Quantification of nitrogen mineralization at the field scale is necessary to improve current N fertilizer recommendations. The objective of this study was to identify a suitable set of explanatory variables and an appropriate modelling framework to estimate the average daily amount of N mineralized (DANM) during a corn (Zea mays L.) growing season. In six field experiments in Flat Inland Pampas, Argentina, DANM was evaluated in three different homogeneous zones (HZ), which were defined based on terrain elevation. Mixed-effect generalized linear models were developed that included growing season as random effect as well as HZ, sand, clay, silt, soil organic carbon, and normalized difference vegetation index (NDVI) as fixed effects. Results showed that DANM ranged from 0.19 to 1.25 kg N ha⁻¹ day⁻¹ with a mean of 0.64 kg N ha⁻¹ day⁻¹, which is equivalent to 110 kg N ha⁻¹ for the corn growing season. The best model included HZ, sand, and early NDVI measurements, and predicted DANM with an RMSE of 0.19 kg N ha⁻¹ day⁻¹. Thus, the model should be improved and tested in the Inland Flat Pampas before it could be used to assess N fertilizer application rates.

中文翻译：

---

野外规模表观氮矿化的评估

田间规模的氮矿化量化对于改进当前的氮肥建议是必要的。本研究的目的是确定一组合适的解释变量和合适的建模框架，以估计玉米 ( Zea mays  L.) 生长季节期间的平均每日矿化氮 (DANM) 量。在阿根廷平原内陆潘帕斯草原的六个现场实验中，DANM 在三个不同的均质区 (HZ) 中进行了评估，这些均质区是根据地形高程定义的。开发了混合效应广义线性模型，其中包括生长季节作为随机效应以及 HZ、沙子、粘土、淤泥、土壤有机碳和归一化差异植被指数 (NDVI) 作为固定效应。结果显示 DANM 的范围为 0.19 至 1.25 kg N ha ^-1 第^-1天 ，平均为 0.64 kg N ha ^-1 天^-1，相当于玉米生长季节的110 kg N ha ^-1。最好的模型包括 HZ、沙子和早期 NDVI 测量值，并以 0.19 kg N ha ^-1 天^-1的 RMSE 预测 DANM 。因此，该模型应该在内陆平原潘帕斯草原进行改进和测试，然后才能用于评估氮肥施用率。