Faultory

Case file 72 · Environment & history

A gas cloud from still water

Cameroon, 1986: where did the carbon dioxide come from, and what released it?

Subject
Lake Nyos gas disaster (1986)
Reading
13 min
Evidence
5 exhibits
Reasoning
3 steps
Level
Intermediate
Pipe and vent structure on the surface of Lake Nyos.
Photographed 2006 An artificial carbon dioxide vent in Lake Nyos, photographed in 2006 after engineers began venting the gas. Photo: Bill Evans · Public domain · Wikimedia Commons.
Illustrative signature. An illustrative strip-chart of dissolved gas at depth shows a slow build-up over years, then a single sudden drop (a sketch, not data). Drawn to explain the case, not plotted from source data.

01 · The briefing

Begin with the question.

Lake Nyos is a crater lake in the volcanic highlands of northwest Cameroon. On 21 August 1986 a sudden release of gas from its deep water sent a dense cloud of carbon dioxide into the surrounding valleys. The people killed lived in nearby towns and villages, and thousands of livestock also died. The 1988 review calls such crater lakes a hitherto unrecognized natural hazard. Survivors were examined and interviewed by a medical team, and the investigators who followed had to reconstruct the event from chemistry, geology and medical records.

Two explanations were under debate from the start. One was volcanic: magma might have released gas directly into the lake. The other was a limnic eruption, in which carbon dioxide that had built up dissolved in deep water comes out of solution and erupts from the lake. A smaller release at Lake Monoun in 1984, which killed 37 people, raised the question of whether the two events shared a cause.

This investigation separates these readings. It asks what the chemistry, the medical evidence and the later measurements actually show, and which parts of the account remain open.

Satellite view of a small lake in hilly terrain.
Where it happened18 December 2014 A 2014 Landsat 8 satellite view of Lake Nyos, taken long after the 1986 release. Photo: NASA · Public domain · Wikimedia Commons.

The timeline

  1. A gas burst from Lake Monoun, a smaller crater lake in Cameroon, kills 37 people.

  2. A sudden release of carbon dioxide from Lake Nyos kills at least 1,700 people in nearby settlements, and thousands of livestock also die.

  3. Kling and colleagues review the chemical, isotopic, geologic and medical evidence and find that it supports four hypotheses: the gas was mostly carbon dioxide stored in the deep water, the victims died of carbon dioxide asphyxiation, the carbon dioxide was magmatic in origin, and there was no significant direct volcanic activity.

  4. Sigurdsson argues that the deep-water geochemistry does not support a sudden injection of volcanic gas, and proposes a gradual build-up of dissolved gas instead.

  5. A volcanic lakes workshop report in Eos notes that after the bottom water recharged with carbon dioxide, the limnic eruption hypothesis became the most likely explanation.

A researched reconstruction for learning. These reasoning steps simplify a complex record; no score can capture the full history.

02 · The evidence

What does the record show?

Top of a vertical pipe and a raft at a lake.
On the evidence tablePhotographed 2001 A pipe top at Lake Nyos, part of the gas-venting system that engineers installed after the 1986 release. Photo: Bill Evans · Public domain · Wikimedia Commons.

Read each exhibit before settling on an explanation. The source links below every exhibit lead to the record behind it.

Exhibit 01

The 1987 review of the evidence

The 1987 review by Kling and colleagues, indexed on PubMed, states that the release on 21 August 1986 caused the deaths of at least 1,700 people. It lists four hypotheses that the chemical, isotopic, geologic and medical evidence support: that the gas was mostly carbon dioxide stored in the hypolimnion, that victims died of carbon dioxide asphyxiation, that the carbon dioxide was magmatic in origin, and that there was no significant direct volcanic activity. This establishes the authors' synthesis at that date. It does not establish what triggered the release.

Source record s1eutils.ncbi.nlm.nih.gov

Exhibit 02

Sigurdsson's rival mechanism

Sigurdsson's 1988 review, indexed on PubMed, gives 1,746 deaths at Lake Nyos and 37 at Lake Monoun. It accepts a magmatic carbon source from isotope data, but argues that the geochemistry of deep water and gases does not support a sudden injection of volcanic gas. It proposes a gradual build-up of dissolved bicarbonate, with steady input from the mantle, and says that asphyxia from carbon dioxide caused almost all of the lethal effects. This is a proposed mechanism inferred from chemistry, not an observed event.

Source record s2eutils.ncbi.nlm.nih.gov

Exhibit 03

Cloud volume and the range of deaths

Wagner and colleagues, in a 1988 medical evaluation indexed on PubMed, describe a carbon dioxide cloud with an estimated volume of about 1 cubic kilometer that caused 1,700 to 2,000 human fatalities and killed thousands of livestock and wild animals. The range shows that early counts were uncertain. The paper establishes the scale of the cloud estimate and the death range. It does not establish how the gas was released, and the volume is an estimate.

Source record s3eutils.ncbi.nlm.nih.gov

Exhibit 04

A regional summary of the two releases

An archived USGS page, carrying a Smithsonian Global Volcanism Program summary, dates the Lake Monoun release to August 1984 and the Lake Nyos release to August 1986. It says that the two releases together resulted in the loss of nearly 1,800 lives. This establishes the regional pattern and a rounded combined toll. It does not separate the two tolls, and it gives only month-level dates.

Source record s4web.archive.org

Exhibit 05

The later recharge observation

A 2016 Eos report from a volcanic lakes workshop in Cameroon, by Rouwet, Tanyileke and colleagues, gives the 21 August 1986 cloud as killing 1,746 people and more than 3,000 livestock. It says that early disagreement was between volcanic and limnic eruption, and that after the bottom water naturally recharged with carbon dioxide, the limnic eruption hypothesis was the most likely. It establishes the later consensus and the recharge observation. It does not give the measurements behind the recharge or the trigger.

Source record s5eos.org

03 · Your reasoning

Follow the strongest explanation.

Choose an interpretation, then compare it with the evidence. You can revise your choice. The goal is understanding, not a perfect score.

Reasoning step 01 of 03

What was the most likely source of the gas that killed people at Lake Nyos?

Evidence-based explanation

The reviews agree that carbon dioxide was the dominant gas and that it had been stored in the deep water. The disagreement is over how it got there and what triggered the release. Sigurdsson argues for gradual build-up and against a sudden volcanic injection, and the 2016 Eos report says the limnic eruption hypothesis became the most likely after the lake recharged. The trigger was not directly observed, so this remains a best inference.

Weigh it against

Exhibit 01The 1987 review of the evidenceExhibit 02Sigurdsson's rival mechanismExhibit 05The later recharge observation

Reasoning step 02 of 03

Which finding best supports the claim that carbon dioxide caused the deaths, rather than heat or another hazard?

Evidence-based explanation

The medical and chemical evidence points to carbon dioxide asphyxia. Sigurdsson reports that the lethal effects are almost entirely due to asphyxia, and the 1987 review lists death by carbon dioxide asphyxiation among its supported hypotheses. The evidence is strong on the mechanism of death, but it is based on the victims and survivors the medical teams examined, so it describes the typical course rather than every case.

Weigh it against

Exhibit 01The 1987 review of the evidenceExhibit 02Sigurdsson's rival mechanismExhibit 03Cloud volume and the range of deaths

Reasoning step 03 of 03

What is the most defensible lesson for other crater lakes?

Evidence-based explanation

The 1987 review argues that hazardous lakes may be identified and monitored, which reduces the danger of future incidents. The Eos report describes pipes that siphon deep water to the surface so that carbon dioxide escapes in safe amounts. The lesson applies to lakes with similar deep-water chemistry, and it does not assume that every volcanic lake behaves in the same way.

Weigh it against

Exhibit 01The 1987 review of the evidenceExhibit 05The later recharge observation

04 · The findings

The chain behind the outcome.

The best-supported account is that Lake Nyos had carbon dioxide dissolved in bottom waters that did not mix with the surface. On 21 August 1986 a large volume of that gas was released suddenly, forming a cloud that killed at least 1,700 people and thousands of livestock in nearby settlements. The reviews attribute the deaths mainly to carbon dioxide asphyxia.

The source of the gas is less settled. The 1987 review of Kling and colleagues supported a magmatic origin for the carbon dioxide and found no significant direct volcanic activity. Sigurdsson's 1988 review accepted a magmatic carbon source too, but argued that the geochemistry does not support a sudden injection of volcanic gas, and proposed slow build-up of dissolved gas sustained by input from the mantle. The 2016 Eos report, from a volcanic lakes workshop, concluded that the limnic eruption hypothesis was the most likely explanation after the lake's bottom water recharged with carbon dioxide.

The death toll appears as 1,700 to 2,000 in some sources and as 1,746 in others, and the USGS summary gives nearly 1,800 for Monoun and Nyos together. The earlier Monoun release, with 37 deaths, shows that the hazard was not unique to one lake. The engineering response, which vents deep-water gas through pipes, followed the understanding that the gas could build up again.

Contributing causes and conditions

  1. Carbon dioxide from deep sources accumulates dissolved in the stagnant bottom water of the lake, where the water is largely isolated from the surface.
  2. The deep water becomes close to saturated, so the gas stays under pressure rather than escaping gradually.
  3. A disturbance whose exact nature was not directly observed releases a large volume of gas at once, forming a cloud.
  4. The cloud moves into the valleys around the lake, where people and livestock breathe it in towns and villages.
  5. Carbon dioxide asphyxia accounts for almost all of the lethal effects, according to Sigurdsson's review.

OutcomeLake Nyos gas disaster (1986)

Take the method with you

  1. Two explanations can share a key fact, so separate the disputed link, here the source of the gas, from the agreed one, here asphyxia.
  2. Early death tolls vary across sources, so report the range and say which source gives which number.
  3. A later observation, such as the lake's recharge, can settle a question that early evidence left open.

You can read findings at any time. Exploring every reasoning step marks this case as complete in your notebook.

Open another investigation

The open record

Sources & further reading

Read the original record and its scope before relying on a conclusion. Source references support the exhibit summaries; Faultory’s reasoning prompts are editorial reconstructions.

  1. s1
  2. s2
  3. s3
  4. s4
  5. s5

Written and curated for Faultory by Yunus Emre Vurgun, with AI-assisted research and drafting. Sources and interpretations can need correction. Read our editorial standards or report an issue.

Keep investigating