The IceCube Observatory at the geographical South Pole is the main instrument I use to study neutrinos and cosmic rays. It is comprised of two sub-detectors: IceCube and IceTop. The former consists of a cubic kilometer of antarctic ice instrumented with more than 5000 optical modules at depths ranging from 1.5 to 2.5 kilometers from the surface of the ice. The later is an array of 160 Ice Cherenkov detectors located on the surface of the ice, right on top of IceCube.
I study high energy cosmic rays. These are sub-atomic particles that come from space and their energy can be a million times the energies achieved at the largest colliders, such as the LHC at CERN. In every-day terms: the energy can be close to the energy of a ball traveling at a speed of about 100 km/h. All concentrated in one subatomic particle! I have done that for several years as member of international collaborations such as IceCube and Pierre Auger.
I have been interested in the composition of cosmic rays since I worked in the Pierre Auger collaboration. The composition of the very high energy cosmic ray flux is extremely difficult to determine since we can not detect the cosmic rays directly but by the air shower they induce. We develop statistical methods to infer the composition based on observable characteristics of air showers.
I am particularly interested in the use of cosmic ray air shower arrays for the detection of astrophysical neutrinos. The main background in the search for astrophysical neutrinos is the flux of atmospheric muons and neutrinos. The identification of astrophysical neutrinos is done by tagging the atmospheric neutrinos/muons through the detection of the accompanying air shower.
We rely on simulations of the air shower process in order to study cosmic ray composition and to estimate the brackground in the searches for astrophysical neutrinos. Our lack of knowledge of interactions at the highest energies can cause systematic deviations that can affect our measurements. In order to reduce these systematic effects, we develop methods to cross-check the simulations with data.