Testing DAYCENT model simulations of corn yields and nitrous oxide emissions in irrigated tillage systems in Colorado.

Agricultural soils are responsible for the majority of nitrous oxide (N(2)O) emissions in the USA. Irrigated cropping, particularly in the western USA, is an important source of N(2)O emissions. However, the impacts of tillage intensity and N fertilizer amount and type have not been extensively studied for irrigated systems. The DAYCENT biogeochemical model was tested using N(2)O, crop yield, soil N and C, and other data collected from irrigated cropping systems in northeastern Colorado during 2002 to 2006. DAYCENT uses daily weather, soil texture, and land management information to simulate C and N fluxes between the atmosphere, soil, and vegetation. The model properly represented the impacts of tillage intensity and N fertilizer amount on crop yields, soil organic C (SOC), and soil water content. DAYCENT N(2)O emissions matched the measured data in that simulated emissions increased as N fertilization rates increased and emissions from no-till (NT) tended to be lower on average than conventional-till (CT). However, the model overestimated N(2)O emissions. Lowering the amount of N(2)O emitted per unit of N nitrified from 2 to 1% helped improve model fit but the treatments receiving no N fertilizer were still overestimated by more than a factor of 2. Both the model and measurements showed that soil NO(3)(-) levels increase with N fertilizer addition and with tillage intensity, but DAYCENT underestimated NO(3)(-) levels, particularly for the treatments receiving no N fertilizer. We suggest that DAYCENT could be improved by reducing the background nitrification rate and by accounting for the impact of changes in microbial community structure on denitrification rates.