2C.2 HFIP Real-Time Parallel Experiments for 2023 and 2024 Hurricane Seasons Using HAFS-A Configurations

Monday, 6 May 2024: 11:00 AM
Beacon B (Hyatt Regency Long Beach)
Bin Liu, Lynker at NOAA/NWS/NCEP/EMC, College Park, MD; and J. Shin, B. Thomas, Z. Zhang, H. S. Kim, B. Li, C. K. Wang, M. F. Aristizabal Vargas, J. D. Steffen, W. Wang, L. Zhu, X. Li, Y. Weng, J. Cheng, A. Mehra, and V. S. Tallapragada

With Hurricane Analysis and Forecast System version 1 (HAFSv1) operationally implemented as NCEP’s regional hurricane forecast system (including two configurations of HAFS-A and HAFS-B), the operational HAFS-A configuration based real-time parallel experiment (HAFSv1.1A) was developed and conducted as one of the Hurricane Forecast Improvement Program (HFIP) real-time parallel demo projects for the 2023 hurricane season. The HAFSv1.1A experiment utilizes a higher resolution regional storm-following moving nest configuration, an upgraded vortex initialization technique being able to relocate and cycle cloud (hydrometeor) variables, and the improved inner-core data assimilation technique by assimilating additional high-resolution GOES-R mesoscale floater Atmospheric Motion Vectors (AMVs). In addition, HAFSv1.1A is featured with the Thompson microphysics scheme as part of the physic upgrades and the newly developed regional Modular Ocean Model version 6 (MOM6) coupling capability. Retrospective tests were conducted for the 2021-2022 North Atlantic basin storms, and real-time parallel experiments were carried out and monitored for the 2023 North Atlantic, Eastern North Pacific, Western North Pacific, and North Indian Ocean storms. Compared to operational HAFS-A configuration, the HAFSv1.1A experiment produced neutral to positive storm track forecast skill, with 5-10% improvement in intensity forecast skill. Development and plans for the operational HAFS-A configuration based real-time parallel experiment for the 2024 hurricane season will also be discussed and presented.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner