Hum... my first observation on that link is that the author 'evaluates' the colour coding, without noticing that the temperature scale changes... The second is that 'hot spot' locations don't seem to be fixed. The third is that according to the AREVA slides, the open reactor well is filled with water.

I also suspect that there is no direct temperature observation of the SFP on the thermograms (possibly due to the water vapour over it): the pond should be boiling or near boiling temperature, not 30-47°C.

_{*Lunatic*, n.
One whose delusions are out of fashion.}