Quarterly Journal Royal Meteorological Society, 125, 2181-2200, 1998

Predictive Skill of a NWP system in the Southern Lower Stratosphere

Darryn W. Waugh¹, Janice M. Sisson, and David J. Karoly
Cooperative Research Centre for Southern Hemisphere Meteorology, Monash University, Clayton, Australia

Abstract

The predictive skill of the Australian Bureau of Meteorology's Global Assimilation and Prediction (GASP) system in the southern lower stratosphere is examined using two different sets of diagnostics: (i) conventional verification statistics used in numerical weather prediction studies (namely, RMS error, anomaly correlation, and bias), and (ii) elliptical diagnostics of the polar vortex (defined using potential vorticity on isentropic surfaces). Both sets of diagnostics indicate the same variation in predictive skill for forecasts during October 1994. The stratospheric forecasts are a large improvement over persistence even at 7 days, with the performance at 7 days being comparable to that in the troposphere of 3 day forecasts. There is large daily variability in the forecast scores for 7-day forecasts, and the days with below-average scores occur when the flow (vortex) is rapidly changing. Examination of the differences in the elliptical diagnostics show that the forecast vortex is weaker, less disturbed (i.e., closer to the pole and less elongated), and rotates faster than the analysed vortex. Consistent with a weaker forecast vortex, the minimum polar temperature and maximum zonal wind are under-predicted in the forecasts. The verification statistics in the stratosphere have a large seasonal variation, although the variation is different for different statistics. The GASP RMS errors are largest (smallest) in late-spring (summer) whereas both the ratio of GASP to persistence RMS error and the anomaly correlation indicate that the performance relative to persistence is best (worst) in late-spring (summer).