VayuDrishti
See the air. Understand the impact.
VayuDrishti (वायुदृष्टि: vayu = air, drishti = sight) is a tool for studying air quality and traffic in a city. You count vehicles and measure dust at a few spots, and it ranks the spots, draws the charts and shows how traffic and dust move together. Next to your own field data it shows live air quality and live traffic for 46 places across Delhi NCR, with a plain-language note on what the air means for people.
No frameworks, no build step, no dependencies: plain HTML, CSS and JavaScript, plus a tiny Node server (or two serverless functions on Vercel) that keeps your API keys off the browser.
What you can do
Take your own readings (Observations)
- Field counter: stand at a spot and tap one button per vehicle (cars, bikes and autos; trucks and buses). A 5, 10 or 15 minute timer starts on the first tap. At the end you get vehicles per hour and the heavy-vehicle share, and one button fills the form.
- Dust calculator: turn milligrams on a plate, its area and the days left out into mg/m²/day.
- Records: add, edit and delete observations; sort by any column; search by place, date or notes. Data lives in your browser.
- CSV export and import: move observations between devices. Imports are checked row by row; duplicates and demo rows are skipped and nothing existing is overwritten.
- Sample data: clearly labelled demo records to explore with, removable at any time without touching your own data.
Understand them (Dashboard, Charts, Hotspots)
- A summary, filters by place and date, and key numbers that count up when they change.
- Charts of traffic and dust over time, traffic against dust, and a comparison of places.
- Every place gets a hotspot score from 0 to 100 and a rank (see below).
See the air right now (Live)
- Pick one of 46 places in Delhi NCR (no latitude or longitude needed) to get the US AQI from IQAir and speed and congestion from TomTom. "Somewhere else" accepts coordinates, and "Use my location" fills them.
- A ring shows the AQI, with a plain-language note for that band based on the US EPA's categories. The page's background haze takes the colour of the air.
- Every reading is saved in your browser and drawn as a trend for that place. An optional 15 minute auto-refresh pauses while the tab is hidden.
Ask people (Health survey)
- An anonymous survey that stores counts only. Export your totals and add other people's files to combine a group.
Share the result
- Print report: a clean, dated report on white paper (or "Save as PDF" from the print dialog), whatever theme you were using.
Comfortable to use
- Light and dark themes, phone to desktop, keyboard friendly, honours reduced motion and reduced transparency. See DESIGN.md.
Quick start
Needs Node.js 20.12 or newer. There is nothing to install.
cp .env.example .env # then add your two API keys (see below)
npm start # same as: node server.js
Open http://localhost:8000. Everything except Live also works by simply opening public/index.html in a browser.
New to the idea? Press Load sample data to see a finished study, then read the user guide.
API keys
The Live section needs two keys, and the Stations section needs a third (all have free plans):
IQAIR_API_KEY=...
TOMTOM_API_KEY=...
XKDR_API_KEY=... # https://airquality.xkdr.org/signup
Put them in .env, which is git-ignored. Without XKDR_API_KEY everything else still works; the Stations section just says the key is missing. The server adds them to the provider calls itself, so they never reach the browser or the repository. When you deploy, set them as environment variables instead (see docs/DEPLOY.md).
- The AQI comes from the nearest IQAir station city, so places in one city often show the same number. The traffic numbers are for the road at the place you picked, so they differ.
- TomTom reports speed and congestion, not vehicle counts. It complements the vehicles per hour you count yourself.
- Free IQAir keys are small. The server reuses a reading for 10 minutes, limits each visitor and stops at a daily cap, so a busy page cannot use up your quota.
How the hotspot score works
score = 40% dust + 40% traffic + 20% heavy-vehicle share
Each part is the place's average measured against a fixed reference (300 mg/m²/day, 2000 vehicles/h, 30% heavy) and capped at 1, so a score does not change just because another place was added. Below 40 is Low, 40 to 69 Moderate, 70 and above High. The references are teaching values you can change (REF in public/app.js); this is not an official air quality index.
Deploy
One small Node process (or two serverless functions) and no database. Vercel is the recommended host: import the repository, add the keys as environment variables, deploy. docs/DEPLOY.md has step-by-step instructions for Vercel, Render, Docker and any Node host, the full list of settings, and what to do when something goes wrong.
This is the beginning of the README. Read the rest at the original. Read the full README ↗
