Pressure-Gradient Forcing Methods for Large-Eddy Simulations of Flows in the Lower Atmospheric Boundary Layer

Round 1

Reviewer 1 Report

Review of Pressure-Gradient Forcing of Precursor Wind Flows in Wildfire Simulations

By Pimont, Dupuy, Linn, Sauer, and Muñoz-Esparza

General comments:

This study investigates the application of a technique that represents a pressure gradient forcing on wind flow. They use four versions of this PGF which are compared to existing techniques using idealised simulations. They show that these new forcings effectively represent  the turbulent flow and structural features are captured. Some of the variants are also effective in reducing streaks/streak locking.  This paper is well-written, with clear explanations and descriptions of the methods used and the results found. The model description was simple to follow and was explained clearly.  As such I recommend this paper be acceptable pending minor revisions.

Specific comments:

Abstract

L17: ‘forcing’ – ‘forcings’?

L23: ‘turbulent’?

L23: ‘The variants can either enable to prescribe …’ – awkward. Perhaps : ‘The variants can either enable researchers/models/… to prescribe …’?

L26: ‘EKB’ - perhaps define this in line 16.

Introduction

L41: ‘For instance, simulations enable to …’ Enable who?

L56: remove comma after ‘domain size’

L61: ‘Modelling phenomenology …’ – not really sure what this is; the study of phenomenons?

L208:replace  ‘latter’ with ‘later’

L338-339: Any particular reason 40m height was chosen?

L375: remove dash between 106-m

L517-518: Think the caption may be wrong!

L526: ‘they enable to..’ – again, enable who?

Author Response

Dear Reviewer,

Please find our replies below in italic.

Best regards

F.

General comments:

This study investigates the application of a technique that represents a pressure gradient forcing on wind flow. They use four versions of this PGF which are compared to existing techniques using idealised simulations. They show that these new forcings effectively represent  the turbulent flow and structural features are captured. Some of the variants are also effective in reducing streaks/streak locking.  This paper is well-written, with clear explanations and descriptions of the methods used and the results found. The model description was simple to follow and was explained clearly.  As such I recommend this paper be acceptable pending minor revisions.

=> We would like to thank you for your positive appraisal of our manuscript and for the specific comments which improved the manuscript. We followed all suggestions, as detailed below.

Specific comments:

Abstract

L17: ‘forcing’ – ‘forcings’?

=> Done

L23: ‘turbulent’?

=> Done

L23: ‘The variants can either enable to prescribe …’ – awkward. Perhaps : ‘The variants can either enable researchers/models/… to prescribe …’?

=> We added “modellers”

L26: ‘EKB’ - perhaps define this in line 16.

 => “The stability of the Ekman balance”

Introduction

L41: ‘For instance, simulations enable to …’ Enable who?

=> We added “researchers”

L56: remove comma after ‘domain size’

=> Done

L61: ‘Modelling phenomenology …’ – not really sure what this is; the study of phenomenons?

=> “Modeling phenomena affected by ambient winds in real events”

L208:replace  ‘latter’ with ‘later’

=> Done

L338-339: Any particular reason 40m height was chosen?

=> We added “This reference height was selected because it is high enough to undergo a limited influence of the canopy (more than 2h), while remaining theoretically measurable from the ground.”

L375: remove dash between 106-m

=> Done

L517-518: Think the caption may be wrong!

=> Yes. The reviewer is right. In fact this whole figure should not have been here (it was a copy of fig 7 and there were two figures 12). Sorry about that.

L526: ‘they enable to..’ – again, enable who?

=> We added “us”

Reviewer 2 Report

In this paper, the authors investigate the effect of different implementations of large-scale pressure gradient forcing on the flow behavior and statistics in Large-eddy Simulations for problems happening in the atmospheric boundary layer (with a specific focus on canopy-flow and fire-related simulations). The authors present a new technique to capture the effect of large-scale pressure gradients that sidesteps the practical limitations of the methods that use a constant pressure gradient or an Ekman balance. The performance of the four versions of the new technique is thoroughly investigated and compared against those of the simple gradient and Ekman balance methods. The authors show that the new method allows forcing of the velocity at a specific height, with two versions of the new method mitigating the streak locking features observed in simulations using a simple pressure gradient. The results seem to be satisfactory.

In general, I enjoyed reading the paper. Besides the value of the new techniques presented in this work, I believe, the details of the investigation shed new light on the consequences of the widely used constant-pressure-gradient forcing (as well as the Ekman balance) on flow features that will be useful for a broad range of LES for the atmospheric boundary layer problems.

I believe that the study is interesting to the fluid dynamics community and therefore is worthy of being published in this journal. I have no major concerns and recommend publication considering the following minor edits:

• In the introduction, the acronym EKB should be defined.
• Figures after Fig. 9 are partly numbered incorrectly
• Suggest adding a discussion relevant to the last figure or remove it
• Line 521: It is mentioned that a two-point correlation study was conducted, while that is not the case in the presented results.
• Line 535: This figure ‘is’…
• I believe the introduction can be improved to better cover the breadth of the study
• I understand that the authors’ focus was on studies relevant to wildfire simulations. However, I think (and this is just a suggestion that) a more general title can benefit those working on other problems relevant to the ABL.

Author Response

Dear Reviewer,

Please find our replies below in italic. We would like to thank you for your positive appraisal and careful reading of our manuscript (thanks for bugs in figures!), and for the specific comments which improved the manuscript. We followed all suggestions, as detailed below.

Best regards

F.

In this paper, the authors investigate the effect of different implementations of large-scale pressure gradient forcing on the flow behavior and statistics in Large-eddy Simulations for problems happening in the atmospheric boundary layer (with a specific focus on canopy-flow and fire-related simulations). The authors present a new technique to capture the effect of large-scale pressure gradients that sidesteps the practical limitations of the methods that use a constant pressure gradient or an Ekman balance. The performance of the four versions of the new technique is thoroughly investigated and compared against those of the simple gradient and Ekman balance methods. The authors show that the new method allows forcing of the velocity at a specific height, with two versions of the new method mitigating the streak locking features observed in simulations using a simple pressure gradient. The results seem to be satisfactory.

In general, I enjoyed reading the paper. Besides the value of the new techniques presented in this work, I believe, the details of the investigation shed new light on the consequences of the widely used constant-pressure-gradient forcing (as well as the Ekman balance) on flow features that will be useful for a broad range of LES for the atmospheric boundary layer problems.

I believe that the study is interesting to the fluid dynamics community and therefore is worthy of being published in this journal. I have no major concerns and recommend publication considering the following minor edits:

• In the introduction, the acronym EKB should be defined.

=> Here, you probably mean “in the abstract”, where EKB was used but not defined (and was replaced by “the Ekman balance” lign 26 for simplicity). In the introduction, the Ekman balance is explicitly defined lign 69.

• Figures after Fig. 9 are partly numbered incorrectly. Suggest adding a discussion relevant to the last figure or remove it

=> Sorry about this mistake that occurred when formatting the draft for MDPI format. We removed the last figure (that was just a copy of fig 7 that should not have been here) and changed the number in accordance with the text.

• Line 521: It is mentioned that a two-point correlation study was conducted, while that is not the case in the presented results.

=> Sorry about this other mistake. Two point correlations were carried out in a former version of the manuscript, but were finally not presented in the submitted as they did not provide valuable information in the present paper. We modify the text consequently.

• Line 535: This figure ‘is’…

=> This finding is…

• I believe the introduction can be improved to better cover the breadth of the study. I understand that the authors’ focus was on studies relevant to wildfire simulations. However, I think (and this is just a suggestion that) a more general title can benefit those working on other problems relevant to the ABL.

• We agree. In fact, the manuscript was submitted to a special issue about coupled-fire atmosphere, which is why we focused on fire applications. We modified the title to broaden the scope “Pressure-Gradient Forcings for Large-Eddy Simulations of Flows in the Lower Atmospheric Boundary Layer”

We also did several modification in the abstract. However, reading the intro, we find that it was relatively general. We simply removed the reference to “, including for wildfire simulations [2][3]” at the very beginning, included more explicitly wind energy in applications regarding the ability to specify a desired wind speed (“as in most wildfire or wind energy applications”), as well as a few other references to wildfires that were not needed.