A globally influential area-condition metric is a poor proxy for invertebrate biodiversity

There is increasing demand for standardised, easy-to-use metrics to assess progress towards achieving biodiversity targets and the effectiveness of ecological compensation schemes. Biodiversity metrics based on combining habitat area and habitat condition scores are proliferating rapidly, but there is limited evidence on how they relate to ecological outcomes. Here, we test the relationship between the statutory biodiversity metric used for Biodiversity Net Gain (BNG) in England — and as the basis for new biodiversity credit systems around the world — and invertebrate richness, abundance, and community composition. We find that the combined area-condition BNG metric does not capture the value of arable farmland and grassland sites for invertebrate biodiversity: invertebrate communities were highly variable across sites that had the same type and condition under the BNG metric. We found no reliable relationship between scores under the metric and either invertebrate abundance or species ..., , # A globally influential area-condition metric is a poor proxy for invertebrate biodiversity

Dataset DOI: [10.5061/dryad.p5hqbzm1b](10.5061/dryad.p5hqbzm1b)

## Description of the data and file structure

The data was collected across a range of parcels within different sites using pitfall traps. Pitfall traps were used to sample ground invertebrate abundance and species richness. In year 1 (2022), just carabid species richness was analysed. In year 2 (2023) DNA metabarcoding was used to look at total ground invertebrate species richness. Invertebrate richness and abundance was analysed against England's statutory biodiversity metric scoring (distinctiveness*condition scoring). 

## Files and variables

### File: JAPPL-2025-00285-Data_v2.xlsx

### **Description:** Datasheets used in the R code for analysis.

### Sheets

#### **Sheet 1: Y1 data:**

Data on ground invertebrate abundance and richness for analysis from year 1 (2022). Includes distinctiveness*condition scoring values from t...,