The ATLAS system combines a comparatively small telescope with a very powerful camera. The telescope will be an equatorially-mounted 50cm diameter f/2 Wright-Schmidt system, while the camera will use a 110 Megapixel CCD array. A major feature of the telescope will be its large 7.4° field of view - about 15 times the diameter of the full moon. With this system the whole night sky visible from Hawaii can be imaged with about 700 separate telescope pointings. At 20 seconds per exposure, plus 5 seconds readout we can scan the whole sky twice each night, reaching to magnitude 20. This is astronomer-speak for "respectably but not extremely faint." It is the equivalent to the light of a match flame in New York viewed from San Francisco
The detailed specification for the ATLAS telescope may be found here
The schedule for the completion of the project may be found here
Images of the sky taken xx minutes apart
show an asteroid moving against the
backgroud of stars.
The key to detecting asteroids is that they continously move against the more or less fixed background of stars and galaxies: a typical asteroid moves something like 30 seconds of arc an hour. ATLAS will therefore record pairs of images of each part of the sky, with exposures separated by about 30 minutes. The computer system in the telescope dome, which is capable of analysing 500Mbytes of data per minute, will make a detailed comparison of the two images immediately after the second one is taken.
The computer will then compile a list of all objects that appear to have changed either their positions in the sky or their brightness and will then work out which of these objects is likely to be an asteroid moving across the sky, and which may be some other astronomical phemenon such as those discussed here. If an object is suspected of being an asteroid, the computer will see if it can be identified with any of the hundreds of thousands of already-known asteroids. If not, the next step is to check if the object also showed up on previous nights' data; two nights of observation can usually provide enough information to calculate the orbit of an asteroid and distinguish between a main-belt asteroid, which is no danger to us, and a Potentially Hazardous Object, (PHO) which might be. All these calculations are made within minutes of the data being taken.
Go to the Impact Response page to find out what happens if we detect an asteroid on a path that may intersect with the Earth.
Under favorable circumstances ATLAS can detect a 100-meter diameter asteroid at a distance of 40 million kilometers (a quarter of the distance to the Sun) and a 10-meter diameter asteroid at a distance of 4 million kilometers (10 times the distance to the Moon). If we assume the asteroid is approaching Earth at a typical speed of about 20 kph, we will get somthing like about 3 weeks warning for the 100m object and 2 days for the 10m one. As we discuss in the Impact Response page this amount of warning should be enough to move people out of dangerous areas but not to deflect the asteroid away from the Earth.
Unfortunately, a large portion of the sky toward the sun – about one quarter of the night sky – is impossible to look into from the ground without seeing daylight glare and losing sensitivity. Also, parts of the southern sky are not visible from the northern hemisphere since they are below the horizon (about one quarter of the southern sky cannot be seen from Hawai'i). Moonlight also spoils ATLAS's sensitivity This leaves half of the entire sky that in principle can be seen over the course of a night from a single mid-northern latitude site.
In order to assess the effect of these blind spots we undertook a computer simulation in which the Earth was threatened with a large number of fictitious dangerous asteroids. The asteroids had a range of plausible orbits, so that some were easy for ATLAS to find and some were difficult. The graph shows the range of warning times that ATLAS would have provided for these asteroids. We see that most 50 m diameter asteroids will be detected between 3 and 9 days before impact, and most 140 meter asteroids will be detected 10 - 40 days before impact.
When commissioned in 2015, ATLAS will consist of two identical telecopes, one at Mauna Loa Observatory on Hawai'i Island, and a second on the summit of Haleakala, about 100 miles to the north-west. Having two separated telescopes means that we can get extra information about an asteroid by the parallax effect; an asteroid which is close to the Earth will appear to be in slightly different position with respect to the background stars when viewed from the two locations, providing us with additional data for calculating its orbit. Having two telescopes has other advantages: it doubles our survey time, it reduces the chances of the sky being obscured by clouds, and it potentially gives us an asteroid's position at four times in the night, rather than two
In the longer term we would like to see ATLAS-like systems built in the southern hemisphere, to look for asteroids coming from that direction, and at a range of longitudes so that we have 24-hour night sky coverage. Multiple systems would also reduce the amount of observing time lost to cloudy weather.
We consider the initial ATLAS system to be a proof of concept and hope to geographically expand the system once we prove it is an effective and cost-effective asteroid detection system.
ATLAS scientists have all worked on the University of Hawai'i Pan-STARRS project, a system designed to find near-Earth objects (NEOs) moving through the Solar System long before they might strike us. Pan-STARRS is a larger telescope than ATLAS, has a larger camera, and can detect fainter objects than ATLA.. But because its field of view is narrower than that of ATLAS, it takes PanSTARRS many nights of observation to image the whole sky even once. Although Pan-STARRS can give us plenty of warning for large asteroids, it is less efficient at finding the far more numerous small asteroids that may be on course to impact Earth.
ATLAS is designed to complement Pan-STARRS by monitoring "shallow but wide" sections of space. By removing the need to look all the way across the Solar System as Pan-STARRS does we can canvas the sky much more frequently. Even a small asteroid will become very bright when it makes its final approach to Earth; ATLAS will be searching for these every night, weather permitting.
We have already set up and commissioned a pilot system we call Pathfinder in order to test the control and analysis software being developed for ATLAS. Pathfinder's telescope and camera are smaller than those under construction for ATLAS, but capable of testing all the important featuresof the final system. Its parameters may be compared with those of ATLAS in the specifications page.
Pathfinder consists of an 18 cm diameter telescope with an 8 megapixel CCD. It is currently installed inside the Mauna Loa dome that ATLAS will eventually occupy. It is currently automatically scanning the sky in the same way that ATLAS will, and is being used to test the control and data reduction software that will be used by ATLAS.