the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 07 May 2024
Vertical Profiles of Liquid Water Content in fog layers during the SOFOG3D experiment
Abstract. Observations collected during the SOuth west FOGs 3D experiment for processes study (SOFOG3D) field campaign are examined to document vertical profile of microphysical and thermodynamic properties of fog layers. In situ measurements collected under a tethered balloon provide 140 vertical profiles of liquid water content (LWC) from an adapted cloud droplet probe (CDP), which allow an exhaustive analysis of the life cycle of 8 thin fogs (thickness < 50 m) and 4 developed layers. We estimate thin-to-thick transition time from remote sensing instruments (microwave radiometer and Doppler cloud radar) and surface measurements, by using thresholds for longwave radiation flux, turbulent kinetic energy, vertical temperature gradient, fog top height and liquid water path (LWP) values. We found that a LWP threshold value of 15 g.m−2 is more suited for the thick fogs sampled at the super-site. CDP data are used to compute the equivalent fog adiabaticity from closure (αclosureeq) and compare to value derived from remote sensing instruments, 2-m height visibility, and an one-column conceptual model of adiabatic continental fog assuming that LWC linearly increases with height. The comparison of αclosureeq shows a large variability that results mainly from the parameterization used to estimate LWC at ground, but their evolution as a function of the fog thickness follows the same trend. We found larger negative values of αclosureeq for thin layers, associated to low LWP values. CDP data reveal that reverse trend of LWC profile (LWC being maximal at the ground and decreasing with altitude) are ubiquitous in optically thin fogs, while quasi-adiabatic features with increasing LWC values with altitude are mainly observed in well-mixed optically thick fogs. We investigate the actual fog adiabaticity and lapse rate fraction by using linear regressions to best fit the vertical profiles of LWC and temperature, respectively. This analysis highlights that reverse LWC profiles, when stable temperature conditions exist during the optically thin phase of fogs, evolve towards quasi-adiabatic features with slightly unstable temperature lapse rate, when fogs become optically thick. We also found that LWC at ground is higher during the thin phase and significantly decreases as the profile is changing from reverse to increasing with height. But this trend could be balanced when collision-coalescence and sedimentation processes redistribute the LWC through the fog layer from the top to the ground. This study provides new insights on the evolution of LWC profile during the fog life cycle, that would help to constrain numerical simulations.
- Preprint
(7624 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2024-1344', Anonymous Referee #1, 05 Jun 2024
Review of ‘Vertical profiles of liquid water content in fog layers during the SOFOG3D experiment’.
5th June 2024
Background
This paper presents results from a facet of the SOFOG3D campaign examining vertical profiles of fog microphysical properties in a number of fog cases. The report concentrates mainly on the adiabaticity of fog, but also demonstrates how the vertical profile of liquid water content differs between shallow and deeper fogs. Very few in situ observations have been made at elevated levels (>50m) within fog, particularly microphysical ones. From this perspective the paper is a welcome contribution to the field.
Review
I found the paper generally well written and suitably laid out. There is a lot of information presented, and a significant number of figures, including those in the appendices. I did not find this a problem but there may be scope to shorten the paper a little, starting with the appendices.
I consider the paper suitable for publication after minor corrections and alterations, but ask the authors to give my main comment serious consideration as I believe that the paper could be significantly improved from one that just reports on the morphology of fog, to one that provides new insight into fog dynamics.
Main comment:
A major element of the paper is to report on the adiabaticity of the observed fog cases. Whilst some background on this parameter is given, I feel the presentation lacks focus and justification as to why studying adiabaticity is important. For example, what new insights into the physics of fog can we learn, and how can this be promulgated into improvements in NWP of fog?
I felt that the choice to reject data at cloud top and only evaluate adiabaticity over the lower to mid-fog region is a little arbitrary, and means that the results are not quantitative, and for indication only. I don’t believe that the method used invalidates the conclusions made, but consider that a better analysis method could have provided more significant results.
For example, the authors may have been able to study the main process creating sub-adiabatic liquid water profiles within adiabatic fog (rather than simply reporting on levels of adiabaticity). This is an important process that has great relevance to NWP of fog since it partly controls the amounts of liquid water in the fog. Clearly, entrainment at fog top is an important process in this regard. Whilst this is mentioned in the paper it is not studied and no attempt to quantify it is made. I suggest that, if possible, the authors consider extending their study to assess how the levels of relative humidity, wind shear and turbulence at fog top, influence the levels of adiabaticity within adiabatic fog.
Other comments:
- The term, ‘Condensation rate’, is used throughout the document but I am wondering if it is the most appropriate term since it implies dql/dt, whereas what is being talked about seems to be better described as the ‘condensation amount’.
- Line 111, ‘SW to NE’ : in which sense, clockwise or anticlockwise?
- Page 6. Table 1. The error quoted for the TKE appears to be an instrumental one. However, would you not expect the random error to be more like 20%?
- Figure 3 caption. Not all the lines and features of the figure are explained in the caption. Please also explain what is meant by ‘local values’.
- Lines 365-368. It appears to be assumed here that it is the Gultepe inversion that is inaccurate and that the measured LWC0 is accurate. However, there is no discussion of the random errors in the measured LWC0 to justify this assertion, or those from the visibility measurement used in the Gultepe inversion. I expect that LWC0 is measured over a relatively shot time scale (not specified, but perhaps a few seconds?). We know from many surface-based observations of fog water (using similar devices to the one used here), that there is rapid variation in Liquid water content on timescales of seconds to minutes. Therefore, a short sampling period will be prone to larger random errors, meaning that the assumption here that the measurement is ‘accurate’ may be erroneous and in fact the Gultepe inversion may not be as inaccurate as assumed here. I think that this conclusion should be re-evaluated once random errors for the measured LWC0 and visibility has been provided. Also, in this regard, when short time scales are used, it is important that the two instruments are closely co-located to meaningfully compare the results between them.
- Line 387, you mean ‘negative’?
- Line 408. The averaging time for the CDP for each reported sample can be set manually. What was it set to?
- Line 446. Units for lapse rate?
- Figure 8 caption. Upward and downward radiation explanation is incorrect.
- Line 491. Suggest not using ‘supersite’ and sticking to Jachere or Charbonniere.
- Line 522. Any suggestions as to why the actual evolution did not match that expected?
- Line 528. I can’t see the upper liquid layer mentioned in fig. 11c. Could you make this clear?
- Line 635. Why should geometric depth be more important than LWP as a driver for the transition to adiabatic fog?
- Line 650. See point 5.
- Line 670. Did you examine the wind and temperature profile in detail for this case? It may be possible that differential advection had mixed two different fog regimes, one over the other, which might explain your observations.
Citation: https://doi.org/10.5194/egusphere-2024-1344-RC1 -
RC2: 'Comment on egusphere-2024-1344', Anonymous Referee #2, 10 Jun 2024
Review of “Vertical Profiles of Liquid Water Content in fog layers during the SOFOG3D experiment” by Costabloz et al. submitted to Atmospheric Physics and Chemistry
This paper studies vertical profiles of liquid water content (LWC) observed in fog layers during a field campaign carried out in southwestern France. The twelve fog events sampled during the campaign concern “thin” fogs occurring in stable conditions, characterized by decreasing LWC and increasing temperature as a function of altitude and vertically developed (or “thick”) adiabatic fog, characterized by increasing LWC and decreasing temperature as a function of altitude.
The work is original in that it relies on a comprehensive set of in-situ measurements of fog microphysics, complemented by remote sensing using a cloud radar and microwave radiometer.
The paper includes several interesting investigations that make it worthwhile to consider this paper for publication in Atmospheric Physics and Chemistry: Vertical profiles of LWC (measured by a cloud droplet probe, CDP) and temperature are analysed at different stages of the fog life cycle. The conditions that lead to thin-to-thick transition are investigated for four fog events using remote sensing measurements, surface measurements and in-situ vertical profiles. This is important to evaluate how the representation of transition can be improved in model simulations. Further, this analysis is compared to a conceptual model of fog adiabaticity proposed and exploited earlier in the literature (Toledo et al. 2021 and Dione et al. 2023) and provides a detailed assessment of the performance of this conceptual model.
The paper is clearly written and well organized. However, two topics require thorough attention and revisions, while others are more minor.
Major comment 1: derivation and uncertainty of fog adiabaticity.
Fog adiabaticity is studied extensively in the paper, both from in-situ LWC measurements and using the conceptual model proposed by Toledo et (2021). The paper presents several methods to derive fog adiabaticity.
- Equivalent adiabaticity from Closure (Eq. 5), where input variables are derived from remote sensing measurements and horizontal visibility
- Equivalent adiabaticity from Closure (Eq. 5), where input variables are derived from in-situ CDP measurements
- Adiabaticity from fitting LWC profiles from the ground to just below the diluted layer.
Section 3.2 and 3.3 are dedicated to comparing the first two definitions,
The balloon ascent (that carries the CDP sonde) takes 15-40 minutes to cover the vertical extent of the fog layer, depending on its depth. In the equivalent adiabaticity formulation (Eq. 5), there is an inherent hypothesis that the LWC0 measured at the surface is consistent with the integral of LWC (the liquid water path, LWP) and the fog layer depth (provided by CTH). We know that CTH, LWP and LWC_0 can vary significantly in 20-30 minutes.
Q1.1 Hence what is the uncertainty in deriving equivalent adiabaticity (definition 2) from CDP measurements given the temporal variations of LWC during the ascent or descent of the balloon?
Q1.2 What temporal smoothing or averaging is used for CTH, LWP and visibility to derive equivalent adiabaticity (definition 1).
Q1.3 What is the impact of these uncertainties on the comparisons of the two alpha-closures made in figure 4?
Q1.4 Line 310-311: the alpha-closure derived for IOP11 (22:09 to 22:23) is 0.53, while the same alpha-closure derived by Dione et al; (2023) ranges -0.9 to -1.1. Can you explain this discrepancy?
Q1.5 Line 310-311: the alpha-closure derived for IOP14 (06:11 to 06:47) is 0.45, while the same alpha-closure derived by Dione et al; (2023) is 0.6. Can you explain this discrepancy?
Q1.6 What is the impact LWC0 measurement variability (from CDP and Visibility) on the results presented in Figure 5?
Section 3.4 compares adiabaticity (definition 3) with equivalent adiabaticity (definition 2).
Alpha (definition 3) does not consider the fog layer at the top that is diluted by entrainment of dry air, hence it is expected to be larger than the equivalent alpha (definition 2) that accounts for the entire for LWP and depth.
Q1.7 What is the impact of the variability in the entrainment layer (depth, shape of LWC profile in the layer) on the comparison shown in Fig. 7 ?
Major comment 2: thin-to-thick fog transition
In Section 2.3, the thin-to-thick fog transition is presented as a time when the transition occurs. And thresholds are defined for five variables to identify the transition time linked to different processes that affect the transition: Net LW radiation, temperature gradient between 50 and 25m, TKE, CTH and LWP. A multi-parameter evaluation of the transition is interesting. However, the transition from thin to thick fog should rather be defined as a time interval with a beginning and an end (as proposed by Dione et al. 2023 using three parameters), rather than as an instant in time. Dione et al. (2023) shows that the duration of the transition phase is variable from one fog event to another.
Here, you propose the transition phase duration to be time interval between the first and the last thresholds of the five variables (L213). Later you mention “scattering” between thresholds (L216) and “period of uncertainty” (L324).
Q2.1 I suggest that you revise the definitions of thin-to-thick transition to include both thresholds for multiple variables, and time of onset of transition, and time of ending of the transition when the fog has reached an adiabatic state. This would allow you to evaluate if multiple thresholds are reach in a short amount of time favours a rapid transition from thin to thick, and reversely if a slow transition can be caused by inconsistencies in the different processes involved in the transition.
Q2.2 Could you propose a more thorough definition of thick-to-thin transition duration and explanation of the duration, based on the values reached by the five variables compared to the threshold that you identified.
L470 (Section 4.2.1) Explain how the time interval between first and last threshold can be compared with the transition phase duration of Dione et al. (2023) for IOP14
L512 (Section 4.2.2) You write “Surprisingly, reverse LWC trends were observed for profiles # 6 and # 7 with α <0 even though the thin-to-thick transition had already occurred one hour earlier.”
In fact, this is not surprising for a slow transition. The fact that thresholds are reached does not mean that transition is over and that fog has become adiabatic.
Discuss how the LWC profiles temporal evolution that you show in can explain the slow transition.
L637, 638 “uncertainty periods” is used again for transition phase duration. Please revise.
Minor comments.
Section 2.1. Check consistency in the tense used throughout this section (past, present, future).
L106 Change “spread” to “distributed”
L119: CBH cannot be derived unambiguously with a cloud radar alone as the signal in the lower part of the cloud can come from cloud droplets or from drizzle below the cloud. CBH is usually derived from a ceilometer.
L129 “However, we analyse here all the data collected during the SOFOG3D experiment, and we then use independant retrieval.” Not clear, please rephrase.
“independant" à “independent”
L137 “aspirates” à “sucks in”
Table 2. What objective criteria do you do define “radiative” vs “radiative-advective” fog types?
L203 “These discrepancies may be explained by the contrasting environment between the two measurement areas (Thornton et al., 2023).” Explain what contrasts you are referring to ?
L307 “These results hightlight that while the adiabatic model correctly represents thermodynamical and microphysical properties of well-mixed fogs, it does not represent the properties of optically thin fogs at all.” Does anyone expect the adiabatic model to correctly represent optically thin fog ?
“hightlight” à “highlight”
L378 “significant values” à “large values”
L393 “This attests that differences observed from our dataset result from the actual properties of the fog sampled during SOFOG3D and not from the measurements used (in situ or remote sensing) to compute the fog adiabaticity from closure.” Not clear, please rephrase or explain.
L634 Replace “if” by “while”.
L703 “period of uncertainty” please rephrase according to Major comment 2 discussions.
L703 “more suited …” than what ?
L711 “ larger negative values …” than what ?
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 287 | 136 | 37 | 460 | 32 | 33 |
- HTML: 287
- PDF: 136
- XML: 37
- Total: 460
- BibTeX: 32
- EndNote: 33
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1