Correcting co-tidal charts produced by numerical models

Much of the tedious task of tuning a 2-dimensional tidal numerical model has been avoided by introducing a corrective procedure. The model is run in a time-stepping mode until a steady state is achieved, and harmonic constants determined; these are then corrected by reference to the measured values at representative ports. The technique has been successfully applied to a fine-mesh model of the English Channel. When it is extended to adjust the tidal streams, the technique fails because of the sensitivity of the computed streams to the spatial gradient of the vertical tide (especially its clockwise component). It would be better to replace the geometrically based corrective technique by a least-squares fit to the measured data, applied to a non-time-stepping model.