the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 20 Jun 2024
Seasonal and annual variability of methane emissions to the atmosphere from the surface of a eutrophic lake located in the temperate zone (Lake Kortowskie, Poland)
Abstract. Despite studies on methane emissions from lakes, there remains considerable uncertainty in accurately estimating global emissions from this source. This uncertainty is related to the diversity of lake types, their conditions, geographical locations, as well as the various research methods employed, typically short measurement series, temporal variability of emissions, and the various forms of emissions: diffusion, ebullition, transport in macrophytes, and storage emission.
In this study, an attempt was made to supplement information on methane emissions through real-time in situ measurements using a measurement chamber connected to a mobile CRDS spectrometer. Measurements were conducted on Lake Kortowskie, which is representative of highly eutrophic lakes in the northeastern region of Poland. The methane emission measurement cycle was carried out over a full four-year period (2019–2022) at weekly intervals, alongside simultaneous observations of water indicators and meteorological measurements.
The average methane emission from the surface of Lake Kortowskie over the entire observation period was 11.79 mg m-2 d-1, with a median of 6.91 mg m-2 d-1, and a maximum of 134.4 mg m-2 d-1 on a single measurement date. During the four-year observation period, slight differences in annual averages were noted, along with significant variability in seasonal emissions. In the years 2019, 2020, 2021, and 2022, the average CH4 emissions were 13.7, 10.1, 11.8, and 11.6 mg m-2 d-1, respectively. Seasonally, average emissions were recorded at 3.2, 12.1, 20.6, and 14.9 mg m-2 d-1 for winter, spring, summer, and autumn, respectively.
The studies indicated that the main environmental factors associated with methane emissions from the lake were primarily water temperature and air temperature. However, water waves height, wind speed and gusts, precipitation totals, Secchi depth, and oxygen concentration in the water also played significant roles. Regression analyses for Lake Kortowskie suggest that only changes in the main climate components, following the current trend of changes, could lead to an increase in methane emissions from the lake by over 30 % by the year 2100.
- Preprint
(1279 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2024-1786', Anonymous Referee #1, 25 Jul 2024
general remarks
This paper presents weekly (!) floating chamber measurements in a eutrophic lake over a period of 4 years – a quite impressive dataset. The author uses this data to calculate monthly and annual CH4 emission from the lake. Using linear correlation analysis he develops an empirical model to predict CH4 emissions from climate variables with a rather high power. That model is then used to predict future CH4 emissions based on extrapolated historical climate data.
Although the topic of aquatic CH4 emissions is subject to intensive research in the last years and decades it is still an actual topic with many unsolved questions. The rather high temporal resolution of this study offers a good potential to contribute to our knowledge about CH4 dynamics in lakes. Unfortunately the manuscript does not really use that chance. There are a number of serious issues which need to be addressed before the paper can be accepted.
The paper lacks a clear research question or hypothesis. Just saying “We want to quantify CH4 emissions from a particular lake” is not sufficient for an international journal.
It is not very clear what new knowledge is provided. It is well known that temperature is a major regulator of CH4 production in lakes and thus, that CH4 emissions will increase with warming. There are numerous studies quantifying CH4 emissions from particular lakes. The author need to make clear what makes his study outstanding. I thought a bit about this. In my eyes the strong points in this paper are:
- Weekly resolution
- 4 years covered
- Special lake with partly anoxic epilimnion
- Wave height measured
The author does not exploit these points. Instead of analysing weekly temporal dynamics he even aggregated the data to monthly resolution.
Your annual budgets are strikingly similar. Does that mean that short term fluctuations are not relevant on the annual scale (Morales-Pineda et al., 2014)?
Weak parts of the paper are the treatment of ebullition and the lack of interesting additional data like weekly probe profiles or CH4 concentration in the water. Although the author observed little evidence for ebullition in his measurements I am not convinced that ebullition can be ruled out as a process in this rather shallow lake. A considerable part of the discussion is rather hypothetically without showing more data from the water column. He writes that he did probe measurements along with the chamber measurements. These data need to be shown! I recommend contour plots (heat maps) of the relevant parameters.
The reference list needs to be updated. The literature on the topic is highly dynamic and there are many more recent papers on the topic.
detailed remarks
l.9: I would replace “Despite studies on methane emissions “ by a statement underlining the relevance of lake CH4 emissions.
l.17: Here a clear research question or hypothesis is missing
l.25: I guess water and air temperature were highly correlated. Maybe just write “temperature” without differentiating air and water.
l.32: there should be a more recent reference than Dlugokencky. I would also remove Oh et al here.
l.47: There are more recent references on global CH4 emissions from Lakes. For example (Rosentreter et al., 2021; Delsontro et al., 2018).
l.50: There is more recent literature than Bastviken 2004.
l.55: Remove “current”. You may also refer to thermal convection here.
l.60: Refer to Rosentreter et al.
l.67: “to attempt to assess” is not a good formulation here. Write which questions you want to answer.
l.71: Maybe replace CRDS by “a portable GHG analyser”. Replace “flow chamber” by “floating chamber”.
l.73: “… where to my knowledge CH4 emissions from lakes were never published before”
l.78: Are these coordinates refer to the lake (then write so) or the site of measurement. Use past tense
l.90: I wonder if the lake treatment can be named “experiment”
l.92: One profile is not enough to prove this statement.
l.93-94: Why is this information relevant for the paper?
l.111: use past tense.
Table 1: May be move to the supplement. Any rational why is humidity reported?
Figure 2: Move to the supplement and add wind speed data.
l.136: I wonder whether there was a diurnal pattern of wind speed. Since all flux measurements were done during the day this could lead to an over-estimation of daily CH4 emissions. Wind speed is often lower in the night.
l.146: Is the equation correct? I cannot reproduce the unit mg/m2 d from it. Don’t you need atmospheric pressure to convert ppm to mole? Please check the equation. You may also have a look at Unesco/Iha (2010).
l.160 This exclusion of 2 of 5 replicates is very unusual. I am sure such a procedure reduces data variability. But you cannot remove data just because they differ a bit.
l.167: I know that fluxes can vary on very short timescales depending on wind fluctuations. That’s why I usually measure wind during each chamber measurement separately.
l-170: You need to explain how you scaled up your measurements at the 5 sites to the entire lake! You need to do this area weighted: The outmost point should be representative for a large part of the lake surface.
l.176: I am not convinced.
l.183: Show the probe data.
Table 2: why ius ther no average pH and two pH values in the other columns. Ranges should be indicated by “-“, not by “+”.
l.217: “between individual years”
l.225: replace “result” by “meand flux”
l.233: Lake area should not be problem if area specific fluxes are reported.
l.235: There should be more recent references – especially on eutrophic shallow lakes.
l.244: Remove “taken”
l.245: Remove “equilibrium state with methane present in the”
l.247: Would be interesting to see a plot of CH4 flux versus wind and/or wave height. The wave height measurements are interesting. Would also be interesting how wave height depended on wind speed.
l.255: The treatment of ice cover is not correct. During ice cover emissions are zero. If CH4 accumulates under the ice there should be higher emissions after ice off. Since you measured weekly you should have flux data very shortly after ice off.
l.269: You say that weather affected CH4 emission. Can you be a bit more specific here saying which weather had which effect?
l.284: Add “or emissions after ice off”.
Figure 4: The range is rather high in some months and hardly to explain by differences in wind speed. I would be suspicious that the very high flux measurements were affected by ebullition. If you have continuous ebullition of very small bubbles you do not see jumps in your chamber data and would interpret it as diffusion.
Figure 4: The boxes cover both between site and between week variability. I recommend that you present the data with weekly resolution. Than the variability can be interpreted as spatial variability and you can look at sub-monthly temporal pattern.
l.313: You can calculate activation energy and Q10 of your temperature dependence of the CH4 flux. This allows the quantitative comparison with temperature dependence in the literature.
l.327: This is very new and interesting. Exploit more. Maybe have a look at marine literature, where the effect of waves on gas exchange have been studied intensively.
Figure 5: Why do we need Figure 5b?
l.364: I do not really see that.
l.370: This can be checked by looking at weather station data.
Section 3.3: it is clear that air temperature and water temperature should be correlated and that water temperature has an effect on methanogenesis in the sediment – but air temperature not. Why did you correlate air temperature here?
Maybe a better way to come up with a statistical model is runing mixed models using R and the use of something like the AIC to compare the performance of different models.l.455: You discuss eutrophication here but your data to not show a correlation with chlorophyll. Does that mean eutrophication is not in important regulator in your lake?
l.457: This is a strong statement. Can you support this by quantitative arguments?
l.471: “Methane emissions from lake Kortowskie exhibited …“
l.471-474: this is not new
l.477: Also the observed seasonality is not new.
l.480-481: This is new.
References
DelSontro, T., Beaulieu, J. J., and Downing, J. A.: Greenhouse gas emissions from lakes and impoundments: Upscaling in the face of global change, Limnology and Oceanography Letters, 3, 64-75, doi:10.1002/lol2.10073, 2018.
Morales-Pineda, M., Cózar, A., Laiz, I., Úbeda, B., and Gálvez, J. A.: Daily, biweekly, and seasonal temporal scales of pCO2 variability in two stratified Mediterranean reservoirs, J Geophys Res-Biogeo, 119, 509-520, Doi 10.1002/2013jg002317, 2014.
Rosentreter, J. A., Borges, A. V., Deemer, B. R., Holgerson, M. A., Liu, S., Song, C., Melack, J., Raymond, P. A., Duarte, C. M., Allen, G. H., Olefeldt, D., Poulter, B., Battin, T. I., and Eyre, B. D.: Half of global methane emissions come from highly variable aquatic ecosystem sources, Nature Geoscience, 14, 225-230, 10.1038/s41561-021-00715-2, 2021.
UNESCO/IHA, (IHA), T. I. H. A. (Ed.): GHG Measurement Guidelines for Freshwater Reservoirs, UNESCO, 138 pp., https://www.hydropower.org/publications/ghg-measurement-guidelines-for-freshwater-reservoirs, 2010.
Citation: https://doi.org/10.5194/egusphere-2024-1786-RC1 - AC1: 'Reply on RC1', Andrzej Skwierawski, 13 Nov 2024
-
RC2: 'Comment on egusphere-2024-1786', Anonymous Referee #2, 25 Oct 2024
The paper “Seasonal and annual variability of methane emissions to the atmosphere from the surface of a eutrophic lake located in the temperate zone (Lake Kortowskie, Poland)” by Skwierawski is looking into the methane emissions from Lake Kortowskie for four years at five different stations, measuring emissions almost weekly. Skwierawski analyses the methane emissions in relation to a variety of environmental factors on all the data (n = 198) and on monthly average emissions using Spearman correlation, principal component analysis and linear modeling. Lastly the author uses the measured values to predict methane emissions from the lake in 2050 and 2100.
The author has collected an interesting dataset with a long time series of data. I am concerned about the placement of the chambers, as all chambers are placed within 1.5–3.5 meters of water depth, which leads the author to conclude negligible amounts of methane ebullition take place. Nonetheless, many papers are starting to conclude that ebullition occurs in the deeper parts of the lake and accounts for large fractions of the total methane emission with increasing emissions. Moreover, the analysis conducted in the paper needs a thorough check, as simple things such as intercorrelation were never considered. Additionally, I find that the extrapolation of methane emissions to 2050 and 2100 should be associated with much uncertainty. It has become a discipline in the papers on methane emissions to try and extrapolate further and wider, more often than not too much. The manuscript is also in need of more references on some of the important statements.
Overall, I find that the paper has a good-fair scientific significance due to the large amount of data collected. However, the scientific quality of the paper is fair. The statistics are missing from the manuscript, making it hard to conclude on the significance statements. Furthermore, I cannot find any information on how the data has been handled beforehand (scaled, intercorrelation, etc.). The presentation quality of the paper is fair; the language is easily understood, yet on several occasions, the reason for using analysis or the results is not revealed until later.
In general
Add statistical information when referring to correlations.
Title should reflect only diffusive emissions are considered.
Abstract
L 14. Please refrain from using abbreviations in the abstract.
L 24. “The studies” should be changed to “The results”
Introduction
L36. Needs a reference
L37. Needs a reference
L45-46. I see your point, but it’s counterintuitive to first state that lakes play a significant role and then state that the magnitude is difficult to estimate.
L70. Belongs in the method.
Figure 1. I don’t understand the partitioning into partial catchments.
Methods
L130. What qualifies as a faulty measurement?
L132. The representativeness, please spell out what the measurements are representative off.
L133. Please indicate the method used to assess the correlation and statistical information (df, F-value, etc.).
L145. Why do you only use C0 and C180, when you have 216 measurements (72*3) of methane increase? You could use the linear increase.
L160-167. I am not convinced that this is a good method. Methane emissions are variable in space and time, which you concluded in the beginning. By removing these observations, you remove some of the noise, but the high emissions may be due to ebullitive emissions and thus very much as relevant as the low fluxes.
L167-170. I believe that the sensor measures the changes in temperature and humidity within the chamber, so it’s possible to determine the change in environmental conditions.
L173. I would like you to point out in the title that you are only measuring diffusive methane emissions and not ebullitive.
L137. As all measurements were done within 1.5–3.5 meters of water depth, you are likely to have low ebullitive emissions due to oxygen reaching the sediment, more wind disturbance and less accumulation of organic material. There might in fact be high ebullition in the deeper parts of the lake. Or in the reed belt where there is also high accumulation of organic material.
L182. Wrong parenthesis at NTU
Table 2. Use – rather than ÷ to display range. Units should be the same as the methods text even though they are equivalent.
Results and discussion
L216. missing mg after 14.6
L220. Also missing mg after 14.9
L224. Remove “prepared”
L224-230. Remove the absolute values. As you are comparing it to your values, which are in unit area, it is irrelevant here what the total area emissions are.
L260. In my opinion, you should discard ice-covered periods, as you will have a buildup of methane underneath the ice, which will eventually be released when the ice breaks, however, by only measuring in a few locations by drilling, you will cause few areas where the methane can actually escape and thus elevated emissions here. This scenario might be the case in the instance with emissions of 24 mg m-2 d-1.
L269. To my knowledge, a lot of people have tried explaining this variation, without luck. To state that the variability is attributed to weather conditions should at least be backed up further than just referring to a figure.
L307. Please explain prior how this skewness is calculated and what it means.
L310. Were tests made to look for intercorrelation between parameters? I would expect a high collinearity between water and air temperature.
L320–322. Please rephrase, it’s very hard to read this sentence.
L323–325. Need reference.
L326. Need references.
L328. Did you calculate wave action, and if so, how? This is the first time we hear about it. It needs to be in the method section as well.
Table 3. I am certain many of the indicators would show collinearity which should be taken into account. What emission values are used here, is it average from each sampling period?
Section 3.2. This section is missing more discussion on the parameters, which group together with methane emissions, and why it is expected or not. What effect do the different parameters have on methane emissions? Right now, it is not used for much, or at least you do not say why you do it. I can read in the next section why you do it, but I would want to know beforehand.
L381. Yes, water temperature affects methane production, but that is different from methane emission. Water temperature also affects methane oxidation.
L411. “much” higher is pushing it here.
Table 4. Please indicate the confidence interval around the predicted values.
L449. It should also be mentioned that the methane oxidation will increase with temperature.
L452. Add reference.
L467. B coefficient? I guess it is the slope.
Section 3.3. Please indicate when values or trends from references are done in-situ or in laboratory experiments.
L480. Unit is wrong 264.4 m-2 d-1.
Data availability
To my understanding Biogeosciences requires data to be uploaded in an online repository, which I encourage.
Citation: https://doi.org/10.5194/egusphere-2024-1786-RC2 - AC2: 'Reply on RC2', Andrzej Skwierawski, 13 Nov 2024
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 197 | 88 | 21 | 306 | 10 | 11 |
- HTML: 197
- PDF: 88
- XML: 21
- Total: 306
- BibTeX: 10
- EndNote: 11
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1