I will add my voice to those saying that an automotive alternator is a poor choice. I have rewound a couple of them for automotive applications where a replacement was not available. They are made to be inexpensive to produce, and just adequate in service. I have also rewound a few motors, and the higher quality was apparent.
Alternators seem attractive because you can regulate in automotive fashion by controlling the field current, but this is a power loss, and the prime mover (windmill?) can run-away if the regulator or field winding, or brushes fail. The claw-pole design that is used is also horrible from an efficiency standpoint, but cheap to produce.
So it would be best to replace the rotor with a permanent magnet assembly. (shown in hiker's post) But now you are no better off than if you had started with a motor, which will have better bearings, more iron, and fewer burrs and other defects in the laminations, and a more weather resistant housing.
But if all you can come up with in your austere location is a car alternator, then have at it. If replacing the armature, make that first. Regardless you will need to wind some test coils and find a way to spin the machine at known RPM, then you can calculate how many turns per phase are needed.
One trick I found was to use lengths of 1/4" (6mm?) PVC plastic tubing in the slots to help keep control of the wire, and to mark which slots to skip, etc. Make these about 2"/5cm longer than the stack so they stick out 1" or so on each end. You can also use nylon zip ties to keep one set of coils (phase) in place while you are working on the next.
After winding, you need to lace and varnish the wire so vibration doesn't eat through the insulation. I used some spray stuff the motor rewinding shop sold me (they no-bid the rewind). I suspect polyurethane varnish would work as well.