The influence of idealized surface heterogeneity on virtual  turbulent flux measurements

De Roo, Frederik; Mauder, Matthias

doi:https://doi.org/10.5194/acp-18-5059-2018

Articles | Volume 18, issue 7

https://doi.org/10.5194/acp-18-5059-2018

© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 3.0 License.

https://doi.org/10.5194/acp-18-5059-2018

© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 3.0 License.

Articles | Volume 18, issue 7

Research article

|

13 Apr 2018

Research article |

| 13 Apr 2018

The influence of idealized surface heterogeneity on virtual turbulent flux measurements

Frederik De Roo and Matthias Mauder

Download

Final revised paper (published on 13 Apr 2018)
Preprint (discussion started on 13 Sep 2017)

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

RC1: 'Review of "The influence of idealized surface heterogeneity on virtual turbulent flux measurements"', Anonymous Referee #1, 08 Oct 2017
- AC1: 'Reply to reviewer 1', Frederik De Roo, 30 Dec 2017
RC2: 'Review', Anonymous Referee #2, 13 Nov 2017
- AC2: 'Reply to reviewer 2', Frederik De Roo, 30 Dec 2017

Peer-review completion

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

AR by Frederik De Roo on behalf of the Authors (30 Dec 2017) Author's response Manuscript

ED: Referee Nomination & Report Request started (03 Jan 2018) by Heini Wernli

RR by Anonymous Referee #1 (15 Jan 2018)

RR by Anonymous Referee #2 (28 Jan 2018)

Suggestions for revision or reasons for rejection

The authors addressed many of my concerns, including a correction of an equation. However, the second read of the manuscript was not satisfying and I will try to give a constructive reasoning below.

Comments on the author's reply to the first revision:

Major comment 1. I commented that the authors should be careful when integrating of large volumes (here 50m height) and not considering the logarithmic temperature profile. They answered that the logarithmic profile only used as boundary condition and thus affects the region from 0 to 5m above ground. I disagree on this point. The logarithmic profile is supposed to be valid in the entire surface layer. With a boundary layer height of 1 km as in the present simulations, the surface layer extents up to say 100 m, whatever it assumed as boundary condition. Please show me the temperature profile between 0 m and 50 m to prove that its shape can be approximated by a linear function.

Major comment 2. The authors state that they performed homogeneous control runs, but actually, I do not see any data analysis for these runs. Hence, my comment is still not addressed. I still want to see that the residual is zero for this case! And if not, it might indicate that the time-averaging was not long enough (see below).

Minor comment 6 and 14. The authors state in their answer that the heterogeneity creates a non-zero friction velocity. Now, strictly speaking the friction velocity is a parameter based on MOST and which does not (strictly speaking) exist when the flow is heterogeneous. I am aware that it is used nevertheless and in LES models it is even used LOCALLY (another violation of MOST). Unfortunately, I have the feeling that the authors use these terms rather loosely and I would be happy to have a more precise use and/or additional comments in the manuscript. In summary: friction velocity is only defined properly when a) the flow is homogeneous and b) when it is defined based on a mean profile. It is zero for free convection. And yes, when it is measured it is > 0 because of the local shear created by thermals, but that is partly because of the violations explained above.

Language 10. I still do not like the sentence "we plot" because the plot was already made. You do not plot something while I am reading your manuscript. You can say "we show" or something similar instead.

Comments on the revised manuscript:

1. Eq. 1 and 2: "H" is used in both with different meaning. Even worse, in Eq. 2 it is not explained that is shall be the heat flux, so the reader is forced to believe it to be the Heaviside function.

2. P8 L19. So at which height is the flow sufficiently resolved? You use 50 m without justification.

3. Fig. 2: This data is at 50 m height, correct? This information is missing in the caption. Then, I am very much surprised that the homogeneous control run displays horizontal variations that large (0.5 m/s is not negligible). You report that you averaged over 4 h of time and ideally one expects zero variance in the resulting fields. The fact that there is so much variance suggests that the averaging time is way from being sufficiently long. This then affects all other runs as well and questions the results in general.

4. P13 L 28: "one magnitude smaller" than what?

5. General: you state you performed many simulations (288 in total), but I gather that the reader is only confronted with data from six runs. Is this correct or did I understand something wrong here? I was expecting at least some analysis for all runs otherwise you should drop them.

6. Again, I do not understand the biplots. You make some effort to describe them but I would have to read other literature to understand them. I do not think that this is educationally good. Maybe there are more capable readers than me, but if you want to reach a larger audience why not include a very short and easy tutorial on how to read these plots.

7. General: Maybe it's just me, but I find the manuscript hard to read. That is also partly because of all the terms like "energy balance ratio" and "available energy (scaled)". Why not use the formula terms in the plots instead?! It would make reading so much easier.

Hide

ED: Reconsider after major revisions (08 Feb 2018) by Heini Wernli

AR by Frederik De Roo on behalf of the Authors (01 Mar 2018) Manuscript

ED: Referee Nomination & Report Request started (06 Mar 2018) by Heini Wernli

RR by Anonymous Referee #1 (19 Mar 2018)

RR by Anonymous Referee #2 (25 Mar 2018)

ED: Publish subject to technical corrections (27 Mar 2018) by Heini Wernli

AR by Frederik De Roo on behalf of the Authors (29 Mar 2018) Manuscript

Short summary

We investigate the mismatch between incoming energy and the turbulent flux of sensible heat at the Earth's surface and how surface heterogeneity affects this imbalance. To resolve the turbulent fluxes we employ large-eddy simulations. We study terrain with different heterogeneity lengths and quantify the contributions of advection by the mean flow and horizontal flux-divergence in the surface energy budget. We find that the latter contributions depend on the scale of the heterogeneity length.