MIKE3-Instability : water surface oscillations in upstream boundary of a channel

Hello every body,

I am running a physical model, a channel scaled 1/6000 in XY plane and 1/400 in Z direction, with MIKE3D. the characteristics of the model are:

0- channel bathymetry is not the same along the channel: maximum depth: 15cm, maximum width: 12 cm and length : 3-4 meter
1- Ten vertical sigma variable layers and rectangular mesh.
2- 10000, 0.01 sec timesteps
3-Navier Stokes-equations
4- minimum and maximum time steps: 0.0001, 0.01 sec
5- Riemann factor and filter coefficient: 1, 0.02
6-Drying/wetting depth: 0.001, 0.002 m
7- Minimum and Maximum eddy viscosities: 2.25e-10 , 0.000125 m2/s
8- Roughness height: 2.425e-05 m
9- Upstream and downstream boundary conditions: 0.000707 m3/s and 0.0054 m
10- initial condition: water surface elevation: 0.0054 m
11- TKE and Dissipation of TKE in upstream and downstream boundaries: 3.75e-09 m2/s2 and 5.625e-09 m2/s3

When I ran the model under above conditions, there is a huge oscillation/fluctuation of water surface in upstream boundary of the channel. This causes the model NOT to get stable.
Any comments/help/guide would be appreciated.
Regards