A spatially explicit framework to assess the environmental impacts of agricultural production on landscape level

Problem & Aim Global agriculture is facing immense challenges due to an increasing demand by a growing population that is more (over)consuming and wasting a substantial proportion of the food produced [1] while environmental impacts at local and global scales increase [2-4]. Life cycle assessments (LCA) are widely used to evaluate the environmental impacts of agricultural production, addressing energy as well as material flows and their environmental impacts [5-6]. Despite of recent advancements [7-11], LCA approaches generally measure eco-efficiency without taking into account local ecosystem boundaries or cross-scale interactions and therefore do not allow for a final conclusion on the sustainability of an agricultural production system within its specific local context. Furthermore, LCA neglect the inherent linkages between farming systems (e.g. whey in pig production), which only emerge once analyses cover the entire food system. We address those needs with the development of a spatially explicit toolbox for assessing nutrient flows and environmental impacts of agricultural production at landscape level. Proposed methodological framework The toolbox will be based on a gridded representation of agricultural production (Fig.1 - 1) with a LCA-based farm model applied to each grid cell (2). This model captures within-cell flows of agricultural products, quantifies N/P flows through plant as well as feed and livestock production and considers, through a cradle to farm-gate approach, upstream processes, and external inputs. Hence, it allows for a quantitative environmental evaluation of farms, both per hectare and per kg of outputs. The application of this farm model to each grid cell will facilitate a high degree of spatially resolved farming activity data. A data processing method (3) will centrally manage each grid cell with regard to all physical exchanges with other cells. The biodiversity depletion potential [12] will be calculated for each cell based on agricultural land use intensity and landscape structure parameters, such as presence of semi-natural habitats (4). Additionally, maps indicating water quality as well as other landscape functions and ecosystem services will be integrated (4). Through the combination of this model with a global mass-flow model [13-14], we will quantify off-site effects as well as trade-offs and synergies across different scales (5). Expected results The above presented framework will provide a reproducible and innovative modelling approach to assess and optimize agricultural activities in their surrounding landscape. Latter will be firmly linked to their location in ecosystems and consistently embedded in regional to global food system dynamics. The framework will thus allow to explore the regional option space of impact mitigation, including trade-offs and synergies between agricultural land use, ecosystem services, and biodiversity conservation at landscape level. Thus, the approach optimally contributes to regionalised systemically consistent policy making.