Tuesday, November 25, 2014

Arxiv to bibdesk

A long time ago (in a Baltimore far far away), I met a summer student named Jonathan Sick. He is a very good Python Programmer and he made a script that allows one to add a astro-ph page or ads page to the bibdesk program. It's clever.

I never got it to work.

Then I saw another tweet by him today and gave it another go.

oh myyyyy how have I been getting along without this?
Smooth ingestion of everything I wanted to save on astroph. Much easier bibliography work.

*staples fingers* eeexcellent

Friday, November 21, 2014

Hubble Cycle-26 TAC results

Thoughts on the Hubble Space Telescope Cycle-26 results.

The results on the HST cycle-26 are out and there has been a lot of hay made of the improvement in gender balance in the results. It made me think about how the decision process in groups goes. I was reading how in times of stress (say a <3 day decision making process) people fall back on “what everyone knows” and “common knowledge”.

The argument was that people let some rote or authoritarian argument sway them if they are overwhelmed. And I think that is a decisive factor in the imbalance in previous cycles. When overwhelmed, consciously or unconsciously, go with the team you’ve heard of. Not necessarily your mates etc but people you know.

So I think it’s great that the information is now not being passed on to the TAC. Other organizations are similarly exploring double blind evaluations. ESO did an experiment with crowd sourced TAC. Interesting and much improved feedback (in my experience). Though no double blind yet.

Oh and only if funding agencies would do the same. They still ask for massive amounts of personal information. Resume, letters, publication lists etc etc. why? Why not simply focus on the science? And just that? I would be *very* interested to see if that would change the gender balance of NASA, NSF and ERC rewards.

But I also want to see what the next Hubble TAC result looks like. This cycle was...a mess. An experiment...perhaps successful...but massively oversubscribed. The experiment needs confirmation.

Thursday, November 20, 2014

z~8 galaxy sizes

Got sick of struggling with SED fits of dwarfs... so I worked on z~8 galaxy sizes. I had done *much* more that I remembered. The paper is 80% done?

And it was Robin's first day at work with UNAWE! woo!

Trip to Tucson is approved and booked. Only now to have a get my talk together.

All I so far is Scrat images... those will come in handy.

Wednesday, November 19, 2014

Clearing the hopper. Co-authors! C'mon. Going for a personal best

latest draft of a small paper on the link between morphology and normalized specific angular momentum with Danail. Email with Australia. so. 24hr cycle.

Trying to figure out what is going on with my SED fits of Brown Dwarfs in CANDELS.

Proofs of second-ever Supernova paper are in. Third supernova paper needs to go back to MNRAS.
(waiting for co-author).

Trip to Tucson is coming down to the wire. hope someone will book me a flight (likely to be through Novisibirsk at this rate...) soon...
I would like to give my first ever invited talk.

Preparing talk. So far I have a lot of images of Scrat.
How I thought of the Philae lander. Yeah that tweet got more retweets than *everything I've done before*. Go scrat...



Monday, November 17, 2014

IAU invite

Sooo. I got an invitation to speak at the IAU in Hawaii next year.
This is nice. I hear Hawaii is a nice place. All is good.

End of announcement. Back to paper writing.

Tuesday, November 4, 2014

4 years in the making...SNIa and their host extinction properties

Back in South Africa (a postdoc, a kid, two moves, a continent, and a lifetime ago) I worked with a summer student, Adam Reynolds, on the possibility of using the SDSS-III data to use Supernovea Type Ia (SNIa) as probes of the ISM. The idea worked in principle on the --then public-- SDSS-III data but to be really convincing, we'd need their final data-release.
This was a few years in the making but finally arrived in 2014 (huzzah!). I dug out Adam's report and converted it to a small paper (script everything folks...). My interest is of course the distribution of extinction values, could not give a fig about SNIa...

The results are interesting. SNIa effectively "see" a dust screen (inclination dependence is cos(i), nothing more fancy) and their distribution is roughly exponential. The latter is weird, based on everything I've seen (PHAT survey results, occulting galaxy pairs), one would expect a log-normal one. I suspect that is what it really is but right now an exponential is used in SNIa measurements.

The gist of the current paper is Host Galaxy Inclination Matters!

So constructing the next version of SNIa cosmology, should start by taking disk inclination into account as the first host observable to be included. My second guess is host mass. But that is the second paper (already submitted...)

After that, one would need to redo the lightcurves with different extinction priors. To provide the templates for that, I got my STARSMOG HST SNAP program. The ultimate goal is here to reduce the errors in SNIa to below 1%. That'll keep them competitive as a cosmological tool.

That'll keep me busy. All that from a side-interest...

you can find the paper here:

SNIa Host Galaxy Properties and the Dust Extinction Distribution

Supernovae Type Ia display a complex relation with their host galaxies. An important prior to the fit of the supernovae's lightcurve is the distribution of host galaxy extinction values that can be encountered. The SDSS-SN project has published light curve fits using both MLCS2k2 and SALT2. We use the former fits extinction parameter (AV) to map this distribution of extinction values.
We explore the dependence of this distribution on four observables; the inclination of the host galaxy disk, radial position of the supernova, redshift of the supernova and host, and the level of star-formation in the host galaxy. The distribution of AV values encountered by supernovae is typically characterised by: N0 eAV/τ, with τ= 0.4 or 0.33.
We find that the inclination correction using an infinitely thin disk for the SNIa is sufficient, resulting in similar exponential AV distributions for high- and low-inclination disks. The AV distribution also depends on the radial position in the disk, consistent with previous results on the transparency of spiral disks. The distribution of AV values narrows with increased star-formation, possibly due to the destruction or dispersion of the dusty ISM by stellar winds prior to the ignition of the supernova.
In future supernova searches, certainly the inclination of the host galaxy disk, should be considered in the construction of the \av \ prior with τ=0.4/cos(i) as the most likely prior in each individual host galaxy's case.