One use for IceCube is to study solar flares, or, more generally, solar weather. Solar weather is important; solar flares are often accompanied by coronal mass ejections (CME), the ejection of plasma which sometimes hit the Earth. This plasma can disrupt radio communications, disable satellites, and endanger astronauts. In 1859, a large CME, the Carrington event, disrupted telegraph communications in the U. S. and Europe, and produced enormous auroras which were visible over much of the globe. Modern electronics is far more susceptible to damage than simple telegraph systems. A similar event occurring today would cause incredible ($1 trillion?) damage. With appropriate warning, we could reduce the damage significantly by turning off and/or shielding as much electronics as possible, grounding airplanes, etc. For this very practical reason, it is important to understand solar weather in more detail.
In addition, sunspots and solar flares contribute to the variation in the Sun's total output. Sunspots and solar flares reduce and increase the solar irradiance directly. So, the lack of sunspots in the current 11-year solar cycle (cycle 24) might be thought to reduce solar irradiance, and hence help combat global warming. However, the story is a bit more complicated than that. Both sunspots and solar flares are governed by the sun's magnetic fields; there is a more direct correlation between the Sun's magnetic activity and irradiance, as can be seen from this plot from the Bartol Institute at the University of Delaware, which compares solar magnetic activity and the rate of neutrons from cosmic-rays reaching Earth. Before ~1950 (i. e. before global warming became significant), these magnetic field variations likely accounted for much of the observed climate variation. The evidence for this comes from comparing carbon-14 dating curves (carbon-14 measures cosmic-ray activity) with climatic data from tree rings and ice cores.
Most of the particles emitted in a solar flare or CME are, by IceCube standards, low energy, protons, neutrons (which mostly decay before reaching Earth) and photons with energies of at most a few billion electron volts (GeV). When the individual particles reach Earth, they leave relatively few direct traces at ground level; only a small fraction of them lead to (via a small air shower) particles reaching the Earths surface. However, a CME contains a very large number of particles, an, if the density is high enough, it can raise the counting rate of terrestrial particle detectors This is known as a ground level event (GLE). Space scientists have deployed neutron detectors at many sites around the globe to monitor these signals. Because of the total sensitive area and low background rates of the 162 IceTop surface array tanks, it is a very sensitive detector for GLE's, and we had expected to detect of order 1 GLE per year.
Unfortunately, Nature has not cooperated. At this week's International Cosmic-Ray Conference, IceCube presented an analysis of GLE's from 2011 to 2016. Only three GLEs were observed, and they were 'quite small by historical standards.' The reason for the low rate is unknown, but it may be connected with the paucity of sunspots during the current solar cycle. This might be expected to lead to a reduction in solar irradiance. Data from the SORCE satellite presents a mixed picture, showing a small (~0.07%) increase in solar irradiance from 2009 and 2015, with a similar sized decrease over the past two years.
We do not understand what is causing these changes; clearly the Sun still has many secrets.
Many thanks to Paul Evenson (Bartol Institute, Delaware) for useful discussions on the relationship between sunspots, flares, CME and magnetic fields. Of course, any errors here are my own.