# The typical reduction procedure

If you read a paper in the technical astronomy journals, you'll often see the phrase "we applied the standard reduction procedure to our images." What does that mean? It means:

1. subtract a (master) dark frame
2. divide by a (master) flatfield frame

Here's the overview of the work required to convert a night's worth of raw CCD images into "clean" images which you can then analyze.

• go to your home directory
• create a new sub-directory for this exercise:
```

mkdir typical

```

and then go to the sub-directory

```

cd typical

```
• copy over all the raw files you'll need for today's exercise
```

cp \$dd/typical/* .

```
• look at several of the raw target images. Get a feeling for the quality of the pictures.
• create a median master dark from the 4-second frames; you'll use this on the 4-second long flatfield images
• create a median master dark from the long dark frames; you'll use this on the target images
• subtract the 4-second dark from all the flatfield images
• create a master flatfield frame
• subtract the long dark from the target image: v585.fit
• look at the dark-subtracted target image. Verify that the hot pixels are gone, and that the images look OK. Write down the typical background pixel level.
• prepare to make the flatfield correction. You need to calculate the mean level in the master flatfield frame, so that we can normalize the division in a moment. If your master flatfield is called master_flat.fit, then run the mn command once, like so:
```
mn master_flat.fit
```
The XVista program mn not only calculates the mean level, and prints that value to the screen; it also stores the mean value (and the stdev) into a secret place for future reference. You can peer into the secret place by typing the command xlet.

• correct the dark-subtracted target frame for the flatfield effect like so:
```
div target.fit  master_flat.fit  flat
```
where you replace "target.fit" by the name of the target image, and "master_flat.fit" by the name of your master flat image. The keyword "flat" must appear after the names of the two images. It signals the div program to perform the division, pixel-by-pixel, but to multiply the result by the mean of the flatfield frame before storing the result back into the "target.fit" image.

• look at your target image after you have divided by the flatfield frame. Verify that the average value is still roughly the same as it was before, and make sure that the overall pattern of brightness variations has decreased.

#### Cleaned images

The result of your work should be a "clean" image: a CCD frame from which the thermal contribution and variations in sensitivity has been removed. You should be ready to perform the next step in your analysis, whether it is photometry, astrometry, or just plain looking.

Before you can move on to measuring the light of the stars in your image, you need to figure out some properties of the data. For example, consider the following "clean" image of the region around the star "V585 Lyr" (click on the image below for a larger version of the picture):

You can download a copy of this clean image by executing the command

```
cp \$dd/v585_clean.fit .
```

1. make a finding chart of the field using the Aladin tool from SIMBAD
2. find the orientation of the images. Which way is North? East?
3. what is the pixel scale? That is, how many arcseconds does each pixel represent?
4. which star is V585 Lyr?
5. what are the catalog magnitudes -- in the standard R passband -- of the stars at (108, 146) and (165, 331)? Which catalog is the source for these values?
6. just what sort of star is V585 Lyr, anyway?

One way to look for information on a particular star is to use SIMBAD's list of stellar properties and references. Another way is to go to the Astrophysics Data Service Abstract Search site. Choose either the "Object name/position" or "Abstract Words/Keywords" box, and type the name of your object. Then click on the "Send Query" button.