After taking a closer look it seems that at least some of the tidal currents Windy is reporting for the US is quite a bit different from what the government NOAA website predicts. This makes me hesitant to use Windy until there's an explanation of what's going on.

As an example, I looked today at tide station Upper Hell Gate, Sasanoa River (ACT0596). Windy has this data for February 11, 2020:
10:16 AM Slack, Flood Begins 1h 7m ago, 0kt
1:11 PM Max Flood in 1h 48m, 1.26kt
5:01 PM Slack, Ebb Begins in 5h 38m, 0kt
11.2 at 7:25 PM Max Ebb, -1.28kt

But NOAA has this data:
2020-02-11 11:47 AM slack -
2020-02-11 01:19 PM flood 1.31
2020-02-11 03:34 PM slack -
2020-02-11 06:59 PM ebb -1.06

As you can see, the times given for slack is 90 minutes too late in Windy. (I also checked a number of other sites which are very close or identical to NOAA, and my experience on the water is that NOAA gets these predictions quite accurately.)

What's the reason for this discrepancy?

I am an avid sea kayaker and make use of tides and currents constantly, pulling from a variety of sources. I already use Windy for weather and swell predictions so it's great that you're now including this information. I have two suggestions for changes that would make what you've got much more useful:

• For current stations only, add ebb and flood headings, that is, the compass direction of the ebb and flood directions. Without this information, the flood and ebb timing is often useless unless the station is in a very simple location (like a river mouth). As one example, where the East River joins the Harlem River in NYC, the definition of ebb/flood direction actually reverses as you go through the river confluence. If you don't know that, you can wind up making exactly the wrong decision! The definitions of ebb and flood are a bit arbitrary for waterways that connect two larger tidal bodies of water, and also for channels between islands where it is hard to guess which current is labeled which. This is why they are included in all NOAA current stations.

• If it were possible to plug an arbitrary time/date into the tide/current stations (rather than just a time from the slider at the bottom) this would turn Windy into a full-blown tidal prediction app, which would make it a "one-stop shop" for those planning a trip in the future. But obviously this is more disruptive to the UI design.

Thanks for your consideration!