Contrary to popular belief, astronomers are not completely nocturnal folk who stay up all night, every night, pondering the stars. We only do this sometimes! Some of us don't even do it at all!

When we observe with telescopes, it is a thoroughly organised affair, because we have to get approval to use the telescope for a certain amount of time. Depending on the telescope and the project, you might be lucky to get half a night allocated to you. And in other cases you might be so fortunate as to have a whole week of observing time.

The preparation for an observing run (what we call time using a telescope) includes deciding what to observe, for how long, and any back-up plans you can think of if the conditions aren't right for your type of observations.

Alas, we are victims of the weather, especially if we are using ground-based telescopes (as opposed to telescopes in space, which don't have the weather problem, but are much more difficult to get time on). That is unless you are a radio telescope observer - in which case you may even be able to observe during the day, cloudy or not!

Another popular myth is that we spend all of our observing time looking through the eye-piece of the telescope. As much fun as this is, it's not usually possible on the really big telescopes. These telescopes are designed to collect their data into a computer, whether it be images, or images that our eyes could not see, or spectra.

A telescope can point at the same target for hours on end, collecting as much light from these distant objects as it can, whereas our eyes can only collect so much. So using computers here is crucial.

Most of our work, like analysing all the images from the telescope, is done on computers.

This means much like other people working, we spend the typical 9-5 hours at our desks with our computers.

There are even astronomers who don't do any observing with telescopes and create whole universes to study with supercomputers and using simulations. These are the theoretical and computational astrophysicists.

As astronomers we can't do the same kinds of experiments as other scientists, because you can't exactly bring a star, planet, galaxy or black hole into a lab, and observations are limited by how well the telescope can see some of these very distant things, so creating computational models helps us to understand what's going on.

These models use different laws of the universe, and then let the system you've created evolve. This can be a baby star or planet, the birth of the galaxy, the evolution of stars and galaxies and even the fate of the universe.

These simulations can even help with observations: sometimes observations confirm the theory, or give suggestions on how it should be changed, or done differently to get a deeper understanding of what is going on.

To do these computational models and simulations, people often need more computing power than a standard computer, and so will connect to supercomputers (no, not like in The Martian where he plugs his computer in directly, but through the magic of networks).

Using a supercomputer isn't too dissimilar to the process of using a telescope - you still have to request time and be approved; the difference is that you don't have to stay up all night running your simulations (or at least you shouldn't have to).

Observational astronomy gets lots of love, because it is so exciting to see the images of what is in space, but theory and computational astrophysics play a very important role in helping us understand what is out there.