Abstract
The accurate estimation of carbon and heat fluxes at global scale is paramount for future policy decisions in the context of global climate change. This paper analyzes the relative relevance of potential remote sensing and meteorological drivers of global carbon and energy fluxes over land. The study is done in an indirect way via upscaling both Gross Primary Production (GPP) and latent energy (LE) using Gaussian Process regression (GPR). In summary, GPR is successfully compared to multivariate linear regression (RMSE gain of +4.17% in GPP and +7.63% in LE) and kernel ridge regression (+2.91% in GPP and +3.07% in LE). The best GP models are then studied in terms of explanatory power based on the analysis of the lengthscales of the anisotropic covariance function, sensitivity maps of the predictive mean, and the robustness to distortions in the input variables. It is concluded that GPP is predominantly mediated by several vegetation indices and land surface temperature (LST), while LE is mostly driven by LST, global radiation and vegetation indices.
| Original language | English (US) |
|---|---|
| Title of host publication | International Geoscience and Remote Sensing Symposium (IGARSS) |
| Publisher | Institute of Electrical and Electronics Engineers Inc. |
| Pages | 4416-4419 |
| Number of pages | 4 |
| Volume | 2015-November |
| ISBN (Print) | 9781479979295 |
| DOIs | |
| State | Published - Nov 10 2015 |
| Externally published | Yes |
| Event | IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015 - Milan, Italy Duration: Jul 26 2015 → Jul 31 2015 |
Other
| Other | IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015 |
|---|---|
| Country | Italy |
| City | Milan |
| Period | 7/26/15 → 7/31/15 |
Fingerprint
Keywords
- carbon
- energy
- feature ranking
- Gaussian process
- global monitoring
- GPP
- regression
ASJC Scopus subject areas
- Earth and Planetary Sciences(all)
- Computer Science Applications
Cite this
Ranking drivers of global carbon and energy fluxes over land. / Camps-Valls, Gustau; Jung, Martin; Ichii, Kazuhito; Papale, Dario; Tramontana, Gianluca; Bodesheim, Paul; Schwalm, Christopher R; Zscheischler, Jakob; Mahecha, Miguel; Reichstein, Markus.
International Geoscience and Remote Sensing Symposium (IGARSS). Vol. 2015-November Institute of Electrical and Electronics Engineers Inc., 2015. p. 4416-4419 7326806.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Ranking drivers of global carbon and energy fluxes over land
AU - Camps-Valls, Gustau
AU - Jung, Martin
AU - Ichii, Kazuhito
AU - Papale, Dario
AU - Tramontana, Gianluca
AU - Bodesheim, Paul
AU - Schwalm, Christopher R
AU - Zscheischler, Jakob
AU - Mahecha, Miguel
AU - Reichstein, Markus
PY - 2015/11/10
Y1 - 2015/11/10
N2 - The accurate estimation of carbon and heat fluxes at global scale is paramount for future policy decisions in the context of global climate change. This paper analyzes the relative relevance of potential remote sensing and meteorological drivers of global carbon and energy fluxes over land. The study is done in an indirect way via upscaling both Gross Primary Production (GPP) and latent energy (LE) using Gaussian Process regression (GPR). In summary, GPR is successfully compared to multivariate linear regression (RMSE gain of +4.17% in GPP and +7.63% in LE) and kernel ridge regression (+2.91% in GPP and +3.07% in LE). The best GP models are then studied in terms of explanatory power based on the analysis of the lengthscales of the anisotropic covariance function, sensitivity maps of the predictive mean, and the robustness to distortions in the input variables. It is concluded that GPP is predominantly mediated by several vegetation indices and land surface temperature (LST), while LE is mostly driven by LST, global radiation and vegetation indices.
AB - The accurate estimation of carbon and heat fluxes at global scale is paramount for future policy decisions in the context of global climate change. This paper analyzes the relative relevance of potential remote sensing and meteorological drivers of global carbon and energy fluxes over land. The study is done in an indirect way via upscaling both Gross Primary Production (GPP) and latent energy (LE) using Gaussian Process regression (GPR). In summary, GPR is successfully compared to multivariate linear regression (RMSE gain of +4.17% in GPP and +7.63% in LE) and kernel ridge regression (+2.91% in GPP and +3.07% in LE). The best GP models are then studied in terms of explanatory power based on the analysis of the lengthscales of the anisotropic covariance function, sensitivity maps of the predictive mean, and the robustness to distortions in the input variables. It is concluded that GPP is predominantly mediated by several vegetation indices and land surface temperature (LST), while LE is mostly driven by LST, global radiation and vegetation indices.
KW - carbon
KW - energy
KW - feature ranking
KW - Gaussian process
KW - global monitoring
KW - GPP
KW - regression
UR - http://www.scopus.com/inward/record.url?scp=84962486937&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84962486937&partnerID=8YFLogxK
U2 - 10.1109/IGARSS.2015.7326806
DO - 10.1109/IGARSS.2015.7326806
M3 - Conference contribution
AN - SCOPUS:84962486937
SN - 9781479979295
VL - 2015-November
SP - 4416
EP - 4419
BT - International Geoscience and Remote Sensing Symposium (IGARSS)
PB - Institute of Electrical and Electronics Engineers Inc.
ER -