Air Quality Analysis Of Delhi
This dataset falls under the category Environmental Data Air Quality Data.
It contains the following data: Daily air quality data across cities
This dataset was scouted on 2022-02-05 as part of a data sourcing project conducted by TUMI. License information might be outdated: Check original source for current licensing. The data can be accessed using the following URL / API Endpoint: https://www.kaggle.com/yashvi/air-quality-analysis-of-delhi/data

Context

Delhi is the capital of India and Second most populous city of the country. This is city situated in northern part of country and home of nearly 2 million people. Delhi is AQI is consider as hazardous category, this city suffering from poor AQI. Now poor air quality impact people health and daily activities. Almost 25000 people died because of poor air quality, people suffering from asthma, lung cancer and many disease. Government of India putting great efforts to tackle this problem, Let's show analysis skill that help agencies to understand the data better ways.

Data Content Hourly data for last one year (1-sep-2021 to 1-sep-2022). dataset contains various parameters of AQI such PM 2.5, PM 10, CO, CO2 etc. This data collected on IGI Delhi center.

Acknowledgement I would like to thank Central Control Room for Air Quality Management - All India for providing data & They are actual owner. Please provide reference in case of any external usage.

Image credit Unsplash by amir hosseini

Inspiration 1. Understand the trend, seasons other factor in given time series. 2. Which parameter of AQI is more dangerous 3. can we forecast the AQI parameter in advance so that agencies can prepare accordingly.

Let's do analysis and find the insights.

This dataset contains hourly PM2.5 measurements from monitoring stations located within 25 km of Delhi, India, collected via the OpenAQ API.

The data covers the period 2010–2025, filtered to PM2.5 only, and includes pollutant values, timestamps, and sensor coordinates. It can be used for air quality analysis, health impact studies, forecasting models, and urban pollution research.

Columns - value → PM2.5 concentration (µg/m³) - parameter.name → pollutant measured (always pm25) - parameter.units → measurement units - period.datetimeFrom.utc → UTC timestamp of measurement

Data Source Data retrieved from the OpenAQ open air quality platform.

The datasets contains date- and state-wise historically compiled data on air quality (by pollution level) in rural and urban areas of India from the year 2015 , as measured by Central Pollution Board (CPCB) through its daily (24 hourly measurements, taken at 4 PM everyday) Air Quality Index (AQI) reports.

The CPCB measures air quality by continuous online monitoring of various pollutants such as Particulate Matter10 (PM10), Particulate Matter2.5 (PM2.5), Sulphur Dioxide (SO2), Nitrogen Oxide or Oxides of Nitrogen (NO2), Ozone (O3), Carbon Monoxide (CO), Ammonic (NH3) and Lead (Pb) and calculating their level of pollution in the ambient air. Based on the each pollutant load in the air and their associated health impacts, the CPCB calculates the overall Air Pollution in Air Quality Index (AQI) value and publishes the data. This AQI data is then used by CPCB to report the air quality status i.e good, satisfactory, moderate, poor, very poor and severe, etc. of a particular location and their related health impacts because of air pollution.

In 2023, *** days in India's capital city Delhi had air quality levels ranging from poor to very poor. ** days witnessed air quality in severe to severe plus category. 2020 was the best year during the recorded period with the highest number of days ranging from good to moderate category, due to nationwide lockdown because of the Covid pandemic.

Dataset

This dataset was created by Hyperion72

Released under Apache 2.0

Contents

According to a survey on air pollution in the national capital Delhi in October 2024, over ** percent of respondents experienced a sore throat or cough, and another ** percent had breathing difficulty or asthma. Nine percent of respondents had burning eyes as a result of air pollution.

In a survey conducted in 2022, ** percent of the survey respondents from Delhi in India believed that detailed plans to drive stubble burning in Punjab to near zero levels is the most effective initiative that could help reduce air pollution in Delhi. A reduction in the number of vehicles by ** percent during the months of October and December was another solution that around ** percent of the respondents considered effective in combating the pollution levels.

This dataset contains air quality data from the national capital of Delhi, India. It includes information on air pollution levels, including particulate matter (PM2.5 and PM10) levels, nitrogen dioxide (NO2), sulfur dioxide (SO2), carbon dioxide (CO2), ozone (O3), and other pollutants. The data was collected from monitoring stations located in various areas of Delhi between November 25, 2020, and January 24, 2023. This dataset is a valuable resource for researchers and policymakers to better understand air quality in Delhi and its impacts on public health.

This dataset consists of all contributions made by Social AQI (SAQI) project. The description of dataset is as below -

Local Sensor Data (hyperlocal-air-quality-sensor-data) - contains all sensors values recorded through local neighbourhood sensors throught the length of the project
Locations for all these sensors are as below - In Najafgarh, Delhi, India : Jharoda Kalan, Nangli Dairy and DTC Bus terminal.
In Okhla : Sanjay Colony, Tekhand, Shaheen Bagh.

Data from Central Pollution Control Board (central-air-quality-sensor-data) - Najafgarh_CPCB.csv, Okhla_CPCB.csv : Contains data provided by CPCB from Najafgarh,Delhi and Oklha, Delhi

PollutionODP.owl : Ontology Design Pattern for pollution - http://ontologydesignpatterns.org/wiki/Submissions:Pollution.

Ontology : SAQI ontology as triples (ttl), xml (rdf) and json-ld (json) serialization format
Ontology documentation : ontology/diagram contains figures describing ontology, ontology/documentation/saqi.html contains LODE documentation for the ontology

ethnographic-survey-data - anonymized survey responses for initial pollution perception and literacy survey as well as SAQI app feedback survey.

SHACL-shapes - for validating against SAQI ontology.

sparql-queries - sample queries to run on our ontology.

setup-rdf-store-script - script to setup rdf store with given data using rml mapper.

📋 Dataset Overview

A comprehensive hourly dataset tracking atmospheric conditions and air quality across 29 major Indian cities (28 states + Delhi) from 2022 to 2025. This dataset provides detailed insights into India's environmental patterns, pollution trends, and meteorological conditions.

🗺️ Geographic Coverage

• Cities Covered: 29 major cities representing all 28 Indian states + Delhi
• Time Period: 2022 to 2025
• Records: 842,160 hourly observations
• Features: 63 parameters covering weather, pollution, and temporal data

📊 Dataset Specifications

• Total Records: 842,160
• Features: 63 columns
• Temporal Resolution: Hourly data
• Period: 2022-2025 (4 years)
• Memory Usage: ~405 MB

🏙️ Indian Cities Included

(Representing all 28 states + Union Territory of Delhi) - Delhi - Mumbai (Maharashtra) - Chennai (Tamil Nadu) - Kolkata (West Bengal) - Bengaluru (Karnataka) - Hyderabad (Telangana) - Ahmedabad (Gujarat) - Pune (Maharashtra) - Jaipur (Rajasthan) - Lucknow (Uttar Pradesh) - And other major state capitals...

🗂️ Data Columns

📍 Geographic & Temporal Context

🌡️ Meteorological Parameters

Temperature & Humidity - Temp_2m_C - Ambient temperature (°C) - Humidity_Percent - Relative humidity - Dew_Point_C - Dew point temperature - Humidity_Category - Comfort levels

Wind Patterns - Wind_Speed_10m_kmh - Surface wind speed - Wind_Dir_10m - Wind direction (critical for pollution dispersion) - Wind_Gusts_kmh - Wind gusts - Wind_Stagnation - Air stagnation events

Precipitation & Pressure - Precipitation_mm, Rain_mm - Monsoon rainfall tracking - Is_Raining, Heavy_Rain - Rain events - Pressure_MSL_hPa - Monsoon pressure systems

☀️ Solar & Cloud Data

• Solar_Radiation_Wm2 - Total solar radiation
• Direct/Diffuse_Radiation_Wm2 - Radiation components
• Cloud_Cover_Percent - Total cloud cover
• Cloud_Low/Mid/High_Percent - Cloud altitude distribution
• Sunshine_Seconds - Bright sunshine duration
• Is_Daytime - Day/night indicator

🏭 Air Quality Metrics (Critical for India)

Particulate Matter - PM2_5_ugm3 - Fine particulate matter (primary concern) - PM10_ugm3 - Coarse particulate matter - PM_Ratio - PM2.5/PM10 ratio (source identification) - Dust_ugm3 - Dust concentrations - AOD - Aerosol Optical Depth

Gaseous Pollutants - CO_ugm3 - Carbon monoxide (vehicular/industrial) - NO2_ugm3 - Nitrogen dioxide (traffic, industries) - SO2_ugm3 - Sulfur dioxide (industrial, power plants) - O3_ugm3 - Ozone (secondary pollutant)

📊 Air Quality Indices

US AQI System - US_AQI - Overall US AQI - US_AQI_PM25, US_AQI_PM10 - PM-specific indices - US_AQI_NO2, US_AQI_O3, US_AQI_CO - Gas-specific indices

EU AQI System - EU_AQI - European Air Quality Index - EU_AQI_PM25, EU_AQI_PM10 - European standards

India-Specific Categories - AQI_Category - Overall air quality category - PM25_Category_India - India-specific PM2.5 categorization

🌪️ Special Features

• Temp_Inversion - Temperature inversion events (critical for winter pollution in North India)

⚠️ Data Quality Notes

• High Completeness: Most columns have 842,160 non-null values
• Minor Missing Values:
  • US/EU AQI columns: ~145 missing records
  • AQI_Category: 2,516 missing
  • US_AQI_NO2: 2 missing
  • US_AQI_O3: 73 missing

🎯 India-Specific Analysis Opportunities

Seasonal Patterns

• Winter (Dec-Feb): Severe PM2.5 pollution in Indo-Gangetic plain
• Summer (Mar-May): Dust storms, high O3 levels
• Monsoon (Jun-Sep): Natural cleansing, improved AQI
• Post-Monsoon (Oct-Nov): Crop burning impacts in North India

Geographic Hotspots

• Northern Plains: Delhi, Lucknow - Winter smog, crop burning
• Coastal Cities: Mumbai, Chennai - Marine influence, humidity
• Southern Plateau: Bengaluru, Hyderabad - Moderate pollution
• Eastern India: Kolkata - Industrial and vehicular pollution

Event-Based Analysis

• Festival Impacts: Diwali fireworks, Holi bonfires
• Agricultural Cycles: Stubble burning seasons
• Meteorological Events: Western disturbances, monsoon progress

🛠️ Suggested Research Topics

Public Health

• C...

The concentration of PM2.5 across the Indian capital city of Delhi was about *** micrograms per cubic meter in 2019. Air pollution has been a major problem across the south Asian country, causing almost *********** deaths every year. Industrial processes, emissions from vehicles, demolition activities and fossil fuel combustion release a significant concentration of particulate matter into the atmosphere.

Ambient air pollution alone accounts for more than 3 million premature deaths worldwide. Low and Middle Income Countries (LMIC) account for most (~87%) of this disease burden. Air pollution in the megacities of these countries has risen to the levels of public health hazards forcing the cities to take emergency measures, such as issuing red alerts and vehicle-rationing interventions (VRI). Using in-situ and high-resolution satellite data, this research examines the efficacy of VRI in Delhi and Beijing, two of the most polluted cities of LMIC. This research shows that VRI reduced particulate matter (PM) ≤ 2.5 μm in aerodynamic diameter (PM2.5) in Beijing during the 2008 Summer Olympics. However, such interventions implemented in 2015 and 2016 in Beijing and in 2016 in Delhi were ineffective in improving air quality. Moreover, the effects of such interventions were short lived, for example 54% of the cleaning in Beijing disappeared within 2 weeks after the Olympics, and Delhi witnessed a 34% increase in PM2.5 during the 2 weeks after the interventions. Both cities observed excess cardiopulmonary mortality even during the interventions. Short- and long-term preventive and mitigation strategies are needed to manage the air pollution disease burden.

1) Data Introduction • The Industrial-Residential Air Quality Classification Dataset is based on daily air pollution data measured in six cities, including Moscow, Delhi, Beijing, Zurich, Vancouver, and Stockholm, in 2024 and is designed to analyze air quality differences and the impact of urban environments by dividing each city into industrial or residential areas.

2) Data Utilization (1) Industrial-Residential Air Quality Classification Dataset has characteristics that: • This dataset consists of daily measurements for each city, including city name, date of measurement, concentration of six types of air pollutants (standard unit: ppm CO₂, µg/m ³) and city type (industrial/residential) labels for all samples. • It is designed to compare the air quality characteristics of industrial districts (three cities) and residential districts (three cities), and the unit of measurement and label criteria are clearly standardized. (2) Industrial-Residential Air Quality Classification Dataset can be used to: • Air Quality Based Urban Type Classification Model Development: Using pollutant concentrations as input variables, it can be used for machine learning classifiers to predict whether the city is an industrial or residential district. • Major pollutant correlation analysis: It can be used for variable importance analysis and correlation studies on how important certain pollutants, such as PM2.5, NO₂, are to distinguish industrial and residential districts. • Comparison of industrial/residential air quality trends: It can be used as a basis for urban planning and environmental policy establishment by comparing and analyzing air quality change patterns throughout the year by city type. • Environmental and Policy Studies: By analyzing pollutant concentrations and their association with city type, city-specific environmental regulations, and urban planning policies, it can be applied to urban environment improvement and policy effectiveness assessment.

New Delhi was the most polluted city in India in 2024, based on an average air quality index (AQI) of ***. The seven most polluted cities in India in 2024 all had AQI levels above ***. An AQI between 151 and 200 is classified as unhealthy. Air pollution in India India was the third most polluted country in the world in 2023, behind only Bangladesh and Pakistan. The South Asian country recorded an average annual fine particulate matter (PM2.5) concentration of ** micrograms per cubic meter of air (µg/m3) that year, more than ** times above the World Health Organization’s recommended limit. Health effects of air pollution Exposure to air pollution can lead to a range of health issues, such as strokes, respiratory conditions, and cardiovascular disease. Air pollution is attributable to millions of premature deaths every year around the world, with India one of the most affected countries.

Delhi faces some of the world’s highest concentrations of PM2.5, the most damaging form of air pollution. Although awareness of outdoor air pollution is rising across the world, there is limited information on indoor air pollution (IAP) levels, particularly in heavily polluted cities like Delhi. Even less evidence exists on how IAP varies by socio-economic status (SES), and whether or not addressing information gaps can change defensive investments against IAP. In this paper, we deploy Indoor Air Quality Monitors (IAQMs) in thousands of Delhi households across varying socio-economic strata in order to document IAP levels during the peak wintertime air pollution period. Across high and low SES households, we document indoor PM2.5 levels that are: (1) extraordinarily high — more than 20 times World Health Organization (WHO) standards; (2) only 10 percent lower in high (versus low) SES households; and (3) significantly higher than levels reported by the nearest, outdoor government monitors, the main source of public information on air pollution in this setting. We then report on a field experiment that randomly assigned IAQMs, as well as an opportunity to rent an air purifier at a subsidized price, across medium and high SES homes during the 2019-20 winter season.

Novel Coronavirus disease (COVID-19), after being identified in late December 2019 in Wuhan city of China, spread very fast and has affected all the countries in the world. The impact of lockdowns on particulate matter during the lockdown period needs attention to explore the correlation between anthropogenic and natural emissions. The current study has demonstrated the changes in fine particulate matter PM2.5, PM10 and their effect on air quality during the lockdown. The air quality before the lockdown was low in New Delhi (India) and Riyadh (Saudi Arabia), among major cities worldwide. The air quality of India is influenced by dust and sand from the desert and surrounding areas. Thus, the current study becomes important to analyse changes in the air quality of the Indian sub-continent as impacted by dust storms from long distances. The result indicated a significant reduction of PM2.5 and PM10 from 93.24 to 37.89 μg/m3 and from 176.55 to 98.87 μg/m3 during the lockdown period as compared to pre lockdown period, respectively. The study shows that average concentrations of PM10 and PM2.5 have declined by -44% and -59% during the lockdown period in Delhi. The average value of median PM10 was calculated at 33.71 μg/m3 for Riyadh, which was lower than that value for New Delhi during the same period. The values of PM10 were different for pre and during the lockdown periods in Riyadh, indicating the considerable influence on air quality, especially the concentration of PM10, from both the natural (sand and dust storms) and the anthropogenic sources during the lockdown periods. However, relatively smaller gains in the improvement of air quality in Riyadh were correlated to the imposition of milder lockdown and the predominance of natural factors over the anthropogenic factors there. The Air Quality Index (AQI) data for Delhi showed the air quality to be ‘satisfactory’ and in the green category during the lockdown period. This study attempts to better understand the impact of particulate matter on the short- and long-term air quality in Delhi during the lockdown. This study has the scope of being scaled up nationwide, and this might be helpful in formulation air pollution reduction and sustainable management policies in the future.

Air Quality Forecast: Contaminant Concentration: PM2.5: India: New Delhi data was reported at 137.209 mcg/Cub m in 22 May 2025. This records a decrease from the previous number of 140.762 mcg/Cub m for 21 May 2025. Air Quality Forecast: Contaminant Concentration: PM2.5: India: New Delhi data is updated daily, averaging 67.523 mcg/Cub m from Oct 2019 (Median) to 22 May 2025, with 2038 observations. The data reached an all-time high of 423.393 mcg/Cub m in 15 Jan 2021 and a record low of 14.475 mcg/Cub m in 10 Sep 2023. Air Quality Forecast: Contaminant Concentration: PM2.5: India: New Delhi data remains active status in CEIC and is reported by CEIC Data. The data is categorized under Global Database’s India – Table CAMS.AQF: Air Quality Forecast: Contaminant Concentration: PM2.5: by Cities. [COVID-19-IMPACT]

Particle counts in New Delhi are much higher than in urban background sites of other cities and towns.

Dataset in "Enhanced Aerosol Particle Growth Sustained by High Continental Chlorine Emission in India" by Gunthe, Liu, et al., Nature Geoscience