Legume cover crops can supply a significant amount of nitrogen (N) for cash crops, which is particularly important for organic farmers. Because N mineralization from cover crop residue depends on the amount of biomass, cover crop quality, as well as environmental conditions such as soil moisture and temperature, predicting the amount of N mineralized and the timing of release has been difficult. We have developed a Cover Crop Nitrogen Calculator based on the N subroutine of the CERES crop model and evaluated the use of the predicted N credits on yields of fall broccoli [Brassica oleracea L. (Italica group)] at a research farm and two working farms. Research farm trials consisted of a cowpea (Vigna unguiculata L. Walp.) cover crop and no cover crop treatments, each at four N rates (0N, 0.5N, 1N and 1.5N, with 1N the target N rate of 112 kg N ha−1 in 2013 and 168 kg N ha−1 in 2014 and 2015) in a randomized complete block design. On-farm trials consisted of a cowpea or sunn hemp (Crotolaria juncea L.) cover crop at four N rates (0N, 0.5N, 1N and 1.5N) and no cover crop treatment at the 1N rate in a completely randomized design. Cover crop biomass and quality (N%, carbohydrates%, cellulose% and lignin%) were measured and used with a 5-yr average soil moisture and soil temperature from a local weather station to predict an N credit. In the cover crop treatments, the N rate was modified by the predicted credit, while the no cover crop treatment received the full N fertilizer rate either as feathermeal (certified organic fields) or as urea (conventional field). At the research farm, broccoli yield increased up to the 0.5N rate, and there was no difference in yield between the no cover 0.5N rate and the cover crop 0.5N rate in 2013, 2014 and 2105. On-farm, we saw an N response in two site-years. In these site-years, there was no difference between the no cover 1N rate and the cover crop 1N rate. At the third site-year, no N response was seen. Overall, our results showed using the cover crop credit predicted by the Calculator did not reduce yields. The use of a decision support tool such as the Calculator may help farmers better manage N fertilizer when cover crops are used, and increase cover crop adoption.

中文翻译：

---

使用覆盖作物 N 计算器进行适应性氮肥管理：概念验证

豆科覆盖作物可以为经济作物提供大量的氮 (N)，这对有机农民尤为重要。由于覆盖作物残茬的 N 矿化取决于生物量、覆盖作物质量以及土壤水分和温度等环境条件，因此很难预测 N 矿化量和释放时间。我们基于 CERES 作物模型的 N 子程序开发了一个覆盖作物氮计算器，并评估了预测的 N 信用对秋季西兰花产量的使用[甘蓝L. (Italica group)] 在一个研究农场和两个工作农场。研究农场试验包括豇豆（豇豆L. Walp.) 覆盖作物和不覆盖作物处理，每种处理采用四种 N 比率（0N、0.5N、1N 和 1.5N，1N 的目标 N 比率为 112 kg N ha-12013 年和 168 kg N ha-12014 年和 2015 年）在随机完整区组设计中。农场试验包括豇豆或太阳麻（芥菜L.) 在完全随机设计中以四种 N 比率（0N、0.5N、1N 和 1.5N）覆盖作物和以 1N 比率不覆盖作物处理。测量覆盖作物的生物量和质量（N%、碳水化合物%、纤维素% 和木质素%），并与当地气象站的 5 年平均土壤湿度和土壤温度一起使用，以预测 N 信用。在覆盖作物处理中，N 率被预测信用修改，而无覆盖作物处理以羽毛粉（经认证的有机田）或尿素（常规田地）的形式获得全氮肥。在研究农场，西兰花产量增加到 0.5N 率，2013 年、2014 年和 2105 年无覆盖 0.5N 率与覆盖作物 0.5N 率之间的产量没有差异。在农场，我们看到两个站点年的 N 响应。在这些网站年代，无覆盖 1N 率和覆盖作物 1N 率之间没有差异。在第三个站点年，没有看到 N 响应。总体而言，我们的结果表明，使用计算器预测的覆盖作物信贷不会降低产量。使用计算器等决策支持工具可以帮助农民在使用覆盖作物时更好地管理氮肥，并提高覆盖作物的采用率。