Ran the cloud sensor and it correlated well with our visual observations. Temps colder than 27C for our site were clear. 22C-27C were partly cloudy and warmer than 22C cloudy and not good for imaging. One of my PROMPT/CTIO colleagues, Kevin Ivarsen made a quick script to graph the results. DOnt think it is as complete as John's but you can see it at:

http://salsa.skynet.unc.edu/skynet/w...mp?range=10800

I think the "PhiCloud" (apologies to Steve Jobs) is gonna be a winner.