I am an atmospheric scientist specializing in severe storm environments, climate variability, and data-driven prediction frameworks at Western Kentucky University. My work integrates understanding of atmospheric processes with high-performance computing and machine learning to quantify how storm-supportive environments evolve across space, time, and future climate scenarios.

My current research focuses on the Ohio River Basin, a region strongly influenced by northward moisture transport from the Gulf of Mexico into the interior United States, which enhances convective instability. Leveraging ERA5 reanalysis, NOAA Storm Events Database, and downscaled CMIP6 climate projections, I develop probabilistic frameworks to estimate the occurrence of tornadoes, large hail, and damaging winds at high spatiotemporal resolution. This work involves Random Forest classification with calibration, event–environment matching, and diagnostic analysis of thermodynamic and kinematic parameters, including CAPE, vertical wind shear, storm-relative helicity, lapse rates, and a large suite (100) of derived environmental predictors.

I conduct large-scale atmospheric data processing on national high-performance computing systems, including NCAR’s Derecho supercomputer and Casper cluster under Dr. Fan, as well as Stampede3 supercomputer at the Texas Advanced Computing Center under Dr. Manmeet Singh. Across these platforms, I generate multi-decadal climatologies of severe weather diagnostics and develop scalable, reproducible workflows for processing multi-terabyte atmospheric datasets. My computational framework integrates Python-based scientific computing (xarray, MetPy, scikit-learn), NetCDF data pipelines, and parallelized batch processing using PBS scheduling.

In addition, I apply numerical weather prediction using the WRF (Weather Research and Forecasting Model) to simulate extreme events, including mesoscale convective systems and flash floods, enabling process-level insight alongside statistical modeling. My broader research interests include severe weather, climate modeling, hydroclimatology, and climate dynamics.

I am particularly interested in advancing physically interpretable machine learning approaches in atmospheric science and contributing to research that improves severe weather prediction, risk assessment, and societal resilience.

Climatology and Future Projection of Mesoscale Severe-Storm Environments in the Ohio River Basin
Western Kentucky University | Aug 2025 - Present
Advisor: Dr. Xingang Fan | NSF Kentucky EPSCoR CLIMBS

My thesis examines environments supportive of tornadoes, large hail, and damaging winds across the Ohio River Basin using ERA5 reanalysis, NOAA Storm Events, and downscaled CMIP6 climate projections. The project develops event-environment matching and probabilistic machine-learning workflows to evaluate how severe-storm-supportive environments vary across space, time, and future climate scenarios.

Core methods and datasets

• ERA5 reanalysis and NOAA Storm Events Database
• Downscaled CMIP6 climate projections
• Random Forest classification and probability calibration
• Thermodynamic and kinematic diagnostics including CAPE, CIN, vertical wind shear, storm-relative helicity, lapse rates, and Significant Tornado Parameter

Computational workflow

• Python, Xarray, MetPy, Cartopy, Scikit-learn, NetCDF
• NCAR Derecho and Casper high-performance computing systems
• PBS batch processing for multi-decadal atmospheric datasets

Languages

Libraries & Frameworks

Atmospheric Science Tools

Platforms & Environments

🌪️  Extreme Weather Events & Severe Storm Prediction
🌡️  Climate Change, Variability & Dynamics
🌧️  Hydroclimatology & Precipitation Extremes
🌀  Tropical Meteorology & West African Climate
🖥️  Numerical Weather Prediction (WRF)
📡  Climate Modeling (ERA5 · CMIP6 · Downscaling)

• Science & Technology Head — Meteorology & Climate Science Students' Association, KNUST (Aug 2023 – Sep 2024)
• Python Instructor — Undergraduate workshops in data analysis, visualization & scientific computing, KNUST
• ForWEB Advocate — Weather forecasting early warning & disaster preparedness program, Ghana (2021 – 2024)
• Daily Weather Forecaster — Operational forecasts for Kumasi & national summary outlooks, KNUST