Closing the energy balance gap at scale

众所周知，陆地表面状况会影响天气和气候。地表与大气之间的水分、能量和气体交换取决于地表的类型（如水、森林、草地）、植被、土壤条件等。为了研究这些交换，科学家们建立了仪器塔来测量不同高度的各种特性，以了解陆地表面和大气是如何相互作用的。然而，在许多地方，陆地表面的不均匀条件使得解释这些数据变得困难。本研究项目将尝试采用先进的数值方法，提供一种具有量化不确定性和低偏差的更好地确定地表-大气通量时空变化的方法。一个成功的项目将对生态学、水文学、地球系统建模、自然气候解决方案以及石油和天然气工业排放产生影响。多名学生和早期职业科学家将参与这项工作，研究小组计划扩大范围，以提高公众对地球系统的了解。本项目的总体目标是通过开发以大涡模拟（LES）模型为基准的整体标度方法，并将其应用于涡相关方差通量、边界层剖面和地表性质观测的密集网络，改进理解区域尺度地表-大气能量、水和碳通量的大小和驱动因素的方法。研究小组将评估环境响应函数（ERF），以测试所有能量平衡组件的综合缩放方法如何既能解决通量测量中的已知偏差，又能利用通量塔数据改进模型数据基准。该技术将与LES和空间涡旋相关方法进行比较，并应用于涡旋相关塔和空气通量的广泛数据集，以及在高密度广泛探测器阵列（CHEESEHEAD-19）野外活动中收集的checamegon异质生态系统能量平衡研究期间收集的其他数据。将测试三个主要假设：1。在与下垫面变率和上覆气象的光谱间隙一致的时空尺度上进行评估时，能量平衡封闭性得到改善。与涡旋相关塔站点的平均能量不平衡相比，湍流热通量、净辐射、土壤热通量以及空气、土壤和生物质中相关的存储项的持续缩放将改善能量平衡闭合。与能量输送、地表热和生态系统异质性相关的一组有限的地表和大气变量可以使用erf虚拟控制体积对涡动相关通量进行缩放，从而增加了与模型、大气反演和基于清单（包括碳）的一致性。该项目由地球科学理事会共同资助，旨在支持人工智能/机器学习在地球科学领域的发展。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。

Land surface conditions are known to impact weather and climate. The exchange of moisture, energy, and gases between the surface and atmosphere depends on the types of surfaces (e.g., water, forest, grassland), vegetation, soil conditions, and more. To study these exchanges, scientists have set up instrumented towers to measure various properties at different heights to get a sense for how the land surface and atmosphere are interacting. However, the heterogeneous conditions of the land surface in many locations makes interpreting this data difficult. This research project will attempt to use an advanced numerical approach to provide a method to better determine the space and time variations of surface-atmosphere fluxes with quantified uncertainties and low bias. A successful project would have implications for ecology, hydrology, Earth system modeling, natural climate solutions, and oil and gas industry emissions. Multiple students and early-career scientists would be involved in the work and the research team plans outreach to enhance public understanding of Earth systems.The overarching goal of this project is to improve approaches to understanding the magnitude and drivers of regional-scale surface-atmosphere fluxes of energy, water, and carbon through development of a holistic scaling approach benchmarked within a large eddy simulation (LES) model and applied to a dense network of eddy covariance flux, boundary-layer profile, and surface property observations. The research team will evaluate the Environmental Response Function (ERF) to test how a comprehensive scaling approach of all energy balance components can both resolve the known bias in flux measurements and improve model-data benchmarking with flux tower data. The technique will be compared to LES and spatial eddy covariance approaches and be applied to an extensive dataset of eddy covariance tower and airborne fluxes and other data collected during the Chequamegon Heterogenous Ecosystem Energy-balance Study Enabled by a High-density Extensive Array of Detectors (CHEESEHEAD-19) field campaign. Three main hypotheses will be tested:1. Energy balance closure improves when evaluated over a space-time scale that is consistent with spectral gaps in underlying surface variability and overlying meteorology.2. Consistently scaled turbulent heat fluxes, net radiation, soil heat flux, and associated storage terms in air, soil, and biomass will improve energy balance closure compared to the average energy imbalance at eddy covariance tower sites.3. A restricted set of land surface and atmospheric variables related to energy transport and land surface thermal and ecosystem heterogeneity can scale eddy covariance fluxes using ERF-Virtual Control Volume in ways that increases consistency with modeled, atmospheric inverse, and inventory-based, including of carbon.This project is co-funded by the Directorate for Geosciences to support AI/ML advancement in the geosciences.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.