Some interesting background information on lake turnover in an active crater lake in Cameroon:
*Article copied from the University of Arizona Geosciences Page: http://www.geo.arizona.edu/geo5xx/geos577/projects/kayzar/html/lake_nyos_disaster.html
The Lake Nyos Disaster
August 1986-Lake Kills 1,700
Lake Nyos is an active crater lake that formed by an eruption about 5 centuries ago. Nyos and Lake Monoun, located 95km to the southeast of Nyos, are the only two volcanic lakes in the world other than Lake Kivu that contain large amounts of CO2 dissolved at depth (Holloway, 2000). Nyos and Monoun both released their gases in the month of August only two years apart from one another.
After many years of study the science community has come to an agreement that the origin of CO2 within Lake Nyos is due to CO2 that rises from volcanic activity. This CO2 is then dissolved into groundwaters and transferred to the lake resulting in the slow saturation of the hypolimnion (Kanari, 1989; Holloway, 2000; Kling, 1989). In most crater lakes, turnover of the stratified waters occurs periodically and harmless amounts of dissolved gases are released; however, the problem with Lake Nyos and Lake Monoun is that these two particular lakes do not periodically turn over (Holloway, 2000). Thus, dissolved gases are allowed to reach much higher concentrations. The setting of Lake Kivu is analogous to Lake Nyos.
The exact cause of the gas release at Lake Nyos is still unresolved. One theory is that a small confined area of the lake released gas allowing for the stratification in Lake Nyos to remain (Kanari, 1989). Another theory describes a slow influx of heat into the system causing instability (Kling, 1989). A landslide within the lake is another possible explanation for the displacement of the bottom CO2 saturated layers in Lake Nyos. Evidence of water surges on the southern shore of the lake suggest a possible seiche motion of the lake waters (Kanari, 1989). In all situations, the possibility of a volcanic injection is ruled out. High concentrations of reduced iron were found in Lake Nyos, the presence of which cannot be explained by the possibility of a volcanic injection into the lake (Kling, 1989). In general, a gradual heating from below the lake is widely accepted as the cause for rollover and or gas release.
The gas killed all living things within a 15-mile (25km) radius of the lake.
A similar emission at Lake Kivu could be possible. Lake Kivu is heated from below by thermal springs from the Virunga volcanic field. It is important to note, however, that because of Lake Kivu’s great depth, the lake is generally thought to be more stable than Lake Nyos or Lake Monoun (Kling, 1989). A more stable lake would require a much larger influx of heat to disturb the stratification and potentially cause rollover.
Degassing Lake Nyos – Hazard Prevention
The studies of Lake Nyos revealed that there was more CO2 forming at depth in the lake that could overturn once again, so an attempt was made in 2001 to degass the lake using an electronic pump that would simulate an eruption.