Near Earth Objects represent a clear and present danger to the people of Earth. The Tunguska event over Eastern Siberia in 1908 is a good example of the kind of destruction that a relatively small object can produce.

That was a lucky escape given that the devastation occurred in one of the remotest spots on the planet. But there is every chance that another Tunguska-like object is heading our way.

So it would seem a good idea to identify asteroids with our name on them so that the threat they pose can be mitigated in some way. Indeed, efforts to catalogue Near Earth Objects have been growing in recent years. To date, astronomers have found some 10,000. Of these, more than a thousand are bigger than a 1 kilometre in size.

But there is a problem. These larger objects are only a tiny fraction of the total; near-Earth asteroids smaller than 50 metres probably number in their millions.

Finding these smaller ones is difficult because they are particularly faint. And even when astronomers do spot them, these objects drift past us so quickly that they are gone before anybody can work out their orbits. That’s why the number of these small near-Earth asteroids on astronomers’ books is tiny.



Today, that looks set to change thanks to the work of Michael Shao at the Jet Propulsion Laboratory in Pasadena and a few buddies who have developed a new way to spot and track faint, fast-moving objects in space. These guys say their new technique could increase the rate of discovery by a factor of a thousand.

Here’s the problem astronomers face in identifying faint asteroids. They begin by taking two photographs of the night sky a few hours apart and then comparing them to see if anything has moved. Because near Earth objects are generally moving very rapidly by the time they become visible, they ought to show up in these kinds of comparisons.

But although close, these objects are extremely faint so the cameras must take exposures of at least 30 seconds to capture enough light. And even over that time, a near Earth asteroid will have moved against the background starfield. For that reason, these objects appear as faint smears in these images, making them even harder to pick out.

Now Shao and co have developed a trick that solves this problem. “We present a new technique that is designed to significantly increase the sensitivity for finding and tracking small and fast moving near Earth asteroids,” they say.

Their idea is simple. Instead of taking a single 30 second picture, they take lots of much shorter exposures which each freeze the position of the near Earth asteroid, like a high speed action photograph.

Of course, these images will each be very dim. So Shao and co add them together in a special way. The trick is to shift each image so that the pixels that record the asteroid are superimposed on top of each other, a process which causes the background stars to become streaked instead.

So in the final image, the asteroid appears as a sharp, pin-prick of light against a background of streaks. Shao and co call this new technique synthetic tracking

The difficult part of this is knowing which way to shift each image. Shao and co have solved this by brute force: they take consecutive images and examine all possible shifts to see which resolves the fast moving asteroid.

Given that a sequence of images may contain 120 pictures, that’s a computationally intensive task. “The shift and add process for 120 images for 1,000 different velocity vectors requires over 10^11 arithmetic operations,” they say.

But that’s chicken feed for state-of-the-art computers. “With current off-the-shelf graphics processing units with up to 2,500 processors and teraFLOPS peak speeds, we were able to analyze 30 sec of data in less than 10 sec,” they say.

The results are impressive. Shao and co have put the new technique through its paces using the 200 inch telescope at the Palomar Observatory in California. They spotted and tracked several faint asteroids in this way proving that the technique has huge potential.

And by repeating their synthetic tracking process later the same night, they can get a handle on the asteroid’s orbit. That allows follow up observations during the following nights that should pin down the orbit accurately enough to predict when the object will return.

That has the potential to revolutionise the process of near Earth asteroid discovery. Shao and oc point out that in the last 5 years, astronomers have identified near Earth asteroids with a size of a few tens of metres at a rate of only 30 per year.

By contrast, they say that with a 5 metres telescope, synthetic tracking should spot about 80 of these objects each night. That’s “almost 1000 times higher than the discovery rate of these small objects over the last 5 years,” they say.

It’s hard to overstate the Importance of this work given the potential threat. Tracking and categorising these objects is a crucial first step in developing serious planetary defence against this threat.

But there are other uses for this data. NASA is hoping to send a crewed mission to a near Earth asteroid in the not-too-distant future and has started a program called the Asteroid Grand Challenge to identify potential targets.

One way or another, we’ll be seeing and hearing a lot more about these objects over the coming years.

Ref:arxiv.org/abs/1309.3248: Finding Very Small Near-Earth Asteroids using Synthetic Tracking