the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 14 Nov 2025
Integrating MethaneAIR aircraft and TROPOMI satellite observations in the Integrated Methane Inversion (IMI) to optimize methane emissions
Abstract. The MethaneAIR aircraft remote sensing instrument observes methane dry air column mixing ratios (XCH4) over ~100×100 km2 scenes with sub-km resolution, from which methane emissions can be inferred by inverse analysis with an atmospheric transport model. It emulates the MethaneSAT satellite instrument launched in March 2024 to quantify emissions from oil/gas production regions. We show here how the single day MethaneAIR observations can be integrated with the global continuous but relatively coarse and sparse observations from the TROPOMI satellite instrument into a common Integrated Methane Inversion (IMI) platform for optimizing methane emissions. The IMI, originally designed for TROPOMI, is used here with 12×12 km2 spatial resolution and lognormal error probability density functions (PDFs) for prior estimates. Application to two scenes in oil/gas production basins of the western US shows remarkable consistency between independent MethaneAIR (single day) and TROPOMI (monthly) inversions including for emission hotspots, with some differences that may reflect temporal variability of emissions. The IMI is able to optimize emissions even when starting from a very poor prior estimate. Using TROPOMI inversion results as prior estimate improves the MethaneAIR inversions by correcting emissions upwind of the MethaneAIR observation scenes and by adding information to the original prior estimate.
- Preprint
(1517 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2025-4626', Anonymous Referee #1, 02 Jan 2026
-
RC2: 'Comment on egusphere-2025-4626', Anonymous Referee #2, 25 Mar 2026
General comments:
This manuscript integrated MethaneAIR aircraft and Tropomi satellite observations in the Integrated Methane Inversion to optimize methane emissions in Permian and Uinta oil/gas production basin. This work can improve methane emission estimates from top-down approach and validate bottom-up emission inventories. This topic is interesting and meaningful to readers.
Special comments:
- Figure 4. Why were posterior methane concentrations from TROPOMI not compared in this study, while methane emissions were compared with TROPOMI in Figure 3? Also, the same question for the following figures.
- Section 3.1. The author posted two figures first in this section, but only a few words about the concentrations shown in Figure 4. I suggest adding more discussion of concentrations, as the spatial distributions driven by posterior and prior emissions were still different.
- Section 3.2, Any differences in the prior emissions between GHGI and Omara et al., 2024 in the region of Uinta?
- Line 328, The data shown here was not consistent with the values in Figure 8.
- Is the framework developed in this study also suitable for other point-source methane observation satellites, such as PRISMA or GHGSAT, rather than MethaneSAT?
- Some small errors. Lines 101, change “10 and 12 local time” to 10:00 and 12:00 local time”, similar to that in line 104. Line 208, change “Total” to “total”.
-
EC1: 'Comment on egusphere-2025-4626', Jason Cohen, 26 Mar 2026
A reviewer’s comments arrived after the discussion phase closed. I am posting them here as an editor comment to ensure they are considered. The reviewer wishes to remain anonymous.
--------
The manuscript presents an interesting extension of the IMI framework to integrate high-resolution airborne methane observations. It is indeed timely, given that high-resolution empirical datasets both from remote sensing and aerial surveys are growing. Therefore, the results discussed are encouraging and indicate a useful application of an existing framework. The paper is also well written and easy to follow. However, I have a few concerns as listed below, which should be addressed to strengthen the paper.
1. The chosen argument for validating the integration
Authors chose to show the robustness of their integration by comparing a one-day airborne observation with monthly mean TROPOMI data. I believe, this is not sufficient to demonstrate that the framework performs equivalently for high-resolution data. Methane emissions from oil and gas systems are known to exhibit substantial temporal variability as also noted by the authors, including intermittency and episodic releases. As a result, the “remarkable” agreement between a snapshot airborne inversion and a monthly mean satellite inversion could be partly coincidental unless the authors can demonstrate that emissions in the study region are relatively stable over time. So mere comparison of posterior totals does not necessarily imply similar observational constraint.
Nevertheless, to strengthen this aspect of the study, I suggest that the authors:
- provide evidence for limited temporal variability in emissions over the study period, OR
- perform additional tests of consistency between the airborne and TROPOMI inversions. For example, running inversions with multiple prior estimates (authors can use 0.75x,1x, 1.25x of the same prior or use multiple different priors altogether) and examining whether the airborne and TROPOMI inversions “respond” similarly to changes in prior magnitude and prior uncertainty. So, instead of just the “posterior totals”, comparing the “behavior across different settings” would provide a more robust demonstration that the airborne observations are being assimilated in a manner consistent with the TROPOMI-based framework.
2. Selection of prior uncertainty (GSD) based on concentration RMSE
The prior uncertainty (GSD) is selected by minimizing the RMSE between GEOS-Chem simulations (using posterior emissions) and the same observations used in the inversion. While this is an interesting approach to balance the regularization and data fit, it can introduce a degree of circularity, as the same dataset is used both to constrain and evaluate the inversion. I would suggest authors to compare their optimal prior GSD with independent estimates (e.g., based on inventory uncertainty or facility-level variability)
Citation: https://doi.org/10.5194/egusphere-2025-4626-EC1 - AC1: 'Author Response To Referee Comments', Jack Bruno, 06 Apr 2026
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 495 | 133 | 37 | 665 | 46 | 48 |
- HTML: 495
- PDF: 133
- XML: 37
- Total: 665
- BibTeX: 46
- EndNote: 48
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
This study presents a timely and strategic application of the Integrated Methane Inversion (IMI) framework, combining high-resolution MethaneAIR observations with the global context provided by TROPOMI. This approach offers a potential pathway for processing data from future missions with high-resolution measurements. The manuscript is well-structured, and the logic is easy to follow. Several points of clarification on the inversion strategy and the consistency between observing systems would strengthen both the methodology and the discussion.
Major comments:
1. The study adopts a step-wise approach, using the TROPOMI posterior as a prior for the MethaneAIR inversion. The authors mention that concatenating data was avoided due to temporal mismatches. However, have the authors considered or tested a joint inversion where both observation vectors are included in a single cost function with appropriate error covariance matrices as used in the paper "Global methane budget and trend, 2010–2017: Complementarity of inverse analyses using in situ (GLOBALVIEWplus CH4 ObsPack) and satellite (GOSAT) observations"(https://acp.copernicus.org/articles/21/4637/2021/)? A discussion on why the step-wise approach is superior or more practical in this context would be valuable.
2. In the Permian case (RF06, Figure 1), TROPOMI XCH4 is significantly lower (11 ppb on average) than MethaneAIR. It is surprising that the independent inversions yield nearly identical total emissions (84.8 vs. 85.9 t/h) despite this substantial systematic offset. Could the authors clarify how the inversion achieves this consistency? Could this result primarily be driven by the adjustment of boundary conditions to compensate for the observational bias?
3. The coefficient 4.7 in equation (3) is unclear. Is this a fitted slope from the resampling experiment? Moreover, given that RF06 and RF08 likely differ in spatial heterogeneity, the authors should justify whether this single empirical constant is applicable to both flights, or if flight-specific coefficients would be more appropriate.
Minor comments:
1. The Introduction (Line 39) states that TROPOMI provides continuous data at a resolution of 5.5×7 km2. While this is true for this study period (2021), TROPOMI XCH4 products were initially provided with a 7×7 km2 resolution and switched to 5.5×7 km2 on August 6, 2019. This context should be corrected.
2. Please provide the specific temporal information and native spatial resolution for the GHGI and Omara et al. (2024) emission inventories.
3. MethaneAIR observations occur in the morning (10:00–12:00 LT), whereas the TROPOMI overpass typically occurs in the early afternoon (~13:30 LT). Apart from the transport errors mentioned, how are diurnal variations in emissions and boundary layer height accounted for when comparing or integrating these two datasets?
4. When aggregating very-high-resolution MethaneAIR pixels to the 12-km grid, is strict mass conservation or column-weighting maintained?