Assessment of Error in Synoptic-Scale Diagnostics Derived from Wind Profiler and Radiosonde Network Data

Gerald G. Mace Department of Meteorology, The Pennsylvania State University, University Park

Search for other papers by Gerald G. Mace in
Current site
Google Scholar
PubMed
Close
and
Thomas P. Ackerman Department of Meteorology, The Pennsylvania State University, University Park

Search for other papers by Thomas P. Ackerman in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

A topic of current practical interest is the accurate characterization of the synoptic-scale atmospheric state from wind profiler and radiosonde network observations. The authors have examined several related and 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 presented where two wind profiler triangles with nearly identical centroids and no common vertices produced strikingly different results during a 43-h period. It is concluded that the uncertainty in objectively analyzed quantities can easily be as large as the expected synoptic-scale signal. In order to quantify the statistical precision of the algorithms, the authors conducted a realistic observing system simulation experiment using output from a mesoscale model. A simple parameterization 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.

Abstract

A topic of current practical interest is the accurate characterization of the synoptic-scale atmospheric state from wind profiler and radiosonde network observations. The authors have examined several related and 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 presented where two wind profiler triangles with nearly identical centroids and no common vertices produced strikingly different results during a 43-h period. It is concluded that the uncertainty in objectively analyzed quantities can easily be as large as the expected synoptic-scale signal. In order to quantify the statistical precision of the algorithms, the authors conducted a realistic observing system simulation experiment using output from a mesoscale model. A simple parameterization 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.

Save