UMass Magazine Online

Fall 2003

Departments

Around the Pond

Features

Experiencing Jeff Corwin

Drawing on the past

Clean-up at the old Davis Mine

Around the Pond

It's a big, big deal
Tracking big winds with a major grant

– Christopher O’Carroll ’97




	Where the action all happens: David McLaughlin ’84, ’89G principal investigator and director of CASA.

EVERY YEAR, AN ESTIMATED 1,000 tornadoes touch down in the United States, killing about 100 of us and costing our nation $1 billion in property damage and economic disruption. Meteorologists, relying on Doppler radar technology, do their best to predict the arrival of these destructive windstorms, and public officials make every effort to alert people who live in threatened areas. But one-third to one-half of each year’s tornadoes go undetected until it’s too late.

“And for those that we do detect, the average warning time is 12 minutes,” says electrical and computer engineering professor David McLaughlin ’84, ’89G, whose areas of expertise include the design and construction of radar sensing equipment. Furthermore, McLaughlin says, tornado forecasting techniques are so imprecise that 75-80 percent of warnings turn out to be false alarms.

Other catastrophic weather events such as hurricanes, floods and thunderstorms multiply the annual death toll and economic devastation. And the shortcomings of today’s weather forecasting technology leave us just as vulnerable to these natural disasters as we are to twisters. McLaughlin, backed by an Engineering Research Center grant from the National Science Foundation (NSF), and aided by dozens of collaborators in meteorology, computer science and other fields, is putting his engineering skill to work to change all that.

McLaughlin is principal investigator and director of a new five-year, $40 million research and development effort that aims at nothing less than revolutionizing the country’s severe weather response system. This ambitious project, dubbed CASA (Center for Collaborative Adaptive Sensing of the Atmosphere), will be headquartered at the UMass College of Engineering with research partners at the University of Oklahoma, Colorado State University, the University of Puerto Rico, Mayaguez and five other campuses. With $17 million in NSF funding, and $5 million from the Massachusetts state government, CASA will also receive substantial support from Raytheon, IBM and numerous other corporate partners.

Sharing leadership of the project with McLaughlin are co-investigator Jim Kurose, computer science professor and associate director of CASA, and managing director Brenda Philips, who will draw on her background in business administration to coordinate the diverse efforts of CASA’s multi-disciplinary research team.

Collaborators in their far-flung laboratories will be reaching across a variety of disciplinary boundaries as engineers, computer scientists and meteorologists working on the system’s hardware and software join forces with sociologists designing effective emergency warning campaigns. “One of the key issues,” Philips says, “is going to be communication among the different disciplines.”

At an October 1 press conference announcing CASA’s launch, Chancellor John V. Lombardi stressed the same point when he hailed the project as “a triumph of collaboration.” McLaughlin told the assembled journalists, state officials, academics and industry executives at the press conference that an NSF Engineering Research Center grant is “a very big deal” for UMass, an indication that “in the marketplace of ideas, we’re performing at top rank.” He is especially pleased to be heading this research effort, McLaughlin said, because the weather forecasting challenge that CASA has taken up is “a complex problem with enormous societal impact.”

The problem is one that would not exist if the earth were flat. Today’s weather forecasting relies on an array of about 150 huge Doppler radar installations, positioned hundreds of miles apart throughout the United States. A radar beam travels in a straight line, so if these sensors were spread out across a flat plane, their beams would stay close to the earth’s surface and would do a thorough job of probing the lowest portion of the atmosphere. On the surface of a globe, however, things work out a bit differently. Near its point of origin, a Doppler radar beam senses activity in the lower atmosphere. But with the planet curving away, the beam’s straight path carries it further and further from the surface; a few miles out from its source, the beam is taking readings high in the atmosphere, far above the region where dramatic weather events occur.

By detecting upper-level disturbances, the Doppler system allows meteorologists to extrapolate what is likely to happen down below, but their forecasts would be more accurate if they had a way of acquiring data directly from that portion of the atmosphere that actually makes contact with land and water. “Today’s weather observing system is not able to see the lower kilometer or thereabouts of the earth’s atmosphere,” McLaughlin explains. “And that’s where all the action happens. That’s where friction causes the atmosphere to drag on the ground and that’s where hazardous weather is spun up.” And that’s where CASA is going to change the rules of the game.

The new weather forecasting system that McLaughlin, Kurose, Philips and the rest of the CASA team envision will replace a small number of large, expensive radars with a large number of small, inexpensive ones. Instead of a few widely spaced Doppler sensing devices, behemoths that can cost as much as $10 million apiece, CASA will use dense clusters of radars, each about the size of a desktop computer and costing an estimated $20,000. Mounted on rooftops and cell phone towers and positioned just a few miles apart, these devices will vastly increase the amount of data meteorologists can gather from a given area.

“The big radars have all of their resources in one beam,” McLaughlin says. “They can only do one thing at a time. Whereas, when you talk about very low-cost radars sitting on top of cell towers, each one is going to have multiple beams. A region that today is covered by a single beam could be covered by 10,000 beams in the system we’re talking about.” So if a single storm system threatens one area with tornadoes and another nearby location with overflowing streams, the CASA system will be able to assign some radar beams to tracking individual whirlwinds while others provide precise rainfall measurements to keep emergency personnel one step ahead of flash floods.

Although no area of the country is immune from heavy weather, UMass is heading up the CASA effort not because of this state’s special vulnerability to meteorological catastrophes, but because of this university’s leadership in relevant areas of technology. “We have the expertise here in sensing and in distributed computer networking to bring to bear on the problem,” McLaughlin says.

Kurose and his team, providing the computer science know-how for the project, confront some formidable software challenges. The CASA system has to be able to process trillions of pieces of data, juggling input from its entire network of sensors, converting the raw data into a three-dimensional picture of atmospheric activity and supplying outputs in real time to forecasters, emergency responders and other system users. “Our system has lots of different users with competing demands,” Kurose says. “We’ve got to move the data from the right places to the right people at the right time.”

McLaughlin sees the system providing not only better response to natural disasters, but also improved national security. The same technology that tracks storm systems through the lower atmosphere, he points out, can be used to predict the wind-borne dispersal patterns from lethal biological, chemical and radioactive releases.

The project’s first “test beds” will be set up in areas associated with disastrous storm activity. One experimental network of sensors will cover 20 percent of the frequently tornado-ravaged state of Oklahoma. Another will focus on the flood-prone city of Houston, Texas. A third will monitor the entire island of Puerto Rico, which lies in the path of many hurricanes and tropical storms. Field testing the new weather forecasting technology in these three locations, CASA planners hope, will lay the groundwork for taking the system nationwide once this five-year project is complete.

[top of page]

It's a big, big deal

Big Deal: Larger photo

Passing the $100 million milestone

Milestone: More images

Speaking of the weather

Weather: More images

Ground Zero: Of memory & memorials

Ground Zero: Larger image

The lotus leaf, the moth's eye and cars that clean themselves

Lotus: Larger image

A good way to live