In the picture sequence which was taken by New Horizons during the approach to Ultima Thule on the 3rd of Dec there can be seen a speck of light in the lower left corner, not as bright as 2014 MU69 but clearly visible.

Animation of LORRI images taken during UT approach
Source: Here (German) or here (English), NASA / Johns Hopkins Applied Physics Laboratory / Southwest Research Laboratory / Henry Throop

Is there any theory what may have caused this signal? A particle from the probe itself, coincidentally passing in front of the lens? A variable star? A small object in the right distance and angle to be glimpsed upon by New Horizons? I think the latter would be an astronomically improbable event...

Here are some single frames of that image sequence, the speck is clearly visible on the 3rd of December.

enter image description here

  • $\begingroup$ Almost unrelated to your Q. What are the pictures depicting? What is left and what is right? Thanks $\endgroup$
    – Alchimista
    Commented Jan 7, 2019 at 20:17
  • $\begingroup$ @Alchimista the sequence was taken by New Horizons approaching 2014 Mu69 Ultima Thule. Speaking as a layperson, I think the left is the original brightness data and the right is a difference image attenuating all static brightness signals and leaving only the signals which increase/decrease in brightness. $\endgroup$ Commented Jan 8, 2019 at 9:29
  • $\begingroup$ @uhoh The article I took it from is at heise.de/newsticker/meldung/… (german) $\endgroup$ Commented Jan 8, 2019 at 9:30

2 Answers 2


It's probably a cosmic ray strike, a common artifact on New Horizons images. The image, however, is of very poor quality which makes it difficult to analyse the exact nature of the spot. I'll go into this more at the end of the answer.


  1. Cosmic ray hit
    A common occurance (see below); the most likely cause without evidence to the contrary.

  2. Background star (e.g. a nova)
    Novæ (and variable stars) brighten and dim over several days. Were this a nova it would have shown in images from previous and later days.

  3. Kuiper belt object
    To not show in images from adjacent days it would need large relative motion and thus a sizable trail would be expected (none is evident).

  4. Aliens
    Possible? Yes. Likely? Far, far, from it. Not worth serious consideration.

Here's the New Horizons team's standard disclaimer:

This image contains one or more objects whose brightness exceeds the detector's saturation level. This sometimes produces a "tail" of bright and/or dark pixels to the right of the object. You may also notice a faint vertical white stripe passing through the saturated object; this is an artifact called "frame transfer smear" and is associated with the incomplete removal of signal produced when the image is transferred from the optically active region of the detector to the storage region of the detector. If the target is badly saturated, you may also notice a faint, X-shaped feature nearly centered on the object; these are optical diffraction spikes.

This image contains one or more streaks associated with cosmic rays passing through the detector. Nearly every LORRI image has at least one cosmic ray strike, but most are "single pixel" events (i.e., they only appear to be in single pixel and can easily be mistaken for stars). But sometimes a cosmic ray is energetic enough that it leaves a "trail" as it passes through the LORRI detector.

Example of cosmic ray hit and saturation trails:
Example of cosmic ray hit and saturation trails
Source: here (via "Learn more..." link), NASA/JHUAPL/SwRI
Note that despite being a PNG, JPEG compression artifacts can be seen in this image

The frame in question:
Extract from GIF of frame in question
Source: Frame K-29 of this image taken from here, NASA/JHUAPL/SwRI/Henry Throop
Enlargement below is from frame K-29 of this image

The image in question is almost certainly a stacked composite of ~10 source images and appears to have been 2×2 binned (compare here). To make faint objects more visible, the image was also likely contrast-enhanced. The image also shows clear signs of lossy compression and quantization (from 12 bits per pixel to ~5), see enlargement below. While these are (mostly) acceptable choices for the image's intended use it tends to obscure detail from the original images.

Crop showing artifacts:
Crop showing artifacts

Looking at the spot in question (enlargement below) a couple of things are noticable:

It is compact and relatively dim. Its overall size in pixels is about half that of the stars and of MU69. In addition, the drop-off in brightness moving away from its center is much more rapid than for the other objects. This is consistent with a bright but compact object appearing in only one image in the K-29 stack.

This object also exhibits a less-blurry structure with two distinctly brighter spots or lines. This is consistent with spray from a cosmic ray hit.

Crop showing mystery spot:
Crop showing mystery spot

All in all, I think a cosmic ray hit is the most likely explanation. Given the very poor quality of the image in question though it's difficult to give a definitive answer. To make a proper determination one would need access to the yet-to-be-released raw imagery.

I've seen a paper with a more thorough discussion of New Horizons image artifiacts but don't seem to have saved a copy. I'll add references from it when I track it down.

  • 1
    $\begingroup$ Stellar variability is another possibility, such as a flaring object. The Point Spread Function doesn't look very cosmic ray-like. $\endgroup$ Commented Jan 7, 2019 at 22:59
  • $\begingroup$ @astrosnapper Do you think a flare, nova, etc. would be brief enough to not show on earlier and later images? I agree though that the pattern isn't a perfect match for cosmic rays but I do think it's currently the best fit. When the original, unprocessed images become available it should make analysis a lot easier. $\endgroup$ Commented Jan 8, 2019 at 4:57
  • $\begingroup$ @uhoh I'll be updating my answer shortly and I'll incorporate your suggestions ("something like "As astrosnapper suggested in the comments...", etc.). My apologies but I've had a rather hectic day. On a totally different note, I only mentioned KBOs since the camera is pointed out of the solar system, not in. What other classes of asteroid could it be? Are you refering to comets? $\endgroup$ Commented Jan 8, 2019 at 10:46
  • 1
    $\begingroup$ @uhoh I've got that info at hand. $\endgroup$ Commented Jan 8, 2019 at 11:46
  • 1
    $\begingroup$ @Vroomfondel it probably is not a cosmic ray. Let's keep an eye on this one, I'm interested to see how this story ends ;-) $\endgroup$
    – uhoh
    Commented Jan 9, 2019 at 13:08

Looking closely, the spot has roughly same slightly noncircular, diffuse, extended shape and size as 2014 MU69 (optical or kinematic artifact) which would not happen from a cosmic ray or hot pixel.

This could be a real object, and simply much closer to New Horizons than MU69 is. That could certainly account for the object being visible on one day only. If its 10x closer than MU69 it can be 100x smaller than MU69 and still have a similar apparent magnitude, and the number of asteroids increases rapidly with decreasing size.

Left is object in question, right is MU69.

enter image description here enter image description here

enter image description here

above: purely unscientific processing in order to perform "blobology": OP's image, blown up and then cropped twice (to get rid of the red arrow) then just plotted in python https://pastebin.com/8KWBHqnf

  • $\begingroup$ What's the source for that image? If self-made, how was it processed? $\endgroup$ Commented Jan 8, 2019 at 8:01
  • $\begingroup$ @AlexHajnal added the info at the bottom. $\endgroup$
    – uhoh
    Commented Jan 8, 2019 at 8:48
  • $\begingroup$ Unscientific processing is a good way to make an image show what you want it to. For example, I'm pretty sure that, through judicious use of contrast enhancement, sharpening, and edge detection, I could demonstrate that the speck is a Borg cube. $\endgroup$
    – Mark
    Commented Jan 13, 2019 at 8:04
  • $\begingroup$ @Mark go for it! btw there is no processing, that word came from another user and you've repeated it. These are the same shape in the original image, I've made it 1) large to make it easier to see for people with bad eyes and 2) also included an old fashioned jet color version. I haven't done any specific processing and included my raw images for anyone to look at. $\endgroup$
    – uhoh
    Commented Jan 13, 2019 at 8:26
  • $\begingroup$ @Mark Being straightforward and modest beats writing long, authoritative-sounding answers in the real world, but this is SE, so... Let's wait and see what the answer turns out to be! $\endgroup$
    – uhoh
    Commented Jan 13, 2019 at 8:29

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .