Ilaria Pietroni

Observed and Modelled Convective Mixing-Layer Height at Dome C, Antarctica

The mixing-layer height is estimated using measurements from a high resolution surface-layer sodar run at the French-Italian station of Concordia at Dome C, Antarctica during the summer 2011–2012. The temporal and spatial resolution of the sodar allows the monitoring of the mixing-layer evolution during the whole diurnal cycle, i.e. a very shallow nocturnal boundary layer followed by a typical daytime growth. The behaviour of the summer mixing-layer height, variable between about 10- and 300 m, is analyzed as a function of the mean and turbulent structure of the boundary layer. Focusing on convective cases only, the retrieved values are compared with those calculated using a one-dimensional prognostic equation. The role of subsidence is examined and discussed. We show that the agreement between modelled and experimental values significantly increases if the subsidence is not kept fixed during the day. A simple diagnostic equation, which depends on the time-averaged integral of the near-surface turbulent heat flux, the background static stability and the buoyancy parameter, is proposed and evaluated. The diagnostic relation performance is comparable to that of the more sophisticated prognostic model.

Use of a High-Resolution Sodar to Study Surface-layer Turbulence at Night

Measurements in the atmospheric surface layer are generally made with point sensors located in the first few tens of metres. In most cases, however, these measurements are not representative of the whole surface layer. Standard Doppler sodars allow a continuous display of the turbulent thermal structure and wind profiles in the boundary layer up to 1000 m, with a few points, if any, in the surface layer. To overcome these limitations a new sodar configuration is proposed that allows for a higher resolution in the surface layer. Because of its capabilities (echo recording starting at 2 m, echo intensity vertical resolution of approximately 2 m, temporal resolution of 1 s) this sodar is called the surface-layer mini-sodar (SLM-sodar). Features and capabilities of the SLM-sodar are described and compared with the sodar. The comparison of the thermal vertical structure given by the SLM-sodar and the sodar provides evidence that, in most cases, the surface layer presents a level of complexity comparable to that of the entire boundary layer. Considering its high vertical resolution, the SLM-sodar is a promising system for the study of the nocturnal surface layer. The nocturnal SLM-sodar measurements have shown that, depending on wind speed, the structure of the surface layer may change substantially within a short time period. At night, when the wind speed is greater than 3 m s−1, mechanical mixing destroys the wavy structure present in the nocturnal layer. Sonic anemometer measurements have shown that, in such cases, also the sensible heat flux varies with height, reaching a peak in correspondence with the wind speed peak. Under these conditions the assumption of horizontal homogeneity of the surface layer and the choice of the averaging time need to be carefully treated.

Wavy Vertical Motions in the ABL Observed by Sodar

Some characteristics of wavelike motions in the atmospheric boundary layer observed by sodar are considered. In an experiment carried out in February 1993 in Milan, Italy, Doppler sodar measurements were accompanied by in situ measurements of temperature and wind velocity vertical profiles using a tethered balloon up to 600 m. The oscillations of elevated wavy layers containing intense thermal turbulence, usually associated with temperature-inversion zones, were studied by using correlation and spectral analysis methods. The statistics of the occurrence of wavelike and temperature-inversion events are presented. The height distributions of Brunt–Vaisala frequency and wind shear and their correlation within elevated inversion layers were determined, with a strong correlation observed between the drift rate of the wavy layers and the vertical velocity measured by Doppler sodar inside these layers. Spectral analysis showed similarities regarding their frequency characteristics. The phase speed and propagation direction of waves were estimated from the time delay of the signals at three antennae to provide estimates of wavelength. Moreover, wavelengths were estimated from the intrinsic frequency obtained from sodar measurements of the Doppler vertical velocity and oscillations of wavy turbulent layers. The two wavelength estimates are in good agreement.

Observations of near surface wind speed, temperature and radiative budget at Dome C, Antarctic Plateau during 2005

Abstract The annual and diurnal behaviours of near surface wind speed, temperature, and the radiative budget at Concordia Station (Dome C) in different seasons are shown. The wind speed was lowest in summer when a daily cycle was also observed. The largest mean values were concurrent with boundary layer growth in the afternoon. In winter and spring the wind speed reached the highest mean values. Perturbations in the wind flow were due to warming events which occurred periodically at Dome C. The lowest temperatures were in April and at the end of August. The coreless winter behaviour was perturbed by warming events which in many cases produced an increase in temperature of c. 20°C. The average temperature profiles show permanent thermal inversion, with the exception of a few hours in the afternoons during the summer. The strongest ground-based thermal inversions were observed in the polar winter. The largest potential temperature gradients were limited to a 30–40 m deep layer close to the surface. The net radiation was negative almost all the time with the exception of the period from mid-December to mid-January.

Erratum to: Observed and Modelled Convective Mixing-Layer Height at Dome C, Antarctica

We have discovered a calculation error in the evaluation of the history scale Q represented by Eq. 3 (Sect. 4.2). We apologise for any confusion or inconvenience that may have arisen. The miscalculation does not affect the general performance of the diagnostic relation proposed in the same section, so that the conclusions remain almost unchanged. After the correction, the values of α (Eq. 4) and of the determination coefficient (R2) retrieved by the linear regression showed in Fig. 8 become 11.20±0.30 and 0.86, respectively. The corrected values of the parameters listed in the last column (Diagnost relation) of Table 2 are mae = 33, rmse = 47, F B = 0.19, and I oA = 0.76. Figures7 and 8 with the correct calculation of Q are shown below.