The atmospheric boundary layer over the Southern Ocean.

Abstract

The thermodynamic structure of the lower troposphere over the Southern Ocean is analysed by employing over 16 years of high resolution upper air soundings from Macquarie Island (54.62 S, 158.85 E). The soundings are analysed to develop an understanding of the structure of the boundary layer and wind shear occurring through the lower levels over this region, and to compare this to European Centre for Medium-Range Weather Forecasts (ECMWF) model level reanalysis data for the Year of Tropical Convection (YOTC). A multiple layered structure is commonly observed in the high resolution soundings, and is also observed in YOTC, but with a lower frequency. The climatological mean and variability of a number of variables are calculated for both data sets, which reveals that YOTC performs well, but has weaknesses in modelling the observed moisture and wind fields, particularly evident in wind shear profiles. A distinction between a number of boundary layer types is made, and the frequency with which they occur is quantified for both data sets. Proxy cloud fields are constructed for the two data sets, and these suggest that clouds are commonly observed in a region between the top of the boundary layer and a secondary temperature inversion. An examination of the wind shear across the cloud boundaries finds wind shear over cloud base occurs more frequently than cloud top, suggesting that the cloud fields are not embedded in a well-mixed boundary layer. Next, an attempt to generalise these results to a broad region across the Southern Ocean is made using the Constellation Observing System for Meteorology, Ionosphere, and climate (COSMIC) Global Positioning System (GPS) Radio Occultation (RO) observations. A direct comparison of temporally and spatially co-located COSMIC profiles and radiosonde profiles from Macquarie Island shows the performance of COSMIC is variable. COSMIC often struggles to reproduce the profiles over strong inversions in temperature or large changes in moisture, and the height and occurrence of the boundary layer, as well as any decoupled layers, rarely agree between the two data sets. A statistical analysis is performed on a large set of COSMIC profiles in the vicinity of Macquarie Island, and shows COSMIC reproduces the height of the boundary layer and decoupled layers well, but the frequency is underestimated. The analysis is extended to a large region of the Southern Ocean to investigate any spatial patterns in the height of the layers. There is a north-south gradient in the height of the boundary layer, as well as the frequency of occurrence of the boundary layer. Decoupling of the boundary layer is found to be a frequent and wide spread feature of the Southern Ocean. Next, the WRF model is used to simulate the boundary layer over the ocean south of Tasmania during a period of observed decoupling. A variety of model runs are performed to investigate the performance of three common PBL schemes. Other parameters investigated include increasing the vertical resolution, and the effect of varying the model initial conditions through 3D Variational data assimilation. An extensive model evaluation reveals the WRF model rarely captures main or secondary temperature inversions in the lowest few kilometers with the strength or frequency of in-situ observations. In reproducing the thermodynamics, the 3D Variational data assimilation model run performs the best. The YSU PBL scheme consistently performs best at simulating the Southern Ocean boundary layer and the associated clouds, and the benefit of improving the model initial conditions through data assimilation is significant. Finally, an analysis of trends in the wind over Macquarie Island is performed with a radiosonde database spanning nearly four decades. The results indicate that the surface wind speed is increasing, with the trend for the upper levels being less well defined. The surface wind is highly correlated with the upper level winds, and the wind at all levels are moderately correlated with the Southern Annular Mode. ECMWF ERA-Interim reanalysis data shows significant trends in wind speed over several levels, however slightly smaller than trends in the soundings over a similar time period. The correlations in ERA-Interim are similar to those in the soundings. A clustering analysis of the wind reveals four distinct regimes, with a trend towards a regime characterised by strong north westerly winds.