CAREER: Peatland Geomorphology: Quantifying Geomorphological Changes across SE Asia Peatlands

该项目将确定人类活动（由于森林砍伐造成的泥炭退化）造成的过程如何影响自然灾害（洪水）和气候变化（二氧化碳排放）。泥炭地是一种湿地，在那里，浸水的条件会减缓植物的分解，以至于死亡的植物积累起来形成泥炭。在自然条件下，泥炭圆顶随着时间的推移而生长，将碳储存在地下，并将沿海洪水的风险降至最低。相反，受损的泥炭地正在压实，这导致地面高度降低，温室气体排放增加。该项目建议将多个卫星和地面数据集与人工智能工具相结合，以解决一个具有高度社会重要性的问题：“数千年来积累的泥炭碳将以什么速度释放到大气中？东南亚大片沿海地区将在什么时间尺度上永久被淹没？”回顾路易斯安那州，研究表明，如果没有之前几十年的泥炭退化，卡特里娜飓风的影响不会那么严重。该项目将主动量化东南亚洪水在何时何地变得最严重，并提高我们对全球泥炭地动态的理解。从这项工作中获得的知识将有助于预测未来的泥炭分解率和洪水灾害，并改进对二氧化碳通量的估计。这些结果将为旨在通过教育赋予当地社区权力的推广工作提供信息，促进科学家参与公共交流，并通过早期接触实际研究来增加本科生的多样性。该项目的目标是建立预测热带泥炭地及其碳的命运以应对环境变化的能力，并制定有效的方法使人们了解泥炭地的重要性。本研究的中心假设是泥炭退化速率受景观形态和土地利用历史的控制。为了验证这一点，该项目将依靠三个主要任务：1)通过结合形态学（光探测和测距数据）、变形（经实地调查验证的干涉合成孔径雷达数据）和土地利用时间序列（多光谱光学数据），确定东南亚泥炭退化率的控制因素；2)开发一个机器学习模型，考虑对退化率的控制，以预测未来退化的时间和幅度、二氧化碳排放以及相关的洪水；3)通过分享社会响应性外展材料，提高对泥炭地重要性的认识。所开发的技术将为全球泥炭地地貌学研究提供一个关键框架。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 determine how a process resulting from human activities (peat degradation due to deforestation) influences natural hazards (flooding) and climate change (carbon dioxide emissions). Peatlands are wetlands where waterlogged conditions slow plant decomposition to such an extent that dead plants accumulate to form peat. Under natural conditions peat domes grow over time, storing carbon underground and minimizing the risk of coastal flooding. In contrast, damaged peatlands are compacting, which results in a lowering of the ground elevation and in greenhouse gas emissions. This project proposes to integrate multiple satellite- and ground-based datasets with artificial intelligence tools to address a question of high societal importance: “At what rates will thousands of years of accumulated peat carbon be released to the atmosphere and on what timescale will large coastal areas in southeast Asia become permanently flooded?” Retrospectively, in Louisiana, it was shown that the impact of hurricane Katrina would have been less severe without the preceding decades of peat degradation. This project will proactively quantify when and where flooding will become the most severe in southeast Asia and improve our understanding of peatland dynamics worldwide. Knowledge from this work will enable forecasting future peat decomposition rates and flooding hazards, and refine estimations of carbon dioxide fluxes. These results will inform outreach efforts aimed to empower local communities through education, boost engagement of scientists in public communication, and increase undergraduate diversity through early exposure to hands on research. The goals of this project are to establish the capability to predict the fate of tropical peatlands and their carbon in response to environmental changes and to develop effective ways to engage the population on the importance of peatlands. The central hypothesis of this project is that peat degradation rates are controlled by the landscape morphology and land use history. To test it, this project will rely on three main tasks: 1) identify the controls on peat degradation rates across southeast Asia by combining morphology (Light Detection and Ranging data), deformation (Interferometric Synthetic Aperture Radar data validated by fieldwork), and land use timeseries (multi-spectral optical data); 2) develop a machine learning model accounting for the controls on degradation rates to predict the timing and amplitude of future degradation, carbon dioxide emissions, and the associated flooding; and 3) increase awareness on the importance of peatlands by sharing socially responsive outreach material. The developed techniques will provide a critical framework for worldwide peatland geomorphology studies.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.