The Heliospheric Imager for the NASA STEREO Mission
Part of the SECCHI Consortium
Sky at Night Competition
In the April 2008 edition of the BBC's Sky at Night (which
featured the STEREO Heliospheric Imager), Sir Patrick Moore
announced a prize for the observation of the first Vulcanoid
asteroid (asteroids orbiting the Sun within the orbit of Mercury)
in the STEREO HI data. The Heliospheric Imagers are capable of
imaging stars down to around 13th magnitude and so it is indeed a
challenge to observe such objects in our data. STEREO has
discovered at least three comets and several known asteroids have
already been identified in the data so far.
As the two spacecraft move away from the Earth, along Earth-like orbits, the Heliospheric Imagers look back at the space between the Sun and the Earth. This design enables scientists to track Earth-directed coronal mass ejections (CMEs) as they travel into space from the Sun's surface. This unique viewpoint also gives us an excellent opportunity to look for Near-Earth Objects (NEOs) that would otherwise be difficult to observe from the Earth (because they would lie very close to the Sun as observed from the surface of our planet). There are two STEREO spacecraft (Ahead and Behind or A and B), each carrying a Heliospheric Imager. Each imager contains two cameras; HI1 and HI2. The HI1 cameras view a 20 degree region of the sky with the sunward edge of the field of view being approximately 4 degrees from the Sun. The HI2 cameras have a 70 degree field of view starting approximately 18 degrees from the Sun.
Traditionally, the detection of planetary objects has been carried out by 'blinking' pairs of images and looking for anthing that moves between them. STEREO team member Dennis Wang has written a great web application to demonstrate this effect. On his webpage at http://secchi.nrl.navy.mil/index.php?p=js_blink, you can compare the first and last images taken each day with the STEREO cameras. January 1st 2008 contains a great example of a planet moving across the background star-field. If you can see the same object in images from both STEREO A and B spacecraft, it means that the object must be between the Sun and the Earth. When trying to identify a planet or asteroid, don't forget that the STEREO spacecraft are moving away from the Earth and so the position of a planet with respect to the starfield observed from STEREO will be different for an Earth-based observer.
If you want to look at a region of any of the images in more detail, you will need to download the files by going to the 'data' tab in the menu on the left of this webpage. Our data are made available via the UK Solar System Data Centre with whom you will need to register first (which is free). The data are in 'fits' files which can be read into many of the standard packages designed to manipulate astronomical images (such as registax). While data from all the STEREO telescopes is available on this website, the HI cameras provide the widest field of view in which to hunt for asteroids. The website provides two versions of the data, L0 and L1. These are NASA-speak for the amount of data processing that has been carried out on the data. L0 data is raw data, more-or-less as it comes from the spacecraft. L1 data has been processed to apply calibration corrections to the cameras and (for HI) correct for the fact that the cameras do not have shutters (it's a long story).
The signal in both L0 and L1 HI images are dominated by the light scattered from the F-corona (or zodiacal light). A CME is less than 1% of the brightness of the F-corona and so it has to be removed before these moving features are revealed. Luckily, the F-corona is very stable and so it can be estimated (from the minimum value in each pixel over a series of images) and removed. Once this is done, the starfield becomes much easier to see (as in Dennis Wang's blink images). Another way of removing the F-corona is to subtract one image from the other (so-called differenced images). These tend to remove anything that doesn't move relative to the cameras (like the F-corona) and leave behind anything that is moving (like CMEs, stars and asteroids). Both background-subtracted and differenced movies are available from the 'movies' tag in the menu to the left of this page. We hope, soon, to supply L2 data via the UKSSDC which will have our best estimate of the background F-corona removed automatically.
Once you have detected an object moving through the data, you may want to look at one of the monthly movies (also in the menu to the left) to see if you can track your object throughout a much longer period.