bupkiss

I did no research today.

Yale

I spent the day at Yale, where I spoke about dynamics and inference.

baryon acoustic volatility

Eyal Kazin (NYU) spoke today in a long group meeting session about the Baryon Acoustic Feature in data and simulations (or mocks, really, which are simulations of particular data sets). He showed that we actually got lucky that we detected the BAF in SDSS-I. There was a better-than-ten-percent chance that we would have seen no feature at all. And yet when we did detect it, we could measure its location (length scale) to a few percent! This is because the correlation function is very stable in shape, but unstable in amplitude; or because the data points in a plot of the correlation function amplitude as a function of scale are incredibly strongly correlated. Prospects for SDSS-III are good, but there is a lot of cosmic variance to disappoint us.

obituaries done, merger rates measured

I finished my two obituaries for Sam Roweis, one for our high school and one for the American Astronomical Society. You can read the submitted version of the AAS obituary here (PDF). I also spent some time writing in my much longer reminiscence, which I am hoping will be a document that captures what I learned from Roweis in a professional and personal way.

I also spent a long time talking to Tao Jiang, the student who is following up Masjedi's work on merger rates, but now in the full SDSS Main Sample, and thinking about what to do with SDSS-III. I am pretty excited about our results which have very high signal-to-noise.

pixel modeling and giving up

I spent the day talking with Bolton and Lang (and, variously, Blanton and Bovy) about modeling images and spectra. We principally debated what advantages you will or could get in data analysis by building good models of the raw data at the pixel level. We all have intuitions that it will be good, but we have few concrete examples. We assigned Bovy one task along these lines: Show that the SDSS astrometry of point sources could or would be much better if we re-centroid using a PSF fit rather than the (approximate) thing done by the pipelines at present.

In the afternoon, I made the difficult decision to drop my robotic telescope proposal for NYU Abu Dhabi and the NYU Global Campus. This is a great project, which we might resurrect, but I am feeling stretched too far and I need more time to understand my research program in the post-Roweis period. Argh.

obituaries, universes, spectrographs

[Just back from a few days of vacation.]

I attended an interesting and extremely interactive talk about the observability of universe–universe collisions by Tommy Levi (UBC). His calculations are approximate, but give a general idea about what would be observed. Interestingly, there is a cold spot in the WMAP map that could be what he predicts.

In the rest of the day I discussed next-generation spectroscopy projects with Adam Bolton (Utah) and worked on two obituaries for Sam Roweis.

GALEX

Schiminovich and I attempted to re-start our GALEX–SDSS projects today.

fiducial field

Brett Mensch (old friend of Sam's) and I discussed the possibility of blessing a good field for an amateur campaign, aimed at demonstrating the power of the crowd for astronomical discovery (and not just monitoring and the like). He had many good ideas for what would appeal to amateurs and what would make the project easier.

[I have noticed that this diary has devolved from "Hogg's research" to "Hogg's conversations about his research"]

two crazy projects

I spent the day at the IAS working with Lang. We started re-coding our celestial mechanics stuff in C for performance. That was painful. We spent quite a bit of the non-coding parts of the day talking to various people about two of our craziest projects: One is the Comet Holmes project (the Web is a sky survey; working with untrusted data; automated discovery), and the other is my project with Sam to revolutionize spectral data analysis (spectral extraction as data rotation; producing functions not spectra; daisy-chaining probabilistic inference). I can't say that everyone at the IAS was impressed!

double redshifts

Tsalmantza and I spoke about our multiple-redshift search in the SDSS spectroscopy. The new technology we bring is a data-driven model of the spectra; the goal is to increase the number of known lenses. We discussed tests of the model, and the hope that increasing the precision of the model will increase the sensitivity of the system to second (and third) redshifts.

stars, pulsars, dark matter

I can't say I did much research today but I saw two beautiful talks, and going to talks does count as research.

At lunch time Dmitry Malyshev (NYU) gave a beautiful talk on millisecond-pulsar and dark-matter contributions to the observed haze at the Galactic Center from Fermi and WMAP. He showed specific pulsar-plus-DM models that explain the spectral properties of the haze beautifully, many of which are natural for both pulsars and the DM. In some, he had to make the electron–positron emission from pulsars very high, but it really is an unknown. He mentioned that 47 Tuc (globular cluster) is a key observable Fermi source for distinguishing these ideas. Malyshev was very cautious and made no strong claims, but my excitement about the possibility that dark matter annihilation is being observed grew during the presentation.

In the afternoon, Nathan Smith (Berkeley) gave an outstanding talk about extremely massive stars as observed in our own Galaxy and nearby galaxies, including their dramatic explosions and mass-loss episodes. These are incredibly rich in their kinematic and chemical properties and have implications for chemical abundance propagation, star formation, supernova prediction, and the evolution of the young universe. He made a comment at the end about extrapolating theories we don't understand into regimes where we have no data which made the astronomers laugh and the particle theorists ask And?.

unit tests failed

Argh. Lang came into town and we added another JPL-ephemeris-based unit test to our code and it failed. It is a coordinate system problem we weren't able to diagnose before we ran out of day. But we started playing with the Canon Digital Rebels that Sam bought to put on the telescopes we have on the roof of 715 Broadway.

meta-up

Fengji Hou (my new student, will be Hou from now on in this diary), his co-advisor Jonathan Goodman (NYU Courant), and I discussed Fengji's start on exoplanet radial velocity fitting using advanced sampling tools. We spent a long time talking about code, but once we were done, Goodman and I spent some time talking about medium-term projects that would be non-trivial and interesting. We discussed the idea that if you are a Bayesian (not always advisable), you don't really want to detect planets per se, you want to pass forward probabilistic information about their existence and properties, and then perform your analysis on those probabilistic outputs. In this world, you might be able to discover and say things about classes of planets that are not detected clearly in any individual stellar radial velocity time series. Approaches like this could greatly increase the number of known expolanets for some kinds of statistical studies.

Sam, numerical stability

I spent the train ride coming back from Queen's drafting an AAS obituary about Roweis. I decided also that I am going to count scientific reminiscence about Roweis as research for the purposes of this diary, lest I end up with nothing to say!

On the flight back to New York, I followed up an intuition floating around in my head that small changes in what we call the orbital parameters could make the code that converts between phase-space position and standard orbital elements much more numerically stable at edge cases. I found some nice things, and I am confident that there are large improvements to be made. Here's a trivial example: Instead of working with inclination of the orbit and the longitude of the ascending node, work with the sine of the first times the sine and cosine of the second. These two products are just the x and y components of the angular-momentum-direction unit vector. A well-known change is to work with the argument of perihelion plus the longitude of the ascending node and not with argument on its own, but I think I can show that an even better one is to work with the eccentricity times the sine and cosine of that angle sum. And so on. Not sure if we will analyze and implement all these before or after we finish our Holmes project.

Queen's

I spent the day at Queen's University. After talking about Gaia-related matters, most of my time was spent with Larry Widrow and students or Stéphane Courteau and students.

Widrow's students are working on a lot of dynamical problems with disk galaxies and with n-body simulations. Three projects that stood out: His group can make disk galaxies that are (numerical) potential–density pairs and then evolve them forward in time or use them in galaxy–galaxy interaction simulations, even ones that are designed to fit real data. His group has revived (or made a new version of) the made to measure method as a new source of realistic galaxy models. And they are looking for tidal streams (or other phase-space structure) in n-body simulations; that is, treating the simulation outputs as data and searching it for structure.

Courteau's students are working on a range of precise measurements on nearby galaxies, including search for incredibly low surface-brightness structure in and around M31; building maps of the age and metallicity over the surfaces of extremely nearby galaxies, or as a function of radius in less nearby ones; and making much more precise the scaling relations that relate size, luminosity, and rotation velocity. In that latter project, his students do a large part of their radial profile fitting by hand—setting fitting ranges, choosing extrapolation techniques—which Courteau deems absolutely essential. Of course both my readers know how I feel about that! My take is this: If you want to do precise fitting of a simple model to a set of complex galaxies, you have to make a substantial number of choices, and somehow an extremely good student is far, far better than any currently available code for making those choices. Now fixing that is an important problem. That said, the scaling relations they get are amazingly precise.