Ecological Similitude and Scaling for Robust Modeling and Predictions of Ecosystem Carbon, Water and Energy Fluxes

1705941（Abdul-Aziz）。本研究旨在揭示生态系统碳、水、能通量的相似模式和尺度规律，并建立稳健的模型来预测不同时空尺度下的碳、水、能通量。基本假设是生态系统通量遵循突发生态相似性和标度律，这将导致稳健的估计和预测模型。具体研究目标是：（1）确定不同时空尺度下生态系统碳（CO2）、潜热（LE）和敏感热（H）通量与相关气候、水文和生态变量的主导驱动因子，并量化其相对联系;（2）研究相似性（参数缩减），制定有意义的无量纲组，并制定标度律（3）建立预测不同生态系统通量的时空稳健经验模型。该研究将利用AmeriFlux数据的通量，气候，环境和生态变量为不同的时间尺度（例如，每小时、每天、每周、每月、每年）和时段（例如，2000-2015）在美国各地的许多网站（超过100），代表水文气候，地球化学和生态过程的梯度。通过量纲分析和经验建模来实现研究目标，该方法已成功地应用于流体力学、水利工程和河流生态学。该研究建立在以前的湿地生态地球化学相似性和缩放的项目，通过使用基于室的现场数据稳健地预测湿地温室气体通量。该项目将使用基于通量塔的涡动协方差数据（AmeriFlux）来概括不同生态系统之间的相似性，缩放和鲁棒建模假设（例如，落叶林、万年青针叶林、雨林、草地、农田、湿地）在不同的时间（小时到年）和空间（地点、区域、大陆）尺度上的变化。该研究将通过对生态系统碳、水和能量通量的出现（相似性和缩放）模式的深入了解、理解和广泛知识来促进环境的可持续性。稳健的、简约的模型可以作为简单的、与尺度无关的工程工具来预测大范围时空尺度上的生态系统通量。研究成果将通过期刊出版物，会议演讲，研究生论文，报告，YouTube视频和开放访问的项目网站广泛传播。研究结果将被用来教两个跨学科的课程，生态水文工程（本科）和生态工程（研究生），使用归纳学习方法。高中学生和教师也将通过外联活动参与研究。

1705941 (Abdul-Aziz). The goal of this research is to discover the similitude patterns and scaling laws of ecosystem carbon, water and energy fluxes, and formulate robust models to predict the fluxes at variable time and space scales. The underlying hypothesis is that the ecosystem fluxes follow emergent ecological similitude and scaling laws, which will lead to robust estimation and predictive models. The specific research objectives are to (1) identify the dominant drivers and quantify the relative linkages of ecosystem carbon (CO2), latent heat (LE) and sensitive heat (H) fluxes with the relevant climatic, hydrologic and ecological variables at different time and space scales; (2) investigate similitude (parametric reductions), formulate meaningful dimensionless groups, and develop scaling laws (emergent patterns) for the ecosystem fluxes; and (3) formulate spatiotemporally robust empirical models for predicting the ecosystem fluxes from diverse ecosystems. The research will utilize the AmeriFlux data of fluxes, climate, environmental, and ecological variables for different time-scales (e.g., hourly, daily, weekly, monthly, yearly) and periods (e.g., 2000-2015) at numerous sites (above 100) across America, representing gradients of hydroclimatic, biogeochemical, and ecological processes. The research objectives will be achieved by conducting dimensional analysis and empirical modeling, which has successfully been applied in fluid mechanics, hydraulic engineering and stream ecology. The research builds on a previous project on wetland biogeochemical similitude and scaling to robustly predict wetland greenhouse gas fluxes by using chamber-based field data. This project will use flux tower-based eddy-covariance data (AmeriFlux) to generalize the similitude, scaling and robust modeling hypothesis across diverse ecosystems (e.g., deciduous forests, evergreen needleleaf forests, rain forests, grasslands, croplands, wetlands) at variable time (hour to year) and space (site, region, continent) scales. The research will promote environmental sustainability by contributing insights, understanding, and broad knowledge into the emergence (similarity and scaling) patterns of ecosystem carbon, water and energy fluxes. Robust, parsimonious models can be utilized as simple, scale-independent engineering tools to predict the ecosystem fluxes across a wide range of spatial and temporal scales. Research outcomes will be broadly disseminated through journal publications, conference presentations, graduate dissertations, reports, YouTube videos, and an open-access project website. The research findings will be utilized to teach two interdisciplinary courses, Ecohydrological Engineering (undergraduate) and Ecological Engineering (graduate), using inductive learning methods. High school students and teachers will also be involved with the research through outreach activities.