Review Reports - Air-Sea Interactions over Eddies in the Brazil-Malvinas Confluence

Round 1

Reviewer 1 Report

In the reviewed manuscript the authors presented multidisciplinary research combining field in situ micrometeorological and oceanographical study, ERA5 results, and satellite observations. The research investigates properties of the Brazil-Malvinas Confluence (BMC) by paying close attention to two different eddies: warm (ED1) and cold (ED2). The paper is important in understanding mesoscale weather systems and have a scientific-sounding. There are however many major mistakes and a lack of proper scientific discussion.

One major editorial issue which makes reading the paper very difficult is a jumbled citation numbering. The authors should fix this issue before submitting the paper.

One of my biggest concerns is the lack of a proper description of micrometeorological settings and data processing. Authors choose a shortcut and just linked to the methodology presented in papers [26] (de Camargo et al., 2013) and [29] (Oliveira et al., 2017). I suppose there is a wrong citation given here. As far as there is no ship’s name provided in the text we do not know is it the same ship as in the previous experiment. It should be discus. Inter alia a way of inertial correction and a range of error reduced or other corrections made. In this case, I can’t evaluate such results (especially with wrong citations).

Another issue is the lack of any statistics to prove the author's hypothesis. If the authors claim that i.e. SST gradients were major local forcing factors modulating the local weather, they should prepare carefully a deep statistical analysis of such results. Without it, it is only a qualitative result. If authors do not like the statistic, it should be the corresponding modeling experiment made, to prove or reject the 0 hypothesis.

In my opinion for such a good journal as Remote Sensing, publication in a present form should be rejected with an opportunity to resubmission. As I stated at the beginning, the topic is interesting and important, the results seem to be valuable. The wrong citations suggest that the authors prepared the paper negligently. It is impossible to evaluate them honestly in the present form.

de Camargo, R.; Todesco, E.; Pezzi, L.P.; Souza, R.B. Modulation mechanisms of marine atmospheric boundary layer at the Brazil-Malvinas Confluence region. J. Geophys. Res. Atmos. 2013, 118, 6266-6280, doi: 10.1002/jgrd.50492.

Sugimoto, S.; Aono, K.; Fukui, S. Local atmospheric response to warm mesoscale ocean eddies in the Kuroshio-Oyashio Confluence region. Sci. Rep . 2017, 7, 11871. https://doi.org/10.1038/s41598-017-12206-9.

Oliveira, F.S.C.; Polito, P.S. Mesoscale eddy detection in satellite imagery of the oceans using the Radon transform, Prog. Oceanogr. 2018, 167, 150-163, doi.org/10.1016/j.pocean.2018.08.003.

Author Response

Answer. Eventual mistakes were corrected as suggested by the reviewer. New processing was made and a new and more complete abstract was written. New citations were added, especially in the field of micrometeorology. New figures were also added or improved. We believe the discussion is quite robust and cover many fields of expertise ranging from oceanography, meteorology, micrometeorology, climatology to remote sensing.

One major editorial issue which makes reading the paper very difficult is a jumbled citation numbering. The authors should fix this issue before submitting the paper.

Answer. Citations were checked and fixed. They are presented in numbers correspondent to the order of appearance in the text.

Answer. We added many new references regarding the methods and analysis of the micrometeorological data, and corrected eventual erroneous citations. The new citations include Hackerott et al. [26], Fratini et al. [27], Edson et al. [38], Miller et al. [39], Fugitani [40]. The new paper by Santini et al. [16], whose 1^st author is also contributing for this paper, presents many techniques used here and for its novelty and publication in a well-known meteorological journal, is widely cited. Santini et al. [16] and other authors leading other publications in prestigious journals also use the same vessel and many micrometeorological instruments used here, although in other INTERCONF campaigns. We explicitly cite this fact in line 243 of our newly revised version of the manuscript.

Answer. We agree with the reviewer in this point. We remade many of our discussion now based upon new ERA5 data collected in an area outside ED1 and ED2 for comparison. We recovered new data in a position inside the Zapiola Rise, a known bathymetric feature in the Southwestern Atlantic Ocean where the Eddy Kinetic Energy is very low, thus serving as a reference for neutral waters only subject to remote, large scale forcing of the atmosphere for not sustaining eddy activity. The differences between measurements made on top pf the eddies with respect to the Zapiola Rise are an indication of the importance of ED1 (warm core eddy) and ED2 (cold core eddy) on locally modifying the lower atmosphere along the eddies’ tracks. Results are presented as percentages of change in the variables in comparison to the ones of the neutral region. We based our analysis in similar methodology used by Leyba et al. [8] and Bharti et al. [53]. Our hypothesis, as stated in lines 217-221 of the revised manuscript is that in the Zapiola Rise region (45 ^oS, 42 ^oW) a region located close to where ED1 and ED1 existed, is subject to the same large scale, meteorological variability that affected the MABL above the eddies along their trajectories, although not being affected by an eventual local forcing caused by the eddies’ presence. In this case, an expressive difference between the time series of ZR with respect to ED1 and ED2 can be expected. The differences are expressed in terms of percentages following Leyba et al. [8] and Bharti et al. [53], and proved to be high (Table 4 – lines 552 to 556 of the revised manuscript).

Oliveira, F.S.C.; Polito, P.S. Mesoscale eddy detection in satellite imagery of the oceans using the Radon transform, Prog. Oceanogr. 2018, 167, 150-163, doi.org/10.1016/j.pocean.2018.08.003.

Answer. We do not think this is a fair comment. The number of citations in our text is quite extensive and we are doing our best to present a good paper using all reference available to our disposal and knowledge. Mistakes, however, when occurred in the previous version of the manuscript, were corrected as the reviewer suggested. The references cited by the reviewer were corrected and suggestions included in the reviewed manuscript.

Submission Date

11 January 2021

Date of this review

27 Jan 2021 19:06:34

Reviewer 2 Report

Review of "Air-sea interactions over eddies in the Brazil-Malvinas Conflu-ence" by Ronald Souza, Luciano Pezzi, Sebastiaan Swart, Fabricio Oliveira and Marcelo Santini.

The authors investigated air-sea interactions over the Brazil-Malvinas Confluence (BMC) region using variety of datasets including in-situ, satellite and atmospheric reanalysis data. Analysis results indicate that the marine atmospheric boundary layer (MABLE) is modulated over mesoscale eddies in the BMC. The mid-latitude air-sea interaction is one of the hot topics in climate study. So I think this study, describing air-sea interactions in the BMC using actual observations, would be beneficial for readers of remote sensing. Thus, this manuscript is acceptable for publication after minor revision.

P3 "Radon transform":
I think there are a lot of algorithms to detect mesoscale eddies (e.g., identification using the Okubo-Weiss parameter).
If possible, it would be better to describe following information, which would be useful for readers.
Why did the authors adopt the Radon transform method? And what is the merit of this method compared to other methods?

P6 "EC method":
Does "EC" stand for Eddy Covariance?
Please specify formal name of this acronym.

Figure 3:
The gray and white dots are almost indistinguishable. The authors should try to draw it a little better.

P9 2nd paragraph:
"0.3 +- 0.07 mm (ED1) and -0.5 +- 0.13 mm (ED2)"
I think the unit should be "m", not "mm". Please check it.

Sec 3.3:
There is no comment for Figure 8b. The time series of air temperature at 10 m (T10) is well synchronized with that of sea surface temperature from 17 October 17 to October 18, when the ship's track across the ED1 and ED2. This implies that warm (cold) water of ED1 (ED2) modifies surface air temperature. So, I think this is important result.

Author Response

Review of "Air-sea interactions over eddies in the Brazil-Malvinas Conflu-ence" by Ronald Souza, Luciano Pezzi, Sebastiaan Swart, Fabricio Oliveira and Marcelo Santini.

Answer. We extended the explanation about why we used the Radon Transform in lines 144-169 of the new version of the manuscript. We made a point that the Radon transform, as described by Oliveira and Polito [28] was compared to other ten methods for eddy detection, including the Okubo-Weiss. The authors listed a series of negative and positive points among the distinct methods, reporting that the Radon transform presents versatility and is amplitude-independent, although requiring a previous knowledge of eddy size range for being properly applied. According to Oliveira and Polito [28], the Okubo-Weiss method was used for supporting the first global statistics of eddies in the ocean by Chelton et al. [32], but is noisy and can promote false positives.

P6 "EC method":
Does "EC" stand for Eddy Covariance?
Please specify formal name of this acronym.

Answer. Yes, EC means Eddy Covariance, as stated in lines 260 and 336 of the reviewed manuscript.

Figure 3:
The gray and white dots are almost indistinguishable. The authors should try to draw it a little better.

Answer. Figure 3 (now Figure 4) was completely remade and this time we opted for a zoomed area in the Southwestern Atlantic Ocean were the eddies traveled. This allowed a better view of the SLA and wind fields in the period when the eddies were present. ED1 and ED2 trajectories are now in black for better visualization and the figure also refers to the previous one where the trajectories are better described.

P9 2nd paragraph:
"0.3 +- 0.07 mm (ED1) and -0.5 +- 0.13 mm (ED2)"
I think the unit should be "m", not "mm". Please check it.

Answer. Corrected to “m” (lines 438-439 of the new version). Thank you for spotting that.

Answer. Indeed it is an important result. We added a new comment on the subject on lines 637-641. Figure 8b is now Figure 9b. It indicates that T₁₀ is well synchronized with SST in the period 17-18 October, when the ship's track crossed the eddies. This implies that warm (cold) surface waters of ED1 (ED2) were important on modifying the surface air temperature in respect to surrounding areas outside the eddies, an important result showing the local modulation of these eddies in the lower levels of the atmosphere.

Submission Date

11 January 2021

Date of this review

25 Feb 2021 08:21:43

Reviewer 3 Report

Comment to remotesensing-1090210:

The paper “Air-sea interactions over eddies in the Brazil-Malvinas Confluence” by Souza et al. investigated influence of ocean eddies on the atmospheric dynamics via local air-sea interactions. This analysis is based on various datasets, including remote sensing, reanalysis, and in-situ data. The results are logical. I found the paper is of interest, and could be accepted for publication after some necessary clarifications.

Major comments:

More detailed information about the two eddies are needed, for example, the size of the eddies, is the ¼ resolution satellite products good enough to resolve the eddy? This could be added in Fig. 2b.
There are multiple eddies in the studied region (see Fig1b), why only these two are chosen for the study? How does the role of other eddies play in shaping the atmospheric dynamics.
The in situ measurement doesn’t show much difference over ED1 and ED2 in terms of heat flux, this contradict to the main conclusion of the paper, which stated ‘static hypothesis is the driving mechanism of the local air-sea interactions from the eddies’. See detailed comment in the file regarding Fig.8.
Since there is no line numbers in the document, I inserted all the detailed comment in the pdf in the attachment.

Author Response

Comment to remotesensing-1090210:

Major comments:

More detailed information about the two eddies are needed, for example, the size of the eddies, is the ¼ resolution satellite products good enough to resolve the eddy? This could be added in Fig. 2b.

Answer. Information about the eddies’ sizes was added in lines 135-136 of the revised manuscript: During their lifespans, the mean diameters of ED1 and ED2, as estimated from their surface signature with respect to surrounding waters in SLA maps where 104 km (ED1 varied between 86-122 km and ED2 varied between 95 and 114 km). Line 180 present the estimates made during the in situ campaign when the eddies were observed from the ship’s instruments: The eddy diameters, as estimated by the maximum SST gradients with respect to surrounding waters using thermosalinographer data obtained in October 2013, were about 150 km for ED1 and 130 km for ED2. The ¼ resolution of the satellite data represents ~25 km in the mean latitude of 41 ^oS, thus able to resolve the eddies’ imprint. The eddies’ diameters are also within the sizes able to be detected by the ERA5 dataset (30 km - see line 223 of the revised manuscript) and by the Radon transform (see lines 173, 180-182, 201 of the revised manuscript). The information was also added to the Figure 2 (now Figure 3, lines 396-403 of the revised manuscript).

There are multiple eddies in the studied region (see Fig1b), why only these two are chosen for the study? How does the role of other eddies play in shaping the atmospheric dynamics.

Answer. We choose to describe eddies ED1 and ED2 in the paper because they were sampled during the INTERCONF-32 campaign, thus allowing us to directly measure the air-sea heat fluxes inside these structures. Lines 122-127 and 136-139 of the revised manuscript provide an explanation on why these particular eddies were studied here. We followed these two structures for they were clearly warm core and cold core eddies clearly identified on the XBT, CTD and thermosalinographer data collected during the in situ campaign (Figure 7 and Figure 9). The objective of this paper was not to describe the air-sea interaction properties of all the eddies in the study area.

The in situ measurement doesn’t show much difference over ED1 and ED2 in terms of heat flux, this contradict to the main conclusion of the paper, which stated ‘static hypothesis is the driving mechanism of the local air-sea interactions from the eddies’. See detailed comment in the file regarding Fig.8.

Answer. We added in lines 681-705 an extensive explanation for the cause of ED1 and ED2 heat fluxes being not very distinct during the INTERCONF-32 campaign. The air-sea heat fluxes over ED1 and ED2, as measured by the EC method, were directly influenced by the atmospheric stability condition occurring during 16-18 October 2013 and the EC method tends to fail when the atmosphere is stable or nearly neutral (lines 363-370 and 684-688 of the revised manuscript). References are: Santini et al. [16], Pattey et al. [49], Yusup et al. [50] and Sun et al. [50].

Since there is no line numbers in the document, I inserted all the detailed comment in the pdf in the attachment.

Answer. We believe we adapted the manuscript line by line following the detailed comments in the original manuscript offered to us in the .pdf document. We also added a new figure (Figure 2, lines 268-271 of the revised manuscript) showing the instruments mounted in the micrometeorological tower that were used here. We thank very much the reviewer for such a great work. A particular and important point detailed in the .pdf file was that the reviewer spotted the presence of a spurious data in Figure 4c (now Figure 5c) that was now removed from the series. That resulted in a small modification of the ASCAT statistics for ED2 presented in Table 2. All the other minor or major questions described in the .pdf document were taken into consideration in the reviewed manuscript.

Submission Date

11 January 2021

Round 2

Reviewer 1 Report

The manuscript was significantly improved. All review suggestions were corrected.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Reviewer comments

Research article “remotesensing-884847-peer-review-v1” entitled:

Air-sea interaction over warm and cold core eddies in the Brazil-Malvinas Confluence Zone,

by Ronald Souza et al.

General Comments:

The study investigates local air-sea interaction processes in the BMC zone, in the presence of ocean eddies monitored by altimetry missions and also by in-situ data during a field campaign. The analysis focuses (a) on the differences in estimating heat fluxes using bulk formulas vs. eddy covariances and (b) on the signal modulation between wind and sea level at synoptic time scales.

The manuscript is well written and the general approach is valid. However, my main concern is that the analysis for the wind and sea level modulations is not very convincing and a more thorough investigation is needed to support the arguments in the text. Overall, I find the manuscript worthy of publication in Remote Sensing only after major revisions. Please find below major and minor comments, that the authors must address in the revised manuscript.

Major comments:

(1) Page 12, Lines 392-393 “(i)… demonstrate for… mesoscale structures in the BMC”: This is a very general statement and I can’t really understand how this was shown in the analysis following Figure 4. Is there a quantitative or qualitative assessment to support this argument? Clarify in the text.

(2) My main concern is that the second focal point of this study, i.e. the wind and sea level modulations discussed in the context of Figures 6, 7, and 8, is written in a rush manner, being very compact compared with other sections of the ms. and with assertions not entirely supported in the text. Please see below a few statements that are not properly posed and need further investigation:

(2a) Page 14, Lines 428-430 “Although the… Figure 7 shows… SLA pattern”: the correlation is relevant here, so must be shown and investigated more thoroughly (cf. also comment 2d), as the visual inspection of the filtered timeseries is not the proper analysis to justify the arguments in the text.

(2b) The authors here should also clarify if the filter is a low-pass or high-pass and the reasons why they applied. From Figures 7 and 8, I understand that this is a low-pass filter and I find later in the conclusions section that is a 5-day low-pass filter. This is where I start to be skeptical about the results shown in this section. How is it possible to remove the high-frequency signal by applying a low-pass 5-day filter and at the same time to discuss modulations for a period lesser than 5 days, i.e. cf. Page 14, Line 431 “… between 13-17 October 2013,…” and Line 436 “… at the 2-3 days period typical of… synoptic…”? This is a major shortcoming in this study that needs to be properly addressed, otherwise, the ms. should be entirely reconsidered.

(2c) In the same paragraph, Page 14, Line 433 “… favors the local SST (directly correlated to SLA)… ”: This statement is not clear; in which way favors the SST? Do you mean that when there are no atmospheric systems the local SST plays an important role in the MABL modulation? Clarify in the text. In addition, I am not sure if you can say that the SST is directly correlated to SLA. For instance, the wind has a thermodynamic effect on the sea surface, e.g. through the heat fluxes changing SST, but also has a dynamic impact through geostrophic and ageostrophic components (i.e. Svedrup and Ekman dynamics) that affect SLA. So, this argument requires a more thorough investigation and I strongly recommend removing it from the text.

(2d) Page 15, Lines 444-447 “… below r=0.5. An interesting… needs to be better investigated.”: First of all, correlations appear to be small and on top of that, the authors do not discuss if those correlations are statistically significant to some interval or not. Then, the authors close their arguments in a rush manner that further investigation is needed (e.g. possible lag between wind and sea level). I strongly recommend the authors to continue their analysis. It is clear here that, there aren't convincing results in this section of the ms.

Minor comments:

1) Page 8, Lines 286 and 289: I think it should be Figure 1b and 1c respectively, instead of Figure 2b and 2c.

2) Page 11, Figure 4h: The ζ variations are not visible; I would recommend using a log scale for the y-axis, e.g. with y-ticks [-10^(0) -10^(-1) 10^(1) 10^(0)] (note that it is possible to use log scale y-axis for zero and negative values).

3) Page 12, Lines 377-378: I think it should be ζ instead of z; in Eq. 5, z is the measurement height.

4) Page 13, Line 415: I can’t see any grey or white lines in the maps of Figure 6.

Best regards.

Reviewer 2 Report

Review of the manuscript “Air-sea interaction over warm and cold core eddies in the Brazil-Malvinas Confluence Zone” by Souza et al.

------------------------------------------------------------

Using observations field campaigns, the manuscript discussed MABL caused by the presence and transition of two ocean eddies in the Brazil Malvinas Confluence Zone. The topic if research is interesting, however, the presentation of the manuscript is very poor. Further, the use of poor sentence structure and the number of grammatical errors makes the text very hard to follow. Further, contrary to the mentioned data period in the method section the authors discussed the results beyond that period. This creates questions and an actual data period. They also used climate model data (CFSv2) for supplement/compare field campaigns, which is not acceptable.

Given the poor representation and overemphasizing the importance of the outcome, I feel this present manuscript may not be suitable for publication at this stage.

Recommendation:

Reject

Major revision:

The writing style is very confusing. The sentences should be written with a complete sense, even there is a reference. Otherwise, it’s hard to read and follow. For, example, the very first sentence of the introduction. These very unusual sentences persist throughout the text.

Please change all the shaded plots with rainbow colormap (jet). This is colormap is very often misleading.

https://blogs.mathworks.com/headlines/2018/10/10/a-dangerous-rainbow-why-colormaps-matter/#:~:text=Taylor%20II%20from%20the%20University,%2Ddata%2Ddependent%20gradients.%E2%80%9D

https://blogs.egu.eu/divisions/gd/2017/08/23/the-rainbow-colour-map/

https://www.climate-lab-book.ac.uk/2014/end-of-the-rainbow/

The introduction section is unnecessarily long, and many portions should be cut to provide a crisp background that is essential for this study. It’s not a review paper.

How did you identify the eddies? Nothing is mention in the text.

Figure 1 is hard to follow. Why did you use different time periods for different variables? How can you compare/relate like this way?

While in the 2.2 (methods) section the authors mention the metrological and oceanographic data were collected during 13-23 Oct 2013, INTERCONF research. And INTERCONF-32 was during 14-23 Oct 2013.Subsequently the results are shown/discussed for 22 Sept-1 Nov, 2013 (L401), 13 Sep to 1 Nov 2013 (L402). How did you get the data beyond the campaign period? It really poses the question of the method and real campaign data.

While we have high-quality reanalysis products, why the authors have used CFSv2 data (which is a model output) is not at all clear and totally unconvincing. The climate model always has biases, and not useful to use in this kind of study with a short period. Why did not you use ERA5 data?

“We are not aware of any other attempt to perform this kind of measurements and comparisons in the BMC region by any other research group in the World so far.”

This type of sentence is totally misleading and unsubstantiated. Despite a long introduction and lengthy text, the authors have not cited or discussed results from many relevant studies. For example,

Boebel, O., Schmid, C., Podestá, G., & Zenk, W. (1999). Intermediate water in the Brazil‐Malvinas Confluence Zone: A Lagrangian view. Journal of Geophysical Research: Oceans, 104(C9), 21063-21082.

Leyba, I. M., Saraceno, M., & Solman, S. A. (2017). Air-sea heat fluxes associated to mesoscale eddies in the Southwestern Atlantic Ocean and their dependence on different regional conditions. Climate Dynamics, 49(7-8), 2491-2501.

Pezzi, L. P., Souza, R. B., Farias, P. C., Acevedo, O., & Miller, A. J. (2016). Air‐sea interaction at the S outhern B razilian C ontinental S helf: In situ observations. Journal of Geophysical Research: Oceans, 121(9), 6671-6695.

Pezzi, L. P., de Souza, R. B., Acevedo, O., Wainer, I., Mata, M. M., Garcia, C. A., & de Camargo, R. (2009). Multiyear measurements of the oceanic and atmospheric boundary layers at the Brazil‐Malvinas confluence region. Journal of Geophysical Research: Atmospheres, 114(D19). (this is cited in the text)

Acevedo, O. C., Pezzi, L. P., Souza, R. B., Anabor, V., & Degrazia, G. A. (2010). Atmospheric boundary layer adjustment to the synoptic cycle at the Brazil‐Malvinas Confluence, South Atlantic Ocean. Journal of Geophysical Research: Atmospheres, 115(D22).(cited in the text)

Minor revision:

1. Please make sure references are inconsistent format. What is “AbrreviatedJounralName....Climate”?

2. What is the ocean’s “mixture layer” temperature? Do you mean mixed layer or are you introducing a new term?

3. L34: confusing. Air sea fluxes are dependent on SST as well.

4. What is INTERCONF?

5. What is the typical sampling period? You have assumed atmospheric conditions did not change abruptly (L160).

6. What is AWS? Write the full form.

7. What type of sonic anemometer was used (L203)?

8. Use proper degree symbol in the text.

9. Why did you use low-frequency channel data (L280)?

10. Figure 3: Too much information in a single plot. Please make the contour lines thick and clearer.

11. What are meteorological data and what are micrometeorological data (L214)? Clearly mention it.

12. What do you mean my visual analysis (L438)?

Editorial:

There are numerous grammatical errors. The authors should revisit the manuscript thoroughly for such corrections.

For example,

L45: changes on -> changes in

L88: sustain -> sustains –sentence needs modification for better clarity

L284: centred in -> centred on

L290: in vicinity -> in the vicinity

L286: that dates -> those dates

L297: as function -> as a function

L303: OML to rise to -> OML rise to

L354: route indicate -> route indicates

L370: was -> were

...and many more

Reviewer 3 Report

This paper describes research on air-sea interaction in the Southwestern Atlantic Ocean by in-situ and satellite observations. The description of observed atmospheric and oceanic physical quantities and calculated surface heat fluxes obtained mainly by in-situ observation campaigns is the main part, and satellite remote sensing was used to understand the spatial distribution. The title and theme seem vague. The authors should also include important information in the title and abstract. There are long introductions, from which I can roughly understand what has been done in past observational studies, but I have not been able to fully understand their relationship to this research. I think it is necessary to make the introduction more concise and emphasize the purpose and significance of this research. What's new about the air-sea interactions on oceanic warm and cold eddies? At this stage it looks like a cruise report. Therefore, it is suggested that the purpose and significance be clarified and rewritten as a research paper. If authors want to focus on the difference between direct observation and bulk estimate of turbulent heat fluxes, it may be more appropriate to submit to other journals. Specific comments are as follows. L201–208 in P6: This paragraph and the following paragraphs provide no information. It is not recommended to cite data processing methods from unpublished papers. At least the authors need to explain what kind of data processing was conducted. L289 in P8: Figure 2c should be Figure 1c L371–375 in P12: I couldn't understand this from Figure 4. I think it is necessary to describe the results more specifically. L373: Is Santini et al. (2020) same as ref.[21]? L405 in P12: Figure 7 and 8 are presented before Figure 6. L442 in P14: SLS (typo, SLA?) L470–472in P16: It's unclear from the results of this research why these rapid changes are responsible for the significant differences. Is it a problem to improve if the sampling frequency is improved? Figure 1: The caption should be rewritten using (a), (b), and (c). Figure 4: Sometimes the latent heat flux indicates large negative values. Is this correct? What mechanism causes negative latent heat flux?