When it comes to hurricanes, the adage, “it only takes one” to be catastrophic, has proven true time and again. For example, the National Oceanic and Atmoshperic Administration (NOAA) reported last year’s Atlantic tropical storm season as below-average with just 14 named storms and only three that made landfall in the U.S. But one of those storms, Hurricane Ian, which made landfall in southern Florida on September 28, became the third costliest weather disaster in U.S. history. Preplanning and positioning response crews are critical to reducing outage duration when storms like Hurricane Ian occur. The challenge is determining how many outages for specific service areas and the severity of the storm impact to right-size response teams several days before landfall. A new utility solution from DTN was tested during Hurricane Ian with promising results that demonstrated using data-driven intelligence to plan for outages before a storm.
The challenge of predicting Hurricane Ian
Ian started as a tropical wave off Africa's west coast, then it moved across the Atlantic, reaching the southeastern Caribbean. It briefly reached Category 5 status before it finally made landfall in southern Florida as a high-end Category 4 hurricane with maximum sustained winds were around 145 mph.
Hurricane Ian's predicted forecast track was expected to be parallel to the western coastline of Florida, but as DTN Risk Communicator Andrew Polk noted, a minor fluctuation in the forecast track of a tropical system can lead to significant changes to the storm impacts on businesses, emergency crews, service providers, and residents preparing for a hurricane.
“While the predicted track for Hurricane Ian within 24 hours was better than average,” Polk said, “the 36 through 72-hour forecasts were slightly below average further adding to the difficult decisions in hurricane preparedness.”
The storm produced catastrophic storm surge, damaging winds, and historic freshwater flooding across much of central and northern Florida. More than 2 million customers were without power that first evening and mandatory curfews were issued for communities along Florida’s west coast. After briefly moving offshore in northeastern Florida and downgrading to a tropical storm, Hurricane Ian quickly regained power and made a second landfall in South Carolina until it entirely dissipated days later in North Carolina. NOAA concluded Hurricane Ian was responsible for over 150 direct and indirect deaths and over $112 billion in damage.
Outage intelligence before the storm
To help incident commanders make more informed and confident decisions about outage planning and response, DTN recently developed the utililty solution Storm Risk Analytics. It was tested during Hurricane Ian with outstanding results that demonstrated the value of accessing data-driven insights before the storm. Nearly two days before landfall, DTN Storm Risk Analytics predicted nearly 4.6 million customers would experience outages during a week-long window when the storm was expected to be over land. While that forecast was within 7% of the actual outage count of 4.29 million customers, a revised prediction a day before landfall reflected the hurricane’s changing intensity and track. New, real-time information was modeled with updated predictions delivered every six hours and came within 3% of 4 million actual outages.
This test around Hurricane Ian combined advanced weather intelligence and machine learning outage prediction and demonstrated the possibilities for making incident command and storm impact decisions more confidently before, during, and after extreme weather events. The solution combines seven years of verified, historical outage and weather data with advanced weather and machine learning models based on a geographical region.
Extreme weather events, like Hurricane Ian, will always challenge the response and restoration services of utilities. With technology solutions that leverage weather intelligence and sophisticated modeling, like Storm Risk Analytics, utilities now have the potential to better predict and respond to weather impacts on their service areas and their customers.
