Performance of a high resolution diagnostic model for short range mesoscale wind forecasts in complex terrain

Approved for public release, distribution is unlimited === This study investigates the feasibility of using a high resolution simple diagnostic model (WOCSS) initialized from a coarser grid full physics prognostic model (COAMPS) to obtain mesoscale winds. This approach using COAMPS 81, 27, and 9 km...

Full description

Bibliographic Details
Main Author: Gallaher, Shawn G.
Other Authors: Miller, Douglas K.
Published: Monterey, California. Naval Postgraduate School 2012
Online Access:http://hdl.handle.net/10945/5242
Description
Summary:Approved for public release, distribution is unlimited === This study investigates the feasibility of using a high resolution simple diagnostic model (WOCSS) initialized from a coarser grid full physics prognostic model (COAMPS) to obtain mesoscale winds. This approach using COAMPS 81, 27, and 9 km forecast model soundings to initialize WOCSS at 3 km is compared to COAMPS forecast at 3km horizontal resolution alone. Four case studies were collected during various weather regimes in Central California. Observations were collected from 5 different agencies and were used for verification of the models. The sensitivity of various WOCSS parameters were also explored. The results showed that overall the COAMPS(9km)/WOCSS approach provides winds as good as COAMPS at 3 km at a greatly reduced computation time. The COAMPS/WOCSS methodology performed particularly well during non-frontal situations where low-level inversions were present. Separation of the surface observation data by agency revealed large errors from data networks with low maintenance, monitoring and site specifications standards. The highest flow surface in WOCSS was the only parameter that displayed any significant sensitivity. Further work is needed to test the advantages of this sensitivity. COAMPS/WOCSS mesoscale forecast winds may prove to be very useful as input to emergency response applications such as dispersion and trajectory modeling.