A satellite data driven biophysical modeling approach for estimating northern peatland and tundra CO₂ and CH₄ fluxes

• Main Authors: J. D. Watts, J. S. Kimball, F. J. W. Parmentier, T. Sachs, J. Rinne, D. Zona, W. Oechel, T. Tagesson, M. Jackowicz-Korczyński, M. Aurela
• Format: Article
• Language: English
• Published: Copernicus Publications 2014-04-01
• Series: Biogeosciences
• Online Access: http://www.biogeosciences.net/11/1961/2014/bg-11-1961-2014.pdf

The northern terrestrial net ecosystem carbon balance (NECB) is contingent on inputs from vegetation gross primary productivity (GPP) to offset the ecosystem respiration (<i>R</i>_eco) of carbon dioxide (CO₂) and methane (CH₄) emissions, but an effective framework to monitor the regional Arctic NECB is lacking. We modified a terrestrial carbon flux (TCF) model developed for satellite remote sensing applications to evaluate wetland CO₂ and CH₄ fluxes over pan-Arctic eddy covariance (EC) flux tower sites. The TCF model estimates GPP, CO₂ and CH₄ emissions using in situ or remote sensing and reanalysis-based climate data as inputs. The TCF model simulations using in situ data explained > 70% of the <i>r</i>² variability in the 8 day cumulative EC measured fluxes. Model simulations using coarser satellite (MODIS) and reanalysis (MERRA) records accounted for approximately 69% and 75% of the respective <i>r</i>² variability in the tower CO₂ and CH₄ records, with corresponding RMSE uncertainties of &leq; 1.3 g C m⁻² d⁻¹ (CO₂) and 18.2 mg C m⁻² d⁻¹ (CH₄). Although the estimated annual CH₄ emissions were small (< 18 g C m⁻² yr⁻¹) relative to <i>R</i>_eco (> 180 g C m⁻² yr⁻¹), they reduced the across-site NECB by 23% and contributed to a global warming potential of approximately 165 ± 128 g CO₂eq m⁻² yr⁻¹ when considered over a 100 year time span. This model evaluation indicates a strong potential for using the TCF model approach to document landscape-scale variability in CO₂ and CH₄ fluxes, and to estimate the NECB for northern peatland and tundra ecosystems.