KiDS-450 Weak Lensing data

The weak lensing data set based on the KiDS DR3 data set is also called the "KiDS-450" data set. Based on the same input data set as the main DR3 release, the KiDS-450 data set encompasses 454 survey tiles. Masking of bright stars, satellite trails and other image defects is already applied in the release catalog, resulting in an effective area of 360.3 sq.deg. Only galaxies with reliable shape measurements are included in this catalog. Unreliable sources have been rejected based on several criteria listed in Hildebrandt & Viola et al. (2017, MNRAS, 465, 1454). The faint limit of the catalog is r = 25.0, corresponding to a typical signal- to-noise of 5σ. The catalog contains a total of 14,650,348 sources, is made up of 455 files (454 data files and 1 metadata file), and has a total data volume of 4.14 GB

For the galaxy shape measurements, the r-band data was processed using a lensing optimized pipeline based on the THELI data reduction system and the lensfit shape measurement software. Source detection was done on stacked r-band images, after which galaxy shapes were measured on individual sub-exposures. The included photometry and photometric redshifts are based on the 4-band ugri image stacks from the main DR3 release.

Note: Hildebrandt & Viola et al. (2017, MNRAS, 465, 1454) found in their cosmic shear analysis that small residual c-terms (non-zero average shear) are present in the catalog. These were subtracted per patch and per tomographic bin before computing the shear-shear correlation functions. Any science analysis for which c-terms are important should determine these from the data, using the lensfit weights to compute the averages.

Cosmic shear data products

A cosmological analysis based on this data set has been published in Hildebrandt & Viola et al. (2017). Several cosmic shear data products are available from the KiDS-450 webpage:

KiDS-450 paper and data access

Release notes

For further details regarding the catalog and its creation please see the release notes.

Data access

The weak lensing data set based on the KiDS DR3 data set is also called the "KiDS-450" data set. The weak lensing shear catalog based on this data set is publicly available via the following web services:

ESO Catalog facility query interface (Weak lensing shear measurements, version 1)
CADC

Via the "Graphical Search Tool" on the CADC query page the THELI-reduced r-band image stacks and their weight, sum and flag maps can also be retrieved. Single exposure data is available on request.

Apart from the above catalog query interfaces, the FITS files with the catalog data (454 data files and 1 metadata file) can be downloaded with wget using the input file linked below. The md5sums file can be used to verify your downloaded files with the command "md5sum -c kids_dr3.1_shear_md5sums.txt"

Catalog format

Please note that the source ID's in this catalog are not always identical to those in the DR3 multi-band catalog, due to the independent astrometric solution used for the lensing analysis.

The table below lists the columns provided in the KiDS-450 or KiDS-DR3.1 lensing shear catalog. For a number of columns additional information is provided:

  • KIDS_TILE: name of the KiDS survey tile in which the source is located. Searching the ESO archive with this OBJECT name will link to further data products for this tile.
  • THELI_NAME: name for the survey tile in the THELI pipeline; for scripting reasons the '.' and '-' characters are replaced with 'p' and 'm', respectively.
  • MASK: bit mask indicating sources affected by different types of defects. Automatic and manual masks produced during data processing flag areas affected by bright stars (e.g. saturated pixels, reflection halos, diffraction and readout spikes) and other severe image defects. In the released catalog the most strongly affected regions are already removed, leaving only sources with reliable measurements. As a result, only the bit mask values 2 (faint stellar reflection halo), 64 (u-band Astro-WISE manual mask), 128 (g-band Astro-WISE manual mask), 256 (r-band Astro-WISE manual mask) and 512 (i-band manual mask) are present.
  • Flag: SExtractor extraction flag. Many sources that are flagged during source detection are removed from the catalog based on the lensfit results because they do not provide reliable shape measurements. The flag values that are still present in the catalog are 1 (the object has neighbors, bright and close enough to significantly bias the photometry, or bad pixels (more than 10% of the integrated area affected), 2 (the object was originally blended with another one) and 16 (objects aperture data are incomplete or corrupted).
  • SG_FLAG: star-galaxy separator based on analysis of the second and fourth order image moments of the source; 0 = star, 1 = galaxy
  • MAG_u/g/r/i: the magnitudes are based on Gaussian Aperture and Photometry (GAaP) measurements and are dereddened and color-calibrated using stellar locus regression. Note: these aperture magnitudes are mainly intended for color measurements, since they only probe the central regions of the source. They are not total magnitudes, except in the case of unresolved or point sources.
  • MAG_LIM_u/g/r/i: local limiting magnitude, defined as the magnitude corresponding to a flux equal to the 1σ flux error.
  • ZPT_offset: the magnitudes reported in this catalog have been color-calibrated, but their absolute calibration has only been homogenized per survey tile, not over the full area. Based on a comparison of the $r$-band magnitudes with the GAIA DR1 \citep{gaia/etal:2016} we provide these additional photometric offsets that can be used to homogenize the photometry over the whole catalog. The reported offsets are with respect to the GAIA photometry, but can be used to calibrate the photometry to the SDSS photometric system as follows:
    mag_u/g/r/i_homogenized = mag_u/g/r/i - ZPT_offset + 0.049
    Note: if used, these offsets must be applied to the magnitudes in all filters!
  • T_B: the best-fit spectral template for each source; these values correspond to the following types, where fractional types can occur because the templates are interpolated: 1=CWW-Ell, 2=CWW-Sbc, 3=CWW-Scd, 4=CWW-Im, 5=KIN-SB3, 6=KIN-SB2 (Capak, 2004, PhD. thesis, Univ. Hawai'i).
  • ODDS: a measure of the uni-modality of the redshift Probability Distribution Function; a higher value indicates a higher reliability of the best photo-z estimate.
  • fitclass: lensfit object class; the only classes included in the catalog are 0 (galaxy, no issues) and -9 (large galaxy, overfills 48 pixel postage stamp size). The latter class is retained to avoid ellipticity selection bias in the brightest galaxy sample.
  • n_exposures_used: the number of r-band sub-exposures for which lensfit measured the shape of the source. Due to the dither pattern, or near tile edges, some sources are only present in a subset of the 5 sub-exposures.
  • e1 or e2: lensfit shear estimators. Note: the e2 component is defined relative to the RAJ2000, DECJ2000 grid; depending on the user's definition of angles in this reference frame, the sign of \verb e2 may need to be changed.
  • PSF_e1 and PSF_e1_exp[k]: model PSF ellipticities at the location of the object, in this case the real part of εPSF. PSF_e1_exp[k] is the PSF ellipticity in sub-exposure number k, while PSF_e1 refers to the average over all exposures used.
  • m: the multiplicative shear calibration correction which should be applied in an ensemble average, rather than on a galaxy-by-galaxy basis \citep[see ][]{fenechconti/etal:2016}. Averaged catalogd ellipticities should be divided by 1+, where the ellipticities should be weighted with the lensfit weight.
LabelFormatUnitDescription
ID25ASource identifier
RAJ2000 1D deg Right ascension of barycenter (J2000)
DECJ2000 1D deg Declination of barycenter (J2000)
Patch 3A Patch (G9, G12, G15, G23 or GS)
SeqNr 1J Running object number within the patch
KIDS_TILE 16A Name of survey tile
THELI_NAME 16A THELI name for the tile
MASK 1J Mask value at the object position
SG_FLAG 1E Star-galaxy separator (0=star, 1=galaxy)
KRON_RADIUS 1E pixel Scaling radius of the ellipse for magnitude measurements
Xpos 1E pixel Object position along x in the r-band THELI stack (non unique)
Ypos 1E pixel Object position along y in the r-band THELI stack (non unique)
FWHM_IMAGE 1E pixel r-band FWHM assuming a Gaussian core
FWHM_WORLD 1E deg r-band FWHM assuming a Gaussian core
Flag 1J r-band SExtractor extraction flags
FLUX_RADIUS 1E pixel r-band half-light radius
CLASS_STAR 1E r-band SExtractor S/G classifier output
MAG_u 1E mag Magnitude in the u-band (GAaP, dereddened)
MAGERR_u 1E mag Magnitude error in the u-band
MAG_g 1E mag Magnitude in the g-band (GAaP, dereddened)
MAGERR_g 1E mag Magnitude error in the g-band
MAG_r 1E mag Magnitude in the r-band (GAaP, dereddened)
MAGERR_r 1E mag Magnitude error in the r-band
MAG_i 1E mag Magnitude in the i-band (GAaP, dereddened)
MAGERR_i 1E mag Magnitude error in the i-band
MAG_LIM_u 1E mag Limiting magnitude in the u-band
MAG_LIM_g 1E mag Limiting magnitude in the g-band
MAG_LIM_r 1E mag Limiting magnitude in the r-band
MAG_LIM_i 1E mag Limiting magnitude in the i-band
ZPT_offset 1E mag Zeropoint offset derived from GAIA DR1 G-band photometry
Z_B 1E BPZ best redshift estimate
Z_B_MIN 1E Lower bound of the 95% confidence interval of Z_B
Z_B_MAX 1E Upper bound of the 95% confidence interval of Z_B
T_B 1E Spectral type corresponding to Z_B
ODDS 1E Empirical ODDS of Z\_B
fitclass 1I lensfit: fit class
bias_corrected_scalelength 1E pixel lensfit: galaxy model scale length
bulge_fraction 1E lensfit: galaxy model bulge-fraction B/T
model_flux 1E counts lensfit: galaxy model flux
pixel_SNratio 1E lensfit: data S/N ratio
model_SNratio 1E lensfit: model S/N ratio
contamination_radius 1E pixel lensfit: distance to nearest contaminating isophote
PSF_e1 1E lensfit: PSF model mean ellipticity e1
PSF_e2 1E lensfit: PSF model mean ellipticity e2
PSF_Strehl_ratio 1E lensfit: PSF model mean pseudo-Strehl ratio
PSF_Q11 1E lensfit: 2nd order brightness moment Q11 of the PSF
PSF_Q22 1E lensfit: 2nd order brightness moment Q22 of the PSF
PSF_Q12 1E lensfit: 2nd order brightness moment Q12 of the PSF
n_exposures_used 1E lensfit: number of r-band exposures used in lensfit measurements
PSF_e1_exp1 1E lensfit: PSF model ellipticity e1 of exposure 1
PSF_e2_exp1 1E lensfit: PSF model ellipticity e2 of exposure 1
PSF_e1_exp2 1E lensfit: PSF model ellipticity e1 of exposure 2
PSF_e2_exp2 1E lensfit: PSF model ellipticity e2 of exposure 2
PSF_e1_exp3 1E lensfit: PSF model ellipticity e1 of exposure 3
PSF_e2_exp3 1E lensfit: PSF model ellipticity e2 of exposure 3
PSF_e1_exp4 1E lensfit: PSF model ellipticity e1 of exposure 4
PSF_e2_exp4 1E lensfit: PSF model ellipticity e2 of exposure 4
PSF_e1_exp5 1E lensfit: PSF model ellipticity e1 of exposure 5
PSF_e2_exp5 1E lensfit: PSF model ellipticity e2 of exposure 5
e1 1E lensfit: galaxy e1 expectation value
e2 1E lensfit: galaxy e2 expectation value
weight 1E lensfit: inverse variance shear weight
m 1E Multiplicative shear calibration