Monique Coleman’s basement was still wet with saltwater when the rallying began. Just days after Superstorm Sandy churned into the mid-Atlantic region, pushing a record-breaking surge into the country’s most densely populated corridor, the governor of New Jersey promised to put the sand back on the beaches.

The “build it back stronger” sentiment never resonated with Coleman, who lived not on the state’s iconic barrier islands but in a suburban tidal floodplain bisected by 12 lanes of interstate highway. Sandy was being billed as an unusual “Frankenstorm,” a one-in-500-year hurricane that also dropped feet of snow. But for Coleman and many residents of the Watson-Crampton neighborhood in Woodbridge Township, the disaster marked the third time their houses had been inundated by floodwaters in just three years. Taxed by the repetitive assault of hydrodynamic pressure, some foundations had collapsed.

As evacuees returned home for another round of sump pumps and mold, Coleman considered her options. Woodbridge sits in the pinched waist of New Jersey, where a network of rivers and creeks drain to the Raritan Bay and then to the Atlantic Ocean. She heard that the Army Corps of Engineers wouldn’t be coming to build a berm or tide gate; the area had recently been evaluated, and such costly protections seemed unlikely. Spurred by previous storms, Coleman had already learned a bit about the ecological history of her nearly 350-year-old township. She discovered that parts of her neighborhood, like many chunks of this region, were developed atop low-lying wetlands, which had been elevated with poorly draining “fill” back around the early 20th century. As Coleman researched more deeply, a bigger picture emerged. “I started to realize that, in a sense, we were victims of a system because we were living in a neighborhood that should have never been built,” she says.

Although she had flood insurance—her mortgage required it—Coleman knew that her premiums would soon go up, and she worried that her property value would go down. She and her husband liked their house, a prewar colonial. Best of all, it was affordable, a rare find in a town so close to New York City. Coleman had only discovered she would be living in a “special flood hazard area” once she was reading the closing paperwork in 2006. That made her nervous. She recalls her attorney waving it off by saying that at the rate we’re going, everyone in New Jersey will live in a floodplain. That might be true in spirit, as a future-looking thought experiment, but it was severely misleading given the circumstances. Desperate to move her family away from a block in Newark with increasing drug activity, Coleman signed away one type of risk for another.

For four uneventful years, the marsh near the bottom of her street was an attractive amenity, a place where her three young sons could play freely. Then the drainages that wrapped around her neighborhood like a wishbone were overwhelmed by a nor’easter in 2010. And by Hurricane Irene in 2011. And again, by Sandy, in 2012.

Land where Monique Coleman's house stood is now planted with saplings. The house to the left will go through the same process. Credit: Grant Delin

When federal recovery money started trickling into New Jersey after Sandy, Coleman learned that she could apply for an elevation grant. But raising her house on stilts seemed silly if her car and the road were still on the ground. During Irene, she had witnessed what happens during a storm surge. “The high tide rushes in, and water envelops the entire area in no time at all,” she says. “The street becomes a river within a river.” Coleman didn’t want to be “made whole,” in the parlance of disaster-recovery law, if it meant rebuilding in place. Her stress levels spiked every time it rained during high tide. She didn’t feel safe, physically or financially.

While commiserating with a neighbor, Coleman heard about a program called Blue Acres. Its premise struck her as radically sensible: The government would “buy out” her repeatedly flooded property at its prestorm value instead of paying to repair it yet again. Demolition crews would then knock down the house and remove other markers of human habitation. She would transfer the deed to the state, and redevelopment would be blocked, forever.

Compared with selling her house, this process seemed overwhelming. But even if she could find a willing buyer, how could she ethically transfer this vulnerability to someone else? “All of us who live in high-risk flood zones were taken advantage of somewhere along the line,” Coleman says. “This was a way to end that cycle.”

RETREATING FROM THE COASTS, in concept or practice, is not popular. Why would people abandon their community, the thinking goes, unless no better alternatives remained? To emergency responders, retreat is a form of flood mitigation. To environmental advocates, it’s ecological restoration. To resilience planners, it’s adaptation to climate change. Everyone agrees, however, that retreat sounds like defeat. It means admitting that humans have lost and that the water has won. “American political institutions, even our national mythology, are ill-suited to the indeterminacy and elasticity of nature,” wrote journalist Cornelia Dean nearly two decades ago in her book Against the Tide. “It would almost be un-American to concede ... that it is we who must adapt to the ocean, not the other way around.”

The U.S. has occasionally experimented with retreat on a tiny scale by offering voluntary buyouts to waterlogged families. The outcome is rarely promising. “Buyouts are extremely expensive, extremely disruptive, and many of the attempts have not gone well,” says Craig Fugate, former administrator of the Federal Emergency Management Agency (FEMA). They invoke fear among citizens in every political stratum, bringing to mind land grabs, racist resettlement projects, class warfare, and, depending on your ideology, either federal overreach or federal abandonment. Because they require coordination among politicians, homeowners, lawyers, engineers, banks, insurers and all levels of government, they are enormously complicated to execute, even poorly. At their worst, buyouts break up community support systems, entrench inequality and leave a checkerboard of blighted lots in their wake. At their best, they avoid these things and still displace people from their homes.

Tidally influenced wetlands abut the Watson-Crampton neighborhood in New Jersey. By 2050 it will likely be underwater. Credit: Grant Delin

Yet anyone who has looked at a map that forecasts sea-level rise can see that in low-lying neighborhoods exposed to the tides, some amount of retreat is inevitable. Regardless of how much and how quickly humans cut greenhouse gas emissions, climate change is already producing effects that cannot be reversed. Within a few decades, as saltwater begins to regularly block roads, kill wetlands, disrupt power supplies, bury popular beaches, undermine houses and turn common rainstorms into perilous floods, the most vulnerable pockets of coastal towns will become uninhabitable. As the National Oceanic and Atmospheric Administration has warned, “today’s flood is tomorrow’s high tide.”

Buyouts, however, are not designed for adapting to climate change. Past beneficiaries were almost exclusively riverine communities in the U.S.’s rural interior—people who lived too close to the overflowing Mississippi and Red rivers, for instance, were relocated nearby. The government didn’t even begin promoting buyouts as a form of disaster recovery until the 1990s, and since then, they have been conducted as one-off reactions to hurricanes. With multiple federal agencies involved, yet no one taking charge, “it’s amazing how much we’re still making this up as we go,” says Alex Greer, an expert in disaster science at Oklahoma State University. Until recently, retreating from the coasts was practically unheard of.

Superstorm Sandy changed that. The hurricane made sea-level rise, an abstract, future problem for far-flung places, manifest in the form of drowned subway lines and a roller coaster tossed into the waves. It communicated both the experience and evidence of future flooding in a way that probabilities never could. “It’s Global Warming, Stupid,” said the cover of Bloomberg Businessweek. Political leaders in New York State and New Jersey, sensing a tonal shift, realized they couldn’t just talk about rebuilding without also talking about resiliency, the rising buzzword of disaster preparedness.

Environmental types were also acknowledging that they could no longer fixate solely on the problem of carbon emissions. Rob Moore was executive director of the Environmental Advocates of New York back in 2012. “We didn’t want to talk about adaptation, because we saw it as a distraction from mitigating climate change,” he says. “But Sandy made it unavoidable.” A few months later Moore took a job at the Natural Resources Defense Council (NRDC) to work on how the country would cope with rising seas. Climate scientists who study the acceleration of sea-level rise felt a similar urgency, and many emerged from their silos to produce better projections. “Now the geophysical people are talking to the atmospheric people and to the economists and the sociologists,” says Robert E. Kopp, director of the Institute of Earth, Ocean, and Atmospheric Sciences at Rutgers University and a lead author on major climate reports. This interdisciplinary approach has led to localized forecasting. Instead of only one number—the global mean—we now know that sea-level rise will vary significantly from region to region.

As flooding worsens, a few massive seawalls will likely be built to protect densely populated economic centers, such as lower Manhattan. But there is only so much money, and time, for cement enclosures. Residents in places such as Tangier Island in Virginia and Isle de Jean Charles in Louisiana—and globally from Bangladesh to the Maldives to Senegal—are coping with the same reality as Coleman and her neighbors in Woodbridge Township: a wall isn’t coming to save them, and the floods are already here.

NEW JERSEY’S CHIEF OF LAND ACQUISITION clasps her hands on the approach to Bay Point, a peninsula that is disappearing into the Delaware Bay. This is the site of the state’s first beachfront buyout, a hard-won success for Fawn McGee. Of the 31 properties she has acquired here, most were recently demolished. Some of the houses were already long gone, leaving behind skeletons of whittled pilings repurposed by nesting ospreys. “It’s bittersweet,” says McGee from atop a mound of makeshift riprap, residents’ last-ditch attempt to stop the erosion. “Even when you can be excited that we bought all these homes, and now we’re going to restore the land, everybody is miserable that they had to leave.”

Fawn McGee, director of the Blue Acres Buyout Program, has purchased hundreds of houses in New Jersey, including this one. It will soon be demolished. Credit: Grant Delin

When it comes to the unsustainable development of the American coastline, New Jersey owns the honor of being the first and worst. It was here that the tempestuous beach environment, long feared and avoided until industrial times, was rebranded as a summer vacationland. Atlantic City and Cape May were tourist destinations by the mid-1800s, escapes from the undignified swelter and infectious diseases of Philadelphia. The Lenape people had long done the same—a seasonal migration to the shore—but they came to fish, not to conquer the sand with an arsenal of hotels and boardwalks. To hold everything in place, New Jersey was the first state to try to wrest control of coastal sediment flows from nature by erecting seawalls and jetties and bulkheads, and today virtually none of its coastline is untouched by human intervention. It’s no surprise that the first speculator to see dollar signs on the sandbar that became Miami Beach was a New Jerseyan.

For the communities McGee works with, retreat has gone from a strategy of last resort to the only option left in the span of about a decade. Climate change drove that shift. But the reason people are in this predicament in the first place is because of the unchecked hubris of coastal engineering, coupled with general human tendencies to love the water and ignore the future. The government gave a huge boost to coastal development in 1968 by introducing the National Flood Insurance Program—an exemplar of moral hazard that allowed homeowners to rebuild over and over in risky areas while keeping their premiums artificially low. Fifty years later that program has accumulated $36.5 billion of debt while effectively trapping people who might prefer to escape to higher ground, NRDC’s Moore explains.

Coastal systems, by their tidal nature, are always dynamic and occasionally volatile. The harder we tried to make them stay put, the less stable they became. In the 1960s, when scientists discovered that beaches armored with hard structures actually eroded faster and recovered slower than natural ones, the Army Corps of Engineers began dredging sand from the continental shelf and pumping it back to the shore. Before long, a storm would wash it away, and the dredging would begin anew. Today the Eastern seaboard has a Sisyphean dependence on the “nourishment” cycle. As sea-level rise rapidly accelerates, beach towns are increasingly desperate for fresh infusions of sand, which the corps must travel farther offshore to find. Geologists warn that we are running out of usable sediment faster than the planet can replace it. Wealthy homeowners in Florida are now stealing sand from public beaches to fill in their private ones.

Just as New Jersey pioneered the quixotic development of the coasts, it is leading an acceptance of what it wrought. Twenty-three years ago the state’s Department of Environmental Protection (NJDEP) launched the Blue Acres Buyout Program, using state money to purchase a handful of routinely flooded houses here and there. McGee, who runs the program now, was its first leader to leverage federal money from FEMA and other agencies, turning Blue Acres into one of the country’s very few standing buyout departments to deal with rising tides. Florida and Louisiana don’t even have their own versions. Blue Acres, McGee says, has three major goals: permanently move people and property out of harm’s way; open the land for public access; and restore the natural ecology so that the ground becomes a sponge, mitigating flood risk for the rest of the community.

After Superstorm Sandy, McGee prioritized buyout eligibility for entire clusters of houses over individual ones. In that way, the open space created would be big enough to have a real impact on managing floods. She looked for towns based on where enough homeowners had expressed interest in the process and the local government was willing to part with a portion of its tax base. After all, when the houses disappear, so, too, goes the tax contribution. Homeowner participation is voluntary, with an ability to pull out at any stage before signing the final contract. That means that families in a “buyout zone” will ultimately have to make a profoundly personal decision about whether or not to uproot in the context of both their neighbors’ choices and their town’s vision for the future.

The year before Superstorm Sandy, McGee tried pulling off her first large round of buyouts in the aftermath of Hurricane Irene. It didn’t go well. “On paper, buyouts make a world of sense, but in practice, that is absolutely not true,” Greer, the disaster scientist, says. McGee scrutinized how the process got disjointed and stalled out, making it nearly impossible for anyone to make decisions. It took more than a year for the funding to align, leaving desperate homeowners in a lurch. Some people were upside down on their mortgages, which disqualified them from the program. Or they couldn’t afford to keep living in temporary housing while waiting for answers. As disaster amnesia began to set in, many decided just to pursue the readily available methods of rebuilding or elevating. “I realized this went way beyond land acquisition,” McGee says. She tasked her tiny Blue Acres team with anticipating the adjacent challenges. “When Sandy hit and the big federal money came in, we were ready,” she says.

But advocating for large buyouts of clustered properties meant McGee was pushing for something novel at a moment when New Jersey was in triage mode. The state administration wanted to recover as quickly as possible. “I think they looked at me as a troublemaker for challenging us to think beyond black-and-white,” she says. McGee was arguing to disrupt the cycle of hasty rebuilding in areas where she knew the water would rise again. After many persistent meetings (her approach was “confidence, not confrontation,” she says) and assurances that she could navigate through the bureaucracy, McGee convinced the state to give her plan the go-ahead.

Immediately McGee asked to borrow 33 staffers from other areas of the NJDEP. “In addition to my GIS [geographical information system] people, I brought in four paralegals, six project managers to do grant writing and deal with the feds, and eight case managers to walk families through the process,” she says. “Then title people. Appraisers. Surveyors. Accountants. We had bookkeepers just to go through shoeboxes of Home Depot receipts from homeowners and cross off things like Snickers bar, water, couch.” With each snag that threatened the buyouts, McGee found herself filling some unexpected role. She played a therapist to storm victims experiencing trauma, illness and divorce. When McGee saw how many homeowners were upside down on their mortgages, she facilitated short sales that amounted to more than $5.4 million of debt forgiveness—which resulted in 67 additional bought-out homes. When the issues went beyond her scope, she outsourced. “Hoarders! They literally can’t get out of their houses,” she says. “We had to bring in specialists to help them let go of their stuff.”

Most important, McGee sensed that buyouts were a deeply communal decision. She figured neighbors would be looking at their options together over a bottle of wine, so she assigned case managers according to social groups. Case studies of buyouts show that individual considerations play only a limited role in whether a family leaves or stays. “The implication here is that residents’ perceptions of their broader community were more influential in the buyout decision than were their experiences of Sandy, or their personal or family characteristics,” wrote Sherri Brokopp Binder, an expert on postdisaster relocation, and her colleagues in a 2015 study in the American Journal of Community Psychology. McGee realized that grassroots support from a few residents could make all the difference between a transformative buyout and a dud.

IT DIDN’T TAKE LONG after Sandy for Monique Coleman to decide that she wanted a buyout. Now she would have to convince her neighbors and the town to want the same. McGee remembers meeting Coleman in 2013, when she hosted a standing-room-only meeting in Woodbridge to tell the residents about Blue Acres. “People would listen to me but then look over at Monique for her to signal it was okay,” McGee says. “We always have police at these meetings because people get heated, but Monique had this calming effect on her neighbors. It was astonishing.”

Coleman, it turned out, had exposure to grassroots organizing dating to her childhood. Her grandmother took her to Black Liberation Movement meetings, and her father brought her door-to-door during his city council run. These days Coleman works as a teacher for blind children, so in a sense her profession is helping others adapt to unfamiliar challenges. When Coleman and a handful of buyout supporters began canvassing their own streets in late 2012, she expected resistance. “You can’t present this idea once, and when you hear that ‘no,’ just say, ‘Okay, I understand,’” she says. She set up a blog and a Facebook group and organized monthly meetings. It was going to take a lot of listening at a time when people were traumatized by Sandy and overwhelmed by the day-to-day choices of recovery.

Blue Acres has been touted as a model for how retreat might evolve. Roy Wright, former resilience lead at FEMA, called New Jersey’s approach to buyouts “masterful.” But in all McGee’s meetings and maps and her talks with mayors and residents, she was not able to evoke the language or data of climate change. Until January 2018, under the administration of Governor Chris Christie, McGee “was not to use those terms,” as she puts it.

So Coleman took it on herself to “learn the scientific side of things.” When she started doing Web searches for the impacts of climate change in 2012, she couldn’t find much that spoke to her area specifically—it simply didn’t exist yet. But what she did find convinced her that the back-to-back floods she had experienced in Woodbridge weren’t a spate of bad luck but signals of a new reality. Coleman wrote and distributed a leaflet of 12 reasons why life in the flood zone wasn’t sustainable in a climate-changed future.

Eventually some of her neighbors stopped slamming their doors when she showed up to chat about buyouts and started asking her questions instead. Coleman was patient but persistent. “It is very hard for people to accept this is really happening,” she says. She explained that no matter how she sliced it, she found no positive paths forward. Leaving for good was simply the least onerous. “Who knows when a flood will come again,” she would tell her neighbors. “But it will be coming.”

THE MODELS SCIENTISTS HAVE MADE to predict the influence of sea-level rise on those future floods have become highly sophisticated, combining global factors such as the thermal expansion of the seas with local variables such as land subsidence and variations in the gravitational pull of land on the ocean around it. But major uncertainties remain. For one thing, we don’t know how quickly and severely societies will cut greenhouse gas emissions. For the next few decades certain effects will occur regardless of how much we mitigate climate change. Rutgers’s Kopp, a leading climate scientist, says that New Jersey will likely experience between one and 1.8 feet by 2050. Even at the low end, numbers like that will reshape life along the coast. After 2050 the rise continues to accelerate, but the picture gets murky: NOAA estimates that New Jersey could see between three and 12 feet of sea-level rise by 2100. That range is overwhelming if you’re a mayor who is trying to come up with an adaptation strategy. “Climate change is a study of probabilities, but the public wants yes or no answers,” says Graham Worthy, director of the National Center for Integrated Coastal Research.

Besides the human element, however, the biggest wild card when it comes to the fate of coastal communities in New Jersey and beyond is the stability of the West Antarctic ice sheet. The rate at which the whole of Antarctica is shedding ice has tripled over the past decade, and West Antarctica, being especially sensitive to the forces of climate change, is one of the most rapidly changing places on the planet. The West Antarctic ice sheet is so voluminous that it will add more than 10 feet of sea-level rise alone if it catastrophically collapses. This scenario might only be avoided with extreme emissions cutbacks in the next decade, according to a June 2018 report in Nature.

NASA’s satellites have been collecting data on this ice sheet and others for more than four decades. Satellites are unrivaled at capturing continuous observations over a wide area, but they can’t pick up details that would make it possible to predict the ice sheet’s fate with a higher degree of certainty. Some of those more granular clues include ice thickness, the grounding line where a glacier’s base meets the sea and the slope of the ice sheet, which is a driving force that sends ice from the interior of the continent to the ocean, says Kenneth Jezek, a glaciologist and retired polar researcher. Because of its sheer size and remoteness, studying West Antarctica up close is a dangerous, logistical nightmare.

One survey project, a NASA airborne mission called Operation IceBridge, has been able to capture some of those details by flying above the region in a retrofitted jetliner. From an altitude of just 1,500 feet, scientists onboard this winged laboratory can see that the top of the ice sheet is textured with signs of movement, such as geometric crevasses, the milky cerulean of ancient ice exposed to sunlight and cracked-up plains that resemble lake beds in drought. These features, they know, are carved by katabatic winds from above and by invisible rivers from below. But it is the ridges of black bedrock that hint at dramatic topography underneath the ice—a hidden world that IceBridge has been mapping extensively over the past decade.

To understand what is happening under the frozen surface, pilots maintain a precise track over the sheet while radar streams data to an onboard computer screen, revealing evidence of entire mountain ranges and valleys that make up the shape of the continent. A gravimeter picks up the depth and size of glacial cavities filled with seawater, a marker of how floating ice shelves might be melting as they interact with the ocean. Along the glaciers’ terminal edges, icebergs float against the inkiness of the Amundsen Sea, a scene photographed every second by two cameras affixed to the belly of the plane. IceBridge has flown some of these tracks year after year, capturing change in unprecedented detail. At a time when the necessity of earth science is being attacked by political leaders, “I can’t emphasize enough that we do not collect this data because we find it scientifically interesting,” says John Sonntag, IceBridge’s mission scientist. “We collect it to try to warn and protect our communities from sea-level changes that are coming their way.”

As raw data from IceBridge, satellites and similar projects have percolated through scientific papers and reports such as the 2017 National Climate Assessment, new tools have emerged. NOAA’s Digital Coast and Climate Central’s Surging Seas, for instance, allows town planners to begin envisioning how sea-level rise will affect flooding in their jurisdictions.

IceBridge data have turned out to be essential for filling in fundamental gaps in polar ice knowledge, “but we still have a way to go in Antarctica,” says Eric Rignot of the University of California, Irvine, and NASA’s Jet Propulsion Laboratory. Rignot was the lead author on a landmark 2014 paper in Geophysical Research Letters that used radar measurements of a large sector of West Antarctica and concluded that it “is undergoing a marine ice sheet instability that will significantly contribute to sea level rise in decades to centuries to come.” That same week a Science paper suggested that, based on modeling, the collapse of the West Antarctic ice sheet had already begun, making extreme sea-level rise inevitable, possibly within two centuries. But Rignot thinks that time line might be too conservative. Real observations of ocean temperature—and how glaciers are responding to those warmer waters—are still “totally lacking” in parts of Antarctica, meaning that “it is a matter of fact that our projections tend to underestimate sea-level changes,” he says.

Neither Rignot nor Jezek thinks that the next NASA ice-monitoring satellite, which launched on September 15, will be enough to narrow the uncertainties coming out of Antarctica. Rignot suggests that it will take more airborne research such as IceBridge, along with shipborne surveys by unmanned subs and multibeam sonars and new forms of intelligent technology—“an entire army of robotic devices”—dispatched to the remotest fringes of the continent.

Next up, science teams led by the U.S. and U.K. will travel by air and icebreaker to West Antarctica’s marquee feature—the massive and infamously unstable Thwaites glacier—to do just that. More than 100 scientists from around the world will study the interaction between warming ocean water and the ice shelf to examine how Thwaites is thinning from below. Thwaites is like a tub stopper holding much of West Antarctica in place; if it’s doomed, so is the ice shelf. The more these researchers and others learn about the shifting dynamics among ice, ocean water and atmosphere, the more factors they can plug into regionally specific sea-level predictions. The data they gather will inform whether coastal populations have centuries, or mere decades, to prepare for the onset of the deluge.

BY THE SUMMER OF 2014, when Coleman signed over the deed to her old house and moved into a new one, Woodbridge Township was on its way to becoming the site of Blue Acres’ biggest ever buyout project. Today a total of 142 homeowners have accepted offers. And about 115 homes from the Watson-Crampton neighborhood alone have been removed, most of them clustered within a grid of streets covering about 30 acres. Millions of people globally will have to move inland to escape the coming floods, so these numbers can sound too tiny to be meaningful. But what happened in Woodbridge upends many of the assumptions traditionally tied to buyouts: that no residents want to leave, that politicians will never get onboard, that ecological health in suburbia will never win out over real estate growth and that no one is planning for a climate-changed future by making painful choices in the present. “What we’re doing here is paving the way for conversations about how to manage retreat,” says Thomas C. Flynn, the town’s floodplain manager.

As houses come down, Thomas C. Flynn, Brooke Maslo and Jeremiah Bergstrom are restoring floodplains in Woodbridge Township, N.J. Credit: Grant Delin

Officially the buyout process ends once the property has been demolished. Woodbridge, however, found itself with an abundance of lots and ambitions beyond mowing the grass. The town reached out to Rutgers ecologist Brooke Maslo, who works with the school’s Cooperative Extension to assist New Jersey communities with science-based projects. The term “resilience” gets tossed around a lot, “but what does it actually translate to?” Maslo asks. She came to see the Watson-Crampton buyout project as a unique opportunity: she could create a floodplain restoration that buffered the remaining neighborhood from sea-level rise. She brought in Jeremiah Bergstrom, a landscape architect with experience managing stormwater in urban environments. “As far as I can tell, this is the first coastal land restoration in the context of residential retreat,” Bergstrom says.

Using nature as infrastructure is a well-established concept—think mangroves and oyster beds as storm-surge absorption—but it is not commonly applied in places as densely developed as the greater New York City area. Liz Koslov, an assistant professor at the University of California, Los Angeles, who did ethnographic research on Staten Island’s post-Sandy buyouts, says she has seen next to no discussion about what happens to the land itself after the houses come down. “Residents said they just want the land to go ‘back to nature,’ but when you get down to it, ‘nature’ can mean a lot of different things,” she says. Karen O’Neill, a Rutgers sociologist who is cataloguing global instances of coastal retreat, says that “you hardly ever see a comprehensive ecological restoration. It just doesn’t exist.”

The Watson-Crampton neighborhood can’t simply return to nature, because it was built on fill. “We have to re-create a new ecology, a new nature,” Bergstrom says. Over the past year the restoration team has ripped out roads, assessed soil quality, and planted more than 950 saplings to increase flood storage capacity and encourage the growth of a biodiverse salt-marsh habitat. Without intervention, the land would become a monoculture of invasive reeds that can break down and form dense mats, which might ultimately make flooding issues worse, Flynn explains. Maslo and her team are softening the hard curvature of the swale so that tidal surges entering from the river below the New Jersey Turnpike don’t rush in at high velocity. They’ll carve a channel that will allow for spillover, with the hope that the depressions will create permanent standing water for wildlife. Maslo wants to prove that a town can recoup its tax losses with new lures, such as parkland trails and a kayak launch. “This is not a wasteland,” she says.

Maslo’s vision helped to convince the Woodbridge mayor and city councillors to change the township’s buyout regions—120 acres in all—from residential zoning and existing marshland to something they named the Open Space Conservation/Resiliency Zone. No development would be allowed. Nineteen households in the Watson-Crampton buyout area dug in their heels. The town warned these “holdouts” that if they ever wanted to sell, their houses would first need to meet new floodplain standards—which would likely mean elevating them higher. Land-use changes like this are controversial because they make what is supposed to be a voluntary process into one that is significantly less so. But without them, developers might be attracted to come in and build bigger and higher properties. Then new people—those who can afford higher flood insurance premiums and the building costs of living in a floodplain—are likely to move in, replacing those who can no longer afford to stay.

COLEMAN RELINQUISHED HER HOME to the forces of nature. Four years later, in 2018, she says she has no regrets about taking the buyout. For moving within the same county, she received an incentive grant of $10,000, which helped her afford another single-family home on higher ground. The process was financially and emotionally stressful, but the way Coleman describes her participation reframes a reaction to misfortune as a deliberate act. In an era of climate refugees who will be forcibly cast out of their homes by either too much water or not enough, Coleman sees herself as more of a retreat pioneer—someone who seized whatever agency she could as she faced an uncertain future. “There’s nothing worse than sticking your head in the sand and resisting all this change going on around you,” she says. “Because then you end up feeling pushed to make a decision that you are not prepared to make.”

McGee, meanwhile, is playing the long game. In the spring of 2018, five and a half years after Sandy, Blue Acres was still submitting new buyout applications to the federal government. Woodbridge is on its third round of buyouts, which involves seven of the 19 holdouts that remain in the resiliency zone. “I don’t close out a grant until we’ve done enough demolition so that the holdouts can digest how the character of their community is changing,” McGee says. The tactic is working. She has spent $172 million of her total funding pot, which has grown as other recovery programs failed and FEMA directed the unused money her way. Blue Acres has facilitated nearly 1,000 buyout offers since Sandy, of which 713 have been accepted by homeowners. “You think it’d be 10,000 families for all the work we’ve done, but it’s not, because it’s so damn hard,” McGee says.

Woodbridge’s resiliency zone is not quite ready to be held up as a demonstration project. The work isn’t done, and the marsh isn’t yet beautified. Over the years, as the houses came down in stages, the scene sometimes looked like an eerie abandonment of the built environment, not a harbinger of progressive adaptation. But now that nearly all the structures are gone and the outlines of formerly paved roads are blurring into grass, “it looks less like a ghost town and more like it’s just land,” says Coleman, who visits every few months to see the transition in progress. “Now it’s the houses that look like they don’t belong.”

No one disagrees that undeveloping certain areas of the coastline will be messy and expensive. But as the science of coastal resilience becomes more collaborative, the how-to of retreat may become less daunting. Retreat, after all, will not mean drawing a line some distance inland from Maine to Florida and removing everything to the east, explains Bryan Jones, a geographer who models climate-induced human migration. Modeling combined with artificial intelligence is now producing tools that allow planners to play out what-if scenarios in their towns. If, say, you buy out 40 houses from one location, restore groundwater storage and run the 100-year flood of the future, would that significantly reduce damage to adjacent houses? What are the social and economic trade-offs of undeveloping one neighborhood to protect another? Can land be designated as a safe relocation spot? These are the kinds of questions that Fugate, who led FEMA during Sandy, and others are working on now. “Just as the quantification of catastrophe risk drove a huge expansion in catastrophe insurance, it is about to drive a great industry of disaster risk reduction,” wrote Robert Muir-Wood in his 2016 book The Cure for Catastrophe.

Retreat is so new that few planners are thinking about the next step: relocation. “Globally, there’s substantial evidence that people end up right back in harm’s way,” Jones says. In a survey of Staten Island families who took buyouts after Superstorm Sandy, Binder, the sociologist, found that 20 percent moved to a home that is equally or more vulnerable to flood risk. As more people begin to flee slightly inland, they will encounter a wave of people still moving toward the coast. Just like backwash hitting the surf, the result could be turbulent.

As U.C.L.A.’s Koslov wrote in 2016 in Public Culture, “the complexity and ambivalence of retreat serves as a reminder that there are no easy solutions and that it is not possible to rebuild forever or to wall ourselves off from the problems we face.” Retreat signals not just the physical movement of recalibrating to the tides but an existential reckoning with our ways of living along the water. The word itself is borrowed from the language of geologic processes, which humans have undeniably hastened. As glaciers and beaches retreat, so, too, will we.