Before the Flood—Predicting the Deluge
“There are flood and drought over the eyes and in the mouth, dead water and dead sand contending for the upper hand. The parched eviscerate soil gapes at the vanity of toil, laughs without mirth. This is the death of the earth.”
“What is the appropriate behavior for a man or a woman in the midst of this world, where each person is clinging to his piece of debris? What’s the proper salutation between people as they pass each other in this flood?”
[Note: this post is the first in a series that describes the Columbia Global Flood Initiative, a new Earth Institute initiative that attempts to better understand, predict and mitigate the impact of extreme floods around the world. For more information about the project, please see our website at the Columbia Water Center].
It’s been a rough year. From bloodbath in Libya and other Arab nations, to the earthquake, tsunami and nuclear crisis in Japan, to the unfolding and ongoing human tragedies in Haiti, Pakistan and Brazil, to the “super-typhoon” Megi that left 200,000 people homeless in the Philippines – 2010 and the first quarter of 2011 has felt like a never-ending cascade of natural and human-made catastrophes. Just last week torrential rains deluged Thailand, causing floods that have killed at least 17 people.
In our frustration we seek someone to blame—or disclaim all responsibility, chalk it all up to an act of God or Mother Nature. Pin it on human-made climate change, rail at the mismanagement of disaster–or else shrug your shoulders, change the channel and think of something else.
The truth is more complicated, of course. We cannot deny that human-made global warming is happening and that it could be an unmitigated catastrophe; but neither can we escape the truth that there is ebb and flow in nature that preceded human beings, and will outlast us. The greatest tragedy would be let our anger or denial or despair obscure our understanding or stun us into complacence and keep us from very real actions that can save lives.
The Columbia Global Flood Initiative, a new joint initiative of Columbia Water Center, the International Research Institute for Climate & Society (IRI), CICAR (the Cooperative Institute for Climate Applications and Research) and the Center for Climate Systems Research, seeks to better understand, predict and plan for extreme floods. The project is based on the conviction that while human beings may not have direct control of where and how much rain falls (the long-term effects of human-caused global warming notwithstanding) there is a great deal more that can be done to manage the risk of extreme flooding around the world.
The project emerges from more than a decade of climate research by Columbia scientists and others on the patterns of global moisture flow in the atmosphere. In that time, there has been a quiet revolution in climatology—a revolution that has been obscured by the debate and concern over long-term human-caused climate change.
Today, scientists have a much better understanding of how the global climate works than they did even a few years ago. As a result, phenomena such as flooding—once thought of as essentially random events–are increasingly understood as the result of predicable (if complex) climate patterns.
What this means is for any given part of the world it may be possible to forecast when and where the next extreme flood will occur, anywhere from a season to a year ahead of time. Global climate patterns that can affect where and when extreme floods will occur include El Nino/La Nina-Southern Oscillation (ENSO), the Pacific Decadal Oscillation (PDO) and other “climate precursors” such as ocean temperatures, or the amount of regional snowpack.
The way these various factors combine differs by region, and determining which factors play a what role in a given region is complicated. To give one example, in 2000 Shaleen Jain and Upmanu Lall published a paper in Water Resources Research on the magnitude timing of spring floods in the Blacksmith Fork River in Utah. What they found was that in spite of conventional climate wisdom that the Blacksmith Fork River was only weakly affected by ENSO variations, a deeper analysis suggested that there was a non-linear, statistically significant connection between the ENSO, the PDO and flooding on the river.
A three-dimensional chart from their paper illustrated the complexity of the connection, and how selected combinations of the extreme phases of ENSO and PDO appears to lead to a higher flood potential for the river.
The Blacksmith Fork River is just one of many studies by Columbia scientists targeting flood-prone river basins in different areas around the globe.
What does this all mean? The implications are vast. Understanding when and where an extreme flood is likely to occur a season or even a year ahead of time could allow everyone from policymakers to reservoir managers to emergency responders better plan for what is coming.
An example of what such an early-warning response might look like occurred in 2008, when the Columbia’s IRI issued a seasonal alert that parts of West Africa faced a higher than average chance of heavy flooding. In response, the International Federation of the Red Cross/Red Crescent, which had partnered with IRI, issued its first-ever emergency appeal based on a seasonal climate forecast.
As a result, the IFRC was able to distribute supplies within 48 hours of the beginning of floods in contrast with an estimated 40-day time lag and three times the per-person cost for a similar incident the year before. The experience of the 2008 West African floods demonstrates that the ability of scientists to forecast floods a season or more in advance can save lives, conserve resources and help those most vulnerable to extreme floods better weather these traumatic events. (For more information about this event, please see IRI’s publication “A Better Climate for Disaster Management” or watch a video on the same topic: Climate Forecasts Improve Humanitarian Decision Making in West Africa).
Real-world applications of the emerging science of flood forecasting are still in development. Emergency responders, reservoir managers and others who stand to benefit from seasonal flood forecasting are understandably wary of acting on forecasts that are by their nature probabilities, not certainties.
This will likely change, however, as forecasting gets more sophisticated. But another way to make flood forecasting more useful is to couch it in a broader, whole-system flood management design that looks at the best way to address different parts of flood risk at different times. I’ll cover how that design might work in future posts.