Predicting carbon dioxide and energy fluxes across global FLUXNET sites with regression algorithms

Abstract. Spatio-temporal fields of land–atmosphere fluxes derived from data-driven models can complement simulations by process-based land surface models. While a number of strategies for empirical models with eddy-covariance flux data have been applied, a systematic intercomparison of these method...

Descripción completa

Detalles Bibliográficos
Autores principales: Tramontana, Gianluca, Jung, Martin, Schwalm, Christopher R., Ichii, Kazuhito, Camps-Valls, Gustau, Ráduly, Botond, Reichstein, Markus, Arain, M. Altaf, Cescatti, Alessandro, Kiely, Gerard, Merbold, Lutz, Serrano-Ortiz, Penelope, Sickert, Sven, Wolf, Sebastian, Papale, Dario
Formato: Journal Article
Lenguaje:Inglés
Publicado: Copernicus GmbH 2016
Materias:
Acceso en línea:https://hdl.handle.net/10568/129400
_version_ 1855542148786552832
author Tramontana, Gianluca
Jung, Martin
Schwalm, Christopher R.
Ichii, Kazuhito
Camps-Valls, Gustau
Ráduly, Botond
Reichstein, Markus
Arain, M. Altaf
Cescatti, Alessandro
Kiely, Gerard
Merbold, Lutz
Serrano-Ortiz, Penelope
Sickert, Sven
Wolf, Sebastian
Papale, Dario
author_browse Arain, M. Altaf
Camps-Valls, Gustau
Cescatti, Alessandro
Ichii, Kazuhito
Jung, Martin
Kiely, Gerard
Merbold, Lutz
Papale, Dario
Reichstein, Markus
Ráduly, Botond
Schwalm, Christopher R.
Serrano-Ortiz, Penelope
Sickert, Sven
Tramontana, Gianluca
Wolf, Sebastian
author_facet Tramontana, Gianluca
Jung, Martin
Schwalm, Christopher R.
Ichii, Kazuhito
Camps-Valls, Gustau
Ráduly, Botond
Reichstein, Markus
Arain, M. Altaf
Cescatti, Alessandro
Kiely, Gerard
Merbold, Lutz
Serrano-Ortiz, Penelope
Sickert, Sven
Wolf, Sebastian
Papale, Dario
author_sort Tramontana, Gianluca
collection Repository of Agricultural Research Outputs (CGSpace)
description Abstract. Spatio-temporal fields of land–atmosphere fluxes derived from data-driven models can complement simulations by process-based land surface models. While a number of strategies for empirical models with eddy-covariance flux data have been applied, a systematic intercomparison of these methods has been missing so far. In this study, we performed a cross-validation experiment for predicting carbon dioxide, latent heat, sensible heat and net radiation fluxes across different ecosystem types with 11 machine learning (ML) methods from four different classes (kernel methods, neural networks, tree methods, and regression splines). We applied two complementary setups: (1) 8-day average fluxes based on remotely sensed data and (2) daily mean fluxes based on meteorological data and a mean seasonal cycle of remotely sensed variables. The patterns of predictions from different ML and experimental setups were highly consistent. There were systematic differences in performance among the fluxes, with the following ascending order: net ecosystem exchange (R2 < 0.5), ecosystem respiration (R2 > 0.6), gross primary production (R2> 0.7), latent heat (R2 > 0.7), sensible heat (R2 > 0.7), and net radiation (R2 > 0.8). The ML methods predicted the across-site variability and the mean seasonal cycle of the observed fluxes very well (R2 > 0.7), while the 8-day deviations from the mean seasonal cycle were not well predicted (R2 < 0.5). Fluxes were better predicted at forested and temperate climate sites than at sites in extreme climates or less represented by training data (e.g., the tropics). The evaluated large ensemble of ML-based models will be the basis of new global flux products.
format Journal Article
id CGSpace129400
institution CGIAR Consortium
language Inglés
publishDate 2016
publishDateRange 2016
publishDateSort 2016
publisher Copernicus GmbH
publisherStr Copernicus GmbH
record_format dspace
spelling CGSpace1294002025-12-08T10:11:39Z Predicting carbon dioxide and energy fluxes across global FLUXNET sites with regression algorithms Tramontana, Gianluca Jung, Martin Schwalm, Christopher R. Ichii, Kazuhito Camps-Valls, Gustau Ráduly, Botond Reichstein, Markus Arain, M. Altaf Cescatti, Alessandro Kiely, Gerard Merbold, Lutz Serrano-Ortiz, Penelope Sickert, Sven Wolf, Sebastian Papale, Dario carbon energy carbon dioxide algorithms Abstract. Spatio-temporal fields of land–atmosphere fluxes derived from data-driven models can complement simulations by process-based land surface models. While a number of strategies for empirical models with eddy-covariance flux data have been applied, a systematic intercomparison of these methods has been missing so far. In this study, we performed a cross-validation experiment for predicting carbon dioxide, latent heat, sensible heat and net radiation fluxes across different ecosystem types with 11 machine learning (ML) methods from four different classes (kernel methods, neural networks, tree methods, and regression splines). We applied two complementary setups: (1) 8-day average fluxes based on remotely sensed data and (2) daily mean fluxes based on meteorological data and a mean seasonal cycle of remotely sensed variables. The patterns of predictions from different ML and experimental setups were highly consistent. There were systematic differences in performance among the fluxes, with the following ascending order: net ecosystem exchange (R2 < 0.5), ecosystem respiration (R2 > 0.6), gross primary production (R2> 0.7), latent heat (R2 > 0.7), sensible heat (R2 > 0.7), and net radiation (R2 > 0.8). The ML methods predicted the across-site variability and the mean seasonal cycle of the observed fluxes very well (R2 > 0.7), while the 8-day deviations from the mean seasonal cycle were not well predicted (R2 < 0.5). Fluxes were better predicted at forested and temperate climate sites than at sites in extreme climates or less represented by training data (e.g., the tropics). The evaluated large ensemble of ML-based models will be the basis of new global flux products. 2016-07-29 2023-03-10T14:34:29Z 2023-03-10T14:34:29Z Journal Article https://hdl.handle.net/10568/129400 en Open Access Copernicus GmbH Tramontana, Gianluca; Jung, Martin; Schwalm, Christopher R.; Ichii, Kazuhito; Camps-Valls, Gustau; Ráduly, Botond; Reichstein, Markus; Arain, M. Altaf; Cescatti, Alessandro; Kiely, Gerard; Merbold, Lutz; Serrano-Ortiz, Penelope; Sickert, Sven; Wolf, Sebastian; Papale, Dario. 2016. Predicting carbon dioxide and energy fluxes across global FLUXNET sites with regression algorithms. Biogeosciences 13: 4291-4313
spellingShingle carbon
energy
carbon dioxide
algorithms
Tramontana, Gianluca
Jung, Martin
Schwalm, Christopher R.
Ichii, Kazuhito
Camps-Valls, Gustau
Ráduly, Botond
Reichstein, Markus
Arain, M. Altaf
Cescatti, Alessandro
Kiely, Gerard
Merbold, Lutz
Serrano-Ortiz, Penelope
Sickert, Sven
Wolf, Sebastian
Papale, Dario
Predicting carbon dioxide and energy fluxes across global FLUXNET sites with regression algorithms
title Predicting carbon dioxide and energy fluxes across global FLUXNET sites with regression algorithms
title_full Predicting carbon dioxide and energy fluxes across global FLUXNET sites with regression algorithms
title_fullStr Predicting carbon dioxide and energy fluxes across global FLUXNET sites with regression algorithms
title_full_unstemmed Predicting carbon dioxide and energy fluxes across global FLUXNET sites with regression algorithms
title_short Predicting carbon dioxide and energy fluxes across global FLUXNET sites with regression algorithms
title_sort predicting carbon dioxide and energy fluxes across global fluxnet sites with regression algorithms
topic carbon
energy
carbon dioxide
algorithms
url https://hdl.handle.net/10568/129400
work_keys_str_mv AT tramontanagianluca predictingcarbondioxideandenergyfluxesacrossglobalfluxnetsiteswithregressionalgorithms
AT jungmartin predictingcarbondioxideandenergyfluxesacrossglobalfluxnetsiteswithregressionalgorithms
AT schwalmchristopherr predictingcarbondioxideandenergyfluxesacrossglobalfluxnetsiteswithregressionalgorithms
AT ichiikazuhito predictingcarbondioxideandenergyfluxesacrossglobalfluxnetsiteswithregressionalgorithms
AT campsvallsgustau predictingcarbondioxideandenergyfluxesacrossglobalfluxnetsiteswithregressionalgorithms
AT radulybotond predictingcarbondioxideandenergyfluxesacrossglobalfluxnetsiteswithregressionalgorithms
AT reichsteinmarkus predictingcarbondioxideandenergyfluxesacrossglobalfluxnetsiteswithregressionalgorithms
AT arainmaltaf predictingcarbondioxideandenergyfluxesacrossglobalfluxnetsiteswithregressionalgorithms
AT cescattialessandro predictingcarbondioxideandenergyfluxesacrossglobalfluxnetsiteswithregressionalgorithms
AT kielygerard predictingcarbondioxideandenergyfluxesacrossglobalfluxnetsiteswithregressionalgorithms
AT merboldlutz predictingcarbondioxideandenergyfluxesacrossglobalfluxnetsiteswithregressionalgorithms
AT serranoortizpenelope predictingcarbondioxideandenergyfluxesacrossglobalfluxnetsiteswithregressionalgorithms
AT sickertsven predictingcarbondioxideandenergyfluxesacrossglobalfluxnetsiteswithregressionalgorithms
AT wolfsebastian predictingcarbondioxideandenergyfluxesacrossglobalfluxnetsiteswithregressionalgorithms
AT papaledario predictingcarbondioxideandenergyfluxesacrossglobalfluxnetsiteswithregressionalgorithms