2.15.05 -- At his desk in the campus trailer that is temporarily housing the UCSD administrative offices of the California Institute for Telecommunications and Information Technology (Calit2), division director Ramesh Rao pulls up the hundreds of emails he received in the 48 hours after the Indian Ocean tsunami that killed more than 220,000 people. The disaster resonated with Rao, a professor of electrical and computer engineering at the Jacobs School of Engineering, for two main reasons beyond the obvious. First, he leads several research projects at UCSD and collaborates with researchers at Calit2's UC Irvine division on ways to improve emergency communications before and after disasters. And second, he grew up and went to college in Tamil Nadu, a tsunami-prone state with 500 miles of Indian Ocean coastline.
Q.Where were you when you heard about the tsunami?
A. I was home in San Diego, but I was in constant touch with many old friends from university who were holding a reunion that day in Chennai on the coast. My e-mail in-box was filled with updates from friends there. What struck me most about the anecdotal reports - not just e-mail, but also phone calls, text messages, and eyewitness accounts from bloggers that I was tracking on the Internet - was the apparent randomness with which the tsunami took its toll on the coast. Whole towns escaped loss of life, while, a few miles away, some villages were wiped out. They didn't get the word in time.
Q. Could information technology have reduced the death toll?
A. Fortunately, wireless technology and the Internet can improve early-warning systems and alert more people more quickly, and every minute counts. Even in parts of Indonesia closest to the epicenter, if residents were given as little as 10 minutes' notice to evacuate, many would have had enough time to escape from the coast or reach higher ground. On the eastern coast of India, they could have had up to three hours' notice.
Q. What is being done by engineers to fix the problem?
A. With $12.5 million in funding from the National Science Foundation, researchers from Calit2 at UCSD and UCI have embarked on a project called RESCUE, which stands for 'Response to Crises and Unexpected Events.' Professor Sharad Mehrotra leads the effort at Irvine, but within Calit2 the two campuses are working extremely closely on a variety of robust information systems to create 'situational awareness' for first responders and the general public. Our goal is to transform radically the ability of responding organizations to gather, manage, use, and disseminate information within emergency-response networks.
Q. Why wireless?
A. Imagine how different the response to the Indian Ocean tsunami would have been if every individual or family in a threatened area had a cell phone. Imagine also that every mobile phone operator was set up to sound the alarm to their subscribers instantaneously, via voice, alarm and/or text message. Going further, if every phone was equipped with a global positioning system, the operator could know their location and therefore push special announcements to subscribers in the most threatened regions. The automated service could even alert individuals to their best route out of a danger zone.
Q. What role does the Internet play in improved disaster response?
A. Camera phones and other data-enabled devices will transform how we respond to crises because all those e-mails, images, blogs, even video are instantly uploaded to the web. One early arrival at a disaster scene may only upload three photos from their phone, but if you have a thousand people doing the same, all of a sudden you have a detailed visual record of the disaster scene.
Q. So how do you pull all that impromptu and disparate data together?
A. For want of a better analogy, I think we need a type of Google that would gather disaster-related information from all official and unofficial sources. The first obligation in a crisis is to 'do no wrong.' Before propagating an alert, you want to have reliable information. In RESCUE, we are looking at ways to validate information without relying on any single source. Out of a large amount of seemingly reliable information, you should be able to pick up that something is definitely happening.
Q. How important is it that emergency systems be designed for customized alerts?
A. Just imagine if there was an earthquake offshore California, a tsunami was headed our way, and coastal San Diegans were given 60 minutes to evacuate. What would happen? Most families would pack up their cars, head out of town, and promptly get stuck in gridlock on highways heading east. In other words, a generalized alert might be the last thing you want to do. But if the alert system is robust enough to know where the commuters are, and to advise them on alternative routes that would prevent the traffic jams, thousands of lives might be saved.
A. How can technology deal with misleading information during a disaster - whether the information is malicious or unintended?
B. The first obligation in a crisis is to 'do no wrong.' Before propagating an alert, you want to have absolutely reliable information. In RESCUE, we are looking at ways to validate information without relying on any single source. Out of a large amount of seemingly reliable information, you should be able to pick up that something is definitely happening. If you have six eyewitness accounts of a 100-foot-high wave heading to the coast, you should be able to infer that a tsunami is going to hit, even if you don't have confirmation from a central government agency.
Q. What about the argument that cellular infrastructure is too vulnerable in a disaster?
A. Batteries die, and cell phone towers can topple -- that's true. Technology under development at research institutes including UCSD's Center for Wireless Communications could lead to much longer battery lives for cell phones, by tackling both sides of the equation: reception and transmission. As for infrastructure, all facilities are vulnerable in a disaster. But across the Indian Ocean region, there was no concerted plan for wireless operators to keep services operating in disaster areas if transmission towers and access points were swept away by flood or tsunami. This is more a failing of policy than of technology.
Media Contact: Doug Ramsey, (858) 822-5825, email@example.com