If you want to measure very small changes, it might seem obvious that you need a very sensitive instrument... but what do we really mean by sensitive? Well, we don't mean that it should make you a cup of cocoa if you're feeling sad, that's for sure! When we talk about the sensitivity of an instrument, we mean the extent to which we can see the signal we want to measure. So for gravitational wave detectors, the sensitivity is a measure of how much of the signal we see is actually from gravitational waves. To get the best sensitivity, we should make the strength of the signal we want to measure as big as possible, and make the strength of the unwanted signals as small as possible. We call the unwanted signals noise.

Let's start with the first bit; maximising the gravitational wave signal. The first way we can do this is by making a very big detector. The interferometers aren't up to 4km long just for fun; having a very long arm length increases the actual length change of the arm under the influence of a gravitational wave, which gives us a larger signal coming out of the detector.

So now that we have made our detector as sensitive to gravitational waves as possible, why isn't this enough on its own? Well, the gravitational wave signals coming from space are still so small that many other signals can get in the way. Imagine that you are sitting on one side of an empty park, and your friend is sitting on the other side. If your friend calls your name, you can probably hear it easily, because there aren't too many other sounds getting in the way. But if you were sitting the same distance apart in the middle of a busy city centre, would it be so easy to hear them? I doubt it! That's because all the other sound signals are getting in the way, and making it hard to pick out the signal you want to hear. This is what happens with gravitational wave detectors; they are in fact such sensitive instruments that they can even pick up tiny noises like the vibrations of atoms on the surface of mirrors, and these can obscure the gravitational wave signal!

Above is a sketch of a sensitivity curve: each line represents a different noise source, and the aim of building a good interferometer is to lower each one as much as possible. So the first thing to do is to try to put your detector in a quiet place, i.e. to choose the park over the city centre. However, there are many other ways in which you can try to make your detector measure less noise.