New data‐driven estimation of terrestrial CO2 fluxes in Asia using a standardized database of eddy covariance measurements, remote sensing data, and support vector regression

• DOI: 10.1002/2016jg003640
• Publication Date: 2017-03-17T16:26:25Z
• Authors: Kazuhito Ichii; Masahito Ueyama; Masayuki Kondo; Nobuko Saigusa; Joon Kim; Ma. Carmelita Alb...; Jonas Ardö; Eugénie S. Euskir...; Minseok Kang; Takashi Hirano; Joanna Joiner; Hideki Kobayashi; Luca Belelli Marc...; Lutz Merbold; Akira Miyata; Taku M. Saitoh; Kentaro Takagi; Andrej Varlagin; M. Syndonia Bret‐...; Kenzo Kitamura; Yoshiko Kosugi; Ayumi Kotani; Kireet Kumar; Sheng‐Gong Li; Takashi Machimura; Yojiro Matsuura; Yasuko Mizoguchi; Takeshi Ohta; Sandipan Mukherjee; Yuji Yanagi; Yukio Yasuda; Yiping Zhang; Fenghua Zhao
• Keywords: upscaling; data-driven model; remote sensing; Asia; eddy covariance data; 13. Climate action; 0401 agriculture, forestry, and fisheries; terrestrial CO flux; 04 agricultural and veterinary sciences; 15. Life on land; 01 natural sciences; 0105 earth and related environmental sciences

Abstract The lack of a standardized database eddy covariance observations has been an obstacle for data‐driven estimation terrestrial CO 2 fluxes in Asia. In this study, we developed such using 54 sites from various databases by applying consistent postprocessing gross primary productivity (GPP) and net ecosystem exchange (NEE). Data‐driven was conducted machine learning algorithm: support vector regression (SVR), with remote sensing data 2000 to 2015 period. Site‐level evaluation the estimated shows that although performance varies different vegetation climate classifications, GPP NEE at 8 days are reproduced (e.g., r = 0.73 0.42 day NEE). Evaluation spatially Global Ozone Monitoring Experiment sensor‐based Sun‐induced chlorophyll fluorescence monthly variations subcontinental scale were SVR ( 1.00, 0.94, 0.91, 0.89 Siberia, East Asia, South Southeast respectively). atmosphere‐land Greenhouse Gases Observing Satellite (GOSAT) Level 4A product these Siberia Asia; meanwhile, inconsistency found Asia Furthermore, differences land SVR‐NEE GOSAT partially explained accounting definition fluxes. These estimates can provide new opportunity assess evaluate constrain models.