The cheerful twinkling of stars at night is one of first things people notice when they look up at the sky after dark. Twinkling is not constant: stars don’t twinkle much on cool, still nights, but fairly dance about when the winds are strong and gusty. Twinkling also differs from place to place. While twinkling makes the stars active and friendly to the casual observer, it causes problems for the sky photographer because twinkling is actually the motion and distortion of starlight as it passes through turbulence in the atmosphere on its way to your eye or camera. These distortions make images of surface features on the planets and stars blurred and enlarged. Take, for example, this short movie of the Moon’s surface as seen through a small telescope:
The image at left is a “movie” made by Philipp Salzgeber out of 8 individual frames taken from a video of the Lunar crater Clavius. The image at right is a satellite image of Clavius. The satellite image, taken from outside Earth’s atmosphere is not distorted by twinkling and shows much finer detail.
Note how small features blur and move about. Now imagine how a point of light, like a distant star, would move and blur. The picture below shows the first five images in a sequence of 24 made by William C. Keel, University of Alabama. Each image was exposed for about1/100 of a second, and each image is taken about a second apart.
Notice how the image of the star moves about and changes shape and brightness. Putting all 24 images into a movie shows how this movement causes “twinkling:”
These visual distortions of a star causing the tinkling typically change about 100 times each second. If we take all of these individual images and average them together with image processing, we get what the star looks like on a long film or CCD exposure:
The image appears “star-like,” but actually distorted and enlarged compared to an image taken from above the atmosphere.
Professional astronomers actually measure the amount of twinkling and blurring of stars in images and refer to it as “seeing.” Astronomical seeing is defined as the average diameter of the image of a star as seen through a telescope such as the star above, and it is measured in seconds of arc. The world’s best seeing from the surface of the Earth is about 0.4 seconds of arc; 1.0 second of arc is considered “good” seeing, but when star images are 4 or 5 seconds of arc or more, the seeing is considered so “bad” that the astronomer throws up his hands and does something besides imaging!
The sequence of simulated star images in the figure below shows how the image changes with worsening seeing:
Seeing changes from excellent to poor from left to right.
Turbulence in the air is caused by uneven heating of the air by surfaces with differing temperatures as warm parcels rise while cool parcels fall. Turbulence is also caused by air flowing over a rough surface such as a rocky ridge. As shown below, each parcel of air bends the light in different ways so that an observer looking at an object through the turbulence sees the object flicker and waver.
Figure by Bob Tubbs
Through experience, astronomers have found that seeing is good to excellent at two types of site: 1) isolated high mountains in temperate (neither tropical nor polar) oceans – like Hawaii and La Palma, and 2) mountains near coastlines with cold ocean water upwind. Each situation provides for smooth flowing (laminar) winds around and above the mountaintops and a low elevation to the top of the turbulent convecting layer near the ground. Clouds, dust, and pollutants (if a city is present) largely stay in the convecting layer, so that a low-altitude layer allows the mountaintop to stick up into the smoother layers of clear air.