After each of six climate-altering eruptions, from Krakatau in 1883 to Pinatubo in 1991, extreme cold in the ocean known as a marine cold spell briefly grew stronger across the global ocean, once natural variability and long-term trends were removed and conditions compared with those before each eruption. Researchers at South Westphalia University of Applied Sciences and the University of Cologne in Germany analysed daily sea surface temperatures for 1850–2024 and published the study in Ocean-Land-Atmosphere Research on 6 November 2026. Extreme ocean cold has generally declined over the past 175 years, but the team thinks major eruptions can temporarily interrupt that trend.
What is a marine cold spell?
A marine cold spell (MCS) is a period when sea surface temperature stays unusually cold, compared with the normal seasonal conditions at that location, for at least five days in a row. This article uses the English term throughout.
The press release lists four factors that shape how strong and how frequent these events are:
- the local seasonal climate and its variability
- how persistent the cold is
- regional ocean currents
- the distribution of sea surface temperature
The team suspected that volcanic eruptions might add an outside influence to this list.
Cold periods after large eruptions are called volcanic winters, and history records several. The year 1816, after the major 1815 eruption of Mount Tambora in Indonesia, became "The Year Without a Summer", and the 1991 eruption of Pinatubo lowered global temperatures by about 0.5°C. Sulfur-containing compounds blasted into the atmosphere form sulfate aerosols, a thin veil around the Earth that reflects part of the sunlight and cools the surface.
What did 175 years of sea temperatures let the team compare?
The team used COBE-SST3, a dataset that reconstructs daily global upper-ocean temperatures for 1850–2024. With it, they examined how the strength of marine cold spells after six climate-altering eruptions compared with conditions before each eruption.
| Volcano | Year of eruption |
|---|---|
| Krakatau | 1883 |
| Santa Maria | 1902 |
| Novarupta | 1912 |
| Agung | 1963 |
| El Chichón | 1982 |
| Pinatubo | 1991 |
According to the abstract, the team removed part of the natural climate variability, such as the El Niño–Southern Oscillation (ENSO), along with long-term temperature trends, and then compared each post-eruption period with pre-eruption conditions defined for that eruption. Strength was measured as "cumulative intensity", but the abstract and press release we could read do not give its exact definition.
Raven Quilestino-Olario, a doctoral researcher at the Institute of Geology and Mineralogy, University of Cologne, and a co-author, explained what motivated the work:
"We noticed that while the cooling effects of major volcanic eruptions have been widely studied, relatively little attention has been given to how these volcanic winters affect extreme cold events in the ocean."
— Raven Quilestino-Olario (University of Cologne, co-author)
How much did ocean cold change after the eruptions?
Across most of the global ocean, both the cumulative intensity and the frequency of marine cold spells have generally declined since 1850. Against that background, global cumulative intensity was higher than before the eruption for a short time after each of the six eruptions. The abstract says this suggests that volcanic forcing can temporarily interrupt the long-term weakening.
Jan Kempert of South Westphalia University of Applied Sciences in Hagen, the first author, summed up the result:
"Our study shows that although cold extremes in the ocean have generally become less frequent and less intense over the past 175 years, major volcanic eruptions may temporarily interrupt this long-term pattern."
— Jan Kempert (South Westphalia University of Applied Sciences, first author)
Neither the abstract nor the press release gives figures for how much stronger the cold spells became or how many years the effect lasted. We have not been able to check the full paper, so this article cannot say more than "briefly stronger".
Why did different ocean regions respond so differently?
Even though the global signal strengthened, the effect of the eruptions was not uniform across the world's oceans. According to the press release, some regions saw strong cold extremes while others showed weaker or even opposite responses. As the abstract notes, the response was strongest in some midlatitude bands, and the volcanic imprint remained even when the tropical Pacific was left out of the calculation.
The abstract and press release do not explain why the regions differ so much. The factors the press release lists for marine cold spells, such as currents and seasonal climate, also differ between regions, but how they relate to the varying volcanic effect is not clear from what we could read.
For the 1991 Pinatubo eruption, the team also compared the long-term 175-year reconstruction with a separate dataset from the satellite era and found close agreement.
The data were developed by Japan Meteorological Agency researchers
The press release calls COBE-SST3 a "newly available dataset" but does not say who made it. COBE-SST3 is a daily sea surface temperature analysis published in 2025 in the Journal of the Meteorological Society of Japan by Masayoshi Ishii of the Meteorological Research Institute of the Japan Meteorological Agency (JMA) and other JMA researchers. The paper's title includes the words "high-resolution" and "daily".
COBE is the name of the sea surface temperature and marine meteorological data that JMA has developed, and its sea surface temperature part is called COBE-SST. According to JMA's website, COBE covers 1891 to the present on a 1-degree grid. It was built from sources including ICOADS, compiled by the US National Oceanic and Atmospheric Administration (NOAA), and the "Kobe Collection". The Kobe Collection consists of historical marine meteorological data gathered by the Kobe Marine Observatory (now the Kobe Local Meteorological Office) and digitised by JMA and the Japan Weather Association.
| Dataset | Features (source) |
|---|---|
| COBE-SST | 1891 to present, 1-degree grid (JMA website) |
| COBE-SST2 | Improved version of COBE-SST (2014 paper); used by JMA since February 2023 to assess long-term global mean change (same) |
| COBE-SST3 | Daily; 1850–2024 analysed in this study (press release, abstract) |
So COBE-SST3, in the line of COBE-SST datasets developed by JMA researchers, underpinned a German team's study of volcanoes and the ocean. The team hopes to use the same data to explore how marine heatwaves have changed over the past 175 years, with greater emphasis on the uncertainties in historical observations.
Among studies of ocean climate and circulation, we have also covered a study of the dissolved organic matter carried by water sinking into the Antarctic deep sea.