Advanced

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

Ichii, Kazuhito; Ueyama, Masahito; Kondo, Masayuki; Saigusa, Nobuko; Kim, Joon; Alberto, Ma Carmelita; Ardö, Jonas LU ; Euskirchen, Eugénie S.; Kang, Minseok and Hirano, Takashi, et al. (2017) In Journal of Geophysical Research - Biogeosciences 122(4). p.767-795
Abstract

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 8days are... (More)

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 8days are reproduced (e.g., r2=0.73 and 0.42 for 8day 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.

(Less)
Please use this url to cite or link to this publication:
author
, et al. (More)
(Less)
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
Asia, Data-driven model, Eddy covariance data, Remote sensing, Terrestrial CO flux, Upscaling
in
Journal of Geophysical Research - Biogeosciences
volume
122
issue
4
pages
767 - 795
publisher
American Geophysical Union
external identifiers
  • scopus:85017578063
  • wos:000401171800003
ISSN
2169-8953
DOI
10.1002/2016JG003640
language
English
LU publication?
yes
id
d73171d4-11b1-4b85-a2dc-6225e5df3187
date added to LUP
2017-05-10 09:04:26
date last changed
2018-05-13 04:31:10
@article{d73171d4-11b1-4b85-a2dc-6225e5df3187,
  abstract     = {<p>The lack of a standardized database of eddy covariance observations has been an obstacle for data-driven estimation of terrestrial CO<sub>2</sub> 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 CO<sub>2</sub> 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 CO<sub>2</sub> fluxes shows that although performance varies in different vegetation and climate classifications, GPP and NEE at 8days are reproduced (e.g., r<sup>2</sup>=0.73 and 0.42 for 8day 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 (r<sup>2</sup>=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 CO<sub>2</sub> 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 CO<sub>2</sub> fluxes from SVR-NEE and GOSAT Level 4A were partially explained by accounting for the differences in the definition of land CO<sub>2</sub> fluxes. These data-driven estimates can provide a new opportunity to assess CO<sub>2</sub> fluxes in Asia and evaluate and constrain terrestrial ecosystem models.</p>},
  author       = {Ichii, Kazuhito and Ueyama, Masahito and Kondo, Masayuki and Saigusa, Nobuko and Kim, Joon and Alberto, Ma Carmelita and Ardö, Jonas and Euskirchen, Eugénie S. and Kang, Minseok and Hirano, Takashi and Joiner, Joanna and Kobayashi, Hideki and Marchesini, Luca Belelli and Merbold, Lutz and Miyata, Akira and Saitoh, Taku M. and Takagi, Kentaro and Varlagin, Andrej and Bret-Harte, M. Syndonia and Kitamura, Kenzo and Kosugi, Yoshiko and Kotani, Ayumi and Kumar, Kireet and Li, Sheng Gong and Machimura, Takashi and Matsuura, Yojiro and Mizoguchi, Yasuko and Ohta, Takeshi and Mukherjee, Sandipan and Yanagi, Yuji and Yasuda, Yukio and Zhang, Yiping and Zhao, Fenghua},
  issn         = {2169-8953},
  keyword      = {Asia,Data-driven model,Eddy covariance data,Remote sensing,Terrestrial CO flux,Upscaling},
  language     = {eng},
  month        = {04},
  number       = {4},
  pages        = {767--795},
  publisher    = {American Geophysical Union},
  series       = {Journal of Geophysical Research - Biogeosciences},
  title        = {New data-driven estimation of terrestrial CO<sub>2</sub> fluxes in Asia using a standardized database of eddy covariance measurements, remote sensing data, and support vector regression},
  url          = {http://dx.doi.org/10.1002/2016JG003640},
  volume       = {122},
  year         = {2017},
}