Nature Climate Change | Climate Feedback

AGU 2009 Joint Assembly: Climate risk research

Today, at the AGU Joint Assembly, I looked into two kinds of risk: of drought leading to civil violence and of the odds that extreme weather events may cause other types of disruption.

It’s always encouraging to see undergraduates engaged in scientific research, the more so when it is interdisciplinary. That’s what surprised me about a poster with the intriguing title, “Climate Change and Civil Violence,” which turned out to be a class project at Princeton University, stretching over three semesters, with three successive classes of students involved in the research.

The students, led by their professor, Gregory van der Vink, focused on a swath of West Africa – 21 countries from Mauritania to the Central African Republic, including most of the semi-arid Sahel region, as well as coastal forests. They looked at the likelihood that climate change, especially increasingly severe drought, will lead to civil violence in each of the countries studied. It is obviously difficult to attribute specific environmental causes to social phenomena, but the students—majoring in politics, various sciences, pre-engineering, and other fields—gave it a good try.

The choice of drought as the key climatic factor was based on the expectation that the Sahel will further dry, shortening the growing season by at least 20% over the next 40 years. Nomadic herdsmen will, for example, have to move further south or adopt a more sedentary life, or both.

Another key factor is the resiliency of the population, its ability to cope with the coming climatic changes. This factor varies considerably from country to country, the researchers found, and is related to population size, political system, natural resources, and other variables. The relative significance of each of these is difficult to assess, of course.

Bottom line: after exhaustive analysis, the students concluded that the five countries in the study that are most vulnerable to drought-induced civil violence are Nigeria, Senegal, Mali, Côte d’Ivoire, and Gambia. Of these, three—Nigeria, Senegal, and Mali—have already experienced conflicts that included an environmental factor, they say. Of the five, only Gambia has not experienced any violent conflict during the past 10 years.

The study, as the students themselves note, was a preliminary effort to correlate environmental and socio-economic risk factors that could lead to civil violence. They are not predicting specific events in particular countries, but there seems little doubt that this topic is one that will assume increasing importance in coming years to policy makers and civilian populations alike, and not just in Africa.


Another kind of risk discussed at Joint Assembly was that of extreme weather events. Are such events, many people ask, caused by climate change, or, specifically, by manmade greenhouse gas emissions? While the short answer is that scientists cannot say that any particular event—a devastating hurricane, for example—resulted from rising atmospheric CO2, an effort is under way to determine whether such emissions are altering the probability of these events occurring during a given time scale.

Dáithí Stone of Oxford and Cape Town universities, UK and South Africa, described a worldwide project, just getting underway, that harnesses the power of ordinary home computers to crunch data in order to assess the changing risk. The project begins by looking backward to 1959, in order to refine the program, which will be followed by simulations of the 2006-2007 period. These simulations will then be run with manmade contribution to atmospheric greenhouse gas levels and related ocean warming removed.

Stone said, in an interview, that the risk analysis will be of the likelihood of events that could occur. If currently there is a 10-year flood, he explained, you have a 10% chance of that flood occurring in any given year during a decade. “But, because of what we are emitting, that one in ten chance may change to one in 20, or one in five, of that happening. We are not affecting the event; we are affecting the odds, like loading the dice.” It seemed an apt metaphor.

Those interested in participating in this study by offering computer time, can learn more here.

Harvey Leifert is a freelance science writer

Comments

There are currently no comments.