ASSESSMENT OF ERROR IN SYNOPTIC-SCALE DIAGNOSTICS DERIVED FROM WIND PROFILER AND RADIOSONDE NETWORK DATA

A topic of current practical interest is the accurate characterization of the synoptic-scale atmospheric state from wind profiler and radios onde network observations. The authors have examined several related a nd commonly applied objective analysis techniques for performing this characterization and considered their associated level of uncertainty both from a theoretical and a practical standpoint. A case study is pr esented where two wind profiler triangles with nearly identical centro ids and no common vertices produced strikingly different results durin g a 43-h period. It is concluded that the uncertainty in objectively a nalyzed quantities can easily be as large as the expected synoptic-sca le signal. In order to quantify the statistical precision of the algor ithms, the authors conducted a realistic observing system simulation e xperiment using output from a mesoscale model. A simple parameterizati on for estimating the uncertainty in horizontal gradient quantities in terms of known errors in the objectively analyzed wind components and temperature is developed from these results.