Assessment of Floodplains in New York City - ASCE Library

World Environmental and Water Resources Congress 2013: Showcasing the Future © ASCE 2013

Assessment of Floodplains in New York City Mohammad Karamouz1, Yaser Altayyar 1, Zahra Zahmatkesh2, Mahama Damei1 1

Department of Civil & Urban Engineering, Polytechnic Institute of NYU, Brooklyn, NY, Email: [email protected] 2School of Civil Engineering, University of Tehran, Tehran, Iran, Email: [email protected]

ABSTRACT Different studies around the world have proved that the mean sea level is rising globally in the last century. The flood plains are mapped to determine the extent of flood hazard and zoning for construction, coastal protection and evacuation if needed. In this study, the mean high high as well as extreme water level in New York City is analyzed to determine how the current flood plain developed by FEMA and NOAA could represent the current hazard of 100 year floods. For this purpose the sea level rise data since 1920 up to 2012 are considered including the water level data during Irene and Sandy hurricanes. The results are employed to adjust the 100-year in New York City. The flood plain boundaries are determined and the evacuation zones are developed and compared to the current 100 year flood plain map. Moreover, an aerial map was used in this study for better assessment of current and future vulnerabilities of coastal cities including New York City. The vulnerable stations and major financial and residential regions are identified and some suggestions are provided to reduce their vulnerabilities. Some recommendations are provided for best management practices (BMPs) application to reduce flood risks and damages. This study shows that rising sea level and high storm surges, especially following hurricane Sandy, have caused the floodplain to shift inland and more efficient flood control measures are needed. Keywords: New York City, Storm surge, Flood plain, Statistical analysis, Best Management Practices (BMPs) INTRODUCTION Flood is a natural disaster that occurs when water body cannot hold the excess water and floods the adjunct areas or by storm surge. The term floodplain is referred to the lands adjacent to these water bodies as well as low-lying areas that can’t transport excess rainfall. Floodplains are designated by how often they can be expected to flood. A “100year flood” describes an area that has a one percent chance of a flood occurring in any given year. Whether or not a flood occurs one year has no bearing on the following year. on how to act during and after The vulnerability of the region is also increasing if no adequate planning and management practices are taken such as flood control management, updating evacuation zone, flood risk assessment, and education of people on how to act during and after floods. These types of practices will help coastal areas to be more resilient to face catastrophic flood events, reduce the damage and loss of life and recover quickly after disasters.

1068


The federal government of the U.S got involved in floodplain management in the 1800s. The great Mississippi River flood of 1927 led the federal government to become a major player in flood control. In 1960’s, structural flood control projects such as such as dams, levees and flood walls became a mandatory way to reduce flood losses. As developments were increasing within the floodplain, structural flood control projects failed to reduce flood losses. In response, federal, state and local agencies began to develop policies and programs with a “non-structural” emphasis, ones that did not prescribe projects to control or redirect the path of floods. The creation of the National Flood Insurance Program (NFIP) in 1968 was a land mark step in this evolution. The NFIP established an insurance program as an alternative to disaster relief, distributed responsibility for floodplain management to all levels of government and the private sector, set a national standard for regulating new development in floodplains, and began a comprehensive floodplain mapping programs. The land area covered by the flood waters of the base flood is the base floodplain. Most coastal floods are caused by coastal storms, usually hurricanes and northeasters. Such storms bring air pressure changes and strong winds that “pile” water up against the shore in what is called a storm surge. McEwen (2002) evaluated the inherent exposure of caravan parks on floodplains to flood risk and the vulnerability of residents in the aftermath of the April 1998 floods in the Midlands, UK. It considers flood warning dissemination and response, flood impacts and related planning and control issues associated with caravan parks and their communities. The paper recommends that owners/managers at existing parks at flood risk should be legally required to check on the flood risk, to fined effective flood action plans, to communicate this information to prospective park users and to consider flood risk in park design. Mori (2012) presented a model of the problem on floodplain development, exploring the conditions that are both necessary and sufficient for development to be optimal. The model is calibrated for a particular catchment, the Ouse catchment in the United Kingdom, and is used both to estimate the expected impact of floodplain development and to explore the impact of alternative policy instruments. They found that the use of price- based instruments that signal the expected flood damage cost of floodplain development has the potential to lead to outcomes close to the social optimum. The finding is robust to two types of uncertainty: model error about the relation between precipitation and flood-risk and measurement error about the benefits of developed floodplains. Simonovic and Akter, (2006), developed a methodology using fuzzy set theory and fuzzy logic which is applied to floodplain management in the Red River Basin, Canada that faces periodical flooding. They have demonstrated that the empowerment of stakeholders can improve the floodplain management process and provide decisions acceptable to a wider group of stakeholders. Barnard, et al (2009) developed a coastal-hazards model to determine the impact of severe winter storms, both in real-time and using prescribed scenarios. The model and forecasts in this project can be updated periodically to include new changes in sea level and other inputs (for example, wave climate, bathymetry, topography), and can be used as the basis for local hazard assessments and real-time warning systems in strategic partnership with numerous

1069


Federal, State, local, and academic partners. This paper deals with assessment of flood plains in NYC. It touches on the one hundred year flood maps as well as Zone A evacuation maps and the way they have been determined and mapped. There is also information present on New York City’s vulnerability to flooding in terms of infrastructure, focusing on the tunnels and subways of NYC. Best Management Practices (BMPs) that can be used to mitigate the effect of flooding on New York City Infrastructure are also mentioned. The paper is followed by description of the proposed methodology. Case study is presented in the following. Finally results and, summary and conclusion are given. METHODOLOGY Mean high high water level and extreme high water level will be considered in this study to update the existing floodplain of NYC. Estimate floodplain based on the data including Hurricane Irene is also reassessed and the impact of Hurricane Sandy will be considered. The sea level Data from the Battery station will be fitted using other distributions. For this purpose the frequency analysis is used to determine the flood elevation with different return periods, which is in turn used to determine floodplain elevations. The U.S. Water Resources Council recommended the log-Pearson III distribution as a base method for flood frequencies. The parameters of the log-Pearson III distribution are determined based on the historical data. Regarding the corresponding probability to different return periods, the sea level heights for different return periods are determined. The estimated flood elevations with different return periods are used to determine the corresponding flooding areas. Then the results of frequency analysis on historical data will be compared with Hurricane Irene and Sandy to evaluate the extent of the floodplain. The results are compared to FEMA-determined flood plains including 100-yrs and 500-year flood zones. To create flood plain maps, Google Earth (available at: http://www. google.com/earth/index.html) is used. FEMA flood plain maps are provided in compatible format with Google Earth in (https://hazards.fema.gov /femaportal /wps/ portal/ NFHLWMSkmzdownload). An attempt will be made to include other factors such as Long Island Sound impact on Upper East Side and East Midtown by using other data stations such as Kings Point data. Using the resulted data, the vulnerable regions of NYC will be reassessed using Google earth map. Using the results of frequency analysis and determining the flood plain boundaries, the evacuation zones are developed and compared to the current 100 year flood plain map. For better assessment of current and future vulnerabilities of coastal regions of New York City, an aerial map is used. Due to the importance of subway stations, the more vulnerable stations are identified and some suggestions are provided to reduce their vulnerabilities. Some recommendations are provided for best management practices (BMPs) application to reduce flood risks and damages.

1070


CASE STUDY New York City is located on one of the world's largest natural harbors (40.71ºN, 74.01ºW). This study mainly focused on the east side of Manhattan. With close to 2400 km of shoreline, development in the region was always related to the ocean. Four out of five of the New York City boroughs are located on islands. More than 2000 bridges and tunnels connect these islands and the mainland. Flood levels of only 0.30–0.61 meters above what was seen during a December 1992 storm could have produced large amounts of inundation and deaths. Rising sea levels would make these types of flooding situations much more common. The sea level time series for the region, from 1920 to 2012, are obtained from the Battery Park station through the National Oceanic and Atmosphere Administration (NOAA) (http://tidesandcurrents. noaa.gov/station_retrieve.shtml). RESULT Based on the collected data, frequency analysis is performed on the highest recorded sea level for each year and extreme storm events when water levels are higher than normal tide levels. Return periods and exceedance probabilities are calculated and the corresponding heights are found using a log-Pearson Type III probability distribution. Table 1 lists the probability distribution statistics for the mean higher high water levels and extreme water levels. Table 1: Statistical values for mean higher high water levels Statistics

Mean Higher High (1920-2012)

Extreme Water Level (1966-2012)

Variance

0.0029

0.0016

Standard Deviation

0.0536

0.0395

Skewness Coefficient

1.7874

2.9674

Using Table 1, the water level for each return period are found using Log-Pearson analysis. The values are tabulated in Table 2. Table 2: Water levels for return periods using mean higher high water levels Return Period, (Years)

Mean higher high (1920-2012)

Extreme water level (1966-2012)

2

K-Value for Skew Coeff = 0.4701 -0.280

Water Level, (ft) 7.070

K-Value for Skew Coeff = 1.775 -0.394

Water Level, (ft) 10.612

5

0.645

7.925

0.427

11.434

10

1.319

8.612

1.185

12.251

25

2.191

9.590

2.278

13.530

50

2.844

10.394

3.146

14.642

100

3.492

11.259

4.039

15.880

200

4.137

12.192

4.948

17.249

1071


As presented in this table, different return periods of water levels are determined by available data during the period of 1920 to 2012 which includes the water level followed by hurricane Sandy. Damei (2011) through studying the water level trend in the period of 1920-2011, determined mean high high and extreme water levels’ of 100 and 200 year return periods as 10.30 and 13.26, and 10.70 and 13.96ft, respectively. Comparison of the results of the present study and Damei (2011) reveals that by considering water level data followed by hurricane Sandy in determining 100 year water level return period, the mean high high and extreme water levels have increased from 10.30 to 11.26 and 13.26 to 15.88, respectively. The results show that rising sea level and storm surges following hurricane Sandy have increased the water level of different return periods. Therefore more efficient flood control measures are needed in the study area. Figure 1 shows the current flood zones in the New York City. As seen, the different flood zones are referred to Zones A, B, and C where zone A has the highest chance of flooding and C has the least. Zone A is the 100-year floodplain, where the chance of a flood occurring is 1% in any given year, and the area with a 26% chance of flooding during a 30-year mortgage. It is the area that is evacuated in the event of a large storm as it is the most likely to get flooded. As defined by FEMA, Zone B is the “area of moderate flood hazard, usually the area between the limits of the 100-year and 500-year floods, which has a 0.2% chance of occurring in any given year. These are also used to designate base floodplains of lesser hazards, such as areas protected by levees from 100-year flood, or shallow flooding areas with average depths of less than one foot or drainage areas less than 1 square mile. This is the area that is evacuated in the event of a hurricane of Category 2 or higher. Zone C has the least risk of flooding from a storm event.

Figure 1: Flood Zones in New York City Figure 2 shows the flood hazard zones for Lower Manhattan, New York City (Google Earth, FEMA, 2012). The map is created using FEMA’s flood map overlays in Google Earth. From this map, it can be seen which parts of the island are considered high risk in the event of storms which produce high levels of storm surge. As seen in Figure 2, the

1072


areas shaded in red are the high-risk zones, or the 100-year floodplain. The lighter pink shaded areas are the 500-year floodplain.

Figure 2: Proposed floodplain extension based on Manhattan water Levels (Google Earth, FEMA, 2012)

The flood plain extends onto the island of Manhattan in some areas, such as the lower part of the borough (Figure 2) as well as the Upper East Side. It is important to see the potential flooding possibilities in these areas as they are the major financial and residential regions.

Figure 3: Maximum water level determined by FEMA at lower Manhattan (Google Earth, FEMA, 2012)

Comparing the results of frequency analysis for extreme water levels using the data of Battery Park, with Figure 3 reveals that water level rise resulting from a 200 year flood, considerably more than the current what FEMA predicted. FLOOD PLAIN ELEVATION IN THE UPPER EAST SIDE

1073


The interaction of water bodies from West, East and North of the city is influencing the elevation of the water level, which makes the flood plain, extended in the Upper East Side (East Harlem). From the west side of the city, the Harlem River is navigable tidal strait that flows 8 miles between the Hudson River and East river, separating the boroughs of Manhattan and the Bronx. From East, The water coming from the Long Island Sound to the East River will lead to a higher water elevation that flow easily in the low elevation areas to reach to Central Park. From North, 24 miles Bronx River is also contributing to the east river flow. It divides the east Bronx from West Bronx and then it flows into the East river. A tidal strait is also connecting East River to the Long Island Sound between the Sound view and Hunts Point. The elevation assessment is an essential step for determining the floodplain. Google earth is used to estimate the elevation of different blocks in the Upper East Side region of NYC. The floodplain is traced using FEMA NFHL application (2012) for the 100-year floodplain map and overlaid in Google Earth (2012). The blocks from 79th to 125th St are selected that are most affected by the above interconnected water bodies. The floodplain in this region is extended to higher elevation of about 18 ft. Moving from north to south, up to the cross-section of Harlem River Dr. and Martin Luther king Blvd at about 125th St the floodplain elevation is 9ft. It increases to 11ft and stays there with the same elevation until close to 108th St. Then, increases drastically to 17-18 ft at 108th St and Park Ave and remain in the same height up to 2nd Ave and 102nd St. Then the elevation reduces to 12 ft at 99th St between 1st and 2nd Ave and stays at 12 ft. until 93rd St and York Ave. The floodplain is gradually reduces to 9 ft beside the FDR and 90th St. The elevation increases at one location to 14 ft. at 79th St. and then stays at 9 ft until 23rd St. The yellow pins represent the shoreline elevation beside the FDR. The elevation ranges from 1- 4 ft high, which is close to the mean sea level. The elevations are pinpointed in Figure 4. Shoreline elevation beside the FDR

Harlem River

Floodplain borders

Long Island Sound East River

Manhattan Queens

Figure 4: Elevation of the 100-yr floodplain and the low-lay regions in Upper East Side (East Harlem) (Google Earth 2012)

1074


Figure 5 shows the elevation level at the Midtown area of NYC. The white pins show the floodplain borders. It can be seen that the elevation start from 9 ft at 38th St. then, it reduces to 6 ft at 37th and 36th St and the FDR. It continues decreasing to 4 ft at 23 St between Ave B and Ave C, and then goes back to 9 ft. The Yellow pins show also the low shoreline elevations beside the FDR. The elevation start from 6 ft at 36 St and decreases to a range of 1 and 2 ft at Ave C and 23rd St and 22nd St. These low-lay areas has no wall protection to prevent the flood from spreading Inland. Therefore, there is a need of implementing a protection measures in this region. All these factors: the interaction of different water bodies, the low elevation of the Upper East Side, and the narrow width of this region makes the floodplain cover more inland than other part of the city. During the hurricanes, the storm surge and the water level will affect the highly Hispanic populated region with low income. This region needs an effective Best Management Practices (BMPs) that help to reduce the runoff and flood. Also, a hospital is located at 1st Ave and 102nd St that will be affected by storm surge and consequent loss of power. Mitigation measures as BMPs have a long way to go in determining how to control flooding in the subways. The proposed floodplain maps also show the areas of Lower Manhattan that are more suseptable to flooding (Mahama 2011). Shoreline elevation beside the FDR Floodplain borders

Midtown Manhattan

Figure 5: Floodplain and Low-lay areas at midtown NYC (Google Earth 2012)

1075


SUMMARY AND CONCLUSION In the present study frequency analysis is performed on the mean high high as well as extreme water level data and the log Pearson type III probability is fitted to sea level data and the corresponding sea levels for different return periods are determined. The sea level data from 1920 to 2012, including the water level data during Irene and Sandy hurricanes, are obtained from the Battery Park station of NOAA. The data are used to determine the 100-year in New York City (Lower Manhattan). Results are compared with the current flooding map of the study area, developed by FEMA. The results show that New York City faces high risk of floods by storm surges and hurricanes that result in a significant shift of floodplain inland. The results also show that the current flood plains are significantly different from the results of this study. An area in the North east of Central part is particularly vulnerable due to intersection of different water bodies colliding at this area. Further studies and revising the flood maps are necessary and reassessments of the evacuation zones are also need to be considered in this region. REFERENCES - Damei, M., (2011) Vulnerability Assessment of Coastal: Revisiting New York City Floodplains, project report Submitted in Partial Fulfillment of the requirements for the degree of MASTER OF SCIENCE (Environmental Science) at Polytechnic Institute of New York UniversityBarnard - Google Earth (http://www. google.com/earth/index.html) - McEwen, L. , Hall, T., Hunt, J., (1998) Flood warning, warning response and planning control issues associated with caravan parks: the April (1998) floods on the lower Avon floodplain, Midlands region, UK. - Mori, K, Perrings. C, (2012), Optimal management of the flood risks of floodplain development, Science of the Total Environment, 431, 109–121. - Reilly P. L., B. O, Maarten van Ormondt, Elias. E, Ruggiero. P, Erikson. L. H., Hapke. C, Collins. B. D., Guza. R. T., Adams. P. N. and Thomas. J, (2009), The Framework of a Coastal Hazards Model—A Tool for Predicting the Impact of Severe Storms, Coastal Hazards Task of the USGS Multi-Hazards Demonstration Project in Southern California. - Simonovic S., Dempsey. T., M. Harrison, (2000), Participatory floodplain management in the Red River Basin, Canada.

1076