Response of sea level to tide, atmospheric pressure, wind forcing and river discharge in the Kitimat Fjord System

Abstract

The response of sea level to tide, atmospheric pressure, wind forcing, and river discharge in the Kitimat Fjord System (KFS) is investigated by using observations from tide gauges and results from a high-resolution ocean circulation model. Tidal harmonic analysis of the observed hourly sea level at the head and mouth of Douglas Channel shows that the largest semidiurnal tide (M2) alone accounts for 79% of the total sea level variance, and the top four major tidal constituents (M2, S2, K1, and O1) account for 94% of the total variance. In addition, the amplitude of M2 increases slightly from 1.586 to 1.647 m between the mouth and the head. For the observed subtidal sea level, the local inverse barometer response due to the variation of atmospheric pressure accounts for 59% of the variance. The variation of subtidal sea level difference between the mouth and the head of Douglas Channel is observable and can be largely attributed to the along-channel wind (54%) with an additional 13% of the variation due to the daily-averaged and seasonal-cumulative river discharges.A fjord-shelf three-dimensional ocean circulation model for KFS is developed to simulate the sea level variability. The model performance in simulating the sea level is assessed by comparing model results with observed harmonic constituents of ~70 tidal stations inside the model domain and time series recorded at the two tide gauges of KFS. The modelled tides are in good agreement with observations with a domain-averaged root-mean-square error estimate of ~0.04 m for M2. For the subtidal sea level, the model is able to capture the observed synoptic variations. Further analysis suggests that the underestimates of subtidal sea level in the ocean circulation model are mainly due to the weaker-than-observed wind forcing taken from the atmospheric model especially during the peak of strong down-fjord katabatic wind events.