Files
Abstract
Although agriculture contributes to four main drivers of biodiversity loss, impact assessment of food products remains limited either to in situ measurements that prevent generalization, or to systematic models that are not validated by in situ data. Here we describe the BVIAS (Biodiversity Value Increment from Agricultural Statistics) model, which allows estimating the biodiversity impact of all major food products based on accountancy data and public statistics. BVIAS is calibrated based on the most relevant large-scale studies and meta-analysis. It is then used to find out whether major Food Quality Schemes (FQSs) have different practices and biodiversity impact than their conventional counterparts. We show that only mandated FQS specifications lead to significant practice differences. Consistent with in situ data, organic farms, as well as those producing Comté (Protected Designation of Origin), have less biodiversity impact on a per hectare basis. This local benefit is offset by lower yields, resulting in a higher impact per ton. However, biodiversity impact gap between animal and plant products (e.g., milk vs. wheat) is far greater than the difference between FQS and conventional versions of the same product. Taking into account the main drivers of biodiversity losses related to agriculture, relying on quantitative data for a large sample of farms and calibrating our model based on relevant large-scale studies and meta-analysis, we therefore propose here an objective, robust and operational method to estimate the impact of food products on biodiversity for use in environmental labeling schemes or other purposes.