Hub Bound Travel Data

Analysis and visualization of NYMTC's "Hub Bound Travel" reports — vehicle and passenger volumes into Manhattan's Central Business District (below 60th St), 2014–2024.

Live site: cbd.hudcostreets.org

Reports are also mirrored in Hudson County Complete Streets' Google Drive.

Charts

NJ→NY (entering)	NY→NJ (leaving)

Each GIF cycles through peak 1hr, peak period (3hr), and 24hr views. Click to open the interactive chart.

	NJ→NY	NY→NJ
Peak 1hr
Peak period
24hr

Data

8 normalized JSON files extracted from NYMTC AppendixII and AppendixIII Excel workbooks, totaling ~133k records. All extractions are idempotent, DVX-tracked, and cached to S3.

File	Records	Source	Description
crossings.json	528	Tables 14A-C, 15A-C	NJ sector per-crossing volumes by mode, direction, time period
modes.json	165	Quick Reference Table	NJ sector day volumes by mode
vehicles.json	2,158	Tables 16-17	All 4 sectors, per-crossing motor vehicle + bus counts
bus_passengers.json	1,191	Tables 18-19	All 4 sectors, local/express bus passenger counts
hourly.json	8,736	Tables 21A-27	Hourly profiles by mode, sector, and aggregate category
peak_accumulation.json	148	Tables 21B-23B	Peak persons in Manhattan Hub, 1973–2024
appendix_iii.json	17,606	AppendixIII A-F	Sector-level hourly data (bus, subway, rail, ferry)
appendix_iii_detail.json	102,581	AppendixIII B-G	Per-facility/line hourly data (subway lines, rail, roads, bikes)

See specs/extraction.md for full schema documentation and extraction notes.

Visualizations

The web app has 5 interactive charts built with Plotly.js and pltly:

Chart	Description	Controls
Unified Chart	NJ→NY crossings: bubble scatter, grouped bars, stacked %, recovery index	View, direction, time period, granularity, palette
Hourly Profile	Stacked area of CBD entries/exits by hour	Mode vs sector, direction, year
Mode Share	Stacked bar of daily persons by mode across years	Absolute vs %, direction
Sector Vehicles	Motor vehicles by sector (stacked or grouped bars)	View, direction, time period
Peak Accumulation	50-year timeline of max persons in Manhattan Hub	(annotated, no toggles)

All charts support legend hover highlight and click-to-solo via pltly.

Extraction

pip install -e .
./extract.py extract    # Extract all tables → data/*.json
dvx push                # Push to S3 cache

Requires the source Excel files in data/source/ (see .dvc files for provenance).

Web app

cd www
pnpm install
pnpm dev                # Dev server on port 3847

License

Data sourced from publicly available NYMTC Hub Bound Travel reports.