------------------------------------------------------------
COSMOS GALAXY AND X-RAY GROUP MEMBERSHIP CATALOGS
NASA/IPAC IRSA Release January 2012
Matt George and Alexie Leauthaud
mgeorge@astro.berkeley.edu, alexie.leauthaud@me.com
------------------------------------------------------------

The reference for this release is George et al. 2011 ApJ, 742, 125. 
There are two catalogs in this release, one for group properties and
one for galaxy properties. Information is drawn from a number of
sources:

-ACS imaging catalog (Leauthaud et al. 2007, updated)
-Photometric redshifts (Ilbert et al. 2009, 
     photoz_vers1.7_010809, pdzBay_v1.7_010809.out.gz) 
-X-ray group properties (Finoguenov et al. in prep. [also 2007],
     Xmass_chaxmm10_mie_ext6_id_update1010_zupdate.cat) 
-X-ray group galaxy members (George et al. 2011, ApJ, 742, 125)
-Spectroscopic redshifts (observed_targets_20110513.dat, M. Salvato, 
     used for "ZBEST" membership; redshifts not included in this release)

We include galaxies with MAG_AUTO<24.2 (ACS F814W band) and groups
with z<1. The galaxy magnitude cut is set by the K-band completeness
limit of ~24 since a detection is required for stellar mass
estimates. Photoz uncertainties also rise steeply beyond this
magnitude.

The galaxy catalog contains ACS photometric parameters, stellar
masses, photoz information, and group membership associations. The
group catalog contains X-ray properties, lensing-calibrated masses,
quality flags, aggregate member galaxy statistics, and coordinates for
the center of each group.


----------GROUP CATALOG INFO----------

This is a COSMOS X-ray group membership catalog, combining X-ray group
properties from Finoguenov et al. in prep. (also 2007) (catalog
Xmass_chaxmm10_mie_ext6_id_update1010_zupdate.cat) with estimates for
masses and radii calibrated from weak lensing (Leauthaud et al. 2010),
and member galaxy information (George et al. 2011). Group redshifts have been 
determined by searching for red sequence overdensities within 500 kpc of the 
X-ray centers and are refined by using spectroscopic redshifts when
available. We use groups with z<1 to ensure good optical
identifications and small photoz uncertainties.


----------BASIC PROPERTIES AND X-RAY MEASUREMENTS----------

GROUP_ID		- Group ID from X-ray catalog
REDSHIFT		- Group redshift (specz when available, red sequence otherwise)
REDSHIFT_ERR		- Group redshift uncertainty
LX			- log10[X-ray luminosity 0.1-2.4 keV], from Alexis' catalog
LX_ERR			- log10[uncertainty in LX], from Alexis' catalog
LX_SCALE		- log10[X-ray luminosity/E(z)], from Alexis' catalog
XRAY_FLUX		- log10[extrapolated flux (0.5-2 keV band) within R500 in ergs/s/cm^2], from Alexis' catalog
XRAY_FLUX_ERR		- log10[error on the flux], from Alexis' catalog
RA			- X-ray RA from Sextractor ellipse (cosmos_m46finalmie_ext5.txt)
DEC			- X-ray Dec from Sextractor ellipse (cosmos_m46finalmie_ext5.txt)
XRAY_POS_ERR		- X-ray position uncertainty (32" for XFLAG>1, 32"/significance for XFLAG=1)
NAME			- IAU formatted name COSMOS CL JHHMMSS.S+DDMMSS using coordinates from Alexis' catalog


----------GROUP QUALITY FLAGS----------

Alexis' catalog includes a quality flag based on the confidence in the
optical identification. We include a few additional flags to identify
cases where two groups are overlapping and membership identification
is unclear (FLAG_MERGER), where we do not find a significant
concentration of photoz members (FLAG_POOR), and where member
selection may be affected by masks (FLAG_MASK). FLAG_INCLUDE is meant to
combine these quality flags into a single selection.

XFLAG			- quality flag from X-ray catalog
			   1 - can use X-ray center, all good
			   2 - good ID, but cannot use X-ray center
			   3 - ID needs to be checked spectroscopically
			   (Higher flags not included in this catalog)
			   4 - projection effects, no unique ID
			   5 - ID is potentially wrong
			   6 - ID is ok, but this is not a group
			   7 - no ID possible
			   8 - no photometry is available 
FLAG_MERGER		- merger flag
			   1 - LENSING_R200 overlaps by more than 25%
			   in radius and dz<0.01 with another group
FLAG_POOR		- poorness flag
			   1 if N_MEM <= 3
FLAG_MASK		- mask flag 
			   1 - >10% of area within LENSING_R200 or area within Rscale masked by stars, image defects, or edge of ACS coverage (ACS + Subaru B, V, I, z masks)
FLAG_INCLUDE		- combined quality flag
			   1 - good, FLAG1<3, FLAG_MERGER=0, FLAG_POOR=0, FLAG_PROJECTION=0, FLAG_MASK=0


----------LENSING-CALIBRATED PROPERTIES----------

M200c			- log10[M200] from WL-calibrated LX-M scaling relation (mass within radius where mean density is 200 times critical density)
R200C_MPC		- R200 in Mpc
R200C_AS			- R200 in arcsec, at group redshift
CONC200c		- C200 concentration from Zhao 2009 M-c (C=R200/Rscale)

----------MEMBER STATISTICS----------

NMEM			- Number of objects in galaxy catalog with P_MEM > 0.5 for this group
NMEM_W  		- Sum of P_MEM for objects in galaxy catalog for this group

----------GROUP CENTERS----------

Multiple "centers" are defined, and corresponding positions, galaxy
IDs, stellar masses, and positional uncertainties are saved. From the
lensing signals stacked on these centers, MMGGs appears to be the
best tracer of the center of mass. These results do not imply that 
MMGGs is at the center of mass of every group, however. To study 
central galaxies, one may wish to confirm that a galaxy is both 
the most massive within the scale radius as well as the
most massive in the group.

MMGGs		- most massive group galaxy within Rscale + X-ray positional error of X-ray center, using photoz selection only
MMGGs_ZBEST	- most massive group galaxy within Rscale + X-ray positional error of X-ray center, using speczs when available, else photozs

Each center has a position associated with it, e.g
ALPHA_MMGGs 	  	- RA (J2000) of MMGGs
DELTA_MMGGs		- Dec (J2000) of MMGGs
...

We save the object ID in the galaxy catalog and the galaxy stellar mass, e.g. 
ID_MMGGs  	        - ID for MMGGs galaxy from galaxy catalog
SM_MMGGs		- stellar mass for MMGGs galaxy
...


----------EXCEPTIONAL CASES AND FAQs----------

- MMGGs is not necessarily the most massive member of a group,
since a more massive galaxy can live in the outskirts. It is also not
necessarily the brightest member in any particular band.

- Selection based on spectroscopic redshifts and photometric redshifts
do not always agree so we use both approaches. If you want a uniformly
selected sample, use the photozs. If you favor purity over uniformity,
use the photoz+specz selection ("ZBEST") which replaces photozs with
speczs when available. 

- When a galaxy can belong to multiple groups, the highest membership
probability is saved to P_MEM and the corresponding group ID to
GROUP_ID. Central galaxies are exceptions. Each group has the most
massive galaxy near its X-ray position with membership probability
P>0.5 (MMGGs) locked into the group. P_MEM and GROUP_ID are not
altered for this galaxy, even if the redshift weighting produces a
higher membership probability for a different group. We see no
instances where a galaxy qualifies as MMGGs for multiple groups.

- Not all groups have a member galaxy near the X-ray position. These
groups are flagged and the algorithm currently selects no members.

- A small number of objects do not have photozs because of saturation
or deblending issues. When a spectroscopic redshift is available, we
replace the photoz PDF with the specz convolved with a gaussian of
width equal to the typical photoz uncertainty for the galaxy's
magnitude. In cases where a photoz does not exist because an object
falls in a ground-based imaging mask (e.g. near a bright foreground
star), we do not replace the redshift. Groups with significantly
masked areas are flagged. 

- To first order, the positional dependence of the membership
probability is a step function at R200. However, because the
probability depends on the relative group and field densities, there
is some positional dependence (a higher group density results in
higher membership probabilities for candidate members). If a group has
different centers chosen by the photoz and photoz+specz selections,
the group density estimate can change. Thus, there can be differences
between P_MEM and P_MEM_ZBEST even if a galaxy does not have
a specz. 


----------COSMOLOGICAL PARAMETERS (WMAP5)----------

Omega b		          0.0438000
Omega m 	       	  0.258000
Omega l  	       	  0.742000
H0    		       	  72.0000
Sigma_8  	       	  0.796000


----------EXTERNAL FILES USED----------

X-ray catalog				  - Xmass_chaxmm10_mie_ext6_id_update1010_zupdate.cat
X-ray SExtractor coords			  - cosmos_m46finalmie_ext5.txt
Photoz catalog				  - photoz_vers1.7_010809
Photoz PDFs				  - pdzBay_v1.7_010809.out.gz
Image mask (FLAG_MASK)			  - COSMOS_borders_F814W_mask.fits
Spectroscopic catalog                     - observed_targets_20110513.dat