the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 15 Sep 2023
A bottom-up emission estimate for the 2022 Nord-Stream gas leak: derivation, simulations and evaluation
Abstract. A major release of methane from the Nord Stream pipelines occurred in the Baltic sea on 26 September 2022. Elevated levels of methane were recorded at many observational sites in northern Europe. While it is relatively straightforward to estimate the total emitted amount from the incidents (around 330 kt of methane), the detailed vertical and temporal distributions of the releases are needed for numerical simulations of the incident. Based on information from public media and basic physical concepts, we reconstructed vertical profiles and temporal evolution of the methane releases from the broken pipes, and simulated subsequent transport of the released methane in the atmosphere. Since we used pure-methane assumption, the inventory total amounts to 290 kt of methane.
The emission rates were calculated with a numerical solution of a problem of a gas leak from a half-opened pressurized pipe. Initial vertical distribution of the released gas was derived from a parametrization for an injection height of buoyant plumes, and validated with a set of large-eddy simulations by means of UCLALES model.
The estimated emission source was used to simulate the dispersion of the gas plume with the SILAM chemistry transport model. The simulated fields of the excess methane led to noticeable increase of concentrations at several carbon-monitoring stations in the Baltic Sea region. Comparison of the simulated and observed time series indicated an agreement within a couple of hours between timing of the plume arrival/departure at the stations with observed methane peaks. Comparison of absolute levels was quite uncertain. At most of the stations the magnitude of the observed and modelled peaks was comparable with natural variability of methane concentrations. The magnitude of peaks at a few stations close to the release was well above natural variability, however the magnitude of the peaks was very sensitive to minor uncertainties in the emission vertical profile and in the meteorology used to drive SILAM.
The obtained emission inventory and the simulation results can be used for further analysis of the incident and its climate impact. They can also be used as a test case for atmospheric dispersion models.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(5660 KB)
-
Supplement
(3066 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(5660 KB) - Metadata XML
-
Supplement
(3066 KB) - BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-732', Anonymous Referee #1, 07 Oct 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-732/egusphere-2023-732-RC1-supplement.pdf
-
AC2: 'Reply on RC1', Rostislav Kouznetsov, 29 Nov 2023
We would like to thank the reviewer for their interest to our paper and for the valuable comments they gave.
The only major thing requested by both reviewers is the quantitative evaluation of the simulations.
As we stated in the response to RC2, standard metrics used for model-measurement comparison
are suboptimal for the case. We have performed evaluation using more robust metrics, and
a section on quantitative evaluation of the model performance will be added to the final manuscript.
We will also account for all the minor points that have been raised raised by the reviewer.Citation: https://doi.org/10.5194/egusphere-2023-732-AC2
-
AC2: 'Reply on RC1', Rostislav Kouznetsov, 29 Nov 2023
-
RC2: 'Comment on egusphere-2023-732', Anonymous Referee #2, 10 Oct 2023
-
AC1: 'Reply on RC2', Rostislav Kouznetsov, 28 Nov 2023
We would like to thank the reviewer for their interest to our paper and for the valuable comments they gave.
The reviewer's comments largely repeat the comments given on a preprint submission stage, and we believe most of them have been mostly addressed already in the published discussion paper. We will add the complete updated response together with the revised paper.
The updated comment has two more points:
>>The conclusions are vague, not providing the values obtained when applying this methodology and how much it deviates from the observations used to validate this study.The problem of the evaluation for the performed simulations is that one has to compare observed methane to the excess methane forecasted by the model. In case of a large excess it is relatively straightforward to select a background value, and compare excesses. These cases constitute a small fraction of the observations, so fitting these data can be achieved with rather unrealistic emission profiles (such as one given by Jia et al., 2022) study. Moreover, due to the small spatial extent of the source, small inaccuracies in the dispersion model and/or driving meteorology can lead to significant discrepancies between model and observations even for perfectly accurate emission profiles. Therefore larger dataset is needed for a robust model evaluation.
For most of the stations the excess is of the same order as the natural variability. For such cases subtraction of background becomes a poorly defined procedure.
To address this difficulty, in the published preprint we showed plots where the modelled methane excesses were plotted in the same scale as observed methane concentrations. The plots qualitatively show that the emission with the dynamic injection height improves the modelling results compared to a fixed injection profile.
For the revised version we involved several metrics that allow for variable background to provide a quantitative measure of the simulation quality at different assumptions of the injection height.
>>Why 3x95,000 tons and not 285 000 or 285kT?
The idea was to point that in our vision there were three identical pipes with equal amount of gas in them. We'll put total 285kT figure in the final paper.Citation: https://doi.org/10.5194/egusphere-2023-732-AC1
-
AC1: 'Reply on RC2', Rostislav Kouznetsov, 28 Nov 2023
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-732', Anonymous Referee #1, 07 Oct 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-732/egusphere-2023-732-RC1-supplement.pdf
-
AC2: 'Reply on RC1', Rostislav Kouznetsov, 29 Nov 2023
We would like to thank the reviewer for their interest to our paper and for the valuable comments they gave.
The only major thing requested by both reviewers is the quantitative evaluation of the simulations.
As we stated in the response to RC2, standard metrics used for model-measurement comparison
are suboptimal for the case. We have performed evaluation using more robust metrics, and
a section on quantitative evaluation of the model performance will be added to the final manuscript.
We will also account for all the minor points that have been raised raised by the reviewer.Citation: https://doi.org/10.5194/egusphere-2023-732-AC2
-
AC2: 'Reply on RC1', Rostislav Kouznetsov, 29 Nov 2023
-
RC2: 'Comment on egusphere-2023-732', Anonymous Referee #2, 10 Oct 2023
-
AC1: 'Reply on RC2', Rostislav Kouznetsov, 28 Nov 2023
We would like to thank the reviewer for their interest to our paper and for the valuable comments they gave.
The reviewer's comments largely repeat the comments given on a preprint submission stage, and we believe most of them have been mostly addressed already in the published discussion paper. We will add the complete updated response together with the revised paper.
The updated comment has two more points:
>>The conclusions are vague, not providing the values obtained when applying this methodology and how much it deviates from the observations used to validate this study.The problem of the evaluation for the performed simulations is that one has to compare observed methane to the excess methane forecasted by the model. In case of a large excess it is relatively straightforward to select a background value, and compare excesses. These cases constitute a small fraction of the observations, so fitting these data can be achieved with rather unrealistic emission profiles (such as one given by Jia et al., 2022) study. Moreover, due to the small spatial extent of the source, small inaccuracies in the dispersion model and/or driving meteorology can lead to significant discrepancies between model and observations even for perfectly accurate emission profiles. Therefore larger dataset is needed for a robust model evaluation.
For most of the stations the excess is of the same order as the natural variability. For such cases subtraction of background becomes a poorly defined procedure.
To address this difficulty, in the published preprint we showed plots where the modelled methane excesses were plotted in the same scale as observed methane concentrations. The plots qualitatively show that the emission with the dynamic injection height improves the modelling results compared to a fixed injection profile.
For the revised version we involved several metrics that allow for variable background to provide a quantitative measure of the simulation quality at different assumptions of the injection height.
>>Why 3x95,000 tons and not 285 000 or 285kT?
The idea was to point that in our vision there were three identical pipes with equal amount of gas in them. We'll put total 285kT figure in the final paper.Citation: https://doi.org/10.5194/egusphere-2023-732-AC1
-
AC1: 'Reply on RC2', Rostislav Kouznetsov, 28 Nov 2023
Peer review completion
Journal article(s) based on this preprint
Model code and software
SILAM v5_8_2 Silam team https://doi.org/10.5281/zenodo.7598284
Video supplement
Methane dispersion form the Nord Stream gas leaks Rostislav Kouznetsov and Evgeny Kadantsev https://doi.org/10.5446/1770
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 471 | 214 | 21 | 706 | 37 | 14 | 14 |
- HTML: 471
- PDF: 214
- XML: 21
- Total: 706
- Supplement: 37
- BibTeX: 14
- EndNote: 14
Viewed (geographical distribution)
| Country | # | Views | % |
|---|---|---|---|
| United States of America | 1 | 179 | 25 |
| Finland | 2 | 98 | 14 |
| Germany | 3 | 53 | 7 |
| Netherlands | 4 | 51 | 7 |
| Estonia | 5 | 38 | 5 |
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
- 179
Rostislav Kouznetsov
Risto Hänninen
Andreas Uppstu
Evgeny Kadantsev
Yalda Fatahi
Marje Prank
Dmitrii Kouznetsov
Steffen Manfred Noe
Heikki Junninen
Mikhail Sofiev
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(5660 KB) - Metadata XML
-
Supplement
(3066 KB) - BibTeX
- EndNote
- Final revised paper