The spectral analysis of the velocity time series show similar peaks for all the ADCP locations and bins.

Even the temperature spectral analysis in a Toronto Islands channel shows a strong peak for a period of *T*=1 hour (bottom right)

**It is interesting to observe the data from the City of Toronto Water Infrastructure Management department (bottom left):**

● They had a sampling period of half hour, which due to Nyquist theorem limits the accuracy of the spectral analysis to one hour period.

● However, we can still see the huge spike in the FFT plot close to the 1 hour period.

The magnitude of the peak for the spectal analysis of the Inner Harbour data suggests that Inner Harbour's natural oscillation frequency (eigen frequency) may have a period *T *= 1 h and its enhanced resonant oscillation may be the main driver of water exchange of the other embayments in Toronto Harbour.