2024 AIChE Annual Meeting
Validation of Sea Surface Latent and Sensible Heat Fluxes in Three Global Reanalysis Products Against Saildrone Observations during 2023 Atlantic Hurricane Season
Many devastating hurricanes, such as Katrina, Harvey, and Maria, have seen a 30+ kt increase in the maximum sustained wind speed in a 24 hour period. This is known as rapid intensification and is challenging for models to predict. One key factor in rapid intensification is the energy transfer from the ocean to the atmosphere. While observations from systems such as saildrones, buoys, and aircrafts can be used to observe the air-sea energy exchange, they are irregular in space and/or time. Reanalysis procedures assimilate all available observations and synthesize them into products that are uniform in both space and time. However, model errors can impact these products. Before reanalysis products are used as proxies for in situ observations, they should be validated, especially under extreme conditions such as hurricanes. This study aims to validate sea surface latent and sensible heat fluxes from three commonly used global reanalysis products (NASA MERRA2, ECMWF ERA5, NOAA CFSR2) against saildrone observations taken during the 2023 Atlantic hurricane season. Hourly averaged saildrone data were paired with reanalysis data from each product. Discrepancies between saildrone observations and reanalysis outputs were mostly within one standard deviation with sporadic large spikes. The largest discrepancies are found under hurricane conditions. CFSR2 demonstrated a systemic negative bias for latent heat fluxes and large negative and positive errors for sensible heat fluxes. These results suggest heat fluxes from MERRA2 and ERA5 are a reliable representation of air-sea interactions in the Atlantic Ocean except under hurricane conditions. Away from hurricanes, the results are consistent with those from a similar study conducted for the Pacific sub-Arctic Ocean.