Long-term tendencies in the MLT prevailing winds and tides over Antarctica as observed by radars at Molodezhnaya, Mawson and Davis

Long-term tendencies in horizontal neutral wind parameters in the southern polar mesosphere/lower thermosphere are presented. The wind data analyzed were obtained from meteor and MF radars situated at Molodezhnaya (45.9°E, 67.7°S), Mawson (62.9°E, 67.6°S) and Davis (78.0°E, 68.6°S). The composite da...

Full description

Bibliographic Details
Published in:Journal of Atmospheric and Solar-Terrestrial Physics
Main Authors: Merzlyakov, E., Murphy, D., Vincent, R., Portnyagin, Y.
Format: Article in Journal/Newspaper
Language:English
Published: Pergamon-Elsevier Science Ltd 2009
Subjects:
Online Access:http://hdl.handle.net/2440/52162
https://doi.org/10.1016/j.jastp.2008.09.024
Description
Summary:Long-term tendencies in horizontal neutral wind parameters in the southern polar mesosphere/lower thermosphere are presented. The wind data analyzed were obtained from meteor and MF radars situated at Molodezhnaya (45.9°E, 67.7°S), Mawson (62.9°E, 67.6°S) and Davis (78.0°E, 68.6°S). The composite dataset covers years from 1970 to 2006. A Bayesian approach in the form proposed by Wang and Zivot [2000. A Bayesian time series model of multiple structural changes in level, trend and variance. Journal of Business and Economic Statistics 8, 374-386] is used for the trend assessment. This approach allows structural breaks to be identified in the trend parameters (slope, mean or variance of residuals) or demonstrates their absence. The results of our analysis have shown persistence through such structural breaks in trends of the winter and summer prevailing winds and in meridional tidal components. It is demonstrated that the wind parameters exhibit different stable states with transitions between the states. Correlations between the southern polar MLT wind and indices of atmospheric variability (Northern annular mode, Southern annular mode, Multivariate El-Niño/Southern Oscillation Index) were then considered. The results show that statistically significant correlations exist during some periods of observations that do not exist during others. © 2008 Elsevier Ltd. All rights reserved. E.G. Merzlyakov, D.J. Murphy, R.A. Vincent and Yu.I. Portnyagin