# What is the meaning of ellipticity of a galaxy being more than 1?

I was recently doing a project on calculating ellipticities of galaxies using SDSS data. I downloaded the stokes parameter values (q and u) from data

$$e = 1 - \frac{b}{a} = 1 - \frac{1 - \sqrt{Q^2 + U^2}}{1 + \sqrt{Q^2 + U^2}}$$ and for few values of stokes parameters, the value of ellipticity goes above 1. What could this possibly mean? Could this mean the data of random galaxies I've taken has irregular agalaxies in it and that's the reason for ellipticity being greater than 1? I would like to hear your opinions and answers.

• Stokes parameters are related to the polarisation of electromagnetic radiation, not galaxy morphology. Where did you get the equation for ellipticity in terms of Q and U?
– user24157
Aug 18 '19 at 21:18
• @mistertribs When it comes to galaxy morphology, "Stokes parameters" has a different meaning than the typical one. Aug 18 '19 at 21:40
• The ellipticity cannot be greater than 1, for any kind of galaxy. (Or equal to 1, really, as that would require an infinitely thin disk.) Aug 18 '19 at 23:31
• It's not the ellipticity in your data that's greater than 1, it's the raw Stokes parameter, which is allowed to be greater than one, or less than zero. Pop that into the equation and you'll see that you cannot get out a number greater than 1. Aug 19 '19 at 19:50
• @SaikrishnaK My guess would be that some of the "Stokes parameter" values are bad (in the link in HDE 22686's comment, there is mention of the Stokes parameters not being very reliable when the S/N is low). E.g., the "q_r" value in the line you highlighted. So I'd just ignore those sources. Aug 20 '19 at 15:13