Agricultural Production Systems Simulation

Agricultural production systems simulation encompasses the use of computational models to represent the complex interactions among crops, soils, climate, management and markets. Such models range in scale from field-level, one-dimensional soil–crop simulators that calculate water and nutrient fluxes, phenology and yield, to regional or global network models that trace trade flows, value chains and environmental footprints. At the field scale, process-based models integrate mechanistic descriptions of soil water balance, carbon–nitrogen turnover and crop physiology—often underpinned by radiation-use-efficiency or photosynthesis–respiration routines—to forecast how changes in planting date, cultivar choice, irrigation or fertiliser rates affect biomass accumulation and harvestable yield. At broader scales, agent-based and multilayer network models capture the decisions of individual producers, processors and traders, and how shocks or policy measures propagate through supply chains. By linking spatial data (soil maps, topography, remote sensing) with long-term weather series and management scenarios, simulation tools support tactical and strategic decision-making: guiding site-specific fertiliser and irrigation applications, evaluating mitigation of greenhouse-gas emissions or nitrate leaching, and exploring trade-off pathways toward sustainable intensification under climate change.

Research from Nature Portfolio

A multilayer network approach has been used to simulate the propagation of a local production shock through both direct trade links and indirect product-conversion pathways. The resulting model quantified how a complete loss of output in a single region can trigger cascading shortfalls in multiple foodstuffs across dozens of countries, underscoring the value of accounting for both trade dependencies and processing chains in resilience planning.

Global flows of nitrogen and phosphorus—both as physical inputs and as virtual nutrients embedded in traded commodities—have been mapped across two decades of international agricultural trade. Analyses reveal that over one-quarter of global nutrient consumption is linked to direct trade, while roughly one-third of total fertiliser inputs are transferred via virtual flows, highlighting opportunities to optimise trade pathways for resource conservation.

A complementary study has revisited global biomass potentials by integrating intensified cropping cycles and spatial reallocation of crops to their most productive areas. Using process-detailed, spatially explicit simulations, it demonstrated that, without further cropland expansion, there remains sufficient potential to meet mid-century demand, provided that cropping intensity and farm management are optimised.

Research from all publishers

In Bangladesh, a stochastic production-frontier model combined with propensity-score matching has been applied to assess the relative efficiency of surface-water and groundwater irrigation in rice systems. Simulations under counterfactual scenarios revealed that while groundwater can enhance short-term productivity and technical efficiency, long-term depletion risks call for policies that also incentivise more sustainable use of surface supplies.

A hybrid agent-based and life-cycle-assessment model for mixed crop–livestock farms in Luxembourg simulated scenarios of reduced stocking rates, altered feed composition and local soybean integration. Results showed that tailored subsidy schemes could cut freshwater eutrophication by up to 25 % and significantly lower climate-change impacts, while maintaining profitability—demonstrating the power of coupling economic and environmental modules in farm-scale simulations.

A purely data-driven agent-based model has been developed to evaluate Portugal’s incentive programme for sown biodiverse pastures. By inferring farmers’ decision rules with machine-learning algorithms and simulating payment scenarios, the model revealed that initial uptake gains were followed by residual dis-adoption effects post-incentive, pointing to the need for policy designs that secure long-term ecosystem benefits.

Agricultural Production Systems Simulation publication trend

The graph below shows the total number of articles in agricultural production systems simulation across all publications each year (not limited to Nature Index journals).

Technical terms

Process-based model: A simulation tool that represents key biophysical processes—such as soil water flow, carbon and nitrogen cycling, and plant growth—based on mechanistic equations.

Radiation-use efficiency (RUE): The ratio of biomass produced per unit of intercepted photosynthetically active radiation, used to convert light capture into growth in many crop models.

Agent-based model (ABM): A bottom-up approach in which individual decision-makers (farmers, herders) are simulated as autonomous ‘agents’ with behavioural rules, whose interactions produce system-level outcomes.

Multilayer network model: A framework that represents different types of relationships—such as trade links and commodity conversion—as overlapping network layers, enabling analysis of direct and indirect propagation of shocks.

Virtual nutrients: Quantities of nitrogen or phosphorus embedded in traded agricultural products, reflecting indirect flows of fertiliser equivalents between producing and consuming regions.

References

1. Shock propagation from the Russia–Ukraine conflict on international multilayer food production network determines global food availability. Nature Food (2023).
2. Physical and virtual nutrient flows in global telecoupled agricultural trade networks. Nature Communications (2023).
3. Global biomass production potentials exceed expected future demand without the need for cropland expansion. Nature Communications (2015).
4. Exploring competitiveness of surface water versus ground water irrigation and their impacts on rice productivity and efficiency: An empirical analysis from Bangladesh. Agricultural Water Management (2023).
5. Sustainable farming strategies for mixed crop-livestock farms in Luxembourg simulated with a hybrid agent-based and life-cycle assessment model. Journal of Cleaner Production (2023).
6. Data-driven agent-based modelling of incentives for carbon sequestration: The case of sown biodiverse pastures in Portugal. Journal of Environmental Management (2023).