STUDY COMPONENTS OF SHALLOW WATER TIDE (OVER AND COMPOUND TIDES) 1 DIMENSIONAL CHANNEL MODEL BY USING VARIATIONAL DATA ASSIMILATION METHOD

Abstract

Shallow water tides are very important to improve accuracy of tidal predictions. It is used by port interest, sea transportation, fishing industry, coastal engineering, etc. Simulation of shallow water tides was obtained from harmonic analysis of 1 dimensional channel model (12 grid) by using variational data assimilation (grid 3 and 8). Two partial tides with angular frequencies σ1 = 1,4x10-4 and σ2 = 1,6x10-4 rad/sec and amplitude A1 = 1x10-8 and A2 = 0,5x10-8 meters are used for defining external forcing in the model domain. When inspecting the amplitudes of both partial tides σ1 and σ2 and some of their dominant over- and compound tides (σ3 = 2σ1-σ2 and σ4 = 3σ1), in general the “to be corrected” solution can be improved significantly. Root mean square (rms) error of tidal constituent σ1 between the “reference” and the “to be corrected” without data assimilation is 0,1075 m/sec, and for σ2 is 0,0440 m/sec, respectively. On the other hand, the harmonic analysis of the phase of tidal constituent σ1 showed a good result (root mean square = 0.0000 m/s) and for σ2 (root mean square = 0.0002 m/s).

Keywords: shallow water tides, data assimilation, harmonic analysis.