A tank of water under a mountain
The Kamioka Observatory sits deep underground in the Mozumi mine near Hida, in Japan’s Gifu Prefecture, a location chosen for a specific reason: a detector built to catch faint neutrino interactions would be overwhelmed at the surface by the far more common cosmic ray particles constantly raining down, and only a thick layer of rock overhead filters those out enough to let the much rarer neutrino signals through. The original Kamiokande detector, built in 1982 and 1983, was a cylindrical tank about 16 metres tall and 15.6 metres across, holding roughly 3,000 tonnes of pure water and watched over by around 1,000 light-sensing photomultiplier tubes lining its walls.
Eleven flashes from a distant supernova
Detection relies on Cherenkov radiation: a neutrino occasionally strikes an electron or nucleus in the water hard enough to send it moving faster than light travels through that medium, producing a faint cone of light the surrounding tubes can register, and from the pattern and timing of that light physicists can reconstruct the direction, energy and type of the particle responsible. In its early operation, Kamiokande measured only about half the neutrino flux expected to be streaming from the Sun based on models of solar fusion, a persistent shortfall that puzzled physicists at the time and only later turned out to be an early hint of the same underlying phenomenon the detector would help confirm more directly two decades on.
Half the neutrinos anyone expected
In February 1987, the upgraded Kamiokande-II detector registered eleven neutrino events within a matter of seconds, arriving from a supernova roughly 160,000 light years away, the first ever detection of neutrinos from an astrophysical source beyond the solar system. That small handful of particles, caught by an instrument built for a different purpose, gave physicists their first direct observational confirmation that a distant, exploding star produces the burst of neutrinos theoretical models of core-collapse supernovae had predicted. The result was recognised in Masatoshi Koshiba’s share of the 2002 Nobel Prize in Physics, awarded jointly with Raymond Davis Jr. and Riccardo Giacconi.
Ten thousand tubes staring into water
Its successor, Super-Kamiokande, began taking data in 1996, scaling the same basic design up to a tank about 41.4 metres tall and 39.3 metres across, holding roughly 50,000 tonnes of ultrapure water and watched by about 11,200 photomultiplier tubes, more than ten times the instrumented capacity of the original detector. In 1998 it delivered a result with consequences reaching well beyond its own field: strong evidence that neutrinos produced in Earth’s atmosphere change type, or oscillate, as they travel, a phenomenon only possible if neutrinos carry a small but nonzero mass, contrary to the assumption of exactly zero mass built into the standard model of particle physics at the time.
Neutrinos have mass after all
The detector’s operation has not been free of setbacks. On 12 November 2001, a chain-reaction implosion destroyed roughly 6,600 of its photomultiplier tubes, more than half the total, forcing an extended, staged rebuild: the detector ran in a reduced configuration from 2001 to 2005 while protective shells were fitted to the surviving tubes, then had around 6,000 new tubes reinstalled by mid-2006, followed by a further major electronics upgrade that brought the detector to its current configuration by 2008. The episode is a reminder that even a discovery as significant as neutrino oscillation depended on physical hardware, thousands of individual glass tubes under water pressure, that could and did fail catastrophically, and that recovering from that failure took years rather than months.
An implosion and two Nobel Prizes
Together the two Kamioka detectors have now earned two separate Nobel Prizes in Physics: Koshiba’s in 2002 for the supernova detection, and Takaaki Kajita’s in 2015, shared with Arthur McDonald, specifically for the atmospheric neutrino oscillation discovery. This is worth an hour for how directly an instrument built essentially as a giant tank of water and light sensors, sunk under a mountain to escape the noise of the ordinary world, managed twice over to answer genuinely open questions about the universe, once by simply being in the right place when a star exploded, and once by patiently tracking a subtle statistical pattern in particles passing through Earth itself.