Which beach will bust? A high-tech flood forecaster predicts water’s next move like no one else

Written by 

rep nazari beach pc 550pxWhen Hurricane Michael slammed into Mexico Beach in October 2018, it wasn’t just the strongest hurricane to hit the Panhandle or the strongest October hurricane on record to land in the United States. Climate scientists consider it a warning of what is to come: Sea levels are rising and the impacts of global warming are becoming more pronounced. The asphalt jungle is spreading, too.

“Soil soaks up 80 to 100% of rainfall; on paved surfaces, it’s 0 to 10% and the velocity of water runoff increases, too,” said Rouzbeh Nazari, Ph.D., co-director of the UAB Sustainable Smart Cities Research Center. “We’re seeing flooding after less than an hour of rainfall in places where heavy rain used to be no big deal. A major disaster used to be thought of as an earthquake or landslide. Now flooding is front and center.”

That is cause for concern for the 123 million Americans who live in counties that directly touch the seashore. But the risks, even for those living right up to the water’s edge, are decidedly unequal. Nazari, an associate professor in the schools of Engineering and Public Health, has spent the past several years creating a system to map flood risks like never before. He started work soon after Superstorm Sandy devastated the New York and New Jersey coast in 2012. His model achieves two-inch resolution, compared with the 30-foot resolution of the flood maps produced by the Federal Emergency Management Agency (FEMA), which allows him to pinpoint flood dangers down to individual houses and plots of land.

Nazari brings up a composite satellite and drone image of a section of Brick Township, one of the largest cities in Ocean County, New Jersey. His mapping model has marked each house with a colored circle. Many of these dots are green, including ones fronting Barnegat Bay, which separates Brick from the Jersey shore. But several clusters of homes stand out as patchworks of yellow and red.

rep nazari neighborhood 1080Image courtesy Rouzbeh Nazari, Ph.D.

‘All these properties will be destroyed’

In a major storm, “all these properties will be destroyed,” Nazari said. This is more than intuition. First off, Barnegat Bay is like a bowl, he explained, with the ocean side “actually at higher elevation than the west side, where these houses are, so that is where the water will go.” That macro-level geography is just the beginning. Nazari used drones, light aircraft, laser-based LIDAR surveys, terabytes of data, hundreds of hours of computing time and thousands of hours of undergraduate effort to build his Q-Peak model and its hyper-detailed Property Risk Reports. Property elevation, underwater contours, wave action, location of utilities and a host of other factors play a role in his system — including whether or not a home has a basement. “You can see how the scenario changes on the ground at different levels of storm — a Category 1 versus a Category 4 hurricane,” Nazari said. “You see which areas go underwater first.”

Those details can be chilling. “Every time I show this, I get a tingle down my spine,” he said. “These are people’s homes. If a group of people on this street say, ‘Sandy wasn’t too bad, we think we’ll stay behind…’”

rep nazari mexico beach slide 800pxImage courtesy Rouzbeh Nazari, Ph.D.

What are the odds?

FEMA maps are the default standard used to determine if a property owner needs to buy flood insurance. “FEMA admits these are outdated, but lots of communities rely on them” for emergency planning, Nazari said. The Mexico Beach, Florida, community devastated in Hurricane Michael is a perfect example. In presentations, Nazari overlays a graphic of FEMA’s A, V and X zones on this once-booming Gulf-front town. The X zone is an area that is only supposed to flood once in every 500 years. You don’t need to buy flood insurance if you live in an X zone. But after Michael came through, “80% of the homes in the X zone were gone,” Nazari said. It isn’t FEMA’s job to provide hyper-accurate estimates of flooding for small areas, he noted. “FEMA is a national agency, providing information on a national or major regional level. They don’t have the capacity or time to look at Mexico Beach or downtown Birmingham or the community you live in. That has been left for other institutions to explore. We’re trying to fill that gap.”

House Fax

The goal isn’t to frighten, but to inform decisions before it’s too late. “Higher resolution lets you make better choices,” Nazari said. Based on his modeling work, he created a website, NJfloodalert.com, that offers information “on how to go from orange to yellow to green” — that is, from a 50% chance of total property loss to 5% or less.

Rouzbeh Nazari“We’re seeing flooding after less than an hour of rainfall in places where heavy rain used to be no big deal," said Rouzbeh Nazari, Ph.D. "A major disaster used to be thought of as an earthquake or landslide. Now flooding is front and center.”Homeowners can use this information to decide to buy flood insurance, build seawalls or take other preventive measures, for example. Banks and insurers can get a better idea of the risks involved in properties they are financing. And prospective homebuyers can avoid literally going underwater on a purchase. “You shouldn’t be able to sell a property for $10 million if it is going to be underwater in 15 years,” Nazari said. “This is an opportunity to pull back from these locations. You have services like Car Fax when you buy a used car. This is a House Fax.”

If that sounds like a good pitch for a business, you are on Nazari’s wavelength. The National Science Foundation has given him an iCorps grant to commercialize his model. “We can rate the resiliency of entire towns, specific zip codes and neighborhoods and down to individual houses,” Nazari said. Creating new maps can take anywhere from six months to a year, he explains. But the cost is far lower than an engineering firm would charge. “We have the luxury of access to armies of students who are willing to learn these methods as part of their education,” he said. “They get hands-on experience and a stipend, as well.” At Rowan University, Nazari’s previous institution, he would often have 30 to 35 undergraduates working as paid interns during the summers.


Get out of town

Nazari also hopes his map can remove another storm threat: massive traffic snarls that can be life-threatening in themselves. “Mass warning systems cause chaos,” Nazari said. “People are stuck on the roads. Then they think the storm has passed and come back too soon. The damage may look minor right after the storm passes by, but that’s because the flooding happens an hour or two later.”

One of Nazari’s latest research papers adds dynamic traffic simulation to his map model, working out the best evacuation routes from the Brick community after predicted flooding of local bridges and roadways.

“Traffic is the most critical aspect of emergency preparedness,” Nazari said. During Hurricane Rita in 2005, a line of stalled traffic 100 miles long trapped residents fleeing from Houston. Evacuees spent more than 20 hours in their cars and 100 people died on the roads. In Hurricane Matthew, which hit Florida, Georgia and the Carolinas in 2016, evacuees spent as long as 40 hours stuck on the roads.

With his model, “we are able to provide targeted warnings so police and emergency services can focus their efforts,” Nazari said. “If you only have 20 trucks to send out to help people evacuate, where do you send them? We can let a local government know, ‘Concentrate on the houses on this block and ask them to leave.’” Existing maps chart the social vulnerability of populations — noting schools and nursing homes, for example. Nazari combines that with structural information. “It will save a lot of money and avoid having everyone rush onto the roads at once.

“We need better maps and technologies and that’s where universities can do more,” Nazari said. “I tell companies, come and get the benefit of our work.”


Flood animation from Dr. Rouzbeh Nazari

Where has all the water gone?

Earth, wind and fire may get all the attention in the natural disaster realm, but water is nothing to dismiss.

Several years ago, Nazari introduced the concept of total water elevation. “Storm surge, sea level rise, high precipitation — we need to stop looking at these as single events,” he said. Often, these factors combine to create disasters greater than the sum of their parts. “They all feed into each other,” Nazari said.

Total water level takes into account storm surge, tides, wave runup, sea level rise and precipitation.

It only takes about six inches (0.15 meter) of water to reach the bottom of a passenger car according to FEMA research. At 0.3 meters, just under a foot, cars start to float, which is why most departments of transportation close roads to traffic after they hit 0.3 meters.

Water depth of two feet, flowing at three feet per second, will carry away the majority of vehicles.

That is what happened during the catastrophic flooding in Houston after Tropical Depression Imelda in September 2019, when the city was buried under three days of rain. “There were 48 inches of rainfall, more than Houston normally gets in a year,” Nazari said.

Eyes on trouble

Nazari’s specialty is remote sensing — “I’m a satellite guy,” he said — which means he’s constantly thinking of ways to use space-based and other imaging data.

In one ongoing project, he is using drones equipped with infrared sensors to monitor hidden dangers buried in landfills. “There are over 10,000 landfills around the country, many of them capped and no longer used,” Nazari said. This doesn’t mean they are just sitting around, however. “Internally they are going through changes.” Heat builds up and, out of sight under a layer of dirt, “they can start smoldering and destroy their protective membrane, leaking waste into the groundwater. The smoke comes out a year or two years later, but that’s too late.” His team has studied a notorious landfill in Missouri that has been burning for 10 years and has already consumed $200 million in remediation costs. “They wanted to know if the fire was heading for a nuclear waste facility next door,” he said. “Unfortunately, it is.”

Nazari isn’t always on disaster drill, though. Several years ago, while he was on the faculty at New York University in Brooklyn, he put a camera on the university’s CUSP Urban Observatory and pointed it north toward Manhattan. He tracked the status of every light source in the shot — thousands of individual office windows and apartments and stores. (The camera was intentionally blurred so that individuals or specific locations were not identifiable.) When he put all the data together, the same patterns recurred. Individual lights varied — someone who usually left the office at 5 p.m. stayed late some nights, and lights in apartments that usually went off at 9 p.m. sometimes stayed on until 2 a.m. But overall, “we saw the same patterns over and over,” he said.

Changes in these overall patterns could lead to surprising results. “Humans as individuals can’t be modeled, but as a whole they are very much modelable,” Nazari said. With one camera looking at a city you could predict a financial downturn or boom, he said. “If you see that a lot of commercial spaces have less light coming from them, that indicates they are going out of business,” he said.

At UAB, Nazari is looking to build collaborations with a wide range of colleagues. As he told a group at the School of Public Health recently, “I’m looking to work together to find new and interesting ways to study human behavior, disease and disaster.”