Collaborative Research: Conservation Tillage for Sustainable Food, Energy and Water Systems: Linked Econometric and Process-based Models

粮食-能源-水系统（FEWs）的可持续性主要取决于农业能否在继续提高作物产量的同时减少其环境和生态足迹。保护性耕作因其具有减少土壤侵蚀、提高土壤质量和农业生产力、节约能源和水、改善水质等优点而受到推崇。然而，CT的采用在美国停滞不前，在包括中国在内的许多其他国家仍然很低，部分原因是对CT的经济和环境影响了解不足，以及农民的行为差距和偏离最佳。该项目旨在采用综合系统方法，将基于过程的模型与基于观测数据的计量经济模型联系起来，这些模型可以模拟耕作对FEWs影响的复杂生物物理和生物地球化学相互作用，这些模型包含了农民的行为和决策。利用中美两国巨大的异质性，将利用美国玉米带实验田数据和数十年的观测数据，以及中国东北丰富的实验田数据和新获得的观测数据对关联模型进行校准。对相关模型的模拟将生成和汇总空间和时间上明确的知识和信息，以便农民更好地了解与生态环境保护有关的生态环境保护相关的最佳实践，使政府能够设计更有效的目标政策，以实现考虑环境外部性的社会最优，并最终实现生态环境保护的最佳采用和推广。该项目涉及中美两国团队之间的国际和多学科合作，将推进一套基于前沿科学和行为研究的FEWs背景下的跨学科CT模型，整合来自不同地区的数据，并适用于广泛的生物物理和社会经济环境。该项目将填补关于CT在FEWs环境中影响的重要知识和模型空白。CT的影响分析大多是通过对照实验研究和基于过程的仿真模型进行的，没有充分考虑农民的行为。使用观测数据的计量经济模型可以揭示农民的行为，但不能做出“样本外”的预测。更重要的是，基于过程的CT模型和计量经济学模型并没有结合生物物理和行为过程。拟议的研究将以新颖的方式将计量经济学模型和基于过程的模拟模型联系起来，以便更好地理解农民的次优行为和决策。相关模型将支持政策情景分析和模拟CT对FEWs的动态影响，针对每种类型的现场天气-政策组合，以了解CT的采用和影响。实地一级的分析可以按比例扩大和汇总到更大的范围，如县和地区。链接模型将提供更可信和可靠的产出，通过空间明确的CT采用和实施建议更好地告知农民，并为社会最优CT采用和实施提供更有效的政策建议。该项目将提出关于最佳CT实践的实用建议和政策，并将其传播给包括美国和中国以及世界其他地区的农民和决策者在内的利益相关者。该研究将被纳入美国和中国院校的各种本科和研究生课程和课程。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

The sustainability of food-energy-water systems (FEWs) depends critically on whether agriculture can reduce its environmental and ecological footprints while continuing to increase crop yields. Conservation tillage (CT) has been advocated as having advantages of reducing soil erosion, improving soil quality and agricultural productivity, saving energy and water, and improving water quality. However, the adoption of CT has stalled in the U.S. and remains low in many other countries including China, due partly to insufficient understanding of CT’s economic and environmental impacts and to farmers’ behavioral gaps and deviations from the optimum. This project aims to take an integrated systems approach to link process-based models that can simulate complex biophysical and biogeochemical interactions of tillage effects on FEWs, with observational data-based econometric models that incorporate farmer behaviors and decisions. The linked models will be calibrated with experimental field data and decades of observational data in the U.S. Corn Belt, and rich experimental field data and newly available observational data in Northeastern China, taking advantage of the vast heterogeneity between the two countries. Simulation of the linked models will generate and aggregate spatially and temporally explicit knowledge and information for farmers to better understand the best practices associated with CT in relation to FEWs, for governments to design more effective targeting policies to achieve a social optimum that accounts for environmental externalities, and ultimately for optimal adoption and diffusion of CT. The project, involving international and multidisciplinary collaboration between teams from U.S. and China, will advance a set of interdisciplinary models of CT in FEWs contexts that are grounded on frontier science and behavioral studies, integrate data from vastly different regions, and are applicable to a wide range of biophysical and social economic settings.This project will fill important knowledge and model gaps about the impacts of CT in FEWs settings. Impact analyses of CT have mostly been conducted through controlled experiment studies and process based simulation models, without taking full account of farmer behavior. Econometric models using observational data can reveal farmer behaviors, but cannot make “out-of-sample” predictions. More importantly, process-based models and econometric models on CT have not been linked to incorporate both biophysical and behavioral processes. The proposed research will link econometric models and process-based simulation models in novel ways that allow better understanding of farmers’ suboptimal behavior and decisions. The linked models will enable policy scenario analysis and simulation of dynamic impacts of CT on FEWs, for each type of field-weather-policy combinations, to understand CT adoption and impacts. The field level analysis can be scaled up and aggregated to greater scales such as counties and regions. The linked models will provide more trustworthy and reliable outputs, better inform farmers about CT with spatially explicit adoption and implementation recommendations, and provide more effective policy recommendations for socially optimal CT adoption and implementation. The project will generate practical recommendations and policies on optimal CT practice, which will be disseminated to stakeholders including farmers and policy makers in the U.S. and China and elsewhere in the world. The research will be incorporated into various undergraduate and graduate programs and courses at the institutions in the U.S. and China.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.