Training courses on the matrix approach to modeling land carbon and nitrogen cycles; 2018-2021: Flagstaff, AZ

该项目将培训早期职业科学家并发展美国劳动力。培训将通过一系列关于土地碳和氮循环主题的短期课程进行。碳和氮对人类福祉和生态系统功能都很重要。陆地生态系统吸收了目前人类排放的大约30%的二氧化碳。我们需要进一步了解这是如何发生的。为了改进能够提供这些细节的模型，科学家们最近开发了一种称为矩阵方法的新技术。矩阵法已被公认为是统一陆地碳循环模型的一种方法。这种方法还有其他好处，包括理解不确定性、减少计算时间和将数据合并到模型中。该方法有可能改进土地碳和氮模型，以更好地预测生态系统服务的未来变化。这些课程将培养新一代生态系统建模者，以实现编码的简单性，更清晰、更高效的模型，更快的结果，以及在模型复杂性和可理解的结果之间更好的桥梁。该项目将在今后三年每年举办一次短期培训班。本课程的主题是模拟土地碳和氮循环。具体的方法称为矩阵方法。本课程将重点培养五项技能：(1)绘制模型的碳流图；(2)写出模型的碳平衡方程；(3)建立碳平衡方程的矩阵模型；(4)添加诊断变量；(5)增加半解析自旋算法。实践培训将与各种主题的讲座相结合。主题将是：土地碳循环动力学理论，数据同化，矩阵运算，微分方程求解和模型结构数据库。培训课程将学员分成几个工作小组，每个小组有一名指导老师，以确保个人帮助。随着学员反馈的纳入，课程将随着时间的推移得到改进，变得更加有效和高效。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。

This project will train early career scientists and develop the U.S. workforce. The training will occur through a series of short courses on the topic of land carbon and nitrogen cycles. Carbon and nitrogen are both important for human well-being and ecosystem function. Land ecosystems absorb approximately 30% of current carbon dioxide emissions by humans. We need further details about how this happens. To improve models that can provide these details, scientists have recently developed a new technique called the matrix approach. The matrix approach has been recognized as a way to unify all the land carbon cycle models. There are additional benefits from this approach that involve understanding uncertainty, reducing computation time, and merging data into the models. The approach has the potential to improve land carbon and nitrogen models to better predict future changes in ecosystem services. The courses will train a new generation of ecosystem modelers to achieve simplicity in coding, cleaner and more efficient models, faster results, and a better bridge between model complexity and understandable results. This project will run an annual short training course for the next three years. The topic of the course is modeling land carbon and nitrogen cycles. The specific approach is called the matrix approach. The course will focus on five skills to be trained: (1) drawing the carbon flow diagram of a model; (2) writing carbon balance equations of a model; (3) developing a matrix model out of carbon balance equations; (4) adding diagnostic variables; and (5) adding semi-analytic spin-up algorithms. The hands-on training will be combined with lectures on a variety of topics. The topics will be: theories of land carbon cycle dynamics, data assimilation, matrix operation, differential equation solvers, and model structure databases. The training course will divide the trainees into several working groups, each with one instructor to ensure individual help. The course will be improved to become more effective and efficient over time as feedback from trainees is incorporated.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.