Mitigating greenhouse gas emissions in subsurface-drained fields in Eastern Canada

In wet regions subsurface drainage is essential in removing excess water in soil and promoting crop growth; however, it may also result in environmental problems. Implementation of agricultural best management practices (BMPs) on subsurface-drained lands can mitigate environmental problems brought on by human activities and climate change. To provide effective mitigation and adaptation measures for the management of subsurface-drained fields, a quantitative assessment of the impact of water table depth, agronomic management practices, and climate-change-driven rises in greenhouse gas (GHG) emissions on water quality and crop production were assessed through an modeling approach (Root Zone Water Quality Model, RZWQM2).Drawing on a comprehensive hydrological dataset (i.e., tile drainage, sub-irrigation, soil water content, sap flow and crop growth) for calibration, the RZWQM2 then accurately simulated crop growth and growing season drainage. However, the model significantly overestimated winter tile flow, indicating its reliability to be compromised by its imperfect winter drainage process. Implementation of Kalman filter technique successfully enhanced model reliability and reduced predictive uncertainties in simulating winter drainage in cold areas. The revised modelling approach could then serve to evaluate water and field management scenarios for subsurface-drained and irrigated fields.A comparison of the abilities of the RZWQM2 and DNDC models to comprehensively simulate both crop growth and the biogeochemical processes occurring within the soil profile, showed both models to accurately estimate soil temperature, but DNDC to perform poorly in simulating the soil water content (SWC) due to the lack of a heterogeneous soil profile, shallow simulation depth and lack of root density functions for crops. Both models showed similar performances in simulating N2O emissions, with predicted cumulative N2O emissions being with ±15% of measured vales for all four treatments; however, RZWQM2 better estimated CO2 emissions (greater R2, lesser root mean square error). Both models accurately (within ±15%) estimated cumulative growing season drainage; however, RZWQM2 was more accurate in predicting daily drainage and DNDC was not equipped to simulate controlled drainage or sub-irrigation. Both models performed satisfactorily in predicting grain yields of corn and soybean. Overall, RZWQM2 proved to be more applicable to simulating the biogeochemical processes in sub-surface drained fields than DNDC.RZWQM2 was used to evaluate different potential BMP's ability to mitigate GHG emissions in a subsurface-drained corn (Zea mays L.) field under water table management. The optimal range of N fertilization to reduce GHG emissions while maintaining high nitrogen use efficiency and crop yields was identified as 125 to 175 kg N ha-1. Splitting N applications was found to reduce total N2O emissions by 11%. Controlled drainage with subirrigation resulted in 21% greater N2O emissions, but 6% lower CO2 emissions compared to free drainage. A corn-soybean rotation reduced GHG emissions by 20% over continuous corn. Climate change impacts on crop production, water quality and GHG emissions from subsurface drained fields at two sites of Eastern Canada were assessed using RZWQM2. Under future climate scenarios, mean drain flow and N losses through drainage would increase by 23-41% and 47-76%, respectively. The N2O emissions would rise by 21-25% due to greater denitrification and mineralization, while CO2 emissions would rise by 16% due to greater crop biomass accumulation, faster crop residue decomposition, and greater soil microbial activity. These simulations further indicated that future corn yields would decline, while soybean yields would increase in the future, and that climate change would exacerbate environmental pollution by increasing the GHG emissions from croplands and N losses in drainage.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.