This is Part 3 of the Neutrino History series. In Part 2 we left the field at the start of the solar neutrino problem and the discovery of the muon neutrino. In Part 3 we watch the 20-year buildup that would eventually lead to the 1998 oscillation discovery.
1970–1985: The solar problem entrenches
Through the 1970s and into the 1980s, Raymond Davis Jr. continued running his chlorine experiment at the Homestake gold mine. Year after year, the count of argon-37 atoms produced by solar neutrinos came in lower than John Bahcall’s standard solar model predicted. The deficit was real, persistent, and statistically significant. Most physicists assumed that either the solar model was wrong or the neutrinos were doing something unexpected.
For decades, the field couldn’t tell which. Davis’s chlorine detector was only sensitive to electron neutrinos. If solar neutrinos were oscillating into other flavors during their 8-minute journey to Earth, the detector would miss them entirely.
Bahcall, meanwhile, was patiently defending the standard solar model. He computed the solar neutrino flux to ever higher precision, with detailed uncertainty budgets, and argued that the model agreed with every other observable property of the Sun (surface composition, helioseismic oscillations). If the model was wrong about something, it would have to be something quite specific to neutrino production.
By the mid-1980s, neither view had won. The field was waiting for new experiments.
1980s: New solar experiments come online
Several new experiments joined Homestake through the 1980s, using different techniques and sensitive to different parts of the solar neutrino spectrum.
Kamiokande in Japan began operating in 1983 — initially built to search for proton decay but quickly repurposed for solar neutrinos. As a water Cherenkov detector, it was sensitive only to the highest-energy ⁸B solar neutrinos. It confirmed Davis’s deficit at about half the standard solar model rate.
GALLEX and SAGE — Italian-led and Soviet-led gallium experiments — began operating in 1990 and 1990 respectively. Using a different interaction (chlorine had a high energy threshold; gallium worked down to ~233 keV), they could catch the dominant pp solar neutrinos that Davis couldn’t see. Both reported deficits, but smaller than Davis’s — about 60% of the standard model prediction.
The energy-dependent pattern was suggestive. Different detectors with different energy thresholds saw different fractions of the predicted flux. That pattern was hard to explain with a single solar-model error but fit neatly with matter-modified oscillation (the MSW effect, proposed in 1985-86).
Still, no single experiment had decisive evidence.
Mid-1980s: The atmospheric anomaly
In parallel with the solar story, a different anomaly was emerging in atmospheric neutrinos. Cosmic rays hit Earth’s upper atmosphere and produce showers that include muon and electron neutrinos in a calculable ratio (roughly 2:1 in favor of muon-flavor, from the chain π → μ → e plus neutrinos).
Several underground experiments measured the actual ratio. Kamiokande, IMB in Ohio, and later Soudan 2 in Minnesota and MACRO at Gran Sasso all observed a deficit of muon-flavored events relative to expectations. The deficit appeared specifically in upward-going neutrinos that had passed through the Earth — exactly the geometry where flavor oscillation over long baselines would produce a signal.
By the late 1980s, the atmospheric anomaly was real but contested. The flux models had substantial uncertainties. The deficit was at the 3σ-4σ level in any single experiment, not 5σ. Multiple explanations were on the table. The community was leaning toward oscillation but couldn’t yet make the definitive case.
February 23, 1987: The supernova
At 07:35:35 UT on February 23, 1987, three underground detectors recorded a sudden burst of neutrino events.
Kamiokande-II in Japan caught 11 events within 13 seconds. IMB in the Ohio salt mine caught 8 events within 6 seconds. Baksan in the Russian Caucasus caught 5 more. All within 23 seconds of each other.
Three hours later, the astronomer Ian Shelton at Las Campanas Observatory in Chile noticed a new bright star in a photograph of the Large Magellanic Cloud. SN 1987A — a Type II supernova in the LMC, 168,000 light-years away.
The 24 neutrino events confirmed long-standing theoretical predictions about core-collapse supernovae:
- About 99% of the released energy went into neutrinos.
- The neutrinos escaped the collapsing core hours before the visible light, which had to diffuse out through the stellar envelope.
- The energy spectrum (10-40 MeV) and burst duration (~10 seconds) matched core-collapse models.
- The combined timing spread set the first laboratory limit on neutrino mass at around 10 eV.
The detection was the founding event of supernova neutrino astronomy. The 2002 Nobel Prize would later recognize Masatoshi Koshiba (Kamiokande’s leader) for the result.
It was — and remains — the only supernova whose neutrinos have been directly detected. The next one will likely come from our own galaxy, and modern detectors will catch tens of thousands of events from it.
Late 1980s: Super-Kamiokande planned
By the late 1980s, the Japanese community led by Masatoshi Koshiba and his student Takaaki Kajita was already planning Kamiokande’s successor: an even larger water Cherenkov detector in the same Kamioka mine, 8 times the size, capable of catching enough atmospheric neutrinos to make the oscillation case definitively.
Super-Kamiokande would begin operations in 1996. Its 1998 announcement of atmospheric neutrino oscillation would be one of the foundational moments of modern physics.
1990: The state of the field
By the start of the 1990s, the neutrino sector looked like this:
- Three flavors confirmed (the tau neutrino itself directly observed only in 2000 by DONUT, but inferred theoretically).
- The neutrino exists and interacts via the weak force as Fermi’s theory predicted.
- Solar neutrino problem unresolved for 22 years, with multiple experiments seeing energy-dependent deficits.
- Atmospheric neutrino anomaly observed at multiple sites, suggestive of oscillation but not yet definitive.
- SN 1987A had been caught, founding supernova neutrino astronomy.
- Pontecorvo’s oscillation hypothesis still untested but increasingly the favored explanation for both anomalies.
The community was ready for the experimental campaign that would resolve everything. Super-K was being built. SNO was being designed. Reactor experiments like KamLAND were in early planning. By the late 1990s, the oscillation picture would be definitive.
In Part 4 we’ll watch that campaign produce two foundational discoveries — Super-Kamiokande’s 1998 atmospheric oscillation, and SNO’s 2001 solar oscillation — that earned the 2015 Nobel Prize.
Frequently asked
Why did the solar neutrino problem persist so long?
Because distinguishing the two possible explanations — wrong solar model versus neutrino oscillation — required an experiment that could measure both the electron-neutrino flux and the total all-flavor flux of solar neutrinos. SNO's heavy-water target was the first to provide this capability. Building SNO took until the late 1990s. The puzzle waited 30 years for the experiment that could solve it definitively.
What was the atmospheric neutrino anomaly?
Through the 1980s, multiple underground detectors observed that the ratio of muon-flavored to electron-flavored atmospheric neutrinos was different from theoretical predictions. The deficit appeared specifically in upward-going neutrinos that had traveled through the Earth. The pattern was consistent with neutrino oscillation but no single experiment had the statistics to make a definitive claim.
Why was SN 1987A so important?
Because it was — and remains — the only supernova whose neutrinos have been directly detected. The 24 events caught within 23 seconds across three detectors confirmed that core-collapse supernovae release ~99% of their energy as neutrinos, set the first laboratory limit on neutrino mass from timing, and validated decades of supernova theory. It also launched multi-messenger astronomy as a discipline.
Which detectors observed SN 1987A?
Three: Kamiokande in Japan (11 events), IMB in Ohio (8 events), and Baksan in the Russian Caucasus (5 events). All three caught the burst within 23 seconds of each other on February 23, 1987. The supernova's optical light arrived about three hours later — the neutrinos had escaped the collapsing core promptly while the photons had to diffuse out through the stellar envelope.
Cite this article 5 formats
APA
Neutrino Times Editorial Team. (2026, February 2). Neutrino History — Part 3: The solar problem builds, atmospheric anomalies, SN 1987A (1970–1990). Neutrino Times. https://neutrino-times.com/articles/neutrino-history-part-3-1970-1990/
Chicago
Neutrino Times Editorial Team. "Neutrino History — Part 3: The solar problem builds, atmospheric anomalies, SN 1987A (1970–1990)." Neutrino Times, February 2, 2026. https://neutrino-times.com/articles/neutrino-history-part-3-1970-1990/.
MLA
Neutrino Times Editorial Team. "Neutrino History — Part 3: The solar problem builds, atmospheric anomalies, SN 1987A (1970–1990)." Neutrino Times, 2 Feb. 2026, https://neutrino-times.com/articles/neutrino-history-part-3-1970-1990/.
BibTeX
@misc{neutrino-times-neutrino-history-part-3-1970-1990,
author = {Neutrino Times Editorial Team},
title = {Neutrino History — Part 3: The solar problem builds, atmospheric anomalies, SN 1987A (1970–1990)},
howpublished = {Neutrino Times},
year = {2026},
month = {feb},
url = {https://neutrino-times.com/articles/neutrino-history-part-3-1970-1990/},
note = {Accessed: 2026-02-02}
} RIS
TY - GEN TI - Neutrino History — Part 3: The solar problem builds, atmospheric anomalies, SN 1987A (1970–1990) AU - Neutrino Times Editorial Team PY - 2026 DA - 2026-02-02 PB - Neutrino Times UR - https://neutrino-times.com/articles/neutrino-history-part-3-1970-1990/ ER -