the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 09 Sep 2024
Spatiotemporal variations in atmospheric CH4 concentrations and enhancements in northern China based on a comprehensive dataset: Ground-based observations, TROPOMI data, inventory data and inversions
Abstract. Methane (CH4) is a potent greenhouse gas with a global warming potential that is 28–36-fold higher than that of CO2 at the 100-year scale. Northern China notably contributes to CH4 emissions. However, high uncertainties remain in emissions, and observation gaps exist in this region, especially in urban areas. Here, we compiled a comprehensive dataset (available at https://doi.org/10.5281/zenodo.10957950) (Han et al., 2024), including ground- and satellite-based observations, inventory data and modeling results, to study the CH4 concentration, enhancement and spatiotemporal variation in this area. High-precision in situ observations from Beijing and Xianghe revealed that obvious seasonal cycles and notable enhancements (500–1500 ppb) occurred at a regional background site (Shangdianzi). We found significant increasing trends in the CH4 concentration over time in both the ground- and satellite-based observations and positive correlations between these observations. Anthropogenic emissions largely contributed to surface concentration variations and their increases in middle and southern Shanxi Province and northern Hebei Province. However, a spatially inconsistent pattern was observed between the results of optimized simulations driven by surface atmospheric inversion data and Tropospheric Monitoring Instrument (TROPOMI) column CH4 observations in summer. Further validation on the basis of this comprehensive dataset indicated that the TROPOMI data may exhibit systematic bias in summer. The posterior concentrations generally agreed well with the surface in situ observations (mean biases ranging from -2.3~80.7 ppb). The posterior surface CH4 concentrations (with a spatial resolution of 0.5°× 0.625°) revealed that southern Shanxi, northern Henan, and Beijing exhibited relatively high levels (an increase of ~300 ppb), which were positively correlated with the PKU-CH4-v2 emission inventory data. This study provides a comprehensive dataset of CH4 concentrations and enhancements in high-emission areas, which can benefit the research community and policy-makers for designing future observations, conducting atmospheric inversions and formulating policies.
- Preprint
(2834 KB) - Metadata XML
-
Supplement
(2220 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2024-2162', Anonymous Referee #2, 01 Oct 2024
Review of “Spatiotemporal variations in atmospheric CH4 concentrations and enhancements in northern China based on a comprehensive dataset: Ground-based observations, TROPOMI data, inventory data and inversions” by Han et al.
This study investigates methane (CH₄) emissions in Northern China, a region recognized for its significant contribution to global greenhouse gas emissions. Although CH4 is a potent greenhouse gas, uncertainties surrounding CH₄ emissions persist, particularly in urban and industrial areas of Northern China. To address these uncertainties, the authors compiled a comprehensive dataset that integrates ground and satellite observations, inventory data, and modeling results. This dataset aims to analyze CH₄ concentrations, enhancements, and their spatiotemporal variations in the region. High-precision in situ observations from locations such as Beijing and Xianghe revealed distinct cycles and variation, with concentration enhancements ranging from 500 to 1500 ppb. The findings indicate a significant upward trend in CH₄ concentrations over time, corroborated by both ground and satellite observations. The study highlights that anthropogenic activities are major contributors to variations in surface CH₄ concentrations, especially in middle and southern Shanxi Province and northern Hebei Province. However, discrepancies were noted between optimized simulations based on surface atmospheric inversion data and observations from the Tropospheric Monitoring Instrument (TROPOMI) during summer months, suggesting potential systematic biases. Their posterior analysis of CH₄ concentrations revealed that areas such as southern Shanxi, northern Henan, and Beijing exhibited elevated levels (an increase of approximately 300 ppb), which correlated positively with existing emission inventory data. This research provides essential insights into CH₄ emissions in high-emission areas of Northern China and offers a valuable dataset for researchers and policymakers aiming to enhance observational strategies and formulate effective emission reduction policies.
In my opinion this paper is a bit weak for ACP. However, since the paper is anchored in measurements, I would like to see it worked on and ultimately published in ACP. If the authors are willing to address the three general points listed here as well as specific details below, I believe that the standard can reach that of ACP and make a high quality and impactful paper.
(a) The authors need to look at more literature and bring in more observations. Knowledge of where coal mine and oil well emissions are located on the ground does not currently match with the modeled emissions fields.
(b) The authors demonstrate that there is extensive day-to-day variability in the observations. They should therefore find a way to evaluate the products more effectively. The way in which they compute month-by-month values and evaluate is out of date, as well as weakens the value of their high frequency observations.
(c) They need to do a bit more work on their underlying explanations of their findings. I found the data to be far more interesting than the conclusions and statements made.
Detailed Questions/Points to consider:- What exactly does the author mean when they state urban emissions of CH4? Do they mean leaky pipes, non-fully combusted CH4 from cooking and water heating, CH4 from urban rubbish, or do they also mean indirect CH4 from the goods and products consumed in urban areas? Similarly, the CH4 from coal and oil are large, but seem to not be included. What about large industrial sources which use coal, oil, or natural gas and continue to leak/emit CH4? I think that better and wider referencing and more extensive literature review can address this issue.
- What is the impact from outdated a priori emissions datasets? Is it related to the magnitude being wrong, the spatial distribution being wrong, the temporal variation being wrong, or some combination of these factors together? (Zhang et al., 2024)
- These satellite platforms measure radiance, which is then used to invert or constrain a column concentration loading. This step involves many assumptions and approximations, and produces uncertainty. There are additional assumptions and steps required to go from concentration to emissions. This second set of steps introduces further uncertainties and errors. In fact, there is a lot of literature recently published which demonstrates that this later step is not error free or fast and simple, and can lead to extremely different emissions end point inversions, especially at high spatial and temporal resolution. (Guanter et al., 2021; Pei et al., 2023; Qin et al., 2023).
- Are you sure that only 30% of anthropogenic CH4 from China was from your region of interest? I would think that the coal based CH4 emissions from Shanxi and Anhui and Oil based CH4 emissions from Shandong would be close to this number on their own, based on some newer studies. Only 5.7Tg/yr of emissions from Shanxi seems too low. Perhaps characterizing a range of values across the literature would be more effective rather than relying on percentages? (Qin et al., 2024).
- What is meant by a methane enhancement? What is the background value used? Does this include instrument minimum reliable observations including uncertainty, or just some other statistical technique? Since concentrations and total column loading are influenced by height, how is this corrected for in terms of deciding where is background? On top of this, how is background considered when making monthly or seasonal plots and comparisons? Day-to-day would be easier to achieve. (Chen and Prinn, 2005; Rivera Martinez et al., 2023).
- All three of your surface stations are near to sea level. Are these representatives of the coal emitting regions in Shanxi, which are found from 1000m to 2000m above sea level. What approach is used to account for this elevation difference? (Li et al., 2023)
- All three of your surface stations also observe ranges of CH4 concentration which is far lower than high values observed in and around coal and oil areas in Shanxi, Anhui, and Shandong. Will this make a difference when doing inverse modeling? (Liu et al., 2022)
- Given the very strong North to South gradient in CH4 background concentration in the Northern Hemisphere (as observed at GAW stations and from AGAGE), would it make sense to use the data from Mona Loa at 19oN to compare to the background at stations located around 40oN? (Prinn et al., 2018)
- Figure 1: What does Emissions (ht) mean? What unit is this?
- In terms of data from TCCON, it is well known that they do not operate well under high aerosol or cloud conditions. This region is known to have high AOD. What percentage of the total observations did not retrieve CH4 due to these conditions? Are there differences in the surface CH4 between when retrievals were successful and not? (Tu et al., 2020; Laughner et al., 2024)?
- Why was TROPOMI data used with QA of 0.5? How many days of missing data are there when aggregating the data in time from day-to-day to your output frequency? How much missing data is there at the three spatial scales you are aggregating to (0.1x0.1? At 0.3x0.3? at 0.5x0.5)? Are there differences in surface concentration between days with TROPOMI data and without TROPOMI data? (Qu et al., 2021; Worden et al., 2012)
- Since you have not used the same retrieval assumptions for TCCON and for TROPOMI, therefore you cannot quantify the accuracy. Please re-word this sentence. Precision can possibly be determined, but would require an explicit treatment of the uncertainties. This work may be heavy, so please at least just talk about how you would do this in the future. (Povey et al., 2015)
- Your a priori data is annual. How is this distributed for both coal mine emissions and agricultural emissions, which are variable throughout different times of the year?
- Since the observations are daily, I would want to see spatial maps of the daily comparisons between TROPOMI and the modeled results. Especially on some highly emitting days. Other studies have already been doing this level of comparison. It could be done by using the data to filter high emissions days, or even some objective approach such as EOF. In the event of a high emissions perturbation, the transport time across the domain is more than 1 day, allowing the emissions to still be within the observed region, while at monthly scale they will have smeared around much of the northern hemisphere. (Lalongo et al., 2020; Verhoelst et al., 2021; Wang et al., 2021)
- Based on figure 2, the variation seems to be at daily or higher frequency, consistent with. Your analysis mentions September and January as being the most extreme, yet I cannot see it from the data. From my eyes, November, December, February, and other months also have very high values. The patterns seem more complex than seasonal.
- Figure 3: hour-by-hour observations require meteorology to interpret. Were there changes in the wind speed or boundary layer height?
- In Figure 3, what happened to December (D)?
- Given the strong global gradient at different latitudes, perhaps comparisons should be made with other large urban areas and industrial areas at a similar latitude?
- Can Figure 4 include comparisons at daily scale?
- In Figure 4 TROPOMI looks very low compared to TCCON at XH, much more so than the uncertainty given in the paper. Please explain.
- In Figure 4, how to explain the cases where TROPOMI is higher than PICARRO in BJ?
- Where daily comparisons are made (In Figure 7) the RMSE error ranges from 110ppb to 185ppb, which is in the range from 5% to 10% of the xCH4. If this is robust, this is an important finding for the satellite community. Please highlight this result.
- The comparison with CAMS is only done monthly, and does not seem to offer any improvement to the story. Plus CAMS has been found to not be a good match with other remotely-sensed datasets over China. If there is no added value, perhaps this part can be left out, so the other parts can be focused on more clearly? (Liu Z. et al., 2024; Zhang et al., 2022).
- Do the high emissions spots in Figure 9 match the known locations of high gas emitting coal mines and oil fields in Shanxi and other areas? How can the artificially low urban signals in Nanjing and other heavily urban parts of the southern edge of the domain, as well as the artificially low coal signals in southwestern Shanxi along the western edge be explained? Is this an artifact of being located near the edge of the domain? (Cohen and Prinn, 2011)
References:
Butz, A., Galli, A., Hasekamp, O., Landgraf, J., Tol, P., and Aben, I.: TROPOMI aboard Sentinel-5 Precursor: Prospective performance of CH4 retrievals for aerosol and cirrus loaded atmospheres, Remote Sens. Environ., 120, 267–276, https://doi.org/10.1016/j.rse.2011.05.030, 2012.
Chen, Y.-H. and Prinn, R. G.: Atmospheric modeling of high- and low-frequency methane observations: Importance of interannually varying transport, J. Geophys. Res., 110, D10303, https://doi.org/10.1029/2004JD005542, 2005.
Cohen, J. B., & Prinn, R. G. (2011). Development of a fast, urban chemistry metamodel for inclusion in global models. Atmospheric Chemistry and Physics, 11(15), 7629–7656. https://doi.org/10.5194/acp-11-7629-2011
Guanter, L., Irakulis-Loitxate, I., Gorroño, J., Sánchez-García, E., Cusworth, D. H., Varon, D. J., Cogliati, S., & Colombo, R. (2021). Mapping methane point emissions with the PRISMA spaceborne imaging spectrometer. Remote Sensing of Environment, 265, 112671. https://doi.org/10.1016/j.rse.2021.112671
Hu, W., Qin, K., Lu, F. et al. Merging TROPOMI and eddy covariance observations to quantify 5-years of daily CH4 emissions over coal-mine dominated region. Int J Coal Sci Technol 11, 56 (2024). https://doi.org/10.1007/s40789-024-00700-1
Kang, H., Qin, K., Fan, L., Wei, H., Qing, X., and Cohen, J.B.: Methane point sources emission characteristics of coal industry in Shanxi Province based on Gaofen-5 satellite, J. China Coal Soc., https://doi.org/10.13225/j.cnki.jccs.2023.1247, 2024.
Kirschke, S., Bousquet, P., Ciais, P. et al. Three decades of global methane sources and sinks. Nature Geosci 6, 813–823 (2013). https://doi.org/10.1038/ngeo1955
Krummel, P. B., Montzka, S. A., Harth, C., Miller, B. R., Mühle, J., Dlugokencky, E. J., Salameh, P. K., Hall, B. D., O’Doherty, S., Steele, L. P., Dutton, G. S., Young, D., Nance, J. D., Elkins, J. W., Arnold, T., Miller, L., Fraser, P. J., Derek, N., Weiss, R. F., … Prinn, R. G. (n.d.). Overview of comparisons of non-CO2 trace gas measurements between AGAGE and NOAA at common sites.
Lalongo, I., Virta, H., Eskes, H., Hovila, J., & Douros, J. (2020). Comparison of TROPOMI/Sentinel-5 Precursor NO2 observations with ground-based measurements in Helsinki. Atmospheric Measurement Techniques, 13(1), 205–218. https://doi.org/10.5194/amt-13-205-2020
Laughner, J. L., Toon, G. C., Mendonca, J., Petri, C., Roche, S., Wunch, D., Blavier, J.-F., Griffith, D. W. T., Heikkinen, P., Keeling, R. F., Kiel, M., Kivi, R., Roehl, C. M., Stephens, B. B., Baier, B. C., Chen, H., Choi, Y., Deutscher, N. M., DiGangi, J. P., … Wennberg, P. O. (2024). The Total Carbon Column Observing Network’s GGG2020 data version. Earth System Science Data, 16(5), 2197–2260. https://doi.org/10.5194/essd-16-2197-2024
Li, X., Cohen, J. B., Qin, K., Geng, H., Wu, X., Wu, L., Yang, C., Zhang, R., and Zhang, L.: Remotely sensed and surface measurement-derived mass-conserving inversion of daily NOx emissions and inferred combustion technologies in energy-rich northern China, Atmos. Chem. Phys., 23, 8001–8019, https://doi.org/10.5194/acp-23-8001-2023, 2023.
Liu, Y., Qin, K., Cohen, J. B., Kang, H., Hu, W., Lu, F., Wu, X., and Yang, C.: Analysis of the characteristics of methane in the coal mining area of southeastern Shanxi with eddy and mobile observation, J. China Coal Soc., 47, 4395–4402, 2022.
Liu, Z., Cohen, J. B., Wang, S., Wang, X., Tiwari, P., & Qin, K. (2024). Remotely sensed BC columns over rapidly changing Western China show significant decreases in mass and inconsistent changes in number, size, and mixing properties due to policy actions. Npj Climate and Atmospheric Science, 7(1), 1–16. https://doi.org/10.1038/s41612-024-00663-9
Pei, Z., Han, G., Mao, H., Chen, C., Shi, T., Yang, K., Ma, X., & Gong, W. (2023). Improving quantification of methane point source emissions from imaging spectroscopy. Remote Sensing of Environment, 295, 113652. https://doi.org/10.1016/j.rse.2023.113652
Plant, G., Kort, E. A., Murray, L. T., Maasakkers, J. D., and Aben, I.: Evaluating urban methane emissions from space using TROPOMI methane and carbon monoxide observations, Remote Sens. Environ., 268, 112756, https://doi.org/10.1016/j.rse.2021.112756, 2022.
Prinn, R. G., Weiss, R. F., Arduini, J., Arnold, T., DeWitt, H. L., Fraser, P. J., … Zhou, L. (2018). History of Chemically and Radiatively Important Atmospheric Gases from the Advanced Global Atmospheric Gases Experiment (AGAGE). https://doi.org/10.5194/essd-2017-134
Qin, K., Hu, W., He, Q., Lu, F., and Cohen, J. B.: Individual coal mine methane emissions constrained by eddy covariance measurements: low bias and missing sources, Atmos. Chem. Phys., 24, 3009–3028, https://doi.org/10.5194/acp-24-3009-2024, 2024.
Qin, K., Lu, L., Liu, J., He, Q., Shi, J., Deng, W., Wang, S., and Cohen, J. B.: Model-free daily inversion of NOx emissions using TROPOMI (MCMFE-NOx) and its uncertainty: Declining regulated emissions and growth of new sources, Remote Sens. Environ., 295, 113720, https://doi.org/10.1016/j.rse.2023.113720, 2023.
Qu, Z., Jacob, D. J., Shen, L., Lu, X., Zhang, Y., Scarpelli, T. R., Nesser, H., Sulprizio, M. P., Maasakkers, J. D., Bloom, A. A., Worden, J. R., Parker, R. J., and Delgado, A. L.: Global distribution of methane emissions: a comparative inverse analysis of observations from the TROPOMI and GOSAT satellite instruments, Atmos. Chem. Phys., 21, 14159–14175, https://doi.org/10.5194/acp-21-14159-2021, 2021.
Rivera Martinez, R. A., Santaren, D., Laurent, O., Broquet, G., Cropley, F., Mallet, C., Ramonet, M., Shah, A., Rivier, L., Bouchet, C., Juery, C., Duclaux, O., & Ciais, P. (2023). Reconstruction of high-frequency methane atmospheric concentration peaks from measurements using metal oxide low-cost sensors. Atmospheric Measurement Techniques, 16(8), 2209–2235. https://doi.org/10.5194/amt-16-2209-2023
Tu, Q., Hase, F., Blumenstock, T., Kivi, R., Heikkinen, P., Sha, M. K., Raffalski, U., Landgraf, J., Lorente, A., Borsdorff, T., Chen, H., Dietrich, F., & Chen, J. (2020). Intercomparison of atmospheric CO2 and CH4 abundances on regional scales in boreal areas using Copernicus Atmosphere Monitoring Service (CAMS) analysis, COllaborative Carbon Column Observing Network (COCCON) spectrometers, and Sentinel-5 Precursor satellite observations. Atmospheric Measurement Techniques, 13(9), 4751–4771. https://doi.org/10.5194/amt-13-4751-2020
Verhoelst, T., Compernolle, S., Pinardi, G., Lambert, J.-C., Eskes, H. J., Eichmann, K.-U., Fjæraa, A. M., Granville, J., Niemeijer, S., Cede, A., Tiefengraber, M., Hendrick, F., Pazmiño, A., Bais, A., Bazureau, A., Boersma, K. F., Bognar, K., Dehn, A., Donner, S., … Zehner, C. (2021). Ground-based validation of the Copernicus Sentinel-5P TROPOMI NO2 measurements with the NDACC ZSL-DOAS, MAX-DOAS and Pandonia global networks. Atmospheric Measurement Techniques, 14(1), 481–510. https://doi.org/10.5194/amt-14-481-2021
Wang, S., Cohen, J. B., Deng, W., Qin, K., & Guo, J. (2021). Using a New Top‐Down Constrained Emissions Inventory to Attribute the Previously Unknown Source of Extreme Aerosol Loadings Observed Annually in the Monsoon Asia Free Troposphere. Earth’s Future, 9(7). https://doi.org/10.1029/2021EF002167
Povey, A. C. and Grainger, R. G.: Known and unknown unknowns: uncertainty estimation in satellite remote sensing, Atmos. Meas. Tech., 8, 4699–4718, https://doi.org/10.5194/amt-8-4699-2015, 2015.
Worden, J., Kulawik, S., Frankenberg, C., Payne, V., Bowman, K., Cady-Peirara, K., Wecht, K., Lee, J.-E., and Noone, D.: Profiles of CH4, HDO, H2O, and N2O with improved lower tropospheric vertical resolution from Aura TES radiances, Atmos. Meas. Tech., 5, 397–411, https://doi.org/10.5194/amt-5-397-2012, 2012.
Zhang, B., Chellman, N. J., Kaplan, J. O., Mickley, L. J., Ito, T., Wang, X., Wensman, S. M., McCrimmon, D., Steffensen, J. P., McConnell, J. R., & Liu, P. (2024). Improved biomass burning emissions from 1750 to 2010 using ice core records and inverse modeling. Nature Communications, 15(1), 3651. https://doi.org/10.1038/s41467-024-47864-7
Zhang, Y., Li, J., Li, J., Pan, X., Wang, W., Zhu, L., Wang, Z., Chen, X., Yang, W., & Wang, Z. (2022). An intercomparison of ozone taken from the Copernicus atmosphere monitoring service and the second Modern-Era retrospective analysis for research and applications over China during 2018 and 2019. Journal of Environmental Sciences, 114, 514–525. https://doi.org/10.1016/j.jes.2022.01.045
Citation: https://doi.org/10.5194/egusphere-2024-2162-RC1 -
RC2: 'Comment on egusphere-2024-2162', Anonymous Referee #1, 01 Oct 2024
Han et al. present an evaluation of various methane measurements from northern China, analyzing the spatiotemporal variability of concentration data from both measurements and models for the 2019-2021 period. In the current era, there are many different kinds of methane observations, and it is not always clear how to effectively use the combined strengths of these datasets. In that sense, this study undertakes an important task by attempting to cross-validate and reconsider the variability in different types of measurements. The region analyzed are also very important, as it includes one of the largest anthropogenic emissions hotspots.
However, the study has some major shortcomings, which is why I am recommending major revisions. The primary issue is the exclusive focus on concentration data. The ultimate goal should be to understand methane emissions (and sinks). Concentration data can be used as a proxy for emissions, but attempts must be made to convert concentrations into quantities that more accurately reflect emissions variability (spatial or temporal). Simply comparing concentration data from different sources is insufficient, especially when the data represent different types of measurements (e.g., in-situ vs. total column).
Major Comments:
1. Emission Representation from Concentration Data: I would like to see a more robust attempt to convert concentration data into a metric that represents emissions. This can be achieved in various ways, but at the very least, a local/regional background should be subtracted from the concentration data. While defining the correct background is challenging, such an attempt would be valuable and may improve the correlation values presented in the paper.
2. Model-Derived Correlations: Using models like CAMS or GEOS-Chem, it should be possible to estimate the expected correlation between different measurement types. For example, I would be interested to see what correlation would emerge from CAMS/GEOS-Chem-simulated observations between TCCON, TROPOMI, in-situ data, and emissions. This would provide a reference for interpreting the correlation analyses and contextualize the low correlation values presented in the study.
3. Inversion Emissions Estimates: It is unfortunate that the paper offers only a very limited analysis of inventory and inversion emissions estimates. These estimates are among the closest to true emissions of all the datasets presented, and I believe there should be a more substantial investigation of the inventory and GEOS-Chem posterior emissions.
4. Column vs. In-situ Concentration Observations: The authors need to first clarify why these different measurements can be directly compared. Column and in-situ observations represent different quantities, and in-situ data, for instance, is strongly influenced by boundary layer height variability.
5. Line 312: A strong correlation is not expected as TROPOMI averages over a larger horizontal spatial area, which is further diluted by column averaging. However, TROPOMI also senses local/regional emissions. A deeper investigation into the very poor correlation needs to be conducted. I wonder if the correlation improves when different spatial and/or temporal averaging methods are applied. Also, how much of the variability of TROPOMI data is dictated by data coverage gaps needs to be discussed.
Minor Comments:
• Figures: Panels within figures need to be consolidated. For example, panels in Figure 6 are on separate pages, and the same issue occurs with Figure 9.
• Line 94: “due mainly to” should be revised to “mainly due to.”
• Line 100: “measurements of trends” should be revised to “measurements.”
• Line 130: “quantify the spatiotemporal CH4 concentrations” should be revised to “quantify the spatiotemporal variations of CH4 concentrations.”
• Line 245: This sentence is overly complex. Consider rephrasing for clarity.
• Line 264: “To determine the importance of understanding the uncertainties in concentrations…” should be revised to “To understand the errors in concentrations…”
• Figure 9 (Annual Panel): Correct “XH4” to “XCH4” in the figure legends.
• Figure Captions: More detailed information is needed. For instance, the caption for Figure 1 is too brief considering the amount of data it presents.
Citation: https://doi.org/10.5194/egusphere-2024-2162-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 245 | 42 | 102 | 389 | 26 | 2 | 4 |
- HTML: 245
- PDF: 42
- XML: 102
- Total: 389
- Supplement: 26
- BibTeX: 2
- EndNote: 4
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1