2016 NYC Gaia Sprint

final wrap-up session

Acknowledgements

• This project was developed in part at the 2016 NYC Gaia Sprint, hosted by the Center for Computational Astrophysics at the Simons Foundation in New York City.
• This work has made use of data from the European Space Agency (ESA) mission Gaia (http://www.cosmos.esa.int/gaia), processed by the Gaia Data Processing and Analysis Consortium (DPAC, http://www.cosmos.esa.int/web/gaia/dpac/consortium). Funding for the DPAC has been provided by national institutions, in particular the institutions participating in the Gaia Multilateral Agreement.
• Details: http://gaia.esac.esa.int/documentation/GDR1/Miscellaneous/sec_credit_and_citation_instructions.html

​

Future Gaia Sprints

• 2017 Heidelberg Gaia Sprint
• tentative dates: 2017 July 17-21
• tentative location: MPIA / Haus der Astronomie
• 2018 Gaia Sprint
• summer 2018 (for after DR2)
• conceivably Leiden or back to Heidelberg

morning wrap-up

Using Gaia parallaxes to break degeneracies in RAVE spectra

Andrea Kunder, Leibniz Institut für Astrophysik Potsdam (AIP)

RAVE using Gaia distances vs external

RAVE DR5 vs external

RAVE-on vs external

external = RAVE stars with High-Res spectra

Strong gravitational lens quasars in Gaia DR1

Leonidas A. Moustakas, JPL/Caltech

Topcat and the Gaia Archive are incredible.

Searching for the 60-ish anticipated four-image quasars in DR1 is promising.

Global AGN-candidate catalogs have a small but important number of stellar sources that Gaia can help identify.

There are 26 stars with parallax putting them <10pc and no proper motion.

There are tons of quasars behind globular clusters, it’ll be fun to predict microlensing events.

The Gaia Archive

Alcione Mora, ESA-ESAC Gaia SOC

Location: http://gea.esac.esa.int/archive/. Helpdesk https://support.cosmos.esa.int/gaia/

• Data: Gaia main table, TGAS, variables, QSO,, Xmatch, external catalogues
• Functionality: TAP+ (data base, user space, sharing), cross-match
• Bring code to the data: select and refine (ADQL, user tables), then download
• Prepare for DR2 (1+ billion sources). Archive might be the only way forward
• Add ADQL queries to your papers (Brown et al. 2016) ⇒ Reproducibility
• Use it. User demand is a key driver for future developments
• Ask us. Via Helpdesk, if additional support is needed. Suggestions welcome

LMC stellar “halo” with Gaia RR Lyrae

Vasily Belokurov

Peculiarities of counter-rotating RAVE-on stars

Ana Bonaca / Harvard

Python tools for 3-D dust map

Gregory Green (via Doug Finkbeiner)

Brand new for Gaia: dustmaps product available via pip install.

Contains 3-D Bayestar, Planck, SFD, BH, etc. See

http://dustmaps.readthedocs.io/en/latest/installation.html

Interactive tool at

http://argonaut.skymaps.info

Note that all values are “SFD” E(B-V), and must be converted to

Various bands with this table:

Schlafly & Finkbeiner (2011)

High-Velocity stars in RAVE-TGAS

Georges Kordopatis / Leibniz Institut fur Astrophysik, Potsdam, + Wyn Evans + Keith Hawkins + Nathan Leigh+ Andrea Kunder

B_T -V_T (deredenned)

Cumulative MDF

Stars with V_Tot > 350 km/s

[M/H]

Toy model

V_Tot

Hercules stream may not be a resonance feature

Chao Liu/NAOC, China

The Her stream are mostly comprise of the metal-rich or old stars, implying that they may together formed in high SFR region, unlike the local-born stars. Therefore, the Her stream seems not a resonance feature induced by the bar!

Fraction of the Her stream members in Age-Metallicity plane

Her stream

afternoon wrap-up

David Spergel CCA

Job opportunities at CCA:

Postdocs

Associate Research Scholar

Joint ARS/AP with Columbia

Joint Group Leader/Tenured Prof. with SBU

Group Leader

Sabbatical

​

​

​

​

​

​

Lectureship (Assistant Professorship) in Astrostatistics at Cambridge

​

http://www.ast.cam.ac.uk/vacancies

​

Also, November AAS webpage.

​

​

A Probabilistic Approach to Fitting Period-Luminosity Relations and Validation of Gaia Parallaxes

Branimir Sesar, Max Planck Institute for Astronomy

No significant global offset in TGAS parallaxes

“Homework for Dynamicists” inspired by Gaia

Kathryn V Johnston, Columbia University

“My” homework - study partners welcome:

• Understanding outside of equilibria: signatures of chaos and regularity apparent over short timescales and as function of phase.
• Tools - modeling: perturbations from integrability in space and time; numerical integrations.
• Tools - model-data comparison: machine learning tuned to expected physical signatures?
• Homework to understand the etiquette of assigning homework

Predicting actions from stellar ages & metallicity

Melissa Ness, MPIA

lz

Jz

Age

Jz

JR

lz

Age

Model made to data by Morgan Fouesneau (MPIA)

Data: 2000 red clump stars: APOGEE-TGAS

Actions, abundances, ages

Oort constants from TGAS

Jo Bovy

A= 14.8, B= -12.2, C= -3.0 +/- 0.7, K= -4.8 +/- 1.7, Vc/R0= 27+/-1 km/s/kpc, Vc = 218 +/- 8 km/s

To do: Measure v_asy and h_sigma to correct A and B

Gaia DR1: the beauty of a limited data release

Anthony G A Brown, Leiden University

• Stimulates creativity!
• Great feedback for DPAC
• Constructive feedback greatly appreciated
• Get in touch if you have questions on the data and/or its interpretation
• Also let your DPAC colleagues know you are happy with their work

Adrian M. Price-Whelan

Princeton University

Fitting Variable Star Period-Luminosity Relations ‘Properly’

Victoria Scowcroft, University of Bath

Now questioning my life choices

Are Red Clump Stars Standard Candles? … with TGAS

Keith Hawkins (simons fellow) Columbia University; with Leistedt; Hogg; Coronado

Model RC as a constant absolute magnitude with some dispersion

(APOGEE RC sample ~250 stars)

Mk = -1.72 +/- 0.03

Sigma_Mk = 0.07 (+0.05, - 0.04)

​

(APOKASC RC sample ~30 stars)

Mk = -1.99 +/- 0.08

Sigma_Mk = 0.09 (+ 0.08,-0.05)

​

​

Laney et al. (2012) found Mk = -1.61 +/- 0.02 ; 0.06 +/- 0.03

Stars and dust in Orion

Eleonora Zari, Leiden Observatory

​

The Galah survey & actions

Sven Buder (MPIA) & The Galah team

GALAH+TGAS 6D information:

RA, Dec, pmRA, pmDec from Gaia DR1 TGAS

(e.g. Michalik et al. 2016)

• D (‘r50’) with MW prior (Astraatmadja et al. 2016)
• Galah radial velocities (Martell et al. 2016)
• Galpy by J. Bovy with tutorial by W. Trick

→ Actions for Galah-TGAS

Gradient for HRD vs. vertical action J_Z?

f(J_Z | T_eff,log(g), [Fe/H])?

How biased are we by the GALAH+TGAS selection?

Trend in RMS!

→ To be found out until next sprint?!

Extra tree model with M. Fouesneau

trying to predict f(J_Z | T_eff,log(g), [Fe/H])

https://

Galah-survey

.org

Gaia Hybrid Catalogs

M. Fouesneau, R. Andrae, C. A. L. Bailer-Jones, (MPIA, Gaia Team)

H.W. Rix (MPIA) & D.W. Hogg (NYU, CCA, MPIA)

Predicting fbol

+ astrophysical parameters

as Gaia/CU8 will

do in DR3

(i.e., MCMC samples)

Exploring Gaia Data

Sergey Koposov (Uni of Cambridge)

• Proper motions
• 3D structure of dwarfs
• Photometry outliers

u-g

u-g

g-r

g-i

G-r

Fitting Stellar Models with isochrones

https://github.com/timothydmorton/isochrones

$ pip install isochrones

• Designed to be easy to use for star-fitting newbies
• Fits stellar properties (mass, age, [Fe/H], distance, extinction) given arbitrary observed quantities (photometric bands, spectroscopic params, parallax…)
• Can handle multiple stars (e.g. can fit blended and/or resolved binaries at the same age, feh, distance, extinction, etc.)

​

• #GaiaSprint: applied to a sample of Kepler stars & other planet hosts (TGAS plax + 2MASS + WISE + APASS)
• Planet hosts & non-hosts
• Wide binary candidates with measured rotation periods

​

Contact Tim (tdm@astro.princeton) with any usage questions!

Timothy Morton (Princeton)

Dan Foreman-Mackey

Sagan Fellow | University of Washington | dfm.io | @exoplaneteer | github.com/dfm

github.com/dfm/gaia-kepler

My week David W. Hogg (NYU) (MPIA) (SCDA) (CCA)

• Run a hack week called the Gaia Sprint.
• Build a data-driven model of the CMD; de-noise parallaxes (with Leistedt).
• requires a challenging Gibbs sampling
• Measure the mid-plane of the disk, tilt and curvature (with Price-Whelan).
• requires a selection function for TGAS
• Determine causal relations among age, [Fe/H], and actions (with Bird, Ness).
• looks like there are differences between low-alpha and high-alpha populations
• “Low-hanging” == “Doesn’t require the selection function”
• We need a selection function!
• input: RA, Dec, G
• output: probability of being in TGAS, and pdf over parallax errors

Attempt at astrometric radial velocity

Semyeong Oh, Princeton University

?

Assumed zero intrinsic velocity dispersion

Jonathan Bird

Vanderbilt University

The Age-Velocity Dispersion Relationship

for APOGEE RC Stars

Data: ~1300 RC stars from APOGEE overlapping with TGAS. Thanks to Ness, Bovy, Gaia.

​

Using generative model for both velocities and ages.

​

Many Thanks: Bovy, Ness, Hogg, Price-Whelan, Sanders; EVERYONE! What a groovy week.

TGAS + RAVE for actions

Johanna Coronado, MPIA

Gradient in vertical action (Jz) across the red clump for LAMOST stars

Current picture for error samples in action space for the distances of RC stars in TGAS + RAVE:

Work in progress: 222 stars with APOGEE labels in TGAS to obtain photometric distances

Photometric Twins in GAIA using APASS

Xmatched with RAVE for stellar parameters

Lauren Anderson Flatiron Fellow, CCA

landerson@simonsfoundation.org

SIMONS FOUNDATION

​

​

TGAS+RAVE-on age-velocity dispersion relation

Age estimates from isochrones for giants and turn-off sample

Jason Sanders, University of Cambridge

Turn-off

Giants

To do: more careful error modelling, remove OB stars that are flagged by RAVE for binarity etc.

Galactic escape speed & axis ratio

Gus Williams, Vasily Belokurov, Andy Casey

Model tail of velocity distribution:

​

p(V) = (Vesc - V)^k (Leonard & Tremaine 1990)

​

Allow Vesc to vary as a function of position => constrain change in

Vesc with radius and the axis ratio of the potential

​

Constrain with BHBs from SDSS with radial velocities and

Sergey Koposov’s revised proper motions...

​

​

​

​

Use photometric BHBs and marginalise over radial velocity (with prior on anisotropy) => factor of 10 more objects so better constraints.

​

​

Escape speed ~400 km/s

Escape speed not varying over volume probed by BHBs => q unconstrained

Vesc

slope

Axis ratio

Andy Casey

A first (bad) attempt at a semi-analytic model for stellar spectra

• APOGEE spectra

​

• Colors give priors on temperature

​

• Gaia gives prior on absolute magnitude

​

• Assume that a linear model is the true model for flux, and use a noise model (not shown):

​

​

​

​

• Require [Fe/H] coefficient to be negative (more MR = deeper lines)

​

• ~50,000 model parameters that need sampling

The Origin of High-velocity Stars

Kevin C. Schlaufman/Carnegie, Princeton, & JHU

Andrew R. Casey/Cambridge IoA

Kinematics-Chemistry-Age / Binaries + Neural Network

Yuan-Sen Ting, Harvard University

Rotating star from Kepler | James Davenport (WWU)

Full paper draft written during #GaiaSprint! github.com/jradavenport/gaia_rot

Low-hanging fruit: match TGAS w/ rotating stars from Kepler. Filter out significant subgiant contamination

Discovery: bimodal period distribution found in K/M dwarfs, extends to F/G’s!

Original finding

(McQuillan+2013)

Lots of future potential: e.g. subgiant rotation evolution,

log(g) calibrations for Kepler targets, improved gyrochronology

Remote participant