The Catalina Surveys Data Release 1 (CSDR1) |
1) If an object is within 10-15 degrees of the Galactic plane it will not have Catalina Surveys photometry. This is because the fields are too crowded for our survey. 2) If an object is below Dec~-30 degrees or above Dec~65 degress, the area is not covered by the data release. 3) If the object is brighter than V = 12, the object will not have good data as such bright objects saturate in the images we take. This is the case for many, if not most, variable star archetypes (eg. RR Lyr, Beta Lyrae, R Cor Bor, etc.). |
All Catalina photometry given here was measured using the SExtractor package. As the photometry is apperture-based it allows for both point and extended sources. SExtractor performs well in fields with low source density. In crowded fields, where many sources blended, aperture photometry is inaccurate. For this reason Catalina avoids dense fields near the Galactic plane. |
The Catalina Sky Survey is an NEO survey. Asteroid magnitudes reported by NEO surveys have historically been poor. Survey observations dominate the dataset of observed magnitudes, creating corresponding uncertainties in our understanding of NEO populations as function of H-value. Desiring to address this problem, and mindful that more precise magnitudes could increase the usefulness of our observations for other purposes, we have resolved to improve our photometry. Obtaining good photometry of asteroids is difficult because of the lack of suitable standard stars in any given survey field. Occasionally, images will include a Landolt or Stetson standard field, but often these stars are too bright to use as comparisons. Our unfiltered data further complicates the analysis by introducing color-dependent sources of uncertainty. We assembled a large number of well-distributed comparison stars by constructing a catalog of G0-G8 dwarfs from the large, all-sky 2MASS Point Source Catalog(1). Most asteroids are rather neutral in color, reflecting the light of the Sun with a roughly constant distribution. This fact, combined with the use of solar analogs as comparison stars, can be used to substantially reduce the uncertaintiesof unfiltered photometry. Applying the transformation equations described by Cutri et. al to relate the 2MASS JHK colors to the "homogenized system" of Bessel and Brett (2) provides the transformation to obtain V magnitudes from J, H and K. This allows selection of comparisons from the 2MASS catalog stars that are well matched to the range of colors of asteroids that are likely to dominate our sample. The 2MASS catalog was vetted to select only those objects for which the corrected photometric uncertainty in all bands was ~0.05 magnitudes. This yields a catalog for which most survey images contain between 10 to 100 suitable comparisons. Multiple comparison stars are then selected along with their J, H, & K magnitudes from the custom subset 2MASS catalog. The stars are located in the calibrated image and their instrumental magnitude determined. Transforming the magnitudes of the comparisons to V, the technique of Henden (3) can be applied, measuring the difference between the instrumental magnitudes and an ensemble catalog magnitude to obtain a mean zero-point magnitude for the frame, valid only for (most) of the asteroids and G-type dwarfs. This procedure has resulted ina zero-point uncertainty in V of ~0.06-0.08 magnitudes from field-to-field on photometric nights. At magnitudes V < 13 raw Catalina photometry exhibits a significant amount scatter due to source saturation and non-linearity of the response. There is an internal correction for this. However, the photometry for sources brighter than V=13 should be viewed with caution. 1 - Cutri et al. Explanatory Supplement to the 2MASS All Sky Data Release. 2 - M.S. Bessel and J.M. Brett, JHKLM Photometry: Standard Systems, Passbands, and Intrinsic Colors, PASP, 1988, 100, 1134, paper. 3 - A. Henden, The M67 Unfiltered Photometry Experiment, 2000, JAAVSO, 29, 35 abstract. |
The photometric uncertainties given have been determined via an emperical relationship between source flux and the observed photometric scatter. This relation was derived based on 100,000 isotropically selected sources that exhibited no siginificant sign of variabilty based on their Welch-Stetson variabilty index. |
The photometric catalog only contains the detections made at a given location. Faint objects are detected less often than bright ones because of differing seeing conditions. If a limit is needed, it is possible to very roughly estimate on how faint an isolated point source was in a given image by using the other isolated point sources that were detected within that image. There are plans to put limits on the non-detection of known sources. However, this will not occur until at least the next data release. |
In order to include all photometric detections for a given object,
it is necesary to use an object matching radius that accounts for
any astrometric uncertainties in the reduced images and poor seeing
conditions where position uncertainties are particularly large.
In good seeing the software used to measure sources (SExtractor) can often detected a close object that was not detected in bad seeing, or in the reference catalog. This often occurs when a faint source is near a much brighter source, and when there are two sources of similar brightness being detected as a single extended source. If more than one object is matched to a single reference object in our catalog, that object is marked as a blend.\ In some cases a second match could be due to an artifact or variable source. However, if many such double detections occur, the photometry the object is very likely a blend of two close sources.
One can attempt to separate the sources by using the coordinates and fluxes in cases when there are two detections. However, unless one object is much fainter than the other, such photometry is very unlikely to be accurate enough for scientific analysis. In cases where objects are within a 2-3 arcseconds, they will always be blended in CSS images. Such cases can often be seen by comparison with higher resolution images from SDSS, or other surveys.
|
As noted above, the accuracy of the photometry is limited
by the tranformation to V from unfiltered values.
As a result the accuracy of the Catalina photometry
in highly dependent on source colour.
For the Cousins filters system more accurate V magnitudes can be found using the following calibrations based on 445 standard stars from Landolt 2007 and Landolt 2009 covered by Catalina:
1) V = V_CSS + 0.31*(B-V)^2 + 0.04 (sigma=0.059) Note: Equation 3 was previously incorrectly stated as (V-I)^2. Clearly the colour correction is small for blue objects but can be very large for red objects. Corrections have not yet been derived for other filter systems. Improved photometry can be obtained using colour information derived from overlapping calibrated sources such as SDSS, or nearby photometric standards. In future releases it will be possible to download and perform photometry on the original reduced images. |
Drake, A.J. et al. First Results from the Catalina Real-time Transient Survey 2009, ApJ, 696, 870 |
The CSS survey is funded by the National Aeronautics and Space Administration under Grant No. NNG05GF22G issued through the Science Mission Directorate Near-Earth Objects Observations Program. The CRTS survey is supported by the U.S.~National Science Foundation under grants AST-0909182. |