Characteristics of the near‐surface atmosphere over the Ross Ice Shelf, Antarctica

Abstract

Two years of data from a 30 m instrumented tower are used to characterize the near-surface atmospheric state over the Ross Ice Shelf, Antarctica. Stable stratification dominates the surface layer at this site, occurring 83% of the time. The strongest inversions occur for wind speeds less than 4 m s À1 and the inversion strength decreases rapidly as wind speed increases above 4 m s À1 . In summer unstable stratification occurs 50% of the time and unstable conditions are observed in every season. A novel aspect of this work is the use of an artificial neural network pattern identification technique, known as self-organizing maps, to objectively identify characteristic potential temperature profiles that span the range of profiles present in the 2 year study period. The self-organizing map clustering technique allows the more than 100,000 observed potential temperature profiles to be represented by just 30 patterns. The pattern-averaged winds show distinct and physically consistent relationships with the potential temperature profiles. The strongest winds occur for the nearly well mixed but slightly stable patterns and the weakest winds occur for the strongest inversion patterns. The weakest wind shear over the depth of the tower occurs for slightly unstable profiles and the largest wind shear occurs for moderately strong inversions. Pattern-averaged log wind profiles are consistent with theoretical expectations. The log wind profiles exhibit a kinked profile for the strongest inversion cases indicative of decoupling of the winds between the bottom and top of the tower. PUBLICATIONS

Modeling the Dynamics of the Atmospheric Boundary Layer Over the Antarctic Plateau With a General Circulation Model

“…In this case, the near‐surface temperature inversion may reach values up to 25 K between 10 m and the surface, as the coupling between the air and the snow is (essentially) of radiative origin. Riordan () (see also Hudson and Brandt, ) and Cassano et al () (see also Wille et al, ) identified similar wind speed thresholds that distinguish a SBL regime with strong near‐surface inversions from a regime with weak near‐surface inversions at the South Pole and over the Ross ice‐shelf, respectively. This suggests that the two‐regime behavior of the SBL may prevail over a large part of the Antarctic continent where the surface slope is weak (ice‐shelves, Plateau).…”

Properties and temporal variability of summertime temperature inversions over Dronning Maud Land, Antarctica

“…The statistics of summertime inversion properties over Dronning Maud Land, Antarctica, were qualitatively close to those reported in previous studies based on six Antarctic radiosonde stations during a 20 year period (Zhang et al ., ), radiosonde soundings over sea ice in the western Weddell Sea in autumn and early winter (Andreas et al ., ), microwave radiometer‐based inversion measurements at Dome C (Pietroni et al ., ), and data from a 30 m instrumented tower (Cassano et al ., ). In all these studies, including ours, the inversion occurrence was high.…”

Automated classification of simulated wind field patterns from multiphysics ensemble forecasts

“…Nigro et al used SOM to characterize seasonal variability of near‐surface wind speeds. Cassano et al used a similar approach to extract patterns of near‐surface temperature inversion and diagnosed their relationships with observed surface wind speeds and simulated sea‐level pressure. In a follow‐up work, Nigro et al systematically utilized the SOM approach to validate the WRF model‐based Antarctic Mesoscale Prediction System (AMPS) against the observational dataset.…”

“…Furthermore, the total number of final clusters (could be primes or nonprimes) are determined in a completely automated manner; no ad hoc prescription is needed. This automation makes the proposed methodology more advantageous compared with the previous application of SOMs to identify patterns in near‐surface or boundary layer wind data …”

“…This automation makes the proposed methodology more advantageous compared with the previous application of SOMs to identify patterns in near-surface or boundary layer wind data. [47][48][49][50] [Colour figure can be viewed at wileyonlinelibrary.com] [Colour figure can be viewed at wileyonlinelibrary.com] FIGURE 15 Frequency distribution of ASTD for the members of the high-shear supercluster FIGURE 16 Frequency distribution of ASTD for the members of the LLJ type A supercluster…”