UFig Simulations and Data Products for the PAUS Collaboration

Summary

Using Ultra Fast Image Generator (UFig; Bergé et al., 2013), we have simulated all the 40 NB images of 12 PAUS fields in the COSMOS area.

See the section below for the conditions of use.

Image simulations

The images are simulated with UFig. UFig uses a hybrid simulation approach, simulating stars directly on the pixel grid, and the galaxies by sampling the distributions photon-by-photon. This approach, numerical optimizations, and tools such as HOPE make the image simulation process ultra-fast. Its performance (~3min for a 0.5 sq. deg image) in simulating one image is comparable to running Source Extractor on this image.

At a given fixed field (e.g., centered on RA:09h 59m DEC:+02 05’), we simulate all the 40 NB images. We also provide the detection images.

Exposure times, gains, saturation levels, magnitude zero-points, seeings, background noise amplitudes and sigmas and astrometric properties for each simulated image are taken from the corresponding PAUS real ones.

Galaxy and stars positions, sizes (rescaled due to the different pixel scale), light profiles and redshifts of each object are taken from the output catalogue of the simulated COSMOS Subaru image overlapping with the considered PAUS image.

Subaru galaxy magnitudes are computed assigning a weight to 5 SDSS template spectra and then integrating in the corresponding waveband. In PAUS, we use, object by object, the same coefficients of the template spectra, but integrating in the correspondent NB.

Stars are only characterized by their magnitude, which is sampled from a distribution obtained through the Besancon stellar population model. Their profile is described by a spatially constant Moffatt function with beta = 3.5. Positions and sizes (rescaled) of the stars are taken from those in the simulated Subaru images (also taken from the Besancon catalogue). Star magnitudes are kept fixed at those of the simulated Subaru images.

More details on the image simulation can be found in Tortorelli et al. 2020, Tortorelli et al. 2018, Herbel et al. 2017 and Bergé et al. 2013.

Data products

Single Epoch Catalogues in FITS format are obtained from SExtractor dual image mode run on PAUS real and simulated images. Source detection is performed using COSMOS Subaru images. Sources which are less than 200 pixels from edges of each image are removed. Photometric zero-points and seeing FWHM for SExtractor runs are taken from the PAUdm using the information of the exposure number + ccd number. Catalogue filenames have the same structure of reduced images filenames. The following are the catalogue parameters:

NUMBER #Running object number
FLUX_ISO #Isophotal flux [count]
FLUXERR_ISO #RMS error for isophotal flux [count]
MAG_ISO #Isophotal magnitude [mag]
MAGERR_ISO #RMS error for isophotal magnitude [mag]
FLUX_ISOCOR #Corrected isophotal flux [count]
FLUXERR_ISOCOR #RMS error for corrected isophotal flux [count]
MAG_ISOCOR #Corrected isophotal magnitude [mag]
MAGERR_ISOCOR #RMS error for corrected isophotal magnitude [mag]
FLUX_APER(5) #Flux vector within fixed circular aperture(s) [count]
FLUXERR_APER(5) #RMS error vector for aperture flux(es) [count]
MAG_APER(5) #Fixed aperture magnitude vector [mag]
MAGERR_APER(5) #RMS error vector for fixed aperture mag. [mag]
FLUX_AUTO #Flux within a Kron-like elliptical aperture [count]
FLUXERR_AUTO #RMS error for AUTO flux [count]
MAG_AUTO #Kron-like elliptical aperture magnitude [mag]
MAGERR_AUTO #RMS error for AUTO magnitude [mag]
FLUX_PETRO #Flux within a Petrosian-like elliptical aperture [count]
FLUXERR_PETRO #RMS error for PETROsian flux [count]
MAG_PETRO #Petrosian-like elliptical aperture magnitude [mag]
MAGERR_PETRO #RMS error for PETROsian magnitude [mag]
FLUX_BEST #Best of FLUX_AUTO and FLUX_ISOCOR [count]
FLUXERR_BEST #RMS error for BEST flux [count]
MAG_BEST #Best of MAG_AUTO and MAG_ISOCOR [mag]
MAGERR_BEST #RMS error for MAG_BEST [mag]
FLUX_WIN #Gaussian-weighted flux [count]
FLUXERR_WIN #RMS error for WIN flux [count]
MAG_WIN #Gaussian-weighted magnitude [mag]
MAGERR_WIN #RMS error for MAG_WIN [mag]
KRON_RADIUS #Kron apertures in units of A or B
PETRO_RADIUS #Petrosian apertures in units of A or B
BACKGROUND #Background at centroid position [count]
THRESHOLD #Detection threshold above background [count]
FLUX_MAX #Peak flux above background [count]
ISOAREA_IMAGE #Isophotal area above Analysis threshold [pixel**2]
XPEAK_IMAGE #x-coordinate of the brightest pixel [pixel]
YPEAK_IMAGE #y-coordinate of the brightest pixel [pixel]
XWIN_IMAGE #Windowed position estimate along x [pixel]
YWIN_IMAGE #Windowed position estimate along y [pixel]
ALPHAWIN_J2000 #Windowed right ascension (J2000) [deg]
DELTAWIN_J2000 #windowed declination (J2000) [deg]
X2WIN_IMAGE #Windowed variance along x [pixel**2]
Y2WIN_IMAGE #Windowed variance along y [pixel**2]
XYWIN_IMAGE #Windowed covariance between x and y [pixel**2]
AWIN_IMAGE #Windowed profile RMS along major axis [pixel]
BWIN_IMAGE #Windowed profile RMS along minor axis [pixel]
THETAWIN_IMAGE #Windowed position angle (CCW/x) [deg]
MU_THRESHOLD #Detection threshold above background [mag * arcsec**(-2)]
MU_MAX #Peak surface brightness above background [mag * arcsec**(-2)]
FLAGS #Extraction flags
FWHM_IMAGE #FWHM assuming a gaussian core [pixel]
ELLIPTICITY #1 - B_IMAGE/A_IMAGE
CLASS_STAR #S/G classifier output
FLUX_RADIUS #Fraction-of-light radii [pixel]

Note: Columns without a value are always set to 99.

We also include the input UFig catalogues (*gal.cat) for each simulated image. The parameters are:

ID #ID object

x #x pixel coordinate

y #y pixel coordinate

z #redshift of the object

sersic_n #sersic index

n_phot #number of photons rendered for each object

int_r50 #input half-light radius from size distribution

int_mag #input apparent magnitude from luminosity function

int_e1 #input ellipticity e1 from ellipticity distribution

int_e2 #input ellipticity e1 from ellipticity distribution

gamma1 #not used

gamma2 #not used

e1 #galaxy ellipticity e1

e2 #galaxy ellipticity e2

mag #apparent magnitude

r50 #half-light radius

kappa #not used

psf_beta #Moffat profile beta

psf_fwhm #PSF fwhm at galaxy position

psf_r50 #PSF half-light radius at galaxy position

psf_e1 #PSF e1 at galaxy position

psf_e2 #PSF e2 at galaxy position

template_coeffs #linear combination coefficients of template spectra

abs_mag #absolute magnitude from B-band luminosity function

blue_red #blue(1), red(0) galaxy separation

RA #Right Ascension (J2000)

DEC #Declination (J2000)

Getting the data

We provide the line of code to download all files:


        wget --wait=0.1 -r -v -nH --cut-dirs=3 --no-parent -nc --show-progress --reject=“index.html*” -o [LOGFILENAME].log https://people.phys.ethz.ch/~ipa/cosmo/pau/data

For individual file download, go to https://people.phys.ethz.ch/~ipa/cosmo/pau/data

Conditions of Use

Luca Tortorelli: torluca@phys.ethz.ch
Alexandre Refregier: alexandre.refregier@phys.ethz.ch

If used in publication, please cite papers in References.

Credits: Luca Tortorelli, Claudio Bruderer, Alexandre Refregier

Contact

For further information on data products and image simulations, please contact:

Luca Tortorelli: torluca@phys.ethz.ch

References

Tortorelli et al. 2020: "Measurement of the B-band Galaxy Luminosity Function with Approximate Bayesian Computation"
Tortorelli et al. 2018: "The PAU Survey: A forward modeling approach for narrow-band imaging"
Bergé et al., 2013: "An Ultra Fast Image Generator (UFIG) for wide-field astronomy"