Agricultural diversification practices are widely used as a pathway to sustainable crop yield, yet integrated and multidimensional assessments of their trade-offs among ecosystem services remain limited. To investigate these trade-offs, we synthesized 78 published meta-analyses with a reported sample size of 54,426 to evaluate the effect of agroforestry, cover crops, cultivar mixture, crop rotation, and intercropping across biodiversity, carbon sequestration, climate-forcing proxies, soil functioning, water regulation, crop yield, and pest, disease, and weed suppression. Overall, agricultural diversification practices improved biodiversity by 35.08
Sustainable crop production requires management that maintains yields while limiting nitrogen losses and greenhouse gas emissions. Yet national assessments rarely compare cover-crop rotations with prevailing systems across future climates while jointly evaluating productivity, nitrogen leaching, soil organic carbon (SOC), and nitrous oxide (N2O). This limits the identification of regional trade-offs and the design of locally appropriate rotations. To address this gap, we used DSSAT to compare diversified 4-year rotations with legume and non-legume cover crops against business-as-usual rotations across German agro-ecological zones under six climate models and RCP4.5 and RCP8.5. We evaluated grain-unit production, nitrogen use efficiency, nitrogen leaching, SOC change, and net greenhouse gas balance. Direct and indirect N2O emissions were estimated with a nitrogen input- and leaching-dependent IPCC Tier 2 approach; net soil CO2 exchange was inferred from simulated SOC change. Cover-crop rotations increased grain-unit production and nitrogen use efficiency and reduced nitrogen leaching across scenarios and zones. Legume cover crops produced the strongest productivity gains, with cereal yield increases of approximately 16–33
Agricultural productivity in the semi-arid region of central Tanzania is low and projected to further decrease due to climate change, exacerbating the already high levels of food insecurity among the rural population. Agroforestry is actively promoted in the region as an adaptation strategy due to its multiple biophysical and socio-economic benefits, but whether it can buffer future climate change impacts on maize yields has not yet been quantified. Previous modelling studies in central Tanzania focused either on maize monocropping or on agroforestry under current climate. Moreover, it remains unclear whether yield gains from agroforestry constitute genuine adaptation. Here, we show for the first time how a comprehensive evaluation framework can be used to disaggregate projected yield changes into four distinct indicators: climate change impact, agroforestry impact, agroforestry adaptation effect, and climate change impact with adaptation. The Agricultural Production Systems Simulator was used to simulate maize yields at six sites in Dodoma under current climate conditions and under a low- and a high-emission scenario (shared socio-economic pathways 1-2.6 and 5-8.5) for two periods centred on 2030 (2016–2045) and 2050 (2036–2065). Above-ground agroforestry effects were represented by 10
Increases in food production must be sustainable, i.e., respectful of the environment, profitable for farmers, and improve human conditions. Companies sourcing agricultural commodities are uniquely positioned to shape global sustainability outcomes by influencing producers’ decisions, yet current sourcing initiatives are fragmented and may not enable a true transformation toward more sustainable crops. Our objectives were to characterize current sustainable commodity sourcing programs, identify methods and metrics for program support, and synthesize results on sustainability indicators for four major crops (wheat, Triticum aestivum L.; rice, Oryza sativa L.; maize, Zea mays L.; and soybean, Glycine max (L.) Merrill). Five major business models emerged among current sustainable sourcing programs: practice-based, outcome-based, carbon-credit, commodity-premium, and performance-based. The majority of the programs were practice-based or hybrid, with outcome- and performance-based requiring substantially greater verification. Among the scientific literature across four major crops, a majority of studies addressed production and environmental or economic indicators, while fewer examined human or social aspects. The translation of research findings to field applications seems limited by the initial cost of adoption (monetary and human capital), technology availability, and education. Future research should (1) include end-use quality parameters and grain nutritional components in sustainability assessments and (2) increase the robustness of small-scale assessments of social and human condition sustainability. To achieve meaningful impact, sustainable sourcing programs should adopt transparent assessment frameworks to evaluate suppliers using measurable environmental and social indicators, incentivize sustainable practices through premiums, and invest in farmers’ capacity and access to technology. They should also integrate social criteria, e.g., labor conditions and nutrition, collaborate across supply chains to harmonize standards and reduce duplication, and prioritize circular, resilient systems that enhance long-term sustainability. Together, these actions can shift sourcing from a compliance exercise to a catalyst for more resilient, equitable, and sustainable crop production.
Over three decades, the Agricultural Production Systems sIMulator (APSIM) has evolved from a cropping systems modelling framework into a widely used platform for analyzing agricultural systems. This evolution has been underpinned by collaboration among agronomists, soil scientists, crop physiologists, software engineers, and modelers, enabling continuous improvements in process representation, management flexibility, and predictive capability. This Virtual Issue highlights recent developments that demonstrate the maturity, breadth of application, and ongoing evolution of APSIM. The papers showcase advances in representing soil properties, crop responses, management interventions, and spatial heterogeneity, together with integration of remote sensing, machine learning, and high-resolution datasets. Applications span scales from intra-field prediction to national assessments and encompass diverse crops, production systems, and environments across Africa, Asia, Australia, and the United States. Collectively, these studies demonstrate APSIM’s capacity to test hypotheses, evaluate strategies, and extrapolate agricultural responses across environments, scales, and production contexts. The Issue further illustrates how APSIM is being applied to address major challenges facing agriculture, including improving resource-use efficiency, enhancing climate adaptation, reducing greenhouse gas emissions, and understanding genotype × environment × management (G×E×M) interactions for crop improvement. The studies also reflect the transition toward data-rich digital agriculture, where process-based models provide a mechanistic framework for integrating diverse data streams, interpreting observations, and supporting decision-making. Despite these advances, important challenges remain. These include improving soil characterization, refining genotype parameterization, enhancing the representation of soil processes and plant responses, strengthening integration with emerging data sources from genomics, high-throughput phenotyping, remote sensing, and georeferenced soil and climate databases, and maintaining usability while preserving scientific rigor and flexibility. Looking ahead, APSIM is well positioned to contribute to next-generation agricultural decision-support systems by linking phenotyping, genomics, remote sensing, and machine learning within a coherent systems framework, helping guide the design and management of productive, resilient, and environmentally sustainable agricultural systems.
