How Clean is Pune's Air? A Data Story

Uncovering air quality patterns across India's 8th largest city using IoT sensor data

The Story

Pune — a bustling city of over 7 million people, home to IT hubs, educational institutions, and growing traffic congestion. But how does all this activity affect the air we breathe?

I analyzed 103,000+ sensor readings from 10 monitoring stations spread across Pune to answer one question:

Where and when is Pune's air most polluted, and what can we do about it?

By The Numbers

103,000+
Data Points

10
Monitoring Stations

28
Parameters Tracked

May - Aug 2019
Time Period

Key Insights

1. Pollution Hotspots Are Concentrated Near Transport Hubs

Hadapsar Gadital and Pune Railway Station consistently show pollution levels 40-60% higher than residential areas. The chart above ranks all monitoring stations by total pollution score.

2. Rush Hour = Danger Hour

Pollution spikes between 8-10 AM — exactly when most people commute. The bar chart shows how PM2.5, PM10, and NO₂ levels change throughout the day.

Time	Risk Level	Recommendation
5-7 AM	Low	Best for outdoor exercise
8-10 AM	High	Avoid outdoor activities
12-4 PM	Moderate	Use caution
10 PM+	Low	Safe for evening walks

3. Weekends Bring Relief

Pollution drops by ~15% on weekends — direct proof that reduced traffic improves air quality. This insight supports policies like odd-even traffic rules.

4. Three Distinct Pollution Zones

Using K-Means clustering, I identified three types of areas:

Zone	What It Means	Action Needed
Green	Consistently clean	Maintain current state
Caution	Occasional spikes	Enhanced monitoring
Hotspot	Chronically polluted	Urgent intervention

5. Pollutants Are Interconnected

PM2.5 and PM10 are strongly correlated (r = 0.89), indicating they share common sources — likely vehicle emissions and construction dust. This means targeting one pollutant can reduce both.

Key Takeaways

Location matters — Pollution varies dramatically across the city
Timing matters — Morning rush hours are the worst
Behavior matters — Weekend patterns prove we can improve air quality
Data can guide policy — Targeted interventions beat blanket rules

Recommendations

For City Planners

Install air purifiers at Railway Station & Bus Depots
Implement 8-10 AM traffic restrictions in hotspots
Use humidity as early warning indicator
Increase green cover along major corridors

For Residents

Exercise before 7 AM or after 8 PM
Use air purifiers if near identified hotspots
Check AQI apps before outdoor plans
Consider cycling on weekends (cleaner air!)

Methodology

Step	Description
Data Cleaning	Handled missing values, parsed timestamps, flagged sensor errors
EDA	Distribution analysis, temporal patterns, geographic mapping
Correlation	Identified relationships between pollutants & environment
Clustering	K-Means to group locations by pollution behavior
Thresholds	Compared against NAAQS safe limits

⭐ Star the repo if you found it useful!

Made with 💚 for cleaner air in Indian cities

Air Quality Analysis