Oliver Schneising

Satellite observations of the column-averaged dry-air mole fractions of carbon dioxide (XCO2) and methane (XCH4) are essential for monitoring their global atmospheric distribution and variability, as well as for related climate applications. To generate long-term, consistent Climate Data Records (CDR), ensemble merging approaches such as the ensemble median algorithm (EMMA) have been used. Here, we introduce a new version of the EMMA algorithm, v5.1, and the four merged datasets that were created with it, namely merged Level 2 (L2) and corresponding gridded Level 3 (L3) products for both XCO2 and XCH4. EMMA v5.1 keeps the core idea from previous EMMA versions of the ensemble median being a good choice for merging multiple satellite data products, while taking a new approach: for each gas and time period, one of the L2 input products is used as a reference, and the ensemble is used to compute an empirical spatio-temporal adjustment for this input dataset. This new design eliminates conceptual limitations of former EMMA versions, making the L2 dataset eligible for hotspots emission estimations in periods in which the reference product has sufficient spatial coverage. Additionally, new and updated input L2 products were included in the ensemble, and the L3 datasets have higher spatial resolution, i.e. monthly 1°x1°, compared to the previous monthly 5°x 5° resolution. The four merged datasets cover the time period from 2003 to 2024. All four merged datasets were validated against ground-based Total Carbon Column Observing Network (TCCON) retrievals, showing good agreement. The XCO2 L2 (L3) merged datasets have a spatio-temporal bias of 0.47 ppm (0.48 ppm). The estimated single-sounding (single-cell) random error is 1.51 ppm (1.82 ppm), very close to the average reported uncertainty of 1.66 ppm (1.65 ppm). The estimated linear drift is 0.00 ± 0.12 ppm y−1 (0.00 ± 0.11 ppm y−1). The XCH4 L2 (L3) merged datasets have a spatio-temporal bias of 7.37 ppb (4.31 ppb). The estimated single-sounding (single-cell) random error is 20.87 ppb (9.01 ppb), very close to the average reported uncertainty of 21.11 ppb (11.05 ppb). The estimated linear drift is 0.42 ± 1.14 ppb y−1 (0.11 ± 0.62 ppb y−1).

Routine, passive monitoring using satellite sensors offers an effective means of understanding the location, frequency and intensity of industrial methane emissions. Satellites have been used largely to identify fugitive point source emissions in sectors such as oil and gas, but their efficacy for more diffuse sources such as landfills is less established, and formal integration into official reporting and compliance frameworks has yet to be achieved. A persistent methane hotspot originating from Madrid was identified using TROPOMI, where it remains one of the largest signals in Europe. Two landfills, Las Dehesas and Pinto, were identified as significant contributors based on GHGSat observations. This study presents a comprehensive analysis of the TROPOMI and GHGSat observational record up to 2024 for Madrid, alongside aircraft observations, all contextualised against previous short-term studies, facility-scale reporting, and gridded inventories. Plume-based city-scale emissions averaged 13 ± 9 t/hr from TROPOMI compared to ~8 t/hr from the gridded inventories. Facility-scale emissions ranged from 2.5 ± 1.3 t/hr (Las Dehesas) and 3.8 ± 2.0 t/hr (Pinto) from GHGSat, and 3.0 ± 1.8 to 4.6 ± 2.7 t/hr respectively from aircraft, compared to reported averages of 0.4 and 1.0 t/hr between 2018–2024. Measurement-derived rates exceeded reported estimates by a factor of 0.5–6, highlighting a severe gap between bottom-up inventory and top-down observational approaches. We discuss challenges associated with satellite interpretation and meteorology, connect emissions to on-site management practices, and make recommendations towards operational use of these datasets.

Satellite observations play a central role in monitoring atmospheric methane (CH₄), a potent greenhouse gas. Recently launched instruments such as TROPOMI provide unprecedented spatial coverage, however, heterogeneous sampling with both dense and sparsely covered regions poses challenges for inverse modeling systems. This uneven sampling can cause very densely observed areas to dominate the inversion and lead to vanishing gradients in sparsely covered regions. We present a parametric regularization method that computes observation-specific weighting factors based on the spatial and temporal distribution of measurements. Our method modifies the observational covariance matrix through a preprocessing step, homogenizing the effective weight of densely and sparsely sampled regions without introducing additional computational cost during the optimization. Since the parametric weighting is determined prior to the inversion, the approach can be used to calculate weights for multiple-instrument inversions, without further additional parameter tuning. The adaptive regularization is applied to global methane emission inversions for 2019, at 1° × 1° resolution, using the TMVar (TM5-MP/4DVAR v1.2.6 LAMOS) system, assimilating TROPOMI column observations. Compared to a constant regularization approach, the presented method reduces the relative variation in grid-cell weights by about 20 %, by strengthening constraints in high-latitude and oceanic regions. Posterior simulations from inversions that either assimilate satellite data only or also assimilate NOAA surface measurements are evaluated against independent TCCON column observations, showing that the method preserves overall inversion performance while improving the spatial balance of satellite influence. The resulting global methane budget is consistent with recent top-down estimates from the Global Methane Budget.

Routine, passive monitoring using satellite sensors offers an effective means of understanding the location, frequency and intensity of industrial methane emissions. Satellites have been used largely to identify fugitive point source emissions in sectors such as oil and gas, but their efficacy for more diffuse sources such as landfills is less established, and formal integration into official reporting and compliance frameworks has yet to be achieved. A persistent methane hotspot originating from Madrid was identified using TROPOMI, where it remains one of the largest signals in Europe. Two landfills, Las Dehesas and Pinto, were identified as significant contributors based on GHGSat observations. This study presents a comprehensive analysis of the TROPOMI and GHGSat observational record up to 2024 for Madrid, alongside aircraft observations, all contextualised against previous short-term studies, facility-scale reporting, and gridded inventories. Plume-based city-scale emissions averaged 13 ± 9 t/hr from TROPOMI compared to ~8 t/hr from the gridded inventories. Facility-scale emissions ranged from 2.5 ± 1.3 t/hr (Las Dehesas) and 3.8 ± 2.0 t/hr (Pinto) from GHGSat, and 3.0 ± 1.8 to 4.6 ± 2.7 t/hr respectively from aircraft, compared to reported averages of 0.4 and 1.0 t/hr between 2018–2024. Measurement-derived rates exceeded reported estimates by a factor of 0.5–6, highlighting a severe gap between bottom-up inventory and top-down observational approaches. We discuss challenges associated with satellite interpretation and meteorology, connect emissions to on-site management practices, and make recommendations towards operational use of these datasets.

Satellite observations play a central role in monitoring atmospheric methane (CH₄), a potent greenhouse gas. Recently launched instruments such as TROPOMI provide unprecedented spatial coverage, however, heterogeneous sampling with both dense and sparsely covered regions poses challenges for inverse modeling systems. This uneven sampling can cause very densely observed areas to dominate the inversion and lead to vanishing gradients in sparsely covered regions. We present a parametric regularization method that computes observation-specific weighting factors based on the spatial and temporal distribution of measurements. Our method modifies the observational covariance matrix through a preprocessing step, homogenizing the effective weight of densely and sparsely sampled regions without introducing additional computational cost during the optimization. Since the parametric weighting is determined prior to the inversion, the approach can be used to calculate weights for multiple-instrument inversions, without further additional parameter tuning. The adaptive regularization is applied to global methane emission inversions for 2019, at 1° × 1° resolution, using the TMVar (TM5-MP/4DVAR v1.2.6 LAMOS) system, assimilating TROPOMI column observations. Compared to a constant regularization approach, the presented method reduces the relative variation in grid-cell weights by about 20 %, by strengthening constraints in high-latitude and oceanic regions. Posterior simulations from inversions that either assimilate satellite data only or also assimilate NOAA surface measurements are evaluated against independent TCCON column observations, showing that the method preserves overall inversion performance while improving the spatial balance of satellite influence. The resulting global methane budget is consistent with recent top-down estimates from the Global Methane Budget.

Routine, passive monitoring using satellite sensors offers an effective means of understanding the location, frequency and intensity of industrial methane emissions. Satellites have been used largely to identify fugitive point source emissions in sectors such as oil and gas, but their efficacy for more diffuse sources such as landfills is less established, and formal integration into official reporting and compliance frameworks has yet to be achieved. A persistent methane hotspot originating from Madrid was identified using TROPOMI, where it remains one of the largest signals in Europe. Two landfills, Las Dehesas and Pinto, were identified as significant contributors based on GHGSat observations. This study presents a comprehensive analysis of the TROPOMI and GHGSat observational record up to 2024 for Madrid, alongside aircraft observations, all contextualised against previous short-term studies, facility-scale reporting, and gridded inventories. Plume-based city-scale emissions averaged 13 ± 9 t/hr from TROPOMI compared to ~8 t/hr from the gridded inventories. Facility-scale emissions ranged from 2.5 ± 1.3 t/hr (Las Dehesas) and 3.8 ± 2.0 t/hr (Pinto) from GHGSat, and 3.0 ± 1.8 to 4.6 ± 2.7 t/hr respectively from aircraft, compared to reported averages of 0.4 and 1.0 t/hr between 2018–2024. Measurement-derived rates exceeded reported estimates by a factor of 0.5–6, highlighting a severe gap between bottom-up inventory and top-down observational approaches. We discuss challenges associated with satellite interpretation and meteorology, connect emissions to on-site management practices, and make recommendations towards operational use of these datasets.

Satellite observations play a central role in monitoring atmospheric methane (CH4), a potent greenhouse gas. Recently launched instruments such as TROPOMI provide unprecedented spatial coverage, however, heterogeneous sampling with both dense and sparsely covered regions poses challenges for inverse modeling systems. This uneven sampling can cause very densely observed areas to dominate the inversion and lead to vanishing gradients in sparsely covered regions. We present a parametric regularization method that computes observation-specific weighting factors based on the spatial and temporal distribution of measurements. Our method modifies the observational covariance matrix through a preprocessing step, homogenizing the effective weight of densely and sparsely sampled regions without introducing additional computational cost during the optimization. Since the parametric weighting is determined prior to the inversion, the approach can be used to calculate weights for multiple-instrument inversions, without further additional parameter tuning. The adaptive regularization is applied to global methane emission inversions for 2019, at 1° × 1° resolution, using the TMVar (TM5-MP/4DVAR v1.2.6 LAMOS) system, assimilating TROPOMI column observations. Compared to a constant regularization approach, the presented method reduces the relative variation in grid-cell weights by about 20 %, by strengthening constraints in high-latitude and oceanic regions. Posterior simulations from inversions that either assimilate satellite data only or also assimilate NOAA surface measurements are evaluated against independent TCCON column observations, showing that the method preserves overall inversion performance while improving the spatial balance of satellite influence. The resulting global methane budget is consistent with recent top-down estimates from the Global Methane Budget.

Satellite observations from the Sentinel-5P TROPOMI instrument, combined with inverse modeling, provide a valuable resource for quantifying regional methane (CH ₄ ) emissions. This study compares the 2019 European emissions estimated from variational inversions assimilating three TROPOMI products of dry-column methane mole fractions (XCH ₄ ). The SRON (v2.4, operational product), BLENDED (v1.0), and WFMD (v1.8) products are retrieved from distinct algorithms. These retrievals differ in coverage, error characterization, and XCH ₄ spatial distribution. Machine learning predictions of XCH ₄ differences point to aerosol scattering and albedo sensitivity as the largest contributors to the differences. The derived 2019 European CH ₄ emission budgets show an increase of +2 % for SRON, and a decrease of −1 %, −33 % and −9 % relative to the prior, respectively, for BLENDED, WFMD and surface-based inversions. The range of budget estimates reveals inconsistencies in the total emissions derived from the inversions. At the national and sub-national scale, spatial emission patterns are similar for the non-independent SRON and BLENDED but differ substantially from WFMD. No inversion provides a systematically closer match to the spatial distribution of emissions derived from independent surface observations. Evaluation at surface stations shows that the residual between the observations and the posterior concentrations is reduced for 37 %, 53 % and 47 % of the stations, respectively, for SRON, BLENDED and WFMD. Observing System Simulation Experiments (OSSEs) are used to disentangle the drivers of differences between the posterior emissions. Results show that aligning coverage and individual observation errors increases the consistency between emission estimates. Residual differences in the OSSE posterior emissions can be attributed to differences in averaging kernels and prior profiles.

South Asia is a major contributor to global methane (CH ₄ ) emissions, yet its emissions remain poorly constrained, limiting targeted mitigation. Current bottom-up inventories do not consistently capture the magnitude and seasonality of CH ₄ emissions in this region, particularly during the monsoon. Here we quantify South Asian CH ₄ emissions for 2020 using column observations from a satellite instrument (TROPOMI), a Lagrangian transport model (FLEXPART), and a Bayesian inversion system (FLEXINVERT+). We estimate a posteriori emission of 73.0 ± 0.7 Tg yr ⁻¹ for South Asia, including 35.6 ± 0.5 Tg yr ⁻¹ for India and 13.2 ± 0.4 Tg yr ⁻¹ for Bangladesh. Agriculture and wetlands contribute substantially to the regional budget, with the flux increments coincident with rice-growing areas and inundated lowlands. The inversion indicates pronounced monsoon-modulated seasonality in South Asia: posterior fluxes are higher than the prior by about 19.3 Tg CH ₄ (an increase of ∼ 70 %) during June–September and lower during January–May by ∼ 46 %. Localized enhancements seen over the lower Indus Basin align with runoff patterns, while the seasonal peaks here are absent in inventories. By resolving monsoon seasonality with satellite constraints, our results point towards key uncertainties in the South Asian CH ₄ budget and underscore the need for process-based, seasonally responsive inventories to inform mitigation strategies and reconcile bottom-up and top-down estimates.