How to find an exoplanet

CCD detectors are useful for more than capturing detailed deep-sky images: you can also use them to detect planets outside the Solar System.

Written by Robin Leadbeater

Skill level: advanced.

A
a
-
Image Credit: 
Robin Leadbeater

In last month’s How to, Gary Poyner explained how to visually track changes in the magnitude of variable stars. But this method can also be used for other purposes. For faint targets, such as supernovae, where the variations are too small to be detected with the naked eye, CCD imaging techniques are a must. This is also the case for exoplanet transits. Exoplanets are planets that orbit stars outside our Solar System. Some pass in front of their parent stars as seen from Earth, causing a dip in magnitude.

Traditionally, this has been the domain of well-equipped, dedicated and experienced amateurs. But with the increasing popularity of inexpensive electronic imaging equipment, there is an opportunity for anyone to have a try.


What you need to get started

Astro imaging equipment

Just about any imaging setup can be used to get started in photometry. A wide field is useful to increase the availability of comparison stars. Good tracking helps reduce variability, and if you are looking for high accuracy, you’ll need a monochrome CCD astro camera with good linearity and dynamic range.

Photometric filters

UBVRI photometric filters are specifically designed for the purpose of photometry. They are quite expensive, but you can get started without them: they are only useful if you have a monochrome camera. If you get serious about photometry and wish to produce accurate data that you can compare with other observers, the visual filter is the one to buy first.

Charts

You’ll need these to identify your target and the magnitudes of the comparison stars you’re going to use. The AAVSO website at www.aavso.org is a good source for charts.

Software

This measures your image and calculates the magnitude of the target relative to the comparison stars. Several image processing software packages have photometry functions. These include AIP4Win, MaximDL, AstroArt and the freeware programs TeleAuto and IRIS.


Measuring the brightness of a star using a CCD camera is deceptively simple. In essence the pixel values forming the star image are summed. However, this does not give a very consistent result, as changes in atmospheric absorption and drifts in camera electronics will always conspire against accuracy. Fortunately, much of this unwanted variability can be eliminated by differential photometry, a process of comparing the star being studied with another of known and constant brightness in the same image.

Choosing a good comparison star is very important. It needs to be close enough to the target to fit in the field of view, and nearby stars should not interfere with it. What is more, the star should not be variable itself, and ideally should be similar in brightness and colour to the target star. Finding such a star can usually be left to the experts: charts showing good comparison stars are available for most objects of interest.

When making measurements, it is important to keep in mind the distinction between accuracy and precision. An accurate measurement will be in agreement with the value obtained by other observers. A precise measurement, on the other hand, is repeatable throughout an observation using the same equipment.


Aspire to accuracy

You can get decent precision with most imaging setups. High accuracy, however, is much harder to achieve; it requires a correctly filtered camera with a linear response, and a thorough knowledge of the many factors that affect accuracy. As a beginner, this is something to aspire to; if you are already imaging, you will most likely have the hardware and skills you need to make a start in photometry.

The imaging technique itself is similar to that used to produce any astronomical image, except that in photometry, the aim is scientific accuracy, not aesthetics. As a result, the only processing you should do is dark- and flat-field correction.

If the linearity of the camera is questionable, as is the case for a webcam or digital camera, try to include more than one comparison star, with magnitudes greater and lesser than the target if possible. Some digital cameras can produce linear raw images. A handy tip with the Philips Toucam webcam and its derivatives is to set the gamma to minimum, where the camera’s response is more linear.

Set the brightness so that the background is not completely black to avoid a falsely suppressed zero level. The stars need to be well exposed but not saturated for good precision. Remember that the brightness might increase during a run, particularly if the elevation of the star is increasing. It is best to limit the maximum pixel value to two-thirds of the camera’s range to minimise linearity problems. For highest precision, do not adjust the exposure during a run.

These requirements are all relatively easy to meet for dedicated astrophotography cameras with a wide dynamic range, but can be tricky with  standard cameras that may have only 256 brightness levels. Stacking images helps with brightness levels and can also reduce scintillation (twinkling) in short exposures. With colour cameras, using an IR blocking filter and only the green channel will approximate – very roughly – the response of a photometric V filter.


Software calculations

Several commercial and freeware image-processing packages have photometric functions, most of which employ the aperture photometry technique. Here, the software calculates the brightness within a circle containing the star. The background brightness, measured in an annulus surrounding the star, is subtracted from it. Measurements are made on the target and comparison stars, and the software calculates the magnitude of the target. By repeating this for images taken at different times, a plot of the star’s variability can be built up.

Professionals do not have the time or resources to keep track of all this variability, so the measurement of variable stars is one area where amateurs can make a real contribution to research. It’s worth remembering that measurements made using standard photometric filters have greater value as they allow direct comparison between different observers.

National associations such as the BAA have observing programmes for variable stars, and predicted transit times are on the Transitsearch website.

If you prefer something easier, try a short period eclipsing binary from the beginner’s CCD target list on the BAA website at www.britastro.org/vss. AD Andromedae, for example, dips by 0.7 from magnitude +10.9 for three hours twice a day, and is predicted to be at minimum at 00:30 UT on 10 December 2005, and 20 minutes earlier on each following night.


How to find the exoplanet TrES-1b

Step 1

We plan to measure a transit of exoplanet TrES-1b. The chart shows the CCD field containing the parent star and four comparison stars of similar magnitude. Using multiple comparison stars helps to reduce non-linearity on your camera. To minimise the effects of atmospheric extinction, time the observation so that the target remains above 30º elevation throughout the transit.

 

Step 2

Take a series of images covering the transit and a period before and after. Here, six-second exposures were taken covering 4.5 hours using a modified webcam. These were aligned and stacked in sets of 20 to reduce variability. TeleAuto is used to measure the magnitudes. Import each image and adjust the appearance on the screen to see the full extent of the stars.

 

Step 3

The photometry program is configured through the ‘Analyse’ menu. Select ’Open’ for aperture photometry and ‘With Linear Regression’ to minimise linearity errors if you use a webcam. Select the open aperture sizes to include the whole star in the inner circle and make sure no adjacent stars are included in the background measurement between the two circles.

 

Step 4

Add the comparison stars. For each comparison star, select ‘Add a Standard Star’. Place the cross hairs on the centre of the star and click to display the measurement circles. Enter the magnitude. For the ‘Linear Regression’ option, enter the catalogue value for one of the stars and the calculated relative values for the other comparison stars, based on measurements of a combined image of all the exposures.

 

Step 5

Next, measure the target magnitude. Select ‘Measure’. Place the cross hairs on the centre of the star and click to display the measurement circles. Check that the circles cover the correct areas. If they do not, return to step three. The calculated magnitude of the target is now displayed. Repeat the process for the other images in the series, logging the target star magnitude in each case.

 

Step 6

Finally, plot the measurements against time. The precision, or error, can be estimated by measuring another non-varying comparison star in the images. In this case the values for the 20 frame images were averaged in blocks of five to produce 10-minute averages. This improved the precision sufficiently to discern the 0.02 magnitude dip in the exoplanet’s parent star during the transit.


For more information:

Robin’s Astronomy Page

Transit Observing Cooperative:


This article first appeared in the December 2005 issue of Sky at Night Magazine
Like this article? Why not:
How to collimate your telescope
previous feature Article
How to make a sky ruler
next feature Article
We use cookies to improve your experience of our website. Cookies perform functions like recognising you each time you visit and delivering advertising messages that are relevant to you. Read more here