|Twinned Rainbow ~ Photographed during an afternoon thunderstorm at Stavanger, Norway by András Uhrin. OPOD has slightly enhanced the images. ©András Uhrin, shown with permission|
|Large raindrops are increasingly flattened. Very large drops eventually break up into smaller ones.|
Colour subtraction enhancement shows a clear twinned primary but only one secondary bow
|A media and children's book raindrop|
|Usually during stormy conditions a rainbow will appear and then, perhaps just for a few seconds, split into two. Then it is all over and you wonder whether it was real. Twinned bows are real and by now there are many photographs. Sometimes the twinning is tantalizingly indistinct. At other times, like here, it is clear.
So far though we have never seen a twinned secondary rainbow. The enhancement at right shows obvious separation into two primary bows but only a single secondary. Single, but perhaps just a trifle widened?
A popular explanation of twinning was that a mixture of ice spheres and water droplets made the the bows. An ice bow would be slightly larger than a water generated bow. But the ice spheres would have to be improbably transparent and near perfect spheres, and why is the secondary not split?
A mixture of raindrops of two different sizes is much more likely to be responsible, either in the same rain sheet or in two separate ones. Small raindrops are spherical. Larger raindrops are increasingly flattened by air resistance as they fall. The surface tension forces pulling a raindrop into a sphere are proportionality weaker compared to air resistance distortion in large drops.
A primary rainbow from slightly flattened drops is itself flattened at its top – but – ray tracing calculations show that the secondary is not. Two populations of drops of different flattening will thus create a double primary but a single secondary.
We need two populations of drops and each almost monosized. That too is improbable. Yet, even ordinary rainbows need such perfect drops and sunlight conditions that they are a wonder.