Simulating soil carbon sequestration, yield, and N2O fluxes with DayCent under long-term no-till and cover crop-based cotton cropping system

Authors: DHALIWAL, JASHANJEET - University Of Tennessee; Del Grosso, Stephen; SAHA, DEBASISH - University Of Tennessee

Submitted to: Agriculture, Ecosystems & Environment
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/21/2025
Publication Date: 8/27/2025
Citation: Dhaliwal, J.K., Del Grosso, S.J., Saha, D. 2025. Simulating soil carbon sequestration, yield, and N2O fluxes with DayCent under long-term no-till and cover crop-based cotton cropping system. Agriculture, Ecosystems & Environment. 395. Article e109926. https://doi.org/10.1016/j.agee.2025.109926.
DOI: https://doi.org/10.1016/j.agee.2025.109926

Interpretive Summary: DayCent is an agro-ecosystem model used to estimate crop yields and nutrient losses at farm, regional, national and global scales. However, it is necessary to compare model outputs with plot level data from research sites to verify model results and identify weaknesses to spur improvement. ARS scientist and colleagues from the University of Tennessee compiled model input data and conducted simulations of a long term (> 40 year) cotton cropping study investigating the impacts of synthetic fertilizer additions and hairy vetch cover cropping. DayCent properly represented lint yields for the treatments receiving fertilizer but under-estimated yields when fertilizer was not applied. Both observations and DayCent showed higher levels of soil carbon with fertilizer addition and cover cropping but the model over-estimated carbon gains from these practices. The model matched the observed low nitrogen gas losses when fertilizer was not applied but over-estimated the observed higher losses with fertilizer. Results suggest that soil nitrogen process algorithms need improvement to accurately represent management interventions in cotton systems.

Technical Abstract: Cover crops are widely promoted for improving soil health through carbon sequestration and show promise as a natural climate solution by reducing greenhouse gases through atmospheric carbon dioxide storage. However, their effect on nitrous oxide (N2O) emissions is more variable and contrasting compared to their influence on soil organic carbon (SOC) changes. To effectively represent cover crop systems and estimate the associated N2O emissions, process-based models must be rigorously evaluated. This study aims to assess the ability of DayCent to simulate cotton lint yield, SOC, and N2O emissions in long-term no-till (NT) cotton cropping systems with cover crops in the Southeastern US. The model was evaluated using long-term data on cotton lint yield and SOC and short-term data on soil mineral N, and N2O emissions from NT cotton plots with two N rates (0 kg N ha'¹, NF, and 67 kg N ha'¹, F) and two cover crop treatments (hairy vetch, HV, and no cover crop, NC). DayCent accurately simulated the effect of N fertilizer on cotton lint yield but underestimated yield in non-fertilized treatments, failing to capture the impact of cover crops in these systems. The model accurately captured the long-term impact of cover crops on SOC in non-fertilized treatments but overestimated SOC in fertilized treatments, regardless of cover crop inclusion. Simulated mean soil ammonium levels were overestimated in fertilized treatments and underestimated in non-fertilized treatments, while nitrate levels were consistently lower than measured values across all treatments. Regardless of cover crop presence, DayCent accurately predicted cumulative N2O emissions in unfertilized treatments but overestimated N2O in fertilized treatments. DayCent performance in cover cropping systems could be improved by reducing denitrification rates and enhancing its ability to simulate soil mineral N.