📊🛰️ Data processing scripts, ML models, and Explainable AI results created as part of my Masters Thesis @ Johns Hopkins

Weather and Disaster Station

A Python Package for Computing Effective Precipitation Using Google Earth Engine Climate Data.

Evaluation of extreme hydrometeorological phenomena such as droughts and low water periods, using the Standardized Precipitation Index (SPI), the Standardized Precipitation-Evapotranspiration Index (SPEI) and the Palmer Drought Severity Index (PDSI), in reference to water scarcity, water stress and water availability.

In this repository, two Root Zone Soil Moisture (RZSM) estimation methods are evaluated

Machine Learning based Drought Prediction

Application of the ARIMA model to forecast rainfall patterns. Leveraging time-series analysis techniques, it predicts future rainfall levels by analyzing historical data specifically from Bahawalnagar District, Punjab, Pakistan.

Spatiotemporal Analysis of Agricultural Drought Severity and Hotspots in Somaliland. It integrates MODIS-derived vegetation indices and CHIRPS precipitation data to identify and assess drought severity and hotspots over time.

Lesson materials for Module 2 (M2), "Open Climate Science for Agriculture"

Application of the ETS model to forecast rainfall patterns. Leveraging time-series analysis techniques, it predicts future rainfall levels by analyzing historical data specifically from Bahwalnagar District, Punjab, Pakistan.

A geospatial data pipeline for agricultural drought monitoring in Africa. This project calculates standardized indices (SPI, SPEI, SSI) using ERA5-Land and CHIRPS data to model crop stress, seasonal correlations, and parametric insurance triggers for Maize and Teff systems.

This project supports economic monitoring in Morocco using alternative data. It currently includes drought monitoring and conflict event analysis using satellite and geospatial data.

ML-powered early warning system for predicting high food insecurity risk across Kenya ASAL counties. Combines IPC food security outcomes, rainfall, NDVI vegetation, and food price indicators to support early risk monitoring and humanitarian planning. #DataScience #FoodSecurity #Kenya

A study of the stress response of vegetation to drought situation through multispectral satellite imagery. Case of study of Como lake, summer 2022.

React operator dashboard and public Talk app for ArdaLink, combining live drought monitoring, ground-truth capture, and browser-based voice conversations.

Biophysical intelligence engine for drought resilience: satellite and environmental data ingest, NDVI scoring, spatial assessments, and livestock journey planning for Isiolo County, Kenya.

a platform designed to provide streamlined, user-friendly, and validated Combined Drought Indicator (CDI) products.

Application of the ARIMA model to forecast PET patterns. Leveraging time-series analysis techniques, it predicts future rainfall levels by analyzing historical data specifically from Bahawalnagar District, Punjab, Pakistan.

Drought monitoring in New Zealand