Authors

Kailun Li, Yuxiao Jiang, Renwei Fang

Research Question

How can climate risk factors such as heatwaves, flooding, and drought be effectively analyzed and communicated to support infrastructure resilience and policy planning in NYC?

This capstone project addresses two interconnected challenges:

1. Risk Identification: Understanding the spatial and temporal dimensions of climate-related hazards and their impacts on critical infrastructure and vulnerable populations across NYC.
2. Knowledge Synthesis and Communication: Leveraging large language models (LLMs) to extract and summarize relevant scientific literature, and creating interactive visualizations that translate complex data into accessible insights for policymakers, city agencies, and the public.

The scope includes both data-driven urban risk analysis and AI-assisted synthesis to inform the NPCC's Fifth Assessment Report and its external communication tools.

Background

NYC faces increasing risks from climate change, including more frequent and intense heatwaves, coastal and inland flooding, and prolonged droughts. These hazards threaten critical infrastructure systems such as transportation, energy, water, and healthcare—especially in densely populated and socioeconomically vulnerable neighborhoods.

The NYC Panel on Climate Change (NPCC) started in 2009 and was codified in Local Law 42 of 2012 with a mandate to provide an authoritative and actionable source of scientific information on future climate change and its potential impacts. The NPCC is an independent advisory body that regularly publishes assessment reports that synthesize several years of scientific research and analysis on climate change and advises City policymakers on local resiliency and adaptation strategies to protect against rising temperatures, increased flooding, and other hazards. The NPCC has been central in advancing localized climate risk assessments and guiding city-wide adaptation strategies.

This capstone project supports the development of visual and textual analyses of key climate risk factors affecting NYC. This work directly informs the NPCC's Fifth Assessment Report and enhances public-facing communication tools hosted on the NPCC website.

Methodology

 This capstone project integrates two methodological components:

1. Urban Climate Risk Analysis: Students analyze publicly available datasets such as NPCC climate projections, FEMA flood zones, NYC Open Data, and infrastructure maps to assess spatial distributions of climate hazards. Geospatial and statistical analysis tools (e.g., QGIS, Python, R) are used to identify hotspot areas, infrastructure interdependencies, and population vulnerabilities under various future scenarios.
    
2. AI-Assisted Literature Synthesis: To support the evidence base for the Fifth Assessment Report, students explore the use of large language models (LLMs) to extract and synthesize scientific findings from peer-reviewed publications, government reports, and gray literature. This involves designing and testing simple LLM-based pipelines to identify relevant studies, extract key insights, and summarize evidence aligned with NPCC research priorities.

Deliverables

• Interactive data visualizations that translate climate risk information into accessible formats for policymakers and the public.

• Spatial and scenario-based analysis of climate hazards such as heatwaves, flooding, and drought, using real-world climate projections and infrastructure data.  

• Prototype of a literature synthesis tool using LLMs to identify and summarize key findings from scientific literature relevant to climate risks in NYC.

Data Sources

• New York City Panel on Climate Change (NPCC) climate projection data for NYC