Here is an interview I found on the web although Dana Riddle is not a name I recognize.
http://www.netpets.com/fish/fishnews...ddlewalch.html
It is sort of a reiteration of what is previously been mentioned and she mentions that there is some research out there but I have yet to find it. I guess the fact of the matter is that while Kelvin and PAR are two separated entities, there is some correlation and as is intuitive, the higher the kelvin, then typically the lower the PAR (more or less anyway).