the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Improving Consistency in Methane Emission Quantification from the Natural Gas Distribution System across Measurement Devices
Abstract. Mobile real-time measurements of ambient methane provide a fast and effective method to identify and quantify methane leaks from local gas distribution systems in urban areas. The objectives of these methodologies are to i) identify leak locations for repair and ii) construct measurement-based emission rate estimates, which can improve emissions reporting and contribute to monitoring emission changes over time. Currently, the most common method for emission quantification uses the maximum methane enhancement detected while crossing a methane plume. However, the recorded maximum depends on instrument characteristics, such as measurement cell size, pump speed and measurement frequency. Consequently, the current approach can only be used by instruments with similar characteristics. We suggest that the integrated spatial peak area is a more suitable quantity that can eliminate the bias between different instruments. Based on controlled release experiments conducted in four cities (London, Toronto, Rotterdam, and Utrecht), emission estimation methodologies were evaluated. Indeed, the integrated spatial peak area was found to be a more robust metric across different methane gas analyzer devices than the maximum methane enhancement. A statistical function based on integrated spatial peak area is proposed for more consistent emission estimations when using different instruments. On top of this systematic relation between actual emission rate and recorded spatial peak area, large variations in methane spatial peak area were observed for the multiple transects across the same release point, in line with previous experiments. This variability is the main contributor of uncertainty in efforts to use mobile measurements to prioritize leak repair. We show that repeated transects can reduce this uncertainty and improve the categorization into different leak categories. We recommend a minimum of three and an optimal range of 5-7 plume transects for effective emission quantification to prioritize repair actions.
- Preprint
(1036 KB) - Metadata XML
-
Supplement
(65616 KB) - BibTeX
- EndNote
Status: open (until 21 Feb 2025)
-
RC1: 'Comment on egusphere-2024-3620', Anonymous Referee #1, 06 Feb 2025
reply
The article Improving Consistency in Methane Emission Quantification from the Natural Gas Distribution System across Measurement Devices uses several controlled release experiments to devise a better way of inferring fluxes from measurements in the case of gas leaks from the natural distribution system. By analyzing all the data, they show that using the area instead of the maximal amplitude of the signal allows a better coherence especially when using different measurement technologies. They also confirm the importance of repeating transects to reduce errors.
The article is generally well written, and the authors spend time examining the weaknesses of their method (testing the linearity hypotheses, discussing how precise is the method...).
I recommend publication after a few minor modifications.
I would suggest to push the appendix in the Supplementary, it is a bit confusing to have both. It would also be nice to harmonize a bit the description of the measurement procedures in the Supplementary material.
In the supplementary material: l229 and 235, the Figure seems to be out of place or missing comas, please check.
In the main text, I would make clear that all data have been used to derive the equation and then the equation has been used to derive the fluxes for all the measurements. There is no training data (at least that I what I understood).
l37 replace plattform by platform
l41 replace are by being
l64 (after "environment") and 66 (after now) and at other places in the text, additional comas would be welcome to facilitate the reading of long sentences, please check
l234 add an in "as an empirical model"
l376 replace But by However
Citation: https://doi.org/10.5194/egusphere-2024-3620-RC1 -
RC2: 'Comment on egusphere-2024-3620', Anonymous Referee #2, 06 Feb 2025
reply
General comments:
This manuscript describes how the quantification methods for methane emissions from natural gas distribution systems (leaks) can be improved and give practical recommendations. The strength of the manuscript is the extended study, which incorporates measurements from a number of different methane instruments placed at different measurements sites (three in Central Europe, one in Canada) carried out for a duration of several days at each site by different research groups. Various statistical methods were applied to analyse the available data set as a whole. The outcome of the study - integrated spatial peak area is more robust to use than the maximum methane enhancement - agrees well with previous findings by Maazallahi et al. (2020). However, it would be important to also highlight more in detail what is new and different in the present study compared to e.g. the Maazallahi et al. study.
The manuscript is well-written and generally logical structured with some exceptions. The number of figures and tables in comparison to the text is appropriate. In general, the figures are of good quality. Sometimes the labelling is missing or could be improved. Proper credit is mostly given to related work.
Details on suggestions for improvements on the above mentioned topics are given in the specific comments below.
The paper addresses scientific questions relevant to the scope of AMT.
For the reasons mentioned above and below the paper is appropriate for publication in AMT after a minor revision described below.
Specific comments:
Are any wind measurements available at least for some of the sites or re-analyses data (London)? Incorporation of such data would be interesting and important to evaluate the dependency of the results on the current wind situation. If the measurements were not performed perpendicular to the wind direction (as mentioned in the manuscript for London), a major part of the plume can be missed. Please discuss this in more detail.
Information on the atmospheric stability is not available from the present data set, however it would be interesting to compare the influence of the timing of the measurements on the results. Recommend to divide all flights into morning / noon / late afternoon flights to find out at what time of the day the best measurements can be achieved. At noon it can happen that the main plume rises straight upward and only a minor part is measured. In the morning / late afternoon accumulation of emissions is possible. Please comment more on this.
Page 16, Sect. 3.5: very important subsection! Improve by adding recommendations on wind direction measurements (see next sentence) and by adding the results from your missing study (my recommendation, see above) on the optimal probing time of the day (in case any useful results come out of that study). Add that the wind direction can be measured by an anemometer on the roof of the car or by a simple flag (tissue) reached out of the car window. Such simple measurements can improve the methane quantifications considerable.
Minor comments and technical corrections:
General minor comment: “London” and “London I” (same meaning) are mixed in the manuscript (same for “Utrecht” and “Utrecht I”). Homogenize throughout the manuscript.
Page 1, line 10-12: Is this true for all sites? In Fig. 1 it is only shown for Rotterdam and Utrecht (does it include all data from those sites?). Toronto is not shown and London (all data?) is presented in Fig. B1.
Page 2, line 24: re-order given references chronologically (and also alphabetically if same year, check throughout manuscript)
Page 2, line 34: “present” à represent
Page 3, line 64: “environments complex” à environments with complex
Page 3, line 72-73: Maazallahi (2020) gave the same recommendation as in your study. What have you achieved in addition to the mentioned study?
Page 3, line 81: After the headline “2 Methods”, briefly explain what subsections the reader can expect next.
Page 4, line 85: “SI” à Supplement Information (SI)
Page 4, line 85: “Table A1” à Table A1 in the Appendix A
Page 4, Table 1 legend: “Controlled release experiment” à Controlled release experiment (CRE)
Page 4, Table 1: “Inlet Level” à Inlet Height
Page 4, Table 1: Was the release height always the same (height of gas bottle?)? Add to legend text or somewhere else.
Page 4, Table 1: For London and Utrecht, add I and II (second already incorporated).
Page 4, line 88: “1a” à 2
Page 4, line 95: “2a” à 2
Page 4, line 98: write out TNO the first time
Page 5, line 103: “1b” à 3
Page 5, line 107: write out IMAU the first time
Page 5, line 108-117: Give here also the release rates for London and Toronto (as above for Utrecht & Rotterdam).
Page 5: line 110, line 111, line 116: Not necessary to mention companies, since listed in Table 1.
Page 5, line 132: “vehicle the” à vehicle, the
Page 5, line 133: “dataset peaks” à “dataset, peaks”
Page 6, line 136: “UU car” à UUAQ
Page 8: Remove the dot after some of the equations.
Page 9, After the headline “3 Results and Discussion”, briefly explain what subsections the reader can expect next.
Page 9, Fig. 1: Avoid placing a figure at the beginning of a subsection.
Page 9, Fig.1: Are all data from Rotterdam and Utrecht included? Same in Fig. B1 for London (it might also be an option to add this figure to the main manuscript since it is illustrative)? Why are not the data from Toronto shown?
Page 9, line 206-207: Discuss why the peak maximum measured by G4302 is clearly underestimated by the other instruments.
Page 10, line 209: “0.63” à not in agreement with values given in Fig. 1
Page 10, line 210: “most” à all
Page 10, line 216: “Figure B1” à Figure B1 in Appendix B
Page 10, Fig. 2: add (a) and (b) to legend text.
Page 11, line 225: At several places comparisons to Maazallahi / Gillespie are made. Discuss more details from these papers in your manuscript and point out what is different/new in your study.
Page 11, line 228: Are there any plans in future to install such monitoring instruments downwind of the main wind direction or are they always randomly installed? Or is it worth adding such recommendations in your conclusions (Sect. 3.5)?
Page 11, line 235-236: Add “[CH4] area” and “re” to the sentence.
Page 12, Fig. 3: What is the black line? Why is it missing in (b)?
Page 12, Fig. 3: “London Day 2” à London I Day 2
Page 12, Fig. 3: In figure legend replace London by London I and Utrecht by Utrecht I.
Page 13, Fig. 4: Add (a) and (b) to the top of the figures instead of repeating it below the figures. Write out the different categories directly into the figure.
Page 13, Fig. 4 legend: “true emission rate rE” à true emission rate rR
Page 13, Fig. 4: “pool which” à pool in (%) for which
Page 13, Fig. 4: to be better comparable to Fig. 3, switch the order of (a) and (b). Add after (a) “Based on Peak Maximum” (as in Fig. 3) and after (b) “Based on Spatial Peak Area”.
Page 13, legend: write out “CRE”
Page 13, line 274: but also underestimated
Page 13, line 275: “22%” à 21%
Page 14, line 276-277: Check the numbers given in the text with the numbers in Fig 4.
Page 14, line 279: Add “(Fig. 4b)” to the sentence.
Page 14, Fig. 5: Avoid placing a figure at the beginning of a subsection. Don´t repeat headlines above and below the figures.
Page 14, line 282: “London” à London I
Page 14, line 285: “considered an” à considered as an
Page 15, line 299: “Figure 5a” à Figure 5
Page 15, line 299: Briefly describe what is shown in Fig. 5a compared to Fig. 5b.
Page 15, line 319: “to the mean” à to the mean (Fig. 5a)
Page 15, line 321: “for N=2” à for N=2 (Fig. 5b)
Page 16: Fig. 6: label the y-axes left and right
Page 16, line 334: “account three” à account one (Fig. 6a), three
Page 16, line 337: Move this sentence to follow-up directly after line 336.
Page 16, line 337: “on a single detection” à on one single transect (N=1),
Page 16, line 338: “are used” à are used (N=3)
Page 16, line 339: “six transects” à six transects (N=6)
Page 16, Line 342: “measurements” à transects
Page 17, line 349: “six transects” à six transects downwind of the leak.
Page 17, line 350: “subtract it” à subtract them
Page 17, line 358: “inverse Area eq.” à inverse Area eq. (Eq.13)
Page 17, line 371: “emission rate” à emission rate (rE)
Page 17, line 372: “enhancement” à enhancement ([CH4]area)
Page 18, line 381: “Including…” à Including three transects instead of one improved correct…..
Page 18, line 382: “and to 62-82%...“ à and to 62-82% when incorporating six transects instead of one transect
Page 19, line 388: „London“ à London I
Page 19, line 389: „Utrecht“ à Utrecht I
Page 19, Table A1: Three times „London“ à London I
Page 20: „Utrecht“ à Utrecht I
Page 20: Has the impact of the different timing been investigated (Toronto)?
Page 22, Fig. B1 legend text: “London” à London I and give the day(s)
Page 24-27: Add more details to some of the references (e.g. ISBN number, http://) given in the following lines: 411, 423, 424, 426, 458, 488, 506
Page 26, line 495: improve spelling mistake
Supplement
Specific comments:
Does the main manuscript contain references to everything shown in the supplement? Either refer to the supplement material in the main document or leave it out completely.
Page 6, line 30: Why was the release done from the surface level here? Normally I assume the release height was the height of the cylinder? Please add the release height at each site into one of the available tables (or state that it is constant). How do these release heights agree with heights of detected leaks in the field?
Page 6, line 51: In general, for all sites with car probing: When the cars with the instruments were driving along the streets, where there also other normal cars using these streets at the same time? If yes, this can also distort the measurements. Please comment. For the routine leak detection surveys described in the main Sect. 3.5, you can also add that the measurements should be carried out with a sufficient distance to previously passed cars.
Page 28: Here you present the Figs. S23-S24 and S25-S26. Also add a few sentences on the interpretation.
Minor comments and technical corrections:
Page 1, line 8: „comparison“ à Comparison
Page 2, line 20: “equation” à equations (Eqs. 1-2)
Page 2: Refer in the text to Table S1
Page 3, 5th column: „Cell V“ à Cell Vnorm
Page 3, Table S1: add precision of all instruments
Page 4: „Utrecht“ à Utrecht I
Page 4: „London“ à London I
Page 5-7: Recommend to replace all text by a table.
Page 9, Fig. S5: Is the day shown in Fig. S5 the same as in Fig. S4? Are the measurements shown for G4302 in both figures the same? The first impression is that it isn´t?
Page 9, Fig. S6: “Utrecht” à Utrecht I
Page 10, Fig. S7: Just left of 14:10, the straight increase is most likely due to missing values. Don´t connect such time periods with a line.
Page 10, Fig. S8: Around 13:00 the red curve shows a strange behaviour with a plateau.
Page 10, Fig. S8: “London” à London I
Page 12, Fig S10(d): Replace below and in figure “Utrecht” à Utrecht I
Page 13, line 112: Refer to Figs. S11-14 in the text.
Page 13, line 113: Specify more in detail what you mean with “noise” and give examples.
Page 13, Fig. S11 legend: “London” à London I
Page 14, Fig. S12: all legends in figure and text “Utrecht” à Utrecht I, same for London in Fig. S13
Page 15, Fig. S14: left figure (a) has the headline Toronto Day 2, in the legend text a different order is given
Page 16, line 156: “rtaes” à rates
Page 17, line 170-171: Does this agree to the experience from other similar studies (give references)?
Page 18, Table S3: “Utrecht” à Utrecht I
Page 19, Table S5: “London à London I
Page 20, line 192: “Utrecht” à Utrecht I
Page 20, line 199: Add that this is related to London I.
Page 20, line 204: Add that this is related to London I.
Page 21-25, Fig. S16-S20: For a better overview, give the name of the release (e.g. Rotterdam) once for every page as a headline of all sub-figures.
Page 22, Fig. S17: “Utrecht”à Utrecht I
Page 24, Fig. S19: “London” à London I
Page 26, line 217: “Figure S22 and Figure S21” à Figs. S21-S22
Page 26, line 217: After referring to the figures, also add few sentences on the interpretation.
Page 26, Fig. S21: labelling of y-axes is missing
Page 28, line 228-229: mention Fig. S23b first at the end of the sentence
Page 28, line 234-235: mention Fig. S24b first at the end of the sentence
Page 34, line 256: Add more details to reference (e.g. ISBN number)
Page 34, line 268: Add more details to reference (e.g. ISBN number)
Citation: https://doi.org/10.5194/egusphere-2024-3620-RC2
Viewed
HTML | XML | Total | Supplement | BibTeX | EndNote | |
---|---|---|---|---|---|---|
219 | 22 | 4 | 245 | 25 | 1 | 1 |
- HTML: 219
- PDF: 22
- XML: 4
- Total: 245
- Supplement: 25
- BibTeX: 1
- EndNote: 1
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1