cpcb

AI-powered air quality prediction system using satellite data and machine learning

Monitoring Air Pollution from Space using Satellite Observations, Ground-Based Measurements, Reanalysis Data, and AI/ML Techniques

Satellite-derived Surface AQI estimation & TROPOMI HCHO hotspot detection/attribution over India using INSAT-3D, Sentinel-5P, ERA5 & ML | ISRO BAH-2026

The following repository contains simple sets of code to clean, restructure and assess data downloaded from Central Pollution Control Board Data

VasAQ: An air quality analytics platform using NMF for source identification and Random Forest Regression for 15-day AQI forecasting. Powered by CPCB (data.gov.in) and Visual Crossing data.

Dhuan (धुआँ): India's smog season, live, in 3D. Punjab's stubble fires, the wind that carries the smoke, and the air Delhi breathes, from NASA FIRMS, NOAA GFS and CPCB. Built on God's Eye View.

BioWaste AI is an AI-powered web application that classifies biomedical waste and provides CPCB-compliant disposal guidance using natural language input

Predicting Water Quality Index (WQI) and categorical classification using Deep Learning Neural Networks on CPCB India environmental monitoring data (2019-2022).

A statistical weather & air-quality instrument for any city on Earth — search or geolocate, CPCB AQI, no backend, no build step, no API keys

Bangalore air-quality hotspot analysis — clustering daily AQI from 14 CPCB/KSPCB stations (2024) with K-Means, hierarchical & DBSCAN

Cloud-native geospatial-ML platform for India: fuses satellite (Sentinel-5P HCHO, MAIAC AOD), reanalysis & ground-station data to estimate surface air quality (CPCB NAQI) and detect HCHO hotspots with fire-transport attribution. BAH2026 PS-3.

Station-level air quality analysis across India from raw CPCB data: daily AQI for 579 stations, quality checks and hotspot detection.

India's air quality on one map — official CPCB AQI breakpoints, weighted state roll-up so bad cities can't hide behind good ones, centroid gap-filling for missing stations.

Air quality decision-support and inspection prioritisation prototype for the Mumbai Metropolitan Region

FirstLight — the 6 AM on-call agent for school air: deterministic CPCB (2014) + CAQM GRAP rules, stubble-plume modelling, consent-gated dispatch, and a hash-chained ledger that proves every call. AWS: Lambda, API Gateway, DynamoDB, Cognito, scoped IAM.

Screening-level BOD-DO compliance modelling for the Wazirabad-Okhla reach of the Yamuna in Delhi: 1D Streeter-Phelps, 2000-run Latin Hypercube Monte Carlo, CPCB Class C compliance probabilities. Built from public CPCB station data 2018-2022.

End-to-end air quality analytics project using CPCB data, Python, PostgreSQL, Power BI, and Excel.

Airpy is a tool to preprocess CPCB data, validated using vizualizations

Hyperlocal Real-Time AQI & 72-Hour Atmospheric Smog Inversion Engine with Copernicus CAMS & NASA Satellite Telemetry (Delhi NCR Focus)

Personal air-exposure agent for India. Every AQI app tells you the air is bad — Saans tells you which hour to go out. Telegram bot, Hindi + English, zero dependencies.