HRD seminar – Xuejin Zhang, CIMAS and AOML/HRD – 19 June 2014

Xuejin presented a seminar on “Representing Multi-scale Interactions in HWRF Modeling System: Design of Movable Nests and Forecast Verification”

Resumen

In this study, we systematically documented the design of the multi-level nesting, multiple movable nests, two-way nesting interactions, and initialization procedure in the Hurricane Weather Research and Forecasting (HWRF) system. By configuring an experimental basin-scale HWRF modeling system based on the above design, we demonstrate the realistic evolvement of a multiple-storm forecast and the superior track forecasts of the Hurricane Sandy (2012). The verification of three-year retrospective/real-time forecasts of 2011-13 hurricane seasons statistically shows it can significantly improve the accuracy of the track forecasts (7-12% improvement in basin-scale HWRF vs. in operational HWRF) in both the Atlantic and the East Pacific basins although the improvement of the intensity forecasts is mixed as anticipated. We also noticed the degradation of intensity forecasts can mainly be attributed to large errors of intensity forecasts for a few individual storms over the ocean. The possible oceanic deficiency may come from the inaccurate representation of Sea Surface Temperatures (SSTs) in the input data caused by those stagnant or slowly wobbling TCs. With the flexible capability built in the HWRF modeling system, the basin-scale HWRF system may be utilized to research on various issues related to TC forecast, such as storm-storm interaction, model bias diagnostics, physical scheme evaluation and improvement, and localized multiple-vortex initialization and data assimilation. It can serve as an experimental tool for advancing the current regional operational HWRF model to the next-generation global-to-local scale TC forecast model.

A video recording of the presentation is available on the anonymous ftp site:

ftp://ftp.aoml.noaa.gov/hrd/pub/blog/seminars/2014/X.Zhang_HRD_seminar_20140619.mp4

Screen Shot 2014-06-20 at 11.36.43 AM