Estimates of Drag Coefficients and Surface Waves under Tropical Cyclones using Subsurface EM-APEX floats

Abstract

Nineteen autonomous EM-APEX floats were air-launched from aircraft to measure the profiles of ocean current velocity, temperature, salinity, and high-frequency velocity variance (σ_u ) ̃^2 under five different tropical cyclones. In the first part of this work, the surface wind stress τ is estimated by integrating the float velocity measurements in the vertical, assuming a linear momentum budget balance. The biases and uncertainties of estimated τ are studied carefully, using the simulations of tropical cyclone-induced ocean response in the Price–Pinkel–Weller (PWP3D) model. The results are reliable only before passage of the storms’ eyes. At wind speeds |U10| = 25–40 m s–1 the downwind drag coefficient C_∥ is 1.1–3.1 × 10–3 in the front-right sector of tropical cyclones and 0.8–1.7 × 10–3 in the front-left sector. At wind speeds |U10| > 40 m s–1, the C_∥ is ~ 1.6 × 10–3. The angle ϕ of drag τ is mostly clockwise from the wind U10 in the front-right sector of storms. A new parameter, the nondimensional effective wind duration ζ^*, is proposed as a function of wind forcing, storm translation, and fetch length. The value of ζ^* is shown to predict the values of C_∥ and ϕ. C_∥ increases with decreasing ζ^*. We propose that surface waves with larger amplitude are forced more efficiently by the wind under faster storms, resulting in stronger surface wave breaking and thus higher C_∥. In the second part of this work, the peak frequency fp and significant wave height Hs of surface waves are estimated under Typhoon Fanapi using a nonlinear least-squared fit of the modeled profiles to the profiles of (σ_u ) ̃^2 measured by the floats, assuming the Joint North Sea Wave Project (JONSWAP) surface wave spectrum. The fp is 0.08–0.10 Hz, with the maximum fp (0.10 Hz) in the rear-left quadrant of Fanapi. The Hs is 6–12 m with the maximum in the rear sector of Fanapi. The results are then assessed and compared to the simulations in the WAVEWATCH III surface wave model. The contributions of this work, including the improvements to previous methods for estimating τ and surface waves using subsurface EM-APEX floats, the estimates of drag coefficients and ϕ under five tropical cyclones, and the data-based parameterization of drag coefficients in terms of surface waves’ effects, are all important to future studies of tropical cyclone–wave–ocean interactions.