December 2005 - January 2006
Predicting hurricane damage
Brian Bush. (Photo by Carlye Calvin.)
The scenario is depressingly familiar: a hurricane rips through a coastal town and knocks out power for weeks. People line up for ice and rely on generators while waiting for emergency crews to restore electricity.
But what if government officials and utility crews could anticipate the specific damage before the storm hit, thereby responding more quickly?
Brian Bush is trying to make this possible. An expert in computer simulations of energy and transportation infrastructure at Los Alamos National Laboratory, Brian is working as a visiting scientist in RAL and CISL. His goal is to incorporate better weather data into the hurricane damage models developed at Los Alamos for the federal government, and ultimately to link NCAR's Weather and Research Forecasting model (WRF) with the Los Alamos models.
Last summer, as one hurricane after another headed toward the southeastern United States, Brian reviewed forecasts constantly and consulted with NCAR hurricane experts. He communicated the information to his Los Alamos colleagues, who combined wind and rain forecasts in weather prediction models such as WRF with storm surge data from the Army Corps of Engineers and input from other models. The team generated experimental damage probabilities for generation stations, substations, and power lines.
"It's been such a busy season that we haven't done the full validation yet, but subjectively the models did pretty well," Brian says.
By working with NCAR scientists—including MMM's Chris Davis, Joe Klemp, and Wei Wang, and RAL's Yubai Liu—Brian is developing insights into the strengths and weaknesses of individual models. "It's sometimes good to look under the hood and see what's going on," he says.
"It's sometimes good to look under the hood and see what's going on."
Predicting damage to energy networks is extremely complex, Brian notes. It involves forecasting windfields and storm surges, as well as mapping out areas where the energy grid is most vulnerable (such as power lines above ground), and anticipating which areas are most likely to be affected by flooding, winds, and flying debris.
Brian, who will be at NCAR for two years, is also planning to work with scientists across several divisions to study how climate change may affect energy use and how, in turn, increased power plant emissions may affect climate. In addition, he is helping NCAR deputy director Larry Winter plan a conference next year on designing energy infrastructure that is resilient to weather and climate impacts.
Brian's work here is part of a broad-based multiyear collaboration between NCAR and Los Alamos. The collaboration will focus on the societal, economic, and physical impacts of weather and climate.
While Brian intends to tackle a range of topics, he spent his first months at NCAR just trying to keep on top of all the data about the many hurricanes and clean up his computer files after the storms. For Hurricane Wilma alone, he amassed over 100 gigabytes of data. "Storms leave digital debris as well as infrastructure debris," he says.
• by David Hosansky and Bob Henson
Also in this issue...
The 2005 Outstanding Accomplishment Awards
Offsite but not unseen:
UCAR's non-Boulder staffers stay in touch
A TWERLE reunion
Predicting hurricane damage
Atmospheric science books for all ages
CGD research shows that permafrost may thaw in this century
New CG library
Just One Look: Santa
Staff Notes home page | News Center