Saturday, July 25, 2009

Don't Pass Gas

A year ago, almost to the day, I posted a piece I also called Don't Pass Gas, in which I reflected on one of my primary concerns regarding subterranean sequestration of CO2: What would happen if the reservoir was breached?
Oddly (and tragically) enough, that is a question that has been answered: on August 21, 1986, Lake Nyos in Cameroon erupted about 1.6 million tonnes (I think that spelling implies metric tons, about 2200 pounds) of carbon dioxide, killing 1700 people and thousands of livestock. A few years later, I argued with a professor over whether pollution was necessarily man-made, and used Lake Nyos as an example. (For the context of atmospheric chemistry, his definition stood) But imagine millions of tons of CO2 erupting near a city. The victims would never know what hit them.
It turns out, according to an article in The Guardian today, that Lake Nyos may have been only a slighly smelly fart compared to the full-out gas warfare threatened by another central-African Lake:
More than two million people living on the banks of Lake Kivu in central Africa are at risk of being asphyxiated by gases building up beneath its surface, scientists have warned.

It is estimated that the lake, which straddles the borders of the Democratic Republic of Congo and Rwanda, now contains 300 cubic kilometres of carbon dioxide and 60 cubic kilometres of methane that have bubbled into the Kivu from volcanic vent.
(...)
Kling has since turned his attention to Lake Kivu, which is more than 3,000 times the size of Nyos and contains more than 350 times as much gas. More worrying is the fact that the shores of Kivu are much more heavily populated. About two million people live there, including the 250,000 citizens of the city of Goma.
Since tropical lakes are always warmer at the top than at the bottom; the water in the lake will tend to stay at the same depth it's at for years to decades until some unusual event (a landslide, earthquake, or volcanic eruption, for example) causes them to overturn. Gases in lower dissolved water never have a chance to escape to the atmosphere- in fact, deep dissolved gases increase the deep water density, reinforcing the stratification.

In temperate lakes, seasonal temperature changes cause regular overturns in the water column. Water very close to freezing is a little less dense than warmer water, so as winter ice melts off, warming surface water sinks deeper into the lake carrying oxygen-rich water to depth, and forcing oxygen-depleted, CO2-rich and nutrient-rich bottom waters to move toward the surface. The same pattern in reverse occurs in the fall and early winter, as cooling surface water becomes denser and sinks. When the surface water gets close enough to freezing to start expanding (as it starts to lock into bonding patterns more like ice, forcing it to expand and become less dense), the process stops; the temperature profile of the lake will be close to freezing throughout, very close within a few few feet of the surface, and (often) frozen on top. The narrow range of temperatures also means that wind-driven circulation can have much deeper effects; density differences are very low throughout the lake. Come spring, the biannual mixing starts again.

This biannual mixing means that temperate lakes most often have a fair amount of oxygen available throughout their depth profile (which further means that methane gets oxidized, or can't even form), regularly have an opportunity to release built-up byproducts of the metabolisms of the critters at depth, and important nutrients (nitrogen and phosphorus, for example), can return to upper, better-lit levels, where they can stimulate the growth at the lower levels of the food pyramid.

In contrast, tropical lakes, with their lack of vertical circulation, tend to accumulate gases and nutrients at depth. The shallower areas may be biologically rich and diverse, but shallow water over deeper areas, and the depths themselves, tend to be biologically impoverished. If there is a source of CO2 emptying into the lake, it can build up and accumulate for years. However, if some event causes a wedge of that water to move toward the surface, dissolved gasses can exsolve (just as soda pop does, especially when its warm and/or agitated), form bubbles (which lowers the density of the water), start rushing faster toward the surface, drawing more deep water behind it, which then starts forming bubbles and so on. In short, a rapid, runaway, self-reinforcing overturn of the lake can cause a catastrophic release of gases that have been accumulating for decades or longer.

I think this would be a very interesting kind of eruption to witness from a plane in an uninhabited wasteland. But the idea of it happening in the vicinity of 2 million peoples' homes is horrifying.

No comments: