Authors

Nissim Ram, Darrel Lung, Shangyi Guo

Research Question

How do different flood mitigation strategies alter the spatiotemporal characteristics of flood impacts across New York City, and how can those effects be quantified to guide infrastructure investment decisions?

Background

Urban flooding is a significant and growing challenge, exacerbated by climate change, aging infrastructure, and expanding urbanization. Evaluating the impact of these floods can be costly, time consuming and lacking in spatiotemporal granularity. This project aims to develop a tool for rapid flood impact evaluation across a varied set of metrics including: critical infrastructure disruption, mobility impairment, neighborhood-level effects, and economic impact. In addition, the tool offers the ability to simulate impact mitigation interventions such as permeable pavements and green infrastructure to quantify future impact reduction potential. The tool leverages a unique 5 year dataset of hyperlocal flood depth profiles from the city’s FloodNet sensor network and the output of a cutting edge hydrodynamic model as well as incorporating open NYC datasets.

Methodology

Flooding in dense urban areas like NYC disrupts transit, damages property, and disproportionately impacts vulnerable populations. With climate change amplifying storm intensity, timely and precise flood mitigation strategies are increasingly necessary. The approach of this capstone project integrates environmental sensor data with geospatial modeling and hydrologic theory. Using the FloodNet dataset, it reconstructs historic flood events at a hyperlocal level and compares these with hypothetical modeled interventions.

The methodology includes:

• Spatial-temporal preprocessing of sensor data using Python (e.g., Pandas, NumPy)
• Terrain and flow modeling using digital elevation maps and open-source GIS tools (e.g., QGIS, SWMM, …)
• Integration of land-use data and porosity models to estimate permeability and runoff
• Simulation of hypothetical interventions using parameterized drainage models
• Impact evaluation using infrastructure overlays (e.g., transit, schools, housing) and metrics like area flooded, time to drain, and network disconnectivity

This hybrid approach balances empirical data with lightweight physics-based modeling to enable scenario-based decision making for climate adaptation.

Deliverables

• Modular tool for simulating and comparing historical flood impacts and mitigation scenarios
• Spatial model of NYC surface hydrology built from elevation and land-cover data
• Visualizations of flood impacts and simulated intervention outcomes
• Case study application over selected neighborhoods or transit zones
• Documentation guiding city planners on how to adapt the tool to new events or strategies

Data Sources

The project requires several datasets, all of which are publicly available or provided through project collaborators:

1. FloodNet flood event profiles: A time series dataset giving approximate minute resolution of hyperlocal flood depth measured across a 200+ sensor network across NYC.
2. NYC DEM data: Digital Elevation Map of NYC at 1 foot XY resolution with a height error margin of approximately 9 inches (absolute height).
3. A set of Flood Mitigation Strategies with an associated hydrological model (e.g. could be as simple as a constant drainage model, or a constant drainage model with a saturation point, etc.). This will require finding empirically driven values/models from literature.
4. NYC Land Cover dataset: to get classifications for impervious surfaces and vegetative cover.
5. Critical Infrastructure Layers: transit networks, bus routes, hospitals, schools, or other essential services from NYC Open Data and MTA’s datasets.