|
363516-191001-021000
18-NGI3-57
16-NGI3-05, 17-NGI3-23
10/1/2016
2019-9-30 0:0:0
Completed
$211,890.00
Examination and Validation of Reconnaissance Field Program Data in Multiple HWRF Framework
Fitzpatrick
Pat
MSU
DM-CH
OAR
A digitized ranking index to assess tropical cyclone forecast products has been developed to facilitate decision-making on allocating resources. The validation technique makes the tedious task of product validation easier and increases model assessment efficiency. The 15 metrics used are a blend of relative scale, absolute scale, similarity measurement, and outlier characteristics which capture a range of unique error attributes. This method can also assess track error by vector direction and magnitude, isolating ambiguous model deficiencies based on the traditional absolute error distance calculation. During this process, an improved Kurihara vortex filter (required for shear calculations) was developed for higher-resolution models. Validation and case study analyses were performed on NOAA Hurricane Weather Research and Forecasting (HWRF) model products and experimental products (e.g., HWRF-HYCOM, HEDAS, and basin-scale HWRFT).
|