Frequently asked questions

Why is KM3NeT at the bottom of the sea and not on land?

We use water for three purposes:

1. To shield our detector from particles coming from space, so-called cosmic rays. On the mainland, we would receive too many particles in our detector, making the science analysis harder.
2. We need a medium denser than air, in order to be able to see the Cherenkov radiation. This radiation is our main probe with which to look for neutrinos, and in air it is more difficult to see it.
3. The Cherenkov radiation mentioned above is really (really!) faint. We can see it better when there are no other sources of light, such as the sun or the moon. The bottom of the sea is an ideal place because it is pitch black there, 24/24 and 7/7!

---

I want to know more about the technical details of the pictures in the workflows. Where can I find useful information?

Right here or in the Field Guide (in the middle of the right side of your screen)! Let's go through all the plots in detail.

First plot:

The first plot you see for each sample is composed of two parts:

The top one is the amplitude of the sound wave as a function of time, in blue. If you look at the left corners, you have the range of the plot around the middle line, which is 0.

The data is displayed on a symlog scale, which means that you have a logarithmic scale above and below zero: each thick horizontal line corresponds to a new power of ten (0.0001, 0.001, etc), and each thinner lines correspond to the units between two power of ten (0.0001, 0.0002, etc). The red curve corresponds to the envelope of the variations. Basically, it peaks where there is a signal of interest.

The bottom one corresponds to the spectrum as a function of time, ranging from 4kHz to 98kHz (vertical axis). The red line corresponds to the limit of the human-audible range. Basically, the yellower a spot is in it, the more intense is the corresponding frequency.

Keep in mind that these two plots share the horizontal axis, so what you see at a particular position in the bottom plot corresponds to what is directly above in the top plot.

Second plot:

The second plot you have is the power spectrum of the click (blue curve), which represents the intensity as a function of the frequency, from 0Hz to 98 kHz. If there is something interesting, it should stand above the red line (noise). In this example, you can see that from 0Hz to around 30kHz, there is a huge bump above the mean noise. There is something of interest here! The green curve corresponds to the blue one minus the red one.

To understand which species it corresponds to, look mainly at the frequencies (horizontal axis) of the bump(s). The recorded animal can be far away, so the amplitude (vertical axis) can be small above the red line. This is why amplitude may be less interesting than the frequencies.

Third plot:

The third plot is a zoom of the first on the center of the window. This can help you see features for shorter signals for example.

Fourth plot:

The fourth plot is the spectrum of the zoomed wave form (third plot). Some features may appear more clearly here because we focus on a shorter time window, and so less noise should be in it.

---

What is a neutrino?

The neutrino is one of the most mysterious of the known elementary particles. Neutrinos are extremely difficult to catch. They are produced in different ways, all of which involve the so-called weak (nuclear) force, the fundamental interaction responsible for the radioactive decay of atoms and other nuclear reactions.

---

Why can’t a machine do this work instead?

They can, but not as precisely as human beings. While computers might outshine us in analysing very large datasets, human eyes and ears are still better than a computer at noticing subtleties. We need (a lot of!) human beings to teach computers to notice these subtleties more efficiently. This is why we need you, and also because we are just not enough in the research world to do it ourselves!

---

What if I do the tasks wrong?

You can't do it wrong (unless you do the tasks randomly!). Why? Because the most important is to do it carefully. Each subject will be classified based on your understanding of the question asked and its context, and this is what is important.

Of course, there is a right answer, but we don't know it! To try to get it, your answer will be averaged with the ones of other volunteers. The statistical properties of this average will tell us something about the difficulty of the task, and hence about the data themselves.