Welcome to DUSTY
Overview
DUSTY is a modular atmospheric analytics engine developed by Muhammad Aqeel Abdulla for tracking and characterizing dust storm events using satellite-based reanalysis datasets. The system is engineered to automate the detection, analysis, and lifecycle tracking of dust storms across both time and space using high-resolution NASA MERRA-2 data.
It supports both classical and fusion-based storm detection, correlating satellite observations of dust mass and aerosol optical depth with large airport locations to assess potential operational impact. DUSTY is optimized for environmental data scientists, atmospheric researchers, and aerospace engineers working on climate-aware aviation systems.
Capabilities
Downloads and converts MERRA-2 NetCDF files for both DUSMASS and AODANA
Identifies and tracks dust storm “blobs” using dynamic spatial thresholds
Associates detected storm centers with the nearest large airport (ICAO coded)
Computes monthly statistical baselines and identifies anomalies
Supports both standalone and fusion-based storm detection modes
Outputs CSV summaries of storm evolution and statistical deviations
Use Cases
DUSTY was developed as part of a broader research effort into the interaction between dust storms and aviation operations, particularly the ingestion risk posed to commercial aircraft engines. It has been applied to:
Quantifying dust activity around airports
Analyzing storm frequency and severity trends over months
Supporting data fusion pipelines for multi-source dust monitoring
Ground-truthing atmospheric model outputs with reanalysis data
Project Layout
The project is organized into modular components:
Installation: Set up DUSTY and its dependencies
Usage: Learn how to run and configure DUSTY
MERRA2AODProcessor: Convert and process DUSMASS files
MonthlyDustAnalyzer Module: Statistical monthly summaries
StormDetector Module: Classical storm detection
FusionStormDetector Module: Fusion-based storm detection