Making Darks and Flats for Mar 30, 2002, data

Your job today is to create master dark frames and master flatfield frames for the images we collected March 30, 2002. You will use these master frames to reduce the images of 1107 Lictoria.

1-second darks

First, you must combine the 1-second dark frames to create a master dark suitable for subtracting from the 1-second flatfield frames. The raw images are called dark1.001, dark1.002, etc.

copy all these dark frames to your directory.
in order to give yourself "write privileges" on the files (so that you can modify or delete them), type
```
       chmod 644 dark1.*
```
display the first, dark1.001, via the tv command
define a box we will use for statistical calculations:
```
       box 1 sr=100 sc=100 nr=50 nc=50
```
show the box in the image by typing
```
       box 1 show
```
calculate some statistics for pixels in the box with the abx command
```
       abx dark1.001 box=1
```
write down the mean and stdev of pixel values
Now, combine all the 1-second dark frames via the median command to create a master dark frame, and compare it to one of the individual dark frames.

create the median image

       median dark1.001 dark1.002 ... dark1.010 outfile=dark1.fts nomean verbose

look at closeups of the master and dark1.001 side-by-side. First, kill any tv windows. Then,
```
       tv dark1.001 box=1 zoom=5 z=150 l=100
       tv dark1.fts box=1 zoom=5 z=150 l=100
```
What do you see? Write down the differences and similarities of the individual and master dark frames.
Calculate the statistics in box 1 of the master dark frame and compare them to those in dark1.001.

Okay, you have a master dark frame for the 1-second exposures. Make sure it is called dark1.fts.

15-second darks

Now, repeat the procedure to create a master dark frame suitable for the 15-second target exposures we took later in the night. You need to combine the dark frames called dark15.001, dark15.002, and so on, to create a master dark frame called dark15.fts. As before, calculate the mean and stdev of pixel values in box 1 (the same box) of one individual 15-second dark frame, and the master 15-second dark frame. Compare the individual and master darks like so:

       tv dark15.001 box=1 zoom=5 z=150 l=200
       tv dark15.fts box=1 zoom=5 z=150 l=200

Describe in your own words the difference between these two frames.

How big is the dark current?

Let's try to calculate how many "counts" are generated by the dark current each second. Find an area in the 15-second master dark frame which is free from hot pixels, and exactly 10 x 10 pixels in size. Use the cursor to pick the coordinates of the upper left-hand corner -- say you choose row=116, col=121. Then

define box number 2 at your chosen coordinates; with my numbers, I would type
```
       box 2 sr=116 sc=121 nr=10 nc=10
```
calculate the mean and stdev of pixel values in the box, for both the 1-second master dark frame, and the 15-second master dark frame.
find the difference between the mean values
divide the difference in counts by the difference in time to produce the rate at which dark current appears (units should be "counts per second")
Now, find a hot pixel somewhere in the master dark frames
measure its value in the 1-second master dark, and the 15-second master dark
find the difference in the hot pixel's value
divide the difference by the time to find the rate at which dark current appears in this hot pixel (once again, units are "counts per second")
By what multiplicative factor (i.e. 3 times, 5 times) is the dark current stronger in the hot pixel?

Making the master flatfield frame

We have a series of 1-second, V-band images of the twilight sky, called flatv.001, flatv.002, etc. You must subtract the dark current from each frame, and the combine them via the median command to create a master flatfield frame.

copy all the raw flatfield frames to your subdirectory
run the chmod command on them to give yourself write access to them
display each of the individual flatfield frames via tv. Note any significant differences between the frames, and also write down a rough "typical" pixel value in each. Is there any systematic trend in the pixel values from frame to frame? If so, explain. What is the feature in frame flatv.010, near row=114, col=341?
if any frame(s) appear "bad" in some way, it's best to discard them. To delete a file, type
```
       /bin/rm filename
```
for each of the remaining raw flatfield frames, subtract the master 1-second dark frame
```
       sub flatv.001 dark1.fts
       sub flatv.002 dark1.fts
           etc.
```
display flatv.001 again. Does it look any different to the eye?
use the median command to create a master V-band flatfield for the night.
```
       median flatv.001 flatv.002 ... outfile=flatv.fts verbose
```
Note that the nomean option does NOT appear this time. The median command will print out some stuff to the screen. Explain what it is doing, and why.
display side-by-side an individual flatfield frame, flatv.010, and the master flatfield frame, flatv.fts. Describe any differences which are apparent to the eye.
Calculate the mean and stdev of pixel values in box 1 in flatv.010 and flatv.fts. Are they significantly different? If so, why?