Hydrographic response to winter shamal winds in the Northwestern Arabian Gulf

Abstract

This study investigates the effect of air–sea heat fluxes associated with cold, dry Shamal wind events on the physical dynamics and vertical mixing of the shallow, semi-enclosed northwestern Arabian Gulf. In situ observations were combined with one-dimensional numerical simulations that used two two-equation turbulence closures, the k – ϵ and k – kl formulations. Field measurements were collected from mid-January to mid-April 2013 at two coastal stations near Qarooh Island, Kuwait, during the initial phase of an extended monitoring program. Continuous records of water temperature, dissolved oxygen, light penetration, and current profiles were obtained, together with meteorological observations used to force the models. Reynolds stresses, vertical eddy diffusivity, turbulent kinetic energy (TKE), and its dissipation rate were estimated from a pulse-coherent Acoustic Doppler Current Profiler (ADCP) and used to evaluate the simulations. Two winter Shamal events, with winds of up to 15 m s -1 , increased surface heat loss through latent heat and drove convective mixing that reached the seabed. Turbulence in the bottom boundary layer was controlled by two processes. Shear instability generated by the semidiurnal tidal currents dominated under low wind conditions, whereas convective overturning during the Shamal peaks homogenized the water column vertically, which reduced the tidal shear and weakened the shear-driven mixing. Model–observation comparisons were evaluated using the mean bias error, root mean square error, Pearson correlation coefficient, and Nash–Sutcliffe efficiency. Under observationally constrained temperature and velocity conditions, both closures represented the general temporal variability of the turbulence quantities, with correlation coefficients ranging from 0.63 to 0.86. Agreement was strongest for the broader tidal modulation, whereas some short-duration turbulence peaks were substantially underestimated. During SE1, for example, the peak dissipation rate and Reynolds stress were underestimated by factors of approximately five and three, respectively, and the TKE Nash–Sutcliffe efficiencies were only 0.37 and 0.08. Because the temperature and velocity fields were relaxed toward observations, these comparisons evaluate the behavior of the closures under measured hydrographic and current conditions rather than provide fully independent model validation.