UMBC Researchers Use IBM Technology to Fight Rising Threats of Forest Fires

New System will Monitor Wildfire and Smoke Patterns, Offer Real-Time Analysis.

ARMONK, N.Y. and BALTIMORE. IBM (NSYE: IBM) today announced researchers at the University of Maryland, Baltimore County (UMBC) are using IBM technology to analyze smoke patterns during wildfires. The goal of the research is to provide fire and public safety officials with a real-time assessment of the fire, allowing them to make more informed decisions on public evacuations and health warnings.

Current analysis of smoke patterns is limited to weather forecasting data that is updated every six hours, observations from front-line workers and low resolution satellite imagery. Using a cluster of IBM BladeCenter servers housed at the UMBC College of Engineering and Information Technology’s Multicore Computational Center (MC2), the team of faculty and student researchers will be able to instantly process the massive amounts of data available from drone aircraft, high-resolution satellite imagery and air-quality sensors to develop more effective models for smoke dissipation.

According to the U.S. National Interagency Fire Center, wildfires present a serious and dangerous threat to commercial buildings, private homes and people. In the last year alone, there have been more than 76,000 individual fires, consuming an approximate 5.8 million acres. And according to the agency’s recent Quadrennial Fire Review 2009, the threats are only expected to increase due to drier conditions associated with climate change and increasing population growth.

“By combining the resources and expertise of IBM and UMBC, together with data acquired from agencies like NASA, NOAA, EPA and the US Forest Service, we not only hope to protect the health and safety of neighboring communities, but also prevent the threats to their lives and property,” said Dr. Milton Halem, director, UMBC’s MC2. “We expect this research will be of great interest for many groups - especially in states where these types of fires are commonplace.”

In addition to the IBM blade servers, researchers also will be using the recently announced IBM InfoSphere Streams, a technology designed to help organizations analyze information from any source; narrow it down what people are specifically looking for; and continuously refine the answer in real-time. Together, the UMBC team will have the first-ever opportunity to capture data from a network of surface, aerial and satellite sensors already established by organizations such as NASA and the National Oceanic and Atmospheric Administration (NOAA), and provide more timely:

* identification of the location and progression of wildfires and smoke coverage;
* model forecast of fire and smoke behavior and plume dispersion for air quality monitoring;
* retasking of sensors to create new measurements;
* assistance by issuing real-time forecasts in the fire-fighting effort; and
* issuance of public safety and health alerts.

The research work is in-part funded by IBM’s Shared University Research (SUR) awards program, which is designed to connect the research and researchers at universities with IBM Research, IBM Life Sciences, IBM Global Services and IBM’s development and product labs. The SUR Awards program’s goal is, among other things, to increase access to and successful use of IBM technologies for research and in curriculum.

“We are pleased to be working with the researchers at UMBC to create a breakthrough Smarter Planet application that will help save lives by predicting the movement of wildfires,” said Bernie Meyerson, vice president of innovation and global university programs, IBM. “UMBC students and faculty will have access to IBM’s latest hybrid computing platforms with incredibly fast predictive analytic capabilities.”

The new system will be based on a set of blade servers designed by IBM, including the JS22, HS22 and QS22 models. Together, the computing power of the blades will help process a massive amount of real-time data streams of smoke plume models, helping discern the scope and projection of wildfire lines.

The UMBC MC2 research team is currently in the design stages of their monitoring technology, and expects to have a prototype available next year.