Very cool. One of the challenges about building an energy system with lots of variable inputs is that you can't just look at the lowest levelised cost and buil

Very cool. One of the challenges about building an energy system with lots of variable inputs is that you can't just look at the lowest levelised cost and build only that, otherwise you'd just build wind and be done with it. Adding sources that are not correlated with wind makes the overall grid management problem easier, even if those sources are more expensive the system cost may decrease. Obviously tide, while very predictable, is not dispatchable so not quite as valuable for grid management, but still shouldn't be written off just because it costs a lot per MWh. One thing that does make it hard is that peak water flow only happens in the middle of a six hour tidal cycle. For every location you have a predictable 6 hourly cyclical power curve. There's about 75 minutes between the high tide times in the North of Scotland and Southern England so on a GB grid basis, you can spread the curve out somewhat but still have that cyclical element.