Mengyao Liu

Wetland methane (CH ₄ ) emission seasonality is a major yet uncertain component of the global CH ₄ seasonal cycle, with its quantification primarily limited by the challenges in accurately characterizing wetland dynamics and emission processes. Here, we proposed a remote sensing–based framework to analyze the seasonality of tropical wetland CH ₄ emission systems by integrating satellite methane inversions with high-resolution wetland inundation mapping (distinguishing open water and inundated vegetation). Our results reveal a strong seasonal amplitude (4.75 teragrams per year) and pronounced seasonal hysteresis between precipitation and CH ₄ emissions (7 months) in Lake Chad. We demonstrate that the cascading chain of precipitation–wetland inundation–vegetation succession–CH ₄ emissions drives emission seasonality, characterized by a 3-month lag from precipitation to wetland inundation and a subsequent 4-month lag from inundation to peak emissions. These findings provide comprehensive and mechanistically generalizable insights into tropical wetland CH ₄ seasonality and offer critical constraints for improving tropical CH ₄ flux estimates.

Wetland methane (CH4) emission seasonality is a major yet uncertain component of the global CH4 seasonal cycle, limited by challenges in accurately characterizing wetland dynamics and CH4 emissions. Advances in satellite observations, methodology and computing capabilities enable the production of tropical basin-scale wetland-CH4 seasonality at a finer scale. Here, we quantify monthly CH₄ emissions in the Lake Chad Basin, Africa—a region with pronounced hydroclimatic variability—using an advanced divergence method and TROPOspheric Monitoring Instrument (TROPOMI) satellite observations. We distinguish open water and inundated vegetation and quantify their monthly area dynamics using high-resolution Sentinel-2 data (10 m). Our results reveal a strong seasonal amplitude (4.75 Tg a⁻¹) and significant seasonal hysteresis between CH4 emissions and wetland inundation (i.e., greater CH4 emissions during the reduction period of inundation, +1.68 Tg a⁻¹). The cascading link of precipitation–wetland inundation–vegetation succession–CH4 emissions is proven to drive emission seasonality, comprising a three-month lag for wetland inundation and a subsequent four-month lag for CH4 emissions. These findings provide new constraints for tropical CH4 flux estimates, which are the dominant sources of uncertainty between the bottom-up models (± 44—65%), and differences from satellite observations. It is also crucial to improve our understanding of the driver(s) of tropical wetland CH4 emissions to better assess the impact of future climate changes on these wetland emissions and potential positive feedback.

Regions with intensive agriculture, e.g. India, Belgium, Netherlands, and the Po-Valley, often suffer from air pollution and acidification/nitrification of the soil. In addition, these regions are often responsible for the release of methane emissions. Methane (CH4), the second most important greenhouse gas (GHG) after carbon dioxide (CO2), is emitted from cattle farms, rice paddies and the use of manure. Excessive anthropogenic emissions of nitrogen compounds to the environment have a major effect on the biogeochemical nitrogen cycle. Agricultural activities produce noteworthy ammonia (NH3) and nitrogen oxides (NOx) emissions. NH3 is mainly emitted from stables and via the spreading of manure and use of fertilizers. NOx emissions mainly stem from fossil fuel combustion, while soil emissions are dominant in remote areas. The role of soil NOx emissions on air quality is usually underestimated. Current methods for estimating emissions of those gases are based on the collection of activity data with associated emission factors having large uncertainties.The AGATE project of ESA aims to provide and improve agricultural emissions of CH4, NH3, and NOx independently by using satellite observations, i.e. without relying on the reported information or a-priori information. Satellite-derived emission estimates are calculated for targeted agricultural regions in Europe and South(east) Asia. The derived emissions are downscaled and validated to provide high-resolution emissions for specific subsectors (crops and livestock) and agricultural hotspots. In addtion, nitrogen deposition modelling is conducted to assess the impact of nitrogen deposition on natural areas within the European regions under study.

The role of tropical wetlands in the rise of atmospheric methane over the past two decades remains unclear. Current models define wetlands by surface water coverage, not making the distinction between high-emitting inundated vegetation and much weaker-emitting open water areas of wetlands. Here we use 30-m Landsat satellite data to identify tropical inundated vegetation, and combine it with chamber- and flux tower-derived emission intensities per unit wetland area to produce a Wetland Methane Emission Inventory (WMEI) at 0.1o ×0.1o spatial resolution and quarterly temporal resolution for 2004-2023. We find that tropical wetland emissions increased steadily by 16.3 Tg a-1 or 18% over the 2004-2023 period, with contributions from Africa (11.3 Tg a-1) and Asia (5.0 Tg a-1) and no net increase from South America. Tropical wetlands thus account for 21% of the global methane rise over that period though they do not appear to have contributed significantly to the 2020-2022 methane surge. The long-term trend and interannual variability of wetland emissions correlates strongly with vegetation activity as measured by solar-induced fluorescence (SIF) but not with temperature or precipitation.

Methane (CH₄) is the second most important greenhouse gas after CO₂, and its emissions from the agricultural sector, particularly rice paddies and dairy farms, remain highly uncertain and challenging to quantify. While recent advancements in satellite technology, such as high spatial resolution instruments, have enabled the detection of methane sources from global to facility scales, agricultural emissions still pose challenges. These emissions are typically diffuse and area-like, making them less detectable by targeted satellites like GHGSat and EMIT, which are better suited for isolated point sources such as oil/gas facilities or landfills. Additionally, agricultural emissions exhibit significant spatiotemporal variability driven by climate conditions, water management practices in rice paddies, and differences in farm types.In the AGATE project of ESA, we apply an improved divergence method to estimate monthly methane emissions using TROPOspheric Monitoring Instrument (TROPOMI) satellite observations at a 0.1° grid resolution. We focus on major agricultural regions, including the Po Valley in Italy, as well as India and Bangladesh, over the period 2019-2022. To better isolate agricultural emissions, we separate area-like sources (e.g., rice paddies) from isolated point sources. The locations of identified big emitters are cross-validated using bottom-up emission inventories and targeted satellite observations (e.g., EMIT, Carbon Mapper) to minimize the influence of non-agricultural sources. Furthermore, to better understand the seasonality of methane emissions, we analyze the correlations between methane emission variations and auxiliary datasets, such as rice paddy maps and ammonia emissions derived from satellites.

Wetland methane emissions are an important source of uncertainty in the methane budget due to their significant spatial and temporal variabilities. The Lake Chad Basin is located in central Africa and comprises a number of transboundary waters, which exhibit dramatic expansion and contraction. However, methane emissions from Lake Chad seem not to be properly captured in bottom-up emission inventories. An improved divergence method has been developed to estimate gridded methane (CH4) emissions from satellite observations of the TROPOspheric Monitoring Instrument (TROPOMI). Significant annual methane emissions over the Lake Chad Basin are identified by both the official reprocessed (S5P_RPRO_L2__CH4) and WFM-DOAS (TROPOMI/WFMD v1.8) XCH4 products. The maximum methane emissions appear from December to February while the minimum emissions are found during June to August. We further extract the monthly surface water areas using Landsat satellite imagery and wetland areas based on the MODIS vegetation index. The monthly variations of methane emissions are consistent with monthly surface water areas and wetlands areas but in contrast to the monthly rainfall. The seasonal emissions during the period of 2018 to 2022 over the Lake Chad Basin have been studied to better understand the role of driving factors such as rainfall, temperature, and waterlogged soils.

We introduce an innovative method to distinguish soil nitrogen oxides (NO_x = NO + NO₂) emissions from satellite‐based total NO_x emissions. To evaluate the approach, we compare the deviation between the tropospheric NO₂ concentration observed by satellite and two atmospheric composition model simulations driven by the newly estimated soil NO_x emissions and the Copernicus Atmosphere Monitoring Service (CAMS) inventory. The estimated average soil NO_x emissions in Europe are 1.7 kg N ha⁻¹ yr⁻¹ in 2019, and the annual soil NO_x emissions are approximately 1.7 times larger than that of the CAMS inventory. The discrepancy originates mainly from the forests, which would mean that the soil NO_x emissions over forest areas in Europe are currently underestimated by the CAMS inventory. The model evaluation indicates that the simulations driven by DECSO‐soil emissions performed significantly better than using CAMS‐soil. Overall, the simulated RMSE% of DECSO‐soil is lower than that of CAMS‐soil, approximately 6% lower in spring and 2% lower in autumn. Our method can easily be extended to other regions in the world despite having monthly variations that are very different from those in Europe.