Flood Susceptibility of Critical Infrastructure in NYC

Madison Buchholz, Ruixin Gan, Yifei Wang

Introduction

According to FEMA, flooding occurs when there is a partial or complete inundation of normally dry land area. Flooding may result from overflow of inland or tidal waters, unusual and rapid accumulation or runoff of surface waters, mudslides, or collapse of land along the shore of a body of water (Federal Emergency Management Agency, 2020). In 2022, flooding in the U.S. caused 2.8 billion USD in damages and caused 91 fatalities. It is important to understand the areas and critical infrastructure most at risk for flooding so mitigation measures can be put in place to reduce damages and fatalities (Burgueno Salas, 2023).

Background

New York City, which is built around different rivers, estuaries, and waterways, is extremely susceptible to flooding (Colle et al., 2008). Hurricane Sandy caused over $5 billion in damages to the NYC metro system, with the largest damages occurring to the subway system (Vermeij, 2016). Flooding from the storm also caused damage to almost every hospital facility in the city with serious damages occurring to Bellevue, South Brooklyn Health, and Metropolitan hospitals (New York Health and Hospitals, 2022).

More recently, on September 29, 2023, intense rainfall and flooding caused schools to close and made travel by subway impossible (Frost, 2023). Flooding is often caused when drainage systems are overwhelmed (McGeehan & Howard, 2023). Trees can help reduce flooding by absorbing water through their roots (“Calculating tree benefits for New York City.” n.d.). Flood risk zones are important to use when analyzing the possible impacts of future flooding and how these impacts may worsen with climate change (“NYC Flood Hazard Mapper,” n.d.).

Past occurrences of flooding have paralyzed New York City and caused significant economic losses (Vermeij, 2016). In the end,  we might formulate policy recommendations for city planners and local governments to integrate green infrastructure into urban development plans. This involves creating guidelines for the implementation of green infrastructure in areas identified as high-risk flood zones.

Case Study 1

Researchers at Georgia Institute of Technology analyzed floodplains intersecting with critical transportation networks such as expressways to determine flood susceptibility (Singh et al., 2023). They researched how different flood mitigation measures have resulted in particular flood outcomes in three different cities (New York City, Tokyo, and Rotterdam).

The researchers obtained GIS shapefiles of floodplains from government sources and used google street maps to identify the expressways. They then used intersect to determine which expressways in New York City were within the flood risk zones (Singh et al., 2023) This study was useful for transportation engineers, urban planners, and policymakers to learn about successful flood resilient infrastructure and mitigation strategies.

In comparison to our study, the researchers in this study only used one factor (the hundred year floodplains) to analyze flood risk in New York City, while we used four different factors. Also, while we examined the flood risk of MTA stations and public transportation within the city, they focused on expressways and travel by car.

We also both examined some of the flood mitigation measures used, as we focused on flood mitigation strategies within five at-risk areas within our study area. While this previous study is useful in analyzing which mitigation strategies could be useful to implement, the researchers could have used additional data to create an at-risk map combining many different risk factors. We combined four different factors in our study to assess the flood risk of New York City. There are many different factors that affect flood-risk and a much more extensive study could be developed combining many different variables.

Case Study 2

It is also valuable to examine flood risk analysis in a geographic location very different from New York City to compare similarities and differences. Researchers in the second case study assessed the flood risk in the Shebelle River Basin in southern Somalia. (Abdinour Osman & Das, 2023). They assessed which areas were at very high, high, moderate, low, and very low risk of flooding.

Their overall approach included using GIS to create two different maps: the flood susceptibility map and the flood vulnerability map. The flood susceptibility map indicated the likelihood that a particular area would be affected by flooding. To create this map, the researchers used raster data on elevation, slope, drainage density, distance to river, rainfall, and soil and geology. Their second map demonstrated flood vulnerability or the degree of damage that can be caused by a flood to a community. This map was created using data including land use, population density, distance to road, global man-made impervious surface, and human built-up area settlement extent.

After cleaning and organizing their data, the researchers used reclassify to group the raster data into certain values. Next, they converted the reclassification layers to a polygon and used dissolve. They assigned weights to each feature and then summed all the weighted variables to find the total score to display on the map. They combined both the flood susceptibility and flood vulnerability maps to create an overall flood risk map and were successfully able to locate the areas of high-risk in the Shebelle River Basin (Abdinour Osman & Das, 2023).

We used some of the same factors in our approach including drainage density, population density, and distance to coast (similar to flood risk zones in our analysis). In a more extensive study, we could include additional factors such as elevation, slope, and land use. However, it is important to note that elevation may be a more significant factor in the case study as Somalia ranges from 3-2821m while New York City has a much smaller elevation range from approximately -9-117m (Topographic-Map, n.d.)

Research Question

Our research question included three parts.

  1. Which areas of New York City are most susceptible to flooding?
  2. How does critical infrastructure such as hospitals, schools, and MTA stations overlap with these high-risk zones?
  3. What flood mitigation strategies are employed in each high-risk zone?

Objectives & Strategy

Our main goal of this project was to create a flood susceptibility map of New York City that could be used to determine which areas of the city are most at-risk for flooding. To accomplish this, we used four different variables: population density, proximity to drainage systems, proximity to parks, and location within flood risk zones. We used the reclassify tool to group each layer on a scale from one to four, with four indicating the most at-risk for flooding.

  • Population Density: We calculated the population density in each census tract by dividing the number of people by the total area. We then converted the data to raster data and reclassified on a scale of one to four. Areas with a value of four are the most population-dense and the most at-risk for flooding. 

  • Parks: We used Euclidean Distance to calculate the distance from a park included in the data set. The data were classified on a scale of one to four with the areas furthest from a park classified as a four for being most at-risk for flooding. 

  • Drainage Systems: We used Euclidean Distance to calculate the distance from drainage infrastructure. The data were classified on a scale of one to four with the areas furthest from the drainage infrastructure receiving a score of four for being most at-risk for flooding. 

  • Floodplains: We used Euclidean Distance to calculate the distance from a floodplain area. The data were classified on a scale of one to four with the areas closest to the floodplains classified as four for most at-risk for flooding.

Although there are many factors that could be included in a flood risk assessment, we chose these four factors because we believe they all are important to consider when assessing flood risk in New York City. For example, flooding in the city is often caused by overwhelmed drainage systems. Also, the distribution of parks and greenspaces is very inequitable within the city, causing certain areas to be more flood susceptible than others (Trust for Public Land, n.d.)  To finish our analysis, we used equal weights for all four factors and summed the values to calculate a total flood susceptibility score.  

Why do we need to use GIS?

It was important to use Geographic Information Systems (GIS) to answer our research question for many reasons. First, we used raster data to show continuous data in a spatial context. Since we used raster data, we were able to use the Euclidean Distance tool to calculate the distance to the closest feature of a specific layer. Next, using raster data allowed us to use the raster calculator with a map algebra expression to calculate the total flood susceptibility score. Lastly, GIS allowed us to overlay the critical infrastructure with the total flood susceptibility score to see spatially which infrastructure was most at-risk for flooding.

Preview: Final Flood Susceptibility Score

Our final map clearly shows the areas of New York City most at-risk for flooding. For symbology, we used a color scale from green to blue. Light green areas signify the areas that are least at risk for flooding while dark blue areas signify the areas most at-risk for flooding. As shown by the map, five areas at-risk for flooding include Lower Manhattan, Southern Brooklyn, Manhattan East Side, East Harlem, and Western Brooklyn. The resulting total flood susceptibility scores ranged from four to thirteen.

Methodology

Data Acquisition and Preparation

 Data Sourcing: Extensive data collection from various sources to gather spatial data and attributes relevant to the project scope.

CSV Data Importation and Display: Importation of CSV files for geospatial display as XY data points.

Population Density

Data Analysis and Processing:

 Geographic Projection: Confirmation and projection of data sets to the NAD 1983 StatePlane New York Long Island FIPS 3104 coordinate system..

Symbology Adjustment: Refinement of data layer symbology for enhanced visualization and interpretability, including graduated colors for population density as depicted in the screenshots.

Population Concentration Analysis:

Population Data Normalization: Standardization of population data against the shape area for consistency across different spatial scales.

Raster Data Generation: Conversion of polygonal census tract data to raster format for spatial analyses, specifying the value field and cell size for the raster data set.

Spatial Analysis:

 Field Calculations: Implementation of field calculations using Python expressions to derive population density.

Raster Attribute Table Creation: Generation of a raster attribute table to facilitate data manipulation and querying.

Population Density Analysis:

 Population Density Calculation: Computation of population density by dividing the population count by the area of the census tract.

Graduated Color Symbology: Application of graduated color symbology to the population density data to visually discriminate between different density values.

Raster Data Classification: Classification of raster data using the Natural Breaks (Jenks) method to categorize the population density into discrete classes for visualization and analysis.

Data Reclassification: Reclassification of the population density raster data to assign new values for improved data representation and analysis.

Final Flood Susceptibility Score Calculation:

Raster Calculator: Calculation of a map algebra expression which adds the scores created by the four previous layers (population density, flood risk zones, parks, and drainage systems). 

Symbology: Use of a stretch color ramp from green to blue to demonstrate the flood susceptibility score of each area. Light green indicates low flood susceptibility and dark blue

Final Visualization: 

Finalization of the visual representation of population density scores for inclusion in the StoryMap, ensuring clear communication of data patterns and findings.

Flow Chart

Population Density

The risk of flooding in densely populated areas could have severe implications for a large number of people, property, infrastructure, and critical services.

Color Coding

  • Darkest Shade (Value 4): This color usually marks the most densely populated areas, often the urban core of the city. In NYC, this would typically correspond to parts of Manhattan and perhaps some dense areas of Brooklyn and Queens.
  • Darker Medium Shade (Value 3): These are probably high-density residential neighborhoods or areas with mixed residential and commercial use. They are more densely populated but not the most dense.
  • Light Medium Shade (Value 2): These areas are more populated than the lightest areas but still not as dense as the urban core. They might be suburban or mixed-use areas.
  • Lightest Shade (Value 1): This likely represents the areas with the lowest population density. These areas may include parks, industrial zones, or low-density residential neighborhoods.

Geographic Distribution

The most densely populated areas (darkest red) are concentrated in Manhattan and parts of Brooklyn and Queens that are closer to Manhattan. This is expected as Manhattan is the most densely populated borough in NYC. Staten Island and the areas further out from the city center show the lowest density values.

Hundred-Year Floodplain Area

Understanding the proximity to the 100-year floodplain is important for zoning, insurance, and urban planning, especially in New York City.

Color Coding

  • Darkest Shade (Value 4): This color indicates areas that are directly within the 100-year floodplain. These locations have the highest risk of flooding, with a 1% chance of a flood event occurring each year. They are most likely to be affected by severe weather events and may include regions at or below sea level, or areas that have historically been prone to flooding
  • Darker Medium Shade (Value 3): Zones with this score are close to the 100-year floodplain and may experience similar risks of flooding, especially during significant weather events. These areas might also be affected by overflow or rising water levels from the nearby high-risk zones.
  • Light Medium Shade (Value 2): These areas are further from the 100-year floodplain but still within a range that could be affected under severe conditions. The risk of flooding is lower compared to the darkest shades, but there is still a need for caution.
  • Lightest Shade (Value 1): This is the lowest risk category on the map and represents areas that are at a significant distance from the 100-year floodplain. The risk of flooding is minimal compared to other categories.

Geographic Distribution

The areas within the floodplain are mainly along the coastlines of the city, including areas around Staten Island, Brooklyn, Queens, and the Lower Manhattan region. The floodplain areas follow the contour of the rivers and the coast, indicating that the primary concern for flooding is from storm surges and rising water levels along these areas.

Drainage System

Areas with fewer drainage facilities are at greater risk of flooding, especially during heavy rainfall, due to the limited capacity to channel and drain excess water efficiently.

Color Coding

  • Darkest Shade (Value 4): This color indicates the areas with the scarcest drainage facilities, implying a higher risk of flooding. These could be regions where development has outpaced the installation of sufficient drainage, or perhaps areas where the existing infrastructure is old and has not been updated or expanded to meet current needs.
  • Darker Medium Shade (Value 3): Regions with this score still face a risk of flooding. The infrastructure may be strained during heavy rainfall or may not cover all the areas adequately.
  • Light Medium Shade (Value 2): These zones have a moderate level of drainage facilities. While there is still a risk of flooding, the infrastructure is likely to be more capable of handling typical rainfall events.
  • Lightest Shade (Value 1): Areas with the most drainage facilities, suggesting the lowest relative risk of flooding.

Geographic Distribution

High-risk areas with a scarcity of drainage facilities and darker shades on the map include parts of Staten Island (Richmond) prone to coastal flooding, Southern Brooklyn neighborhoods like Coney Island and Sheepshead Bay near the waterfront, and Eastern Queens areas such as Jamaica and Flushing Meadows, which are close to water bodies. Moderate-risk areas cover some regions along the Harlem and Bronx Rivers in the Bronx and neighborhoods like Inwood and Washington Heights in Northern Manhattan, where drainage infrastructure is more present but may not fully prevent flooding.

Parks

In urban environments, parks and green spaces are critical for sustainable stormwater management and can significantly reduce flood risks by absorbing rainwater.

Color Coding

  • Darkest Shade (Value 4): These areas, being the furthest from parks, could have higher flood risk as there are fewer green spaces to absorb rainfall. This could lead to greater runoff during storm events, increasing the likelihood of flooding.
  • Darker Medium Shade (Value 3): Regions with this score have reduced access to parkland, which may correlate to a somewhat increased flood risk due to less natural land for water management.
  • Light Medium Shade (Value 2): Areas with moderate proximity to parks might experience a lower flood risk.
  • Lightest Shade (Value 1): The closest to parks, these areas are likely to have the lowest flood risk in the urban environment.

Geographic Distribution

In New York City, the distribution of parks correlates with flood risk assessment, where higher scores indicate fewer parks and potentially higher flood risk due to less water absorption capacity. Areas like Central and Southern Queens, parts of the Bronx, and Northern Brooklyn, marked by darker shades indicating higher scores, may face elevated flood risks due to the scarcity of green spaces.

Final Map: Flood Susceptibility Score with Critical Infrastructure

01 / 06

5 Areas Identified for Potential Flood Risk

Lower Manhattan

The infrastructure here is highly developed with many MTA stations, but faces challenges due to its proximity to water bodies, making it vulnerable to flooding.

  • LMCR

The Lower Manhattan Coastal Resiliency (LMCR) Project is an integrated coastal protection initiative aimed at reducing flood risk due to coastal storms and sea level rise in Lower Manhattan. The LMCR Project area spans the Lower Manhattan coast and seeks to increase resiliency while preserving access to the waterfront and integrating with public space (Lower Manhattan Coastal Resiliency (LMCR), n.d.).

Southern Brooklyn

The area is largely residential, with some key neighborhood commercial corridors, destination attractions, and large institutional presences. Southern Brooklyn also houses critical infrastructure for transportation, education, and wastewater treatment.

  • Southern Brooklyn Community Rebuilding and Resiliency Plan  
    • Parks and Beaches Resilience: The city is working on restoring and nourishing beaches. Shoreline parks are being modified to better withstand flooding, with measures like hardening infrastructures.
    • Healthcare System Enhancement: This involves protecting essential systems like electrical and emergency power. 
    • Transportation Infrastructure Improvement: Key steps are being taken to rebuild and resurface roads damaged by Sandy, integrating resiliency features to prevent future damage (Island & Brooklyn, n.d.).

Manhattan East Side

The East Side of Manhattan, a densely populated area of New York City, is characterized by a mix of residential, commercial, and cultural spaces. The infrastructure there is mainly public transit and educational institutions.

  • ESCR

The East Side Coastal Resiliency (ESCR) Project is aimed at reducing flood risk due to coastal storms and sea level rise on Manhattan's East Side from East 25th Street to Montgomery Street. The boundaries of this project correspond with the natural "pinch-points" in the 100-year floodplain: areas where the land is higher along the coastline, making it easier to close the system off from water entering from the north and south. The project design integrates flood protection into the community fabric, improving waterfront open spaces and access (East Side Coastal Resiliency, n.d.).

East Harlem

In East Harlem, a resiliency report that cost $1 million to produce,  remains shelved. In it, experts envisioned the possible rebuilding of East 106th Street as a “green corridor” with a mid-road stream or ditch that would allow better stormwater drainage. The neighborhood has seen some resiliency projects move forward, including building flood walls around Metropolitan Hospital on First Avenue. However, there has been no comprehensive approach to protecting the entire area (Smith, 2022).

Western Brooklyn (historical "South Brooklyn")

The infrastructure here, while robust in terms of transport connectivity, faces challenges due to its industrial past and ongoing redevelopment projects.

  • RHCR

The Red Hook Coastal Resiliency Project is a $100 million initiative aimed at protecting Red Hook, Brooklyn, from coastal flooding and sea level rise. The plan involves installing flood walls, gates, and regrading streets to enhance drainage (Foley & Mayor, 2022).

  • Brooklyn Greenway Initiative (BGI) 

The Brooklyn Waterfront Greenway is envisioned as a 14-mile multi-use trail from Greenpoint through Sunset Park; connecting the proposed Queens waterfront greenway with Brooklyn’s Shore Parkway Esplanade  (Brooklyn Waterfront Greenway, n.d.).

Limitations

After completing the very first GIS project in our life, as potential GIS analyst, our study has certain limitations:

While focusing on the impact of parks on flooding, we also need to look at green lands. 

Also, the current research focuses exclusively on parks and their influence on flood mitigation. This approach potentially overlooks the significant role of green lands, which may offer different or complementary ecological benefits in flood management. Future studies could benefit from incorporating diverse types of green spaces to provide a more comprehensive understanding of urban flood dynamics.

Factor Selection: Flooding is a complex phenomenon influenced by lots of factors beyond the scope of our study. Future research could enhance the analysis by incorporating a broader range of variables, such as soil permeability, urban infrastructure, climate patterns, and vegetation types, to develop a wider understanding of flood risks and mitigation strategies.

Interactive Story Map Features: While the current story map provides a narrative of the findings, incorporating interactive elements such as pop-ups, as seen in other projects, could significantly enhance user engagement and understanding. These interactive features can provide additional context, data insights, or case studies, thereby enriching the storytelling aspect of the GIS analysis.

Predictive Modeling: The study currently lacks a predictive component, which limits its applicability in proactive flood management. Incorporating flood regression models in future research could enable predictions of flood risks under varying scenarios. This predictive approach would be valuable in urban planning and disaster preparedness, allowing for the simulation of flood events under different environmental and developmental scenarios.

Conclusion

The report offers a spatial analysis of flood risk of critical infrastructure across New York City. It emphasizes the role of drainage systems, proximity to parks, and location within flood risk zones in shaping this susceptibility. Through GIS analysis and case studies, it provides insights into flood mitigation strategies and recommends integrating green infrastructure into urban planning to enhance resilience, especially in the context of climate change and rising sea levels.

Further Information

Data List

References

Abdinour Osman, S. & Das, J. (2023). GIS-based flood assessment using multi-criteria decision analysis of Shebelle River Basin in southern Somalia. SN Applied Sciences, 5, 134-151.   https://doi.org/10.1007/s42452-023-05360-5 

Burgueno Salas, E. (2023, November 1). Economic damage caused by floods in the United States 1995-2022. Statistica. https://www.statista.com/statistics/237420/economic-damage-caused-by-floods-and-flash-floods-in-theus/#:~:text=In%202022%2C%20floods%20 and%20flash,damage%20in%20the%20previous%20year

City of New York Parks and Recreation (n.d.). Calculating tree benefits for New York City.   https://www.nycgovparks.org/sub_your_park/trees_greenstreets/images/treecount_report.pdf

Colle, B.A., Buonaiuto, F., Bowman, M.J., Wilson, R.E., Flood, R., Hunter, R., Mintz, A., & Hill, D. (2008). New York City’s vulnerability to coastal flooding: Storm surge modeling of past cyclones. Bulletin of the American Meteorological Society, 89 (6), 829-842.  https://doi.org/10.1175/2007BAMS2401.1 

Federal Emergency Management Agency. (2020, July 7). Flood. https://www.fema.gov/glossary/flood

Frost, M. (2023, September 29). Brooklyn hit hard in historic NYC rainfall: Mayor Adams criticized for a “delayed and insufficient” response. Brooklyn Daily Eagle. https://brooklyneagle.com/articles/2023/09/29/brooklyn-hit-hard-in-historic-new-york-city-rainfall/

Jones, E. (2023). Cars in floodwater on the FDR highway in Manhattan, New York on September 29, 2023. [Photograph].  https://www.foxweather.com/weather-news/flash-flooding-new-york-city-northeast 

McGeehan, P., & Howard, H. (2023, September 29). Why New York City keeps flooding: When the city’s drain network is overwhelmed, “it backs up,” experts say. New York Times. https://www.nytimes.com/2023/09/29/nyregion/nyc-sewer-system-infrastructure.html

New York City Department of City Planning (n.d.). NYC Flood Hazard Mapper. https://www.nyc.gov/site/planning/data-maps/flood-hazard-mapper.page#:~:text=A%20product%20of%20the%20New,the%20future%20with%20climate%20change

New York Health and Hospitals. (2022, October 28). Ten years after Hurricane Sandy, NYC Health + Hospitals makes progress on majority of resiliency projects. https://www.nychealthandhospitals.org/pressrelease/ten-years-after-hurricane-sandy-nyc-health-hospitals-makes-progress-on-majority-of-resiliency-projects/#:~:text=The%20most%20significant%20physical%20damage,power%20outages%20and%20wind%20damage

Singh, P., Amekudzi-Kennedy, A., Woodall, B., & Joshi, S. (2021). Lessons from case studies of flood resilience: Institutions and built systems. Transportation Research Interdisciplinary Persceptives, 9, 1-9.  https://doi.org/10.1016/j.trip.2021.100297 

Shabelle Foundation (n.d.). [Shabelle River Basin]. [Photograph].  http://shabellefoundation.org/wp-content/uploads/2013/09/flooding9.jpg 

Trust for Public Land (n.d). 2023 ParkScore index: New York, NY.  https://parkserve.tpl.org/downloads/pdfs/New%20York%20City_NY.pdf 

Vermeij, D.A.A.N. (2016). Flood risk reduction interventions for the New York City subway system (Publication No. 1382284) [M.Sc. thesis, Delft University of Technology]. http://resolver.tudelft.nl/uuid:db184167-b1b7-4dd1-b7f0-623df2352fe0

Williams, B. (2023). Cars lie in floodwater on the FDR Drive just south of E. Houston St. on Sept. 29, 2023, in Manhattan. [Photograph].  https://www.nydailynews.com/2023/09/29/new-york-city-torrential-downpours-flash-floods/ 

Lower Manhattan Coastal Resiliency (LMCR). (n.d.). Www.nyc.gov. https://www.nyc.gov/site/lmcr/index.page

Island, C., & Brooklyn. (n.d.). A STRONGER, MORE RESILIENT NEW YORK. Retrieved December 20, 2023, from https://www.nyc.gov/assets/sirr/downloads/pdf/Ch17_SouthernBrooklyn_FINAL_singles.pdf

East Side Coastal Resiliency. (n.d.). Www.nyc.gov. https://www.nyc.gov/site/escr/index.page

Smith, R. H. (2022, October 28). Low-Lying East Harlem Dodged Sandy’s Worst But Neighborhood’s Still Not Ready for Next Storm. THE CITY - NYC News. https://www.thecity.nyc/2022/10/28/east-harlem-dodged-sandy-not-ready-next-storm/

‌Foley, T., & Mayor, E. (2022). RED HOOK COASTAL RESILIENCY (RHCR) PUBLIC DESIGN COMMISSION PRESENTATION FOR PRESENTATION PURPOSES ONLY PDC PRELIMINARY REVIEW. https://www.nyc.gov/assets/designcommission/downloads/pdf/08-08-2022-pres-DDC-p-RedHookResiliency.pdf

‌Brooklyn Waterfront Greenway. (n.d.). RPA. Retrieved December 20, 2023, from https://rpa.org/work/reports/brooklyn-waterfront-greenway-a-concept-plan-for-community-boards-2-6

‌

‌

Flow Chart