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

Kazuhito Ichii*, Masahito Ueyama, Masayuki Kondo, Nobuko Saigusa, Joon Kim, Ma Carmelita Alberto, Jonas Ardö, Eugénie S. Euskirchen, Minseok Kang, Takashi Hirano, Joanna Joiner, Hideki Kobayashi, Luca Belelli Marchesini, Lutz Merbold, Akira Miyata, Taku M. Saitoh, Kentaro Takagi, Andrej Varlagin, M. Syndonia Bret-Harte, Kenzo KitamuraYoshiko 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

*Corresponding author for this work

Research output: Contribution to JournalArticleAcademicpeer-review


The lack of a standardized database of eddy covariance observations has been an obstacle for data-driven estimation of terrestrial CO2 fluxes in Asia. In this study, we developed such a standardized database using 54 sites from various databases by applying consistent postprocessing for data-driven estimation of gross primary productivity (GPP) and net ecosystem CO2 exchange (NEE). Data-driven estimation was conducted by using a machine learning algorithm: support vector regression (SVR), with remote sensing data for 2000 to 2015 period. Site-level evaluation of the estimated CO2 fluxes shows that although performance varies in different vegetation and climate classifications, GPP and NEE at 8 days are reproduced (e.g., r2 = 0.73 and 0.42 for 8 day GPP and NEE). Evaluation of spatially estimated GPP with Global Ozone Monitoring Experiment 2 sensor-based Sun-induced chlorophyll fluorescence shows that monthly GPP variations at subcontinental scale were reproduced by SVR (r2 = 1.00, 0.94, 0.91, and 0.89 for Siberia, East Asia, South Asia, and Southeast Asia, respectively). Evaluation of spatially estimated NEE with net atmosphere-land CO2 fluxes of Greenhouse Gases Observing Satellite (GOSAT) Level 4A product shows that monthly variations of these data were consistent in Siberia and East Asia; meanwhile, inconsistency was found in South Asia and Southeast Asia. Furthermore, differences in the land CO2 fluxes from SVR-NEE and GOSAT Level 4A were partially explained by accounting for the differences in the definition of land CO2 fluxes. These data-driven estimates can provide a new opportunity to assess CO2 fluxes in Asia and evaluate and constrain terrestrial ecosystem models.

Original languageEnglish
Pages (from-to)767-795
Number of pages29
JournalJournal of Geophysical Research: Biogeosciences
Issue number4
Publication statusPublished - 1 Apr 2017


This study was supported by the JSPS KAKENHI grants 25281003, 25281014, and 16H02948; the Environment Research and Technology Development Funds (2-1401) from the Ministry of the Environment of Japan; the Global Change Observation Mission (GCOM RA6 PI#111) of the Japan Aerospace Exploration Agency (JAXA); and the Asia-Pacific Network for Global Change Research (ARCP2015-01CMY-Miyata). Joon Kim and Minseok Kang were supported by the Korea Meteorological Administration Research and Development Program under grant KMIPA 2015-2023 and the Weather Information Service Engine (WISE) project KMA-2012-0001-A. Jonas Ard? was supported by the Swedish National Space Board. Andrej Varlagin was supported by the RSF grant 14-24-00113. Joanna Joiner was supported by funding from NASA. Funding for the eastern Siberian flux tower near Cherskii was provided by the National Science Foundation Division of Polar Programs Arctic Observatory Network and by the TCOS Siberia European Union Project 2002?2004 (EU project N EVK2-2001-00143). The directors of CSIR-CMMACS, Bangalore and GBPNIHESD, Kosi-Katarmal, India are acknowledged for establishing the flux monitoring station at Kosi-Katarmal, India and facilitating the data collection. Databases of EC measurement (AsiaFlux, FFPRI, and European Fluxes Database Cluster) facilitated this study: LIBSVM, A Library for Support Vector Machines (, and levmar Levenberg-Marquardt nonlinear least squares algorithms in C/C++ ( facilitated this study. SVR model outputs are available freely for both site-level outputs and gridded outputs through EC observation data are available at each database: AsiaFlux (, FFPRI Fluxnet (, and European Fluxes Database ( EC data processing software is available at

FundersFunder number
Korea Meteorological Administration Research and Development ProgramKMIPA 2015-2023, KMA-2012-0001-A
National Science Foundation Division of Polar Programs Arctic Observatory NetworkN EVK2-2001-00143
National Science Foundation1026843, 1503912
National Aeronautics and Space Administration
Asia-Pacific Network for Global Change ResearchARCP2015-01CMY-Miyata
Forestry and Forest Products Research Institute
Japan Society for the Promotion of Science26292076, 25281003, 25281014, 16H02948, 2-1401
Swedish National Space Agency
Japan Aerospace Exploration Agency
Ministry of the Environment, Government of Japan111
Russian Science Foundation14-24-00113


    • Asia
    • data-driven model
    • eddy covariance data
    • remote sensing
    • terrestrial CO flux
    • upscaling


    Dive into the research topics of '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'. Together they form a unique fingerprint.

    Cite this