DATA SCIENCE
EARTH OBSERVATION
GEOSPATIAL ML

Aycha Tammour.

Data Scientist working at the intersection of machine learning, Earth observation, and geospatial data.

↓   EXPLORE MY WORK

I build machine learning pipelines for satellite and sensor data: air quality, land surface temperature, and environmental monitoring. Formerly an astronomer; now pointed at Earth instead of quasars.

Toolkit

Geospatial & EO

Rasterio GeoPandas Xarray STAC QGIS GEE

Machine Learning

scikit-learn XGBoost TensorFlow/PyTorch Time-Series Forecasting

MLOps & Cloud

AWS Airflow GitHub Actions Docker

Programming & Data

Python SQL PostGIS DuckDB

Selected projects

Wildfire Impact Analysis Using Satellite Imagery

Burn severity analysis of a 2025 wildfire near Latakia, Syria, using Sentinel-2 imagery and the dNBR index to classify fire impact. Burn extent validated against Copernicus Emergency Management Service data, achieving 89% accuracy and ~86% IoU.

Sentinel-2 dNBR Case Study
View writeup →

ECOSTRESS LST Downscaling

Sharpens ECOSTRESS land surface temperature from 70m to 10m resolution using Sentinel-2 and elevation data, validated across seven cities from Toronto to Cairo.

Random Forest Sentinel-2 Rasterio
View writeup →

Geospatial Environment Monitor

An MLOps pipeline for environmental monitoring built on Sentinel-2 spectral indices, with experiment tracking and automated deployment.

XGBoost MLflow GitHub Actions
View writeup →

Background

Education

Western University

PhD, Astronomy  ·  London, ON, Canada  ·  2016

Insights from Unsupervised Clustering and Composite Spectral Analysis into the Physical Properties Driving Emission and Absorption in Quasar UV/Optical Spectra

Publications

Astro research

A summary of the research conducted during my PhD using machine learning to analyze quasar optical/UV spectra. View research →