| 19 Aug 2025
A multi-year global methane data set obtained by merging observations from TROPOMI and IASI
Abstract. Data products of atmospheric methane (CH4) with improved vertical sensitivity in the lower troposphere are crucial for gaining a more comprehensive understanding of the impact of anthropogenic emissions. This study presents a CH4 data product derived from the synergetic combination of level 2 (L2) data from TROPOMI (Tropospheric Monitoring Instrument) and IASI (Infrared Atmospheric Sounding Interferometer), specifically CH4 total column and CH4 profiles, respectively. IASI enables high-quality observation of CH4 mixing ratios in the upper troposphere and lower stratosphere, and TROPOMI observations excel in providing sensitivity to the total column-averaged mixing ratio of CH4. By combining the IASI and TROPOMI L2 products synergetically, we can detect tropospheric CH4 (mixing ratios averaged over a layer from the surface up to 450 hPa) that is not significantly affected by the strong CH4 variations around the tropopause. This is not achievable by using IASI or TROPOMI data alone.
For the synergetic L2 data combination, we use the method as presented in detail in Schneider et al. (2022b), and apply it to combine about 444 million individual and high-quality TROPOMI observations with about 805 million individual and high-quality IASI observations made globally over 42 months (from January 2018 to June 2021). The combination method is fast; it uses a tool designed for efficient geo-matching between large data sets and a computationally cheap Kalman filter for calculations and for merging the data sets. We show that the combined data set has a good global coverage. Moreover, we document that the sensitivity (response of the combined data product to real atmospheric CH4 variations) is extremely satisfactory throughout the globe, and the uncertainties are generally below 12–15 ppbv. Furthermore, we demonstrate the increased scientific value of the combined data product when compared to the two individual data products.
The data set of the combined product consists of about 289 million individual data points, and it is provided as NetCDF files. One file has a typical size of 280 MB and contains all data for observations made in one day (the universal time of the TROPOMI observations are taken as the reference time). For review, the data are accessible at https://radar.kit.edu/radar/en/dataset/wq583rnzpmd83m5g?token=UEuECSWHlGgWBdoPVsvI (Shahzadi et al., 2025) and made freely available at https://www.imk-asf.kit.edu/english/CH4-synergy-IASI-TROPOMI_RemoTeC.php.
This is a review of the manuscript titled “A multi-year global methane data set obtained by merging observations from TROPOMI and IASI” submitted to ESSD by Shahzadi et al. The authors use the algorithms proposed by Schneider et al. to produce a daily, global, long-term methane dataset from TROPOMI and IASI satellite retrievals. Overall, this work continues their previous studies, fits the scope of the journal, and is timely because methane has recently risen sharply and exerts a strong warming influence on the climate. Therefore, I recommend publication in ESSD after the authors address several issues related to key technical validation details and the presentation of results.
Main comments:
1. Although the authors have assessed the dataset accuracy using TM5 outputs and EDGAR emissions, ground measurements were not used for a more independent validation. I suggest a qualified comparison of the dataset with in situ/ground based results (e.g., TCCON).
2. The manuscript states that IASI is “adjusted to TROPOMI a priori (TM5)”, but it is not explicit whether this is (a) a re retrieval with a different a priori, (b) an additive/multiplicative post processing correction, or (c) a linearization/offset mapping. The precise operation matters because it affects bias propagation and degrees of freedom.
3. The Kalman update uses instrument posterior covariances as input. It is unclear how those covariances were pretreated (regularized, truncated, or inflated), whether they include forward-model uncertainty, and how numerical inversion (or pseudo-inversion) is handled for large matrices.
4. Displacement errors are crucial for tropospheric products and are only described by reference to prior work. The current manuscript lacks the exact formulas, parameter choices (time/space scales), and tests showing sensitivity to those choices.
5. The manuscript notes lower DOFS and increased noise in high-altitude areas, but does not quantify the implications for users (e.g., systematic smoothing, negative bias, or overconfidence).
Minor / Technical issues
1. Add a detailed NetCDF variable list (names, units, dimensions), an explanation of averaging kernel format/shape, and a short example (Python/xarray snippet) showing how to read XCH4, troXCH4, the averaging kernel, uncertainty, and the quality bitmask. Provide a recommended QC selection for typical use.
2. The selection of matching windows (50 km / 6 h, then normalized distance minimization) should be justified quantitatively. If possible, add sensitivity tests for different spatial/time windows (e.g., 25 km / 3 h, 75 km / 8 h) to show tradeoffs between coverage and displacement error.
3. In the Methods section, define each symbol and its dimension immediately below the equation. For Eq. (4), (5), (6), and (9), clarify matrix sizes and transpose conventions, and explicitly state when operations are performed in log space versus linear space (you mention omitting log-transform details for brevity, please explicitly state in Methods whether final arrays are reported on the linear or log scale).
4. For Fig. 10, ensure aerosol/snow masking and other flags are clearly labeled in the panels or legend, not only in the title; please check and apply the same clarity to other figures as needed.