3.1.1 Overview

The first ice fog event we focus on occurred between 7 and 9 March 2018. A picture showing the landscape of Dome C in the shallow and thin fog during sunset on 8 March 2018 is shown in Fig. 1. Figure 2 shows a time–height plot of the lidar backscattering signal during the event. A clear shallow fog layer starts to develop from about 06:00 LT on 8 March and grows from the ground until it reaches a maximum height of 200 m at 01:00 LT on 9 March. Its depth then sharply decreases until 06:00 LT, and the decay continues until dissipation at 00:00 LT on 10 March.

The radiosounding reveals a layer supersaturated with respect to ice from 40 to 150 m a.g.l. The maximum RHi value of 113 % is reached at 130 m a.g.l. which roughly corresponds to the upper limit of the fog layer and of the boundary-layer top inversion at this time. The temperature between the ground and 120 m ranges between 216 K at the bottom and 230 K at the top (Fig. 2b), which confirms that the fog develops in cirrus-like temperature conditions.

The 3 m wind slightly veers from a southwesterly to a southeasterly direction, and its magnitude increases throughout the event from 2 to 7.5 m s⁻¹. Libois et al. (2014) identify drifting snow events at Dome C when the 10 m wind speed exceeds 7 m s⁻¹. Assuming a logarithmic wind speed profile between the surface and 10 m and an aerodynamic roughness length value of 1 mm (Vignon et al., 2016), this corresponds to a 3 m wind speed threshold value of 3.5 m s⁻¹. Note that at 22:00 LT on 7 March and after 06:00 LT on 8 March, the 3 m wind speed exceeds the threshold, and some snow drift is therefore possible during those periods.

Figure 3(a) The 2 d back trajectories ending at 50 (blue) and 100 m a.g.l. (red) at the four grid points closest to Dome C. Panels (b) and (c) show the time series of the wind speed and origin at three levels on the mast. The dashed red line in (b) indicates the 3 m wind speed threshold value for snowdrift triggering (see details in the main text). The grey band in (c) delimits the wind direction range in which the flow can possibly be affected by station exhaust. Black arrows indicate the times at which the fog visually appears and disappears near the surface in the lidar measurements.

As the air is coming from the southern sector, we do not expect this fog to be associated with an oceanic intrusion. It rather corresponds to a cloud that forms locally or slightly upstream over the Antarctic Plateau. This is confirmed by the analysis of air mass back trajectories arriving at 50 and 100 m a.g.l. shown in Fig. 3a. As the temperature along the trajectories never exceeds 235 K (not shown), an advection of supercooled liquid droplets towards Dome C is unlikely, which confirms that the fog does not have a liquid origin.

Figure 4Time series of the temperature (dashed black line) and relative humidity with respect to ice (RHi, solid red line) and liquid water (RHl, solid blue line) at 3 (a), 18 (b) and 42 m a.g.l. (c). The red and blue shading shows the RH uncertainty ranges as estimated in Appendix A. The purple line shows the homogeneous freezing RHi threshold value of Koop et al. (2000). The 100 % RH value is highlighted with a solid black line. Vertical coloured bars indicate the times at which vertical profiles are analysed in Fig. 5. Black arrows indicate the times at which the fog visually appears and disappears near the surface in the lidar measurements.

Download

Temperature measurements on the mast remain below 235 K throughout the event, and one can notice a weak diurnal cycle at 3 m that is almost totally damped at 42 m (Fig. 4). A close investigation of the vertical profile of the potential temperature – very close to absolute temperature over the mast depth – reveals that the boundary layer transits between a weakly convective state during daytime to a stable state during nighttime. The fog does not have a clear radiative signature at the surface even though a slight decrease in downward longwave radiation is noticeable during the evening of 7 March (Fig. B2a), which concurs with the overall decrease in air temperature (Fig. 4).

3.1.2 Fog initiation: 15:00 LT, 7 March, to 03:00 LT, 8 March

We now focus on the initiation period from 15:00 LT on 7 March to 03:00 LT on 8 March. At the 42 m level, RHi first exhibits an increase from 106 % to 143 % which coincides with an increase in vapour partial pressure. At the 3 m level, a later but sharper increase in RHi up to 139 % at 22:00 LT on 7 March is noticeable, and unlike at 42 and 18 m, this increase is partly explained by a decrease in temperature during the day–night transition. The maximum RHi reached at the three levels ranges between 139 % and 143 %. RHl remains lower than 78 % at 3 m, but it exceeds 90 % at 42 m at 23:00 LT on 7 March.

RHi then gradually decreases and reaches 100 % at the three levels at 08:00 LT on 8 March (Fig. 4). The decrease starts at 22:30 LT on 7 March at the 3 m level, a half hour later at 18 m and 2 h later at 42 m.

Figure 5Vertical profiles of temperature (a), relative humidity with respect to ice (b) and vapour partial pressure (c) at different times during the fog development phase.

Download

Given the maximum RHl and RHi values attained, the aerosol deliquescence and solution droplet freezing at 3 m a.g.l. are not very likely, but their occurrence cannot be completely excluded since Baumgartner et al. (2022) show that homogeneous freezing can start at RHi values slightly lower than the Koop et al. (2000) threshold. As the station exhaust does not affect our measurements (see grey shading in Fig. 3c), the decrease in 3 m relative humidity from 22:00 LT can therefore be explained by (i) ice nucleation and vapour deposition onto newly formed ice crystals, by (ii) a net downward turbulent moisture flux towards the ground surface as the vertical gradient of vapour partial pressure is positive (see Fig. 5c), and/or by (iii) the vapour deposition of drifting ice crystals in the surface layer, although the erosion threshold is barely attained.

At 18 m, the decrease in RHi from 22:00 LT is co-incidental with a sharp decrease in vapour partial pressure, as well as a decrease in gradient of vapour partial pressure (Fig. 5c) between 3 and 18 m. The 18 m drying from 22:00 LT is therefore due to the turbulent mixing near the surface with a net flux of a moisture oriented downward.

At 42 m, a similar decrease in relative humidity and temperature is observed, but the particularity of this level, and probably that of the overlying air layers, is that RHi gets very close to the Koop et al. (2000) RHi threshold for homogeneous freezing of solution droplets in the middle of the night (see time series in Fig. 4c and how the system evolves in a RH–temperature space in Fig. B1). The difference between the RHi value and the homogeneous freezing threshold is even smaller than the measurement uncertainty (red shading in Fig. 4c). We can therefore assume that the ice fog forms a few tens of metres above the ground surface through homogeneous nucleation, and the subsequent decrease in RHi at 42 m is therefore partly explained by the vapour deposition onto newly formed ice crystals. A few patches of ice fogs are indeed noticeable during the night between 7 and 8 March (Fig. 2a).

3.1.3 Growth and decay: 03:00 LT, 8 March, to 23:00 LT, 9 March

From 03:00 to 08:00 LT on 8 March, the temperature vertical profile shows a clear night–day transition, i.e. a transition from an “exponential” shape characteristic of very stable boundary layers to a convective profile with a well-mixed layer capped by a shallow temperature inversion whose height further increases during the day (Fig. 5a). A close inspection of the vertical profile of specific humidity between 3 m and the surface at 08:00 LT (not shown) – assuming that the specific humidity at the snow surface equals the saturation-specific humidity at the surface temperature – reveals that the vertical gradient of specific humidity and subsequently the surface flux of water vapour reverse sign and become oriented upward. The supply of water vapour from the snow surface – and possibly of drifting ice crystals since the surface wind speed exceeds the erosion threshold (Fig. 3b) – can therefore deposit onto the ice crystal embryos nucleated in the early morning and make them grow and become visible in the lidar signal.

Figure 2a shows that the depth of the fog layer gradually increases from 06:00 LT on 8 March up to about 80 m at 18:00 LT on 8 March as the daytime convective boundary layer deepens in $\approx \sqrt{t}$ (Stull, 1990). The growth of the fog is possible in the higher part of the boundary layer as its top is supersaturated with respect to ice (Fig. 2c). Ice crystals can hence grow by vapour deposition and sediment down to the near-surface layers where they probably partly sublimate (Figs. 2 and 4). Concurring with Genthon et al. (2022b) (see their Fig. 8), the near-surface air becomes subsaturated with respect to ice particularly during daytime when the near-surface air warms by convective mixing. As night falls around 19:00 LT on 8 March, one can then note a sharper deepening of the fog up to 200 m. The mechanisms responsible for this sudden growth cannot be directly identified from our measurements, but one can nonetheless make some assumptions. As the wind speed increases in the late afternoon (Fig. 3b), one could expect an enhanced vertical mixing of ice particles, but a close examination of the wind speed profile from the radiosonde reveals a weak local wind shear at the top of the boundary layer. Nonetheless, in the late afternoon, the decrease in the shallow convection intensity leads to a decrease in the capping temperature inversion strength (Ricaud et al., 2012). The weak vertical gradient of temperature at the top of the evening convective boundary layer is visible in Fig. 2b (see black arrow), whereas a second higher and stronger temperature inversion is visible at the top of the fog layer and probably induced by the cloud-top radiative cooling. Izett and van de Wiel (2020) show that the growth of a radiative liquid fog layer can suddenly accelerate when the vertical gradient of temperature decreases (i.e. when the vertical gradient of saturation-specific humidity decreases; see their Eq. 7). If we assume a homogeneous freezing process, the ice nucleation occurs at the Koop et al. (2000) threshold; that is, according to an increasing function of temperature. Subsequently, we can reasonably think that the conceptual model of Izett and van de Wiel (2020) is qualitatively valid for the present ice fog event. We can therefore assume that during the evening of 8 March, the ice fog grows radiatively, and the depth of the layer sharply increases due to the decrease in the temperature vertical gradient. It is worth noting that the growth rate of the fog layer (about 120 m in 7 h from visual inspection in Fig. 2a, which is ≈17 m h⁻¹) is quite similar to the growth rate of the radiative liquid fog portrayed in Izett and van de Wiel (2020) during the acceleration phase.

The fog grows until it reaches the top of the atmospheric layer supersaturated with respect to ice. From 18:00 LT on 8 March to 06:00 LT on 9 March, the RHi at the three levels is ≤100 % (Fig. 4) – in agreement with the radiosounding shown in Fig. 2c – while the lidar detects ice condensates from the surface up to 200 m a.g.l. The top height of the fog layer then drops down to about 50 m at 07:00 LT on 9 March and then slightly increases back at a time which corresponds to the deepening of the shallow-convective diurnal boundary layer. The fog layer then becomes shallower and shallower from 15:00 LT on 9 March as the turbulent mixing in the boundary layer weakens. The fog vanishes during the evening of 9 March, and RHi and temperature at 42 m increase up to their values before fog formation.

3.2.1 Overview

The second case study occurred between 25 and 27 August 2018. No picture is available since the sun is almost always below the horizon during this period of the year. At 15:00 LT on 25 August a very shallow – about 20 m deep – ice fog starts to be visible in the lidar profiles (Fig. 6a). Its depth then oscillates between 25 and 50 m a.g.l., and the fog vanishes at around 14:00 LT on 26 August. This event is particularly interesting since its vertical extent is almost the same as the one of the meteorological mast. Note that highly reflective bands are observed during 26 August, and most of them correspond to diamond dust precipitation “peaks” falling from the mid-troposphere. The radiosounding reveals a supersaturated layer with respect to ice below 30 m a.g.l. with a maximum RHi value of 147 %. The relative humidity with respect to liquid water also exceeds 75 % near the surface (Fig. 6c).

Figure 7Same as Fig. 3 but for Event 2.

Fig. 7b shows that the 3 m wind speed remains below 4 m s⁻¹ throughout the event and never exceeds the threshold value to trigger drifting snow. Its direction slightly changes from southwesterly to a southeasterly, suggesting that the air parcels reaching Dome C come from the high Antarctic Plateau (Fig. 7b) and that our measurements are not affected by station exhaust except during the very end of the event (see grey shading in Fig. 7c). The back trajectories shown in Fig. 7a confirm the continental origin of the flow, and as the temperature measured at the station is much lower than 235 K, a liquid origin of the fog can be ruled out. The time series of observational data along the mast reveal a near-constant 3 m temperature of about 200 K during the event and confirm the absence of a diurnal cycle during this period of the year.

Figure 8Same as Fig. 4 but for Event 2.

Download

Figure 9Same as Fig. 5 but for Event 2.

Download

3.2.2 Fog initiation: 06:00 to 15:00 LT, 25 August

A significant drop in temperature at 09:00 LT is observed at 18 m. This cooling is due to an abrupt transition between a very stable to a weakly stable dynamical regime of the boundary layer (van de Wiel et al., 2017; Baas et al., 2019; van der Linden et al., 2019) driven by an increase in the synoptic wind speed (note that the wind at 42 m, i.e. at the top or above the shallow boundary layer, is a reasonable proxy of the synoptic wind; see Fig. 8c). Subsequently the temperature profile evolves from a typical “exponential” shape to a concave shape (Vignon et al., 2017) owing to a more vigorous turbulent transport of heat towards the surface (Fig. 9). At 18 m the cooling rate is particularly intense and equals 5 K h⁻¹ between 08:00 and 11:00 LT on 25 August. As an analogy with typical air-parcel cooling within updraft in the middle and high troposphere, such a cooling rate roughly corresponds to an adiabatic air ascent of 0.10 m s⁻¹.

At 42 m, RHi approaches the Koop et al. (2000) threshold between 04:00 and 05:00 LT on 25 August, and some preliminary crystal nucleation can already occur at this time. However at 09:00 LT on 25 August RHi at 18 m suddenly increases as temperature drops and clearly exceeds the Koop et al. (2000) homogeneous freezing threshold. RHl exceeds 90 % at the same time (see Figs. 8b and B1). The relative humidity does not substantially change at the two other levels at this time, leading to a well-marked maximum in the vertical profile (Fig. 9b). It is worth mentioning that Genthon et al. (2022b) show that the climatological RHi along the mast also exhibits a clear maximum at 18 m due to the non-linear dependence of relative humidity on temperature and atmospheric water content. Therefore, the initiation of the second fog event seems mainly due to a homogeneous freezing process starting between 3 and 42 m a.g.l. around 09:00 LT on 25 August. The density and size of ice crystals become sufficiently large to be well visible in the lidar signal a few hours later. From 10:00 LT on 25 August, RHi at 18 m starts a pronounced decrease due to the vapour deposition onto newly formed crystals and to a downward turbulent flux of water vapour as the vertical gradient of vapour partial pressure between 3 and 18 m is positive (Fig. 9c).

Figure 10Conceptual scheme of the formation and growth of the two ice fogs studied in the paper.

Download

3.2.3 Growth and decay: 15:00 LT, 25 August, to 15:00 LT, 26 August

The decrease in RHi at 42 m occurs at 15:00 LT, and it coincides with a decrease in temperature associated with the deepening of the weakly stable boundary layer (Figs. 8c and 9a). The decrease in the 42 m RHi also concurs with the increase in fog layer depth visible in the lidar data between 15:00 and 18:00 LT on 25 August (Fig. 6a) and can therefore be attributed to the deposition of water vapour onto ice crystals. The saturation is reached at 19:00 LT on 25 August.

From the evening of 25 August, RHi remains close to but slightly below saturation at 3 and 18 m probably owing to a net flux of vapour towards the surface in the very shallow boundary layer. Ice crystals at these two heights can therefore not grow very close to the surface, but their detection at 18 m by the lidar may be rather explained by some sedimentation from higher layers.

At 12:00 LT on 26 August, the fog starts to dissipate from the top, and the air at 42 m becomes supersaturated with respect to ice (Fig. 8c). A few hours later a decrease in RHi back to saturation occurs probably due to vapour deposition onto the diamond dust precipitation streaks that suddenly fall down from the low and middle troposphere (Fig. 6a).