Machine-learning-based estimates of global natural vegetated wetland methane emissions (2000–2025)

Abstract. Wetlands are the largest natural source of atmospheric methane (CH4), yet comprehensive global budgets are typically delayed by years, preventing a timely understanding of CH4 sources, sinks, and trends. To reduce this delay, we present a model emulator-driven framework and accompanying workflow that enable timely, continuous emission updates using a machine-learning emulator to reconstruct spatially explicit monthly emission fields at 1° × 1° resolution. We apply this framework to a global dataset of natural vegetated wetland CH4 emissions to extend the most recent Global Methane Budget (GMB; Saunois et al., 2025) record that covers the 2000–2020 emissions through 2025. In the test data (∼ 30 % of the total dataset), the emulator achieved a global R2 of 0.65 ± 0.003 (mean ± 95 % CI, hereafter) and an RMSE of 5.49±0.12×10-3 Tg CH4 yr−1. The emulator is trained on 35 GMB model estimates, including 22 process-based models and 13 atmospheric inversions, paired with 10 ensemble realizations of 11 gridded climate predictor variables from atmospheric reanalyses. Our results show that the global mean predicted wetland CH4 emissions for 2021–2025 (157.8 ± 2.4 Tg CH4 yr−1) are not significantly higher (∼ 0.05 Tg CH4 yr−1) than the 2000–2020 baseline. However, this stability masks a significant hemispheric redistribution of emissions. We detect an increase in Northern Hemisphere (NH) emissions in 2021–2025, with mid- and high-latitudes increasing by 0.76 ± 0.07 and 0.35 ± 0.03 Tg CH4 yr−1, respectively, while the tropics and Southern Hemisphere (SH) extratropics show offsetting negative trends (−0.95 ± 0.19 and -0.11±0.02 Tg CH4 yr−1, respectively). The predicted emissions are able to capture the low emissions in 2023 in South America linked to El Niño-related drought, as reported by recent studies (Ciais et al., 2026; Quinn et al., 2025). Furthermore, we identify a distinct seasonal amplification of global emission trends that peaks in late boreal summer. This new modeled dataset and operational framework bridge the gap between the latest updated budgets and low-latency monitoring, providing a scalable capacity to frequently update global emission estimates and critical early warnings of regional wetland feedback loops. The data are publicly available at https://doi.org/10.5281/zenodo.18870108 (Li et al., 2026).

Authors

Institutions

Publication Details

Journal
Earth system science data
Published
2026-05-22
DOI
https://doi.org/10.5194/essd-18-3507-2026
Primary Topic
Atmospheric and Environmental Gas Dynamics
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Machine-learning-based estimates of global natural vegetated wetland methane emissions (2000–2025)

Zhangcai Qin, Philippe Bousquet, Dmitry Belikov, Adrien Martinez et al.
Earth system science data
Atmospheric and Environmental Gas Dynamics
article

Machine-learning-based estimates of global natural vegetated wetland methane emissions (2000–2025)

Zhangcai Qin, Philippe Bousquet, Dmitry Belikov, Adrien Martinez, Nicola Gedney, Qianlai Zhuang, Thomas Kleinen, Jurek Müller, David J. Beerling, T F Li, Bo Zheng, Shamil Maksyutov, Yosuke Niwa, Naveen S Chandra, Aki Tsuruta, Atul Jain, Akihito Ito, Kyle McDonald, Alison M. Hoyt, Etienne Fluet-Chouinard, Qiuan Zhu, Fortunat Joos, Changhui Peng, Shushi Peng, Marielle Saunois, Masayuki Kondo, Danilo Custodio, Zhen Zhang, Qing Zhu, Hanqin Tian, Kunxiaojia Yuan, Avni Malhotra, Wenxin Zhang, Katherine Jensen, Xiangyu Liu, Rona L. Thompson, William Riley, Prabir K. Patra, Shufen Pan, Robert B. Jackson, Joe R. Melton, Xinyu Dou, Zutao Ouyang, Ting Zhang, Ben Poulter, Fa Li, Peter O. Hopcroft, Peter Raymond, Mengze Li, Philippe Ciais, Yi Xi, Arjo Segers, Kazuhito Ichii, Josep G. Canadell
article en

Abstract

Abstract. Wetlands are the largest natural source of atmospheric methane (CH4), yet comprehensive global budgets are typically delayed by years, preventing a timely understanding of CH4 sources, sinks, and trends. To reduce this delay, we present a model emulator-driven framework and accompanying workflow that enable timely, continuous emission updates using a machine-learning emulator to reconstruct spatially explicit monthly emission fields at 1° × 1° resolution. We apply this framework to a global dataset of natural vegetated wetland CH4 emissions to extend the most recent Global Methane Budget (GMB; Saunois et al., 2025) record that covers the 2000–2020 emissions through 2025. In the test data (∼ 30 % of the total dataset), the emulator achieved a global R2 of 0.65 ± 0.003 (mean ± 95 % CI, hereafter) and an RMSE of 5.49±0.12×10-3 Tg CH4 yr−1. The emulator is trained on 35 GMB model estimates, including 22 process-based models and 13 atmospheric inversions, paired with 10 ensemble realizations of 11 gridded climate predictor variables from atmospheric reanalyses. Our results show that the global mean predicted wetland CH4 emissions for 2021–2025 (157.8 ± 2.4 Tg CH4 yr−1) are not significantly higher (∼ 0.05 Tg CH4 yr−1) than the 2000–2020 baseline. However, this stability masks a significant hemispheric redistribution of emissions. We detect an increase in Northern Hemisphere (NH) emissions in 2021–2025, with mid- and high-latitudes increasing by 0.76 ± 0.07 and 0.35 ± 0.03 Tg CH4 yr−1, respectively, while the tropics and Southern Hemisphere (SH) extratropics show offsetting negative trends (−0.95 ± 0.19 and -0.11±0.02 Tg CH4 yr−1, respectively). The predicted emissions are able to capture the low emissions in 2023 in South America linked to El Niño-related drought, as reported by recent studies (Ciais et al., 2026; Quinn et al., 2025). Furthermore, we identify a distinct seasonal amplification of global emission trends that peaks in late boreal summer. This new modeled dataset and operational framework bridge the gap between the latest updated budgets and low-latency monitoring, providing a scalable capacity to frequently update global emission estimates and critical early warnings of regional wetland feedback loops. The data are publicly available at https://doi.org/10.5281/zenodo.18870108 (Li et al., 2026).

Earth system science dataVol. 18(5)
Boston College (US), Hiroshima University (JP), University of Bern (CH), Finnish Meteorological Institute (FI), Commonwealth Scientific and Industrial Research Organisation (AU), Met Office (GB), Centre National de la Recherche Scientifique (FR), Oeschger Centre for Climate Change Research (CH), Planetary Science Institute (US), Japan Agency for Marine-Earth Science and Technology (JP), Climate Central (US), Environment and Climate Change Canada (CA), Pacific Northwest National Laboratory (US), Lawrence Berkeley National Laboratory (US), Netherlands Organisation for Applied Scientific Research (NL), University of Illinois Urbana-Champaign (US), Sun Yat-sen University (CN), Université du Québec à Montréal (CA), Chiba University (JP), Hohai University (CN), National University of Singapore (SG), Hunan Normal University (CN), City University of New York (US), Lund University (SE), Université de Versailles Saint-Quentin-en-Yvelines (FR), Chinese Academy of Sciences (CN), National Institute for Environmental Studies (JP), Peking University (CN), Purdue University West Lafayette (US), Commissariat à l'Énergie Atomique et aux Énergies Alternatives (FR), Université Paris-Saclay (FR), University Town of Shenzhen (CN), Yale University (US), Southern Marine Science and Engineering Guangdong Laboratory (Guangzhou) (CN), Tsinghua–Berkeley Shenzhen Institute (CN), Beijing Municipal Ecology and Environment Bureau (CN), Laboratoire des Sciences du Climat et de l'Environnement (FR), CEA Paris-Saclay (FR), Stanford Medicine (US), Institute of Tibetan Plateau Research (CN), Institute of Atmospheric Physics (CN), Max Planck Institute for Meteorology (DE), Southern Marine Science and Engineering Guangdong Laboratory (Zhuhai) (CN), University of Houston (US), The University of Tokyo (JP), University of Glasgow (GB), University of Birmingham (GB), Auburn University (US), The University of Texas at Austin (US), University of Sheffield (GB), Stanford University (US)
National University of Singapore
Climate action
Openalex Percentile: Top 6%
Atmospheric and Environmental Gas Dynamics
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.