ucar Highlights 2007

top nav

A Community of Peers

UCAR nourishes connections among scores of universities and laboratories, engaging thousands of people, all working to understand the exquisite complexity of the atmosphere and its relation to Earth, ice, sea, Sun, and life.

Download this chapter as a PDF (2.18 MB)

The personal touch | Africa in mind | Servicio en español | From passing interest to passion |

Online pathways | Extra computing | Whatever the weather


web iconForecasting System Provides Flood Warnings to Vulnerable Residents of Bangladesh (news release)
line separatorClimate Forecast Applications in Bangladesh (Georgia Tech)
line separator

Whatever the weather: dealing with atmospheric impacts

Two great rivers cutting through the heart of Bangladesh hold the fate of millions in their flow. Every few years, major floods engorge the Brahmaputra and Ganges for periods ranging from a few days to a month or more. At least 500 people died and five million were displaced in the summer of 2007 as floods inundated much of central Bangladesh. Farmers or fishers can easily lose a year’s worth of income in a single flood.

U.S. scientists have worked with Bangladeshi officials on developing better flood warnings with support since 1999 from the U.S. Agency for International Development and the relief agency CARE. The system became operational in 2003, and in 2007, a new dissemination program brought warnings directly (sometimes through door-to-door contact) to more than 100,000 residents. Thomas Hopson, who helped develop the forecasting techniques, is now a postdoctoral researcher at NCAR, where he’s watching the research bear fruit. “It’s hugely satisfying to see these forecasts being used to help Bangladeshi citizens,” he says.
thomas hopkins

Thomas Hopson

A day’s notice before a flood isn’t much help to poverty-stricken residents with little access to TV or radio. But with 10 or more days of warning, farmers can hasten harvesting or delay planting as needed, and people and property can be moved to higher ground. The problem is that weather forecast skill goes down with increasing lead time, especially in the tropics. However, by creating forecast ensembles—computer forecast models run dozens of times with tiny differences in the starting points—scientists can see how quickly these forecasts diverge from one another, thus revealing how confident they can be in any day’s forecast. When the ensembles are used to drive hydrologic models, confidence in a flood forecast can be expressed quantitatively.

Hopson and lead scientist Peter Webster (Georgia Institute of Technology) forged a partnership with the European Centre for Medium-Range Weather Forecasts to use ECMWF’s highly regarded ensembles, which span periods of up to 15 days for weather and up to six months for seasonal trends. Hopson combines these outlooks with river observations from Bangladesh to issue daily odds that river heights will exceed critical levels. In a 2007 pilot program organized by the Asian Disaster Preparedness Center, residents of five rural provinces along the Brahmaputra received probabilistic flood forecasts going out to 10 days. In coming years the forecasts will be extended to cover intervals of 15 to 30 days and eventually out to six months, providing the greater lead times that residents say are essential.

This happy marriage of high-end science and user-tailored warnings is the type of work Michael Glantz has advocated for more than 30 years. As NCAR’s first political scientist, Glantz studied El Niño and its impacts well before the phenomenon became widely known. He’s now working with campuses nationwide to establish curricula in climate, weather, and water affairs. These blends of physical and social science are urgently needed, says Glantz, to help societies adapt to the normal vagaries of weather as well as the possible repercussions of climate change.

NASA satellite photos captured severe 2007 floods along the Brahmaputra on August 3 (left) as compared to May 1 (right).

power lines

Power lines

“Most of my work has been about weather and climate variability rather than change,” says Glantz.

People in prosperous countries often think they’re free of weather impacts until the lights go out or the tap runs dry. Utilities are a critical buffer between everyday American life and hardships that can last days or even weeks, as seen in the aftermath of 2005’s Hurricane Katrina. Brian Bush and his colleagues at Los Alamos National Laboratory are examining ways to provide government stakeholders with forecasts of damage to utility infrastructure and estimates of subsequent repair as part of a broad, multiyear collaboration. For the last two years, Bush has been stationed at NCAR, linking output from tools such as the Weather Research and Forecasting model (WRF) with hurricane damage models developed at LANL for the U.S. departments of Homeland Security and Energy.

In one outgrowth of this work, high-resolution wind and rain forecasts from WRF
kathy and david

Kathleen Miller and David Yates

were used in the analysis of Katrina, where they provided damage probabilities for generation stations, substations, and power lines across the central Gulf Coast. The outlook had to consider not only the time required to restore systems but also the time needed for workers to get to the damaged equipment in the first place—a major consideration for a storm as large and intense as Katrina.

Electric companies aren’t the only utilities affected by extreme weather. Water utilities must contend with the twin threats of floods and droughts, both of which appear likely to be exacerbated by climate change. Two NCAR scientists are helping utilities to see how they might best adapt and prepare. Economist Kathleen Miller and hydrological engineer David Yates coauthored the 2006 book Climate Change and Water Resources: A Primer for Municipal Water Providers. With case studies ranging from saltwater intrusion in Miami and the Netherlands to wildfires near Denver, “the focus is on usable information,” says Miller. “We wanted to make the book very accessible to people from the industry and to involve them directly in identifying vulnerabilities and options for adaptation.”

While urging utilities to recognize the serious risks posed by climate change, Yates cautions them about relying on the most fine-grained outlooks just yet: “I tell them that results from climate models are still problematic because we don’t do a good enough job of modeling precipitation.” Rather than taking model output as gospel, he advises utilities to look at what models suggest is most likely and examine their vulnerabilities in that light. “It’s simply best-management practice, with climate change being a new layer of uncertainty.”

The personal touch | Africa in mind | Servicio en español | From passing interest to passion |

Online pathways | Extra computing | Whatever the weather