A framework for capturing indirect impacts in site-level screening for biodiversity risks

This Zenodo record contains data generated for the associated paper published in Methods in Ecology and Evolution: Narain, D., Bedford, J., Grace, E., Muge, A., Rankin, A., Jones, M. I., & Dunnett, S. (2025). A framework for capturing indirect impacts in site-level screening for biodiversity risks. Methods in Ecology and Evolution, 00, 1–11. https://doi.org/10.1111/2041-210X.70162 Code used to generate these data has been archived here. Abstract Site-based industrial operations such as mining, oil and gas extraction, and renewable energy development are associated with many direct and indirect impacts on biodiversity. Consideration of the full range of these impacts when selecting the Area of Influence (AoI) of a project is critical for effective biodiversity risk screening. Indirect impacts, however, are elusive despite often being more extensive than direct ones, both temporally and spatially. There is also limited clarity on the distinction between direct and indirect impacts, leading to the latter either being missed from screening analyses, or misclassified as direct impacts. Here we propose a definition of indirect impacts and a framework for incorporating them in risk and impact analyses. We define indirect impacts as those that are triggered by wider socio-economic and demographic changes induced by the project and not directly by project operations. Indirect impacts manifest through three pathways: increased access to intact ecosystems, increased in-migration and settlement, and increased viability of other economic activity. Literature on the exact spatial extent to which indirect impacts manifest is sparse. A range of factors, however, can act as predictors of the likelihood and intensity of indirect impacts and form the basis of a decision-making framework to select an AoI that captures them effectively. Data There are three gridded datasets, all in Equal Earth (EPSG:8857) projection, at 1km, 10km and 50km resolution: fz_screening_layer_*.tif: six input datasets are transformed to follow a logistic distribution centred on a pre-defined threshold. The mean of these six datasets is then taken to produce a gridded layer with values 0-1 showing the likelihood of needing a larger, precautionary buffer to account for indirect effects when evaluating project impacts. std_screening_layer_*.tif: six input datasets are standardised to a range of 0-1. The mean of these six datasets is then taken to produce a gridded layer with values 0-1 showing the likelihood of needing a larger, precautionary buffer to account for indirect effects when evaluating project impacts. binary_screening_layer_*.tif: six input datasets are transformed to binary TRUE/FALSE grids based on their values vs a pre-defined threshold. If any of the datasets present TRUE for a particular grid cell, the combined layer will also show TRUE, indicating a larger, precautionary buffer is needed to account for indirect effects when evaluating project impacts. For more detailed information on methodology, please see the paper.