Studying boundary layer methane isotopy and vertical mixing processes at a rewetted peatland site using an unmanned aircraft system

Lampert, Astrid; Pätzold, Falk; Asmussen, Magnus O.; Lobitz, Lennart; Krüger, Thomas; Rausch, Thomas; Sachs, Torsten; Wille, Christian; Sotomayor Zakharov, Denis; Gaus, Dominik; Bansmer, Stephan; Damm, Ellen

doi:https://doi.org/10.5194/amt-13-1937-2020

Articles | Volume 13, issue 4

https://doi.org/10.5194/amt-13-1937-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/amt-13-1937-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 13, issue 4

Research article

|

17 Apr 2020

Research article |

| 17 Apr 2020

Studying boundary layer methane isotopy and vertical mixing processes at a rewetted peatland site using an unmanned aircraft system

Astrid Lampert, Falk Pätzold, Magnus O. Asmussen, Lennart Lobitz, Thomas Krüger, Thomas Rausch, Torsten Sachs, Christian Wille, Denis Sotomayor Zakharov, Dominik Gaus, Stephan Bansmer, and Ellen Damm

Download

Final revised paper (published on 17 Apr 2020)
Preprint (discussion started on 17 May 2019)

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

RC1: 'comments', Anonymous Referee #1, 17 Jul 2019
- AC1: 'Answers to Referee 1', Astrid Lampert, 18 Sep 2019
RC2: 'Review Lampert et al. “Studying boundary layer methane isotopy and vertical mixing processes at a rewetted peatland site by unmanned aircraft system”', Anonymous Referee #2, 22 Jul 2019
- AC2: 'Answers to Referee 2', Astrid Lampert, 18 Sep 2019

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

AR by Astrid Lampert on behalf of the Authors (18 Sep 2019) Manuscript

ED: Referee Nomination & Report Request started (26 Sep 2019) by Christof Ammann

RR by Anonymous Referee #1 (08 Oct 2019)

Suggestions for revision or reasons for rejection

The authors have done a good job in revising the manuscript. Most problems of the earlier version have been solved. The only major issue remaining is that differences in delta 13C between samples taken simultaneously at a horizontal distance of 0.13 m on a descending quadrocopter a few 100 m above ground level are considered "features" (page 11, lines 6-9).

Air movement above the inversion layer is not laminar but turbulent. Further, turbulence next to the parallel sample inlets is enhanced by the four propellers surrounding the sampler. Horizontal sample integration due to natural wind is up to 4 m (page 10, lines 24 and 25). Vertical integration during descent at 2 m/s is at least 2.6 m (2m/s * 1.3 s). Even without accounting for the additional flow and mixing induced by the propellers, only a minute fraction of the sample volumes collected in parallel can originate from different air parcels that have not been subject to mixing since they had been in contact with different methane sources at the land surface more than 100 m below the copter. Let's be optimistic and say that 10% of the volumes sampled in parallel were from different, unmixed air parcels. Then, the difference in delta 13C between these parcels would have had to be 4 permil to produce a 0.4 permil difference between the parallel samples. I find such horizontal gradients more than 100 m above ground highly implausible. The difference between below and above inversion layer is less than half that value. Although the source of methane on the ground may vary by a few permil across distances of a few metres, the probability that such a difference is conserved in air during transport across and well beyond the inversion layer is certainly much smaller than is the probability of obtaining by pure chance the results shown in 10b. Anything smaller than 0.5 permil is anyway within the limits of uncertainty (precision level) and therefore not a "feature". To have a chance of eventually observing a real horizontal gradient, samples would need to be taken upwind and downwind of a strong local source (e.g. the rewetted peatland). An example of horizontal concentration gradients between up- and downwind a local CH4 source is shown in Denmead et al (Atmos. Environ. 32, 3679–3688, 1998).

The rather peculiar argument continues on the same page, lines 21 to 24:

"An ideally vertically stratified delta value would not be sampled by the present system as the very dynamic circulation process around the quadrocopter does not result in a homogenously mixed air at the sample ports on the quadrocopter. On the contrary, this circulation process can even amplify natural inhomogeneity. It is assumed that the differences in delta values indicate natural inhomogeneity..."

How should natural inhomogeneities be amplified by circulation processes, such as visualised in Fig. 3? Is entropy in a fluid decreased by stirring? Can I un-mix coffee and milk by stirring it?

Page 10, lines 18 and 20: "dilusion": do you mean "delusion" or "dilution"?

Please add a scale bar to Figures 4 and 11.

Hide

RR by Anonymous Referee #2 (25 Oct 2019)

Suggestions for revision or reasons for rejection

The study by Lampert et al. describes an approach by which air samples can be obtained from different altitudes reaching far into the atmospheric boundary layer (ABL) using an unmanned aerial system. The authors have added important information and I really appreciate the RANS modelling concerning rotor downwash. However, the authors have stopped short with fully exploiting the information provided, and turning them into valuable information for others in the discussion. For my understanding, this requires further elaboration. My most important points are as follows:

1. Still lack of precision in formulations, quite often I was confused at first and had to read several times to understand the meaning. For instance, the term origin sometimes refers to spatial allocation, and sometimes to the production mechanism. I suggest to use spatial origin and source process respectively.
2. Determination of origin at different heights above ground is mentioned as one of the two most important aspects of the study this i) not the same aims as described in the introduction and ii) the abstract never gets back to a conclusion on this issue. The implications of aim 2 given in the introduction remain unclear. What new methods need to be developed?
3. The RANS simulations are really interesting. Thank you for adding. However, it remains unclear to me how the calculation of the spatial allocation of the sampled air arises. Please clarify as I am not sure if I understood correctly: The authors have taken the largest relative air speed to calculate the travel distance during the 1.3 seconds it takes to fill the glass vessels. To my understanding, the sample air inlet was above the rotors, and I don’t see these relative speeds above the rotors. This is because air from a spherical half-space above the rotor is washed down by the rotor. The calculation would make sense if air exclusively straight above the rotors would be washed down, but the figure does not support this notion. It seems like the highest relative speed above the rotors is in the order of 12-15ms-1. Wouldn’t it be more accurate to take the highest speed above the rotors to estimate the maximum travel distance? In addition it is not clear to me if only the downward pointing velocities are shown. Further, this figure suggests that having a sideways pointing tube as air inlet that reaches 25cm beyond the rotor would be a good option for further uas based systems, or a tube that reaches 50 cm above the rotors. Also the discussion if hovering would be the best option for taking a sample is not discussed at all. This is a pity and deserves a section in the discussion.
4. Figure 1 doesn’t tell anything

See some more detailed comments below.
Title
ok
Abstract
P1L2: “quadrocopter-borne air sampling and of methane isotopic analyses, is applied to determine the origin of methane”: I am not sure what the authors mean exactly with origin. Methane isotopic composition can give hints on the process that had produced the methane, but does not tell anything about the location. It could have been produced anywhere in the landscape where the respective production process had taken place. In addition, I don’t see where the authors get back to the origin of methane in the abstract. There should be a reference to the origin if this is mentioned at the foremost place in the abstract.

Introduction
P2L5: one reference can be removed
P2L19: Transformation of the isotope ratio to the delta scale _13C‰ results in values that allow to distinguish the source: Also the isotope ratios allow this process identification. The conversion to delta scale is just more intelligible for us, but is no other information. Just stick to : Isotopic composition allows to distinguish different source categroies.
P2L24: please change from predominantly to “an increasing share of ”
P3L36: I suggest adding that this is typically happens during night.
P5L15: please change from content to volume
P8L8: heated away is colloquial, please rephrase

Materials and Methods

Results
Table 1: the data shown is not isotope ratio but delta value.

Discussion
P10L15: vertical mixing is possible, but hampered.
L18: please change from dilusion to dilution. If the assumption is that air from above the inversion mixes with air below as the inversion climbs up and is dissolved, I suggest calling it mixing. Dilution alone, in the sence of mixing with for instance ch4 free air would not alter the isotopic composition. What you observed is the temporal variability of CH4 isotopic composition which is caused by transport and mixing.
L20: dilusion again. Is this a real technical term that I am not aware of?
Conclusions:
P12L5: in agreement with atmospheric stability: what does this mean? Maybe caused by atmospheric stability?
P12L16: isn’t this bullet point the same as above?

Hide

ED: Reconsider after major revisions (10 Nov 2019) by Christof Ammann

Dear authors,

the two Referees have reviewed the revised version of your manuscript and found significant improvements. However, according to their assessment (see the individual referee reports in the manuscript overview) there are still important shortcomings that need further improvement.
In addition, I have identified some issues as listed in the following comments.

1) I consider the micrometeorological interpretation as insufficient and partly problematic. The objective of the study (as stated in the introduction) was to identify the isotopic composition of vertical layers in the boundary layer and to test the capability of the system to provide reliable vertical profiles of the isotopic composition.
The authors should consider that a vertical profile in the micrometeorological sense is always a time-averaged profile over typical averaging intervals of about 30 min. In this way, the turbulent fluctuations (temporally and spatially) can be integrated. Alternatively to a temporal average, a spatial average (on constant heights) may yield the same result, but only if the surface is sufficiently homogeneous.
Therefore, the authors should consider and discuss, whether the instantaneous point sampling is sufficient and useful for micrometeorological profile analysis. It should be discussed whether a longer sampling time (e.g. using a capillary at the inlet, e.g. with repeated valve openings) would be more useful and possible with the sampling platform.

2) The issue addressed in comment 1 also has implications for the interpretation of the differences between the two parallel samples. I do not think that this instantaneous and small-scaled differences represent "systematic features". If they are not due to sampling/analysis problems, they represent local turbulent fluctuations. But two instantaneous samples are probably not enough to provide insight into the turbulence dynamics.
Please consider this in the formulation of the second objective in Section 1 and in the discussion in Section 4.

3) page 8, line 15/16: Indicate the (horizontal) position of the UAV profile measurements in Fig. 4 and/or Fig. 11

4) The newly introduced Fig. 3 is quite interesting but I have the following suggestions for improvement:
a) I am not sure if the indicated sequence in the figure caption "hover, climb, descent" corresponds to the A, B, C sequence of the figure panels. Please give precise attribution in the Figure caption.
b) It would be useful to indicate the position of the sample inlets relative to the propeller blades and the simulated induced flow fields.
c) Indicate the (simulated) ascent and descent speed in the Figure caption.

5) According to Figs. 5 and 6, the wind direction during all flights was between 0 and 90 deg corresponding to north-easterly wind directions. This would mean that the Eddy Covariance tower measured flux footprints outside of the polder area shown in Figs. 4 and 11 (and distant from the water sampling). Please comment on that.

6) Add length scales to Figs. 4 and 11. This is important to relate the horizontal extension of the polder to the profile measurement heights.

Please consider the above comments and the referee comments (see separate reports) in your further revision.

Best regards
Christof Ammann

Hide

AR by Astrid Lampert on behalf of the Authors (22 Dec 2019) Author's response Manuscript

ED: Publish subject to minor revisions (review by editor) (23 Jan 2020) by Christof Ammann

Dear authors,

You made some improvements in the recent revised version. However, I am not satisfied with some author responses and revisions. I still consider some micrometeorological interpretation as insufficient and problematic. The text modifications partly went into the wrong direction. This necessitates some considerable improvements of the manuscript text as detailed below.
The following comments mainly refer to the editor and referee comments of the previous review (with some additional comments at the end). The page and line numbers refer to the newest revised manuscript version.

EDITOR COMMENTS 1&2 (of previous review)

a) With one (or a few) dozen of grab air samples, it is not possible to determine small-scale variabilities in the boundary layer in a meaningful and quantitative way. As mentioned in the abstract (page 1, line 20), the parallel samples should mainly be considered as a mean to check the reliability and reproducibility of the profile measurements (see also Comment of Referee#1 below). Thus, I recommend to remove the 2nd objective (page 4, lines 3-4) and the corresponding statements elsewhere in the text.

b) page 3, line 1-6: Remove this newly introduced paragraph. It is erroneous or misleading. You totally misinterpreted my original comment. The typical 30 min interval is generally used in micrometeorology to measure and calculate quantities of the boundary layer and turbulence (not just means but also variations, covariances, fluxes, spectra, etc.). This interval comprises most occurring eddy sizes and therefore prevents undersampling.

c) It still needs to be discussed whether the instantaneous point sampling is sufficient and useful for micrometeorological profile analysis.

d) page 3, line 4-6 (and page 12, line 8): This is an erroneous/misleading interpretation of the results of Schaller et al. (2019). They clearly concluded that the episodic 'outburst' events are not connected to soil emissions but to atmospheric dynamics (non-stationary or intermittent turbulent mixing).

REFEREE#1 COMMENTS (of previous review)

e) I do not accept the response of the authors about the "different perceptions of the mixing process". From a micrometeorological perspective, mixing never amplifies concentration gradients or differences but always acts to reduce them. Thus, I want to enforce the critics of Referee#1 concerning the statements on page 12, lines 15-17. These statements should be rephrased or omitted.

REFEREE#2 COMMENTS (of previous review)

f) I want to enforce Comment 2 of Referee#2 that the second objective (page 4, line 3-4) is unclear. The statement "...which require the development of new methods for understanding turbulent mixing processes" is not formulated as an objective. Considering my comment a) above, the second objective should be removed.

ADDITIONAL COMMENTS

g) page 4, line 2: Remove or rephrase the new last part of this objective ("...and therefore the spatial origin ..."). The identification of the spatial origin of the samples/layers is not provided in this study.

h) page 4, line 10: Change "constant" to "near constant"

i) Fig. 3 caption: Rephrase to "Simulation of the relative vertical flow velocity (Vrel) induced
by ..."

j) page 6, line 15: I guess this should read "3913 min^-1"

k) page 11, line 24: Correct to "...except for the last ..."

l) page 11, line 30: Clarify to "...that are not mixed across the inversion."

m) page 13, line 1: What is the specific meaning/purpose of "instantaneous profiles"? Did you mean "simultaneous profiles" here?

n) page 13, lines 5-6: Better write: "to determine vertical differences ..."

Please duly consider these comments in your further (hopefully final) revision.

Best regards
Christof Ammann

Hide

AR by Astrid Lampert on behalf of the Authors (02 Feb 2020) Author's response Manuscript

ED: Publish subject to minor revisions (review by editor) (25 Feb 2020) by Christof Ammann

Dear authors,

you performed the specifically requested modifications in the text. But it should be clear that other closely related parts of the text also need to be adjusted (in particular concerning my previous comments (a) and (e). Especially Section 4.2 still contains misleading and confusing statements as well as repetitions. The response to my previous comment (c) was not at all satisfying. Therefore, more improvements are necessary before publication.
It is not possible to give detailed suggestions for modifications in every case. However, I try to be as specific as possible in the following list of comments.

Christof Ammann
AMT Associate Editor

SPECIFIC COMMENTS

1) page 10, line 27/28: I suggest the following modification: "Small-scale differences in methane isotopic composition can be introduced by intermittent vertical mixing processes, which may also be induced by the quadrocopter system itself."

2) page 11, line 1-3: This is a confusing statement. The previous sentence describes the systematic vertical difference (layering) in the delta values. Such a vertical gradient is not primarily due to small-scale inhomogeneity but rather due to a significant local ground source in combination with stable stratification (i.e. a lack of vertical mixing).
I suggest to modify to: "This shows that the observed systematic vertical differences are not caused by measurement uncertainty, but are due to local emissions in combination with limited mixing due to stable stratification."

3) page 11, line 13/14: Rephrase to: "However, that latter value has been determined experimentally for long laboratory time series, while the uncertainty may be lower for subsequent analyses."

4) page 11, line 16: Omit "systematic"

5) page 11, line 18/19: This statement is a repetition of lines 11-15 and thus can be omitted.

6) page 11, line 20/21: The statement in the first sentence is not fully true (only for the first four flights) and, as it applies to only one set of four values, it may be purely random. The second sentence is confusing and not scientifically sound. Thus, I recommend to omit both sentences.

7) page 11, line 22-25: This paragraph should be integrated into the normal text flow.

8) page 11, line 27/28: "...and the differences in delta values show a similar height profile for subsequent flights." I cannot see these 'similar height profiles' in Fig. 9 and 10. Thus, I suggest to delete this part of the sentence (or to be more specific in the explanation).

9) It still needs to be discussed whether the short sampling time of 1.3 s can yield representative information of the methane and isotope concentration (profile).
E.g. in page 11, line 30 it could be added: "However, the very dynamic circulation process around the quadrucopter and the short sampling time of only 1.3 s have to be taken into account."

ADDITIONAL TECHNICAL AND LANGUAGE CORRECTIONS

10) page 2, line 30/31: "... to determine contributions from isotopically different sources to the dominating source". This is a confusing formulation, please rephrase.

11) page 10, line 21/22: Change to "...during stable atmospheric conditions (Wolf
et al., 2017), as these prevent mixing with..."

12) page 13, line 27: For clarification please modify to: "...the associate editor Christof Ammann..."

13) page 14, line 34: Remove Denmead et al. (1998) from the reference list, as it no longer occurs in the text.

Hide

AR by Astrid Lampert on behalf of the Authors (04 Mar 2020) Author's response Manuscript

ED: Publish as is (13 Mar 2020) by Christof Ammann

AR by Astrid Lampert on behalf of the Authors (16 Mar 2020) Manuscript

Short summary

Methane has high climate warming potential. Sources of methane can be distinguished by the isotopic composition. To investigate the origin of methane, an airborne sampling system has been developed that can take air samples worldwide and at various altitudes. The article shows the performance of the overall system, from taking samples to laboratory analyses. As known methane source, a rewetted peatland site, was studied, and the vertical distribution of the isotopic composition is investigated.