As for the date, a more conventional way would be more appropriate to avoid confusion with tide value in meter.

For Canada, you could use the following web services to provide observations, predictions and forecast
http://www.tides.gc.ca/eng/info/WebServicesWLD

3 as it could improve the forecast for the Atlantic coast as well.

This is a good news to read that you're providing tidal information. First of all it is not forecast tides but predicted tides as it is based on harmonic calculation. Forecast tides would require observations to recalculate the predictions. Also, you are providing tide info for non tidal areas, upstream of Trois-Rivieres in Canada. There are waterlevel stations going up to the Great lakes. I also noted a bug for some stations as the information appear twice for low and high tides with slight differences (Rimouski). Thank you for considering my comments as Windy have been very innovative.

I noticed that the time for the tide on the St Lawrence River is not adjusted for daylight saving time thus making it 1 hour late.

3 as it could improve the forecast for the Atlantic coast as well.

For Canada, you could use the following web services to provide observations, predictions and forecast
http://www.tides.gc.ca/eng/info/WebServicesWLD

As for the date, a more conventional way would be more appropriate to avoid confusion with tide value in meter.

This is a good news to read that you're providing tidal information. First of all it is not forecast tides but predicted tides as it is based on harmonic calculation. Forecast tides would require observations to recalculate the predictions. Also, you are providing tide info for non tidal areas, upstream of Trois-Rivieres in Canada. There are waterlevel stations going up to the Great lakes. I also noted a bug for some stations as the information appear twice for low and high tides with slight differences (Rimouski). Thank you for considering my comments as Windy have been very innovative.