Why is the Sun Green?

Question posed by Jennie.

I know, I know: the Sun isn't green, but in my why is the sky purple? post I boasted that I could offer a fairly convincing argument that it is, and Jennie has called me on that. So here goes:


What is colour?

Well, that's easy, isn't it? You look at stuff, and you can easily tell what colour it is, right? Look at the picture to the right, for example, and most of us will see an image that our brains tell us contains a lot of pink.

By Edward H. Adelson, via Wikimedia Commons

Unfortunately, when we're doing science, it's not always good enough to just look at something and say "that's sulphur springs yellow 5, that is." This is because we can get it wrong; our brains can be fooled fairly easily.

If you don't believe me, take a look at the image to the left. This is a fairly well known illusion published by Edward H. Adelson in 1995. If you don't know what the illusion's about, then take a look at the squares marked A and B. They are exactly the same colour*.

Don't believe me? You could try saving the file to your hard drive, opening it up in an image editor (even MS Paint will do) and using the eyedropper tool. If you couldn't be bothered doing that, here's an alternative demonstration that I found on youtube...

... so you can see that our eyes (and brain) sometimes deceive us.

Also, something's colour can change drastically depending on lighting levels and surrounding environment, yet our brains keep its colour fairly constant. We need something that doesn't change that much.

One way we can define an object's colour is by measuring its spectrum. The 'spectrum' of an object (in short) is a measure of the intensity of each wavelength of light measured coming from it. It would make sense, then, to label an object with the colour that corresponds to the wavelength at the peak of this spectrum; that is, the wavelength that is measured to have the greatest intensity.

Get to it, then: why is the Sun green?

The Sun's observed electromagnetic spectrum looks a lot like this:

Forget about the annotations and just look at the black curve- this is the 'blackbody**' curve for an object with a temperature that is about the same as that of the Sun. The curve peaks at a wavelength of a bit more than 500nm*** (even if you don't know what "spectral irradiance" is, you can find the highest bit of the black curve and look straight down to the axis marked "wavelength", and read the number). Light with a wavelength between 490 and 560nm corresponds to the colour that we perceive as 'green'.

Because of this, we'd have a decent case to put forward for claiming that the Sun is green!

Why doesn't it look green, then?

Firstly...

DO NOT LOOK DIRECTLY AT THE SUN. EVER.

You can see from the graph that although the Sun's spectrum peaks around the wavelength for green light, it actually kicks out a fair amount of light across the entire visible range (the dotted lines show this range). When you mix all the different wavelengths of light together, it looks white.

Why it doesn't look white is another matter, but it's linked to the reason why the sky is (or isn't) purple.

* O.k, for the pedants, they are exactly the same shade of grey.
** A blackbody is an ideal rather than an actual thing. It's a model of a perfect absorber and emitter of radiation that can be adjusted for bodies with different temperatures. The Sun is not a perfect blackbody (if it was, the edge of the yellow squiggly bit of the graph would match up perfectly with the black curve), but as can be seen from the graph, it is a pretty close approximation and helps us to simplify things.
*** "nm" stands for "nanometres". There are 1,000,000,000 (1 billion) nanometres in one metre.