Orrell - The future of everything : the science of prediction : from wealth and weather to chaos and complexity

FROM WEALTH AND WEATHER
TO CHAOS AND COMPLEXITY

INTRODUCTION
THE SCIENCE AND SOCIOLOGY OF FORECASTING

The term natural disaster has become an increasingly anachronistic misnomer. In reality, human behavior transforms natural hazards into what should really be called unnatural disasters.

Kofi Annan, Secretary-General of the United Nations

Prediction is very difficult, especially if its about the future.

Niels Bohr, Nobel laureate in physics

By December 15, 1999, the run-up to the new millennium had begun in earnest. People around the world were getting the champagne in, ready to uncork it the moment 1/1/2000 did its giant, slow march into their time zone. The NASDAQ stock-market index was also acquiring a champagne-like froth, as the Internet rewrote the rules of the world economy. The only cloud on the horizona potential stormwas the Millennium Bug, a software error caused when programmers rounded off the computers internal date from four digits to two. Many predicted it would bring chaos, or even collapse, to the world economy.

In the slums, or ranchos, perched high on El Avila mountain north of Caracas, Venezuela, the Millennium Bug was not a major worry. Most of the residents had electricity, acquired by tapping illegally into the network, but Internet access was limited, to say the least. They were more excited about that days referendum, which their hero, the new president, Hugo Chavez, had called to get the constitution approved. Voter turnout was high, and 78 percent in favour, despite the heavy rains, which had been falling for days and seemed to be getting worse.

Normally, the rainy season lasted only until October, but this year was an exception. The weather was out of synch. Some in the government suggested that the more vulnerable areas of Caracas be evacuated, in case the steep slopes became unstable. Perhaps not wishing to disrupt the referendum, however, the government took no action.

Early the next morning, December 16, the northern side of El Avila, which faces the coastal resorts near the airport, simply gave way. Witnesses said that huge waves of water, six metres high, cascaded down the mountain, carrying away everything in their waketrees, cars, houses, people. Giant boulders hurtled down narrow gullies, not stopping until they fell into the ocean or smashed into the luxury apartment blocks that lined the coast. One survivor, a young woman, described waking up in the middle of the night to the sound of the water and the rocks crashing down the mountain and people yelling, The river is coming! The water filled half her house before she could get out the door. Many others werent so lucky.

At first, the scale of the disaster was not comprehended. Estimates were for 100 dead, then 500. By December 22, this had grown to 30,000, with perhaps ten times that many left homeless. Many of the bodies were swept out to sea or buried under the mud, and were impossible to recover. Survivors gathered in halls and stadiums, desperately searching for lost relatives. Rescue workers were completely overwhelmed. One told a journalist from the Independent how he had rescued a three-year-old girl: Every time she saw water, she screamed. Many feared that the lack of drinking water and sanitation would lead to disease outbreaks, magnifying the human impact, but fortunately this did not come to pass.

Almost immediately, the storm was politicized. The presidents opponents castigated him for proceeding with the referendum. On Christmas Day, Chavez distributed gifts to hundreds of orphaned children at the Poliedro sports arena, in Caracas. When a reporter suggested the ex-paratrooper was culpable for the disaster, he replied: They should shoot me if I have any personal responsibility in this.

Could this storm, with all of its social, economic, and medical repercussions, have been predicted? Was it a random, unforeseeable event, or was someone responsible? How about the weather forecasters? The conditions for the storm began to develop in early December, when a cold front encountered a southwesterly flow of moist air, resulting in precipitation over the northern coast. There was one week of moderate rain, followed by two days of extremely heavy rain on December 15 and 16. The daily totals recorded at the nearby Maiquetia International Airport for these days were so excessive that, in theory, they wouldnt be repeated for 1,000 yearstruly a millennial storm. Almost by definition, such events do not get predicted; no forecaster likes to predict something he has never seen happen.