the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 24 Jan 2023
A simple model to assess the impact of gravity waves on ice crystal populations in the tropical tropopause layer
Abstract. The role of gravity waves on microphysics of tropical cirrus clouds and air parcel dehydration was studied using the combination of Lagrangian observations of temperature fluctuations and a 1.5 dimension model. High frequency measurements during isopycnal balloon flights were used to resolve the gravity wave signals with periods ranging from a few days to 15 min. The detailed microphysical simulations with homogeneous freezing, sedimentation and a crude horizontal mixing represent the slow ascent of air parcels in the Tropical Tropopause Layer. A reference simulation describes the slow ascent of air parcels in the tropical tropopause layer, with nucleation occurring only below the cold point tropopause with a small ice crystals density. The inclusion of the gravity waves modifies drastically the low ice concentration vertical profile and weak dehydration found during the ascent alone, with the increased ice crystal number and size distribution agreeing better with observations. Numerous events of nucleation occur below and above the cold point tropopause, efficiently restoring the relative humidity over ice to equilibrium with respect to the background temperature, as well as increase the cloud fraction in the vicinity of the cold-point tropopause. The corresponding decrease in water vapor is estimated at 2 ppmv around the cold point tropopause.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(2933 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(2933 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2022-1444', Anonymous Referee #2, 15 Feb 2023
This manuscript describes a model of in-situ ice nucleation in cold cirrus clouds and applies that model to cirrus using temperature fluctuations from long-duration high-altitude balloon flights.
The manuscript is mostly correct but needs a change in emphasis to focus more on what is new rather than what is already well described in the literature. I also feel that the analysis stops just when it starts getting interesting.
Much of the manuscript is about comparing the case with temperature fluctuations with a case with constant, slow ascent. It is already well described in the literature that a constant, slow ascent is unrealistic. This manuscript cites some of the relevant literature for cirrus clouds. In addition to that, there is a similar literature for polar stratospheric clouds that goes even further back (for example, Bacmeister et al., 1999). This paper may not need to cite that PSC literature. Instead, I mention it to show just how well-known it is that temperature fluctuations control the formation of ice at low temperatures.
So what is new in this manuscript? Except for some fixable issues mentioned below, the model itself does not look that much better or worse than the models in, for example, the Jensen or Murphy papers that are already cited. There are new temperature histories based on balloon data, whereas much of the older literature used more synthetic temperature histories based on assumed gravity waves or power spectra.
What this manuscript needs to do in order to be novel is to compare not only to a constant cooling case (which should be a quick mention, not a major focus), but also to a variety of wave cases. Some questions are:
- How do the fluctuations from the balloon data compare with the earlier literature? Is there something special about the observed waves for ice nucleation and growth, or is any old fluctuation with a standard deviation of 1.5K food enough?
- Could you run your model with fluctuations from some of the earlier literature? That would separate the effect of temperature fluctuations from the effect of the details of the model.
- What happens with different amplitudes of waves?
- How do your “wave” cases compare to earlier literature with waves?
- Have you run just one model case, or have you run ensembles of your model using different subsets of the observed temperature fluctuations?
You don’t have to pursue all of these questions, but at I think that pursuing at least some of them is necessary to have strong paper.
Fixable model issues:
- The 15 meter vertical resolution is almost certainly too coarse. You should run with a finer resolution and see if the results change.
- I’m not sure that it is appropriate to filter the data to remove temperature fluctuations faster than 15 minutes. From Figure 5 an aerosol that freezes can significantly change its diameter in 15 minutes. Once the ice crystals are 5 or 10 microns fast fluctuations don’t matter much, but for the crucial growth from a freezing particle up to 1 or 2 microns fast fluctuations may matter.
You should run a case with faster fluctuations. From the manuscript line 84 the data may be valid to about 5 minutes – why not run a 5 minute filter and compare that to a 15 minute filter.
Specific comments:
- line 32: This isn’t stated quite right. It is the competition between homogeneous and heterogeneous freezing that is highly sensitive to cooling rate.
- line 109: The manuscript cites Jensen and Toon (1994) for the assumption of using a monomodal aerosol, but that paper was for constant cooling rates. A distribution of sizes can become more important when there are temperature fluctuations.
- line 362: The comment about reasons for dehydration above the tropopause is interesting.
- Figure 2: The cooling rate is only meaningful if the averaging time is specified. Given the power spectrum of atmospheric temperature fluctuations, shorter intervals have higher cooling rates. This figure cannot be interpreted without more information.
Citation: https://doi.org/10.5194/egusphere-2022-1444-RC1 - AC1: 'Reply on RC1', Milena Corcos, 03 May 2023
-
RC2: 'Comment on egusphere-2022-1444', Anonymous Referee #1, 02 Mar 2023
The authors describe a simple model with 1.5 dimensions to simulate the effect of a gravity wave on homogeneous nucleation of ice around the tropical cold point tropopause. They present their model, and discuss two contrasting scenarios, i.e. one where no gravity wave is included and a second with a passing gravity wave. In the latter case they observe a much more realistic-looking pattern of the ice crystals and they also compare that with in-situ measurements.
The manuscript describes an interesting approach and definitely fits into the scope of ACP, however I think it could be extended to yield an even stronger result.
General comments:- Although I like your strategy to present the model simulations, i.e. contrasting the case of no gravity wave with the case including a gravity wave, I think that the no-wave description could be significantly shortened since this case is well-known. In addition, you describe the "preliminary experiment" in section 3.2.1 but I see no reason why this is needed. What do we learn from that experiment? I suggest to remove that preliminary experiment or, if you prefer to keep it, move it to an appendix.
- In the discussion section you also address the dehydration of air parcels. I was pretty unsure if this discussion (and the figure 13) is only based on the single W experiment that you describe in prior sections or if it is based on a whole ensemble of simulations. In any case, I think that you should carry out more than one W experiment, i.e. a whole ensemble of simulations, and base the discussion on the results of the ensemble. To construct the ensemble you could play around with some initial conditions and (maybe more importantly) with the balloon-based timeseries of temperature perturbations. Using the ensemble, you could produce a "mean spatial ice crystal pattern" and illustrate that as in figures 8 and 9, a "mean ice crystal size distribution", a "mean ice supersaturation", and also a "mean dehydration".
- Description of the model physics in section 2.2.1: In each timestep, you need to (i) compute the sedimentation and (ii) apply the wind-shear module. Which order of these two steps do you use? Is there a dependency on the order of these two operations?
Specific comments:- Lines 128 - 131: What do you mean by "has moved to that immediately above or below"? In my understanding, the horizontal wind causes the given row to move horizontally (as depicted in figure 1), but how do you then rearrange the rows to have again all parcels (vertically) aligned (i.e. again arrive at the configuration of t_0 depicted in figure 1)?
- Lines 143 - 145: Given that there is no correlation between the timeseries of two neighboring parcels, do you observe any artifacts? Is there any information about the (de)correlation length of these perturbations?
- I would appreciate an indication of the cold-point tropopause in the figures that show a height-coordinate, i.e. figures 2, 3, 4, 8, 9
- Lines 186 - 189: Is this true? If the solution particles do not sediment out of the layer (as in your model) then homogeneous nucleation should always be possible given that the relative humidity is high enough.
- Line 206: Where can you see in figure 5 that the sedimentation timescale is 1/10hr? The diagram only extends to a radius of about 12 microns where the timescale is about 1/100hr?
- Line 207: I don't understand that sentence. After the nucleation the ice crystals virtually do not sediment. What exactly means "50h=450m"?
- Figures 4, 9: I would prefer having the number concentration indicated by colors since the mass is already related to the radius.
- Figure 5: In the figure you write "half life" and in the caption it is "half time".
- Lines 276 - 281: Note that the background aerosol is kept constant with height, hence nucleation is exclusively controlled by the competition of existing ice crystals for water vapor and the cooling from the gravity wave (+ascend).
Technical corrections:- Line 23: The bracket of the reference only encloses the year. There are further occurrences in the following lines.
- Line 60: It should read "homogeneous".
- Lines 63-66: I would prefer having the numbers of the section instead of, e.g., "the following section".
- Table 1: The row "Duration of simulation" has a missing entry in the column WFT.
- Line 289: It should read "zero" instead of "null".
- Line 406: I think "cloud condensation nuclei" should read "ice nuclei".
Citation: https://doi.org/10.5194/egusphere-2022-1444-RC2 - AC2: 'Reply on RC2', Milena Corcos, 03 May 2023
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2022-1444', Anonymous Referee #2, 15 Feb 2023
This manuscript describes a model of in-situ ice nucleation in cold cirrus clouds and applies that model to cirrus using temperature fluctuations from long-duration high-altitude balloon flights.
The manuscript is mostly correct but needs a change in emphasis to focus more on what is new rather than what is already well described in the literature. I also feel that the analysis stops just when it starts getting interesting.
Much of the manuscript is about comparing the case with temperature fluctuations with a case with constant, slow ascent. It is already well described in the literature that a constant, slow ascent is unrealistic. This manuscript cites some of the relevant literature for cirrus clouds. In addition to that, there is a similar literature for polar stratospheric clouds that goes even further back (for example, Bacmeister et al., 1999). This paper may not need to cite that PSC literature. Instead, I mention it to show just how well-known it is that temperature fluctuations control the formation of ice at low temperatures.
So what is new in this manuscript? Except for some fixable issues mentioned below, the model itself does not look that much better or worse than the models in, for example, the Jensen or Murphy papers that are already cited. There are new temperature histories based on balloon data, whereas much of the older literature used more synthetic temperature histories based on assumed gravity waves or power spectra.
What this manuscript needs to do in order to be novel is to compare not only to a constant cooling case (which should be a quick mention, not a major focus), but also to a variety of wave cases. Some questions are:
- How do the fluctuations from the balloon data compare with the earlier literature? Is there something special about the observed waves for ice nucleation and growth, or is any old fluctuation with a standard deviation of 1.5K food enough?
- Could you run your model with fluctuations from some of the earlier literature? That would separate the effect of temperature fluctuations from the effect of the details of the model.
- What happens with different amplitudes of waves?
- How do your “wave” cases compare to earlier literature with waves?
- Have you run just one model case, or have you run ensembles of your model using different subsets of the observed temperature fluctuations?
You don’t have to pursue all of these questions, but at I think that pursuing at least some of them is necessary to have strong paper.
Fixable model issues:
- The 15 meter vertical resolution is almost certainly too coarse. You should run with a finer resolution and see if the results change.
- I’m not sure that it is appropriate to filter the data to remove temperature fluctuations faster than 15 minutes. From Figure 5 an aerosol that freezes can significantly change its diameter in 15 minutes. Once the ice crystals are 5 or 10 microns fast fluctuations don’t matter much, but for the crucial growth from a freezing particle up to 1 or 2 microns fast fluctuations may matter.
You should run a case with faster fluctuations. From the manuscript line 84 the data may be valid to about 5 minutes – why not run a 5 minute filter and compare that to a 15 minute filter.
Specific comments:
- line 32: This isn’t stated quite right. It is the competition between homogeneous and heterogeneous freezing that is highly sensitive to cooling rate.
- line 109: The manuscript cites Jensen and Toon (1994) for the assumption of using a monomodal aerosol, but that paper was for constant cooling rates. A distribution of sizes can become more important when there are temperature fluctuations.
- line 362: The comment about reasons for dehydration above the tropopause is interesting.
- Figure 2: The cooling rate is only meaningful if the averaging time is specified. Given the power spectrum of atmospheric temperature fluctuations, shorter intervals have higher cooling rates. This figure cannot be interpreted without more information.
Citation: https://doi.org/10.5194/egusphere-2022-1444-RC1 - AC1: 'Reply on RC1', Milena Corcos, 03 May 2023
-
RC2: 'Comment on egusphere-2022-1444', Anonymous Referee #1, 02 Mar 2023
The authors describe a simple model with 1.5 dimensions to simulate the effect of a gravity wave on homogeneous nucleation of ice around the tropical cold point tropopause. They present their model, and discuss two contrasting scenarios, i.e. one where no gravity wave is included and a second with a passing gravity wave. In the latter case they observe a much more realistic-looking pattern of the ice crystals and they also compare that with in-situ measurements.
The manuscript describes an interesting approach and definitely fits into the scope of ACP, however I think it could be extended to yield an even stronger result.
General comments:- Although I like your strategy to present the model simulations, i.e. contrasting the case of no gravity wave with the case including a gravity wave, I think that the no-wave description could be significantly shortened since this case is well-known. In addition, you describe the "preliminary experiment" in section 3.2.1 but I see no reason why this is needed. What do we learn from that experiment? I suggest to remove that preliminary experiment or, if you prefer to keep it, move it to an appendix.
- In the discussion section you also address the dehydration of air parcels. I was pretty unsure if this discussion (and the figure 13) is only based on the single W experiment that you describe in prior sections or if it is based on a whole ensemble of simulations. In any case, I think that you should carry out more than one W experiment, i.e. a whole ensemble of simulations, and base the discussion on the results of the ensemble. To construct the ensemble you could play around with some initial conditions and (maybe more importantly) with the balloon-based timeseries of temperature perturbations. Using the ensemble, you could produce a "mean spatial ice crystal pattern" and illustrate that as in figures 8 and 9, a "mean ice crystal size distribution", a "mean ice supersaturation", and also a "mean dehydration".
- Description of the model physics in section 2.2.1: In each timestep, you need to (i) compute the sedimentation and (ii) apply the wind-shear module. Which order of these two steps do you use? Is there a dependency on the order of these two operations?
Specific comments:- Lines 128 - 131: What do you mean by "has moved to that immediately above or below"? In my understanding, the horizontal wind causes the given row to move horizontally (as depicted in figure 1), but how do you then rearrange the rows to have again all parcels (vertically) aligned (i.e. again arrive at the configuration of t_0 depicted in figure 1)?
- Lines 143 - 145: Given that there is no correlation between the timeseries of two neighboring parcels, do you observe any artifacts? Is there any information about the (de)correlation length of these perturbations?
- I would appreciate an indication of the cold-point tropopause in the figures that show a height-coordinate, i.e. figures 2, 3, 4, 8, 9
- Lines 186 - 189: Is this true? If the solution particles do not sediment out of the layer (as in your model) then homogeneous nucleation should always be possible given that the relative humidity is high enough.
- Line 206: Where can you see in figure 5 that the sedimentation timescale is 1/10hr? The diagram only extends to a radius of about 12 microns where the timescale is about 1/100hr?
- Line 207: I don't understand that sentence. After the nucleation the ice crystals virtually do not sediment. What exactly means "50h=450m"?
- Figures 4, 9: I would prefer having the number concentration indicated by colors since the mass is already related to the radius.
- Figure 5: In the figure you write "half life" and in the caption it is "half time".
- Lines 276 - 281: Note that the background aerosol is kept constant with height, hence nucleation is exclusively controlled by the competition of existing ice crystals for water vapor and the cooling from the gravity wave (+ascend).
Technical corrections:- Line 23: The bracket of the reference only encloses the year. There are further occurrences in the following lines.
- Line 60: It should read "homogeneous".
- Lines 63-66: I would prefer having the numbers of the section instead of, e.g., "the following section".
- Table 1: The row "Duration of simulation" has a missing entry in the column WFT.
- Line 289: It should read "zero" instead of "null".
- Line 406: I think "cloud condensation nuclei" should read "ice nuclei".
Citation: https://doi.org/10.5194/egusphere-2022-1444-RC2 - AC2: 'Reply on RC2', Milena Corcos, 03 May 2023
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 225 | 75 | 13 | 313 | 7 | 5 |
- HTML: 225
- PDF: 75
- XML: 13
- Total: 313
- BibTeX: 7
- EndNote: 5
Viewed (geographical distribution)
| Country | # | Views | % |
|---|---|---|---|
| United States of America | 1 | 118 | 35 |
| France | 2 | 54 | 16 |
| Germany | 3 | 42 | 12 |
| China | 4 | 22 | 6 |
| Russia | 5 | 8 | 2 |
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
- 118
Milena Corcos
Albert Hertzog
Riwal Plougonven
Aurélien Podglajen
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(2933 KB) - Metadata XML