American Museum of Natural HistoryThe Lyot Project


The earth's atmosphere has undesirable effects on the light that comes to us from space. These effects can be avoided by placing the telescope above the atmosphere, such as with the Hubble Space Telescope, or if the effect can be adequately measured, then we can endevour to correct it.

Adaptive optics corrects the effect of atmospheric turbulence. The atmosphere is constantly changing, with warm air rising, cold air sinking, and winds mixing this all together. Light is affected by the medium in which it propagates, so starlight passing through the turbulent atmosphere is disrupted. We see this effect with the naked eye as the twinkling of the stars. A long exposure image of a star taken through a telescope reveals this disruption by the fact that the star looks blurred out and indistinct, especially compared with the telescope's diffraction limit, the theoretical limit of the telescope's image resolution. Adaptive optics, a technique now in routine operation at many of the world's observatories, enables a telescope to achieve extremely sharp, essentially diffraction-limited images.

To understand how adaptive optics works, imagine the light from the star as waves. When these waves arrive at the top of the Earth's atmosphere, the wavefronts are essentially perfectly flat. The turbulent atmosphere crumples the wavefront. If the telescope receiving the light simply makes an image with this crumpled wavefront, a blurry image is formed. In adaptive optics, the crumpled wavefront is reflected off of a deformable mirror, a mirror with hundreds of small actuators glued to the back of it. The actuators are commanded to apply the exact shape to the deformable mirror such that the wavefront, upon reflection emerges flat again, the way it was before it entered the atmosphere. The result is an image nearly as good as if the atmosphere were not present. This system operates in closed loop, measuring the residual error (the residual crumple left on the wavefront after it has reflected off of the deformable mirror) using a "wavefront sensor." Because the atmospheric turbulence rapidly changes, this correction must be updated thousands of times each second.

The schematic at right appears courtesy of C. Max, Center for Adaptive Optics.

Example Wavefronts and Images

On the right is a simulation of uncorrected and corrected wavefronts and images. On the upper left is a representation of the corrupted wavefront entering a telescope and corresponding distorted image is shown below it. The center panel shows the deformable mirror crunched into the best fit shape to correct the wavedfront. On the right is the residual errors in the wavefront (top) after reflection off of the deformable mirror and that wavefront's corresponding image (bottom). To see this in motion, click on the image to download a 2 MB Mpeg movie file.

Making the Image Even Better

To see a faint planet next to a bright star, the wavefront must be carefully controlled. Even small wavefront errors produce a bright cloud of "speckles" surrounding the star. If the speckles are brighter than the planets next to the star, then we cannot see the planet. The image at right shows the effect of increasing the fidelity of the adaptive optics, by using many more actuators on the deformable mirror and updating the correction more rapidly. Since the image is "cleaned up" over a much larger field of view, the prospects are greatly improved for detecting planets. There is still a lot of light remaining in the field of view, in the vicinity of the planets orbiting this star, but much of that can be removed with a coronagraph.

Clicking on the image at right will download a 2 MB Mpeg movie file to show this process in motion.

Copyright © 2003 by Ben R. Oppenheimer