morphology on large surveys with sextractor + psfexmorphology on large surveys with sextractor +...
TRANSCRIPT
-
Morphology on large surveys with SExtractor + PSFEx
B. Moraes, M. Elidaiana, M. Makler, E. Ber1n,
and the CS82 collabora1on
Aldée Charbonnier
Valongo Observatory Rio de Janeiro
ESO, Chile, November 22th, 2013
-
Outline
SExtractor + PSFEx CS82 survey COSMOS field
-
General procedure
SExtractor: hGp://www.astromaLc.net/soMware/sextractor PSFEx: hGp://www.astromaLc.net/soMware/psfex
3
-
SExtractor v.1
image background filtered
apertures segmentaLon
4
-
PSFEx ID card
Solving in direct space (≠ Fourier) AutomaLcally selects the point like sources PSF modeled as a linear combinaLon of basis funcLons Weighted χ2 minimizaLon PSF varia1ons may be decomposed on a polynomial basis HomogenizaLon is available
5
-
PSF modelling
Automa1c selec1on of point
like sources
PSF modeled over the
field of view
6
-
PSFEx checkplots
Distribu1on of point like sources FWHM
+ others (chi2, ellip1city..)
7
-
SExtractor v.i
Automated detecLon of objects
2D Model-‐fiUng of stars and galaxies -‐ Point sources modeled as local PSF + amplitude -‐ Galaxies: size and orienta1ons convolved with the PSF -‐ Background
PSF modeled using PSFEx Pixel grid size depends on sampling and on the object Models: background, point source, de Vaucouleurs, Sérsic, exponen1al Levenberg-‐Marquardt minimizaLon algorithm IniLal parameter guesses based on SEx-‐v.1 measurements Residuals are non linear chi2 modified for improved robustness
towards outliers
MulL galaxy fit/deblending
BerLn & Arnouts 1996
BerLn (in prep.)
BerLn (in prep.)
8
-
SExtractor v.2 sex -dp!
9
-
SExtractor v.2
Residuals Exponen1al profile CFHT/MegaCam
Output structural parameters: -‐ MAG_MODEL -‐ REFF -‐ ELLIPTICITY -‐ errors -‐ chi2 of fijng -‐ PSF properLes -‐ star/galaxy
discriminators
10
-
Star/Galaxy classifier
CLASS_STAR SPREAD_MODEL
11
-
Pros-‐cons
Fast SExtractor and PSFEx compaLble Robust PSF modeling developed for large area tested on simulaLons (DES …) tested on true data (BCS, CS82 …)
background treatment (e.g. GALAPAGOS
Barden et al. 2012) brightness profiles available
12
-
CFHT – ‘Stripe 82’ data
-
CS82 survey in a nutshell France-‐Canada-‐Brazil project CFHT, MegaCam 173 science quality poinLngs with slight overlaps ‘Stripe 82’ goal: weak and strong lensing survey i-‐band service mode insuring good seeing condiLon
-
CS82 observing condi1ons
CS82 mean seeing = 0.6’’ Depth: ~24.1 in 2-‐arcsec circular apertures Source number density: ~17.25 per arcmin2 (standard cuts)
15
-
CS82 first results 16
-
Morphology catalog produc1on ~ 16,700,000 objects (~ 150,000/1le) Deeper than SDSS Choice of configura1on
v deblending v deepness v computaLon Lme v size of catalogs 4 profiles: de Vaucouleurs, exponenLal, Sérsic, deV+exp 177 Lles x 4 profiles x ~ 12h/profile/Lle Masking Cuts: S/N, flagged objects, star/galaxy
Aspect ra1o before … … and ajer model-‐fiUng
0 0 1 1
Moraes et al. in prep
17
-
SEx-‐PSFEx configura1ons
VIGNET image cut size
es1ma1on of the internal sta1s1cal errors due to the choice of the configura1ons
PSF_SIZE Image size of the PSF
Modeled magnitude (deV) Effec1ve radius (deV)
MINSN PSF star selecLon …
18
-
SPREAD_MODEL as a sanity check
coadds: different seeings, bad es1ma1on of the PSF Point like sources selec1on crucial
19
-
Lensfit: ellip1city comparison
Lensfit: Miller et al. 2007
Effec1ve radius S/N
Magnitude
Very good reconstruc1on (single exposure vs coadds)
20
-
Comparison with SDSS: exp
Galaxies – S/N > 5 – SDSS DR9: fracDeV_i < 0.02 (matching) Area of 14 sq. degrees
Good agreement for MAG < 21
21
-
Comparison with SDSS: deV
Galaxies – S/N > 5 – SDSS DR9: fracDeV_i > 0.98 (matching) Area of 14 sq. degrees
Good agreement for MAG < 21
22
-
BOSS/CS82 galaxies
SDSS BOSS sample
14 sq. degrees
Bias in the BOSS sample
23
-
COSMOS field
-
The COSMOS field CFHT Deep2 field -‐ MegaCam (y band) Hubble COSMOS -‐ ACS (I band)
seeing ~ 1’’ seeing ~ 0.09’’
25
-
CFHT-‐D2: catalog produc1on
# objects aMer masking and S/N > 5: -‐ point-‐like: ~ 38,700 -‐ extended: ~ 174,200
SExtractor run: ~ 30.5 h Same diagnosLcs as CS82
Chi 2 – PSF modelling SPREAD_MODEL
26
-
Catalogs comparison
• COSMOS catalog: produced with GIM2D • CFHT D2 catalog: our pipeline • Objects selec1on: eliminaLon of stars, good quality fit
Good agreement
27
-
Catalogs comparison
Ellip1city and effec1ve radius similar The PSF is correctly reconstructed
Ellip1city Effec1ve radius
28
-
Conclusions: 2 points
SExtractor + PSFEx: EsLmaLon of the PSF, variaLon over the field of view The model-‐fijng is working well Very compeLLve in term of computaLon Lme Extensive tests to assess its quality CFHT – Stripe 82 data: Data of very good quality (mean seeing 0.6”) deep (mag limit ~ 24) Originally a lensing survey Morpholgy catalogs produced (deV, exp, Sersic, deV+exp) RedshiM z=[0.2;0.7] Galaxy evoluLon/morphology studies to come…
-
kecélebackup
-
The ‘Stripe 82’
-
40 30 20 10 0RA
1
0
1
DEC
S82m0m
S82m0p
S82m10m
S82m11m
S82m11p
S82m12m
S82m12p
S82m13m
S82m13p
S82m14m
S82m14p
S82m15m
S82m15p
S82m16m
S82m16p
S82m17m
S82m17p
S82m18m
S82m18p
S82m19p
S82m1m
S82m1p
S82m20p
S82m21m
S82m21p
S82m22m
S82m22p
S82m23m
S82m23p
S82m24m
S82m24p
S82m25m
S82m25p
S82m26m
S82m26p
S82m27m
S82m27p
S82m28m
S82m28p
S82m29m
S82m29p
S82m2m
S82m2p
S82m30m
S82m30p
S82m31m
S82m31p
S82m32m
S82m32p
S82m33m
S82m33p
S82m34m
S82m34p
S82m35m
S82m35p
S82m36m
S82m36p
S82m37m
S82m38m
S82m39m
S82m3m
S82m3p
S82m4m
S82m4p
S82m5m
S82m5p
S82m6m
S82m7m
S82m7p
S82m8m
S82m8p
S82m9m
W4m0m1
W4m0m2
W4m1m1
W4m1m2
W4p1m1
W4p1m2
W4p2m1
W4p2m2
0 10 20 30 40RA
1
0
1
DEC
S82p0m
S82p0p
S82p10m
S82p10p
S82p11m
S82p11p
S82p12m
S82p12p
S82p13m
S82p13p
S82p14m
S82p14p
S82p15m
S82p15p
S82p16m
S82p16p
S82p17m
S82p17p
S82p18m
S82p18p
S82p19m
S82p19p
S82p1m
S82p1p
S82p20m
S82p20p
S82p21m
S82p21p
S82p22m
S82p22p
S82p23m
S82p23p
S82p24m
S82p24p
S82p25m
S82p25p
S82p26m
S82p26p
S82p27m
S82p27p
S82p28m
S82p28p
S82p29m
S82p29p
S82p2m
S82p2p
S82p30m
S82p30p
S82p31m
S82p31p
S82p32m
S82p32p
S82p33m
S82p33p
S82p34m
S82p34p
S82p35m
S82p35p
S82p36m
S82p36p
S82p37m
S82p37p
S82p38m
S82p38p
S82p39mS82p39p
S82p3m
S82p3p
S82p40m
S82p40p
S82p41m
S82p41p
S82p42m
S82p42p
S82p43m
S82p43p
S82p44p
S82p45p
S82p46p
S82p47p
S82p48p
S82p4m
S82p4p
S82p5m
S82p5p
S82p6m
S82p6p
S82p7m
S82p7p
S82p8m
S82p8p
S82p9m
S82p9p
Strategy of observa1on
-‐40° < RA < +45° and -‐1° < Dec < +1° avoiding massive stars
-
CS82 magnitudes
MAG_AUTO – MAG_PSF MAG_MODEL – MAG_PSF
CalibraLon of the data done with MAG_AUTO MAG_AUTO is missing flux at the wings (Annunziatella et al. 2012) offset MAG_AUTO – MAG_PSF observed
-
CS82/SDSS
Point like sources + exp/deV profiles: offset due to the calibraLon with MAG_AUTO deV profile: addiLonal offset due to the truncaLon of the profile by the SDSS pipeline (info. By Claire Lackner)
-
Ellip1city: Lensfit vs Sextractor
Sersic De Vaucouleurs exponen1al
Lensfit
SExtractor
Comparison on CS82 1 1le (p28m)