the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 12 May 2026
Tracking organic compounds in smoke plumes using infrared satellite-based measurements
Abstract. We apply new measurements of methanol, ethene, ethyne, and HCN from the Cross-track Infrared Sounder (CrIS) to explore the quantitative use of satellite-based thermal infrared (IR) observations for fire studies. We focus analysis on the western U.S. during the 2018–2019 timeframes of two fire-focused aircraft campaigns, and use the GEOS-Chem model to guide interpretation. The CrIS data reveal large in-smoke enhancements and species:species correlations for targeted volatile organic compounds (VOCs), especially during the more active 2018 fire year. Spectral enhancements are strongest for methanol and ethene. For VOCs with similar vertical sensitivities the in-smoke correlations are height-independent and can be converted to column enhancement ratios without plume altitude information. For VOCs with dissimilar vertical sensitivities, spectral index correlations change coherently with altitude and may constrain injection or plume height changes. The mean (± σ) ethene:methanol ratio measured by CrIS across an ensemble of plumes (0.64 ± 0.24 mol/mol) matches bottom-up emission ratios (0.63 ± 0.08 mol/mol), but satellite-based and aircraft data both reveal greater variability than is predicted by GEOS-Chem. We propose that fire pyrolysis conditions are one driver of this variability and use in-situ data to show that near-field ethene:methanol ratios track pyrolysis conditions and hence inform the abundance of other emitted VOCs. Finally, we apply CrIS ethene:methanol ratios to estimate the high-temperature pyrolysis fraction for the same plume ensemble; the resulting fraction correlates with fire radiative power in a manner not well-captured by models.
Competing interests: At least one of the (co-)authors is a member of the editorial board of Atmospheric Chemistry and Physics.
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.- Preprint
(5411 KB) - Metadata XML
-
Supplement
(3634 KB) - BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2026-1669', Anonymous Referee #1, 08 Jul 2026
-
RC2: 'Comment on egusphere-2026-1669', Anonymous Referee #2, 31 Aug 2026
The authors present an exploratory study illustrating the potential use of CrIS observations for constraining biomass-burning pyrolysis conditions and plume dynamics. For this study, the authors combine CrIS data with in situ measurements from the WE-CAN and FIREX-AQ campaigns, conducted in 2018 and 2019, respectively. They also compare their results with GEOS-Chem simulations as a baseline to evaluate how their findings differ from current parameterizations.
The study begins by illustrating the capability of the CrIS instrument to observe biomass-burning plumes and by establishing criteria to identify these plumes in an automated way. Next, the authors investigate the instrument's sensitivity to different compounds emitted by fires and measured by the instrument. This analysis serves as the basis for their claims regarding the potential use of these observations to constrain pyrolysis conditions and plume dynamics. When the instrument has different sensitivity profiles for specific VOCs, their ratio can be used as an indicator of plume localization. For VOCs with similar vertical sensitivity profiles, the authors reason that their ratio is representative of the in-plume ratio, independent of plume localization. Building on the work of Sekimoto et al., they assert that the ratio of methanol to ethene can be used to infer the contribution of high-temperature pyrolysis.
Historically, significant efforts have been made to characterize VOC emission ratios based on fuel type, whereas relatively less is known about the effect of fuel pyrolysis conditions due to limited experimental data. The current study proposes a methodology to address this gap using satellite observations.
The study is well written and provides convincing arguments to support its claims. Nonetheless, a few clarifications are still needed to explain some of the choices made and to provide statistical quantification of claims made throughout the text. As such, I recommend publication with minor corrections.
Minor comments:
- Section 2.4: Smoke selection criteria.
Where do the thresholds of 7 and 2 for Ethene HRIs come from? From Tables S1 and S2, I note that these thresholds are significantly above the mean HRI values for their respective years. However, the tables do not include information on the spread of the distributions, making it difficult to assess the statistical significance of the thresholds and, consequently, the potential impact of false-positive results. The double trigger with CO will help reduce this effect, but from visual inspection, I cannot infer a statistically significant difference between the ethene HRI distributions for smoke and non-smoke measurements. - Section 3.1: Please be more precise when contrasting measurements and avoid terms such as "clear impact," "modestly higher," or "strong enhancements" without quantified evidence (see technical comments).
- Section 3.2: What is the function of Fig. 3 and its discussion in the paper? It seems to deviate from the central storyline.
- Section 3.4:
- The analysis includes all plumes meeting the dense-smoke criteria, with four additional requirements. Why were requirements (ii) and (iii) needed? You have shown that both are well correlated on the regional scale. How and why do individual plumes deviate from this relationship? What are the implications for your analysis?
- Fig. 8b: Why did you choose to fit the function to the median values? What are the uncertainties in the fitted parameters? It seems unusual that you quantify the uncertainties for the GEOS-Chem and FIREX-AQ (Fig. 10) fits but not for the CrIS fit.
- Why could widespread regional smoke be an issue for the analysis of the WE-CAN data but not for the analysis of the CrIS observations?
Technical comments:
- L 274: "All four VOCs reveal a clear impact from fires." This is not clear without statistical evidence. For example, the range of ethene HRI values in plumes in 2019 is smaller than that of non-plume measurements in 2018. As it is a relatively short-lived tracer, the high background values should not be as significantly affected when the CO trigger has not been activated.
- L 300–305: Please quantify this type of statement. As above, the impact for ethene in 2019 is an increase in HRI of 0.27, which is lower than the increase for HCN (0.28) in the same year. Stating that the fire impact for ethene shows strong enhancements, but is less clear for HCN, therefore seems inconsistent with the data.
- Fig. 3: The gradients in the color scales are not representative of the gradients in the data, which makes any conclusion other than "2018 was more affected by fires" difficult to draw.
- L 389: "Chemical transport model simulations ...": Are these your simulations with GEOS-Chem, or was another model/study used to reach this conclusion?
- L 565: What does "statistically robust" mean?
- Fig. 8b: The dashed line is not explained in the legend or in the description of the figure. Why are both the boxplots and the data associated with each box shown on their right side?
- Figures 8, 9, and 10: The y-axis label for Figure 10 is not consistent with those of Figures 8 and 9, even though I understand the variable to be the same. Please make the labels consistent or clarify the difference.
- L 594: Fig. 3 is the figure for both WE-CAN and FIREX-AQ. The more relevant figure appears to be Fig. S3.
Citation: https://doi.org/10.5194/egusphere-2026-1669-RC2 - Section 2.4: Smoke selection criteria.
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 432 | 119 | 39 | 590 | 66 | 41 | 40 |
- HTML: 432
- PDF: 119
- XML: 39
- Total: 590
- Supplement: 66
- BibTeX: 41
- EndNote: 40
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
This manuscript presents an analysis of several VOCs observed by the CrIS satellite instrument in wildfire plumes above North America during the summers of 2018 and 2019 (when the intensive campaigns WE-CAN and FIREX-AQ occurred).
More specifically, the authors use hyperspectral range index (HRI) observations of methanol, ethene, ethyne and HCN. To support interpretation, they use simulations performed using the GEOS-Chem chemistry-transport model including biomass burning emissions from the GFAS inventory, and radiative-transfer simulation with the LBLRTM model.
The authors aim at deriving information on emissions as a function of fire characteristics using enhancement VOC ratios, exploiting the different vertical sensitivities of the instrument to each compound.
The dependence of emission factors depending on the type of fuel burned and the burning efficiency is well established. While the type of fuel burned can be estimated using databases of land cover and vegetation, the fraction of a wildfire that is in flaming or smouldering phase is very difficult to estimate. Due to the lack of systematic constrain, emission inventories usually do not differentiate emission factors depending on the combustion phase, which leads to major uncertainties. The study presented here is therefore very interesting as it discusses the possibility to infer information of the pyrolysis conditions using VOC ratios observed from satellite in freshly emitted dense plumes. They show that observed ethene:methanol ratios can be linked to the fire radiative power (FRP).
The paper is well written and well structured. The figures and supplementary information clearly support the analysis. The large uncertainty and need for further investigation is also well discussed. For these reasons, I recommend publication with minor corrections.
Main comments.
I think that it would be important to better explain the methodology and uncertainties in the conversion from HRI enhancement ratios to molar ratios. From what I understand, the authors assume a linear dependence in order to avoid a full retrieval of VOC columns but it seems too simplified. What is the uncertainty of this method compared to the ratio of L2 columns?
The possible contribution of other sources to the observed HRI is not discussed. Does a seasonal average remove all possible contributions from biogenic sources? Could the model be used to check this assumption? The ΔHRI is sometimes negative in the maps…
Specific comments.
Section 2.1.
Section 2.3.
Section 3.1.
Section 3.3.
Section 3.4.