Basics of Imaging Data

Imaging runs

Schematic of the arrangement of the CCDs and filters on the SDSS camera.

The SDSS camera worked in drift scan mode, opening its shutter for extended periods and imaging a continuous strip of the sky. In the coordinate system of the illustration shown at right of the SDSS camera focal plane, the sky drifts downwards. Each such drift scan is referred to as a run and there is an associated integer identifying the run. For science quality runs, the lowest run number is 94, and the highest is 8162.

Because of the layout of the CCDs in the focal plane, such an imaging run results in six continuous images associated with the six columns of CCDs. Each of these six are known as camcols and are numbered from 1 to 6. Each camcol is 2048 pixels wide (the width of the CCDs). The gaps between the camcols in most cases are filled by another, somewhat overlapping, drift scan run.

Each camcol is artificially broken up into a series of overlapping fields, each 1489 pixels long, and 2048 pixels wide. They each overlap by 128 pixels with the adjacent fields, meaning there are 1361 unique rows in each. These fields are the basic unit which the photometric pipeline processes. They are designed to be overlapping even within a run so that objects are not misdetected due to being too close to the edge of a field.

Imaging run 756.

Finally, there have been multiple reprocessings of the data over the years. Each reprocessing has been denoted by an integer (the first being rerun 0, the latest being rerun 301). Each rerun consists only in a change to the photometric pipeline, not to the data itself.

The overall result is shown at right for a small section of one run in the SDSS. There are six continuous camcols, broken up into fields. Each image is uniquely identified by its run, camcol, field and rerun number. You can explore the JPG images of a run to get a sense of the geometry.

Bandpasses

In the top figure above, note that there are five rows of CCDs, labeled u, g, r, i and z. As the description of the SDSS camera notes, this results in images in each of these bandpasses nearly simultaneously (in detail, separated by approximately 71.72 seconds). As shown in the illustration above, the observations are in the time sequence r, i, u, z and then g.

The multiple bands allow a determination of the colors of the detected objects. For example, in the JPGs shown above are a composite of the g, r, and i images. In the flat-file versions of the catalogs on SAS, the bandpasses are given in order of increasing wavelength. In the CAS, parameters that are specific to a given bandpasses are named accordingly (e.g., "petroflux_u", "petroflux_g", etc.).

Deblending and resolve basics

Illustration of "parent" and "child" objects.

In the final catalog, there are a number of important pieces of nomenclature to note regarding deblending, resolve, and quality flags. The glossary gives a full description of SDSS nomenclature, and the algorithms section gives a detailed account of the methods. Here we simply try to highlight some of the most relevant points.

The most important regards determination of unique objects in the catalog. The full catalog contains two classes of objects that are to be ignored entirely: "bright" objects, which are detected in a first-pass object detection step; and "parent" objects, which are deblended into "child" objects. "Bright" objects should be ignored entirely, and in almost all cases so should parents, since they are usually successfully separated into their component children.

For example, in the image on the right, the island of high-resolution pixels indicates the region associated with a "parent" object. The yellow diamond is the "center" of the parent. The red diamonds are the centers of the component children. The child objects are the ones to use for any analysis of the catalog. Additionally, there is overlap between each field within a run, and also between fields in different runs. Thus, any source on the sky can in principle appear as a child detection in a number of different runs. The resolve procedure determines which entry is "primary." The resolve documentation describes how to select primary children.

Finally, not all objects are "good" even when they are unique. These objects can be identified using the objc_flags bitmask. Some are associated with saturated pixels, generally bright stars (SATUR), or are close to the edge of a frame (EDGE), or are possible misclassified cosmic rays (MAYBE_CR).