Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Statistical upscaling of ecosystem CO2 fluxes across the terrestrial tundra and boreal domain : Regional patterns and uncertainties

Virkkala, Anna Maria ; Aalto, Juha ; Rogers, Brendan M. ; Tagesson, Torbern LU ; Treat, Claire C. ; Natali, Susan M. ; Watts, Jennifer D. ; Potter, Stefano ; Lehtonen, Aleksi and Mauritz, Marguerite , et al. (2021) In Global Change Biology 27(17). p.4040-4059
Abstract

The regional variability in tundra and boreal carbon dioxide (CO2) fluxes can be high, complicating efforts to quantify sink-source patterns across the entire region. Statistical models are increasingly used to predict (i.e., upscale) CO2 fluxes across large spatial domains, but the reliability of different modeling techniques, each with different specifications and assumptions, has not been assessed in detail. Here, we compile eddy covariance and chamber measurements of annual and growing season CO2 fluxes of gross primary productivity (GPP), ecosystem respiration (ER), and net ecosystem exchange (NEE) during 1990–2015 from 148 terrestrial high-latitude (i.e., tundra and boreal) sites to analyze the... (More)

The regional variability in tundra and boreal carbon dioxide (CO2) fluxes can be high, complicating efforts to quantify sink-source patterns across the entire region. Statistical models are increasingly used to predict (i.e., upscale) CO2 fluxes across large spatial domains, but the reliability of different modeling techniques, each with different specifications and assumptions, has not been assessed in detail. Here, we compile eddy covariance and chamber measurements of annual and growing season CO2 fluxes of gross primary productivity (GPP), ecosystem respiration (ER), and net ecosystem exchange (NEE) during 1990–2015 from 148 terrestrial high-latitude (i.e., tundra and boreal) sites to analyze the spatial patterns and drivers of CO2 fluxes and test the accuracy and uncertainty of different statistical models. CO2 fluxes were upscaled at relatively high spatial resolution (1 km2) across the high-latitude region using five commonly used statistical models and their ensemble, that is, the median of all five models, using climatic, vegetation, and soil predictors. We found the performance of machine learning and ensemble predictions to outperform traditional regression methods. We also found the predictive performance of NEE-focused models to be low, relative to models predicting GPP and ER. Our data compilation and ensemble predictions showed that CO2 sink strength was larger in the boreal biome (observed and predicted average annual NEE −46 and −29 g C m−2 yr−1, respectively) compared to tundra (average annual NEE +10 and −2 g C m−2 yr−1). This pattern was associated with large spatial variability, reflecting local heterogeneity in soil organic carbon stocks, climate, and vegetation productivity. The terrestrial ecosystem CO2 budget, estimated using the annual NEE ensemble prediction, suggests the high-latitude region was on average an annual CO2 sink during 1990–2015, although uncertainty remains high.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; ; ; ; ; ; and , et al. (More)
; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; and (Less)
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
Arctic, CO balance, empirical, greenhouse gas, land, permafrost, remote sensing
in
Global Change Biology
volume
27
issue
17
pages
4040 - 4059
publisher
Wiley-Blackwell
external identifiers
  • scopus:85107431149
  • pmid:33913236
ISSN
1354-1013
DOI
10.1111/gcb.15659
language
English
LU publication?
yes
id
5eb7eac8-80c0-4c81-8b3d-6621ae9ad588
date added to LUP
2021-07-12 12:12:42
date last changed
2022-08-11 22:52:54
@article{5eb7eac8-80c0-4c81-8b3d-6621ae9ad588,
  abstract     = {{<p>The regional variability in tundra and boreal carbon dioxide (CO<sub>2</sub>) fluxes can be high, complicating efforts to quantify sink-source patterns across the entire region. Statistical models are increasingly used to predict (i.e., upscale) CO<sub>2</sub> fluxes across large spatial domains, but the reliability of different modeling techniques, each with different specifications and assumptions, has not been assessed in detail. Here, we compile eddy covariance and chamber measurements of annual and growing season CO<sub>2</sub> fluxes of gross primary productivity (GPP), ecosystem respiration (ER), and net ecosystem exchange (NEE) during 1990–2015 from 148 terrestrial high-latitude (i.e., tundra and boreal) sites to analyze the spatial patterns and drivers of CO<sub>2</sub> fluxes and test the accuracy and uncertainty of different statistical models. CO<sub>2</sub> fluxes were upscaled at relatively high spatial resolution (1 km<sup>2</sup>) across the high-latitude region using five commonly used statistical models and their ensemble, that is, the median of all five models, using climatic, vegetation, and soil predictors. We found the performance of machine learning and ensemble predictions to outperform traditional regression methods. We also found the predictive performance of NEE-focused models to be low, relative to models predicting GPP and ER. Our data compilation and ensemble predictions showed that CO<sub>2</sub> sink strength was larger in the boreal biome (observed and predicted average annual NEE −46 and −29 g C m<sup>−2</sup> yr<sup>−1</sup>, respectively) compared to tundra (average annual NEE +10 and −2 g C m<sup>−2</sup> yr<sup>−1</sup>). This pattern was associated with large spatial variability, reflecting local heterogeneity in soil organic carbon stocks, climate, and vegetation productivity. The terrestrial ecosystem CO<sub>2</sub> budget, estimated using the annual NEE ensemble prediction, suggests the high-latitude region was on average an annual CO<sub>2</sub> sink during 1990–2015, although uncertainty remains high.</p>}},
  author       = {{Virkkala, Anna Maria and Aalto, Juha and Rogers, Brendan M. and Tagesson, Torbern and Treat, Claire C. and Natali, Susan M. and Watts, Jennifer D. and Potter, Stefano and Lehtonen, Aleksi and Mauritz, Marguerite and Schuur, Edward A.G. and Kochendorfer, John and Zona, Donatella and Oechel, Walter and Kobayashi, Hideki and Humphreys, Elyn and Goeckede, Mathias and Iwata, Hiroki and Lafleur, Peter M. and Euskirchen, Eugenie S. and Bokhorst, Stef and Marushchak, Maija and Martikainen, Pertti J. and Elberling, Bo and Voigt, Carolina and Biasi, Christina and Sonnentag, Oliver and Parmentier, Frans Jan W. and Ueyama, Masahito and Celis, Gerardo and St.Louis, Vincent L. and Emmerton, Craig A. and Peichl, Matthias and Chi, Jinshu and Järveoja, Järvi and Nilsson, Mats B. and Oberbauer, Steven F. and Torn, Margaret S. and Park, Sang Jong and Dolman, Han and Mammarella, Ivan and Chae, Namyi and Poyatos, Rafael and López-Blanco, Efrén and Christensen, Torben Røjle and Kwon, Min Jung and Sachs, Torsten and Holl, David and Luoto, Miska}},
  issn         = {{1354-1013}},
  keywords     = {{Arctic; CO balance; empirical; greenhouse gas; land; permafrost; remote sensing}},
  language     = {{eng}},
  number       = {{17}},
  pages        = {{4040--4059}},
  publisher    = {{Wiley-Blackwell}},
  series       = {{Global Change Biology}},
  title        = {{Statistical upscaling of ecosystem CO<sub>2</sub> fluxes across the terrestrial tundra and boreal domain : Regional patterns and uncertainties}},
  url          = {{http://dx.doi.org/10.1111/gcb.15659}},
  doi          = {{10.1111/gcb.15659}},
  volume       = {{27}},
  year         = {{2021}},
}