Hurricane Irene: New York City dodges a potential storm surge mega-disaster

On August 24, 2011, I had good reason to fear New York City's worst-case storm surge disaster might be named Irene. As Category 3 Hurricane Irene ripped through the Bahamas on its way to North Carolina and New England, our most reliable hurricane forecasting model--the European (ECMWF) model--predicted that Irene would intensify to Category 4 strength with a 912 mb central pressure as it grazed the Outer Banks of North Carolina, then slowly weaken to a Category 3 hurricane before hitting southern New Jersey. Just a small perturbation from this scenario would bring Irene over New York City as a Category 2 hurricane. Since Irene was an exceptionally large storm with winds that covered a huge stretch of ocean, the storm had a much larger storm surge than it peak winds would suggest, and could have easily brought a storm surge of 15 - 20 feet to New York City. The storm would arrive during the new moon, when tides were at their highest levels of the month, compounding the storm surge risk.

Thankfully, the ECMWF model was wrong, and Irene's eyewall collapsed before the hurricane reached North Carolina. Irene made a direct hit on New York City as a tropical storm with 65 mph winds. Irene's storm surge reached 4.3 feet at the Battery on the south shore of Manhattan, which was high enough to top the city's seawall and flood low-lying park lands and roads near the shore. Fortunately, the water was not high enough to flood New York City's subway system, which could have easily occurred had Irene's winds been just 5 - 10 mph stronger.


Figure 1. Wind forecast for August 28, 2011 made on August 24, 2011 by the ECMWF model for Hurricane Irene. The model predicted that Irene would be a Category 3 or 4 hurricane with a 936 mb central pressure four days later, just south of New Jersey.

New York City: my number one storm surge disaster concern
I met last year with the head of the National hurricane Center's storm surge unit, Jaime Rhome, and asked him what his number one concern was for a future storm surge disaster. Without hesitation, he replied, "New York City." I agreed with him. Strong hurricanes don't make it to New York City very often, since storms must hit the city from the south or southeast in order to stay over water, and most hurricanes are moving northeast or north-northeast when they strike New England. New York also lies far to the north, where cold water and wind shear can tear up any hurricane that might approach. But if you throw the weather dice enough times, your number will eventually come up. New York City's number came up on September 3, 1821, when what was probably a Category 2 hurricane with 110 mph winds struck the city. The water rose 13 feet in just one hour at the Battery, and flooded lower Manhattan as far north as Canal Street--an area that now houses the nation's financial center. The maximum storm surge from this greatest New York City hurricane is unknown, but could have been 15 - 20 feet, which is what NOAA's SLOSH model predicts could occur for a mid-strength Category 2 hurricane with 100-mph winds.


Figure 2. The height above ground that a mid-strength Category 2 hurricane with 100 mph winds would would create in New York City in a worst-case scenario. The image was generated using the primary computer model used by the National Hurricane Center (NHC) to forecast storm surge--the Sea, Lake, and Overland Surge from Hurricanes (SLOSH) model. The accuracy of the SLOSH model is advertised as plus or minus 20%. This "Maximum Water Depth" image shows the water depth at each grid cell of the SLOSH domain. Thus, if you are inland at an elevation of ten feet above mean sea level, and the combined storm surge and tide (the "storm tide") is fifteen feet at your location, the water depth image will show five feet of inundation. This Maximum of the "Maximum Envelope of Waters" (MOM) image was generated for high tide and is a composite of the maximum storm surge found for dozens of individual runs of different Category 2 storms with different tracks. Thus, no single storm will be able to cause the level of flooding depicted in this SLOSH storm surge image. Consult our Storm Surge Inundation Maps for the U.S. coast for more imagery.

New York City's storm surge history
During the December 12, 1992 Nor'easter, powerful winds from the 990 mb storm drove an 8.5-foot storm surge into the Battery Park on the south end of Manhattan. The ocean poured over the city's seawall for several hours, flooding the NYC subway, knocking out power to the entire system. One train had to be backed out of a tunnel that was filling with water, and hundreds of passengers were rescued from stranded trains. Portions of the Port Authority Trans-Hudson Corporation (PATH) train systems in New Jersey were shut down for ten days, after low points in the rail tunnels flooded and major damage occurred to the control signals. Passengers had to be rescued from a train stalled in the PATH tunnel. Surges only one to two feet higher may have caused massive flooding of the PATH train tunnels. La Guardia Airport was closed due to flooded runways. Roadway flooding was also widespread—FDR Drive in lower Manhattan flooded with 4 feet of water, which stranded more than 50 cars and required scuba divers to rescue some of the drivers. Battery Park Tunnel held six feet of water. Major parkways were flooded in Nassau County, Westchester County, and New Jersey. Mass transit between New Jersey and New York was down for ten days, and the storm did hundreds of millions in damage to the city. The situation was similar in September 1960 during Hurricane Donna, which brought a storm surge of 8.36 feet to the Battery, and flooded lower Manhattan to West and Cortland Streets. The November 25, 1950 Nor'easter brought sustained easterly winds of up to 62 mph to LaGuardia Airport, and pushed a large storm surge up Long Island Sound that breached the dikes guarding the airport, flooding the runways.


Figure 3. Water pours into the Hoboken, New Jersey underground PATH mass transit station during the December 12, 1992 Nor'easter. Image credit: Metro New York Hurricane Transport Study, 1995.


Figure 4. Flooded runways at New York's La Guardia Airport after the November 25, 1950 Nor'easter breached the dikes guarding the airport. Sustained easterly winds of up to 62 mph hit the airport, pushing a large storm surge up Long Island Sound. The storm's central pressure bottomed out at 978 mb. Image credit: Queens Borough Public Library, Long Island Division.

Sea level rise and New York City
According to tide gauge data, sea level at the Battery at the south end of Manhattan has risen about 1 foot since 1900. This is higher than the mean global sea level rise of 7 inches (18 cm) that occurred in the 20th century. Global sea level rise occurs because the oceans are expanding as they heat up, and due to melt water from glaciers. The higher sea level rise in New York is due to the fact the land is sinking along the coast. These processes will continue during the coming century.

The U.N.'s Intergovernmental Panel on Climate Change (IPCC), predicted in 2007 a 0.6 - 1.9 foot global average sea level rise by 2100. However, they did not include melting from Greenland and Antarctica in this estimate, due to the large uncertainties involved. A paper published by Pfeffer et al. (2008) in Science concluded that the "most likely" range of sea level rise by 2100 is 2.6 - 6.6 feet. Three major sea level papers published since the IPCC report was issued in 2007 all agree that the IPCC significantly underestimated the potential sea level rise by 2100. In a 2009 interview with New Scientist magazine, sea level expert/ glaciologist Robert Bindschadler of the NASA Goddard Space Flight Center in Greenbelt, Maryland, commented, "most of my community is comfortable expecting at least a meter (3.28') by the end of this century." Sea level expert Stephan Rahmstorf added, "I sense that now a majority of sea level experts would agree with me that the IPCC projections are much too low." However, he cautioned that the popular media tend to focus on the upper limits of these newer projections (1. 5 - 2.0 meters), and "reaching the upper limits is, by definition, extremely unlikely.""

The sea level rise situation will be worse in areas where ocean currents have a large impact on the local sea level. This is the case along the Northeast U.S. coast, where the balance of forces required to maintain the very strong and narrow Gulf Stream Current means that sea water is sucked away from the coast, lowering the relative sea level from North Carolina northwards. During the coming century, the addition of large amounts of heat and fresh water into the North Atlantic due to higher precipitation, river runoff, and increased melting of glaciers is expected to weaken the Meridional Overturning Circulation (MOC) (also referred to as the thermohaline circulation), a global network of density-driven ocean currents. Weakening the thermohaline circulation will allow the Gulf Stream to spread out, resulting in sea level rise along the Northeast U.S. coast. Hu et al. (2009) found that a slow-down of the Meridional Overturning Circulation by 48% may occur by 2100, resulting in a 0.1 - 0.3 meter (0.25 - 1.0 ft) rise in sea level along the U.S. Northeast coast and Canadian Atlantic coast. This rise would be in addition to global sea level rise from melting glaciers and thermal expansion of the waters. A similar study by Yin et al. (2009) found a slow-down of the Meridional Overturning Circulation of 41% by 2100. New York City was in the region with the highest expected sea level rise from this ocean current effect--a rise of about 0.2 meters (0.75 feet) by the year 2100. If the Atlantic thermohaline circulation were to totally collapse, the authors predict a 4 ft (1.2 meter) rise in sea level along the U.S. Northeast coast solely due to the change in ocean currents. The IPCC predicts that such an abrupt climate change event (rather ridiculously depicted in the movie The Day After Tomorrow) will not occur in the coming century, though.

The future: Stronger hurricanes for New York City?
According to a summary statement endorsed by 125 of the world's top hurricane scientists at the World Meteorological Organization (WMO) Sixth International Workshop on Tropical Cyclones, in San Jose Costa Rica, in November 2006. "it is likely that some increase in tropical cyclone intensity will occur if the climate continues to warm." This makes intuitive sense, since hurricanes are heat engines that convert the heat of the ocean waters into the mechanical energy of wind. Turn up the thermostat, and you increase the energy available to make the strongest storms stronger. One major reason hurricanes weaken quickly when they approach New England is that the coastal waters cool dramatically north of North Carolina. As ocean waters warm during the coming century, hurricanes will be more able to maintain their strength farther to the north. One of the reasons the ECMWF model was simulating a 936 mb Hurricane Irene hitting New Jersey was because ocean temperatures off the mid-Atlantic coast were 1°C (1.8°F) above average during August 2011--the 7th warmest in recorded history. These high ocean temperature were due to the exceptional heat wave that gripped much of the mid-Atlantic during the summer of 2011--every state along the coast from Florida to New Jersey had a summer that ranked in the top four warmest summers since 1895. Such heat waves and warm ocean temperature are expected to become the new normal by mid-century, resulting in increased chances for strong hurricanes to make it to New England.


Figure 5. Summer temperatures along the U.S. Atlantic coast during 2011 ranked as 2nd - 4th warmest on record from Florida to New Jersey, resulting in exceptionally warm waters along the coast for Hurricane Irene to feed off of in late August. Image credit: National Climatic Data Center.

The other major reason that strong hurricanes have trouble making it to New England is that wind shear generally increases as one gets closer to the pole, due to the presence of the powerful winds of the polar jet stream. However, climate change theory predicts that the jet stream should migrate poleward during the coming decades, potentially reducing the amount of wind shear hurricanes arriving in New England will experience. A 2008 study by Archer and Caldeira found that the jet stream moved northwards 125 miles per decade during the 22-year period 1979 - 2001, in agreement with climate change theory. However, the migration of the jet stream northwards may also mean that hurricanes will be less likely to be caught up in a trough of low pressure embedded in the jet stream, resulting in fewer hurricanes swinging northwards to impact New England. At this point, it is hard to say whether or not changes to the jet stream due to climate change will alter the frequency of strong hurricanes reaching New England.

New York City's inadequate sea wall
The floodwalls protecting Manhattan are only five feet above mean sea level. At high tide, the water is only 3.5 feet below the top of the seawall, so clearly Manhattan is going to have a serious storm surge problem by the end of the century if sea level rise reaches the 3-foot plus figure many sea level rise scientists are predicting. As Ben Straus of Climate Central pointed out in a blog post on Irene, "sea level rise will amplify the impact of future hurricanes and Nor'easters. If we replay the 20th century but add an extra foot of sea level at the start (the extra foot we indeed started with in 2000, compared to 1900), about six events would produce higher water levels than the Nor'easter of 1992." Remember, the 1992 Nor'easter crippled the city's transportation system for ten days and caused hundreds of millions in damage. A Category 2 hurricane like the 1821 hurricane would be far worse, and could result in severe global economic consequences. A 15-foot storm surge from such a hurricane would swamp JFK and La Guardia airports. Manhattan would flood north to Canal Street, shutting down Wall Street and New York City's Financial District. The Holland Tunnel, much of the NYC subway system, and the New Jersey PATH mass transit systems would all flood. Many of the power plants that supply the city with electricity might be knocked out, or their docks to supply them with fuel could be destroyed. Nearly half a million people and almost 300,000 jobs lie within the Federal Emergency Management Agency (FEMA) 0.2-percent-annual-chance flood zones that would be inundated. As New York Times columnist Nate Silver wrote, such a disaster would likely cost near $100 billion. Furthermore, he makes the point, "Keep in mind that New York City's annual gross domestic product is about $1.4 trillion, one-tenth of the nation's gross domestic product, so if much of the city were to become dysfunctional for months or more, the damage to the global and domestic economies would be almost incalculable."


Figure 6. The seawall protecting Manhattan at Battery Park is only 5 feet above mean sea level. Tidal range at the Battery is plus or minus 1.5 feet, so at high tide a storm surge of just 3.5 feet is needed to send water over the seawall and into Manhattan.

Flooding of the NYC subway system
The U.S. Federal Transit Administration released a report in October 2011 called, "Flooded Bus Barns and Buckled Rails: Public Transportation and Climate Change Adaptation". The report says that with three feet of sea-level rise, the flooding produced by a 100-year storm at current sea levels will require only a 10-year storm, in other words, a tenfold increase in the frequency of flooding. Even without sea-level rise, a 100-year flood (an 8-foot storm surge) would inundate substantial portions of the subway system, whose tunnels generally lie twenty feet below street level. With sea-level rise though, the flooding occurs more rapidly and is more severe. A 100-year flood with a four foot rise in sea level would flood a large fraction of Manhattan subways, including virtually all of the tunnels crossing into the Bronx beneath the Harlem River and the tunnels under the East River. Flood waters enter the subway tunnels mostly vertically via ventila­tion grates and entrances as the streets flood, but also via inclined rail and road tunnels. Hydraulic computations show complete flooding takes only 40 minutes. Recovery would require obtaining huge quantities of pumps and hoses, awaiting restoration of power to the electrical grid, pumping out the flood waters, cleaning out miles of muddy and debris-filled platforms, stairs, tunnels and trackway, assessing the damage, and repairing problems. Much of the signal equipment and controls in the tunnels would be damged by salt or brackish water and would need to be disassembled, cleaned, and repaired or replaced to avoid corrosion and irreparable long-term damage. This specialized equipment, some of it 100 years old, is difficult to obtain and in many cases no longer manufactured. Researchers estimate a minimum recovery time of three to four weeks to reach 90 percent capacity, although when engineers were presented with the question, they believed that it could take one to two years to recover fully. This also assumes trains were moved to portions of the system with elevations above flood levels, in anticipation of the storm and were thus not damaged. Additional problems could result if the flood waters were contami­nated with toxins. Combined economic and physical damage losses from subway tunnel flooding under a 100-year storm surge were estimated at $58 billion at current sea levels and $84 billion with four feet of sea-level rise, assuming a linear recovery and an estimated subway outage time of three to four weeks.


Figure 7. New York City Subway vulnerability to a 100-year flood of 8 feet, with a 4-foot sea level rise. Blue lines are flooded subway tunnels. Orange areas have elevation less than 30 feet at present. Subway tracks are typically 20 feet below street level. Image credit: New York State Energy Research and Development Authority (NYSERDA), ClimAID: Responding to Climate Change in New York State, Draft Version, 2010.

What to do? Build a storm surge barrier
As I discussed in Part One of this series on U.S. storm surge risk, three cities in New England--Stamford, Providence, and New Bedford--have already built hurricane storm surge barriers that have more than paid for the cost of their construction in damages saved. Many coastal cities will need to substantially improve their flood defenses in coming decades due to rising sea levels. For New York, the best solution is to place three barriers at strategic "choke points"—the Verrazano Narrows, Throgs Neck, and the Arthur Kill, argues Douglas Hill of Stony Brook University's Storm Surge Research Group. I'll present his arguments in a guest post in Part Three of this series on storm surge risk in the U.S., coming up sometime in the next week.

Resources and references
Storm surge barriers: the New England experience: Part One of this series on U.S. storm surge risk.

The National Hurricane Center's Interactive Storm Surge Risk Map, which allows one to pick a particular Category hurricane and zoom in to see the height above ground level a worst-case storm surge may go.

Wunderground's Storm Surge Inundation Maps for the U.S. coast.

Climate Change Adaptation in New York City: Building a Risk Management Response: New York City Panel on Climate Change 2010 Report

Climate change information resources for NYC from Columbia University.

Landstrike is an entertaining fictional account of a Category 4 hurricane hitting New York City.

Colle, B.A., et al., 2008, New York City's vulnerability to coastal flooding: storm surge modeling of past cyclones, Bull. Am. Meteor. Soc. 89, 829–841 (2008).

Hu, A., G. A. Meehl, W. Han, and J. Yin (2009), "Transient response of the MOC and climate to potential melting of the Greenland Ice Sheet in the 21st century", Geophys. Res. Lett., 36, L10707, doi:10.1029/2009GL037998 29 May 2009

Rignot, E., and P. Kanagaratnam (2006), Changes in the velocity structure of the Greenland Ice Sheet, Science, 311, 986. 990.

Yin, J., M.E. Schlesinger, and R.J. Stouffer, 2009, "Model projections of rapid sea-level rise on the northeast coast of the United States", Nature Geoscience 2, 262 - 266 (2009).

Lady Liberty not at risk from a storm surge
As a side note, the Statue of Liberty is not vulnerable to a storm surge, since the good lady stands atop a 65-foot high foundation and 89-foot high granite pedestal. However, the 305' height of the lady's torch above the foundation means the statue will experience winds a full Saffir-Simpson category higher than winds at the surface. The statue is rated to survive a wind load of 58 psf, which is roughly equivalent to 120 mph winds (Category 3 hurricane). However, a mid-strength Category 2 hurricane with 105 mph winds will be able to generate 120 mph winds at a height of 300 feet, and would theoretically be capable of toppling the Statue of Liberty.

Jeff Masters