The IPCC Predicts Future Flood Disasters - Who Will Bear the Risk?

The second working group of the Intergovernmental Panel on Climate Change (IPCC) is charged with estimating the impact of global warming on specific regions of the world. Their summary report for policymakers (pdf), released in Brussels today, predicts that "Drought-affected areas will likely increase in extent. Heavy precipitation events, which are very likely to increase in frequency, will augment flood risk." In other words, droughts and floods will get more severe in areas that are already prone to such disasters.

To understand this problem more clearly, let's consider the problem of "augmented flood risk." In physical terms, this simply means that floods will become more common and heavier in volume, and will cover more territory. In human terms, it means that people in flood-prone areas will bear greater risks of injury, homelessness and other damage. This human factor motivates most discussions of flood risk, although potential damage to ecosystems is also an important problem.

Since the IPCC predicts that future flood damage will be like past damage, only worse, analyzing the historical record can provide useful insights. Fortunately, this record is available from another locale in Brussels - the Center for Research on the Epidemiology of Disasters (CRED) at the Universite Catholique de Louvain. Using the CRED database, we can track the history of severe floods and their consequences by country. With data for the period 1960-2002, I have constructed a weighted human risk measure based on people killed (weight 1000), rendered homeless (10) or otherwise affected (1) by each flood. I've divided by population in 1980 (the period midpoint) to obtain an index of human flood damage risk in each country. I've also developed a physical flood risk index, by dividing the total number of severe floods from 1960-2002 by population in 1980.

With these indices, we can explore the relationship between flood risks and development status using the World Bank's per-capita income categories (country counts in parentheses): low income (54), lower middle income (48), upper middle income (25) and high income (22). Figures 1 and 2 graph the relationship between development status and median index values for countries' physical and human flood risks, respectively.

Figure 1 doesn't reveal a systematic relationship: Physical flood risk increases somewhat from low income to upper middle income status and then drops back for high income countries. Lower physical risk for the richest countries may well reflect both more benign climates and advanced flood control capabilities. Among the other 127 countries, there is no indication that floods are more common in the poorest.

Obviously, Figure 2 tells a very different story for human damage risk: It falls sharply and continuously with income status, and people in low-income countries bear damage risks about 23 times higher than those in high-income countries (in contrast to physical flood risks that are about 1.6 times higher).

For the development community, these results suggest three lessons for interpreting the IPCC's second report:

• The focus should be on human risk, not physical risk.
• The best insurance against human risk is development, with stronger protective institutions, greater resources for flood protection, and affordable insurance for those who suffer damage.
• Most critically, this problem is not new. People in flood-prone poor countries already suffer much more than people in flood-prone rich countries. Flood protection assistance now is important, and its benefits will compound as the rains begin falling more heavily in a warmer world.