Assessing Hurricane Sandy's Damage to the New Jersey Coast

Seaside Heights. Photo: Google/NOAA

After Hurricane Gloria damaged the New Jersey coast in 1985, the Federal Emergency Management Agency (FEMA) asked the state’s Department of Environmental Protection to estimate the coastal effects of the storm. The state agency wasn’t sure, because they didn’t have an accurate baseline. “I mean, they had some Polaroid pictures, and that was it,” says coastal researcher Dr. Stewart Farrell. “They had no data, no surveys, no map, no nothing.”

Farrell put together a proposal to start surveying the coast, got the go ahead, and founded the Coastal Research Center at the Richard Stockton College of New Jersey in 1986. He has conducted coastal surveys every spring and fall since. By late October, he was nearing the end of the 2012 fall monitoring. “But Sandy hit,” he says. “Now we’re going back and seeing how much dune and beach are missing, because somebody’s gotta come up with a number for how many cubic yards of sand we’re going to need to fix things.”

Every morning since the hurricane struck, Farrell has driven and walked the coast to survey the damage. That includes weekends. So far, he’s captured Atlantic, Cape May, and Ocean counties. His groundwork will be combined with aerial surveys and computer models that offer a fuller sense of the damage, but even right now the effects of the storm are clear. “It’s the worst event in my career, which goes back to the 1960s,” he says.

We called up Farrell this past Friday afternoon, after he returned from surveying damage in the borough of Avalon, to find out more.

Do you have any idea what percentage of coastline has been damaged or eroded?
Well, it’s all still there. It’s just been rearranged. The sand hasn’t gone anywhere. It’s just not where we want it. A lot of it, in some places, was washed inland on to the barrier islands, and that’s going to be tough to get back, because it’s in among the houses and bushes and trees and everything. They’re moving it off the streets and the roads and putting it back on the beach, but that’s not going to put it all back by a long shot. Some of it is offshore and is forming into bars that will eventually move back onto the shoreline. They always do. That’s not going to fix the dunes. The dunes take decades to build naturally. So we—that’s a collective we—are going to have to bring in sand from somewhere and rebuild a dune system.

Seabright before and after. Photo: Google/NOAA

How does a rebuilt dune system compare to a natural dune system as a buffer to storms?
They’re both basically piles of sand. Most of them were manmade anyway. The natural dune process has not been ... if you go to Island Beach State park, those are natural dunes. The only difference is that the plant root system is all through the dune so they are kind of wired together better than a manmade dune.

How has you work evolved over the years?
We work closely with the Bureau of Coastal of Engineering in Tom’s River. As a result of the work we do with them, the effect has been to work with the federal government for beach nourishment projects. In 1994 the state passed what they called the Stable Coastal Funding Act. In the last decade or so they’ve funded $25 million a year for coastal projects. They use it as leverage to acquire federal coastal projects through the Army Corps of Engineers. And the Army Corps of Engineers will undertake a project for shore protection—it’s what they call them; they’re not called beach nourishment, they’re called shore protection projects—if the benefits of doing it outweigh the costs by 1.25. So if you divide the benefits by the cost, and the resulting fraction is 1.25 or greater, they’ll agree to do it. The benefits are essentially damage reduction. So they have done a fairly good degree of projects. Between the state and the feds, 55 percent of New Jersey’s 97 miles of shoreline had a beach project since 1986. All of this was primarily driven by the fact that New Jersey is the only state that has a stable funding law providing $25 million annually.

So the reason for 55 percent of the state’s 97 miles to get this protection is because of the development? By nourishing, or protecting, the beaches you're limiting the cost of damages. Is that the way to look at it?
That’s the way it’s looked at. They take each and every segment of the New Jersey coast—by the Army Corps of Engineers—in terms of historical storm damage, development, cost, and potential for future storm damage and they come up with a design for a project. Then it’s up to the state and the locals to partner with the feds to pay the other 35 percent and acquire the rights to the shoreline for the real estate issues.

Now, one of the reasons it was never done for some regions is because the individual property owners own to the high tide line. Getting all of them on the same train looked to be like herding cats. You couldn’t do it, and the Army Corps was not about to do the project without easements. So that’s why they piecemealed projects in places like Long Beach Island. The areas that were done did not suffer wave damage.

Can you describe the economics of nourishing or protecting these beaches? How much does it cost to do that every year?
Take, for example, Long Beach Island. The whole island, 18 miles of it, received a $79 million Corps estimate. They did about five miles of it.  They did Brant Beach, Surf City, and Harvey Cedars. Not counting the money they spent sieving the sand in Surf City, they spent about $20 million, so that’s about $4 million a mile. Since there was no damage to the homes, they probably did a pretty good job.

A lot of pictures in the media are showing damaged home or flooded areas.
Well, there’s a big difference. Flooded areas don’t count. That’s going to happen no matter what happens with beach projects.

What caused sand erosion to be the worst situation on the coastline since you started working in the '60s?
Well, north of the eye—that would be Long Beach Island, Ocean County, Northern Ocean County, and Monmouth County, New York, of course, Long Island—was more intense. It was less intense south of the eye, in Atlantic County, Cape May County. So the intensity factor was more: higher storm surge, bigger waves, more of them.

Can you describe the storm surge and the power at the some of those beaches?
The buoys offshore said 25 feet. Now, they’re smaller by the time they get to shore, but the storm surge was in excess of 12 feet. Some places in the New York area were 14 feet. The previous record was 10’6”. Once they broke they were probably about six feet high. They came rolling up and they did not treat the buildings well. The buildings got one every 10 seconds. So for six hours you’ve got a wave hitting a house every 10 seconds, leaving not much left of the house. The houses built on pilings generally did much better. There are a lot of them these days. The houses built on a foundation, or piers, or footings? Destroyed or seriously damaged. So how the house was built has a lot to do with it. Now, those are going to be replaced, and I’m sure they’ll all be on pilings.

Everything around the buildings, of course, has been wiped away. You need an extension ladder to get to the front door now. So they’re not terribly useful, but they’re still there.

Belmar before and after. Photo: Google/NOAA

What do you think the major lessons have been from Sandy?
I have no idea. Some people are going to throw in the towel. Some of them are going to build tougher and stronger. And some of them are just going to try to replace what they have. I think that the state is going to have a few more regulations: on how high the decks are above sea level, where the house can be placed relative to the shoreline, and these beach nourishment projects may get extended statewide, regardless of the property ownership. They could decide to take the beachfront lot and say, "It's public now, sorry." I haven’t heard anything about that, and it’s definitely going to be a volatile issue if it is brought up. It’s property rights versus public access, and it’s been going on for decades.

Do you think all of this beach nourishment is a good idea?
Well, where it was not done you had virtually total destruction. Where it was done the physical action from the wave damage was far, far less. I mean, Atlantic City survived. No damage to the boardwalk on the oceanfront. They squeeked by, but they still survived. The dune was just high enough. Avalon, Stone Harbor, no damage. Strathmere, no damage. Cape May City, no damage. Harvey Cedars, no damage. Holgate, lots of damage. So there you go.

Is there another option? Some people are calling for a reassessment of how much building is done on the coast, near the coast, considering the amount of damage.
Well, that’s been suggested by many people. The biggest proponent of that was a professor down at Duke, Dr. Orrin Pilkey. He has since retired, but is still around. He was a big proponent of 'Get out of the barrier islands.' Ain’t happening.

Why do you think that won’t happen?
It’s where the money is. People like the beach, and property ownership at the beach is not going to be made illegal. So managing it, regulating it, is something that gets more stringent after Sandy. But I do not see any way—I mean ... look, New Jersey earns $36 billion a year off of it. So the sales tax is pretty impressive. The jobs. There are 400,000 jobs related to the coast. Fifty million people live within an hour of it, and it’s still there. You can go swimming anywhere in New Jersey if you choose to, no problem. The water’s still wet, still salty, still there. It was a beautiful day at the beach today—just a little chilly to be strutting around in your bikini. But definitely still a gorgeous day at the beach, and everything that’s broken can be fixed.

Mantoloking before and after. Photo: Google/NOAA

And how long do you think it’s going to take to fix things based on what you’ve seen?
Two, three years. I mean, you got a quarter of the houses in Mantoloking destroyed. They’re all 3,000- to 4,000-square-foot homes. Some of them are 100 years old. So there’s going to be some time. There’s just not enough tradesmen around to build them all. You’re going to see the egregious evidence go away in six months—the debris, the sand in the street, the other stuff—but the damaged homes are going to take a while to fix. There are only so many guys that can fix them. I mean, they’re going to be coming from the Midwest. The pick-up trucks are going to be everywhere.

Things weren’t normal after the 1962 Northeast storm until 1968, and by then they were beginning to forget about it. Properties changed hands. Right now, you could probably pick up an oceanfront property for maybe a million two. Maybe less, I don’t know. I haven’t seen any sales figures, but that’s what happened last time. You could buy an oceanfront lot for $5,000 in 1962 in May. And that investment, had you held it, would be $4 million—until last week.

So how much of this short- versus long-term thinking?
Well, long term we’re going to have to eventually move landward and upward, because sea level is rising. And so if sea level rises two more feet, you are going to have to go to the house at low tide, unless they raise everything up, but that’s going to take a lot of dirt.

This is the fourth in a series of interviews on the effects of Sandy.
Part 1: Hurricane Researcher Brian McNoldy on the Science Behind Sandy
Part 2: Andrew Revkin on the Lessons Learned From Sandy
Part 3: Dr. Ralph Ternier Talks From Haiti About Sandy and Cholera

—Joe Spring

Filed To: Science / Nature
More Adventure
Pinterest Icon