Normalized Burn Ratio (dNBR), on the assemblages of Rove Beetles (Coleoptera: Staphylinidae). The file Staphylinidae_Data.xlsx is organized into the following components: • Taxonomic Data: Abundance and richness counts for 1,741 specimens across 34 species, identified from 30 pitfall traps distributed along a fire severity gradient (Burned, Transition, and Unburned sites). • Environmental Covariates: Site-specific radiometric fire severity values (dNBR) derived from Sentinel-2 imagery. • Soil Biochemical Indicators: Measurements of Soil Organic Matter (SOM), Dehydrogenase activity (DHA), and Catalase activity (CAT) for each sampling unit. • Ecological Traits: Information on species-specific traits, including dispersal ability (fliers vs. flightless) and habitat specialization (old-growth specialists vs. generalists) used for IndVal and multivariate analyses. These data support the "biotic floor" hypothesis, identifying a critical ecological threshold at dNBR ≈ 0.25–0.50 where soil biodiversity suffers functional homogenization. Keywords biotic floor; Staphylinidae; dNBR; old-growth forest; wildfire severity; soil bioindicators R Script for ReproducibilityThe file Analysis_Scripts.pdf provides the fully annotated R code used to perform the statistical analyses and generate the figures presented in the manuscript. It is structured into the following sections: Environment Setup: Loading of required libraries including glmmTMB, vegan, and indicspecies. Data Pre-processing: Steps for month assignment, species filtering (excluding taxa <1% for multivariate tests), and matrix preparation. Statistical Modeling: Implementation of the GLMM (Negative Binomial) to model the abundance decay ($\beta = -0.971$) in relation to dNBR. Community Analysis: Code for NMDS ordination and PERMANOVA to test the impact of fire severity and soil biochemical drivers (Catalase and DHA) on community divergence. Bioindicator Identification: Application of the IndVal method to identify sensitive specialists and resilient taxa.