Revealing trade-offs in cropping systems sustainability by piecing together pedo-climatic datasets and agronomic knowledge with fuzzy logic

Agriculture is among the sectors most affected by climate change and, simultaneously, is the primary responsible for anthropogenic non-CO₂ greenhouse gas (GHG) emissions worldwide. Nonetheless, the actual estimation of global GHGs emissions due to agricultural activities is still subject to significant uncertainty.  The Intergovernmental Panel on Climate Change (IPCC) and the Water Footprint Network (WFN) provide a set of equations and emission factors, the so-called Tier 1, for separately estimating carbon stock changes, N₂O emissions, and NO₃ leaching losses under a limited set of pedo-environmental conditions and agricultural Soil Management Strategies (SMS, i.e., crop choice, fertilization, irrigation, and soil tillage). However, the Tier 1 methodologies do not explicitly consider the multi-faceted interactions among SMS, pedo-environmental variability, and nitrogen and carbon fluxes. Indeed, current research demonstrated that increased N₂O emissions and NO₃ leaching due to these interactions could offset potential climate-positive effects due to carbon farming.

A Europe-wide policy approach is urgently needed to assess the effects of SMS on nitrogen and carbon fluxes and fulfil the Sustainable Development Goals of the UN's 2030 Agenda. ∑ommit project, financed by the European Joint Programme SOIL, aims to generate a robust indicator to predict trade-offs and synergies between soil carbon sequestration, GHGs emissions, and N leaching losses to identify the best SMS according to their application context. The project harmonized multiple data sources as the 2019 Refinement to the 2006 IPCC Guidelines for National Greenhouse Gas Inventories (IPCC 2019), the WFN Guidelines for grey water footprint accounting (Franke et al 2013), and the Italian National Institute of Statistics (ISTAT), in order to estimate the effects of SMS on four trade-off components: crop yield, soil carbon sequestration, N₂O emission, and NO₃ leaching losses. Each row of the data frame, the case scenario, represented a combination of soil × climate × SMS, whose effect on the trade-off components was either inferred from the Tier 1 methodologies or derived from meta-analyses and experts’ opinions. Trade-off components were aggregated into a composite index (∑ommit index) using a fuzzy logic model. The index ranges from 0 (bad) to 1 (good), providing a synthetic and comprehensive measure of the environmental impact of SMS in different pedo-climatic conditions. Besides giving an immediate measure of the system performance and facilitating the comparison of alternative SMS, the index lends itself to be disentangled to trace the effect of specific management practices on key aspects of system sustainability. The trade-off system has been also tested with data from long-term experiments and model simulations (STICS and Armosa) to identify the best-performing SMS in different application contexts. The system demonstrated its sensitivity to modulate the expression of trade-off components coherently using simulated and observed data as input. The aggregation procedure smoothed the magnitude of the model errors in reproducing experimental data, leading to lower uncertainty in the index values than in the single trade-off components. This result paves the way for an advised use of crop models to speed up the identification of the best SMS and design more sustainable cropping systems.