Temperature and output

I have been running the turbine without ventilating the electrical side to try to reduce condensation.  As a consequence, the temperature inside the compartment is higher.

In recent days, the temperature has been higher than usual because outside temperatures have been warm, and I began to notice that power output seemed to be drifting down as the temperature inside the compartment went up.

So I did an experiment.  Without changing any other factor which might alter power output, I re-established ventilation.  Over the next 40 minutes, the power output went up by by 6 watts.

Next day, to be sure this wasn't a fictitious observation, I reversed the experiment and blocked up the ventilation louvres again: the temperature promptly rose back to 30.4℃ and output dropped back to 616 watts.

OK, - so it's not an earth shattering observation, but nevertheless it's of interest.  The same effect is noticed and commented on (see page 5) in the report written for EcoInnovation's compatibility test for a Powerspout GE 400 coupled to an SMA inverter.

22 Dec 2015 Note added later: 
For the type of permanent magnet used in the SmartDrive rotor, which is ferrite ceramic, the flux density decreases linearly with increasing temperature. The obverse of this statement is that flux density increases with decreasing temperature. This would explain the above observation. 
3 Feb 2017 A further note:
The resistance of copper wire increases linearly with temperature; the impedance of the stator coils will thus increase with a rise in temperature and this too will affect the power leaving the alternator.