CCD Photometry Applet
The Applet
Photometry is the measurement of the apparent brightness of objects in the sky. We will use a JAVA applet developed by Greg Bothun at the University of Oregon to simulate CCD observations of a field of stars. These simulations will be used to study the accuracy with which we can measure stellar brightnesses under varying observing conditions.
- From cosmos:
- The first time that you try to use the applet in cosmos you will have tell Netscape how to handle it.
- From the Edit menu of Netscape, choose Preferences.
- When a new window opens, highlight Helper Applications under the Navigator category on the left.
- Press the New Type button and enter the following under the appropriate fields in the new window:
- Description of type: application/javaws
- File extension: jnlp
- MIME type: application/javaws
- Application to use: /usr/local/java5/jre/javaws/javaws
- Press OK when finished.
- Note that if you would like to configure Netscape to automatically open PDF files (like the ones on the class webpage!), you can do so by pressing the New Type button again and filling in the fields as follows:
- Description of type: application/pdf
- File extension: pdf
- MIME type: application/pdf
- Application to use: /usr/local/bin/acroread
- Press OK when finished until all of the popup windows have been closed.
- From home:
- You may need to install the JAVA JRE plug-in. If you cannot get the applet to load, then follow this link and download and install "J2SE v1.4.2_07 JRE" for your home OS. Then try running the applet again.
Your browser should now be able to open the applet. Some important features of the CCD photometry applet are (click on the link to open):
- An exposure time slider, which allows you to choose an exposure time between 1 and 100.
- A Measure button. After you have chosen an exposure time, press this button to "expose" the CCD. Each time you adjust the exposure time you have to press Measure to re-run the simulator.
- Red and green readout boxes. Select an exposure time and press Measure. Now run the cursor over the image. Notice that the green readout boxes now show the number of counts at the current cursor position.
- Sample Size text box. Click and drag on a region of the image. Red and green boxes should appear near the cursor. You can resize these boxes with the mouse or with the Sample Size text box (e.g., to make a 20x20 pixel box, type "20" into the Sample Size text box and press enter).
- Red and green boxes in the image. Once created, the red and green boxes are used to measure the counts in a given region of the image. You can move these boxes by clicking INSIDE a box and dragging it to a new location. The red and green readout boxes display the mean, standard deviation, and total number of counts in the red and green boxes, respectively. For these exercises we will be ignoring the standard deviation boxes. The standard deviation that we will use is the square root of the number of counts.
CCD Photometry
In the following exercises you will measure the brightness of stars in a CCD image and its associated error to determine the accuracy with which you have made your measurement.
Exercise 1:
- If you haven't opened the applet yet, do so now.
- Set the exposure time to 50 and press Measure.
- Draw a 20x20 box on the image. Place both boxes in a background region devoid of stars.
- Record the Mean in each of the boxes and calculate its associated error. Remember, the standard deviation is the square root of the Mean, NOT the value listed in the Standard Deviation readout boxes. However, in this case the error is not equal to the standard deviation. When you calculate the average value of an ensemble of measurements, the error in the mean is given by error = standard deviation/sqrt{N} (Bevington, Eq. 4-14), where N is the number of points being averaged. Since we are using a 20x20 pixel box, 400 pixels are being averaged together to calculate the mean. Therefore, the error in the mean value in each box is error = sqrt{Mean}/20.
- Now move each of the boxes to a new background region. Record the Mean in each of the boxes and its associated error.
- Repeat until you have 10 data points. Calculate the mean background level in the image. Do the error bars from each of your individual measurements overlap with this mean background level? Is this the result that you expected?
Exercise 2:
- Vary the exposure time until one of the dim stars (Star 6 or 8) is barely detected. Just use your eye for this and find an exposure time where you are just starting to see the star above the background. What exposure time did you choose?
- Change the box size to 10x10 pixels. Place one of the boxes over the barely-detected star and one in a background region. Calculate the total number of counts from the star using the formula star counts = TotalValue(starbox) - TotalValue(skybox).
- Calculate the error in this value using the formula noise = sqrt{TotalValue(starbox) + TotalValue(skybox)}. We will formally derive an analogous error formula in lecture next week.
- Calculate the signal-to-noise ratio and fractional error in your measurement of the star's brightness.
- Double the exposure time and press Measure. Calculate the total number of counts from the star. What is the signal-to-noise ratio and fractional error of your measurement in this image?
- How do the total star counts and signal-to-noise ratio in the second image compare to the signal-to-noise ratio in the first image (i.e., how do they scale with time)? What exposure time would you need to reach a signal-to-noise ratio that is twice what you measured in the first image?
Exercise 3:
- Now set the exposure time to 50 and the box size to 5x5 pixels. Place one of the boxes over Star 1.
- Record the box size, total number of counts from the star, and the signal-to-noise ratio of your measurement.
- Change the box size to 10x10 pixels. Record the new box size, total number of counts from the star, and signal-to-noise.
- Repeat with box sizes of 15x15 and 20x20 pixels. Which box size allowed for the highest signal-to-noise measurement? Why? Which box size would you choose if you wanted to measure the brightness of Star 1? Is this different than the box size that facilitated the highest signal-to-noise measurement? If so, why?
Varying Observing Conditions
The final exercises will study the effects of varying observing conditions on the accuracy of your photometry. In each of the five cases below the detector is "imaging" the same field of stars, so in each case there are 8 stars on the detector. However, the background level and seeing vary from case to case, so the detector will not always detect all 8 stars.
Exercise 4:
- Load each of the five applets below individually. In each case, use an exposure time of 20 and note the background level, relative seeing (i.e., how fuzzy are the stars?), how many stars you can see in the image, the faintest star visible to you, and the signal-to-noise ratio of Star 2.
- Which images had the highest and lowest background levels? What are some causes of background noise in images? Which image had the worst seeing? Which observing condition makes it harder to detect faint stars, high background or poor seeing? Which has a greater impact on the signal-to-noise of your measurement?
- In each of the cases above, what exposure time is necessary to reach a signal-to-noise ratio of 100 in Star 2? This is the signal-to-noise ratio required to measure the brightness of the star to within 0.01 magnitudes.
