The Green flash observed: Three groups of observers report on different aspects of this fascinating and elusive phenomenon.
JUST AS THE SUN'S top rim vanishes at sunset or emerges at sunrise, it sometimes appears a vivid green for about a second--a phenomenon called the green flash. Few people have seen it, but it often becomes an obsession for those who have.
Green flashes are caused principally by atmospheric refraction, which makes celestial objects appear higher than they actually are. The effect is quite strong near the horizon, where the apparent position of an object is raised roughly a half degree--one full Sun width. Refraction is stronger for shorter wavelengths (blue and green) than it is for longer ones (orange and red), so bright objects viewed through a telescope almost always show a blue or green fringe on top and a red fringe on bottom when they're low in the sky. By the time an object reaches the horizon, the fringe on top is usually green rather than blue, because blue light is strongly absorbed and scattered by the many miles of dense atmosphere through which the light must pass. When the green fringe is the only part of the Sun above the horizon, you see a green flash. On the rare occasions that the atmosphere is clear enough to let the blue light through, you may even see a blue flash.
Sunset over the Sea
But the mechanics of the green flash are more complicated than that. Marco Meniero and Andreina Ricco of Rome, Italy, observed 500 sunsets from the Adriatic coast between January 1997 and October 2000, using a 4-inch telescope, 20 x 60 binoculars, and their unaided eyes. During the most favorable months (May, June, September, and October), one sunset out of eight included a green flash, but most of those were visible only with optical aid. A naked-eye green flash occurred only once for every 40 sunsets. They found that telescopic green flashes were most common when a high-pressure system had been in place for three days or less, bringing a cold north wind off the European landmass. Naked-eye green flashes occurred only when the Sun was heavily misshaped as it approached the horizon, but still bright and white.
What's going on here? The most obvious prerequisite for a green flash is good transparency. If the air contains too many aerosols, the entire disk of the setting Sun appears deep red, and both blue and green are removed from the fringe on top. No doubt that explains the importance of a high-pressure system freshly arrived from the north--precisely the conditions that favor good transparency at night. But why does the naked-eye green flash correlate with a heavily distorted Sun?
Astronomer Andrew T. Young explains this on his Web site at http://mintaka.sdsu .edu/GF, the starting point for any serious study of the green flash. He claims that naked-eye green flashes are always associated with mirages--nonstandard refraction caused by steep thermal gradients in the lower atmosphere. The most common and familiar example is an inferior mirage, produced when relatively cool air passes over a hot surface, like sunlit asphalt. In such cases, the air adjacent to the ground can be several degrees warmer than the air just one meter (three feet) higher. Hot air is less dense than cool air, so it refracts light less, and the net effect is that normal atmospheric refraction is reversed, and an object near the horizon appears to be lower than it actually is. You end up seeing an upside-down image of the object underneath the normal upright image. In the case of the classic desert or road mirage, you see the upside-down sky superposed on the ground, mimicking a pool of water. Nakedeye green flashes are most often seen when an inferior mirage of the Sun's green upper fringe merges with the normal erect image, producing an elongated lenticular blob.
Sunrise over Mountains
It's not clear that mirages can account for the numerous green flashes recorded by Jay H. Norman and Claribel T. Norman, who live near the northwestern edge of Las Vegas, New Mexico. They enjoyed watching the Sun rise over the jagged ridge that bounds the Las Vegas basin on the east and decided to record the precise spot of sunrise every morning for a year. Claribel was paying particularly close attention when sunrise reached its southermost point on December 22, 2002, and it was then that she noticed the Normans' first green flash. They recorded some 80 sunrise flashes over the next year, rating each as either OK, good, great, or excellent. Remarkably, two or three green flashes sometimes appeared simultaneously, as different parts of the Sun rose through clefts in the mountains. Jay Norman notes that this phenomenon is a precise analogue of Baily's Beads, where the Sun appears through clefts in lunar mountains during a total solar eclipse, and he has dubbed it "Clari's Beads" in honor of his wife's observations.
A Little Altitude Goes a Long Way
A surprising feature of the green flash (or any sunrise or sunset seen over a level surface) is how sensitive the timing is to the altitude of the observer. Former astronaut Owen K. Garriott discovered this serendipitously in 2002 aboard the Russian oceanographic vessel Akademik Mstislav Keldysh, which was exploring hydrothermal vents 2,300 meters below the ocean surface with submersible capsules. He describes the scene:
Some observers were on an open portion
of Deck 7 and others on Deck 8,
nearby but 2.85 meters higher. Five of us
on Deck 7 saw the usual red disk of the
Sun change to solid green just before
disappearance. We broke into spontaneous
applause. The people on Deck 8
were still seeing a red disk, and one of
them looked down at us, wondering why
we were so excited. He missed the green
flash, but his two partners did see it--several
seconds after we had.
Who would guess that a few meters could make such a difference when the Sun is disappearing over the rim of a planet that is 13,000 kilometers (8,000 miles) across? Yet you can prolong the green flash substantially simply by standing up from a seated position (see S&T: May 1997, page 111). The effect is easy to calculate if you ignore refraction, as shown below. Garriott and his shipmates proceeded to verify this calculation on several separate evenings by stationing observers on up to five different decks. As you can see from the graph at right, the correspondence between calculation and measurement was good but not perfect--possibly because their calculation ignored refraction. Information on atmospheric refraction is widely available on the Web; can you use it to figure out the appropriate correction factor?
Green flashes are a font of contradictions and exceptions. Their underlying causes are very simple, but the details are complex and controversial. They have been studied widely, even obsessively, by both amateur and professional scientists, though they have little inherent scientific value. The great majority appear as brief pulses of green, but some are blue or even yellow and some last for many seconds. They are beautiful and ephemeral, easy to see but even easier to overlook. Have you seen a green flash?
Computing the Effect of Elevation
If you were swimming in a perfectly still ocean with your eyes right at the water line, the Sun would set level with your eyes. But on top of a mountain, you have to look downward to see the horizon, so the Sun is actually lower than your eyes when it sets. To compute the delay in sunset due to elevation, you have to divide the angle between the horizon and the horizontal (the angle of depression) by the rate at which the Sun gets lower.
The angle of depression is easy to calculate by elementary geometry, as shown at right. The Sun travels 1[degrees] every 240 seconds, so it would be losing 0.00417[degrees] (0.0000727 radian) every second if its path were perpendicular to the horizon--which is nearly the case in the tropics. The actual rate of descent at any given latitude and date can be calculated geometrically, but Owen Garriott found it easier to measure the interval from when the Sun first touched the horizon to when it disappeared. That was almost precisely 3 minutes (180 seconds). Dividing the Sun's mean apparent diameter of 0.533[degrees] by that interval yields a rate of descent of 0.00296[degrees] (0.0000517 radian) per second. The net delay works out to 10.8 seconds times the square root of the elevation measured in meters.
TONY FLANDERS hopes to see his first green flash one of these days.
|Printer friendly Cite/link Email Feedback|
|Publication:||Sky & Telescope|
|Date:||Dec 1, 2004|
|Previous Article:||Unsung star clusters.|
|Next Article:||Perseids peak as predicted.|