CaltechAUTHORS
Published June 14, 2023 | Version public
Journal Article Metadata-only

Observational constraints reduce model spread but not uncertainty in global wetland methane emission estimates

Creators

  • 1. ROR icon Lawrence Berkeley National Laboratory
  • 2. ROR icon Oak Ridge National Laboratory
  • 3. ROR icon University of Illinois at Chicago
  • 4. ROR icon ETH Zurich
  • 5. ROR icon University of British Columbia
  • 6. ROR icon University of California, Berkeley
  • 7. ROR icon Stanford University
  • 8. ROR icon Goddard Space Flight Center
  • 9. ROR icon Laboratoire des Sciences du Climat et de l'Environnement
  • 10. ROR icon Japan Agency for Marine-Earth Science and Technology
  • 11. ROR icon Environment and Climate Change Canada
  • 12. ROR icon National Institute for Environmental Studies
  • 13. ROR icon University of Bern
  • 14. ROR icon Max Planck Institute for Meteorology
  • 15. ROR icon University of Sydney
  • 16. ROR icon European Centre for Medium-Range Weather Forecasts
  • 17. ROR icon Lund University
  • 18. ROR icon Japan Meteorological Agency
  • 19. ROR icon Agriculture and Food
  • 20. ROR icon Chiba University
  • 21. ROR icon Hunan Normal University
  • 22. ROR icon University of Quebec at Montreal
  • 23. ROR icon Peking University
  • 24. ROR icon Netherlands Organisation for Applied Scientific Research
  • 25. ROR icon Boston College
  • 26. ROR icon Finnish Meteorological Institute
  • 27. ROR icon East China Normal University
  • 28. ROR icon California Institute of Technology
  • 29. ROR icon Earth System Science Interdisciplinary Center
  • 30. ROR icon Institute of Tibetan Plateau Research
  • 31. ROR icon Hohai University
  • 32. ROR icon Purdue University West Lafayette

Abstract

The recent rise in atmospheric methane (CH₄) concentrations accelerates climate change and offsets mitigation efforts. Although wetlands are the largest natural CH₄ source, estimates of global wetland CH₄ emissions vary widely among approaches taken by bottom-up (BU) process-based biogeochemical models and top-down (TD) atmospheric inversion methods. Here, we integrate in situ measurements, multi-model ensembles, and a machine learning upscaling product into the International Land Model Benchmarking system to examine the relationship between wetland CH₄ emission estimates and model performance. We find that using better-performing models identified by observational constraints reduces the spread of wetland CH₄ emission estimates by 62% and 39% for BU- and TD-based approaches, respectively. However, global BU and TD CH₄ emission estimate discrepancies increased by about 15% (from 31 to 36 TgCH₄ year⁻¹) when the top 20% models were used, although we consider this result moderately uncertain given the unevenly distributed global observations. Our analyses demonstrate that model performance ranking is subject to benchmark selection due to large inter-site variability, highlighting the importance of expanding coverage of benchmark sites to diverse environmental conditions. We encourage future development of wetland CH₄ models to move beyond static benchmarking and focus on evaluating site-specific and ecosystem-specific variabilities inferred from observations.

Additional Information

© 2023 John Wiley & Sons This study was funded by the RUBISCO SFA of the Regional and Global Modeling Analysis (RGMA)and the E3SM program in the U.S. Department of Energy Office of Science under contract DE-AC02-05CH11231. This work was also conducted as a part of the Wetland FLUXNET Synthesis for Methane Working Group supported by the John Wesley Powell Center for Analysis and Synthesis of the U.S. Geological Survey. The compilation of the FLUXNET-CH₄ data is supported by the Gordon and Betty Moore Foundation through Grant GBMF5439 "Advancing Understanding of the Global Methane Cycle" to Stanford University supporting the Methane Budget activity for the Global Carbon Project (globalcarbonproject.org). We acknowledge the FLUXNET-CH₄ community product (Delwiche et al., 2021) and Global Carbon Project CH₄ modeling group (Saunois et al., 2020) for the data provided in this analysis. We thank Peter Bergamaschi for sharing the TM5-CAMS model data used in this study. FJ acknowledges support by the Swiss National Science Foundation (#200020_200511). FM and CP acknowledge the National Computational Infrastructure of the National Computational Infrastructure of the Australian Government through the NCMAS Allocation Scheme (grant NCMAS-2021-78), and the Sydney Informatics Hub HPC Allocation Scheme supported by the Office of the Deputy Vice-Chancellor (Research). N.G. acknowledges support from the Newton Fund through the Met Office Climate Science for Service Partnership Brazil (CSSP Brazil). Code Availability: The source code, including usage tutorials, of the ILAMB system can be downloaded from https://github.com/rubisco-sfa/ILAMB. Data Availability Statement: The FLUXNET-CH₄ community product can be downloaded from https://fluxnet.org/data/fluxnet-ch4-community-product/. The modeling data contributed to the 2008–2017 global CH₄ budget are available from ICOS (doi: 10.18160/gcp-ch4-2019). The benchmarking and modeling data analyzed in this study can be downloaded at https://zenodo.org/record/7880014.

Additional details

Identifiers

Eprint ID
121889
Resolver ID
CaltechAUTHORS:20230612-735873000.62

Funding

Department of Energy (DOE)
DE-AC02-05CH11231
USGS
Gordon and Betty Moore Foundation
GBMF5439
Swiss National Science Foundation (SNSF)
200020_200511
National Computational Infrastructure (Australia)
NCMAS-2021-78
University of Sydney
Climate Science for Service Partnership Brazil

Dates

Created
2023-06-14
Created from EPrint's datestamp field
Updated
2023-06-14
Created from EPrint's last_modified field