ABSTRACT Agricultural production must approximately double to support global population by 2050. Practices optimizing crop yield, nutrient use efficiency (NUE; fraction of applied nutrient taken up by crop) and environmental quality are essential. Nutrient management plans specify needed crop nutrients and practices for mitigating losses, however seasonal dynamics are generally poorly accounted for. Given the profound impact of weather on crop growth and nutrient dynamics, more widespread use of models driven by weather and spatially explicit soil processes offers an opportunity to improve NUE and mitigate farm risk. Maize and sugarcane require large amounts of nitrogen (N) and both advanced tools and routine testing are needed to improve NUE. Real-time crop yield, weather and soil data for precision agriculture is becoming commonplace on large farms. Progress with dynamic simulation models for maize suggest that including weather and soil processes into predictive models improves NUE compared to static N rate recommendation. Whole-farm dairy system models aim to simulate all major biophysical farm components to better quantify dynamic nutrient loss paths. Greater adoption of precision-based soil nutrient management practices offers the potential to enhance agroecosystem NUE, environmental quality and mitigate crop production risk.

中文翻译：

---

土壤养分管理：促进农业生态系统的可持续性

摘要 到 2050 年，农业产量必须大约翻一番才能支持全球人口。优化作物产量、养分利用效率（NUE；作物吸收的施用养分比例）和环境质量的做法至关重要。养分管理计划详细说明了减少损失所需的作物养分和做法，但通常没有充分考虑季节性动态。鉴于天气对作物生长和养分动态的深远影响，更广泛地使用由天气和空间明确的土壤过程驱动的模型为改善 NUE 和减轻农场风险提供了机会。玉米和甘蔗需要大量的氮 (N)，需要先进的工具和常规测试来提高 NUE。实时作物产量，用于精准农业的天气和土壤数据在大型农场中变得司空见惯。玉米动态模拟模型的进展表明，与静态氮肥量推荐相比，将天气和土壤过程纳入预测模型可提高 NUE。全农场乳品系统模型旨在模拟所有主要的生物物理农场组件，以更好地量化动态养分流失路径。更广泛地采用基于精确的土壤养分管理实践，为提高农业生态系统 NUE、环境质量和降低作物生产风险提供了潜力。全农场乳品系统模型旨在模拟所有主要的生物物理农场组件，以更好地量化动态养分流失路径。更广泛地采用基于精确的土壤养分管理实践，为提高农业生态系统 NUE、环境质量和降低作物生产风险提供了潜力。全农场乳品系统模型旨在模拟所有主要的生物物理农场组件，以更好地量化动态养分流失路径。更广泛地采用基于精确的土壤养分管理实践，为提高农业生态系统 NUE、环境质量和降低作物生产风险提供了潜力。