Strange Rainbow

Reinhard Nitze observed at Barsinghausen (Niedersachsen, Germany) on the morning of July 7, 2007 a strange rainbow produced by a passing shower. Owing to the low solar elevation it had a predominantly reddish colour. When he saw the photos he remarked on the strange step change in brightness, colour saturation, radius and width near the left-hand base of the bow (photo with unsharp mask). The change is marked on the photo by an arrow and is not an artefact because it is present on other images (1 2 3). The upper part of bow is considerably brighter and more strongly coloured although blues are weak and violet is completely absent. This is not easy to explain. A cloud might be shadowing the weaker part of the bow but that does not explain the width or radius change. Smaller drop sizes in the lower region might be responsible and could be obscuring rays from larger raindrops further away. Effects of reflected light bows can be ruled out because no water was nearby. Somewhat similar unusual rainbows were imaged by the Japanese observer Yuji Ayatsuka.

Purple sun?

These pictures were taken on a flight from Geneva to Warsaw on 18 July 2007. Shortly, after the sun disappeared behind a bank of clouds, I was surprised to notice that the LCD display of my Lumix TZ-1 camera was showing a purple sun. As purple is not a colour not normally associated with the sun, my first thought was that my camera had been damaged by taking pictures of the sunset. However, it soon became clear that the camera really could "see" the sun - despite the fact that the sun was completely invisible to my eyes! In fact, the sun remained visible to the camera for about 2 minutes after it disappeared according to my eyes.

Notice that the sun appears to have been squashed vertically (due to atmospheric refraction).

The explanation for this phenomenon is that the CCD sensors used in digital cameras have their peak sensitivity in the infra-red - typically at a wavelength of about 1000 nm, which is well beyond visible spectrum of 400 - 700 nm. You can test the infra-red performance of your own digital camera by pointing a TV remote control at the lens of the camera from a distance of about 15 cm (6 inches). Most remote controls transmit infra-red at wavelengths of 850 - 1000 nm. Your camera viewfinder will probably show a purple light when you press a button on the remote control. The purple colour suggests that the red and blue sensors in the camera are sensitive to infra-red - but not the green sensor. Hence, the purple sun .....