Document Type

Conference Paper


Available under a Creative Commons Attribution Non-Commercial Share Alike 4.0 International Licence



Publication Details

XXIst. TELEMAC-MASCARET User Conference, Grenoble, France, 15-17th. October.


A review of the literature shows that most model calibrations involve the adjustment of the bottom friction coefficient to minimise the error between predicted and measured tidal elevations. In this study, an alternative procedure is adopted when calibrating a Telemac2D model covering an area on the Eastern coast of Ireland. The model is forced with eight principal tidal constituents derived from the MIKE 21 global model. It is calibrated separately for the two principal constituents, M2 and S2. The field data comprises tidal elevations recorded at five locations over a full lunar cycle in October 1998 and Spring and Neap current data. At each node on the open boundary, the M2 amplitude supplied by Mike21 is adjusted by one of five possible amplitude multipliers (0.8,0.9,1.0,1.1,1.2) while the phase is shifted by eight possible values (-15°,-10°,-5°,0°,+5°, 10°,15°,20°). An amplitude multiplier of 1.05 and phase shift of 12.5° corresponding to the lowest error is determined by plotting the minimum error for a total of 40 simulations. The S2 tide calibration also requires 40 simulations. Alternative strategies are investigated to reduce the total number of simulations. The application of the method of steepest descent reduces the number of required simulations to eleven. In another approach, the amplitude modifier and the phase shift can be calibrated separately as it is shown that the amplitude modifier had minimal effect on the phases and the phase shift has negligible effect on the amplitudes. The calibration requires just five simulations to find the optimum amplitude multiplier and eight for the optimum phase shift. The friction parameter is calibrated separately using measured Spring and Neap tidal currents.