Wind stress measurements from the QuikSCAT-SeaWinds scatterometer tandem mission and the impact on an ocean model

We examine diurnal signals captured by the QuikSCAT-SeaWinds scatterometer tandem mission (April-October 2003) and their impact on ocean model simulation. The diurnal variability captured by twice-daily scatterometer wind from the tandem mission is substantially larger than that estimated by the National Centers for Environmental Prediction (NCEP) reanalysis product (even with a 6-hourly interval in the latter). Consequently, the impact of diurnal wind on model sea surface temperature (SST) is significantly larger with scatterometer than with NCEP winds because of stronger vertical mixing caused by the twice-daily scatterometer wind. This is consistent with previous studies that high-frequency wind at the ocean's inertial frequencies enhances vertical mixing through resonant inertial oscillations. The weak vertical mixing associated with daily scatterometer winds causes warm bias of SST (relative to that resulting from twice-daily scatterometer wind) and larger deviation from observations. The warm bias reaches several degrees Celsius in midlatitude oceans during summertime and can accumulate with time. Heat flux correction that attempts to account for the feedback of SST would propagate the error in wind and vertical mixing to heat flux without correcting the error source. Because of this error compensation, caution is needed in the interpretation of SST budget resulting from ocean models and data assimilation outputs based on wind products that do not adequately resolve diurnal variability. Our findings highlight the need to resolve diurnal wind in future scatterometer missions. Our study assumes that wind differences between the two scatterometers are primarily due to the sampling of diurnal cycle at different times. However, potential bias between the two needs further investigations.