The modelled and measured significant wave heights coincided rather well – the corresponding correlation coefficient was 0.89 and the scatter index (root mean square difference between the simulated and measured significant see more wave height divided by the mean measured significant wave height) was 0.28.

The correlation coefficient was 0.76 and the scatter index was 0.24 for peak periods. The horizontal distribution of wave-induced shear velocities at the peak of the strong (15 m s−1) southerly wind event showed great variability (Figure 9). Shear velocities were the highest, exceeding 6 cm s−1, in the southern part of the Suur Strait and were less than the critical value for resuspension in its deepest area. Thus, wave-induced shear velocities were generally related to the bottom topography of the strait. The flow velocity measurements in the Suur Strait in November–December were used for the validation of the circulation models (100 m and 400 m grid step). Only the validation of the high resolution 2D circulation model with the 100 m grid step is presented because the model with the 400 m grid step gave approximately the same result. Since the models do not contain semidiurnal tidal currents, the measured flow velocity data series were

smoothed with a 12 h moving average. It can be seen in Figure 10 that the rapid change in the wind field on 23 November (Figure 2) with the consequent Lumacaftor in vivo sea level change (Figure 4) caused remarkable changes in the flow regime. The high-speed flow reversed within a short space of time. The coincidence of the measured and simulated along-strait flow speed was high – the correlation coefficient was 0.88. A certain difference can be seen in the case of higher flow speeds. Re-calculations with the circulation model were performed for November by taking into account wave stress as forcing additional to the wind. The wave stress was obtained from wave model simulations. The r.m.s. difference of the simulated along-strait flow

component v with and without wave stress was 0.01 m s−1 over the Suur Strait model area. The estimate was found for 19 November when the significant wave height was the highest ( Figure 5a). Wind-induced mafosfamide currents were much stronger in the strait area, reaching values of up to 60 cm s−1 (Figure 11), when wave-induced currents were negligible. However, wave-induced currents were essential in the flow field near the south-eastern tip of Muhu Island. The influence of wave stress on water exchange in the Suur Strait is insignificant. Wave-induced currents should be considered when modelling sediment transport in a shallow sea in the case of a long fetch. The simulations with a validated high resolution circulation model were performed for the whole of 2008. Satellite imagery (Envisat, MERIS, ASAR) showed that the Väinameri region was practically ice free in 2008.