Boosting the use of spectral heterogeneity in the impact assessment of agricultural land use on biodiversity

No consensus has been yet achieved among Life Cycle Assessment (LCA) practitioners on how to assess the impact on biodiversity due to land uses and land use changes, in particular with regard to agricultural areas. In the domain of nature conservation and landscape ecology, spectral heterogeneity (SH) derived from remotely sensed imagery is considered a viable proxy for species diversity detection. The assessment rationale is based on the 'spectral variation hypothesis': the higher the spectral variability, the higher the ecological heterogeneity and species community diversity, occupying different niches. Our hypothesis is that SH can be effective to improve or complement current Life Cycle Impact Assessment−LCIA practice on biodiversity loss evaluation driven by land use. Hence, we aim here to explore this assumption by computing SH at a local scale of crops cultivation in Southern Alps (Trentino province, Italy), and then combining this information with land use over 30 years. We observe and analyse the relationships between land cover maps and habitat heterogeneity at different time and spatial resolutions. This allows us to argue about the robustness of SH to be a potential surrogate of environmental nuances for species variability detection in LCIA.