C over crop use in the mid-Atlantic and other regions of the United States has expanded in recent decades because they are an eff ective tool for reducing erosion and nutrient leaching and because they are a source of recycled N and fi xed N 2 for ecological nutrient management (Drinkwater and Snapp, 2007;Hively et al., 2015). However, it remains a signifi cant challenge to explain the variability and predict the outcome of N cycling processes across a range of cover crop management practices (Ketterings et al., 2015), leaving farmers with few tools to support adaptive N fertility management when using cover crops. While simple mathematical models have been developed to predict N mineralization from crop residue decomposition in laboratory incubations and greenhouse studies (Vigil and Kissel, 1991; Th orup-Kristensen, 1994), these models have limited utility for N fertility management because they are not calibrated to predict relevant agronomic outcomes, such as crop yield response or N fertilizer replacement value. We have developed a model-data fusion approach that uses easily collected fi eld data to calibrate a simple, ecologically based, lumped-parameter model of the N cycle to predict how cover crops will aff ect the yield of a subsequent corn crop.Cover crops have the potential to aff ect subsequent crop yields and/or N fertilizer requirements through the processes of N mineralization and N immobilization that occur during cover crop residue decomposition, as well as the pre-emptive competition for soil inorganic N that occurs during cover crop growth (Th orup-Kristensen et al., 2003). Th e quantity and directionality (positive or negative) of net N mineralization from cover crop residues depends on the total N content (kg ha -1 ) and the C/N ratio of the cover crop biomass (Vigil and Kissel, 1991;Starovoytov et al., 2010). Because the N content and C/N ratio of cover crop residues can vary widely across different cover crop species and mixtures as well as across diff erent environments and management practices (Cherr et al., 2006; Poff enbarger et al., 2015;Finney et al., 2016), it is diffi cult to generalize the N supply potential of cover crops (Ketterings et al., 2015). Rather, site-specifi c measurements of cover crop biomass C and N contents may be necessary to accurately predict the eff ect of cover crop N cycling on subsequent crop yields and N fertilizer requirements. However, a major constraint to
aBstractOne potential benefi t of cover crops (CCs) is that N mineralization from decomposing CC residues may reduce the N fertilizer requirement of a subsequent crop, but predicting this credit remains a signifi cant challenge. Th is study used a model-data fusion approach to calibrate a model of CC residue N mineralization and pre-emptive competition for soil NO 3 -that occurs during CC growth to predict the yield response of an unfertilized corn (Zea mays L.) crop. Th e model was calibrated with a data set of 199 observations from four CC experiments in central Pennsylvania. Th e most parsimonio...