loading page

Disaggregating the carbon exchange of degrading permafrost peatlands using Bayesian deep learning
  • +6
  • Norbert Pirk,
  • Kristoffer Aalstad,
  • Erik Schytt Mannerfelt,
  • François Clayer,
  • Heleen Agnes de Wit,
  • Casper Tai Christiansen,
  • Inge Althuizen,
  • Hanna Lee,
  • Sebastian Westermann
Norbert Pirk
Department of Geosciences, University of Oslo

Corresponding Author:[email protected]

Author Profile
Kristoffer Aalstad
University of Oslo
Author Profile
Erik Schytt Mannerfelt
University of Oslo
Author Profile
François Clayer
Norwegian Institute for Water Research
Author Profile
Heleen Agnes de Wit
Norwegian Institute for Water Research
Author Profile
Casper Tai Christiansen
University of Copenhagen
Author Profile
Inge Althuizen
NORCE Norwegian Research Centre
Author Profile
Hanna Lee
Unknown
Author Profile
Sebastian Westermann
University of Oslo
Author Profile

Abstract

Extensive regions in the permafrost zone are projected to become climatically unsuitable to sustain permafrost peatlands over the next century, suggesting transformations in these landscapes that can leave large amounts of permafrost carbon vulnerable to post-thaw decomposition. We present three years of eddy covariance measurements of CH4 and CO2 fluxes from the degrading permafrost peatland Iskoras in Northern Norway, which we disaggregate into separate fluxes of palsa, pond, and fen areas using information provided by the dynamic flux footprint in a novel ensemble-based Bayesian deep neural network framework. The three-year mean CO2-equivalent flux is estimated to be 106 gCO2 m-2 yr-1 for palsas, 1780 gCO2 m-2 yr-1 for ponds, and -31 gCO2 m-2 yr-1 for fens, indicating that possible palsa degradation to thermokarst ponds would strengthen the local greenhouse gas forcing by a factor of about 17, while transformation into fens would slightly reduce the current local greenhouse gas forcing.
11 May 2023Submitted to ESS Open Archive
13 May 2023Published in ESS Open Archive
28 May 2024Published in Geophysical Research Letters volume 51 issue 10. 10.1029/2024GL109283