mosdac

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

AI-powered Help Bot for ISRO's MOSDAC satellite data portal. Knowledge Graph (Neo4j) + RAG + Groq LLM with React chat interface.

MOSDAC AI is an intelligent Q&A system for weather data from [mosdac.gov.in](https://mosdac.gov.in) (ISRO's Meteorological and Oceanographic Satellite Data Archival Centre).

MOSDAC AI Assistant

Explainable AI system for satellite-based heavy rainfall prediction using real INSAT-3DR data, temporal 3D-CNNs, Grad-CAM, and SHAP.

Hybrid RAG + Knowledge Graph API for intelligent access to ISRO's MOSDAC satellite data. Combines FAISS vector search, Neo4j knowledge graph, and Google Gemini synthesis behind a FastAPI service with Firebase auth.

This repository contains codes supplementing the results presented in "On the observed low-level wind variability at Jaisalmer and its representation in ERA5 reanalysis".

AI-powered satellite data intelligence interface — chat with ISRO's MOSDAC datasets using hybrid RAG + knowledge graph, explore satellites visually, and search documentation semantically. Built with React 19, TypeScript, Tailwind CSS v4, and Firebase.

This project addresses the challenge of accessing and querying vast amounts of meteorological and oceanographic satellite data from MOSDAC (Meteorological & Oceanographic Satellite Data Archival Centre). Users often struggle to navigate complex scientific data portals, understand satellite data products, and find relevant information quickly.

Hybrid RAG + Knowledge Graph API for intelligent access to ISRO's MOSDAC satellite data. Combines FAISS vector search, Neo4j knowledge graph, and Google Gemini synthesis behind a FastAPI service with Firebase auth.

Minor Project on the topic of "AI Weather Prediction Model for Extreme Weather Events." Version 4.2.2