US-SEER
US Spatial Environmental Exposure & Respiratory Risk Index — mapping PM2.5, toxic releases, and health outcomes across 3,142 U.S. counties.
US-SEER (US Spatial Environmental Exposure & Respiratory Risk Index) is a high-performance web platform and spatial analytics engine designed to investigate the complex relationships between ambient air quality (), toxic industrial releases, socioeconomic vulnerability, and respiratory disease mortality across all 3,142 U.S. counties.
Overview & Core Mission
What does US-SEER do? US-SEER maps and analyzes environmental health hazards across the United States at the county level. It connects multi-source environmental exposure data with public health outcomes to reveal environmental injustice, demographic disparities, and community risk factors.
- Multi-Source Data Aggregation: Unifies federal environmental, epidemiological, and demographic datasets indexed by 5-digit county FIPS codes.
- High-Resolution Spatial Analytics: Interactive choropleth maps, bivariate overlays, and CDC SVI scatter analysis across 2,953+ reporting counties.
- Simulated Policy Interventions: Real-time policy simulator modeling how pollutant reductions translate into projected lives saved and health improvements.
- Empirical Correlation: Statistical models assessing whether ambient particulate matter directly correlates with respiratory mortality after controlling for smoking prevalence and poverty.
Geospatial Architecture & Data Pipeline
Pulls and parses ambient sensor readings from the EPA Air Quality System (AQS), industrial release metrics from the EPA Toxic Release Inventory (TRI), and CDC respiratory registries.
Normalizes heterogeneous county definitions, boundary changes, and non-reporting tracts into standardized 5-digit FIPS codes using GeoPandas and TopoJSON.
Evaluates bivariate correlations between exposure and COPD/asthma mortality while controlling for smoking rates and median household income.
Streams vectorized choropleth geometry via react-simple-maps and executes client-side policy simulations with dynamic parameter sliders.
Core Capabilities
High-performance county-level vector map rendered with react-simple-maps and d3-scale, featuring smooth pan/zoom controls, dynamic color scales, and instant hover tooltips showing multi-variable breakdowns per county.
Toggle seamlessly between environmental exposure variables (, Toxic Release Inventory), health outcomes (COPD mortality, asthma prevalence), and CDC Social Vulnerability Index / demographic metrics (median household income, poverty rate, uninsured percentage).
Fast fuzzy search engine across 3,000+ U.S. counties, major metropolitan areas, and states with automated viewport centering and metric extraction.
Customizable parameter sliders allowing users to model hypothetical environmental policy interventions (such as reducing ambient emissions by 15%) and immediately project demographic-stratified health outcomes and lives saved.
Client-side PDF compilation suite enabling researchers, policy advocates, and educators to export custom county reports, map snapshots, and correlation findings.
Multi-Source Data Architecture
Analytical Approaches: Observation vs. Intervention
- Vector Choropleth: Visualizes spatial clustering of environmental hazards and respiratory mortality.
- CDC SVI Scatter: Identifies outlier communities where high socioeconomic vulnerability amplifies the physical harm of air pollution.
- Multivariate Regression: Controls for confounding variables such as regional smoking prevalence and healthcare access.
- Targeted Reduction Sliders: Model the epidemiological impact of lowering ambient by arbitrary percentages.
- Demographic Projections: Quantifies lives saved and preventable hospitalizations stratified by vulnerability quintiles.
- Evidence-Based Advocacy: Translates complex spatial statistics into actionable policy briefs for local regulators.
Technology Stack
| Layer | Technology | Purpose |
|---|---|---|
| Frontend Framework | Next.js 16 (App Router) | Server components, streaming rendering, and routing |
| Language | TypeScript & Python | Strict frontend type safety & data pipeline processing |
| Styling & Design | Tailwind CSS v4 & Lucide | Sleek dark-mode aesthetic with custom palette tokens |
| Mapping & Geospatial | React Simple Maps & D3-Scale | Interactive TopoJSON county rendering and choropleth color scales |
| Data Pipelines | Pandas, GeoPandas & NumPy | Automated data cleaning, FIPS alignment, and statistical modeling |
| Deployment | Vercel Edge Network | Optimized serverless distribution with .vercelignore payload limits |