Satellite-based modeling of wetland methane emissions on a global scale (SatWetCH4 1.0)

Geoscientific model development, Journal Year: 2025, Volume and Issue: 18(3), P. 863 - 883

Published: Feb. 14, 2025

Abstract. Wetlands are major contributors to global methane emissions. However, their budget and temporal variability remain subject large uncertainties. This study develops the Satellite-based Wetland CH4 model (SatWetCH4), which simulates wetland emissions at 0.25° × monthly resolution, relying mainly on remote-sensing products. In particular, a new approach is derived assess substrate availability, based Moderate-Resolution Imaging Spectroradiometer (MODIS) data. The calibrated using eddy covariance flux data from 58 sites, allowing for independence other estimates. At site level, effectively reproduces magnitude seasonality of fluxes in boreal temperate regions but shows limitations capturing tropical sites. Despite its simplicity, provides simulations over decades produces consistent spatial patterns seasonal variations comparable more complex land surface models (LSMs). Such an independent data-driven products intended allow future studies intra-annual addition, our highlights uncertainties issues extent datasets need seamless satellite-based future, there potential integrate this one-step into atmospheric inversion frameworks, thereby optimization parameters concentrations as constraints hopefully better estimates

Language: Английский

---

Atmospheric Transport Modeling of CO₂ With Neural Networks

Journal of Advances in Modeling Earth Systems, Journal Year: 2025, Volume and Issue: 17(2)

Published: Feb. 1, 2025

Abstract Accurately describing the distribution of in atmosphere with atmospheric tracer transport models is essential for greenhouse gas monitoring and verification support systems to aid implementation international climate agreements. Large deep neural networks are poised revolutionize weather prediction, which requires 3D modeling atmosphere. While similar this regard, subject new challenges. Both, stable predictions longer time horizons mass conservation throughout need be achieved, while IO plays a larger role compared computational costs. In study we explore four different (UNet, GraphCast, Spherical Fourier Neural Operator SwinTransformer) have proven as state‐of‐the‐art prediction assess their usefulness modeling. For this, assemble CarbonBench data set, systematic benchmark tailored machine learning emulators Eulerian transport. Through architectural adjustments, decouple performance our from shift caused by steady rise . More specifically, center input fields zero mean then use an explicit flux scheme fixer assure balance. This design enables conserving over 6 months all network architectures. study, SwinTransformer displays particularly strong emulation skill: 90‐day physically plausible multi‐year forward runs. work paves way toward high resolution inverse inert trace gases networks.

Language: Английский

---

Quantification of Methane in Water at Parts Per Billion Sensitivity Using a Metal–Organic Framework-Functionalized Quartz Crystal Resonator

ACS Applied Nano Materials, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 26, 2025

Language: Английский

---

Temperature and Water Levels Collectively Regulate Methane Emissions From Subtropical Freshwater Wetlands

Global Biogeochemical Cycles, Journal Year: 2025, Volume and Issue: 39(3)

Published: March 1, 2025

Abstract Wetlands are the largest and most climate‐sensitive natural sources of methane. Accurately estimating wetland methane emissions involves reconciling inversion (“top‐down”) process‐based (“bottom‐up”) models within global budget. However, estimates from these two model types inherently interdependent often reveal substantial discrepancies. To enhance reliability both approaches, we need a comprehensive understanding an independent high‐resolution long‐term flux data set. Here, employed data‐driven random forest approach to identify key variables influencing subtropical freshwater wetlands in Southeastern United States. The model‐estimated monthly mean fluxes fit well with measured ( R 2 = 0.67) at four representative FLUXNET‐CH4 sites across region. Variable importance analysis highlighted sensitivity variations temperature water levels. High temperatures facilitate methanogenesis by enhancing microbial activities, while elevated levels maintain anaerobic conditions necessary for production. Notably, response level fluctuations is contingent on conditions, vice versa. Moreover, constructed first high‐spatial‐resolution (∼1 km × 1 km) (1982–2010) gridded regional product States, annual region 4.93 ± 0.11 Tg CH 4 yr −1 1982–2010. This new benchmark holds promise validating parameterizing uncertain emission processes bottom‐up provides improved prior information top‐down models.

Language: Английский

---

The North American Greenhouse Gas Budget: Emissions, Removals, and Integration for CO₂, CH₄, and N₂O (2010–2019): Results From the Second REgional Carbon Cycle Assessment and Processes Study (RECCAP2)

Global Biogeochemical Cycles, Journal Year: 2025, Volume and Issue: 39(4)

Published: April 1, 2025

Abstract Accurate accounting of greenhouse‐gas (GHG) emissions and removals is central to tracking progress toward climate mitigation for monitoring potential climate‐change feedbacks. GHG budgeting reporting can follow either the Intergovernmental Panel on Climate Change methodologies National Greenhouse Gas Inventory (NGHGI) or use atmospheric‐based “top‐down” (TD) inversions process‐based “bottom‐up” (BU) approaches. To help understand reconcile these approaches, Second REgional Carbon Cycle Assessment Processes study (RECCAP2) was established quantify carbon dioxide (CO 2 ), methane (CH 4 ) nitrous oxide (N O), ten‐land five‐ocean regions 2010–2019. Here, we present results North American land region (Canada, United States, Mexico, Central America Caribbean). For 2010–2019, NGHGI reported total net‐GHG 7,270 TgCO ‐eq yr −1 compared TD estimates 6,132 ± 1,846 BU 9,060 898 . Reconciling differences between NGHGI, approaches depended (a) lateral fluxes CO along land‐ocean‐aquatic continuum (LOAC) trade, (b) correcting land‐use loss‐of‐additional‐sink capacity (LASC), (c) avoiding double counting inland water CH emissions, (d) adjusting area match definition managed‐land proxy. Uncertainties remain from inland‐water evasion, conversion nitrogen fertilizers N O, less‐frequent non‐Annex‐1 countries. The RECCAP2 framework plays a key role in reconciling independent GHG‐reporting support policy commitments while providing insights into biogeochemical processes responses change.

Language: Английский

---

The net ecosystem carbon balance (NECB) at catchment scales in the Arctic

Frontiers in Environmental Science, Journal Year: 2025, Volume and Issue: 13

Published: April 7, 2025

The Net Ecosystem Carbon Balance (NECB) is a crucial metric for understanding integrated carbon dynamics in Arctic and boreal regions, which are vital to the global cycle. These areas associated with significant uncertainties rapid climate change, potentially leading unpredictable alterations dynamics. This mini-review examines key components of NECB, including sequestration, methane emissions, lateral transport, herbivore interactions, disturbances, while integrating insights from recent permafrost region greenhouse gas budget syntheses. We emphasize need holistic approach quantify incorporating all their uncertainties. review highlights methodological advances flux measurements, improvements eddy covariance automatic chamber techniques, as well progress modeling approaches data assimilation. Key research priorities identified, such improving representation inland waters process-based models, expanding monitoring networks, enhancing integration long-term field observations approaches. efforts essential accurately quantifying current future budgets rapidly changing northern landscapes, ultimately informing more effective change mitigation strategies ecosystem management practices. aligns goals Monitoring Assessment Program (AMAP) Conservation Flora Fauna (CAFF), providing important policymakers, researchers, stakeholders working understand protect these sensitive ecosystems.

Language: Английский

---

Global Methane Budget 2000–2020

Earth system science data, Journal Year: 2025, Volume and Issue: 17(5), P. 1873 - 1958

Published: May 9, 2025

Abstract. Understanding and quantifying the global methane (CH4) budget is important for assessing realistic pathways to mitigate climate change. CH4 second most human-influenced greenhouse gas in terms of forcing after carbon dioxide (CO2), both emissions atmospheric concentrations have continued increase since 2007 a temporary pause. The relative importance compared those CO2 temperature change related its shorter lifetime, stronger radiative effect, acceleration growth rate over past decade, causes which are still debated. Two major challenges factors responsible observed arise from diverse, geographically overlapping sources uncertain magnitude temporal destruction by short-lived highly variable hydroxyl radicals (OH). To address these challenges, we established consortium multidisciplinary scientists under umbrella Global Carbon Project improve, synthesise, update regularly stimulate new research on cycle. Following Saunois et al. (2016, 2020), present here third version living review paper dedicated decadal budget, integrating results top-down emission estimates (based situ Greenhouse Gases Observing SATellite (GOSAT) observations an ensemble inverse-model results) bottom-up process-based models estimating land surface chemistry, inventories anthropogenic emissions, data-driven extrapolations). We recent 2010–2019 calendar decade (the latest period full data sets available), previous 2000–2009 year 2020. revision this 2025 edition benefits progress inland freshwater with better counting lakes ponds, reservoirs, streams rivers. This also reduces double across wetland and, first time, includes estimate potential that may exist (average 23 Tg yr−1). Bottom-up approaches show combined average 248 [159–369] yr−1 decade. Natural fluxes perturbed human activities through climate, eutrophication, use. In estimate, component contributing emissions. Newly available gridded products allowed us derive almost complete latitudinal regional based approaches. For estimated inversions (top-down) be 575 (range 553–586, corresponding minimum maximum model ensemble). Of amount, 369 or ∼ 65 % attributed direct fossil, agriculture, waste biomass burning 350–391 63 %–68 %). period, give slightly lower total than 2010–2019, 32 9–40). 2020 highest reaches 608 581–627), 12 higher 2000s. Since 2012, trends been tracking scenarios assume no minimal mitigation policies proposed Intergovernmental Panel Climate Change (shared socio-economic SSP5 SSP3). methods suggest 16 (94 yr−1) larger (669 yr−1, range 512–849) inversion period. discrepancy between budgets has greatly reduced differences (167 156 2020) respectively), time uncertainties overlap. Although bottom-up, source uncertainty attributable natural especially wetlands freshwaters. tropospheric loss methane, as main contributor at 563 [510–663] chemistry–climate models. These values due impact rise remaining large (∼ 25 sink 633 [507–796] 554 [550–567] However, use same OH distribution, introduces less likely justified. agriculture contributed 228 [213–242] 211 [195–231] budget. Fossil fuel 115 [100–124] 120 [117–125] Biomass biofuel 27 [26–27] 28 [21–39] identify five priorities improving budget: (i) producing global, high-resolution map water-saturated soils inundated areas emitting robust classification different types ecosystems; (ii) further development inland-water emissions; (iii) intensification local (e.g. FLUXNET-CH4 measurements, urban-scale monitoring, satellite imagery pointing capabilities) scales (surface networks remote sensing measurements satellites) constrain inversions; (iv) improvements transport representation photochemical sinks (v) integration 3D variational systems using isotopic and/or co-emitted species such ethane well information super-emitters detected (mainly oil sector but coal, landfills) improve partitioning. presented can downloaded https://doi.org/10.18160/GKQ9-2RHT (Martinez al., 2024).

Language: Английский

---

Characterizing Performance of Freshwater Wetland Methane Models Across Time Scales at FLUXNET‐CH₄ Sites Using Wavelet Analyses

Journal of Geophysical Research Biogeosciences, Journal Year: 2023, Volume and Issue: 128(11)

Published: Nov. 1, 2023

Abstract Process‐based land surface models are important tools for estimating global wetland methane (CH 4 ) emissions and projecting their behavior across space time. So far there no performance assessments of model responses to drivers at multiple time scales. In this study, we apply wavelet analysis identify the dominant scales contributing uncertainty in frequency domain. We evaluate seven 23 eddy covariance tower sites. Our study first characterizes site‐level patterns freshwater CH fluxes (FCH different A Monte Carlo approach was developed incorporate flux observation error avoid misidentification that dominate error. results suggest (a) significant model‐observation disagreements mainly multi‐day (<15 days); (b) most can capture variability monthly seasonal (>32 days) boreal Arctic tundra sites but have bias temperate tropical/subtropical sites; (c) errors exhibit increasing power spectrum as scale increases, indicating biases <5 days could contribute persistent systematic on longer scales; (d) differences pattern related structure (e.g., proxy production). evaluation suggests need accurately replicate FCH variability, especially short scales, future developments.

Language: Английский

---

WetCH₄: A Machine Learning-based Upscaling of Methane Fluxes of Northern Wetlands during 2016–2022

Published: April 3, 2024

Abstract. Wetlands are the largest natural source of methane (CH4) emissions globally. Northern wetlands (>45° N), accounting for 42 % global wetland area, increasingly vulnerable to carbon loss, especially as CH4 may accelerate under intensified high-latitude warming. However, magnitude and spatial patterns remain relatively uncertain. Here we present estimates daily fluxes obtained using a new machine learning-based upscaling framework (WetCH4) that applies most complete database eddy covariance (EC) observations available date, satellite remote sensing informed environmental conditions at 10-km resolution. The important predictor variables included near-surface soil temperatures (top 40 cm), vegetation reflectance, moisture. Our results, modeled from 138 site-years across 26 sites, had strong predictive skill with mean R2 0.46 0.62 absolute error (MAE) 23 nmol m-2 s-1 21 monthly fluxes, respectively. Based on model estimated an annual average 20.8 ±2.1 Tg yr-1 northern region (2016–2022) total budgets ranged 13.7–44.1 yr-1, depending map extents. Although 86 budget occurred during May–October period, considerable amount (1.4 ±0.2 CH4) winter. Regionally, West Siberian accounted majority (51 %) interannual variation in domain emissions. Significant issues data coverage remain, only sites observing year-round 11 Alaska 10 bog/fen Canada Fennoscandia, general, Western Lowlands underrepresented by EC sites. results provide high spatiotemporal information cycle possible responses climate change. Continued, all-season tower improved moisture products needed future improvement upscaling. dataset can be found https://doi.org/10.5281/zenodo.10802154 (Ying et al., 2024).

Language: Английский

---

Unraveling Spatially Diverse and Interactive Regulatory Mechanisms of Wetland Methane Fluxes to Improve Emission Estimation

Environmental Science & Technology, Journal Year: 2024, Volume and Issue: unknown

Published: Aug. 12, 2024

Methane fluxes (FCH

Language: Английский

---