Joule also observed that “Just at the upper edge, where bands of the sun's disk are separated one after the other by refraction, each band becomes coloured blue just before it vanishes.” This second form of flash is associated with a mock mirage, which is caused by a thermal inversion below eye level; so it is mainly seen from elevated positions.
There are also several less-common forms, but no photographs of them are available. I have made animated simulations of the common forms.
There is also a page showing the range of observed colors of flashes, if you just want a quick look at some good examples.
There are a lot of links to other peoples' sites here; if you find a broken link, please tell me about it.
The pictures shown here were all taken with long-focus lenses, and show considerably more detail than is visible to the naked eye. Many of the nice flashes here would in fact pass unnoticed without some optical assistance; see the page on how to observe green flashes.
Please respect the copyrights of the original photographers.
I'll start off with a pointer to Les Cowley's atmospheric-optics website, which has a fine section on green flashes. He's found a good selection of pictures to illustrate the common types.
Now, on with a list of flashes scattered all over the Web. . . .
These flashes and low-sun distortions were photographed from Pic du Midi in France by Eric Frappa. There is a good account of the phenomena in French, as well.
Lu Rarogiewicz's pictures taken at Mt. Wilson, with a pretty decent explanation. These very thin forms are typical of flashes seen from high above the sea.
I was sorry to hear that Lu died suddenly on July 2, 2007. He was very helpful to me when I first started working on green flashes.
Another fine set of blue and violet rims photographed from a mountain observatory, provided by Christophe Gervier. These full-resolution images are needed to see all the beautiful details; but if you find them too large and slow-loading, there's also a reduced set available.
Patrick Muller's extensive sequence shows the progression of the sunrise very well, including the sequence of colors in this mock-mirage flash. (Note that some colors in the later, overexposed images are artifacts.) The first, pre-sunrise image showing aerosol scattering in the valleys below, and the final triangular shadow of the mountain are a nice bonus. (Thanks to Luc Dettwiller for providing the URL!)
There's also a nice GIF animation of a mock-mirage flash at sunset on his Web pages here. The individual frames can be examined as well.
Guillaume Blanchard's photographs are arranged as a slideshow. Both mock-mirage and inferior-mirage flashes appear here. Images 4-7 show the progression of a fine blue flash; image 7 clearly shows the true (red) color of the low Sun in the scattered light along the distant ridge, demonstrating that the white color of the overexposed disk in image 4 is an artifact. (Thanks to Luc Dettwiller for providing the URL!)
Page back to the “panorama” that precedes it, to see how the flash developed. Note the strong extinction gradient, which helpfully dimmed the (normally much brighter) lower image.
Photographed from the top of Bantiger mountain (950m above SL) near Berne, Switzerland, above a sea of clouds at around 800m, by Alfred Weiersmüller. The same site has some other pictures of interest, taken by other photographers, such as a nice series showing the formation of an inferior-mirage flash.
One must be careful here; many of the photographers don't understand what they have taken, so many of the images are mislabelled. Some just show artifacts, not natural phenomena.
A nice flash I took at Alpine, CA, about 800 m above sea level. As you'd expect from that height, this is of the “Mock Mirage” type. This photograph accompanies my 2000 paper in JOSA A.
A nice flash taken from Torrey Pines, CA (about 100m above the sea). This is a classical form seen from seaside cliffs, associated with the “Mock Mirage.”
Danilo Pivato's similar mock-mirage flash. Click on the arrow to see the full sequence as a gif animation.
A San Francisco sunset sequence, photographed by Mila Zinkova, who has a fine collection of GF photos. This sunset has mock-mirage flashes, due to a series of inversion layers. The green color in the upper right-hand image almost approximates the vividness of the color one sees with the eye.
This sunset shows two inversions: a surface-based inversion that produces the line just at the sea horizon, and a higher one that made the mock-mirage flash. George Howard was the photographer. [The original page has vanished from the Web, but is preserved on the Wayback Machine at http://web.archive.org/web/20131028235240/http://www.pbase.com/warthog2100/image/52487119 ; it's also shown on the SpaceWeather website.] Notice how the sky darkens rapidly near the sea horizon, just where the yellow line appears; both are due to the rapid increase in refraction just above the sea surface.
This page of pictures has links to two shots of the same sunset flash (the two numbered 7781 and 7782, at the bottom of the page). With the ship on the horizon, it's like the frontispiece of O'Connell's book.
Notice that the inferior mirage also produces an inverted image of the distant clouds. The compression of the inverted images is quite obvious here.
A page by a French pilot, with a beautiful inferior-mirage flash, partly truncated by another ship at the horizon. The color here is gorgeous. (Click on the image to see a larger version.)
A very picturesque flash framed by the Golden Gate Bridge, seen from the Berkeley Marina; taken by Paul Kamen. He has a video of this flash, as well! (Compare his video to my simulation of such a flash; his proceeds a bit more slowly, because of his latitude.) Yet another flash with a vessel nearby.
One of Mila Zinkova's recent GF pictures. She has been collecting flashes since late 2005 in San Francisco. This one turns out to be a little more complicated than a simple inferior-mirage flash (I've seen her other frames of this sunset).
Notice the gradation in color within the flash: yellowish in the center, bluish at the extreme ends. This picture also shows the effect of atmospheric reddening well in the prominent contrail. The orangish hue of the sky, and the visibility of blue in the flash, are indications of extremely clear air.
In addition, there's a small cloud beyond the horizon, miraged by the same layers near the sea surface that stretch the normally invisible green rim into a green flash. It's to the left of the flash, and about the same size; notice that it's exactly the same height above the horizon as the flash.
Mila has links to several more of her pictures of green flashes, sunset mirages, halos, rainbows, glories, and other atmospheric-optics phenomena here.
Jon Adamson's inferior-mirage flash sequence, taken at 5 frames/second with a Canon EOS 20D digital SLR with a 300 MM Image Stabilized lens, from a ship. The flash is apparent in the last few frames (on the second page of thumbnails), but you have to go to the larger images to see it. Part of the sequence is repeated at higher magnification on the 3rd page of thumbnails (and their corresponding full-sized images).
A naked-eye flash taken March 30, 1998, from near Scripps Pier, about 11 meters above sea level. A larger high-resolution scan of the same flash is also available. Note the yellow flash disappearing in the middle of the green one.
I also use this flash as an example on my Omega page, where the flash is compared with the image of the whole Sun when its lower limb was miraged (instead of the upper limb, as in this picture).
Several pictures taken by Steve Rossi with a fairly short focal length give a good approximation to the visual appearance of these phenomena. There is a cloud-top flash as well as several inferior-mirage flashes.
David Dilworth's excellent example of a cloud-top flash seen from near sea level. As the dip of the horizon is small here, you can see that these flashes occur well above the astronomical horizon. (See the green-rim page for details.)
John Coveney caught this mock-mirage flash while on holiday in Sardinia. The green tips on the lower (yellow) part of the solar image suggest there may have been an inferior mirage as well. (Click on the "View on black" link just beneath the image for an enlarged and much better view of it.)
The red sunset sky — which is the same color as the Sun at the horizon — is well captured, showing that the yellow color of the (overexposed) lower image is an artifact.
Martin Rietze photographed these green and blue flashes in Hawaii. Be sure to click on the small photos to see the details; the last picture (in which the vanishing blue flash is hardly seen in the small version) shows the very last, almost violet stage of his blue flash. The clouds to its left show the same vertical stretching as the sliver of the Sun's upper limb — both due to the inferior mirage, which is the usual cause of flashes seen in Hawaii.
A fine set of good examples taken by Robert Wagner from Roque de los Muchachos on La Palma in the Canary Islands. The blue and violet colors here are helped by the great height above sea level (2200 meters), which means less atmospheric reddening when the flashes occur above the clouds instead of the sea horizon.
The photographer has called my attention to another interesting photograph, which shows the “silver lining” of a cloud with green and blue bits along its upper edge, and a red lower rim — as well as a nice cloud-top blue flash from the Sun! The upper “green rim” of a cloud is ordinarily too attenuated by atmospheric reddening to be observable; but the photographer is high enough here to be above much of the atmospheric reddening, making this rare view of dispersion effects on cloud rims visible.
An inferior-mirage flash, showing some gradation of color.
One of Pekka Parviainen's fine green flashes is down at the end of this long page of mirage images. He also has a video of the GF sunset. There's also the bonus of a selection of typical distorted sunsets, showing many beautiful types of sunset mirages.
Still more of Pekka's mirage and distorted-sunset pictures, with several green flashes, are available on his own pages. Click on the "Sun" item in the menu at the left side of his main page to find the green-flash pictures.
Finally, a very nice double mock-mirage flash of Pekka's was chosen by the “Astronomy Picture of the Day” page for Nov. 10, 2002.
Paolo Candy has combined a beautiful inferior-mirage flash here with a solar eclipse, photographed from Madagascar. He has put his experience with green flashes to work, capturing just the moment when the film registers green over the whole flash. There's also a good description of what's going on (in Italian).
This one shows a lot of internal structure, thanks to a long focal length. There's also a good study of the predictability of flashes, mostly in Italian, with some English.
This disappeared for several months after he died. It was temporarily resurrected, but now seems to exist only on the Wayback Machine. There is an interesting perturbation of the flash (visible really only in the enlargement at the bottom of the page) caused by the wake of an airplane.
These interesting pictures were taken at the South Pole sunset of the year 2000. They show a type of flash produced by strong thermal inversions. Notice that the blue parts are generally much fainter than the green ones, and lie outside them. For context, see the accompanying Sunset pictures.
Another example (polar sunset, 2001), taken by Charlie Kaminski. (Thanks to Wayne Davidson for pointing this out.)
A more recent example (polar sunset, 2005), taken by Guillaume Dargaud. He uses the term “green ray” because that's the literal translation into English of the term used in his native French. There's also a nice example of a distorted moonrise, with green upper rim and red features on the lower one. Ignore the remarks about the Sun being reflected by a mirage on the ground; it's seen through a leaky duct, whose upper surface he confuses with the true horizon. The best pictures show complex features due to waves on the inversion overhead. Unfortunately, they were taken with a digital camera and are probably affected by artifacts. (Thanks to Andy Beeler for pointing out this site.)
The same observer has now provided another page that contains an even more interesting picture.
Laurent Laveder's nice examples of the green and red rims on the rising full Moon, with a small mock-mirage flash due to waves on a lifted inversion — an example of Fraser's mechanism. Click on the images to see larger versions, and don't miss the second page of images.
In fact, he has yet another page of images, showing a fine time-series of an Omega sunset and the gradual disappearance of the upper blue rim, as shown in a series of pictures that ends in an inferior-mirage flash. (Again, click on the pictures to display larger versions.)
A mock-mirage red flash. The green rim is visible at the upper limb; this, together with the small extinction gradient across the disk, and the asymmetries of the red flash (indicating waves on a inversion) suggests that the inversion was above the observer (which would make this an example of Alistair Fraser's type of flash). (Thanks to Luc Dettwiller for pointing out this page.)
This picture was taken at Shonaihama, Yamagata prefecture, by photographer Hiroyuki Murata, who takes sunset photos there more than 300 days a year. He says the atmosphere at Shonaihama is unstable and mirages are often produced. (Thanks again to Takayuki Kono for pointing out this page, and for translating the text.)
Park McGraw's series of four pictures, showing how an inferior-mirage flash developed. [as archived on the Wayback Machine]
A couple of nice inferior-mirage flashes and one small mock-mirage flash, accompanied by good clear discussion. Jim is no relation, but I used to work with him at Table Mountain.
Oscar Medina's very nice inferior-mirage flash. Notice the very obvious gap between the flash and the apparent horizon. (Pacific Beach is the part of San Diego coastline that's nearest to me; it's just south of La Jolla.) His original inferior-mirage flash is now accompanied by several more recent flashes, including some mock-mirage flashes.
His pictures also show some interesting variations in the dip of the sea horizon, which I discuss elsewhere.
A couple of nice inferior-mirage flashes about halfway down the page, with other sunset pictures. Follow the links by clicking on the pictures for the May 18 and Sept. 7 sunsets to see the full sequence; the latter nicely shows the progress of an Omega due to the inferior mirage.
Hank Hogan's sunset sequence with both the commonest kinds of flash: first a mock-mirage flash, then an inferior-mirage flash.
Ken Rockwell's inferior-mirage flash. A little jiggled, but nice color. The GF is down about the middle of this longish page.
A fine example of a “cloud-top” flash. The extreme flattening of this flash shows that it is caused by a thermal inversion; the fine diagonal structure within the flash is due to waves on that inversion.
Three images of an overexposed inferior-mirage flash, followed by a much better example of a “cloud-top” flash.
This sequence of 10 images shows the development of a flash associated with strong thermal inversions. The red band below the bright yellow image is a mock mirage of the Sun's lower limb; the discontinuity of colors at the level where the upper image tails off to sharp points at the sides marks the top of a duct. I have simulations of the phenomena produced by ducts; compare this picture with the third image in the views of a deeper duct seen from 65 m.
Explanation in German and pictures taken by Peter Pammer, on the Web pages of the Arbeitskreis Meteore of the Verein der Sternfreunde. Follow the “Bildarchiv” link at the bottom of the page to see a couple of mock-mirage flashes, and a fine mock-mirage lunar red flash.
You don't often see such finely recorded visual observations as these; the GF drawings are down near the end of this page. Don is a professional artist who specializes in astronomical subjects; his pages also contain some unusual halo phenomena.
This is a gif animation of a sunset sequence. The mock mirage shows up very nicely in the 3rd frame of the series, and there is the corresponding green flash in the 9th frame. The animation gives a good feel for the progression of forms in this typical mock-mirage sunset. Associated pages show several fine rainbows and a beautiful example of anticrepuscular rays.
Ron Wolf's mock-mirage flash. Actually, two mock mirages close together here, with waves on the inversions.
A couple of sunsets with strong ducting features are shown down near the end of this page, one with a small green flash in the upper part. This is one of the less-common forms that are not yet well understood.
At the end of Journal entry #20 on this page, you'll find an Antarctic flash very similar in form to the one mentioned just above, photographed from Mauna Kea. Evidently these geographically distant flashes — one from the tropics, one from the polar regions — have a common origin: a strong thermal inversion.
A small blue flash of the mock-mirage type. Nice colors.
A small inferior-mirage flash, with a slightly yellower center. This is the end of a sequence showing the progress of a sunset. The pictures are typical, rather than striking.
Juan Carlos Casado's picture. The flash is small, but the colors are good. Seems to be a mock-mirage flash, probably associated with waves. [Lost; but saved on the Wayback Machine.]
A little green shred above a setting Sun. No text. This is probably one of Fraser's “mountain-wave” flashes.
Nice images of the green and red upper and lower rims. Compare with my rim simulations.
What's more notable in this wide-angle image is the brightness gradient in the sky, which illustrates the extinction gradient. (When the Sun is high, the horizon is brighter than the sky above it — the opposite of what you see here.)
Not a green flash; but such nice examples of the strong refraction associated with thermal inversions that a link here seems appropriate. Caution: some of the pictures show artifacts due to an overexposed solar image.
Now that you've seen some modern pictures, you might be interested in the history of green-flash photography. I have a short page describing a few “firsts” in this field.
You might also like to look at some animated simulations I have prepared to illustrate the effects of different kinds of atmospheric structures. Some of these reproduce the general features of certain common types of green flash remarkably well.
Occasionally, I'm asked if I have a video of a green flash. No, I don't. The first person to have posted a good video recording of a GF sunset is the remarkable Finnish photographer Pekka Parviainen, who has provided a downloadable version (as a *.ram file) on the Virtual Finland Web pages. Pekka seems to be allergic to e-mail, but his address and phone number are provided on that Web page. He is quite capable in English, so don't worry that you might need to speak Finnish to communicate with him.
There is a second video of a GF sunset at The Short Fat Guy On Vacation but unfortunately for me, this, too, is a RealVideo file, which I can't display. I have looked at it on the machines in our library; the flash appears to be an ordinary inferior-mirage flash, and the color (as displayed on the Macs in the library) is rather pale and unsaturated. The whole sequence is overexposed, which may partly account for the lack of vivid color in the flash. This clip seems impressive to people who are not familiar with green flashes. Pekka's sequences (see previous paragraph) are better.
Paul Kamen has now made a 7-second *.avi clip available, nicely showing an inferior-mirage flash as it unfolds. A link to the video is also hosted by Les Cowley.
Though YouTube contains a couple of rather feeble green flashes, most of the clips there show nothing resembling a green flash. There are a few obvious fakes; several video-camera quirks due to the interaction of wobbly hand-held cameras with autoexposure algorithms (e.g., when the dark sea fills most of the frame, the camera brightens up the picture, making ignorant photographers mistakenly think some kind of “flash” has occurred); a number of common overexposure artifacts of digital cameras; several sunsets in which the sky is so cloudy, or so murky with reddening haze, that a green flash could not have been seen; and many sunsets taken with such a short focal-length lens that no flash could have been captured, even if one was present. An amazing number of the clips claiming to show green flashes just show ordinary sunsets, with nothing visible to suggest why the photographer made such a claim.
Don't waste your time looking for needles in this haystack of pathetic junk. The only authentic green flashes I've found there are a couple of very typical inferior-mirage flashes photographed by Andrew Penketh in 2007 and 2008 at St. Lucia. YouTube also has several copies of Paul Kamen's good GF clip, mentioned above.
There are a few movies that purport to have a green flash, but most show only clumsy fakes done with crude special effects. However, the 1986 Eric Rohmer film “Le Rayon Vert” (available with English subtitles as “Summer”) actually has a textbook example of an inferior-mirage flash, photographed by Philippe Demard. This film is also available on videotape. The film has many references to the Jules Verne novel, and a clear exposition of the textbook explanation. Not bad!
Crepuscular rays are shadows of objects (usually cumulus clouds, but sometimes a mountain). If the object is beyond your horizon, you won't see the obstacle that casts the shadow, just its shadow in the hazy air above you. (In the picture above, notice that the ray is darker than the adjacent sky at the same altitude above the horizon.) If there are lots of clouds, there may be only a few gaps between them; then you see bright “rays” between the darker, bluish shadows.
Crepuscular rays typically have time scales of tens of minutes, as opposed to a few seconds for green flashes. Also, crepuscular rays have angular lengths of tens of degrees, while an ordinary green flash is always smaller than the size of the Sun (half a degree), and usually only just above the resolution threshold of the eye — a few minutes of arc.
While there is a “green ray” rarely associated with green flashes, even these are usually only a few degrees long, and share the brief time scale (a few seconds) of the commoner flashes. And as these green rays are brighter than their surroundings (unlike the crepuscular rays, which are darker), even these rare green-ray displays should not be confused with the much more common crepuscular rays — most of which are blue rather than green.
There are some spectacular examples of crepuscular rays on the Web, as well as many more insipid pictures; here are a few of the better ones:
Greg Smith has photographed another striking crepuscular ray; and as he has no website, has granted permission for me to post it here. He says, “The photo was taken just after sunset on July 20, 2000, on the Waikaloa coast of Hawaii (just north of Kona).” You can see, in the upper part of his picture, some clouds that are probably similar to the one just over the horizon that is casting a deep blue shadow on the hazy air. Outside this cloud shadow, the haze is illuminated by the orange light of the setting Sun. At the lower left in his picture, just above the horizon, another ray is faintly visible; the two rays converge to the point just below the horizon where the Sun is.
Crepuscular rays are the shadows of clouds or distant mountains, often just over the horizon, cast on the hazy lower atmosphere. Their apparent divergence from the Sun's position is merely an effect of perspective, as these shadows are all nearly parallel lines in the air.
Philip W. McKee has sent me another fine example that clearly shows the shadows of clouds. He says, “I live 22 miles south of Cape Canaveral and our condo is on the first dune. I try to catch sunrise often. The attached pix were taken off our balcony in early July, 2001.” He adds that the picture was taken more than 10 minutes before nominal sunrise; so these rays, too, are literally “crepuscular” — that is, twilight.
Ten minutes corresponds to a solar depression of about 2.5° at low latitudes. The dip of the horizon is 2.5° at a height of a little over 7 km, so the parts of the cumulus tower higher than that are in full sunlight. You can see from the shape of the tower that it has flattened out near the tropopause (probably about 14 km) — notice the weakly developed “anvil” at the right side. The haze layer is higher than that; it might be just about at the tropopause. The lower atmosphere is still in the Earth's shadow, but the high clouds are in sunlight. If you extrapolate the rays until they converge at the Sun, you find it's just off the left edge of the picture, and just a little below the horizon. Here, a haze layer just above the clouds is illuminated by the same reddened light of the low Sun as the cumulus towers, whose shadows on the haze appear blue. You can see that the “blue rays” here are just the blue sky, uncontaminated by the reddish haze, which contrasts nicely with them.
Recently, the remarkable photographer and dye-transfer printmaker Ctein has put up an interesting picture that nicely illustrates how confusion can arise. It shows some (relatively weak) crepuscular rays, as well as a piece of sky near the Sun that appears quite greenish at first glance — though if you mask it off, you can see it's really a greenish yellow, and the perceived green is mostly due to simultaneous contrast with the purple arc above it. There is a lot of interesting optics in this picture: the purple arc, the distortions of the reflected sky by the ripples in the water, clouds that appear brighter or darker than the surrounding sky depending on location, the crepuscular rays, and the appearance of green in the sky — so it's worth waiting for the large (161 kB) image to download. And it's a beautiful picture, to boot. Go take a look!
An interesting discussion of the “green Sun” phenomenon was given by
C. Michie Smith
Observations on a Green Sun and Associated Phenomena
Trans. Roy. Soc. Edinburgh 32, 389–409 (1884)
following the eruption of Krakatoa, which produced a strong stratospheric haze of nearly constant particle size the previous year. (Contemporary accounts refer to this aerosol layer as “ash” or “dust”, but it was almost certainly made up of photochemically generated droplets of sulfuric acid, resulting from the great quantities of sulfur dioxide injected into the stratosphere by the eruption.) The “green Sun” phenomenon was widely observed, and accounts appeared in all the major scientific journals of the day, as well as popular versions in newspapers; but I think Michie Smith's paper is one of the best first-hand accounts. (He was also an experienced green-flash observer, by the way.) Unfortunately, he concluded that the coloration was due to absorption by water vapor.
For a modern treatment of the subject, see the Meinels' book. They cite the full report compiled by the Royal Society, and discuss the explanation that is accepted today for these curious and somewhat rare displays. As they point out, both bluish and greenish colors can appear.
Finally, sometimes green errors occur in pictures that can fool the unwary. These imaging artifacts are particularly common in digital images and videos. You can't accept pictures at face value — particularly if they show features that were not seen by the eye of the observer.
© 1999 – 2014 Andrew T. Young
or the alphabetical index
or the website overview page.