Analysis-error statistics are crucial to many aspects of predictability, including ensemble forecasting, targeted observations, and data assimilation. However, current predictability work rests upon assumptions of analysis-error statistics. In particular, the largest problem is the vertical profile of analysis errors. For example, studies often show that short-term forecasts are most sensitive to analysis errors in the lower troposphere. My recent findings suggest that errors in the middle and upper troposphere are more important.
I estimated the dominant vertical structures for analysis and forecast errors in midlatitudes using a small ensemble of operational analyses. The focus was the meridional wind and temperature at 40�N because they capture essential aspects of midlatitude baroclinic waves. For meridional wind, analysis errors exhibited upshear tilt, with peak amplitude just below the tropopause. The 12-72-h forecast errors and analyses shared a similar vertical structure. For temperature, analysis errors exhibited a sharp peak in the lower troposphere, with secondary structure near the tropopause. In contrast, forecast errors and analyses showed a dipole with a smooth vertical structure straddling the tropopause, consistent with the meridional wind results and with the idea that these features are related to potential vorticity anomalies due to variance in tropopause position.
Calculations show that forecast errors originate from analysis errors in the middle and upper troposphere, and then rapidly expand in the vertical to span the troposphere, with a peak at the tropopause. A linear regression of forecast errors onto analysis errors over the western North Pacific shows that analysis errors near the tropopause rapidly develop into a spreading wave packet, with a group speed that matches the mean zonal-wind speed of 31 m s^sup -1^ (see figure on previous page).
[Author Affiliation]
-GREGORY J. HAKIM (UNIVERSITY OF WASHINGTON). "Vertical Structure of Midlatitude Analysis and Forecast Errors," in the March Monthly Weather Review.

No comments:
Post a Comment