DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Partitioning net carbon dioxide fluxes into photosynthesis and respiration using neural networks

Abstract

The eddy covariance (EC) technique is used to measure the net ecosystem exchange (NEE) of CO2 between ecosystems and the atmosphere, offering a unique opportunity to study ecosystem responses to climate change. NEE is the difference between the total CO2 release due to all respiration processes (RECO), and the gross carbon uptake by photosynthesis (GPP). These two gross CO2 fluxes are derived from EC measurements by applying partitioning methods that rely on physiologically based functional relationships with a limited number of environmental drivers. However, the partitioning methods applied in the global FLUXNET network of EC observations do not account for the multiple co-acting factors that modulate GPP and RECO flux dynamics. To overcome this limitation, we developed a hybrid data-driven approach based on combined neural networks (NNC-part). NNC-part incorporates process knowledge by introducing a photosynthetic response based on the light-use efficiency (LUE) concept, and uses a comprehensive dataset of soil and micrometeorological variables as fluxes drivers. We applied the method to 36 sites from the FLUXNET2015 dataset and found a high consistency in the results with those derived from other standard partitioning methods for both GPP (R2 > .94) and RECO (R2 > .8). High consistency was also found formore » (a) the diurnal and seasonal patterns of fluxes and (b) the ecosystem functional responses. NNC-part performed more realistic than the traditional methods for predicting additional patterns of gross CO2 fluxes, such as: (a) the GPP response to VPD, (b) direct effects of air temperature on GPP dynamics, (c) hysteresis in the diel cycle of gross CO2 fluxes, (d) the sensitivity of LUE to the diffuse to direct radiation ratio, and (e) the post rain respiration pulse after a long dry period. In conclusion, NNC-part is a valid data-driven approach to provide GPP and RECO estimates and complementary to the existing partitioning methods.« less

Authors:
ORCiD logo [1];  [2];  [2];  [2];  [3];  [4]; ORCiD logo [5];  [4];  [6]
  1. Univ. of Tuscia, Viterbo (Italy). DIBAF Department for Innovation in Biological Agro‐food and Forestry Systems; Universitat de València, Paterna (Spain). Parc Científic. Image Processing Laboratory (IPL)
  2. Max Planck Inst. for Biogeochemistry, Jena (Germany)
  3. Univ. of California, Berkeley, CA (United States). Dept. of Environmental Science, Policy and Management; Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States). Earth and Environmental Sciences Area
  4. Universitat de València, Paterna (Spain). Parc Científic. Image Processing Laboratory (IPL)
  5. INRAE, UMR, Villenaved’Ornon (France)
  6. Univ. of Tuscia, Viterbo (Italy). DIBAF Department for Innovation in Biological Agro‐food and Forestry Systems; Euro‐Mediterranean Center on Climate Change (CMCC), Viterbo (Italy)
Publication Date:
Research Org.:
Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States)
Sponsoring Org.:
USDOE Office of Science (SC)
OSTI Identifier:
1816106
Grant/Contract Number:  
AC02-05CH11231
Resource Type:
Accepted Manuscript
Journal Name:
Global Change Biology
Additional Journal Information:
Journal Volume: 26; Journal Issue: 9; Journal ID: ISSN 1354-1013
Publisher:
Wiley
Country of Publication:
United States
Language:
English
Subject:
54 ENVIRONMENTAL SCIENCES; 97 MATHEMATICS AND COMPUTING; carbon dioxide fluxes partitioning; ecosystem respiration (RECO); eddy covariance; gross primary production (GPP); machine learning; net ecosystem exchange; neural network

Citation Formats

Tramontana, Gianluca, Migliavacca, Mirco, Jung, Martin, Reichstein, Markus, Keenan, Trevor F., Camps‐Valls, Gustau, Ogee, Jerome, Verrelst, Jochem, and Papale, Dario. Partitioning net carbon dioxide fluxes into photosynthesis and respiration using neural networks. United States: N. p., 2020. Web. doi:10.1111/gcb.15203.
Tramontana, Gianluca, Migliavacca, Mirco, Jung, Martin, Reichstein, Markus, Keenan, Trevor F., Camps‐Valls, Gustau, Ogee, Jerome, Verrelst, Jochem, & Papale, Dario. Partitioning net carbon dioxide fluxes into photosynthesis and respiration using neural networks. United States. https://doi.org/10.1111/gcb.15203
Tramontana, Gianluca, Migliavacca, Mirco, Jung, Martin, Reichstein, Markus, Keenan, Trevor F., Camps‐Valls, Gustau, Ogee, Jerome, Verrelst, Jochem, and Papale, Dario. Thu . "Partitioning net carbon dioxide fluxes into photosynthesis and respiration using neural networks". United States. https://doi.org/10.1111/gcb.15203. https://www.osti.gov/servlets/purl/1816106.
@article{osti_1816106,
title = {Partitioning net carbon dioxide fluxes into photosynthesis and respiration using neural networks},
author = {Tramontana, Gianluca and Migliavacca, Mirco and Jung, Martin and Reichstein, Markus and Keenan, Trevor F. and Camps‐Valls, Gustau and Ogee, Jerome and Verrelst, Jochem and Papale, Dario},
abstractNote = {The eddy covariance (EC) technique is used to measure the net ecosystem exchange (NEE) of CO2 between ecosystems and the atmosphere, offering a unique opportunity to study ecosystem responses to climate change. NEE is the difference between the total CO2 release due to all respiration processes (RECO), and the gross carbon uptake by photosynthesis (GPP). These two gross CO2 fluxes are derived from EC measurements by applying partitioning methods that rely on physiologically based functional relationships with a limited number of environmental drivers. However, the partitioning methods applied in the global FLUXNET network of EC observations do not account for the multiple co-acting factors that modulate GPP and RECO flux dynamics. To overcome this limitation, we developed a hybrid data-driven approach based on combined neural networks (NNC-part). NNC-part incorporates process knowledge by introducing a photosynthetic response based on the light-use efficiency (LUE) concept, and uses a comprehensive dataset of soil and micrometeorological variables as fluxes drivers. We applied the method to 36 sites from the FLUXNET2015 dataset and found a high consistency in the results with those derived from other standard partitioning methods for both GPP (R2 > .94) and RECO (R2 > .8). High consistency was also found for (a) the diurnal and seasonal patterns of fluxes and (b) the ecosystem functional responses. NNC-part performed more realistic than the traditional methods for predicting additional patterns of gross CO2 fluxes, such as: (a) the GPP response to VPD, (b) direct effects of air temperature on GPP dynamics, (c) hysteresis in the diel cycle of gross CO2 fluxes, (d) the sensitivity of LUE to the diffuse to direct radiation ratio, and (e) the post rain respiration pulse after a long dry period. In conclusion, NNC-part is a valid data-driven approach to provide GPP and RECO estimates and complementary to the existing partitioning methods.},
doi = {10.1111/gcb.15203},
journal = {Global Change Biology},
number = 9,
volume = 26,
place = {United States},
year = {Thu Jul 02 00:00:00 EDT 2020},
month = {Thu Jul 02 00:00:00 EDT 2020}
}

Works referenced in this record:

Drying and wetting of Mediterranean soils stimulates decomposition and carbon dioxide emission: the "Birch effect"
journal, July 2007


Linking plant and ecosystem functional biogeography
journal, September 2014

  • Reichstein, M.; Bahn, M.; Mahecha, M. D.
  • Proceedings of the National Academy of Sciences, Vol. 111, Issue 38
  • DOI: 10.1073/pnas.1216065111

Upscaled diurnal cycles of land–atmosphere fluxes: a new global half-hourly data product
journal, January 2018


Photosynthetic physiology of eucalypts along a sub-continental rainfall gradient in northern Australia
journal, November 2011


Comprehensive comparison of gap-filling techniques for eddy covariance net carbon fluxes
journal, December 2007


Drivers of inter-annual variability in Net Ecosystem Exchange in a semi-arid savanna ecosystem, South Africa
journal, January 2009

  • Archibald, S. A.; Kirton, A.; van der Merwe, M. R.
  • Biogeosciences, Vol. 6, Issue 2
  • DOI: 10.5194/bg-6-251-2009

Ecosystem CO 2 /H 2 O fluxes are explained by hydraulically limited gas exchange during tree mortality from spruce bark beetles : CO
journal, June 2014

  • Frank, John M.; Massman, William J.; Ewers, Brent E.
  • Journal of Geophysical Research: Biogeosciences, Vol. 119, Issue 6
  • DOI: 10.1002/2013JG002597

Physiological basis of the light use efficiency model
journal, March 1998


Annual net ecosystem carbon exchange by a sugar beet crop
journal, September 2006


Land use regulates carbon budgets in eastern Germany: From NEE to NBP
journal, July 2010

  • Prescher, Anne-Katrin; Grünwald, Thomas; Bernhofer, Christian
  • Agricultural and Forest Meteorology, Vol. 150, Issue 7-8
  • DOI: 10.1016/j.agrformet.2010.03.008

Seasonality of temperate forest photosynthesis and daytime respiration
journal, June 2016

  • Wehr, R.; Munger, J. W.; McManus, J. B.
  • Nature, Vol. 534, Issue 7609
  • DOI: 10.1038/nature17966

Effects of seasonal drought on net carbon dioxide exchange from a woody-plant-encroached semiarid grassland
journal, January 2009

  • Scott, Russell L.; Jenerette, G. Darrel; Potts, Daniel L.
  • Journal of Geophysical Research, Vol. 114, Issue G4
  • DOI: 10.1029/2008JG000900

Long term carbon dioxide exchange above a mixed forest in the Belgian Ardennes
journal, July 2001


Forest and agricultural land-use-dependent CO2 exchange in Thuringia, Germany
journal, December 2004


Prolonged suppression of ecosystem carbon dioxide uptake after an anomalously warm year
journal, September 2008

  • Arnone III, John A.; Verburg, Paul S. J.; Johnson, Dale W.
  • Nature, Vol. 455, Issue 7211
  • DOI: 10.1038/nature07296

Complexity in water and carbon dioxide fluxes following rain pulses in an African savanna
journal, July 2009


Reviews and syntheses: Carbonyl sulfide as a multi-scale tracer for carbon and water cycles
journal, January 2018

  • Whelan, Mary E.; Lennartz, Sinikka T.; Gimeno, Teresa E.
  • Biogeosciences, Vol. 15, Issue 12
  • DOI: 10.5194/bg-15-3625-2018

Increasing net CO2 uptake by a Danish beech forest during the period from 1996 to 2009
journal, July 2011


Predicting carbon dioxide and energy fluxes across global FLUXNET sites with regression algorithms
journal, January 2016

  • Tramontana, Gianluca; Jung, Martin; Schwalm, Christopher R.
  • Biogeosciences, Vol. 13, Issue 14
  • DOI: 10.5194/bg-13-4291-2016

Savanna fires and their impact on net ecosystem productivity in North Australia
journal, May 2007


Revisiting the choice of the driving temperature for eddy covariance CO2 flux partitioning
journal, May 2017


An Emulator Toolbox to Approximate Radiative Transfer Models with Statistical Learning
journal, July 2015

  • Rivera, Juan; Verrelst, Jochem; Gómez-Dans, Jose
  • Remote Sensing, Vol. 7, Issue 7
  • DOI: 10.3390/rs70709347

The response of ecosystem carbon fluxes to LAI and environmental drivers in a maize crop grown in two contrasting seasons
journal, July 2015

  • Vitale, Luca; Di Tommasi, Paul; D’Urso, Guido
  • International Journal of Biometeorology, Vol. 60, Issue 3
  • DOI: 10.1007/s00484-015-1038-2

On the choice of the driving temperature for eddy-covariance carbon dioxide flux partitioning
journal, January 2012


Potential and limitations of inferring ecosystem photosynthetic capacity from leaf functional traits
journal, September 2016

  • Musavi, Talie; Migliavacca, Mirco; Weg, Martine Janet
  • Ecology and Evolution, Vol. 6, Issue 20
  • DOI: 10.1002/ece3.2479

An improved isotopic method for partitioning net ecosystem–atmosphere CO2 exchange
journal, December 2015


Seasonal fluxes of carbonyl sulfide in a midlatitude forest
journal, November 2015

  • Commane, Róisín; Meredith, Laura K.; Baker, Ian T.
  • Proceedings of the National Academy of Sciences, Vol. 112, Issue 46
  • DOI: 10.1073/pnas.1504131112

A perspective on Gaussian processes for Earth observation
journal, March 2019

  • Camps-Valls, Gustau; Sejdinovic, Dino; Runge, Jakob
  • National Science Review, Vol. 6, Issue 4
  • DOI: 10.1093/nsr/nwz028

The carbon balance pivot point of southwestern U.S. semiarid ecosystems: Insights from the 21st century drought
journal, December 2015

  • Scott, Russell L.; Biederman, Joel A.; Hamerlynck, Erik P.
  • Journal of Geophysical Research: Biogeosciences, Vol. 120, Issue 12
  • DOI: 10.1002/2015JG003181

Impact of severe dry season on net ecosystem exchange in the Neotropical rainforest of French Guiana
journal, August 2008


Modeling carbon dynamics in two adjacent spruce forests with different soil conditions in Russia
journal, January 2008


Multiple-constraint inversion of SCOPE. Evaluating the potential of GPP and SIF for the retrieval of plant functional traits
journal, December 2019

  • Pacheco-Labrador, Javier; Perez-Priego, Oscar; El-Madany, Tarek S.
  • Remote Sensing of Environment, Vol. 234
  • DOI: 10.1016/j.rse.2019.111362

Carbon sequestration in a high-elevation, subalpine forest
journal, May 2002


Carbon dioxide exchange in a semidesert grassland through drought-induced vegetation change
journal, January 2010

  • Scott, Russell L.; Hamerlynck, Erik P.; Jenerette, G. Darrel
  • Journal of Geophysical Research, Vol. 115, Issue G3
  • DOI: 10.1029/2010JG001348

Evaluating the predictive power of sun-induced chlorophyll fluorescence to estimate net photosynthesis of vegetation canopies: A SCOPE modeling study
journal, April 2016

  • Verrelst, Jochem; van der Tol, Christiaan; Magnani, Federico
  • Remote Sensing of Environment, Vol. 176
  • DOI: 10.1016/j.rse.2016.01.018

Comparing methods for partitioning a decade of carbon dioxide and water vapor fluxes in a temperate forest
journal, October 2016


Diagnostics for Gaussian Process Emulators
journal, November 2009


Evaporation of perennial semi-arid woodland in southeastern Australia is adapted for irregular but common dry periods: Evaporation from Mallee Woodland
journal, March 2015

  • Meyer, W. S.; Kondrlovà, E.; Koerber, G. R.
  • Hydrological Processes, Vol. 29, Issue 17
  • DOI: 10.1002/hyp.10467

A new assessment of European forests carbon exchanges by eddy fluxes and artificial neural network spatialization
journal, April 2003


Widespread inhibition of daytime ecosystem respiration
journal, February 2019

  • Keenan, Trevor F.; Migliavacca, Mirco; Papale, Dario
  • Nature Ecology & Evolution, Vol. 3, Issue 3
  • DOI: 10.1038/s41559-019-0809-2

Annual variation in soil respiration and its components in a coppice oak forest in Central Italy
journal, September 2002


Deep learning and process understanding for data-driven Earth system science
journal, February 2019


A decade of carbon, water and energy flux measurements of an old spruce forest at the Anchor Station Tharandt
journal, January 2007


Partitioning forest carbon fluxes with overstory and understory eddy-covariance measurements: A synthesis based on FLUXNET data
journal, May 2007


On the Temperature Dependence of Soil Respiration
journal, June 1994

  • Lloyd, J.; Taylor, J. A.
  • Functional Ecology, Vol. 8, Issue 3
  • DOI: 10.2307/2389824

How do leaf and ecosystem measures of water-use efficiency compare?
journal, June 2017

  • Medlyn, Belinda E.; De Kauwe, Martin G.; Lin, Yan-Shih
  • New Phytologist, Vol. 216, Issue 3
  • DOI: 10.1111/nph.14626

Patterns and processes of carbon, water and energy cycles across northern Australian landscapes: From point to region
journal, November 2011


Experimental validation of footprint models for eddy covariance CO2 flux measurements above grassland by means of natural and artificial tracers
journal, August 2017


Partitioning carbon dioxide and water vapor fluxes using correlation analysis
journal, January 2010


Fluxes all of the time? A primer on the temporal representativeness of FLUXNET: FLUXES ALL OF THE TIME?
journal, February 2017

  • Chu, Housen; Baldocchi, Dennis D.; John, Ranjeet
  • Journal of Geophysical Research: Biogeosciences, Vol. 122, Issue 2
  • DOI: 10.1002/2016JG003576

Optimal inverse estimation of ecosystem parameters from observations of carbon and energy fluxes
journal, January 2019

  • Dutta, Debsunder; Schimel, David S.; Sun, Ying
  • Biogeosciences, Vol. 16, Issue 1
  • DOI: 10.5194/bg-16-77-2019

Scaling carbon fluxes from eddy covariance sites to globe: synthesis and evaluation of the FLUXCOM approach
journal, January 2020

  • Jung, Martin; Schwalm, Christopher; Migliavacca, Mirco
  • Biogeosciences, Vol. 17, Issue 5
  • DOI: 10.5194/bg-17-1343-2020

On the separation of net ecosystem exchange into assimilation and ecosystem respiration: review and improved algorithm
journal, September 2005


Cross-site evaluation of eddy covariance GPP and RE decomposition techniques
journal, June 2008


Ecosystem-level CO2 fluxes from a boreal cutover in eastern Canada before and after scarification
journal, November 2006


Spatiotemporal Variations in Growing Season Exchanges of CO 2 , H 2 O, and Sensible Heat in Agricultural Fields of the Southern Great Plains
journal, October 2007

  • Fischer, Marc L.; Billesbach, David P.; Berry, Joseph A.
  • Earth Interactions, Vol. 11, Issue 17
  • DOI: 10.1175/EI231.1

Remote estimation of carbon dioxide uptake by a Mediterranean forest
journal, December 2008


Characterization of ecosystem responses to climatic controls using artificial neural networks: CHARACTERIZING ECOSYSTEM RESPONSES
journal, August 2010


Emulation as an Accurate Alternative to Interpolation in Sampling Radiative Transfer Codes
journal, December 2018

  • Vicent, Jorge; Verrelst, Jochem; Rivera-Caicedo, Juan Pablo
  • IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Vol. 11, Issue 12
  • DOI: 10.1109/JSTARS.2018.2875330

Modelling carbon budget of Mediterranean forests using ground and remote sensing measurements
journal, December 2005


Effect of spatial sampling from European flux towers for estimating carbon and water fluxes with artificial neural networks: SAMPLING EFFECT ON FLUXES UPSCALING
journal, October 2015

  • Papale, Dario; Black, T. Andrew; Carvalhais, Nuno
  • Journal of Geophysical Research: Biogeosciences, Vol. 120, Issue 10
  • DOI: 10.1002/2015JG002997

Model-based analysis of the relationship between sun-induced chlorophyll fluorescence and gross primary production for remote sensing applications
journal, December 2016


A simple two-dimensional parameterisation for Flux Footprint Prediction (FFP)
journal, January 2015

  • Kljun, N.; Calanca, P.; Rotach, M. W.
  • Geoscientific Model Development, Vol. 8, Issue 11
  • DOI: 10.5194/gmd-8-3695-2015

Climatic controls and ecosystem responses drive the inter-annual variability of the net ecosystem exchange of an alpine meadow
journal, September 2011


Sustained carbon uptake and storage following moderate disturbance in a Great Lakes forest
journal, July 2013

  • Gough, Christopher M.; Hardiman, Brady S.; Nave, Lucas E.
  • Ecological Applications, Vol. 23, Issue 5
  • DOI: 10.1890/12-1554.1

The response of ecosystem carbon fluxes to LAI and environmental drivers in a maize crop grown in two contrasting seasons
journal, July 2015

  • Vitale, Luca; Di Tommasi, Paul; D’Urso, Guido
  • International Journal of Biometeorology, Vol. 60, Issue 3
  • DOI: 10.1007/s00484-015-1038-2

Ecosystem-level CO2 fluxes from a boreal cutover in eastern Canada before and after scarification
journal, November 2006


Partitioning forest carbon fluxes with overstory and understory eddy-covariance measurements: A synthesis based on FLUXNET data
journal, May 2007


Partitioning carbon dioxide and water vapor fluxes using correlation analysis
journal, January 2010


Climatic controls and ecosystem responses drive the inter-annual variability of the net ecosystem exchange of an alpine meadow
journal, September 2011


An improved isotopic method for partitioning net ecosystem–atmosphere CO2 exchange
journal, December 2015


Experimental validation of footprint models for eddy covariance CO2 flux measurements above grassland by means of natural and artificial tracers
journal, August 2017


Multiple-constraint inversion of SCOPE. Evaluating the potential of GPP and SIF for the retrieval of plant functional traits
journal, December 2019

  • Pacheco-Labrador, Javier; Perez-Priego, Oscar; El-Madany, Tarek S.
  • Remote Sensing of Environment, Vol. 234
  • DOI: 10.1016/j.rse.2019.111362

Gap filling strategies for defensible annual sums of net ecosystem exchange
journal, March 2001


CO2 and water vapour fluxes for 2 years above Euroflux forest site
journal, June 2001

  • Berbigier, Paul; Bonnefond, Jean-Marc; Mellmann, Patricia
  • Agricultural and Forest Meteorology, Vol. 108, Issue 3
  • DOI: 10.1016/s0168-1923(01)00240-4

A New Data Set to Keep a Sharper Eye on Land-Air Exchanges
journal, April 2017


Widespread inhibition of daytime ecosystem respiration
journal, February 2019

  • Keenan, Trevor F.; Migliavacca, Mirco; Papale, Dario
  • Nature Ecology & Evolution, Vol. 3, Issue 3
  • DOI: 10.1038/s41559-019-0809-2

Deep learning and process understanding for data-driven Earth system science
journal, February 2019


Linking plant and ecosystem functional biogeography
journal, September 2014

  • Reichstein, M.; Bahn, M.; Mahecha, M. D.
  • Proceedings of the National Academy of Sciences, Vol. 111, Issue 38
  • DOI: 10.1073/pnas.1216065111

Savanna fires and their impact on net ecosystem productivity in North Australia
journal, May 2007


Characterization of ecosystem responses to climatic controls using artificial neural networks: CHARACTERIZING ECOSYSTEM RESPONSES
journal, August 2010


A decade of carbon, water and energy flux measurements of an old spruce forest at the Anchor Station Tharandt
journal, January 2007


How do leaf and ecosystem measures of water-use efficiency compare?
journal, June 2017

  • Medlyn, Belinda E.; De Kauwe, Martin G.; Lin, Yan-Shih
  • New Phytologist, Vol. 216, Issue 3
  • DOI: 10.1111/nph.14626

An Emulator Toolbox to Approximate Radiative Transfer Models with Statistical Learning
journal, July 2015

  • Rivera, Juan; Verrelst, Jochem; Gómez-Dans, Jose
  • Remote Sensing, Vol. 7, Issue 7
  • DOI: 10.3390/rs70709347

Optimal inverse estimation of ecosystem parameters from observations of carbon and energy fluxes
journal, January 2019

  • Dutta, Debsunder; Schimel, David S.; Sun, Ying
  • Biogeosciences, Vol. 16, Issue 1
  • DOI: 10.5194/bg-16-77-2019

Works referencing / citing this record:

Will Artificial Intelligence supersede Earth System and Climate Models?
text, January 2021