AQI trends - full timeline

PM2.5-based AQI for major Indian cities (mean across all stations)

Show:

Timeline:

District-level air quality map

Select pollutant and statistic to view the latest district-wise data across India

Pollutant:

Statistic:

Loading district-level air quality map...

All Districts Rankings (by latest value)

Rank	District	State	Latest AQI

All monitoring stations across India

Interactive map showing locations of 560+ CPCB monitoring stations. Click on markers for station details.

Compare cities - AQI trends (full timeline)

Select up to 12 cities to compare PM2.5-based AQI trends (mean across all stations)

Add Cities to Compare

0 / 12 cities selected

Select cities from the dropdown above to start comparing

Show:

Timeline:

Compare districts - AQI trends (full timeline)

Select up to 12 districts to compare PM2.5-based AQI trends (mean across all stations)

Add Districts to Compare

0 / 12 districts selected

Select districts from the dropdown above to start comparing

Show:

Timeline:

District Rankings (by latest AQI)

Rank	District	State	Latest AQI	Category

Compare stations within state - AQI trends (full timeline)

Select a state to compare all monitoring stations within that state

Select State

No state selected

View by:

Timeline:

Station Rankings (by latest AQI)

Rank	Name	Latest AQI	Category

Compare stations within city - AQI trends (full timeline)

Select a city to compare all monitoring stations within that city

Select City

No city selected

Timeline:

Station Rankings (by latest AQI)

Rank	Name	Latest AQI	Category

How to use this data

Access air quality data programmatically or browse the dataset

Python

Read gzipped CSV files directly from GitHub URLs using pandas:

Python

import pandas as pd # Read directly from GitHub URL url = "https://saketkc.github.io/vayuayan-archive/data/delhi__anand_vihar__anand_vihar__site_244.csv.gz" df = pd.read_csv(url, compression='gzip') # Display data print(df.head()) print(f"Total records: {len(df)}")

R

Use readr to load compressed data directly:

library(readr) # Read directly from GitHub URL url <- "https://saketkc.github.io/vayuayan-archive/data/delhi__anand_vihar__anand_vihar__site_244.csv.gz" df <- read_csv(url) # Display data head(df)

cURL

Download raw compressed files:

Bash

# Download and decompress curl -L https://saketkc.github.io/vayuayan-archive/data/delhi__anand_vihar__anand_vihar__site_244.csv.gz | gunzip > data.csv

Browse Data

Search and filter station data:

Understanding AQI Calculation

Learn how the Air Quality Index is calculated

What is Air Quality Index (AQI)?

Air Quality Index (AQI) is a number used to convey the quality of air by the government to the general public. Air quality deteriorates with an increase in the concentration of pollutants. The Air Quality Index represents the severity of pollution for ordinary people.

Indian (CPCB) AQI

According to the Indian Government (CPCB), Indian AQI range is from 0-500, from 0 being good and 500 being severe. There are eight major pollutants to be taken into account for AQI calculation: particulate matter (PM 10 and PM 2.5), carbon monoxide (CO), ozone (O₃), nitrogen dioxide (NO₂), sulfur dioxide (SO₂), ammonia (NH₃), and lead (Pb).

AQI Categories and Health Impacts

AQI Range	Category	Health Impact
0-50	Good	Minimal or no impact on health
51-100	Satisfactory	Breathing difficulty in sensitive groups
101-200	Moderate	Breathing discomfort for children and elderly people
201-300	Poor	Health effects on long-term exposure
301-400	Very Poor	Respiratory illness on long exposure
401-500	Severe	Severe health impacts on all

How to Calculate AQI

The formula to calculate AQI is the same as per the Indian CPCB and US-EPA. The AQI is calculated using the equations separately for each parameter. For example, if you wish to calculate AQI on the basis of four parameters, use the equation four times, and the worst sub-index communicates the AQI.

Formula:

                                Ip = [(IHi - ILo) / (BPHi - BPLo)] × (Cp - BPLo) + ILo
                            
Ip = Index of pollutant p
Cp = Truncated concentration of pollutant p
BPHi = Concentration breakpoint ≥ Cp
BPLo = Concentration breakpoint ≤ Cp
IHi = AQI value corresponding to BPHi
ILo = AQI value corresponding to BPLo

Example Calculation

If the current concentration of PM2.5 is 110 µg/m³, then referring to AQI range as per Indian standards:

BPHi = 120 (upper breakpoint for 201-300 range)
BPLo = 91 (lower breakpoint for 201-300 range)
IHi = 300 (AQI upper value)
ILo = 201 (AQI lower value)

Putting the values in equation:

                            Sub Index = [(300 - 201) / (120 - 91)] × (110 - 91) + 201

                            Sub Index = [99 / 29] × 19 + 201

                            Sub Index = 3.41 × 19 + 201

                            Sub Index = 265.86

Similarly, calculate the sub-index for other parameters, and the worst sub-index represents the overall AQI.

Note: To calculate AQI, a minimum of three parameters should be taken, of which one must be either PM10 or PM2.5. To calculate sub-indices, 16 hours of data is needed.

PM2.5 Breakpoints Reference

This website calculates AQI primarily from PM2.5 concentrations using the following CPCB breakpoints:

Source: Central Pollution Control Board (CPCB)

Station Data Completeness

Ranking of stations based on historical data availability

Scoring Methodology

Each station is scored based on three factors: Consistency (50%) - proportion of days with data within the station's active period; Coverage (30%) - total days of data relative to global date range; Span (20%) - how long the station has been collecting data.

0 stations

Rank	Station	City	State	Score	Avg AQI	AQI Rank	Days	First Date	Last Date
Loading...

All Districts Rankings (by latest value)

District Rankings (by latest AQI)

Station Rankings (by latest AQI)

Station Rankings (by latest AQI)

Python

R

cURL

Browse Data

What is Air Quality Index (AQI)?

Indian (CPCB) AQI

AQI Categories and Health Impacts

How to Calculate AQI

Example Calculation

PM2.5 Breakpoints Reference

Scoring Methodology

Loading...