How Hurricane Sandy Tamed the Bureaucracy

By Adam Parris

Remember Hurricane Irene? It pushed across New England in August 2011, leaving a trail of at least 45 deaths and $7 million in damages. But just over a year later, even before the last rural bridge had been rebuilt, Hurricane Sandy plowed into the New Jersey–New York coast, grabbing the national spotlight with its even greater toll of death and destruction. And once again, the region—and the nation—swung into rebuild mode.

Certainly, some rebuilding after such storms will always be necessary. However, this one-two punch underscored a pervasive and corrosive aspect of our society: We have rarely taken the time to reflect on how best to rebuild developed areas before the next crisis occurs, instead committing to a disaster-by-disaster approach to rebuilding.

Yet Sandy seems to have been enough of a shock to stimulate some creative thinking at both the federal and regional levels about how to break the cycle of response and recovery that developed communities have adopted as their default survival strategy. I have witnessed this firsthand as part of a team that designed a decision tool called the Sea Level Rise Tool for Sandy Recovery, to support not just recovery from Sandy but preparedness for future events. The story that has emerged from this experience may contain some useful lessons about how science and research can best support important social decisions about our built environment. Such lessons are likely to be of increasing importance as predicted climate change brings the inevitability of extreme weather events.

A story of cooperation

In the wake of Sandy, pressure mounted at all levels, from local to federal, to address one question: How would we rebuild? This question obviously has many dimensions, but one policy context cuts across them all. The National Flood Insurance Program provides information on flood risk that developers, property owners, and city and state governments are required to use in determining how to build and rebuild. Run by the Federal Emergency Management Agency (FEMA), the program provides information on the height of floodwaters, known as flood elevations, that can be used to delineate on a map where it is more or less risky to build. Flood elevations are calculated based on analysis of how water moves over land during storms of varying intensity, essentially comparing the expected elevation of the water surface to that of dry land. FEMA then uses this information to create flood insurance rate maps, and insurers use the maps to determine the cost of insurance in flood-prone areas. The cost of insurance and the risk of flooding are major factors for individuals and communities in determining how high to build structures and where to locate them to avoid serious damage during floods.

But here’s the challenge that our team faced after Sandy. The flood insurance program provided information on flood risk based only on conditions in past events, and not on conditions that may occur tomorrow. Yet coastlines are dynamic. Beaches, wetlands, and barrier islands all change in response to waves and tides. These natural features shift, even as the seawalls and levees that society builds to keep communities safe are designed to stay in place. In fact, seawalls and levees add to the complexity of the coastal environment and lead to new and different changes in coastal features. The U.S. Army Corps of Engineers implements major capital works, including flood protection and beach nourishment, to manage these dynamic features. The National Oceanic and Atmospheric Administration (NOAA) helps communities manage the coastal zone to preserve the amenities we have come to value on the coast: commerce, transportation, recreation, and healthy ecosystems, among others. And both agencies have long been doing research on another major factor of change for coastlines around the world: sea-level rise.

Any amount of sea-level rise, even an inch or two, increases the elevation of floodwaters for a given storm. Estimates of future sea-level rise are therefore a critical area of research. As Sandy approached, experts from NOAA and the Army Corps, other federal agencies, and several universities were completing a report synthesizing the state of the science on historic and future sea-level rise. The report, produced as part of a periodic updating of the National Climate Assessment, identified scenarios (plausible estimates) of global sea-level rise by the end of this century. Coupled with the best available flood elevations, the sea-level rise scenarios could help those responsible for planning and developing in coastal communities factor future risks into their decisions. This scenario-planning approach underscores a very practical element of risk management: If there’s a strong possibility of additional risk in the future, factor that into decisions today.

Few people would argue with taking steps to avoid future risk. But making this happen is not as easy as it sounds. FEMA has to gradually incorporate future flood risk information into the regulatory program even as the agency modernizes existing flood elevations and maps. The program dates back to 1968, and much of the information on flood elevations is well over 10 years old. We now have newer information on past events, more precise measurements on the elevation of land surfaces, and better understanding of how to model and map the behavior of floodwaters. We also have new technologies for providing the information via the Internet in a more visually compelling and user-specific manner. Flood elevations and flood insurance rate maps have to be updated for thousands of communities across the nation. When events like Sandy happen, FEMA issues “advisory” flood elevations to provide updated and improved information to the affected areas even if the regulatory maps are not finalized. However, neither the updated maps nor the advisory elevations have traditionally incorporated sea-level rise.

Only in 2012 did Congress pass legislation—the Biggert-Waters Flood Insurance Reform Act—authorizing FEMA to factor sea-level rise into flood elevations provided by the flood insurance program, so the agency has had little opportunity to accomplish this for most of the nation. Right now, people could be rebuilding structures with substantially more near-term risk of coastal flooding because they are using flood elevations that do not account for sea-level rise.

Of course, reacting to any additional flood risk resulting from higher sea levels might entail the immediate costs of building higher, stronger, or in a different location altogether. But such short-term costs are counterbalanced by the long-term benefits of health and safety and a smaller investment in maintenance, repair, and rebuilding in the wake of a disaster. So how does the federal government provide legitimate science—science that is seen by decisionmakers as reliable and legitimate—regarding future flood risk to affected communities? And how might it create incentives, financial and otherwise, for adopting additional risk factors that may mean up-front costs in return for major long-term gains?

After Sandy, leaders of government locally and nationally were quick to recognize these challenges. President Barack Obama established a Hurricane Sandy Rebuilding Task Force. Governor Mario Cuomo of New York established several expert committees to help develop statewide plans for recovery and rebuilding. Governor Chris Christie of New Jersey was quick to encourage higher minimum standards for rebuilding by adding 1 foot to FEMA’s advisory flood elevations. And New York City Mayor Michael Bloomberg created the Special Initiative on Risk and Resilience, connected directly to the city’s long-term planning efforts and to an expert panel on climate change, to build the scientific foundation for local recovery strategies.

Right now, people could be rebuilding structures with substantially more near-term risk of coastal flooding because they are using flood elevations that do not account for sea-level rise.

The leadership and composition of the groups established by the president and the mayor were particularly notable and distinct from conventional efforts. They brought expertise and emphasis that focused as strongly on preparedness for a future that is likely to look different from the present, as on responding to the disaster itself. For example, the president’s choice of Shaun Donovan, secretary of the Department of Housing and Urban Development (HUD), to chair the federal task force implicitly signaled a new focus on ensuring that urban systems will be resilient in the face of future risks.

New York City’s efforts have been exemplary in this regard. The organizational details are complex, but there is one especially crucial part of the story that I want to tell. When Mayor Bloomberg created the initiative on risk and resilience, he also reconvened the New York City Panel on Climate Change (known locally as the NPCC), which had been begun in 2008 to support the formulation of a long-term comprehensive development and sustainability plan, called PlaNYC. All of these efforts, which were connected directly to the Mayor’s Office of Long-term Planning and Sustainability, were meant to be forward-looking and to integrate contributions from experts in planning, science, management, and response.

Tying the response to Sandy to the city’s varied efforts signaled a new approach to post-disaster development that embraced long-term resilience: the capacity to be prepared for an uncertain future. In particular, the NPCC’s role was to ensure that the evolving vulnerabilities presented by climate change would play an integral part in thinking about New York in the post-Sandy era. To this end, in September 2012, the City Council of New York codified the operations of the NPCC into the city’s charter, calling for periodic updates of the climate science information. Of course, science-based groups such as the climate panel should be valuable for communities and decisionmakers thinking about resilience and preparedness, but often they are ignored. Thus, another essential aspect of New York’s approach was that the climate panel was not just a bunch of experts speaking from a pulpit of scientific authority, but it also had members representing local and state government working as full partners.

Within NOAA, there are programs designed to improve decisions on how to build resilience into society, given the complex and uncertain interactions of a changing society and a changing environment. These programs routinely encourage engagement among different scales and sectors of government and resource management. For example, NOAA’s Regional Integrated Sciences and Assessments (RISA) program provides funding for experts to participate in New York’s climate panel to develop risk information that informs both the response to Sandy and the conceptual framework for adaptively managing long-term risk within PlaNYC. Through its Coastal Services Center, NOAA also provides scientific tools and planning support for coastal communities facing real-time challenges. When Sandy occurred, the center offered staff support to FEMA’s field offices that were the local hubs among emergency management and disaster relief. Such collaboration and interactions between the RISA experts, the center staff, and the FEMA field offices fostered social relations that allowed for coordination in developing the Sea Level Rise Tool for Sandy Recovery.

In still other efforts, representatives of the president’s Hurricane Sandy Rebuilding Task Force and the Council on Environmental Quality were working with state and local leaders, including staff from the New York City’s risk and resilience initiative. The leaders of the New York initiative were working with representatives of NOAA’s RISA program, as well as with experts on the NPCC who had participated in producing the latest sea-level rise scenarios for the National Climate Assessment. The Army Corps participated in the president’s Task Force and also contributed to the sea-level rise scenarios report. This complex organizational ecology also helped create a social network among professionals in science, policy, and management charged with building a tool that can identify the best available science on sea-level rise and coastal flooding to support recovery for the region.

We have to reconcile what we learn from science with the practical realities we face in an increasingly populated and stressed environment.

Before moving on to the sea-level rise tool itself, I want to point out important dimensions of this social network and the context that facilitated such complex organizational coordination. Sandy presented a problem that motivated people in various communities of practice to work with each other. We all knew each other, wanted to help recovery efforts, and understood the limitations of the flood insurance program. In the absence of events such as Sandy, it is difficult to find such motivating factors; everyone is busy with his or her day-to-day responsibilities. Disaster drew people out of their daily routines with a common and urgent purpose. Moreover, programs such as RISA have been doing research not just to provide information on current and future risks associated with climate, but also to understand and improve the processes by which scientific research can generate knowledge that is both useful and actually used. Research on integrated problems and management across institutions and sectors is undervalued; how best to organize and manage such research is poorly understood in the federal government. Those working on this problem themselves constitute a growing community of practice.

Communities need to be able to develop long-term planning initiatives, such as New York’s PlaNYC, that are supported by bodies such as the city’s climate change panel. In order to do so, they have to establish networks of experts with whom they can develop, discuss, and jointly produce knowledge that draws on relevant and usable scientific information. But not all communities have the resources of New York City or the political capacity to embrace climate hazards. If the federal government wishes to support other communities in better preparing people for future disasters, it will have to support the appropriate organizational arrangements—especially those that can bridge boundaries between science, planning, and management.

Rising to the challenges

For more than two decades, the scientific evidence has been strong enough to enable estimates of sea-level rise to be factored into planning and management decisions. For example, NOAA maintains water-level stations (often referred to as tide gages) that document sea-level change, and over the past 30 years, 88% of the 128 stations in operation have recorded a rise in sea level. Based on such information, the National Research Council published a report in 1987 estimating that sea level would rise between 0.5 and 1.5 meters by 2100. More recent estimates suggest it could be even higher.

Of course, many coastal communities have long been acutely aware of the gradual encroachment of the sea on beaches and estuaries, and the ways in which hurricanes and tropical storms can remake the coastal landscape. So, why is it so hard to decide on a scientific basis for incorporating future flood risk into coastal management and development?

For one thing, sea-level rise is different from coastal flooding, and the science pertaining to each is evolving somewhat independently. Researchers worldwide are analyzing the different processes that contribute to sea-level rise. They are thinking about, among other things, how the oceans will expand as they absorb heat from the atmosphere; about how quickly ice sheets will melt and disintegrate in response to increasing global temperature, thereby adding volume to the oceans; and about regional and local processes that cause changes in the elevation of the land surface independent of changes in ocean volume. Scientists are experimenting, and they cannot always experiment together. They have to isolate questions about the different components of the Earth system to be able to test different assumptions, and it is not an easy task to put the information back together again. This task of synthesizing knowledge from various disciplines and even within closely related disciplines requires interdisciplinary assessments.

The sea-level rise scenarios that our team used in designing the Sandy tool, which derived from the National Climate Assessment prepared for Congress every four years to help synthesize and summarize the state of the climate and its impacts on society, varied greatly. The scenarios were based on expert judgments from the scientific literature by a diverse team drawn from the fields of climate science, oceanography, geology, engineering, political science, and coastal management, and representing six federal agencies, four universities, and one local resource management organization. The scenarios report provided a definitive range of 8 inches to 6.6 feet by the end of the century. (One main reason for such different projections is the current inadequate understanding of the rate at which the ice sheets in Greenland and Antarctica are melting and disintegrating in response to increasing air temperature.) The scenarios were aimed at two audiences: regional and local experts who are charged with addressing variations in sea-level change at specific locations, and national policymakers who are reconsidering potential impacts beyond any individual community, city, or even state.

But wasn’t the choice of the experts who prepared the scenarios to present such a broad range of sea-level rise estimates simply adding to policymakers’ uncertainty about the future? The authors addressed this possible concern by associating risk tolerance—the amount of risk one would be willing to accept for a particular decision—with each scenario. For example, they said that anyone choosing to use the lowest scenario is accepting a lot of risk, because there is a wealth of evidence and agreement among experts that sea-level rise will exceed this estimate by the end of the century unless (and possibly even if) aggressive global emissions reduction measures are taken immediately. On the other hand, they said that anyone choosing to use the highest scenario is using great caution, because there is currently less evidence to support sea-level rise of this magnitude by the end of the century (although it may rise to such levels in the more distant future).

Thus, urban planners may want to consider higher scenarios of sea-level rise, even if they are less likely, because this approach will enable them to analyze and prepare for risks in an uncertain future. High sea-level rise scenarios may even provide additional factors of safety, particularly where the consequences of coastal flood events threaten human health, human safety, or critical infrastructure—or perhaps all three. The most likely answer might not always be the best answer for minimizing, preparing for, or avoiding risk. Framing the scenarios in this fashion helps avoid any misperceptions about exaggerating risk. But more importantly, it supports deliberation in planning and making policy about the basis for setting standards and policies and for designing new projects in the coastal zone. The emphasis shifts to choices about how much or how little risk to accept.

In contrast to the scenarios developed for the National Climate Assessment, the estimates made by the New York City climate panel addressed regional and local variations in sea-level rise and are customized to support design and rebuilding decisions in the city that respond to risks over the next 25 to 45 years. They were developed after Sandy by integrating scientific findings published just the previous year—after the national scenarios report was released. The estimates were created using a combination of 24 state-of-the-art global climate models, observed local data, and expert judgment. Each climate model can be thought of as an experiment that includes different assumptions about global-scale processes in the Earth system (such as changes in the atmosphere). As with the national scenarios report, then, the collection of models provides a range of estimates of sea-level rise that in total convey a sense of the uncertainties. The New York City climate panel held numerous meetings throughout the spring of 2013 to discuss the model projections and to frame its own statements about the implications of the results for future risks to the city arising from sea-level rise (e.g., changes in the frequency of coastal flooding due to sea-level rise). These meetings were attended by not only physical and social scientists but also by decisionmakers facing choices at all stages of the Sandy rebuilding process, from planning to design to engineering and construction.

As our team developed the sea-level rise tool, we found minimal difference between the models used by the New York climate panel and the nationally produced scenarios. At most, the extreme national scenarios and the high-end New York projections were separated by 3 inches, and the intermediate scenarios and the mean model values were separated by 2 inches. This discrepancy is well within the limits of accuracy reflected in current knowledge of future sea-level rise. But small discrepancies can make a big difference in planning and policymaking.

New York State regulators evaluating projects proposed by organizations that manage critical infrastructure, such as power plants and wastewater treatment facilities, look to science vetted by the federal government as a basis for approving new or rebuilt infrastructure. Might the discrepancies between the scenarios produced for the National Climate Assessment and the projections made by the NPCC, however small, cause regulators to question the scientific and engineering basis for including future sea-level rise in their project evaluations? Concerned about this prospect, the New York City Mayor’s Office wanted the tool to use only the projections of its own climate panel.

The complications didn’t stop there. In April 2013, HUD Secretary Donovan announced a Federal Flood Risk Reduction Standard, developed by the Hurricane Sandy Rebuilding Task Force, for federal agencies to use in their rebuilding and recovery efforts in the regions affected by Sandy. The standard added 1 foot to the advisory flood elevations provided by the flood insurance program. Up to that point, our development team had been working in fairly confidential settings, but now we had to consider additional questions. Would the tool be used to address regulatory requirements of the flood insurance program? Why use the tool instead of the advisory elevations or the Federal Flood Risk Reduction Standard? How should decisionmakers deal with any differences between the 1-foot advisory elevation and the information conveyed by the tool? We spent the next two months addressing these questions and potential confusion over different sets of information about current and future flood risk.

Our team—drawn from NOAA, the Army Corps, FEMA, and the U.S. Global Change Research Program—released the tool in June 2013. It provides both interactive maps depicting flood-prone areas and calculators for estimating future flood elevations, all under different scenarios of sea-level rise. Between the time of Secretary Donovan’s announcement and the release of the tool, the team worked extensively with representatives from FEMA field offices, the New York City climate panel, the New York City Mayor’s Office, and the New York and New Jersey governors’ offices to ensure that the choices about the underlying scientific information were well understood and clearly communicated. The social connections were again critical in convening the right people from the various levels of government and the scientific and practitioner communities.

During this period, the team made key changes in how the tool presented information. For example, the Hurricane Sandy Rebuilding Task Force approved the integration of sea-level rise estimates from the New York climate panel into the tool, providing a federal seal of approval that could give state regulators confidence in the science. This decision also helped address the minimal discrepancies between the long-term scenarios of sea-level rise made for the National Climate Assessment and the shorter-term estimates made by the New York climate panel. The President’s Office of Science and Technology Policy also approved expanding access to the tool via a page on the Global Change Research Program’s website. This access point helped distinguish the tool as an interagency product separate from the National Flood Insurance Program, thus making clear that its use was advisory, not mandated by regulation. Supporting materials on the Web site (including frequently asked questions, metadata, planning context, and disclaimers, among others) provided background detail for various user communities and also helped to make clear that the New York climate panel sea-level rise estimates were developed through a legitimate and transparent scientific process.

The process of making the tool useful for decisionmakers involved diverse players in the Sandy recovery story discussing different ideas about how people and organizations were considering risk in their rebuilding decisions. For example, our development team briefed a diverse set of decisionmakers in the New York and New Jersey governments to facilitate deliberations about current and future risk. Our decision to use the New York City climate panel estimates in the tool helped to change the recovery and rebuilding process from past- to future-oriented, not only because the science was of good quality but because integration of the panel’s numbers into the tool brought federal, state, and city experts and decisionmakers together, while alleviating the concerns of state regulators about small discrepancies between different sea-level rise estimates.

In 2013, New York City testified in a rate case (the process by which public utilities set rates for consumers) and called for Con Edison (the city’s electric utility) and the Public Service Commission to ensure that near-term investments are made to fortify utility infrastructure assets. Con Edison has planned for $1 billion in resiliency investments that address future risk posed by climate change. As part of this effort, the utility has adopted a design criteria that uses FEMA’s flood insurance rate maps that are based on 100-year flood elevations, plus 3 feet to account for a high-end estimate of sea-level rise by mid-century. This marked the first time in the country that a rate case explicitly incorporated consideration of climate change.

New York City also passed 16 local laws in 2013 to improve building codes in the floodplain, to protect against future risk of flooding, high winds, and prolonged power outages. For example, Local Law 96/2013 adopted FEMA’s updated flood insurance rate maps with additional safety standards for some single-family homes, based on sea-level rise as projected by the NPCC.

Our development team would never have known about New York City’s need to develop a rate case with federally vetted information on future risk, if we had not worked with officials from the city’s planning department. Engaging city and state government officials was useful not just for improving the clarity and purpose of the information in the tool. It was also useful for choosing what information would be included in the tool to enable a comprehensive and implementable strategy.

The key difference in the development of the Sandy recovery tool was the intensive and protracted social process of discussing what information went into it and how it could be used.

Different scales of government—local, state, and federal—have to be able to lead processes for bringing appropriate knowledge and standards into planning, design, and engineering. Conversely, all scales of government need to validate the standards revealed by these processes, because they all play a role in implementation.

Building resilience capacity

This complex story has a particularly important yet unfamiliar lesson: Planning departments are key partners in helping break the cycle of recovery and response, and in helping people adopt lessons learned from science into practice. Planners at different levels of government convene different communities of practice and disciplinary expertise around shared challenges. Coincidentally, scientific organizations that cross the boundaries between these different communities—such as the New York City climate panel and the team that developed the sea-level rise tool—can also encourage those interactions. As I’ve tried to illustrate, planning departments convene scientists and decisionmakers alike to work across organizational boundaries that under normal circumstances help to define their identities. These are important ingredients for preparing for future natural disasters and increasing our resilience to them over the long term, and yet this type of science capacity is barely supported by the federal government. How might the lessons from the Sandy Sea Level Rise Recovery Tool and Hurricane Sandy be more broadly adopted to help the nation move away from disaster-by-disaster policy and planning? Here are two ideas to consider in the context of coastal resilience.

First, re-envision the development of resilient flood standards as planning processes, not just numbers or codes.

Planning is a comprehensive and iterative function in government and community development. Planners are connected to or leading the development of everything from capital public works projects to regional plans for ecosystem restoration. City waterfronts, wildlife refuges and restored areas, and transportation networks all draw the attention of planning departments.

In their efforts, planners seek to keep development goals rooted in public values, and they are trained, formally and informally, in the process of civic engagement, in which citizens have a voice in shaping the development of their community. Development choices include how much risk to accept and whether or how the federal government regulates those choices. For this reason, planners maintain practical connections to existing regulations and laws and to the management of existing resources. Their position in the process of community development and resource management requires planners to also be trained in applying the results of research (such as sea-level rise scenarios) to design and engineering. Over the past decade, many city, state, and local governments have either explicitly created sustainability planner positions in high levels (such as mayors’ or governors’ offices) or reframed their planning departments to emphasize sustainability, as in the case of New York City. The planners in these positions are incredibly important for building resilience into urban environments; not because they see the future, but because they provide a nucleus for convening the diverse constituencies from which visions of, and pathways to, the future are imagined and implemented.

If society is to be more resilient, planners must be critical actors in government. We cannot expect policymakers and the public to simply trust or comprehend or even find useful what we learn from science. We have to reconcile what we learn from science with the practical realities we face in an increasingly populated and stressed environment. And yet, despite their critical role in achieving resilience, many local planning departments across the country have been eliminated during the economic downturn.

Second, configure part of our research and service networks to be flexible in response to emergent risk.

The federal government likes to build new programs, sometimes at the expense of working through existing ones, because new initiatives can be political instruments for demonstrating responsiveness to public needs. But recovery from disasters and preparation to better respond to future disasters can be supported through existing networks. Across the span of lands under federal authority, FEMA has regional offices that work with emergency managers, and NOAA supports over 50 Sea Grant colleges that engage communities in science-based discussions on issues related to coastal management. Digital Coast, a partnership between NOAA and six national, regional, and state planning and management organizations, provides timely information on coastal hazards and communities. These organizations work together to develop knowledge and solutions for planners and managers in coastal zones, in part by funding university-based science-and-assessment teams. The interdisciplinary expertise and localized focus of such teams help scientists situate climate and weather information in the context of ongoing risks such as sea-level rise and coastal flooding. All of these efforts contributed directly and indirectly to the Sea Level Rise Tool before, during, and after Hurricane Sandy.

The foundational efforts of these programs exemplify how science networks can leverage their relationships and expertise to get timely and credible scientific information into the hands of people who can benefit from it. Rather than creating new networks or programs, the nation could support efforts explicitly designed to connect and leverage existing networks for risk response and preparation. The story I’ve told here illustrates how existing relationships within and between vibrant communities of practice are an important part of the process of productively bringing science and decisionmaking together. New programs are much less effective in capitalizing on those relationships.

One way to support capacities that already exist would be to anticipate the need to distribute relief funds to existing networks. This idea could be loosely based on the Rapid Response Research Grants administered by the National Science Foundation, with a couple of important variations from its usual focus on supporting basic research. Agencies could come together to identify a range of planning processes supported by experts who work across communities of practice to ensure a direct connection to preparedness for future natural disasters of the same kind. These priority-setting exercises might build on the interagency discussions that occur as part of the federal Global Change Research Program. Also, since any such effort would require engagement between decisionmakers and scientists, recipients of this funding would be asked to report on the nature of additional, future engagement. What further engagement is required? Who are the critical actors, and are they adequately supported to play a role in resilience efforts? How are those networks increasing resilience over time? Gathering information about questions such as these is critical for the federal government to make science policy decisions that support a sustainable society.

Working toward a collective vision

The shift from reaction and response to preparedness seems like common sense, but as this story illustrates, it is complicated to achieve. One reaction to this story might be to replicate the technology in the sea-level rise tool or to apply the same or similar information sets elsewhere. The federal government has already begun such efforts, and this approach will supply people with better information.

Yet across the country, there are probably hundreds of similar decision tools developed by universities, nongovernmental organizations, and businesses that depict coastal flooding resulting from sea-level rise. The key difference in the development of the Sandy recovery tool was the intensive and protracted social process of discussing what information went into it and how it could be used. By connecting those discussions to existing planning processes, we reached different scales of government with different responsibilities and authority for reaching the overarching goal of developing more sustainable urban and coastal communities.

This story suggests that the role of science in helping society to better manage persistent environmental problems such as sea-level rise is not going to emerge from research programs isolated from the complex social and institutional settings of decisionmaking. Science policies aimed at achieving a more sustainable future must increasingly emphasize the complex and time-consuming social aspects of bringing scientific advance and decisionmaking into closer alignment.