The Atmospheric Boundary Layer and Surface Conditions during Katabatic Wind Events over the Terra Nova Bay Polynya

Round 1

Reviewer 1 Report

The paper presents a detailed analysis of the surface and atmospheric conditions in Terra Nova Bay between 18 and 25 September 2012 based on the numerical modelling simulations, satellite images and in-situ atmospheric measurements. Observations show that a characteristic feature of this area is the presence of strong (up to 30 m/s) winds (conditionally called “katabatic” winds) which influence the dynamics of sea ice coverage. The authors conclude that numerical simulations may provide incorrect representation of vertical ABL properties over the TNBP due to an overestimated sea ice concentrations used as a model input. All-in-all, the paper provides a wealth of raw data, satellite imagery and NWP results which would be of interest to a wide audience of readers of “Remote sensing”.

The paper is interesting and deserves publication. Prior to publication I would like to draw the authors attention to an overusage of comas in the text (some cursory remarks are listed below).

Line #1: correct to Antarctica

Line # 132: extra the

Line #562: at the same time (?)

Line #574: M.b. rephrase as : Whereas the winds flowing from southwest exceed 20 m/s near the coast and in the northern part of the bay…

Line #578: I think the coma is extra

Line #585: extra coma

Lines ##596,597,599:---

Author Response

First of all, we would like to thank the Reviewer for his/her insightful comments and valuable suggestions for corrections and modifications to the manuscript. Furthermore, we would like to acknowledge the Editors who selected the reviewers whose expertise covers both theoretical and numerical aspects of the manuscript.

The mistakes pointed out by the Reviewer have been corrected and the phrases at lines 562 (now 568) and 574 (now 582) changed. We also reduced the number of commas used in the article.

Reviewer 2 Report

In this paper the authors investigate the boundary layer properties over the Terra Nova polynya during strong katabatic wind events on September of 2012. The authors use surface observations, satellite and aerosonde measurements as well as the output of a numerical model to analyze the boundary layer properties and the influence of katabatic winds on the polynya development. The results presented are very interesting and worth publishing in Remote Sensing. There are however some points that should be addressed before publication. The first is that coverage of the literature regarding polynya development and boundary layer properties over polynya’s should be expanded. The second is that the argument on the contribution of the synoptic and mesoscale circulation on the katabatic winds should be strengthened using the output of the numerical model considered. And the last is to elaborate on the absence of a strong inversion in the beginning of the time period and the absence of a deeper convective layer in the middle of the time period considered. Specific comments follow.

MAJOR POINTS

1. Reference of existing studies on the boundary layer properties over the Antarctic and the development of polynya should be expanded and the connections of the results of this study to previously obtained results should be strengthened, therefore making this work more attractive and relevant. For example:

- Studies covering convective conditions over polynya should be referenced (such as Heinemann 2008, Meteor. Zeit. 17). In addition, it is stated in l. 361 that the deepening of the ABL observed is “a classic picture of air mass modification above polynya” without referencing any studies supporting this “classic picture”.

- While the article covers extensively in the introduction the influence of synoptic and mesoscale circulation on the extreme katabatic winds, the influence of katabatic winds on the polynya development is partially covered. In addition, the other factors influencing polynya development such as ocean currents (see Ebnar et al., 2014, Antar. Sci. 25) and frazil ice growth (see De Pace et al. 2019, Cryosph) should be covered as well.   

2. There is some observational evidence presented for the contribution of synoptic and mescoscale circulation in the katabatic winds that are present in the days covered by this study (for example the correlation with the SLP and the change in the wind direction on the 25^th). However, more information on the contribution of the pressure gradient and the buoyancy forcing producing the winds could be gained by the numerical model considered that seems to roughly capture the observed winds. Inclusion of such evidence would further strengthen the argument presented.
3. There are some questions regarding the thermodynamic structure of the ABL that are not adequately addressed. The first is: why isn’t there a classic strong inversion on the 18^th that accompanies the katabatic winds? Only the 0 profile shows one and this is from a site without the strong winds present. The second is: why isn’t there a strong convective signal on the 22^nd? The thermal properties of the ice (coverage, temperature) seem very similar to the 25^th of September, yet there is a shallow convective layer (it seems about 100-200m) compared to 700-1000m on the 25^th.

MINOR POINTS

1. In l. 443 it is stated that there is a low level inversion for profiles 0 and 4. However, there is no such inversion in profile 4.
2. In Figures 6-9, profile #4 is with a really light color that is not readable especially when printed.
3. There are minor typos throughout the text:

• 132. ….profiles”.” instead of “,” and double “the” before “outcome”
• 175. …and “is” applied
• 176. ….Next, “the” program
• 230. ….by “a” strong
• 231. “A” gradual increase
• 243. …consequence, “the” pressure
• 244 …and “the” wind speed at Manuela AWS “slows” down
• 251. …”The” presented
• 252. …that “the” pressure
• 275. …significant “changes” in the air
• 286 ….double “the” before direction
• 324. … mean that “convection”
• 325. … The top of “the” convective
• 339. … Fig. ????
• 369. …in almost “the” whole profile
• 384. Something is missing before TNBP
• 546. ….development of “a” mesoscale
• 561. …causing “a” drop
• 562. …same “time”
• 608. …simulates “the” temperature lapse rate and underestimates “the” upper

Author Response

First of all, we would like to thank the Reviewer for insightful comments and valuable suggestions for corrections and modifications to the manuscript. Furthermore, we would like to acknowledge the Editors who selected the reviewers whose expertise covers both theoretical and numerical aspects of the manuscript.

We will refer to specific comments made by the reviewer, our responses are colored blue:

MAJOR POINTS

Reference of existing studies on the boundary layer properties over the Antarctic and the development of polynya should be expanded and the connections of the results of this study to previously obtained results should be strengthened, therefore making this work more attractive and relevant. For example:

- Studies covering convective conditions over polynya should be referenced (such as Heinemann 2008, Meteor. Zeit. 17). In addition, it is stated in l. 361 that the deepening of the ABL observed is “a classic picture of air mass modification above polynya” without referencing any studies supporting this “classic picture”.

References mentioned by the Reviewer and additional ones have been added in the following lines in the manuscript: 367, 571, 580. 

While the article covers extensively in the introduction the influence of synoptic and mesoscale circulation on the extreme katabatic winds, the influence of katabatic winds on the polynya development is partially covered. In addition, the other factors influencing polynya development such as ocean currents (see Ebnar et al., 2014, Antar. Sci. 25) and frazil ice growth (see De Pace et al. 2019, Cryosph) should be covered as well.   

References mentioned by the Reviewer have been added in the following lines in the manuscript: 58, 66.

There is some observational evidence presented for the contribution of synoptic and mescoscale circulation in the katabatic winds that are present in the days covered by this study (for example the correlation with the SLP and the change in the wind direction on the 25th). However, more information on the contribution of the pressure gradient and the buoyancy forcing producing the winds could be gained by the numerical model considered that seems to roughly capture the observed winds. The inclusion of such evidence would further strengthen the argument presented.

The aim of our article was to analyze how AMPS resolves vertical changes of temperature and wind speed over  Terra Nova Bay Polynya. Therefore, the analysis of the pressure gradient and buoyancy forcing is beyond the scope of this paper. Furthermore, in our opinion such analysis may not bring anything valuable to the article, considering poor AMPS performance in resolving the vertical properties of the atmosphere. What is more, as mentioned in the article, the contribution of synoptic and mesoscale circulation to the development of the katabatic wind has already been recognized in other studies (e.g. Turner et. al., 2009; Ebner et. al. 2014) and the analysis of MERRA results agrees with the results from those articles.

There are some questions regarding the thermodynamic structure of the ABL that are not adequately addressed. The first is: why isn’t there a classic strong inversion on the 18th that accompanies the katabatic winds? Only the 0 profile shows one and this is from a site without the strong winds present. The second is: why isn’t there a strong convective signal on the 22nd? The thermal properties of the ice (coverage, temperature) seem very similar to the 25th of September, yet there is a shallow convective layer (it seems about 100-200m) compared to 700-1000m on the 25th.

The lack of strong inversion in the profiles from 18 September can be explained by very strong, katabatic flow. Bromwich (1989) shown that katabatic flow signature is present in thermal infrared wavelengths due to vertical mixing generated by turbulence. We expect that it is this mechanism that is acting in the strong katabatic flow of TNB that results in no surface based inversion being evident even in the near coastal profile within the katabatic flow. Later, on 19 September the situation is different due to changes in surface conditions. This is explained in the article in lines: 306-309. 

Whereas on 25 September, although the surface conditions are similar to the ones from 22 September, the origins of the airflow are different. On 22 September the downslope flow is largely of katabatic origin, due to north-northwest direction, while on 25 September the winds are weaker and come from the south-southwest. This affects the ABL properties as the ongoing transport of cold air from the interior of the continent, associated with katabatic winds, is much weaker. Of particular note, is the difference in air temperature above the boundary layer on 22 and 25 September. On 25 September the air aloft, near 500 m, has cooled by ~5 deg C from conditions on the 22nd, with the pronounced inversion near 500 m on the 22nd no longer evident on the 25th. This allows the near surface air to convectively rise to greater depths on the 25th. This is explained in the article in lines: 579-586.

MINOR POINTS

1. In l. 443 it is stated that there is a low level inversion for profiles 0 and 4. However, there is no such inversion in profile 4.

The inversion is present only in the first few meters. We changed the color of the profile and now it is easier to notice the curvature of profile 4 from 18 September. 

1. In Figures 6-9, profile #4 is with a really light color that is not readable especially when printed.
2. There are minor typos throughout the text:

• 132. ….profiles”.” instead of “,” and double “the” before “outcome”
• 175. …and “is” applied
• 176. ….Next, “the” program
• 230. ….by “a” strong
• 231. “A” gradual increase
• 243. …consequence, “the” pressure
• 244 …and “the” wind speed at Manuela AWS “slows” down
• 251. …”The” presented
• 252. …that “the” pressure
• 275. …significant “changes” in the air
• 286 ….double “the” before direction
• 324. … mean that “convection”
• 325. … The top of “the” convective
• 339. … Fig. ????
• 369. …in almost “the” whole profile
• 384. Something is missing before TNBP
• 546. ….development of “a” mesoscale
• 561. …causing “a” drop
• 562. …same “time”
• 608. …simulates “the” temperature lapse rate and underestimates “the” upper

The mistakes pointed out by the Reviewer has been corrected according to the suggestions.