Simulations of Green Flashes

Introduction

For the benefit of people who have never seen a green flash, I have prepared some animated cartoons of a few typical sunsets. Please bear in mind that these are not very realistic: it is impossible to represent the true colors of green flashes on a computer monitor; the limited dynamic range available forces me to show flashes (which are actually fairly dim, relative to the disk of the Sun) much brighter than they really are; and many important real-world effects, such as solar limb-darkening, and the strong extinction gradient near the horizon, have been ignored to keep the file sizes and download times as small as possible. (Slightly more realistic simulations are being added; but these cannot yet be animated.)

Nevertheless, these animations do show the correct outlines for different colors of light, given a particular atmospheric model. They suffice to show how small the “textbook” flash (due simply to isolating the normal green rim at the horizon) is, and they give a realistic indication of the progression of each sunset in time.

Scales of space and time

There are really two scales on which important features are to be seen:

The overall view

One is the whole progress of a sunset from beginning to end, with the premonitory features that tell the experienced green-flash observer what to expect, and when to look for a flash. As most people will not want to wait the full two minutes and some seconds required for even the fastest sunset, these are speeded up by a factor of 17, and pass before your eyes in less than 10 seconds.

Because of the limited resolution of computer screens, I have made the images appear a little larger at normal working distance than the Sun appears in the sky. If you stand back about 3 meters (10 feet) from the screen, the images will appear the same size as the Sun does in the sky, and what you see will correspond roughly to a naked-eye view of the sunset. At normal viewing distance, the images look bigger — roughly what you would see through low-power field or “opera” glasses.

The details of a flash

The other scale involves the development of a flash in detail, comparable to what is seen through binoculars, or in photographs taken with telephoto lenses. In this case, the images are displayed on a scale 3 times larger, which corresponds roughly to what one sees in binoculars, and in real time (for a low-latitude observer; those nearer the poles would see a more leisurely unfolding of events). The last 8 or 10 seconds of most sunsets is sufficient to give some idea of the way in which each flash appears, but some cases run longer. Note that some browsers cannot display these animations at the correct speed.

What you will see

These simulations work best with a black background, so when you move to the pages where the movies are shown, the text should appear white on black. Don't panic when your screen goes black at first. The simulated sunsets take place over the sea, partly because a sea horizon is favorable, and partly because its simplicity makes the cartoons simpler and easier to prepare.

Each image is flanked by two vertical scales, calibrated in minutes of arc. The horizontal distance between the scales is about 47 minutes of arc, or 1/73 of a radian; so if you measure the separation of the scales and multiply by 73, you'll have the distance to stand back from the screen to see a true naked-eye view of things. If you don't like multiplying by 73, and are willing to settle for less accuracy, measure the horizontal diameter of the Sun and multiply by 100. For example, if the Sun is 3 cm wide on your screen, you should stand back 3 m from the screen to see it “life size.”

The zero-point of these vertical scales is the astronomical horizon, marked by a little extra tick. Because the eye must be above the sea, the apparent horizon is below the astronomical horizon. The apparent sea horizon is the boundary between the dark-red sky and the dark-blue sea in these simulations.

There is also a title displayed with each animated cartoon to identify it. The title appears for 2 seconds, which may not be enough to read it all the first time. You probably don't need to read the title anyway; it's just there as an identifier, and to provide a break between repetitions of the loop. The loop should repeat at least 5 times if your browser can display gif animations. (Some versions of Netscape allow the loop to play over and over again forever.)

Most of the animation files are between 50 and 100kB long. Unless you have a very fast connection to the Net, the animation will take a minute or so to download the first time, during which you can look at each frame as it loads. The loop plays at correct speed after it has loaded completely (unless you have one of the Mac versions of Netscape that fails to insert the proper delay between frames).

Note that the Sun descends vertically in each simulation. That's what an observer at the Equator would see. The timing of the green flashes is also correct for an equatorial observer. If you live at even a moderate latitude and are accustomed to watching sunsets, this will look a bit odd to you.

What's available

At this stage, only a few sunset “movies” have been produced. Eventually, there will be more examples. What we have so far are:

The Standard Atmosphere or “textbook” sunset.

This is the model espoused by Arthur Rambaut, the younger Lord Rayleigh, and many textbook writers; it was rather hesitantly endorsed by Minnaert, though I think he had doubts about it. The simulations here should convince you that this is not how green flashes are formed.

Nevertheless, this model displays the important red (lower) and green (upper) rims, whose miraged, magnified images really do appear in some common flashes. So it's important to understand the “textbook” case, even though it is not the full explanation of real green flashes.

The inferior-mirage or “Omega” sunset.

Simulated inferior-mirage flash This is the flash most commonly seen from near sea level. Here's a considerably magnified (but fairly realistic) simulation, showing what such a flash might look like in 8× binoculars. The animated example is rather extreme: you are unlikely to see such a large and pronounced flash yourself. An atmospheric model was chosen to illustrate the effects about as strongly as they have ever been observed in Nature. Nevertheless, large flashes are seen occasionally, and this example makes all the characteristic features easy to see.

This flash is really the green upper limb of the textbook model, but greatly magnified in the vertical direction at the “fold” where the inverted image of the inferior mirage joins the erect image above it.

I also have a page showing how the computed colors of inferior-mirage flashes change with varying aerosol extinction.

The inferior-mirage or “Omega” sunRISE.

This is the same as the sunset immediately above, only the frames are run backwards. There is a little lead-in of a few seconds of empty sky, so you can see how the Sun appears unexpectedly with little or no warning at sunrise. You can see how easy it would be to miss the sunrise flash, especially if you weren't looking at just the right spot on the horizon.

If you would like to see where the rays go in these mirages, some ray diagrams are available.

Mock-mirage sunsets and flashes

Simulated mock-mirage flash A mock mirage is caused by looking down through a thermal inversion, and then (thanks to the curvature of the Earth) back out through it again beyond the horizon. There are simulations of two mock-mirage sunsets, one with a weak thermal inversion, and one with ducting. This latter sunset shows a more prominent red flash that is shown in detail separately. (The versions shown here now are more realistic than the ones I showed from 1999 to 2002.)

Like the inferior-mirage flash, the mock-mirage flash is essentially the green rim, magnified at the transition where erect and inverted images join. These flashes are best understood by applying Wegener's principle of cutting the atmosphere into two parts: one above the observer that produces the green rim of the setting Sun, and one below eye level that produces the magnifying mirage at a fixed altitude above the horizon. When the green rim reaches that magnifying zone, the observer sees a green flash.

There's also a page showing the effects of varying extinction on the colors of mock-mirage flashes. The results are similar to those for inferior-mirage flashes, as mentioned above.

Eventually, other flashes will be added, like the sub-duct flash at the left; the polar flash, and the cloud-top flash. But the ones here already are enough to give you a notion of what to expect, if you've never seen a green flash.

Please remember that there are several very different kinds of flash; I can only simulate the ones I understand. You might see some flashes that look very different from anything here.

I have also described how the sunset movies were made, for those with a technical interest in such matters.

Back to the GF home page,

or the alphabetic index,

or the Overview page