Impact of Assimilating Atmospheric Boundary Observations on the Rapid Intensification of Hurricane Idalia (2023) in CADRE Self‐Cycled HAFS‐JEDI System

Abstract

This study investigates the impact of assimilating dynamic and thermodynamic atmospheric boundary layer (ABL) observations on the simulation of tropical cyclone (TC) rapid intensification (RI) using the atmosphere–ocean coupled Hurricane Analysis and Forecast System (HAFS) with the Joint Effort for Data assimilation (DA) Integration (JEDI) framework. Despite the recognized importance of ABL processes for TC evolution, the influence of assimilating ABL observations on hurricane prediction remains insufficiently quantified. To address this gap, fully self‐cycled DA experiments were conducted for Hurricane Idalia (2023), which underwent RI from tropical storm to Category 4 TC under favorable oceanic but dry atmospheric conditions. The All ABL experiment assimilated all available observations, including dropsondes sampling the ABL, whereas No ABL withheld those data to isolate their contribution. During DA cycling, All ABL substantially reduced background and analysis errors in ABL temperature and winds, recovering high‐entropy air and strengthening inflow near the inner‐core region. Forecasts initialized from these analyses captured the observed RI (Category 4), whereas No ABL remained mostly below Category 2 intensity. The assimilation of ABL observations enhanced entropy, radial inflow, and tangential wind, supporting stronger secondary circulation and a pronounced warm‐core structure consistent with observations. Air–sea interaction diagnostics further showed that All ABL maintained higher entropy through enhanced wind‐induced enthalpy fluxes, sustaining continuous RI and compact storm organization. These results demonstrate that realistic ABL representation is essential for predicting RI. The findings highlight the need for strongly coupled atmosphere–ocean DA systems to improve future hurricane forecasting.