American Scientist

When Lake Palcacocha, a glacial lake in Peru, suddenly burst in 1941, releasing 8 to 10 million cubic meters of water, the resulting flood killed an estimated 6,000 people, and the water and debris destroyed about a third of the town of Huaraz, 23 kilometers away. Although Peru implemented a lauded engineering program in the decades that followed to stabilize glacial lakes, climate change and political shifts in the region have brought the danger level back up. “That town has been rebuilt in almost exactly the same location, and now it’s much more populous. That’s particularly scary,” says Thomas Robinson, a disaster risk specialist at the University of Canterbury in New Zealand. This example shows how the danger from such glacial lake outburst floods (or GLOFs) isn’t just a matter of a physical hazard, but varies with the local population level and the availability of government support in the region. “Without adding any human dimensions, hazard parameters aren’t very accurate or useful,” says Caroline Taylor, a PhD student in physical geography at Newcastle University in the United Kingdom. Taylor, Robinson, and their colleagues have used satellite data to identify 1,089 river basins across the globe that contain glacial lakes. They factored in the exposed populations, and used known indices of corruption, human development, and social vulnerability to rank each of these basins on the danger that a GLOF would pose to people in the vicinity (see the maps below). As the team reported in Nature Communications, they found that about 15 million people worldwide are exposed to potential GLOF impacts. More than half of those people live in four countries: India, Pakistan, Peru, and China. (The 2022 floods in Pakistan, however, were triggered mostly by abnormally large monsoons.)

The high mountain regions of the Himalayas are well known for GLOFs, but the team found similar dangers in the Andes, although there are relatively fewer studies of that region. Other surprising results from the study include potential vulnerability in places such as Bhutan (including the glacial lake at the foot of the mountain called Jomolhari, shown in the photo below). The number of people who would be affected by a GLOF in Bhutan is low compared with larger countries, but it is a high percentage of the population, upping the danger score. “The way to think about that is, if you have a GLOF, how many people have you got to respond? In Bhutan, you could have a single GLOF that affects a large proportion of the country. Suddenly, their ability to respond is massively impacted,” Robinson says.

GLOFs are tightly linked to climate change. Worldwide, glaciers are projected to lose up to half of their mass to melt by 2100. But as glaciers recede, growing populations move closer to them, looking for farmland, water, and hydroelectric power. Limiting climate change will help slow the growth of glacial lakes, but Taylor notes that each location varies in geology and politics, requiring a different combination of early warning systems, engineering solutions, and land-use planning. “Unfortunately, there is no single solution, and what is appropriate in one location may not work in another,” she says. Robinson notes that the team’s goal is to help focus research efforts on the most potentially dangerous glacial lakes. “If we can go there and show that actually those lakes are not going to burst, great,” he says. “But until we’ve done that, we don’t know.”