🌿 Earth Pulse is live from Baner, Pune · 37,000+ climate records · — bird species · Building Phase 2 now
Citizen science · Open data · Western Ghats

Help us observe
the living Earth

Earth Pulse is a hyperlocal environmental observatory in Baner, Pune, growing into a network across the Sahyadri foothills. Join us.

Open source · Free data · Started January 2022 · Get involved →

—
Climate records
—
Bird detections
—
Species detected
3+
Years of data
1
Active node
678m
Baner Hill elevation
Why this place matters

Baner Hill is the third highest point within Pune city limits, at 678m (2,224 ft). It is part of the Western Ghats — a UNESCO World Heritage Site and one of Earth's eight biodiversity hotspots. These mountains are older than the Himalayas, contain 30% of India's species in under 6% of its land, and are being monitored by almost nobody at the hyperlocal level.

Earth Pulse is the first continuous, open, citizen-science environmental observatory on this edge — recording hourly air quality, temperature, acoustic bird diversity, and hill photography since January 2022.

228
Bird species
possible here
Project timeline

Where we are, where we're going

Earth Pulse is being built in phases. Each phase adds new sensors, new data streams, and new community capability.

✓
Phase 1 · Complete · Jan 2022 – Present
Air quality, microclimate & acoustic bird monitoring
Hourly EU AQI, India AQI, PM2.5/PM10, temperature, humidity, wind, solar radiation, and precipitation — logged continuously from one Pi 3B+ in Baner. BirdNET acoustic AI running on a second Pi 3B, detecting and identifying birds by their calls in real time. 37,000+ hourly records. 35+ species detected. All data is public and downloadable.
Raspberry Pi 3B+ BirdNET AI Google Air Quality API Open-Meteo ECMWF Supabase · Cloudflare
▶
Phase 2 · In Progress · April 2026
A Second sensor array — facing the Sahyadri
A second Raspberry Pi 3B deployed directly facing Baner Hill (South). This node adds four new data streams simultaneously: a second BirdNET mic capturing the hill-facing soundscape; hourly PM2.5/PM10 from an SDS011 sensor for cross-validation; a timelapse camera capturing the hill every 3 hours through all seasons; and a Soundscape Index (NDSI) that measures the ratio of natural to human sounds — a direct proxy for ecosystem health.
Pi 3B · South balcony Second BirdNET mic SDS011 PM2.5 sensor Pi Camera · 3h timelapse NDSI soundscape index
Baner Hill Bioacoustic Index & Bird Species Radar
With two mics running at different orientations, we will compute the Baner Hill Bioacoustic Index — a weekly 0–100 score combining soundscape health (NDSI), species richness, and detection frequency. We will also build the first acoustic bird direction radar for the Baner-Pashan biodiversity corridor, showing which species are detected from the hill direction vs the city. This will be published weekly.
Bioacoustic Index Bird species radar Dual-mic triangulation Baner–Pashan corridor
Remote water & soil sensors + community network
A Pi Zero 2W node, solar-powered, at Baner Lake — measuring water level, turbidity, temperature, and conductivity. Correlating water health with bird detections at the lake (waterbird species are directly tied to water conditions). Simultaneously, open the platform to community contributors who host their own sensor nodes, forming the first distributed environmental mesh across the Baner–Pashan–Aundh ridge.
Pi Zero 2W Solar-powered Baner Lake water sensors Community node network
Research data API & Western Ghats biodiversity archive
A public JSON API at earthpulse.pages.dev/api — live AQI, temperature, bird detections — consumable by researchers, journalists, and citizen apps. Long-term phenology records (when does the Flame of the Forest bloom? when do the swifts arrive?) becoming a climate change signal dataset unique to this latitude and elevation. Formal collaboration with Pune-based ornithology and ecology groups.
Public research API Phenology archive Academic collaboration Open dataset CC-BY 4.0
Coming soon

What's being built right now

These features are in active development as part of Phase 2.

🏔️
Baner Hill timelapse
A Pi Camera on the south balcony clicking one photo every 3 hours — 6am, 9am, noon, 3pm, 6pm. Every photo annotated with AQI, temperature, humidity, and rainfall at the exact moment it was taken. Watching the hill turn brown in April, explode green in July, go blue-grey with monsoon fog.
Phase 2 · In progress
🎙️
Soundscape health index (NDSI)
The Normalized Difference Soundscape Index measures biophony (natural sounds: birds, insects, rain) vs anthrophony (human noise: traffic, construction). A score near +1 means a healthy natural soundscape. Near -1 means urban noise dominance. Computed hourly from the hill-facing mic. The first long-term soundscape health record for Baner.
Phase 2 · In progress
🧭
Bird species radar
With two mics at different orientations, we will build the first acoustic bird direction radar for the Baner–Pashan biodiversity corridor. Which species come from the hill? Which from the city side? A circular chart showing spatial bird activity — never done acoustically at this resolution in Pune.
Phase 3 · Planned
📊
Baner Hill Bioacoustic Index
A single weekly score (0–100) combining NDSI soundscape health, bird species richness, and detection frequency from the hill-facing node. Easy to share, easy to understand, meaningful over years. Is Baner's natural soundscape getting healthier or sicker? This answers it.
Phase 3 · Planned
🔌
Research data API
A public JSON endpoint — live AQI, temperature, top bird species detected — consumable by researchers, other apps, and journalists. earthpulse.pages.dev/api/latest returns current conditions. Free, open, no key required.
Phase 5 · Planned
💧
Baner Lake water monitor
A solar-powered Pi Zero 2W at Baner Lake measuring water level (cm), turbidity (clarity), temperature, and dissolved solids. The lake is a migratory bird stop — when its level drops, waterbird detections change. This will be the first continuous water health record for the lake.
Phase 4 · Planned
Get involved

Three ways to contribute

You don't need to be a developer. All three paths matter.

📸

Contribute photos

Have photos of Baner Hill, the Pashan lake area, or the Sahyadri foothills? Upload them to the hill archive. EXIF metadata is read automatically for the date and time. Every photo is annotated with the climate conditions at that moment.

Upload a photo →
🖥️

Host a sensor node

If you live in Baner, Pashan, Aundh, or nearby — and have a balcony or outdoor space — you can host a sensor node. Total hardware cost around ₹8,000–15,000 (one-time). Your data joins the network and appears on the map. Setup guide below.

See setup guide →
❤️

Support the project

Running costs cover hardware and cloud services. Every rupee goes directly to the project.

Donate via UPI →
Host a sensor node

Hardware you'll need

Everything is off-the-shelf. No soldering required. Most parts are available on Robu.in, Amazon India or local electronics stores.

ComponentPurposeEst. costNotes
Raspberry Pi 3B or 3B+ Main computer — runs all scripts ₹3,500–4,500 Pi 3B+ recommended. Check OLX Pune for refurbished (₹1,000–1,500).
USB microphone (cardioid) BirdNET acoustic AI detection ₹800–1,200 Any USB mic works. Cardioid pattern preferred. Mount outdoors, aim at trees.
SDS011 PM2.5/PM10 sensor Particulate matter air quality ₹1,400–1,800 Requires USB-to-serial adapter (CH340, ₹150). Avoid fake FTDI chips.
Pi Camera Module v2 or v3 Hill timelapse every 3 hours ₹800–1,500 Connects via CSI ribbon. Mount south-facing if possible.
MicroSD card (32GB+) OS and data storage ₹300–500 Class 10 / A1 rated. Samsung or SanDisk recommended.
Power supply (5V 3A USB-C) Power the Pi reliably ₹350–500 Official Pi PSU or quality third-party. Underpowered PSUs cause crashes.
Weatherproof enclosure Protect electronics outdoors ₹200–500 Junction box from hardware store works. Drill ventilation holes for sensors.
Small heatsinks (4-piece set) Thermal management ₹50–80 Stick onto CPU, RAM, and USB chip. Prevents throttling in summer heat.
Estimated total (new) ₹7,400–11,080

Setup steps

Flash Raspberry Pi OS Lite (64-bit) onto your microSD
Use the official Raspberry Pi Imager. Enable SSH and Wi-Fi during the flash process. Boot the Pi and connect via SSH.
ssh pi@[your-pi-ip]
Clone the Earth Pulse repository
All scripts, cron jobs, and configuration files are in the public GitHub repository.
git clone https://github.com/fangchu-in/earth-pulse
Create your .env file with your credentials
You'll need a Google Air Quality API key (free tier) and a Supabase project (free tier). The repository README has step-by-step instructions for each service.
cp .env.example .env && nano .env
Run the setup script
The setup script installs Python dependencies, configures cron jobs, tests each sensor, and verifies connectivity to the database. Takes about 10 minutes.
bash setup.sh
Register your node on the network
Once your data is flowing, contact us via the form below. We'll add your node to the Earth Pulse map and your data will appear alongside ours on the site.
Note: The detailed setup guide is being finalised as Phase 2 hardware is tested. Sign up below and we'll notify you when it's ready — including a full video walkthrough and a troubleshooting guide for common issues.
Register interest

Want to host a node?

Leave your details. We'll get in touch when the setup guide is ready, or sooner if you have questions.

Open source · Open data · Open science

All Earth Pulse code is public on GitHub. All data is downloadable free under CC BY 4.0. If you build something with this data — a research paper, a visualisation, a school project — we'd love to hear about it.

View on GitHub