Diagnosis of numerical analyses and forecasts from the perspective of quasi-geostrophic dynamics

"The conventional statistical measures used by NMC for evaluating numerical analyses and forecasts provide little, if any, insight into the reasons and significance of the end result. Thus, for example, RMS errors and S1 scores convey virtually no information on the type and importance of differences between a forecast and verifying analysis. In contrast, the parameters generated by the oft used Collins and Miller (1975) energy code¹ (e.g., available potential energy) are physically meaningful and readily interpreted. They are extremely valuable, therefore, in diagnosis of such questions as the nature of systematic differences between analyses produced with and without remote sounding data. In that large area and volume averages are considered, however, the energy quantities computed with the Collins and Miller routine can not isolate the effects of individual meteorological features"--Introduction. M. Steven Tracton. "June 1978." "This is an unreviewed manuscript, primarily intended for informal exchange of information among NMC staff members." System requirements: Adobe Acrobat Reader. Includes bibliographical references (page 25). 1978 NWS (National Weather Service) Library Public Domain 1931