Hurricane Sandy Predictions Vs. Actual

This map shows CL&P's predicted trouble spots vs. actual damage due to Hurricane Sandy. (CL&P / May 3, 2014)

A new weather model is being used to show where a severe storm is mostly likely to damage the electrical grid in Connecticut.

The system is a collaboration between the University of Connecticut and Connecticut Light & Power. A team of UConn environmental engineers overlaid weather forecasting data of wind and precipitation with CL&P data of downed power lines and other damage to the power grid from more than 100 storms since 2001.

The information allows CL&P to be better prepared in the hours before a storm and, potentially, to respond more quickly when electricity gets knocked out because of a damaged transformer or a tree limb breaking a power line.

"With this new technology, we have critical guidance ahead of a storm about severity and where we may see the most significant damage to our electric system, which can help us determine where to pre-stage crews and equipment," Peter Clarke, senior vice president of Emergency Preparedness at CL&P's parent company, Northeast Utilities, said in a statement.

UConn and CL&P started working together in 2011 — before Tropical Storm Irene and the October 2011 snowstorm which left some homeowners and businesses without power for more than a week. At the time, the system was only being tested.

"Hurricane Sandy, we had developed our initial model, and that was one of our first predictions trying to use it operationally," said Dave Wanik, a Ph.D. candidate at UConn's environmental engineering program, who helped build the model.

The model is still being revised, but it has reached a stage of reliability that it can be used for severe thunderstorms, snow, ice or hurricanes. The model was used for some smaller winter storms this year.

CL&P is using the model to estimate the expected strength of impending storms, and to estimate the location and extend of damage the storm could cause. However, the power company cautions against any optimism that the model will fix all electrical problems resulting from wind, rain and snow.

"There is no way for a utility to prevent outages in the face of Mother Nature's wrath. This new technology is just another tool CL&P can use to gather more information to improve emergency preparedness and response," the company said.

In addition to the weather modeling, CL&P and UConn also are working together on Stormwise, a forest vegetation management program to reduce the likelihood that trees limbs will knock down power lines. CL&P has a cyclical tree trimming program that's on a four-year schedule, said Mike Zappone, manager of resource acquisition at CL&P.

There are some places that have just been trimmed and others yet to be trimmed on their cycle, Zappone said.

CL&P, UConn and the Connecticut Agricultural Experiment Station are working together to research and develop the best approach for growing different species depending on its location. Trees with adequate space to grow are healthier and more resistant to high winds. The tree management plan is using a "right tree, right place" approach, which incorporates the idea that some trees can be near to power lines and others are better off with more space to grow.