Featured Content: Planetary Features Illustrated Glossary

# Ice Investigators: Tutorials

The Science The Mission The Team
Tutorials FAQ Resources

Published Ice

Help us find Kuiper Belt Objects! In these images they look like small white, well-defined blobs. Not all images will have them, but every one you find has a chance of being the final target of 'New Horizons'. In addition to possible KBOs, these images also contain asteroids that appear as streaks or sets of small white dots lined up in a row.

# Quick Start Guide

Help us find Kuiper Belt Objects! In these images they look like small white, well-defined blobs. Not all images will have them, but every one you find has a chance of being the final target of 'New Horizons'. In addition to possible KBOs, these images also contain asteroids that appear as streaks or sets of small white dots lined up in a row.

### Step 1: Mark Transients

Look for a nice blob of white (not black) in the messy residuals. This could be a Kuiper Belt Object! (Or a variable star.) Trick: The easiest way to distinguish a badly subtracted star from the sought-after transient is to look for a solid center.

You may see some black transients in the residuals. Ignore them. Only mark the white transients. Mark every transient - every potential KBO - by clicking on it to place a diamond around it. The faintest transients are the most likely to be what KBOs in the right place, so make sure you look hard for the faintest transients.

### Step 2: Mark Streaks

In addition to blobs, you may see white streaks and sets of 3 white dots close together. These are most likely asteroids. To mark them, click on the button labeled "Mark Asteroid". Click on the center of each streak. You may see black streaks too - don't mark them. Only mark the white asteroids.

A quick note on asteroids: Many people ask if they can name the asteroids they find. We can only name asteroids if we get enough data to determine their orbits. This doesn't happen most of the time. If we ever do have sufficient data, you'll be notified.

### Step 3. Repeat until done

Repeat Steps 1 and 2 until you see no more solid transients! Most images will have 0 - 2 transients, but a few will have more. Look carefully, and mark everything.

If you make a mistake, click on the eraser to select it, and then click on the mark you want to get rid of. Click 'Done Working' when you're done. (And remember, it is better to err on the side of marking too much than on the side of missing potential objects!)

# Where do the images come from?

The images you will be examining were mostly taken by the 8-meter Subaru telescope on Mauna Kea or the 6.5-meter Magellan telescope in Chile.

### Acquiring Images

Image 1: Original from 2004-Jun-09 at 11:40 UT (this is 3 images combined)
Image 2: Original from 2004-Jun-09 at 13:26 UT (this is 3 images combined)

Each image that you see on this site is the result of six or more exposures by the telescope! Let's consider the images below. They are both images of the exact same field. The first image (top) is a combination of 3 images taken 2004-Jun-09 at 11:40 UT. (UT is Universal Time, a common time used around the world by astronomers and often by navigators. You may have heard it referred to as Greenwich time or Zulu Time.) The second image (bottom) is a combination of 3 images taken the same day, but a couple hours later at 2004-Jun-09 at 13:26 UT. Multiple images are taken to allow more light to be collected over a longer period of time, while making it possible to reject momentary events that might effect the image, like cosmic rays (see more below).

The two images below aren't identical in two fairly significant ways. Most importantly for this project, there is a Kuiper Belt Object in these two images that has moved during those two hours. Most annoying for you and I, the sky conditions changed during those two hours, and the second image isn't as clear as the first. In trying to find that one moving object against the background wealth of stars you are very much looking for a needle in the haystack. Can you see the KBO that moved? If you can't, don't worry! The person writing this tutorial couldn't find it immediately either!

### Making things match: Image Blurring

Image 1: Blurred version of image from 2004-Jun-09 at 11:40 UT
Image 2: Image 2: Original from 2004-Jun-09 at 13:26 UT

To make it easy to find that moving KBO, we want to find a way to somehow subtract away all the stars. Imagine if you could simple say, "Stars, Go Away!" and reveal the KBOs. Mathematically, we can actually do that if the images are the same. You simply take the two images and subtract the values of each pixel in one image from the values of the same pixels in the other image, and just like 1-1 = 0, the stars go to zero and go away. The only problem is, these images aren't the same; the sky changed and the stars are more smeared out in one image than in the other. In order to subtract the images, we have to blur out the crisper image taken earlier in the night. At right, we show the blurred out image (top) compared to the image that is being matched (bottom).

This step is the reason the images you see in Ice Investigators have a variety of different appearances. While it is easy to blur stars so they appear the same number of pixels across from one image to another, it is very hard to get the way the light is blurred out to appear the same. One image may have stars that appear brighter in their centers, while in another the stars are blobby disks. When two images with different distributions of light are subtracted, the results are very odd looking donuts. The exact shape of the donut shaped light residuals varies with these different ways of blurring out the light.

### Image Subtraction

Result of Subtracting the blurred Image 1 from Image 2

To make the pesky stars disappear the two images are subtracted from one another. This doesn't make the stars go entirely away. Even with the blurring step, the two images aren't exactly identically and instead of having 1-1 = 0, we have something more like 1.01432 - 1 is not 0. Where the first image was brighter, a bit of white shows in the image. Where the second image was brighter, there is a bit of black.

The KBO stands out in this image as both a dark blob (from being in the second image) and a white blob (from being in the first image).

In this project you are only marking white blobs.

A lot can be learned from just this one really ugly image. The separation between the white and black blobs tells us how far across the sky the KBO moved in just a couple hours. This movement corresponds to an object out around 40 Astronomical Units. This is a distance that puts it out near Pluto, which has a distance that varies from 29 AU to 49AU. (1 AU is the average distance from the Earth to the Sun, and 29AU is equal to 4.4 billion km or 2.75 billion miles.)

# Transients vs Asteroids

### Transients: Small White Blobs (aka KBOs and Variable Stars)

As you can tell in the examples above, KBOs (and variable stars) appear as solid blobs of white in an image filled with messy residuals from the subtraction process. The variable stars are still there because they are changing in brightness and the subtraction process can't tell the difference between a KBO moving into the image (like a lamp being carried into a room) and a variable star brightening (like a lamp that was there all along being turned on).

### Asteroids: White Streaks and Cigars

One of the curses faced by people making astronomy images is the occational passage of an asteroid through a field of view. These nearby rocks move much faster across the sky than our target KBOs. This means that in the three to six telescope exposures that go into each of the images that are subtracted we can actually see the passage of the asteroid across the sky. In each exposure, the asteroid makes a small smear, and there is a gap where the asteroid passed in the moments between exposures. This is essentially a stop-action moving of an orbiting object. While New Horizons has no plans to visit an asteroid (it would be an awfully long journey back toward Earth to reach one!), asteroids are nonetheless interesting and worth noting. If you're lucky, you may discover an asteroid rich in iron that is easy to access for mining in the future. If you are unlucky, maybe you'll discover an asteroid on a collision course with Earth! If you get lucky enough to find an asteroid that we have enough data from to calculate its orbit (something we normally can't do), then you and any other discoverers will get to share the honor of naming it!

# Weird Stuff

### Cosmic Rays, Zapped Pixels, and Other Artifacts

The cameras used to take the images that you see on this site are sensitive to a variety of problems. The most common problem is high energy particles (either from radiation coming from the Earth, or from sources in space as exotic as supernovae). These high energy particles can show up as either 1 or more saturated (completely white) pixels that either form a small square or as a streak of white pixels that is about 1 pixel wide. The shape of the pattern depends on how the high energy particle hits the detector. If it is a nearly head-on collision, you get just 1 or a small square of saturated pixels. If instead, the high energy particle glances across the surface of the detector, you get the long streak. Below are some examples of cosmic rays.

Do not mark cosmic rays and saturated particles. The features you are interested in will all be much wider.

### Fluffy Stuff

Things not to mark...

While going through the images you may come across light gray arcs, rings, spotches, or other large fuzzy objects.

You may think: Is this a comet? A nebula? A galaxy? Interplanetary dust?

Sadly, it is most likely just dust on the telescope's optics.

The arcs you see come from really bright objects off the side of the image. The arcs are called internal reflections.

# Marked Example Images

Below are a collection of marked images with possible transients (some may be non-KBOs, but all are worth marking!) Click on any image to view it larger.