CAREER: Quantifying Multi-Scale Climate-Smart-Agriculture Management for Triple Wins in Food production, Climate Mitigation, and Environmental Sustainability

密西西比河拥有世界第三大流域，是世界上产量最高的农业区之一，产量占美国玉米和大豆总产量的80%，占全国农产品出口的92%。大规模工业化农业带来了显著的社会经济收益，但代价是该地区的环境代价(土壤侵蚀、养分污染和水生酸化)。气候智能农业(CSA)管理实践被提出作为这些成本的解决方案，因为它们不仅可以增加作物产量，还可以减少温室气体排放，并维持土壤和水的质量。然而，CSA实践的有效性在不同的气候和土地利用条件下各不相同，并涉及紧密耦合的碳、水和养分循环。这些相互作用还没有得到很好的研究，这种知识差距阻碍了对CSA做法的理解和有效应用，以实现提高粮食产量、缓解气候变化和环境可持续性的好处。该项目的总体目标是开发一个综合的生态系统监测、建模和机器学习框架(EcoM3)，其中包括实地观测、卫星遥感数据、基于过程的建模和深度学习方法，以便在多个尺度上系统地研究CSA实践(免耕和覆盖作物)对关键农业生态系统指标(作物产量、土壤碳储量、温室气体和碳/氮淋失)的具体影响。该项目将使用肯塔基州的一个长期现场现场(连续观测超过50年)作为一个试验场，调查从每日到季节、年度、十年尺度的CSA实践效果；检查密西西比河流域不同气候和土壤条件下多个地点的不同CSA影响；并预测CSA实践在整个河流流域尺度上的潜在影响。多尺度数据和模型结果将被集成到EcoM3框架的学习平台中，以与不同的利益相关者和政策制定者交流时间和空间CSA的有效性。这项研究解决了一个具有挑战性的问题：增强的系统方法是否足以促进我们对农业生态系统、气候和环境系统之间相互关联的关系的理解，从而使我们能够同时管理多个目标(粮食安全、碳封存和环境可持续性)？这项研究是一种系统的方法，在不同的气候和土壤条件下，从地点和区域两个尺度上调查CSA措施在农业系统中的综合效应。拟议的EcoM3框架纳入了CSA管理，旨在促进对气候、土地利用/管理和生态系统之间的相互作用和反馈循环的概念和操作方面的理解。这项研究的产品将改进农业生态系统在环境系统模型中的机械表示，以更准确地预测生物地球化学循环和未来的气候变化，并将为农民提供可行的建议，并为制定关于建设可持续和气候适应性农业的循证政策提供科学依据。研究成果将通过当地推广会议和多州农民峰会(密西西比河流域各地区的代表)与农民交流。项目产品将提高人们对CSA管理在建立气候适应性农业生态系统和保护土壤和水健康方面的重要性的认识。多尺度数据集将公开用于研究和教育。该项目由CBET环境可持续发展计划和既定的激励竞争研究计划(EPSCoR)共同资助。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

The Mississippi River has the third-largest drainage basin and represents one of the most productive agricultural regions in the world, yielding 80% of US total corn and soybean production and 92% of the nation’s agricultural exports. Large-scale industrial agriculture has led to significant socio-economic gains, but at environmental costs (soil erosion, nutrient pollution, and aquatic acidification) in this region. Climate-smart agriculture (CSA) management practices have been proposed as solutions to these costs, as they not only increase crop yield, but also reduce greenhouse gas emissions, and sustain soil and water quality. However, the effectiveness of CSA practices varies under diverse climate and land use conditions and involves tightly coupled carbon, water, and nutrient cycles. These interactions have not been well studied, and this knowledge gap has hindered understanding and efficient application of CSA practices to achieve the benefits of enhancing food production, climate mitigation, and environmental sustainability. The overall goal of this project is to develop an integrated ecosystem monitoring, modeling, and machine learning framework (EcoM3) that incorporates field observations, satellite remote sensing data, process-based modeling, and a deep-learning approach to systematically investigate specific effects of CSA practice (no-tillage and cover crops) on key agroecosystem indicators (crop yield, soil carbon storage, greenhouse gases, and carbon/nitrogen leaching) at multiple scales. This project will use a long-term field site in Kentucky (continuous observations over 50 years) as one testing site to investigate CSA practice effects from daily to seasonal, annual, decadal scales; examine varied CSA effects at multiple sites with diverse climate and soil conditions across the Mississippi River basin; and predict the potential impacts of CSA practices at the entire river basin scale. Multi-scale data and model results will be integrated into the learning platform of the EcoM3 framework to communicate temporal and spatial CSA effectiveness with diverse stakeholders and policy-makers.This study addresses a challenging question: Will an enhanced systems approach advance our understanding of the interconnected relationships among agroecosystems, climate, and environment systems sufficiently to allow us to simultaneously manage multiple goals (food security, carbon sequestration, and environmental sustainability)? This study represents a systematic method to investigate the comprehensive effects of CSA practices in agricultural systems at both site and regional scales under heterogeneous climate and soil conditions. The proposed EcoM3 framework incorporates CSA management that is targeted to advance conceptual and operational understanding of interactions and feedback loops among climate, land use/management, and ecosystems. Products derived from this study will improve the mechanistic representation of the agroecosystem in Environmental System Models toward a more accurate prediction of biogeochemical cycles and future climate change and will provide viable recommendations for farmers and a scientific basis for making evidence-informed policy about building sustainable and climate-resilient agriculture. Research findings will be communicated with farmers through local extension meetings and the Multi-state Farmer Summit (representatives across regions in Mississippi River basin). Project products will enhance awareness about the importance of CSA management in building climate-resilient agroecosystems and preserving soil and water health. Multi-scale datasets will be made publicly available for research and education.This project is jointly funded by the CBET Environmental Sustainability program and the Established Program to Stimulate Competitive Research (EPSCoR).This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.