challenges in scientific computing and data analysis

Matthias Steinmetz

Modelling the Milky Way:Modelling the Milky Way:

316 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

Can

we

form disk

galaxies?

416 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

Not really

Formation of disks has been notoriously difficultFeedback?Resolution?Numerical Methods?

AMR vs

SPH

Remember: on galacticscales, hydrodynamicsis an approximation (probably) !What is the mass of MW-typeDM halo

Abundance M~2.51012 MStellar dynamics M~1012 M

Agertz

et al, 2010

516 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

Vcirc

and Vesc

from SDSS

Xue

et al 2008

RAVE

12

21.018.0

10

82.0

= +ACnoM

M12

40.030.0

10

21.1

= +ACMM

616 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

Compare SDSS stellar mass function with DM halo mass function

M12105.2 =MWM

Forero

et al 2009Guo

et al 2009

716 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

We still do not really understand how to form disk galaxies

Guo

et al 2009

Agertz

et al, 2010

816 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

Most of the angular momentum comes in late and thus at low density

Navarro & MS 1997

916 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

The local universe is not a representative part of the universe. The MW is situated in a region of relatively low density and with large nearby mass concentrations like Virgo, the local supercluster, Perseus

Pisces and Coma

1116 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

Local Group alikes

are pretty rare!

analysis of 90 Mpc box, constrained simulationWMAP5 normalizationone excellent candidate

Virgo:

Mass: 1.21014 M

(more massive)

distance: 14.9 Mpc

(19.0 Mpc

NED)

Fornax:

mass: 4.2 1013 M

(7.01013 M

ApJ

548, L139)

distance: 19.2 Mpc

(17.6 Mpc

NED)

Local group

mass: 3.01013M

(lower mass end)

MW/Andromeda distance: 690kpc (700kpc)

1216 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

1316 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

CLUES

1416 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

Metz, Kroupa

& Jerjen

2009

complete!

1516 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

Libeskind

et al 2005

6 DM only simulations, populated with galaxies semi-

analytically.

1616 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

MWM31

Preferential

infall

of satellites

Libeskind

et al, 2011

1716 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

MWM31

Preferential

infall

of satellites

Quadrupole: infall

along

a filamentPointing

towards

Virgo

1816 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

The

MW is

not

a typical

galaxy

Forero

et al, 2011

1916 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

Doumler

et al, 2011

2016 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

Doumler

et al, 2011

2116 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

How about observations?

2216 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

The Milky Way is all around us!

2316 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

How about observations?

The substructure crisisSystematic searches using large imaging surveys (SDSS) reveal a considerable (sufficient?) number of new satellites

Substructure in phase space tracing the formation history of the Milky Way

Lot of activities following the discovery of Sagittarius dwarf

ImagingSDSS, PanStarrs, Euclid, LSST

SpectroscopySDSS, RAVE, GAIA, HERMES, 4MOST

AstrometryHipparcos & GAIA

2416 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

Velocity Substructure in RAVE

Williams et al, 2010

2516 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

Velocity Substructure in RAVE

Williams et al, 2010

2616 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

Illustrative simulations

Williams et al, 2010

2716 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

Recent disruption event

Williams et al,2010

2816 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

Mary Williams http://www.rave-survey.aip.de/rave/

2916 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

Outlook: The GAIA epoch

3016 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

The Future: GAIA

Cornerstone mission of ESAScheduled for launch in late 2012Main objective: To create the largest and most precise three dimensional chart of our Galaxy by providing unprecedented positional and radial velocity measurements for about one billion stars in our Galaxy and throughout the Local Group.

3116 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

Schedule

ProposalConcept

& Technology StudyMission Selection

Re-Assessment

StudyPhase

B1

Scientific

operation

Launch late 2012

Final

Studies

Mission Data Processing

Implementation

Data Processing

Definition

Operation

Mission Products Intermediate

Selection of Prime Contractor (EADS Astrium)

Phase

B2Phase

C/D

Software Development

1995

2000

2005

2010

2015

2020

19941993

1997

1998

1999

2019

20182017

2016

2014

2013

2012

2011

2009

2008

2007

2006

2004

2003

20022001

1996

2021

NowFigure courtesy Michael Perryman and Franois Mignard

3216 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

Light Bending in Solar System

Light bending in microarcsec, after subtraction of the much larger effect by the Sun

3316 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

The Future: GAIA

RAVERAVE

3416 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

GAIA

3516 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

One Billion Stars in 3-D will Provide

in our Galaxy the distance and velocity distributions of all stellar populationsthe spatial and dynamic structure of the disk and haloits formation historya detailed mapping of the galactic dark-matter distributiona rigorous framework for stellar structure and evolution theoriesa large-scale survey of extra-solar planets (~15,000)a large-scale survey of Solar System bodies (~250,000)

and beyonddefinitive distance standards out to the LMC/SMCrapid reaction alerts for supernovae and burst sources (~20,000)QSO detection, redshifts, microlensing structure (~500,000)fundamental quantities to unprecedented accuracy: to 210-6 (210-5 present)

Fundamental physics - reference frame - solar system - extrasolarplanets -stellar systems - stellar physics - Galactic astronomy - quasars and galaxies.

3616 D

ec

2010

The

Futu

re o

f Ast

roco

mpu

ting

challenges in scientific computing and data analysis Slide Number 2Can we form disk galaxies?Not really Vcirc and Vesc from SDSSCompare SDSS stellar mass function with DM halo mass functionWe still do not really understand how to form disk galaxiesMost of the angular momentum comes in late and thus at low densitySlide Number 9Slide Number 10Local Group alikes are pretty rare!Slide Number 12CLUESSlide Number 14Slide Number 15Preferential infall of satellitesPreferential infall of satellitesThe MW is not a typical galaxy Slide Number 19Slide Number 20How about observations?The Milky Way is all around us!How about observations?Velocity Substructurein RAVEVelocity Substructurein RAVEIllustrative simulationsRecent disruption eventSlide Number 28Slide Number 29The Future: GAIAScheduleLight Bending in Solar SystemThe Future: GAIAGAIAOne Billion Stars in 3-D will ProvideSlide Number 36