Climate Analysis - Chapter 2: A Stronger, More Resilient New York
CHAPTER 2 | CLIMATE ANALYSIS 20Climate AnalysisNear Bylot Island, CanadaCredit: Susan van Gelder
22Sandy Inundation, Bronx and Northern QueensDeﬁning DatumsSource: Stevens Institute of TechnologySimulated estimate of flooding by the Stevens Institute of Technology’s NYHOPS model.Note that these results are hypothetical.Tide Cycle Peak Water LevelStorm SurgeSandy hits at low tideHighest observed waterlevel at Kings PointSandy hits at high tideHypothetical highestobserved water levelat Kings PointSandy Inundation Simulated 9 Hrs. Earlier, Bronx and Northern QueensIllustrative Shift in Tide CycleA vertical datum is a base reference point for determiningheights or depths. Vertical datums set a consistent zeropoint so elevations can be compared with one another atdifferent locations with different physical characteristics. Forexample, flood levels can be measured relative to mean sealevel, or relative to ground levels that may be well abovemean sea level.Tidal datums, such as Mean Lower Low Water (MLLW), arestandard elevations defined by a certain phase of the tide.Tidal datums are used as references to measure local waterlevels and therefore vary over different areas. For example,the MLLW tidal datum is determined by averaging the lowerof the two low waters of any tidal day for a particular tidegauge over a period of time. There are tide gauges in theNew York City area at multiple locations, including atthe Battery and Kings Point. MLLW is a useful datum forcomparing water levels at a specific point to “normal”water levels, but is less helpful for comparing waterelevations in different locations, since they may experiencevery different MLLW levels.Gravity-based datums, such as the North AmericanVertical Datum of 1988 (NAVD88) are referenced to a fixedpoint in the ground. NAVD88 is the national standard, largelybecause it allows for comparisons of water levels acrossmany locations that have different tidal characteristics.In order to facilitate comparisons across different locations,this report refers to all water elevations in NAVD88 unlessotherwise specified. MLLW is used selectively to highlightlocation-specific water levels and typically shows higher valuesthan NAVD88. Flood depths, which are measured fromground level and vary with terrain, also are used to describethe flooding experienced in different neighborhoods.CHAPTER 2 | CLIMATE ANALYSISSource: FEMA (MOTF 11/6 Hindcast surge extent)The peak water level during a storm is a combination of the tide plus storm surge.Actual HypotheticalFEETFEETTIME TIMEA STRONGER, MORE RESILIENT NEW YORK21Although New York City has been hit bycoastal storms before, Sandy was anhistoric event by many measures. Since1900, 14 hurricanes and countless nor’eastershave struck the area. Sandy, however,exceeded them all—not only in terms of stormsurge height, but also in the scale and scope ofthe devastation it caused. (See sidebar: StormsThrough New York City History)Of course, Sandy was not just an historic storm.It was also idiosyncratic. As discussed inChapter 1 (Sandy and Its Impacts), a set ofcircumstances—timing, size, and path—allcame together to cause unprecedentedimpacts, primarily on the southern, coastal-facing areas of the city.As devastating as Sandy was, however, noteverything about the storm was unprecedented.Its 80- mile-per-hour (mph) peak wind gusts fellwell short of other storms that have hit New YorkCity, including Hurricane Carol in 1954 (up to125-mph gusts) and Hurricane Belle in 1976(up to 95-mph gusts). Previous storms alsobrought much more rain with them. Sandydropped a scant inch in some parts of New York,far less than the 5 inches of rain dropped on thecity during Hurricane Donna in 1960 or the7.5 inches during the April 2007 nor’easter.With greater winds and more rain, Sandy couldhave had an even more serious impact on theareas of Staten Island, Southern Brooklyn, andSouth Queens that experienced the most devas-tation during the storm. And while Sandybrought the full force of its impact at high tide forthese southernmost areas of the city, it hit thearea around western Long Island Sound almostexactly at low tide. As a consequence, parts ofthe Bronx, Northern Queens, and East Harlemwere not as affected as they could have been.In fact, the same storm, arriving at a slightlydifferent time, likely would have had significanteffects on New York’s northernmost neighbor-hoods. According to modeling undertaken by thestorm surge research team at the Stevens Insti-tute of Technology, if Sandy had arrived earlier—near high tide in western Long Island Sound,rather than in New York Harbor and along theAtlantic Ocean—the peak water level in the west-ern Sound, measured at the King’s Point gauge,which hit more than 14 feet above Mean LowerLow Water, or MLLW (over 10 feet above datumNAVD88) during Sandy, instead could havereached almost 18 feet above MLLW (almost 14feet above NAVD88). (See maps: Sandy Inunda-tion, Bronx and Northern Queens and SandyInundation Simulated 9 Hours Earlier, Bronx andNorthern Queens; see sidebar: Defining Datums;see graph: Illustrative Shift in Tide Cycle)The result would have been devastating forinfrastructure providing critical services tothe rest of the city. Flooding could have over-whelmed parts of the Hunts Point FoodDistribution Center in the Bronx, thereby threat-ening facilities that are responsible for handlingas much as 60 percent of the city’s produce.Meanwhile, the power plants in Astoria,Queens, which are responsible for almost one-third of the city’s installed generation capacity,could have been inundated as well. At La-Guardia Airport, which was flooded to about 14feet above MLLW (about 10 feet above NAVD88)during Sandy, this could have resulted in awater level of about 17 feet above MLLW (13feet above NAVD88) or up to 12 feet of waterabove ground level. Additional, four waste-water treatments plants and 29 water pumpingstations could also have been affected.Clearly, while Sandy was historic, it was not, infact, a worst-case scenario for all of New YorkFlooding at West and Cortlandt Streets, Hurricane Donna, 1960 Credit: Allyn Baum/The New York TimesSandy may have been the latestcatastrophic storm to hit New York City, butit certainly was not the first. Throughouthistory, the city has suffered fromhurricanes and other coastal storms, suchas nor’easters. Hurricanes and tropicalstorms strike New York infrequently,relative to other types of coastal storms(generally arriving during hurricaneseason, June 1 to October 31), and canproduce large surges, heavy rains, andhigh winds. Nor’easters, by contrast, arecold weather storms that have strongnortheasterly winds blowing in from theocean ahead of them. Compared tohurricanes, nor’easters generally bringsmaller surges and weaker winds but cancause significant harm because they tendto last longer, resulting in extended periodsof high winds and high water that can besustained through one or more high tides.In 1821, a hurricane made a direct strikeon New York City, bringing winds of about75 mph and a reported 13-foot stormsurge that flooded Lower Manhattan asfar north as Canal Street. In 1938, a stormknown as the Long Island Express—because the fast-moving eye passed overLong Island—hit with no warning, leadingto over 600 deaths, including 10 in NewYork City, while 100-mph wind gustsknocked out electricity north of 59th Streetin Manhattan. In 1960, Hurricane Donnahad wind gusts of up to 90 mph and a10-foot (above MLLW) storm surge thatcaused extensive pier damage. Majorstorms have been showing up in the NorthAtlantic with greater frequency in the lastfew decades. Examples of recent stormshaving significant impacts to New York Cityinclude: Agnes in 1972, Belle in 1976,Gloria in 1985, a nor’easter in 1992, Berthain 1996, Floyd in 1999, Isabel in 2003,Ernesto in 2006, a nor’easter in 2007, andIrene and Lee in 2011—which madeback-to-back appearances just 14 monthsprior to Sandy.Storms ThroughNew York CityHistory
CHAPTER 2 | CLIMATE ANALYSIS 24certain building codes temporarily, allowedNew Yorkers to begin rebuilding after the storm tostandards that better reflected actual flood risks.In June 2013, FEMA issued Preliminary WorkMaps (PWMs) for New York City that incorporatedeven more accurate wave modeling. Thoughsimilar in many cases to the ABFEs released inJanuary, the revised maps differed significantly incertain respects—they showed, for example,substantially smaller areas of the city at risk ofdestructive wave action. These PWMs will beconsidered best-available information until FEMAreleases Preliminary FIRMs (by the end of 2013),the first official product of the FEMA map updateprocess launched in 2009. After a public reviewand appeals period, the Preliminary FIRMs will berevised and released as new, final Effective FIRMs(replacing the 1983 maps) likely in 2015. The newFIRMs will inform a variety of flood-relatedrequirements, including flood insurance andflood-protective construction standards. Thoughsome adjustments may occur, it is currentlybelieved that the new FIRMs will tell a similar storyabout the city’s vulnerability to coastal storms aswas told by the PWMs. (See map: 2013 FEMAPreliminary Work Maps (PWMs)1983 FIRMs 100-Year FloodplainSandy Inundation Area1983 FEMA FIRMs and Sandy Inundation Area Comparison100-Year Floodplain500-Year Floodplain2013 FEMA Preliminary Work Maps (PWMs)100-Year FloodThe term “100-year” flood can be misleading,and perhaps even provides a false sense ofsecurity. This report uses the term “100-year”flood or floodplain because it is the mostcommonly used phrase and one with whichthe public is familiar. Nevertheless it isimportant to understand what the termmeans. A 100-year flood is not the floodthat happens once every 100 years. Rather,it is the flood that has a 1 percent chance ofoccurring in any given year. Experiencing a100-year flood does not decrease thechance of a second 100-year flood occurringthat same year or any year that follows.Even the 1 percent concept can bemisleading—because when the years addup, so too does the probability. A 1 percentchance each year may not seem like much,but when the public or private sectors aremaking decisions, it matters. Determiningwhether to buy a particular house or whereto build a power plant has long-termimplications. For example, a 100-year floodtoday, without considering future impactsfrom sea level rise or climate change, has a26 percent chance of occurring at leastonce over the life of a 30-year mortgage.Similarly, a 100-year flood today has a45 percent chance of occurring over the60-year life of a power substation.Lest anyone think the probability of aso-called 100-year storm is too remote toworry about or plan for, consider what itmeans for the children of New York today. Achild born today with the average lifeexpectancy of a New Yorker (80.9 years)faces a 56 percent probability (without sealevel rise) of witnessing today’s 100-yearflood within her lifetime.Source: FEMA (MOTF 11/6 Hindcast surge extent)Source: FEMAA STRONGER, MORE RESILIENT NEW YORK23City. And as the climate changes, raising theprospect of stronger storms coming morefrequently, the risks that New York City faceswill only intensify.Of course storms are not the only climatethreats New Yorkers face. The city is alsovulnerable to other “extreme” events, such asheavy downpours, heat waves, droughts, andhigh winds. Chronic conditions, such as risingsea levels, higher average temperatures, andincreased annual precipitation, also have directimpacts on the city and can make the effects ofextreme events worse. That is why this reportis not about preparing New York for the nextSandy or even the next coastal storm, but isinstead about how New York can adapt to thefull spectrum of future challenges posed byclimate change—whatever they may be.New York’s Current VulnerabilitiesSince 1983, New York’s vulnerability to coastalstorms has been reflected in flood maps pro-duced by the Federal Emergency ManagementAgency (FEMA), which describe the Federalgovernment’s assessment of flood risk. CalledFlood Insurance Rate Maps (FIRMs) becausethey are used by the National Flood InsuranceProgram (NFIP) and trigger certain floodinsurance requirements, the maps show howmuch land lies within the “100-year floodplain”(the area that has a 1 percent or greater chanceof flooding in any given year) and the “500-yearfloodplain” (the area that has a 0.2 percent orgreater chance of flooding any year). They alsodefine different zones of vulnerability within the100-year floodplain, including areas that are atrisk of destructive wave action, and thatgenerally require flood-protective constructionstandards (see Chapter 3, Coastal Protection;Chapter 4, Buildings; and Chapter 5, Insurance).These 1983 FIRMs show that a full 33 squaremiles of New York City—almost half ofBrooklyn—are within the equivalent of the100-year floodplain. As of 2010, there wereabout 218,000 New Yorkers living in those areas.All 14 of the city’s wastewater treatment plantsand 12 out of 27 power plants, representing37 percent of the city’s generation capacity, arewithin the 100-year floodplain as reflected in the1983 FIRMs, many of these critical facilitiesplaced on the coast out of operational necessity.There are also vibrant neighborhoods andcommercial districts in this area that contain ap-proximately 35,500 buildings, 377 million squarefeet of floor area, and 214,000 jobs. (See map:1983 FEMA Flood Insurance Rate Maps, FIRMs)However, even before Sandy, the City and FEMAhad known that the flood maps did not adequatelyreflect New York’s risks. Although FEMA convertedthe maps to digital form in 2007, their content hadnot changed meaningfully since 1983. As such,this report refers to the maps as 1983 FIRMs. Inthe intervening three decades, many changes hadbeen made to the city’s shoreline and significantdevelopment had occurred on the waterfront. Inaddition, sea levels had continued to rise as theyhad since the beginning of the 20th century (overa foot since 1900), more accurate coastalmodeling and mapping techniques had beendeveloped, and 30 years of additional data onstorms were available.Recognizing the need for updated information onNew York’s flood risks, in 2007, the City formallyrequested that FEMA update its flood maps forNew York—a multiyear process that FEMA kickedoff in 2009. In 2010, to help inform FEMA’smapping process, the City acquired the mostdetailed elevation data ever gathered for NewYork, known as LiDAR (light detection andranging) data. To collect these data, the City flewan airplane equipped with a laser scanner overthe five boroughs to measure land elevationswith tremendous precision. This allowed the Cityto create a detailed, three-dimensional picture ofthe shape and characteristics of New York’ssurface area—which in turn could be used byFEMA for substantially better flood mapping.Hurricane Sandy demonstrated the importanceof regular coastal updates to FEMA’s maps. Thearea that flooded during the storm was morethan one and a half times larger than the100-year floodplain defined on FEMA’s 1983FIRMs. In certain communities, the areas thatflooded were several times larger than thefloodplains outlined on the maps. In Brooklynand Queens, for example, the combined amountof land flooded was roughly equal to the amountof land in the entire citywide 100-year floodplainas mapped in 1983 (both about 33 square miles).Meanwhile, about 60 percent of all buildings andmore than half of the residential units in areasthat Sandy inundated were outside the 100-yearfloodplain, as were approximately 25 percent ofthe buildings tagged by the Department ofBuildings (DOB) as having been seriouslydamaged or destroyed as of December 2012. Inthese areas, not only were residents unaware ofthe risks that they faced, but the buildings inwhich they lived and worked had not beensubject to the flood-protective constructionstandards that generally apply within thefloodplain (see Chapter 4). (See map: 1983 FEMAFIRMs and Sandy Inundation Area Comparison)Just three months after Sandy, in January 2013, aspart of an effort to give New Yorkers betterinformation about their flood risks from coastalstorms, FEMA issued interim maps for New York,just as it had done for other communities that didnot have up-to-date maps following major storms(for example, it did so for Louisiana and Mississippiafter Hurricane Katrina in 2005). These interimmaps—called Advisory Base Flood Elevationmaps, or ABFEs—together with a set of emer-gency measures enacted by Mayor Bloomberg tosuspend certain zoning restrictions and modify100-Year Floodplain500-Year Floodplain1983 FEMA Flood Insurance Rate Maps, FIRMsSource: FEMA
CHAPTER 2 | CLIMATE ANALYSIS 26Heavy downpours also present risks to the transitsystem. A single rainstorm in 2007 severelydisrupted 19 major segments of New York City’ssubway system during morning rush hour,forcing much of the system to shut down andaffecting as many as 2.3 million subway riders.Impacts to the subway system created furthercongestion and delays on flooded roadways andon the bus system, as subway riders tried to finda ways to get to work.Meanwhile, heat waves—defined here as threeor more consecutive days of temperatures at orabove 90 degrees—are another extreme weatherthreat to New York. These events can be evenmore severe in New York due to the Urban HeatIsland (UHI) effect that can cause the city’s airtemperature to be more than seven degreeswarmer than in neighboring counties, particularlyat night, disproportionately impacting certainneighborhoods. The UHI effect is caused in partby a greater concentration of buildings and pavedareas, and affects energy use, comfort and qualityof life, and exposure to heat stress. Heat wavesstrain the city’s power grid and cause deaths fromheat stroke and exacerbate chronic healthconditions, particularly for vulnerable populationssuch as the elderly. In fact, heat waves kill moreAmericans each year than all other naturaldisasters combined. For example, a heat wave inNew York in July 2006 resulted in 140 deaths.Going forward, a more severe and persistent heatwave, or one coupled with a major power outage,could cause even more deaths.Another extreme event that impacts New Yorkis drought. Droughts can lower reservoir levelsand thus have an obvious and significant impacton the city’s drinking water supply. Severaldroughts have occurred over the last 50 years,with the most intense lasting from 1963 to 1965,during which time residents and businessessignificantly reduced water use throughvoluntary and mandatory restrictions. Since thatA September 2004 storm flooded 9th Street in Brooklyn.Patients being treated for heat exhaustion at the MaimonidesMedical Center in Brooklyn during the July 2006 heat waveCredit: Seth Wenig/The New York TimesCredit: James Estrin/The New York Times2009FEMA initiatesNew York/New JerseyCoastal Flood Study.2010New York City and FEMA form a partnership.The City acquires highly accurate topographicaldata, known as LiDAR.October 29, 2012Sandy hits.2015New FIRMs expected tobe adopted by the Cityafter FEMA’s process ofpublic appeals and response.June 2013FEMA releases PWMsfor New York City.January and February 2013FEMA releases ABFEs.Mayor Bloomberg signs an executiveorder providing zoning relief for New Yorkersrebuilding to FEMA’s new standards.2010sA STRONGER, MORE RESILIENT NEW YORK251981FEMA completes a CoastalFlood Study for New York City.1983FEMA issues the firstFIRMs for New York City.1991-2007FEMA revises FIRMs with updatedwetland and stream modeling, andminor adjustments to the floodplain.2007New York City calls onFEMA to conduct a fullupdate of the FIRMs.2007FIRMs are convertedto digital formand posted online.1980s 1990s 2000s 2Updating FEMA FIRMs for New York CityCityPopulation in the100-Year FloodplainShare of TotalPopulationLand Area of100-Year Floodplain(Square Miles)Population Densityof 100-Year Floodplain(People per Square Mile)New York 398,100 5% 48 8,300Houston 296,400 14% 107 2,800New Orleans 240,200 70% 183 1,300Miami 144,500 36% 18 8,000Fort Lauderdale 83,200 50% 21 4,000San Francisco 9,600 1% 3 3,200Floodplain Comparison of Major American CitiesOverall, the story told by the PWMs isunsurprising but nonetheless troubling. Thenew 100-year floodplain, roughly correspondingto the areas flooded during Sandy, is larger thanindicated on the 1983 maps by about 15 squaremiles, or 45 percent. The new floodplainincludes larger portions of all five boroughswith significant expansion in Brooklyn andQueens. Citywide, there are now 67,700 build-ings in the floodplain (an increase of 90 percentover the 1983 FIRMs) encompassing over 534million square feet of floor area (up 42 percent).The number of residential units in the floodplainhas increased to 196,700 (a jump of over61 percent), with the majority of thoseresidences in Brooklyn, Manhattan, andQueens. Almost 400,000 New Yorkers now livein the floodplain (up 83 percent)—more livingin the floodplain than in any other American city(though some cities, such as New Orleans, havea much higher share of their populations in the100-year floodplain). (See timeline: UpdatingFEMA FIRMs for New York City; see table: Flood-plain Comparison of Major American Cities)While the information contained in the PWMshas been critical for assessing current risks andinforming rebuilding, the city’s experience bothbefore and after Sandy highlights areas forimprovementinthecurrentFEMAflood-mappingprocess. The lack of regular updates, the timeinvolved in performing such updates, and thecommunication to stakeholders regardingthose updates have made it challenging forgovernments, infrastructure operators,residents, and business owners to understandand address their coastal flood risks.Storms are not the only weather challenges toNew York City. Another is heavy downpours—which have increased over the last half-centuryacross the Northeast. These heavy rainsthreaten the city’s critical infrastructure,especially the water and transit systems. Forexample, in 2011, back-to-back Tropical Storms,Irene and Lee, produced elevated turbidity(murkiness resulting from stirred sediment) andhigh bacteria counts in several of the City’sUpstate reservoirs that supply drinking water.During and immediately following the storms,turbidity levels remained high in the Catskill Sys-tem and in the Catskill Aqueduct, which carriesdrinking water from the Ashokan Reservoir tothe Kensico Reservoir before delivering it to thecity. As a result, special treatment continued foralmost nine months, the longest such treatmentperiod ever recorded. With treatment andoperational measures, the City ensured that thedrinking water delivered to the public remainedin compliance and safe for consumption.Source: NOAAs Spatial Trends in Coastal Socioeconomics, Demographic Trends (1970-2011); 2010 US Census Tiger Files, and population data; floodplain census data gathered from Miamis Chiefof Community Planning, Houstons City Engineer, and Fort Lauderdales Planning Department; New York population data was obtained from the Department of City Planning Population Division.
recognized that even updated FEMA flood maps,because they are based on historic data, will notprovide information about the changes that arelikely to threaten New York in the future.To ensure that the City would always haveaccess to the latest information about futureclimate risks, in September 2012 New York Cityformally codified the NPCC and the ClimateChange Adaptation Task Force when it wrotethose two entities into law—the first bill passedby any local government in the country toinstitutionalize a process for updatinglocal climate projections and identifying andimplementing strategies to address climaterisks. The new law requires that the NPCC meettwice a year, advise the City and the ClimateChange Adaptation Task Force on the latestscientific developments, and update climateprojections at least every three years, startingfrom March 2013.Of course, in the wake of Sandy, waitinganother three years would have been too long.That is why, in January 2013, the Cityreconvened the NPCC on an emergency basisto update its projections to inform planning forrebuilding and resiliency post-Sandy. NPCCmembers agreed to participate on anaccelerated timetable, setting aside otherimportant research to focus on updating theprojections to help New York plan for the future.Drawing on the latest climate models, recentobservations about climate trends, and newinformation about greenhouse gas emissions,the NPCC updated its 2009 projections—in adocument called Climate Risk Information2013, which it has released concurrent with thisreport. These projections tell a dire story aboutNew York’s future. (See table: NPCC 2013Climate Projections; see sidebar: How NewYork’s Climate Projections are Developed)The NPCC now projects that, by mid-century,sea levels could rise by more than 2.5 feet,especially if the polar ice sheets melt at a morerapid rate than previously anticipated. Thatmagnitude of sea level rise would threatenCHAPTER 2 | CLIMATE ANALYSIS 28Source: NPCC; for more details, see Climate Risk Information 2013.1Baseline period for sea level rise projections is 2000-2004.Like all projections, the NPCC climate projections have uncertainty embedded within them. Sources of uncertainty include data and modeling constraints, the random nature of some partsof the climate system, and limited understanding of some physical processes. The NPCC characterizes levels of uncertainty using state-of-the-art climate models, multiple scenarios of futuregreenhouse gas concentrations, and recent peer-reviewed literature. Even so, the projections are not true probabilities, and the potential for error should be acknowledged.NPCC 2013 Climate ProjectionsExtreme EventsBaseline(1971-2000)2020s 2050sMiddle Range(25th - 75th percentile)High End(90th percentile)Middle Range(25th - 75th percentile)High End(90th percentile)Heat Wavesand Cold EventsNumber of days peryear at or above 90°F18 26 to 31 33 39 to 52 57Number of heat wavesper year2 3 to 4 4 5 to 7 7Average duration (days) 4 5 5 5 to 6 6Number of days per yearat or below 32°F72 52 to 58 60 42 to 48 52IntensePrecipitationDays per year with rainfallexceeding 2 inches3 3 to 4 5 4 5Coastal Floodsat the Battery1Future annual frequency oftoday’s 100-year flood1.0% 1.2% to 1.5% 1.7% 1.7% to 3.2% 5.0%Flood heights from a 100-yearflood (feet above NAVD88)15.0 15.3 to 15.7 15.8 15.9 to 17.0 17.6Chronic HazardsBaseline(1971-2000)2020s 2050sMiddle Range(25th - 75th percentile)High End(90th percentile)Middle Range(25th - 75th percentile)High End(90th percentile)Average Temperature 54 ºF +2.0 to 3.0 ºF +3.0 ºF +4.0 to 5.5 ºF 6.5 ºFPrecipitation 50.1 in. +0 to 10% +10% +5 to 10% +15%Sea Level Rise10 +4 to 8 in. +11 in. +11 to 24 in. +31 in.A STRONGER, MORE RESILIENT NEW YORK27time, water demand has dropped, reducing therisk to New York from drought. However,the City continues to take steps to reduce waterdemand, such as identifying and repairingleaks, encouraging the use of more efficient“low flow” plumbing fixtures, and installingmore than 830,000 automatic meter readingdevices across the city to allow customers tomanage their water use better. While theseefforts have significantly increased droughtresilience, the City continues to monitor andmanage water demand.Finally, New York also faces the threat of highwinds—especially in connection with coastalstorms. High winds can down trees andoverhead utility lines, damaging property andcausing power outages. At high enoughspeeds, winds can even damage buildings.Category 1 hurricanes come with sustainedwind speeds of at least 74 mph, and Category2 hurricanes bring sustained winds of 96 to110 mph—far greater than Sandy’s 80-mphwind speeds at landfall in New Jersey. In fact,in 1954, Hurricane Carol brought sustainedwind speeds of up to 100 mph to the New Yorkarea, causing extensive damage.New York Vulnerabilitiesin the FutureAlthough New York clearly is at risk today,long-term changes in climate will make manyextreme events and chronic conditions worse.These changes have, in fact, been underway forsome time. As noted earlier, over the lastcentury, sea levels around New York City haverisen by more than a foot. Temperatures, too,are climbing. In fact, the National WeatherService and National Oceanic and AtmosphericAdministration (NOAA) labeled 2012 thewarmest year on record in New York City and inthe contiguous United States, with averagetemperatures in the US 3.2 degrees Fahrenheitabove normal and a full degree higher than theprevious warmest year ever recorded.Globally, all signs indicate that these changeswill accelerate. Atmospheric concentrations ofheat-trapping carbon dioxide have reachedlevels that have not been seen on earth formillions of years. Since the onset of theindustrial revolution, combustion of fossil fuelsand land use changes have led to a roughly40 percent increase in carbon dioxide levels.Because the key greenhouse gas, carbondioxide, stays in the atmosphere for 100 yearsor longer, the climate is essentially “locked in”to some additional warming. Meanwhile, sincethe late 1970s, global average temperatureshave increased by approximately 1 degreeFahrenheit and the volume of sea ice inthe Arctic during the month of September hasdeclined by almost 80 percent. Oceantemperatures have also warmed and the vastmajority of glaciers have retreated.Long-term changes in climate mean that whenextreme weather events strike, they are likelyto be increasingly severe and damaging. As sealevels rise, coastal storms are likely to causeflooding over a larger area and to cause areasalready at-risk to flood more frequently thantoday. As temperatures get warmer, heat wavesare expected to become more frequent, lastlonger, and intensify—posing a seriousthreat to the city’s power grid and NewYorkers’ health.Through PlaNYC, the City has been making aconcerted effort to understand the effects thatclimate change will have on New York. A criticalpart of this effort began as far back as 2008,when Mayor Bloomberg convened the New YorkCity Panel on Climate Change (NPCC)—one ofthe first American cities to create a body ofleading climate and social scientists charged withdeveloping local climate projections. Withrepresentatives from leading scientific institutions,such as the NASA Goddard Institute for SpaceStudies and Columbia University’s Earth Institute,the NPCC brought to bear state-of-the-art globalclimate models and local observations to analyzefuture local vulnerabilities.In 2009, the NPCC released its findings in agroundbreaking report that made predictions fora set of chronic hazards and extreme eventslikely to confront the city in the future. Thereport—entitled Climate Risk Information2009—described a New York that would be farmore exposed to climate-related impacts goingforward than it is today. For example, the NPCCprojected that by mid-century New York couldexperience sea levels (under a “middle range”scenario) that are up to a foot higher, causingflooding from what is today a 100-year storm tooccur two to three times as often. The NPCC alsoprojected that by the 2050s New York was likelyto experience more frequent heavy downpoursand many more days at or above 90 degrees.To begin addressing these risks, in 2008 theMayor convened more than 40 public and privateinfrastructure operators as part of the ClimateChange Adaptation Task Force, another PlaNYCinitiative. Task Force members used the NPCCprojections to evaluate the risks to theirinfrastructure and identify strategies to addressthem. For instance, Con Edison assessed howchanges in extreme heat would impact futurepeak electrical load demand, to determine whenadditional capacity might be required.The City also took action to strengthen its builtenvironment. For example, the City required newwaterfront development to design for the futurerisk of sea level rise and coastal storms, andpassed regulations allowing buildings to elevateelectrical equipment to their roofs withoutspecial permits. The City also launched theNYCºCool Roofs Program to paint rooftops white,thereby minimizing heat gain.The work of the Climate Change Adaptation TaskForce and City agencies demonstrates the powerof accurate information to drive thoughtfulplanning and decision-making. That is why theCity has continued to advocate for better andmore current information on the risks New Yorkfaces. As mentioned earlier, the City pushed foran update to FEMA’s flood maps for New York sothe City and its residents and businesses couldbetter understand the existing risks from floodingduring coastal storms. However, the City alsoCredit: Earl Wilson/The New York TimesWind damage from Sandy in Brooklyn
CHAPTER 2 | CLIMATE ANALYSIS 30low-lying communities in New York with regularand highly disruptive tidal flooding, and makeflooding as severe as today’s 100-year storm atthe Battery up to five times more likely. TheNPCC also predicts it is more likely than not(more than 50 percent probability) that therewill be an increase in the most intensehurricanes in the North Atlantic Basin.Meanwhile, the NPCC also predicts that, by the2050s, the city could have as many days at orabove 90 degrees annually as Birmingham,Alabama has today—a threefold increase overwhat New York currently experiences. Heatwaves could more than triple in frequency,lasting on average one and a half times longerthan they do today. Similarly, it is also very likely(more than 90 percent probability) that the NewYork City area will see an increase in heavydownpours over this time period.These projections have been subjected torigorous peer review, and represent thebest-available climate science for New York City.However, they are not yet officially recognizedby the State or Federal governments becausethere is no formal mechanism for them to do so.As planning for resiliency moves forward in NewYork, it will be necessary to make sure that allstakeholders addressing climate change in NewYork City are using common projections basedon the work of the NPCC to avoid confusion orconflicting standards.The City also has worked with the NPCC todevelop a series of “future flood maps” for NewYork that will help guide the city’s rebuilding andresiliency efforts. These forward-looking mapsare created by using a simplified approach thatcombines the NPCC’s “high end” sea level riseprojectionswithFEMA’sPWMs.Themapsillustratehow the 100-year floodplain could increase overthe next several decades with these high endprojections. Because these maps were notdeveloped using advanced coastal modeling, theaccuracy of the flood projections is limited andthey are not suitable for evaluating risks to indi-vidual properties. However, they are extremelyuseful for understanding the general extent offuture flood risks. (See map: Future Flood Mapsfor the 2020s and 2050s; see sidebar: PossibleLinks Between Sandy and Climate Change)The new maps show that the area that might beflooded in a 100-year storm in the 2020s couldexpand to 59 square miles (up 23 percent fromthe PWMs) and encompass approximately88,800 buildings (up 31 percent). With morethan 2.5 feet of sea level rise, New York City’s100-year floodplain in the 2050s could be72 square miles—a staggering 24 percent ornearly a quarter of the city—an area that todaycontains approximately 114,000 buildings2013 PWMs 100-Year FloodplainProjected 2020s 100-Year FloodplainProjected 2050s 100-Year FloodplainFuture Flood Maps for the 2020s and 2050sLike all environment-related projections and associated map products, the NPCC future flood maps have uncertainty embedded within them.In this case, uncertainty is derived from a set of data and modeling constraints. Application of state-of-the-art climate modeling, best mappingpractices and techniques, and scientific peer review was used to minimize the level of uncertainty. Even so, the map product should be regardedas indicative of the general extent of future flood risks based on high end sea level rise projections and not of the actual spatial extent offuture flooding.Source: FEMA; CUNY Institute for Sustainable CitiesPossible Links Between Sandy and Climate ChangeSandy has brought public attention to the climate hazards of the New York area. But didclimate change cause the storm? While it is impossible to attribute any one event suchas Sandy entirely to climate change, higher sea levels certainly did increase the extentand magnitude of coastal flooding caused by the storm. Since 1900, sea levels haverisen more than a foot in New York City, primarily due to climate change. As sea levelscontinue to rise, coastal storms will cause flooding over a larger area and at increasedheights than they otherwise would have.Sandy is also thought to have gained strength from unusually warm upper oceantemperatures in the North Atlantic. As the planet warms, upper ocean temperatures areexpected to increase, which could fuel storms. Although hurricanes depend on a rangeof climate variables and it is not clear how these other variables will change, recentstudies suggest that the most intense hurricanes may increase globally. And, it is morelikely than not (greater than 50 percent probability) that such hurricanes also willincrease in the North Atlantic Basin.Loss of sea ice as the Arctic warms may possibly have influenced Sandy’s path andintensity. The volume of sea ice in the early fall has decreased 75 percent since 1980,and some researchers have linked this to changes in the atmospheric steering currentsknown as the jet stream—changes that may be increasing the frequency and intensityof extreme weather events. The dip in the jet stream that contributed to Sandy’s“westward” turn that resulted in its striking New Jersey was unusual. Whether thereduction of sea ice played a role in that particular configuration remains unknown, butclimate scientists believe it is worthy of further research.Source: NPCC; for more details, see Climate Risk Information 2013A STRONGER, MORE RESILIENT NEW YORK29The New York City Panel on Climate Change (NPCC)develops climate projections using global climatemodels. These models are mathematical representa-tions of the earth’s climate system (e.g., the interactionsbetween the ocean, atmosphere, land, and ice). Theyuse estimates of future greenhouse gas and pollutantconcentrations to project changes in climate variablessuch as temperature and precipitation. Becausefuture emissions are uncertain, scientists use a rangeof scenarios that can be linked to assumptionsabout future population and economic growth andtechnological change.To develop the most recent set of climate projections,the NPCC used the latest climate models developedfor the upcoming Intergovernmental Panel on ClimateChange Fifth Assessment Report. The NPCC also usedestimates of future atmospheric concentrations ofgreenhouse gases called Representative ConcentrationPathways (RCPs), selecting two RCPs (4.5 and 8.5) forwhich the greatest number of climate modelsimulations were available and which span a range ofpotential future concentrations. To produce localtemperature and precipitation projections, the NPCCused these two RCPs and 35 global climate models forthe land-based grid box covering New York City. Togenerate sea level rise projections, the NPCC used24 global climate models and the same two RCPs. Forsea level rise, the NPCC also included additional globalfactors and local factors.The results provide a range, or distribution, ofoutcomes. Local projections are presented for the“middle range” (the middle 50 percent of thatdistribution) and the “high end” (the 90th percentileof that distribution). The high end is presented as amore extreme outcome and would be appropriate forthose with lower risk tolerances—such as criticalinfrastructure operators.How New York’sClimate ProjectionsAre DevelopedCredit: Center for MultiscaleModeling of Atmospheric ProcessesSource: NPCC; for more details, see Climate Risk Information 2013.
Initiative 1Work with FEMA to improve theflood-mapping processThe nearly three-decade gap between theintroduction of FIRMs for New York in 1983 andthe launch of a map update process in 2009meant that the City and other stakeholdershad to rely upon outdated and inaccurateinformation to assess coastal flood risks. TheCity will work with FEMA to improve the floodmap update process—seeking to requirecoastal analysis updates every 10 years. Toensure that FEMA’s maps are not just morecurrent but also more accurate and informa-tive, the City will continue to work with FEMA toreview the analysis leading to the production ofPreliminary FIRMs by the end of 2013. The Cityalso will call on FEMA to implement a seriesof technical and process improvements—including more appropriate application of wavemodeling, thorough documentation of all work,and the use of an external quality assurancecontractor to review completed work. Thiswork is technically complicated and checksshould be built into the process at every step.With participation from FEMA and the Office ofLong-Term Planning and Sustainability (OLTPS),this joint work can begin immediately.Initiative 2Work with FEMA to improve thecommunication of current flood risksDespite FEMA’s best efforts, many residentsand business owners in vulnerable areas havefound both the flood-mapping process and themaps themselves to be confusing. In fact, eventoday, many New Yorkers in the floodplain arenot aware of the existence of FEMA’s maps.The City, through OLTPS, will call on FEMA toincrease the transparency of its mappingprocess, to improve the user experience inaccessing online flood maps, and to expandefforts to make all affected property ownersaware of the maps. Subject to available funding,this may include joint development of a newinteractive platform for communicating flood-related risk information, insurance availability,and steps New Yorkers can take to protectthemselves from flood risks.Initiative 3Call on the State and Federalgovernments to coordinate with theCity on local climate change projectionsUsing multiple sets of climate changeprojections for New York City across differentlevels of government would cause confusionamong stakeholders and would potentially leadto conflicting standards for protecting againstfuture risks. To address this concern, the City willwork with State and Federal partners to agree ona uniform set of projections for New York Cityand a consistent approach for presenting thoseprojections, based on the work of the NPCC. TheCity, through OLTPS, also will call on the Federalgovernment to establish a policy that wouldrecognize local climate projections if they meetrigorous scientific standards.Initiative 4Continue to refine local climate changeprojections to inform decision-makingAlthough the NPCC’s 2013 work represents themost current view of the risks that New Yorkfaces, there remains more work to be done, asis always the case with such efforts. The Citywill work with the NPCC and key stakeholdersin 2013 and beyond to develop additionalclimate change projections and to make theseprojections even more useful. For example,OLTPS will work with the NPCC to includeadditional extreme climate events and chronichazards, such as high winds and humidity, inthe scope of the NPCC’s work. OLTPS and theNPCC also will work to identify a set of metricsthat can help the City and others measureactual climate changes against predictedclimate change, in order to adjust policies andinvestment decisions in the future.Initiative 5Explore improved approachesfor mapping future flood risks,incorporating sea level riseAlthough the City and the NPCC havedeveloped future flood maps to show how sealevel rise could change flood zones goingforward, the methodologies for developingthese maps can be improved with betterscience and intergovernmental coordination.To plan for future coastal risks more effectively,the City will work with the NPCC and Federalpartners to evaluate alternative approaches tomapping future risks. OLTPS will continue todevelop improved future flood maps and willwork with FEMA to develop recommendationsfor how FEMA can incorporate the futureimpacts of sea level rise into its ongoingnon-regulatory mapping efforts.Initiative 6Launch a pilot program to identify andtest strategies for protecting vulnerableneighborhoods from extreme heathealth impactsOn average, heat waves cause more deathsthan any other type of extreme weather event.Going forward, more intense, longer, and morefrequent heat waves will increase this risk,especially to seniors, those with chronic disease,and those without access to air conditioning.Subject to available funding, the City will:1) develop updated UHI models and maps tomeasure air temperature and evaluate landscape-based strategies to mitigate UHI effects; 2) workin two high-risk neighborhoods to identifyvulnerable populations, residential facilities,walking and transit routes, existing andpotential locations of UHI mitigation measures,and air conditioned spaces that could be madeaccessible as cooling shelters; and 3) engagewith community stakeholders and City agenciesto develop and implement enhanced Heat-Health Warning Systems, targeted UHI mitigationmeasures, and expanded access to air condi-tioned spaces during heat waves. The projectwill produce a replicable model for heat illnessprevention strategies to roll out to otherhigh-risk neighborhoods, and to inform citywidecooling messages and strategies. The projectwill be led by DOHMH, building upon studiesand communications strategies developed aspart of a Centers for Disease Control-fundedClimate-Ready Cities project. DOHMH will workin coordination with OLTPS and the Departmentof Parks & Recreation on the development ofUHI models and maps. The goal is to launch theproject in late 2013 and complete it by 2015.INITIATIVES FOR IMPROVING THE QUALITY OF CLIMATE ANALYSISThis chapter contains a series of initiatives thatare designed to strengthen the City’s ability tounderstand and prepare for the impacts ofclimate change. In many cases, these initiativesare both ready to proceed and have identifiedfunding sources assigned to cover their costs.With respect to these initiatives, the City intendsto proceed with them as quickly as practicable,upon the receipt of identified funding.Meanwhile,inthecaseofcertainotherinitiativesdescribed in this chapter, though theseinitiatives may be ready to proceed, they stilldo not have specific sources of fundingassigned to them. In Chapter 19 (Funding), theCity describes additional funding sources,which, if secured, would be sufficient to fundthe full first phase of projects and programsdescribed in this document over a 10-yearperiod. The City will work aggressively onsecuring this funding and any necessarythird-party approvals required in connectiontherewith (i.e., from the Federal or Stategovernments). However, until such time asthese sources are secured, the City willproceed only with those initiatives for whichit has adequate funding.CHAPTER 2 | CLIMATE ANALYSIS 32A STRONGER, MORE RESILIENT NEW YORK31(almost twice as many as indicated by thePWMs). This area currently accounts for97 percent of the city’s power generationcapacity, 20 percent of its hospital beds, and alarge share of its public housing. Over 800,000New Yorkers, or 10 percent of the city’s currentpopulation, now live in the 100-year floodplainprojected for the 2050s—a number of flood-vulnerable residents that is greater than thetotal number of people living in the entire cityof Boston.Building on the information contained in thesefuture flood maps, the City also commissionedan analysis of the economic impacts ofprojected changes in the city’s vulnerability tocoastal storms. This work was completed bySwiss Re, one of the world’s largest reinsurers(a company that, because it provides its clientswith reinsurance and insurance protectionagainst natural catastrophe risks, has devel-oped expertise in projecting the probability ofextreme weather and the resulting damage).Unlike the risk represented in FEMA’s maps,Swiss Re took into account the potentialdamage caused by both flooding and highwinds. Their analysis shows that the combinationof rising sea levels and more intense storms isexpected to come with significant costs—coststhat will be measured in many billions of dollars.(See sidebar: Expected Loss Modeling andCost-Benefit Analysis)With analytical tools such as the Swiss Re model,the City has yet another way of assessing thelikelihood and impact of coastal storms onNew York. Still the model does not assess theimpact of extreme events beyond coastalstorms (which include both storm surge andwind), nor does it assess potential public healthimpacts of coastal storms and other extremeweather events such as heat waves.The City, however, has been working to fill thisgap in understanding the public health risksposed to New York by climate change. As partof the Climate-Ready Cities and States Initiative,the City’s Department of Health and MentalHygiene (DOHMH) has been estimating healthrisks, identifying vulnerable populations, anddeveloping public health adaptation strategiesfor extreme heat and other climate hazards. Forexample, without mitigation, hotter summerspredicted for the 2020s (based on the NPCC2009 projections), could cause an estimated30 to 70 percent increase in heat-relateddeaths, or about 110 to 260 additional heat-related deaths per year on average in New YorkCity compared to the baseline period for theanalysis (1998–2002). Additional work will benecessary to refine these projections andidentify strategies with which to respond, butthis analysis is an important starting point thatillustrates, in yet another way, the stakesassociated with climate change.The remainder of this report outlines specificinitiatives to address the current and futureclimate change-related vulnerabilities faced byNew York as outlined above. But theseinitiatives will be most effective only if theycontinue to be informed by the best-availablescience. And while New York has been a globalleader in this area, there is still more that theCity can do—on its own and with the Federalgovernment—to improve the quality of thedata and tools available to it.Early morning view of the support dock on Liberty Island, damaged by the storm surge during Sandy. Credit: NPS/Rannow
—— 2050s—— 2020s—— Today050100150200Asset Damage andLost Economic Activity($ in Billions)Likelihood of Damage (1/50 = 50-year loss event)1/50 1/60$19B eventtoday: 1/70$35B$90B1/20 1/40 1/60 1/80CHAPTER 2 | CLIMATE ANALYSIS 34Based on these inputs, Swiss Re models producea “loss frequency curve” for each of threescenarios: 2012, the 2020s, and the 2050s. Eachcurve indicates the probability that a given levelof loss—in terms of both asset damage andlost economic activity, expressed in billions ofcurrent dollars—will be met or exceeded in anygiven year (known also as the “probability ofexceedance”). As sea levels rise and hurricanepatterns change, the loss curves move up,demonstrating both that the chance ofexperiencing a given level of loss grows over timeand the amount of loss increases if the probabilityof occurrence is kept constant.For example, according to the Swiss Re analysis,a storm today that causes the same magnitudeof infrastructure and property damage andeconomic loss as Sandy ($19 billion) isconsidered a once-in-70-year “loss event” (or hasa 1.4 percent chance of happening in any givenyear). This reflects a range of storms includingthose that, unlike Sandy, could result in very littledamage due to flooding but major damage dueto wind. With the impact of climate change (andassuming no additional development in thefloodplain), the models suggest that thisprobability will grow—causing a $19 billion lossevent (in current dollars) to become a once-in-60-year loss event by the 2020s (or an event witha 1.7 percent chance of happening in any givenyear), and a once-in-50-year loss event by the2050s (or an event with a 2 percent chance ofoccurring in any given year).In addition, by keeping the probability ofoccurrence constant, the Swiss Re analysisfurther shows that a once-in-70-year loss eventtoday is expected to cause in the futuresignificantly more damage than Sandy caused.The models suggest that a storm of thisfrequency would cause $35 billion (in currentdollars) of damage by the 2020s, an increase of1.8 times the actual damage caused by Sandy.Meanwhile, by the 2050s, with rising sea levelsand more intense storms, a once-in-70-year lossevent would cause an estimated $90 billion (incurrent dollars) of damage, or almost five timesthe asset damage and economic loss caused bySandy, even if it is assumed that no additionaldevelopment happens in the floodplain.Loss Frequency CurvesSource: Team AnalysisApproachThe City applied Swiss Re’s natural catastrophe models to New York City to help understand thepotential impacts of wind and storm surge on the city (FEMA’s FIRMs do not model the impacts ofwind), assuming a world of rising sea levels and more intense storms. In order to do so, the Cityand Swiss Re combined three sets of inputs:1. Hurricane models: As a seller of large-scale natural catastrophe reinsurance products, SwissRe has built simulations of hurricanes based on robust historical data. Swiss Re uses data fromthe National Hurricane Center that includes nearly 1,200 observed tropical storms andhurricanes in the Atlantic Basin between 1851 and 2008. The Swiss Re model then “tweaks”each of these historical storms hundreds of times to create over 200,000 storms that couldform in the area, and then uses established models for atmospheric pressure, speed, size, andangle of landfall to assess the resulting storm surge and wind fields.2. Climate change scenarios: The City provided Swiss Re with guidance on projected sea levelrise in the 2020s and 2050s, based on work of the New York Panel on Climate Change (NPCC).Specifically, the City instructed Swiss Re to assume of sea level rise by the 2020s, and the 2050s,based on the NPCC’s climate projections. In addition, Swiss Re adjusted the future frequency ofdifferent categories of hurricanes (tropical storm through category 5) based on academic research.3. City-level asset and economic activity: The consultants worked closely with City agencies todevelop a working model of asset value divided into several categories, including, among otherthings, buildings, transportation, telecommunications, and utilities. These asset values were furtherbroken down by zip code as was the city’s economic activity (gross city product).It is important to note several key limitations to this approach. First, while the Swiss Re modelsassess the potential impact of surge and wind resulting from coastal storms, they do not reflectthe risk from other climate impacts—heat waves, drought, heavy downpours, and more. As aresult, the analysis does not provide a holistic assessment of risk. Second, the analysis assumesthe city as it exists today, not as it may change in the future. Thus, impacts to major new buildingsor infrastructure that may exist in the 2020s or 2050s are not reflected in projected losses. Finally,and most importantly, the Swiss Re models only seek to estimate losses that can be readily measuredin dollars—namely, physical damage to assets, such as buildings and tunnels, and reductions inincome and loss of use due to physical damage (for example, if people in unimpacted areas couldnot travel to work due to transportation outages). Using this approach total losses caused by Sandy,an estimated $19 billion (according to the City’s analysis provided to the Federal government), couldbe broken down into over $13 billion of physical damage and almost $6 billion of lost economicactivity. But of course, not every potential impact can or should be quantified by such a simplemetric. For example, the Swiss Re models do not predict loss of life or injury. Nor do they highlightpotentially disproportionate impacts on disadvantaged populations such as the elderly ormedically vulnerable. These and other non-financial impacts should be and have been critical inputsin the development of the initiatives in this report.Expected Loss Modeling and Cost-Beneﬁt AnalysisA STRONGER, MORE RESILIENT NEW YORK33In setting out to define plans forstrengthening New York City’s re-siliency to climate change, it wascritical to anchor the development ofthose strategies in the best possibleunderstanding of the magnitude of therisks facing New York—including itsinfrastructure and its neighborhoods.Moreover, in a world of finite resourcesand competing priorities, a properlydeveloped resiliency strategy shouldassess potential initiatives in part byrelating the costs of those initiatives,including capital and operating costs,to the benefits of those initiatives—namely the reduction in risk.Although it is impossible to quantifyfuture risks to New York or the cost-benefit ratio of any specific interven-tion with precision, the insuranceindustry has developed probabilisticmodels that rely on analyticaltechniques to provide quantitativeguidance on these topics. In order toground its work in the best-availableanalysis, the City engaged Swiss Re, areinsurance company. Swiss Reuses probabilistic models to assessboth the frequency and severity ofan event (such as a coastal storm) aswell as the magnitude of loss likely tobe suffered if such an event wereto occur. Working with the City,the company applied the samemodels used for their internal under-writing and risk analysis activities tothe assessment of the risks facingNew York.Overview
CHAPTER 2 | CLIMATE ANALYSIS 36CHAPTER 2 | CLIMATE ANALYSIS 36In addition to calculating expected losses, theSwiss Re models also enable cost-benefitestimates of proposed interventions. Throughanalysis of the costs (including capital costsand ongoing operating costs) of specificinterventions, the models estimate the benefit ofthese actions in terms of avoided (or mitigated)damage to assets and losses to economicactivity. Although this model is not designedspecifically to measure the costs and benefits ofresiliency measures, it can provide helpfulguidance. For example, in evaluating proposals,the City generally concluded that an interventionwith a cost-benefit ratio of greater than two(projected costs twice as large as projectedbenefits) was unlikely to be attractive on a cost-benefit basis, even with refined assumptions.By contrast, a measure with a cost-benefit ratioof less than 0.5 (projected benefits twice aslarge as projected costs) was consideredhighly likely to be an attractive investment. Thechart above is an illustration of how generalinterventions were evaluated.Of course, as noted earlier, certain interventionsthat perform well or poorly on a cost-benefitanalysis might nonetheless be worthwhilepublic investments as a result of other, less easilyquantifiable attributes (such as the protectionor lack of protection provided to vulnerablepopulations). For this reason, cost-benefitanalyses were an important tool, but not theonly tool employed by the City in selectingamong resiliency strategies for this report.IllustrativeCost/BenefitRatio2.52.01.51.00.50.0Illustrative2050s Loss Averted ($ in Billions)Measures above 2 arelikely not justified bycost-benefit analysis(but may be justifiedby other criteria)Measures between0.5 and 2.0 requiremore detailedanalysis to determinecost-benefit viabilityMeasures below 0.5 arelikely attractive from acost-benefit perspectiveColumn width represents the totalimpact of that measure in the 2050sEach column represents a different resiliency measureCost-Benefit Analysis OutputA STRONGER, MORE RESILIENT NEW YORK35Expected Loss Modeling and Cost-Beneﬁt Analysis (Continued)While the loss frequency curves map differentlevels of loss to their exceedance probabilities,another way to understand the risks to New Yorkis to consider expected annual losses. This is gen-erated by multiplying the different exceedanceprobabilities by the amounts of loss associatedwith them and adding up the results (or putdifferently, by calculating the area under the losscurve). The resulting number indicates theexpected annual average impact to assets andeconomic activity, recognizing that in some yearsthe actual losses may be zero (if no coastalstorms strike New York) while in other years thelosses may be significant (if, for example, a Sandy-level loss event were to strike). The Swiss Remodels project that expected annual losses inNew York City of $1.7 billion today will grow to$4.4 billion in current dollars by the 2050s. As thechart indicates, this growth in expected losses isattributable in roughly equal proportions to risingsea levels (which make flooding from coastalstorms more damaging) and to the increasedfrequency of intense hurricanes.2050s Total2050s Additionalimpact from increasedfrequency of intensehurricanes2050s Additionalimpact fromsea level riseCurrentscenario188.8.131.52.2$ in BillionsYet another way to understand the projectedeconomic loss to the city due to sea level rise andthe increased frequency of intense hurricanes isby conducting a geographical analysis, takinginto account the physical locations of assets andeconomic activity. For example, the Swiss Remodels break these losses down by zip codeover time. Today, expected losses are concen-trated in many of the same areas of the city thatwere impacted during Sandy (such as the Eastand South Shores of Staten Island, SouthernBrooklyn, South Queens, the Brooklyn-QueensWaterfront, and Southern Manhattan), but alsoin other, less-impacted areas such as NorthernQueens and the Bronx. In the future, theexpected losses cover a significantly wider swathof the city. It is also important to note that whilethe maps divide the city by zip code (which maycover reasonably large areas, including inlandareas), actual losses generally will be concen-trated in the waterfront areas of those zip codes.<$10M (77%)$10 to $30M (18%)>$30M (5%)Today 2020s<$10M (71%)$10 to $30M (19%)>$30M (10%)2050s<$10M (53%)$10 to $30M (23%)>$30M (24%)Growth in Expected Annual Losses from Storm Surge and WindTotal Asset and Economic Activity Losses