Don't get your hopes up. I calculated that to obtain a power output of 2 kW at the specified efficiency of 0.7 and a hydrodynamic head of 18 inches (= 0.45 m), you need a water flow rate of 0.65 m^3/s. You can make do with less, as the water not only creates downwards inertia by travelling down 0.45 m, but also has a sizable amount of horizontal intertia too. Using a clever design, you can probably tap some of this energy too without immediately creating a reservoir.
Nevertheless, 0.65 m^3/s is a pretty big stream to have in your back yard. I looked up some flow rates of rivers and then discovered that the river Semme which flows alongside the vacation house of my girlfriend in central France has the right
débit. Here's a picture of the Semme at a wider point; you should think of the width at the place of the visible kayak:
Not a big stream, but not exactly small either, and definitely not something we all have in our backyard. The inventor is quite a smart bloke to extract energy from the stream at such a small hydrodynamic head---that is certainly a very clever invention---but it unfortunately won't be a device which will see much use.
