The Data Analytics of Hurricane Irene

Author:Amanda Brandon, Spotfire Blogging Team. 

At the time I’m writing this, nearly 4 million people are without power due to the damage caused by Hurricane Irene, which made landfall this past Saturday morning. A Category 1 hurricane at her U.S. mainland debut, Irene wasn’t nearly as destructive or as powerful as the infamous Category 5 Hurricane Katrina which celebrates her sixth anniversary today.

However, Irene’s damage has been estimated to be about $2.6 billion in the United States, Kinetic Analysis Corp. told Bloomberg on Sunday. The total economic impact of the killer storm is expected to settle somewhere around $7 billion, about half of last week’s projection of as much as $14 billion. Katrina’s impact was  $133.8 billion in damages and more than 1,800 fatalities, according to the Christian Science Monitor.

While this data is staggering, what’s even more staggering is the data analytics that powers these estimates. 

More Read

“Average” Statistics that Bruise Our Ears

R project named in Intelligent Enterprise 2010 Editor’s Choice Awards

Customer Feedback and Data Analysis: The Keys to a Good Customer Retention Rate

Analytics at the Intersection of Humans and Machines

Earth Day: How Green BI has Become a Reality

According to a report in the International Business Times, the data modelers were projecting damages and the costs right alongside meteorologists warning residents of the storm’s path.

There’s a niche market for this type of data analysis, a spokesman for the Cornell University Theory Center told the IB Times. The buyers of such data are insurance companies and government agencies. One way these companies use this data is to fuel decisions on contracts with Bahamas-based reinsurers, which is a means of risk management for major casualty insurers.

Data modelers use multiple public data sources to build their projections. Sources include the National Weather Service and the National Oceanographic and Atmospheric Administration. According to the IB Times, the risk management analysis companies feed this public data into their proprietary models to crunch the numbers.

The types of data they use to make projections stem from past hurricane paths and damage reports. Then, they combine this data with the current hurricane’s projected path and strength to make predictions on the destruction and economic impact.

The insurers rely on this data to ease their burden, especially in a year with industry losses expected to exceed $90 billion, reports the IB Times. The losses stem from widespread tornado and flood damages across much of the southeastern U.S. this past spring.

Real-time computing and analysis have made it easier to provide more accurate weather and damage forecasts. The advances in this field over the past six years are as staggering as the impact these storms make on the communities they affect. Here’s to even more advancement in this crucial area of data analytics.