The reliability of using dropsondes for high-wind air-sea flux estimates

Direct measurements of high-wind air-sea fluxes are notoriously difficult to make within tropical cyclones, which presents challenges in parameterizing or constraining these quantities in numerical models. Due to the scarcity of data in extreme conditions, hypotheses about how surface fluxes of energy and momentum behave in high winds, including the role of spray and/or waves, remains speculative. In recent years, one method which attempts to estimate air-sea transfer in high winds relies on the so-called flux profile method, which has been used in several instances to retrieve surface momentum fluxes and the corresponding drag coefficient from mean velocity profiles obtained from either dropsondes or meteorological towers. In this work, we critically evaluate the reliability of this method for obtaining both air-sea momentum and enthalpy fluxes by performing a Monte-Carlo-based sensitivity analysis. We also utilize “virtual sondes” released throughout a turbulence-resolving large eddy simulation of an idealized tropical cyclone, in order to test whether prescribed (i.e. known) surface fluxes can be recovered from mean vertical profiles. We conclude that at best, estimates of the drag coefficient are accurate within 50% using the flux profile method, while estimates of the enthalpy flux coefficient are accurate within 200%.