Long-term trends in corn yields and soil carbon under diversified crop rotations

Abstract

Agricultural practices such as including perennial alfalfa (Medicago sativa L.), winter wheat (Triticum aestivum L.), or red clover (Trifolium pratense L.) in corn (Zea mays L.) rotations can provide higher crop yields and increase soil organic C (SOC) over time. How well process‐based biogeochemical models such as DeNitrification‐DeComposition (DNDC) capture the beneficial effects of diversified cropping systems is unclear. To calibrate and validate DNDC for simulation of observed trends in corn yield and SOC, we used long‐term trials: continuous corn (CC) and corn–oats (Avena sativa L.)–alfalfa–alfalfa (COAA) for Woodslee, ON, 1959 to 2015; and CC, corn–corn–soybean [Glycine max (L.) Merr.]–soybean (CCSS), corn–corn–soybean–winter wheat (CCSW), corn–corn–soybean–winter wheat + red clover (CCSW+Rc), and corn–corn–alfalfa–alfalfa (CCAA) for Elora, ON, 1981 to 2015. Yield and SOC under 21st century conditions were projected under future climate scenarios from 2016 to 2100. The DNDC model was calibrated to improve crop N stress and was revised to estimate changes in water availability as a function of soil properties. This improved yield estimates for diversified rotations at Elora (mean absolute prediction error [MAPE] decreased from 13.4–15.5 to 10.9–14.6%) with lower errors for the three most diverse rotations. Significant improvements in yield estimates were also simulated at Woodslee for COAA, with MAPE decreasing from 24.0 to 16.6%. Predicted and observed SOC were in agreement for simpler rotations (CC or CCSS) at both sites (53.8 and 53.3 Mg C ha−1 for Elora, 52.0 and 51.4 Mg C ha−1 for Woodslee). Predicted SOC increased due to rotation diversification and was close to observed values (58.4 and 59 Mg C ha−1 for Elora, 63 and 61.1 Mg C ha−1 for Woodslee). Under future climate scenarios the diversified rotations mitigated crop water stress resulting in trends of higher yields and SOC content in comparison to simpler rotations.