Modeling impacts of N sources on N2O emissions in long-term crop rotation system in Brazil’s Cerrado using DNDCv.CAN

• DNDCv.CAN model fairly simulated the N 2 O emissions in 10-yr crop rotation in Brazil • N application increased N 2 O emissions, in which the model had similar pattern • DNDCv.CAN effectively predicted higher NH 3 losses from UR than other N sources • Soil NH 4 + and NO 3 - , WFPS and grain yield were acceptably simulated by DNDCvCAN • The WFPS dynamics for tropical soils may need further adjustments in the model Predicting N 2 O emissions can help identify crop production practices and environmental conditions for mitigation. The objectives were to calibrate and validate the DeNitrification-DeComposition (DNDCv.CAN) model to simulate N 2 O emission in crop rotation (CR) systems in Brazil’s Cerrado biome, using different N fertilizer sources. Field data were obtained from a CR experiment with maize and common beans cultivated under no-till for 10 years. Treatments were: Zero-N, calcium ammonium nitrate (CAN), urea (UR), and ammonium sulfate (AS), with 5 replicates. Measured data including N 2 O fluxes, NH 3 volatilization, soil NH 4 + and NO 3 - , water-filled pore space (WFPS), and grain yields were used to calibrate and validate the DNDCv.CAN. The model acceptably captured the daily N 2 O emissions associated with different N sources, although episodic peaks were over or under-predicted. Correlation coefficient was 0.62 during calibration, with a mean absolute error of 17 g N 2 O-N ha -1 day -1 , close to average measured (18 g ha -1 d -1 ), and simulated (17 g ha -1 day -1 ) emissions. Accumulated measured N 2 O-N emissions were 1.19, 4.38, 3.17, and 2.56 kg N 2 O-N ha -1 for Zero-N, CAN, UR, and AS, respectively, whereas the simulated were 1.13, 3.44, 2.24, and 3.32 kg N 2 O-N ha -1 . The NH 3 , NH 4 + , NO 3 - , WFPS, and yields were also fairly simulated. Soil hydrologic parameters were adjusted in the model using a built in pedotransfer function to improve the simulations, which should be further investigated. The DNDCv.CAN effectively simulated cumulative N 2 O emissions from different N sources applied to CR under tropical conditions, making it valuable for evaluating potential emissions and mitigation strategies.