气候变化领域集成服务门户(CCPortal): How Machine Learning and High-Resolution Imagery Can Improve Melt Pond Retrieval From MODIS Over Current Spectral Unmixing Techniques

CCPortal

DOI	10.1029/2019JC015569
	How Machine Learning and High-Resolution Imagery Can Improve Melt Pond Retrieval From MODIS Over Current Spectral Unmixing Techniques
	Wright N.C.; Polashenski C.M.
发表日期	2020
ISSN	21699275
卷号	125 期号:2
英文摘要	Meltwater that pools on the surface of Arctic sea ice enhances solar absorption and accelerates further ice melt. The impact of melt ponds on energy absorption is controlled primarily through their influence on ice albedo, which is, in turn, governed in large part by the ponds' spatial coverage. This work seeks to observe the spatial coverage of melt ponds across the Arctic basin with sufficient accuracy to investigate pond-albedo feedback and presents an improved technique to achieve this goal. We approach the problem by using the Open Source Sea Ice Processing algorithm to classify surface features in submeter resolution optical satellite imagery over select sites where such imagery is available. These data establish “true” estimates of pond coverage and the ponds' spectral reflectance. This information is then used to inform, improve, and test spectral unmixing and machine learning techniques that seek to determine melt pond coverage from more widely available, but lower resolution, optical satellite imagery (e.g., Moderate Resolution Imaging Spectroradiometer). The new machine learning approach improves accuracy from prior work and can contribute to improved efforts to validate melt pond models or understand trends in pond coverage. Nevertheless, we encounter and carefully document significant challenges to retrieving melt pond fractions from low-resolution optical imagery. These limit accuracy to levels below that necessary for resolving climatologically important trends. We conclude that greatly expanding the collection of high-resolution satellite imagery over sea ice is necessary to monitor melt pond coverage with the accuracy needed by the scientific community. ©2020. American Geophysical Union. All Rights Reserved.
英文关键词	machine learning; melt ponds; remote sensing; sea ice; spectral unmixing
语种	英语
来源期刊	Journal of Geophysical Research: Oceans
文献类型	期刊论文
条目标识符	http://gcip.llas.ac.cn/handle/2XKMVOVA/186992
作者单位	Thayer School of Engineering, Dartmouth College, Hanover, NH, United States; U.S. Army Cold Regions Research and Engineering Laboratories, Hanover, NH, United States
推荐引用方式 GB/T 7714	Wright N.C.,Polashenski C.M.. How Machine Learning and High-Resolution Imagery Can Improve Melt Pond Retrieval From MODIS Over Current Spectral Unmixing Techniques[J],2020,125(2).
APA	Wright N.C.,&Polashenski C.M..(2020).How Machine Learning and High-Resolution Imagery Can Improve Melt Pond Retrieval From MODIS Over Current Spectral Unmixing Techniques.Journal of Geophysical Research: Oceans,125(2).
MLA	Wright N.C.,et al."How Machine Learning and High-Resolution Imagery Can Improve Melt Pond Retrieval From MODIS Over Current Spectral Unmixing Techniques".Journal of Geophysical Research: Oceans 125.2(2020).