Urban Air Quality Dataset for Jaipur (2019–2025): PM2.5, PM10, and Meteorological Variables from Three CPCB Monitoring Stations

This dataset accompanies the manuscript, "A Small-Data Machine Learning Framework for Forecasting Urban Air Pollution Spikes in Jaipur," submitted to Applied Artificial Intelligence (Taylor & Francis).

The dataset contains hourly and 8-hour averaged PM2.5 and PM10 measurements collected from three Central Pollution Control Board (CPCB) air quality monitoring stations in Jaipur, Rajasthan, India: Police Lines, Shastri Nagar, and Adarsh Nagar. The data span January 2019 through March 2025.

Meteorological variables, including temperature, relative humidity, wind speed, and precipitation, were obtained from the NASA POWER Project at hourly resolution and were used as auxiliary features for air pollution forecasting.

Dataset Files

• Police_2019_2024.csv - PM2.5 and PM10 measurements from the Police Lines station (2019-2024)

• Police_2025_clean.csv - PM2.5 and PM10 measurements from the Police Lines station (January-March 2025)

• Shastri_2019_2024.csv - PM2.5 and PM10 measurements from the Shastri Nagar station (2019-2024)

• Shastri_2025_clean.csv - PM2.5 and PM10 measurements from the Shastri Nagar station (January-March 2025)

• Adarsh_Nagar_2019_2024_combined.csv - PM2.5 and PM10 measurements from the Adarsh Nagar station (2019-2024)

• Adarsh_2025_clean.csv - PM2.5 and PM10 measurements from the Adarsh Nagar station (January-March 2025)

• POWER_POINT_HOURLY.csv - Hourly meteorological data obtained from the NASA POWER Project for Jaipur

Analysis Code

• Urban_Air_Pollution_Spikes_in_Jaipur.ipynb - Jupyter Notebook containing the data processing, analysis, and machine learning workflow used in the study.

The dataset is provided to support transparency, reproducibility, and future research on urban air quality forecasting using machine learning techniques.