Skip to main content


Front. Earth Sci.
Sec. Atmospheric Science
Volume 12 - 2024 | doi: 10.3389/feart.2024.1399409
This article is part of the Research Topic Tropical Cyclone Modeling and Prediction: Advances in Model Development and Its Applications View all 9 articles

Ocean component of the first operational version of Hurricane Analysis and Forecast System: Evaluation of HYbrid Coordinate Ocean Model and hurricane feedback forecasts

Provisionally accepted
  • 1 Atlantic Oceanographic and Meteorological Laboratory (NOAA), Miami, United States
  • 2 Lynker in support of NOAA/NWS/NCEP/EMC, College Park, MD, United States
  • 3 National Weather Service, Silver Spring, Maryland, United States
  • 4 Global Systems Laboratory, Boulder, United States

The final, formatted version of the article will be published soon.

    The first operational version of the coupled Hurricane Analysis and Forecast System (HAFSv1), accepted in 2023, consists of the HYbrid Coordinate Ocean Model (HYCOM) and Finite-Volume Cubed-Sphere dynamic atmosphere model. The system is a product of efforts by improving and updating the system for the four year period (2019)(2020)(2021)(2022), through extensive collaboration between the Environmental Modeling Center at the US National Centers for Environmental Prediction and the NOAA Atlantic Oceanography and Meteorology Laboratory. In this paper, we evaluate HYCOM performance with two official configurations, HFSA and HFSB, with the objective to document the ocean forecast skill in the dynamically coupled systems. We chose Hurricane Laura (2020) as an exemplary case that demonstrates interactions between the storm and oceanic mesoscale features. Comparisons performed with available in-situ observations from gliders, Argos and National Data Buoy Center moorings show that HYCOM's simulations agree better for weak winds than for high winds (greater than Category 2). Skill metrics indicate that model sea surface temperature (SST) and mixed layer depth (MLD) have relatively low correlation. SST, MLD, mixed layer temperature (MLT) and Ocean Heat Content (OHC) are negatively biased. For high winds, SST and MLT become further negative, while MLD becomes closer to the observations by about 8 -19% improvement. The OHC discrepancy is proportional to predicted wind intensity. To the contrary, mixed layer salinity (MLS) uncertainties are smaller positive for higher winds, probably due to deepening MLD. The less negative bias of MLD for high winds implies that the wind-force mixing is less effective, owing to deepened MLD and high buoyancy stability (~1.5 -1.7 times) than the observation. The heat budget analysis suggested that the maximum heat loss by Laura was O(< 3 o C per day). The main contributor is the advection, followed by the entrainment; and they act against or with each other, depending on the storm quadrant. We also found relatively large unaccountable heat residuals for the in-storm period, and the residuals notably led the heat tendency, which means that further improvement of subscale simulations is warranted. In summary, HYCOM simulations showed no systematic differences forced by either HFSA or HFSB.

    Keywords: Earth System modeling1, coupled ocean-hurricane modeling2, ocean forecast modeling3, hurricane forecast4, upper ocean responses to a Tropical Cyclone5, Operational modeling6

    Received: 11 Mar 2024; Accepted: 10 Jun 2024.

    Copyright: © 2024 KIM, Liu, Thomas, Rosen, Wang, Hazelton, Zhang, Zhang and Mehra. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

    * Correspondence: HYUN-SOOK KIM, Atlantic Oceanographic and Meteorological Laboratory (NOAA), Miami, United States

    Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.