Antarctic atmospheric Richardson number from radiosonde measurements and AMPS

Monitoring a wide range of atmospheric turbulence over the Antarctic continent is still tricky, while the atmospheric Richardson number (Ri; a valuable parameter which determines the possibility that turbulence could be triggered) is easier to obtain. The Antarctic atmospheric Ri, calculated from the potential temperature and wind speed, was investigated using the daily results from the radiosoundings and forecasts of the Antarctic Mesoscale Prediction System (AMPS). Radiosoundings for a year at three sites (McMurdo - MM, South Pole - SP, and Dome C - DC) were used to quantify the reliability of the AMPS forecasts. The AMPS-forecasted Ri can identify the main spatiotemporal characteristics of atmospheric turbulence over the Antarctic region. The correlation coefficients (Rxy) of log10(Ri) at McMurdo, the South Pole, and Dome C are 0.71, 0.59, and 0.53, respectively. The Ri was generally underestimated by the AMPS and the AMPS could better capture the trend of log10(Ri) at relatively unstable atmospheric conditions. The seasonal median of log10(Ri) along two vertical cross-sections of the AMPS forecasts are presented, and it shows some zones where atmospheric turbulence can be highly triggered in Antarctica. The Ri distributions appear to be reasonably correlated to some large-scale phenomena or local-scale dynamics (katabatic winds, polar vortices, convection, gravity wave, etc.) over the Antarctic plateau and surrounding ocean. Finally, the log10(Ri) at the planetary boundary layer height (PBLH) were calculated and their median value is 0.316. This median value, in turn, was used to estimate the PBLH and agrees well with the AMPS-forecasted PBLH (Rxy>0.69). Overall, our results suggest that the Ri estimated by AMPS are reasonable and the turbulence conditions in Antarctica are well revealed.