An astute viewer (such as yourself) may have asked, "How can the universe be 78 billion light years across when the age of the universe is only about 13 billion years?"
Good question, how can something be larger than then distance travelled at the speed of light? Since light from the beginning of the universe has only had 13 billion years to travel (not 78 billion), then shouldn't the universe be only 13 billion light years across? That's a pretty intuitive thought.
But it doesn't take into account that the entire universe itself is also expanding. When a photon of light leaves it's point of origin, it does so at the speed of light, so in a universe that doesn't expand, a photon travelling for 13 billion years traverses 13 billion light years.
In a universe that DOES expand, all of the distance covered by the photon gets increased by a scale factor equal to the rate of expansion of the universe.
Since the universe has expanded some since it left 13 billion years ago, we have the apply a scale factor to account for the expansion. Keeping in mind that the universe is expanding continually, it's not stopping and starting, you have to do some calculus to solve the problem. When you do that, you come up with the size of the universe being 78 billion light years in radius, 156 billion in diameter.
Until recently, it was thought that the rate of expansion of the universe was slowing down. Recent measurements of the cosmic microwave background radiation have shown that the universe is actually accelerating, not slowing down.
So, what I was referring to in the video is called the Comoving Distance, or 'proper distance'. You can get a more detailed definition from the above link, but the comoving distance is a more accurate measure of the size of the universe because it takes into account the fact that the observer (us, the earth), is moving. It also takes into account that the universe has been expanding since its beginning.
Here is a another great article about the size of the universe to get more info.
Where Did Hubble Look?
So where exactly did the Hubble look to take the deep field images? Here is a photo with the region of sky I referred to in the video. As you can see, the area in the L-shaped outline is devoid of all stars. This was done on purpose. Astronomers didn't want any stars from the Milky Way galaxy to get in the way, so they selected this region.
It's important to realize that dedicating so much Hubble telescope time to this little project was a risky move. Time on this telescope is expensive, with very long waiting lists of astronomers who want to use it. It was risky because no one knew what they were going to see if they did this. I think taking the risk paid off, in a HUGE way.
This section of sky is located in the constellation of Ursa Major. This constellation lies outside of the disk of the galaxy so there are fewer stars to dodge by looking here. It was important that the image not be contaminated with foreground stars from our own galaxy. To me, that made the image all that more amazing, because every single point of light in that picture was sure to be a galaxy.
The irregular shape of the area outlined above corresponds to the fact that the complete Hubble deep field image was pieced together from three individual images taken with the telescope pointed in adjacent areas of sky. The detector on the Hubble Space Telescope employs a really old CCD that is 800x800 pixels square. To cover more area, they took many sets of images and moved the telescope around as they did so. Then they stitched them together to make the final image.
The diagram at left is a schematic of where the hubble looked for the 1995 Deep Field image. The constellation outlined is the Big Dipper, or Ursa Major the Great Bear.
I mention in the video that the Hubble stared into this region of sky for a little over 10 days. This was not done all at once. Many individual images were taken over the course of weeks, and then all of them were added together.
Adding images together like this is common in astronomical imaging. If you take, for example, 10 images with exposure times of 10 seconds each and then add them all together, it produces one image equal to an exposure time of 100 seconds.
The advantage of doing it this way has to do with the way images are produced by a CCD detector. CCD's produce more noise or 'grain' to an image if you just let them sit there collecting light. Less grainy images are obtained if you just add a bunch of shorter exposures together. With each image added, the light from the galaxies increases by the amount that the image was exposed, but the graininess increases by a lesser amount (the square root of the number of images for those technically motivated).
The result is an image that is sharper and has more detail.
To take the Ultra Deep Field, the Hubble Space Telescope looked in the direction of Orion, in the constellation Fornax. I'm afraid I don't have any pictures of this area yet, I'm still trying to find some, but the region in the animation on the video are accurate. It's just a little harder to see exactly where the Hubble was pointing when it took those pictures.
By the time they took the second image that became the Ultra Deep Field, the Hubble had been outfitted with an infrared camera called NICMOS, a 256x256 pixel camera that allowed them to see even more galaxies.
They also imaged using both cameras for a little longer, over 11 days this time, to produce the Ultra Deep Field. That image represents the farthest we have ever seen into the universe. Keep Looking Up!