Monday, January 25, 2016

Well that is going to speed things up

Ever since I discovered how to make a progress bar in my python scripts, most of them seem to have them now.

It shows how far along it is but also the estimated time till completion. Brilliant. It also made me slightly more worried about how long something takes.

For example, a (older) script I wrote to fit SED models to suspected brown dwarfs. It works well but it's a little heavy on for loops. I wanted to apply it to LOTS of sources. So I ask Matt if he knows any tricks to speed things up. Matt shows me a trick using arrays.

Script goes from weeks-till-completion to mere seconds. wow. Usually I am not too fussed if a script runs for a little while (2 min? fine, I'll check twitter or something) but cutting down from weeks of runtime. nice.

Go Matt.

Now I'm going to stare at a progress bar run across the screen.

Tuesday, January 12, 2016

Hunting GHOSTS-I

This weekend was spent at La Palma and specifically the William Herschel Telescope (WHT) to hunt for the spectral signature of GHOSTS-I a super-faint and very isolated dwarf galaxy (see the discovery paper, Monachesi+ 2014).

It was a dark and thankfully not stormy night. On Sunday, the weather looked good but clouds poured out of the caldera (a beautiful sight) right over the WHT (boo!). So it felt like I may just be out of luck with the humidity (clearly) over 100%. It could well be a night where we'd sit & listen to the water drip off the dome (exciting!).






To add to the feeling of a lost expedition, the electronics on the blue side of the spectrograph were acting up in a decidedly odd way. No real data was being read out!

But the support team got to work and got the Blue side behaving again. In the meantime, the night had started, things cleared up and we could actually go and hunt for GHOSTS-I!

It is so faint that it is invisible in any target acquisition image so we positioned the WHT/ISIS slit on the sky using a nearby star and checking it against the visible objects nearby. Tricky.

And then we integrated. For a total of 7 hours. 5 one hour exosures and 4 half hour ones. Seeing fluctuated so some exposures may be more useful than others. We kind-of-sort-of see a very faint continuum signal already in the raw data (we squint...) but no clear emission line just yet.

I now need to reduce the data and see how good of a spectrum I can construct. Hopefully I can tease out a redshift for this galaxy. It will be interesting: this was definitely one of the faintest things ever observed with WHT/ISIS.




Then on Monday morning, i went down the mountain, give a short talk at the IAC office on GHOSTS-I and the overall GHOSTS survey, off to the airport and home for sleep.