trying to design a turbine that will power a geyser seems fraught with pitfalls. Here's another one:
The power input follows a 3rd order polynomial curve of the windspeed. The resulting rpm is more like a linear curve.
The output voltage is proportional to the rpm, so - since P = Volt²/R, in which R = the resistance of the heating element - the output power follows a second order curve of the windspeed, and there is no way these two curves can be matched (as illustrated by the picture)
In this case, the output curve lies above the input curve. That's - of course - impossible, since you cannot get more out than was put in. So, what will happen?
What will happen is that this turbine wont even get out the starting blocks. Any bit of electricity is consumed by such a power hungry element that nothing is left to turn the turbine.
The only way this turbine will start is when we can get the yellow curve BELOW
the blue curve for all windspeeds,(see the red dotted line) which puts a severe restraint on the usefulness of this design