Mysterious comets and igniting planets? No, they're mostly just ghosts...
Regular viewers of SOHO and STEREO data are well familiar with the variety of strange artifacts we see in the satellites images sometimes. We see various strange blobs, reflections and streaks, and we frequently get emails about them (which is something we strongly encourage: you learn by asking questions, so ask away!). Of course, all of these things we see in the data are completely explainable when armed with the appropriate knowledge of CCD detectors (like in digital cameras) and instrument optics (telescopes, lenses, etc). So after over 13 years of SOHO/LASCO images, we have seen and explained every weird artifact that has appeared in the data, and occasionally responded to a few popular myths. More recently (October 2006), we launched the STEREO/SECCHI mission and began send back data from that too. As expected, the STEREO/SECCHI 'COR2' telescopes see exactly the same blobs and streaks (dust, cosmic rays, etc) that we see in LASCO. But the Heliospheric Imagers (HI) are a new kind of telescope and with that comes a new set of strange image effects. So here we will address the two most commonly questioned artifacts that we see in the HI images and explain what they are, and why we see them.
Did that planet just explode?
A few hours before this image was taken, Venus was typical looking planet in the HI-1 field of view: it was bright, as expected, and had the vertical saturation spikes like those seen in Mercury in the image (the other bright spot). But then as it reached the edge of the images, it got huge! And prior to that, it seemed to throw a mass ejection (a "VME" perhaps??) of its own out at the Sun. What happened?
Check out the movie here (~8Mb .mov) or here (~5Mb mpeg4). Again, the other bright dot is Mercury.
On a certain popular Internet video sharing website there are a couple of videos of this particular time sequence and a number of... interesting... interpretations of what's happening, the primary conclusion being that Jupiter had just been ignited by a passing CME! It's an intriguing prospect but a little wide of the mark for a couple of reasons:
1. It's Venus, not Jupiter. This image shows the inner solar-system planets on January 31, 2009 (the particular date in question), and Venus is clearly in the field-of-view of the STEREO-B spacecraft, just to the 'right' of Mercury as seen from the instruments.
2. Coronal Mass Ejections (CMEs) cannot, and do not, "ignite" things. They are not made of fire or any other flammable substance! They are an extremely sparse and diffuse cloud of plasma with an embedded magnetic field. The only reason we can see them is because we have these extremely sensitive cameras that can detect sunlight scattering off of electrons in the CME.
3. Planets do not ignite under ordinary circumstances (and these were very ordinary circumstances!). Eventually the Sun will expand and engulf the inner planets of the solar system, but that is a few billion years away yet, and an entirely different situation.
So what is this? Well, this is an artifact of optics and reflections. The 'Venus Mass Ejection' is what we call a ghost, and is the result of bright light reflecting off part of the instrument and entering the optical system of the telescope. (Another example is marked in this neat 1.8Mb .avi movie sent to us by SOHO/STEREO comet hunter Alan Watson.) The 'exploding/igniting' is slightly different, and is caused by light diffracting at the very edge of the instrument optical system. We will give a little more detail about these in a moment, but first will address the other strange feature that we are often asked about.
Where did that bright comet come from?
At first glance, you would say that the image opposite shows a comet. It certainly looks like one. Check out the movies here (~5Mb .mov) or here (~3Mb mpeg4). Once again, the other bright dot is Mercury.There's a crucial clue that proves it isn't a comet: in this image, the Sun is on the left. A comet's tail always points away from the Sun, because it is formed by the solar wind flowing out from the Sun, so clearly this can not be a comet. So what's happening here is essentially the same thing as described above, namely that a very bright object is at a certain point outside of the field of view of the camera, and is shining brightly on just the correct spot on the instrument such that its light is reflected in the detector and lens barrel cavities, creating the artifact you see.
In this particular example, this artifact was seen immediately after Venus left the field of view, so it's quite obvious that the two are related. But the same thing can happen if a bright object is about to enter the images, giving the appearance of an abruptly appearing and disappearing bright "comet".
Slightly more technical explanation
So why did Venus appear to explode? What of the apparent ejection from Venus? As noted, they are a result what we call "stray light" and "ghosting", and are basically due to a very bright object just creeping out of direct view of our cameras and the light from that object reflecting off of various parts of the instrument. 
To explain it better, it's helpful to look at the design of the HI instrument. The Heliospheric Imager instrument has a series of lenses designed to give us the optimum performance and sensitivity while minimizing the effect of extremely bright objects in the field of view. Remember: we're trying to detect the incredibly faint signal of solar outflow and corona (atmosphere), so the telescope has to be really, really sensitive to faint signals. In addition, this camera uses a very long exposure times -- approximately 40-minutes per image, in fact! Planets like Venus are incredibly bright, especially when you're up in space with no atmosphere to diffuse their light, and so when they shine into our camera, there is only so much we can do to stop 'bad' reflections.
As you can probably image, before launching a mission like STEREO it is vital that we design our instruments so that the stray light does not blind us. To help do this, we do 'ray tracing' simulations on computer models of our telescopes to simulate how beams of light will react as they enter the instrument. If you click on the image to the right, it will show an expanded diagram that (very roughly) illustrates how a bright object is causing the 'ghosts' described above. At the top of the image is a 'ray tracing simulation' that shows bright light from some object being focused onto a point away from the CCD detector. This bright light then reflects back into the lens system and ultimately reaches the CCD as a very faint signal.
The example given in the image is more specifically for the ring that emanated from Venus (the "VME"!). In the case of the 'mysterious comet', the explanation is partly as above but is more related to diffraction, which in this instance is the 'bending' of light waves around the edge of the telescope and instrument. Refraction is very common in optical systems but the effect is amplified in the HI instrument because of the very sensitive optics and the fact that our exposure times are so long. Any faint refraction (or reflection) that reaches the camera is going to accumulate in brightness for the entire duration of the exposure (~40-mins). So it's is actually an incredible feat of design and engineering that our telescopes have so very few artifacts in them!
So as cool as it may seem for comets to suddenly appear, and planets to suddenly ignite, the true explanation is a little less exciting but hopefully, at least, a little interesting...
(Many thanks to Jean-Philippe Halain (Centre Spacial de Liege, Belgium) for the ray-tracing diagram and technical assistance with this article!)