Foggy Nights

The last three nights of my run were totally fogged out. We weren't up in clouds the whole time, but the humidity was always near 100%. No way I can use a telescope in those conditions. Instead of using the scope, I spent basically all of my time behind the computer processing the data from Mon & Tues. I got sidetracked by a simulation I had been meaning to do for a while now - basically I could see the possibility of an error producing a fake "polarization" and I wanted to make sure that what I was seeing was real. I had tried this simulation before on the data I'd published, but it was only a quick look. Now that I've got over 40 stars, I felt I needed to do it right. And good thing I did. It's a very significant effect in stars with strong emission. So, my last three days has been spent behind my laptop in the kitchen. At least I got a good sunset one night.

Kaho'olawe at sunset.
Daytime clouds behind Panic-starrs. I really liked the cap-clouds as the top of the cumulus would hit an unstable layer.
The convection shifted to being behind MSSS, more toward the north than the east. I always like the zoom shots of these massive clouds with the scopes looking tiny.
One of my simulations - the bottom is the spectrum of a star and the three lines above it show the calculated polarization for different simulated continuum errors. Basically, I try to get everything as close to zero as I can. I'm usually right to 1% or so. If variation of 2% or 4% produces something fake, I start to worry.

A fake signature - the proper polarization is the flat one, but a 4% error produces that bump on top.

A different star - my best guess is the curve that's mostly flat with the -1% dip where the spectrum is lowest. If I make a 4% error , the dip becomes a bump. There's also a broad downward trend across the entire emission part of the line. Looks like I need to be pretty careful with this stuff..... It also makes me wonder why nobody else mentions this, and the other errors I simulate, when they've published their codes.