Robust filling of extra-long gaps in eddy covariance CO2 flux measurements from a temperate deciduous forest using eXtreme Gradient Boosting

• Published in: Agricultural and forest meteorology Vol. 364; p. 110438
• Main Authors: Liu, Yujie, Lucas, Benjamin, Bergl, Darby D., Richardson, Andrew D.
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.04.2025
• Subjects: algorithms; Ameriflux; carbon; carbon dioxide; carbon dioxide production; deciduous forests; eddy covariance; Gap filling; Machine learning; meteorology; Net ecosystem exchange; PhenoCam; photons; photosynthesis; Vegetation indices; PhenoCam; Net ecosystem exchange; Gap filling; Ameriflux; Vegetation indices; Machine learning
• ISSN: 0168-1923
• DOI: 10.1016/j.agrformet.2025.110438

•Gaps in eddy covariance measurements need to be filled to calculate annual budgets.•XGB, a machine-learning algorithm, effectively filled gaps longer than 30 days.•XGB outperformed the standard MDS for different gap lengths and locations.•Vegetation indices provide critical information and improve...