← history
Read stories →
May 20, 1883

Krakatoa

Krakatau's 1883 eruption begins at Perboewatan

Disasters
Krakatoa View full image ↗
Illustration generated by Ideogram 4 (fp8) (20 steps, 1216x832) on an NVIDIA RTX 5090 · 2026-08-27
How was this image made?

The creative brief sent to the image model. These are instructions, not a record of what actually happened.

Original prompt & settings (JSON) ↗

A German warship supplied the first warning: its captain saw an ash column over an island believed quiet for roughly two centuries.

Early morning, 20 May 1883

The first recorded eruptive activity of the 1883 episode came on 20 May at Perboewatan, the northernmost of the island's three aligned volcanic centres, in the Sunda Strait in what was then the Dutch East Indies.

Early that morning the captain of the German warship Elisabeth reported an ash-and-dust cloud rising above the previously uninhabited island — documenting its reawakening after roughly two centuries without a recorded eruption.

How high, and what kind

The plume's size is still argued. A contemporary observation put the cloud at about 11 km; a modern reconstruction of the column reaches about 20 km.

The style is contested too. One modern reconstruction characterizes the opening activity as Vulcanian to Sub-Plinian, while another peer-reviewed reconstruction describes the early eruptions more broadly as phreatic and phreatomagmatic.

Sightseers, a week later

Explosive activity and ash emission continued after 20 May but diminished after the 22nd. Visitors who landed on 27 May still watched explosions at intervals of about ten minutes.

The precise trigger was never observed directly; later petrological research suggests the early eruptions began evacuating a shallow, silicic melt-rich part of the magmatic system. The caldera collapse and the tsunamis were still more than three months away.

Sources

Researched 23 Aug 2026 6 sources 2 audit passes

How this was checked

Researched from the web into a fact sheet, rewritten from that sheet, then audited against it by a different model. How the pipeline works →

What the sources leave uncertain

  • The May 20 start date is well supported, but the eruption’s formal end is less certain; the Smithsonian catalogue gives October 21 with a question mark.
  • Published plume estimates differ substantially: an approximately 11 km contemporary observation and an approximately 20 km modern reconstruction.
  • The opening eruption style is described variously as Vulcanian–Sub-Plinian or phreatic/phreatomagmatic. “Strombolian” should not be presented as the uncontested classification.

Checked against