Sakurajima's Ash Has Reached Kumamoto Before — No Record Yet Shows It Stopping a Chip Fab

As recovery from the Kumamoto earthquake continues, some readers have started asking about a second concern: Sakurajima. The volcano sits in Kagoshima Prefecture, more than 100km south of Kumamoto's fab cluster in a straight line, and it remains active as of August 2026. Does the ash actually reach Kumamoto? Are the chip fabs there at risk? We answer both using only primary sources — the Japan Meteorological Agency (JMA) and Japan's Cabinet Office.

How active is Sakurajima right now?

As of August 7, 2026, JMA maintains Sakurajima at Eruption Warning Level 3 — a five-tier classification meaning entry is restricted near the crater because an eruption affecting that immediate area has occurred or is expected. The restricted zone extends roughly 2km from the Minamidake summit crater and the Showa crater, and JMA has not raised the level to 4, which would signal a risk extending toward residential areas.

Activity has nonetheless picked up somewhat. Sulfur dioxide (SO2, an irritant gas found in volcanic emissions) output roughly doubled, from 2,200 tons per day on July 13 to 4,600 tons per day on August 4, and an eruption in the early hours of August 1 sent a plume as high as 2,800 meters. Because the warning level can change, everything in this article reflects the situation as of August 7, 2026.

Has Sakurajima's ash ever actually reached Kumamoto?

The short answer is yes. JMA's Fukuoka District office keeps a historical record of Sakurajima's activity, and its entry for August 1995 reads: the volcano "erupted actively, causing transportation disruption, with ashfall reaching northern Kyushu" — a JMA regional designation that includes Kumamoto Prefecture. More direct records exist too. The same document notes that in 2019, ashfall was "confirmed in... part of Kumamoto Prefecture," and that an eruption on August 9, 2020 caused ashfall confirmed in "Miyazaki Prefecture and part of Kumamoto Prefecture" — two entries that name Kumamoto Prefecture explicitly.

What does the 1914 Taisho eruption mean for Kumamoto?

Among Sakurajima's eruptions, the 1914 Taisho eruption stands apart. According to JMA, its magma output ranks among the largest of any 20th-century eruption in Japan, close to the second-highest tier on the Volcanic Explosivity Index (VEI), a scale of eruption magnitude. Fine ash reportedly reached as far as Sendai — evidence that a large-enough eruption can carry ash well beyond Kyushu.

Most of that ash, however, did not travel toward Kumamoto. According to a report from Japan's Central Disaster Prevention Council (Cabinet Office), winds on the day of the eruption blew from the east, and the bulk of the ashfall landed on Sakurajima's eastern side, in the Osumi Peninsula's Kimotsuki district — the opposite direction from Kumamoto. In other words, the Taisho eruption's main ashfall axis pointed away from Kumamoto, so citing the Sendai record alone as evidence for heavy ashfall reaching Kumamoto would be inaccurate. Separately, the eruption's death toll of 58 (and 112 injured) combines casualties from the eruption itself and a magnitude-7.1 earthquake that struck roughly eight hours later.

If ash ever did stop a fab, what would actually cause it?

What the record so far shows is: ash reaches Kumamoto, but neither its quantity nor any resulting harm has been documented. So if damage to a fab were ever to occur, what mechanism would actually be behind it? What we could confirm in this reporting points to external infrastructure, not the inside of a factory.

The best-documented mechanism is a power outage triggered by insulators — ceramic or glass components that electrically isolate power lines from towers and poles. Dry volcanic ash is essentially non-conductive, but when it gets wet from rain or mist, soluble salts in the ash dissolve into the moisture and sharply raise its conductivity. Japan's domestic disaster-planning literature cites cases where roughly 0.3cm of ash, combined with rain, degraded insulator performance enough to cause outages: a 1979 Sakurajima eruption left about 2mm of wet ash and cut power to 1,600 households on the island for six hours, and a 1990 eruption of Mount Aso caused outages to roughly 3,700 households in an area with about 1mm of ash.

Water-treatment risk is similarly conditional. Cabinet Office documentation assumes that slow-sand-filtration plants (an older method that filters water gradually through a sand layer) stop functioning once ashfall reaches roughly 1cm, as their filter beds clog. But the same documentation explicitly excludes plants with covered intakes or those drawing from groundwater from that risk model, and Kagoshima City has reported no supply stoppages since it installed covers over its filtration ponds. Whether Kumamoto's chip fabs draw their ultrapure water (highly purified industrial water used in wafer-cleaning steps) from a source that would fall under this exclusion is something we could not confirm in this reporting. For how ultrapure water itself works, see what ultrapure water is (in Japanese).

Is a major eruption imminent, and how does that differ from the official view?

We don't want to overstate the risk, but the expert view is worth including. Kyoto University volcanologist Masato Iguchi told a local newspaper in January 2024 that "Sakurajima is prepared to produce an eruption on the scale of the Taisho eruption" — based, in essence, on ground-deformation data suggesting the magma volume lost from underground during the 1914 eruption, roughly 110 years ago, has largely been replenished.

That is, however, one expert's personal assessment as of January 2024. As of August 2026, we found no record of JMA or Japan's Coordinating Committee for Prediction of Volcanic Eruption issuing an official assessment that a large-scale eruption is imminent. The expert's warning and the official government assessment should always be read as two separate things. For the broader risk landscape around Kumamoto's semiconductor cluster, see our coverage of the Kumamoto earthquake's impact on chipmakers and the Day 11 disaster-designation update.

What are the fabs themselves doing about it?

Finally, an honest gap to flag. What kind of volcanic-ash disclosures or countermeasures JASM (the joint venture that operates TSMC's Kumamoto fab), Renesas, or other Kumamoto-based chipmakers have in place is something we could not confirm with primary sources in this reporting — nor could we confirm how clean-room air intake systems are managed during ashfall, or exactly where each fab's ultrapure-water source draws from. Rather than write around that gap as if it were resolved, we're stating it plainly: this is what we don't know, and it's why we left it out of the rest of this article.