Coupled carbon and nitrogen mass balances in agricultural production and implications for global environmental change

Objective: Agriculture is an important contributor to global environmental changes due to, for example, its large share of global greenhouse gases emissions and nitrate leaching. Most available models to assess the... [ view full abstract ]

Objective: Agriculture is an important contributor to global environmental changes due to, for example, its large share of global greenhouse gases emissions and nitrate leaching. Most available models to assess the environmental impacts of agricultural activities are either too details and applicable only locally, or global and lacking resolution and regionalization. We develop a conceptual mass balance carbon (C) and nitrogen (N) model linking agricultural practices, local characteristics (soil, climate) and agricultural outputs and emissions. We use a scale-consistency approach by combining and matching a bottom-up approach, i.e. calculating local/regional C and N mass balances, with a top-down approach, i.e. disaggregating global/national-level data and ensuring that the sum of smaller-scale inputs and outputs is equal to larger scale aggregates.

Method: The land system was divided into four interconnected sub-systems (soil, animal, plant and machinery). The animal sub-system is only applicable to pasture land use classes. All sub-systems had C and N fluxes from and/or to outside of systems boundaries, e.g. GHG emissions to atmosphere and nitrates to groundwater. The N balance was inspired in the model by the Organization for Economic Cooperation and Development. The two models were connected since C and N are chemically interlinked in fluxes and stocks (e.g. C:N ratio in SOM). For the animal sub-system, we used the Intergovernmental Panel on Climate Change framework. The plant sub-system balances plant growth, intakes and emission fluxes to the atmosphere and soil. The machinery sub-system included fuel consumption and emissions to atmosphere from fuel combustion during agricultural operations. Data used had different scales, local (e.g. groundwater leaching), regional (e.g. fertilizer quantity applied by crop) or national (e.g. nitrogen deposition). Local and regional-level flows were aggregated into larger regions using consistent multi-scale data transference. National-level data was disaggregated using allocation rules. The main data sources used were FAOSTAT and Eurostat for agricultural production, and the National Inventory Report from the United Nations Framework Convention on Climate Change for atmospheric emissions.

Results: We provide regionalized C and N balanced fluxes quantified, calculated at cell scale and aggregated at country and global scale, using a scale-consistent approach. The scale-consistency of the model ensures that data can be aggregated at any level of detail. The models highlight the site dependency and the connection between key variables for emissions such as crop yield, soil organic matter (SOM) accumulation in soils, agricultural operations and fertilization. In regions where precipitation tends to be higher, fertilizer application tends to be higher to compensate for losses from N leaching. N application rates are significantly region-specific. The model also highlights the mechanisms of SOM accumulation in grasslands, which is significantly higher than in croplands.

Conclusion: The approach presented here enables scale aggregation and/or desegregation of data which can be transferred consistently from global scale to regions or countries, or to calculate environmental impacts of individual or groups of agricultural products. The analysis was limited by data availability, but despite several constraints and necessary assumptions the results are compatible with prior findings.