Scott Kelly Email & Phone Number

- Developed an automated tailings pond monitoring tool using Google Cloud Platform delivering weekly area estimates for 6 tailing ponds, increasing annual data frequency by 26x and enabling automated email alerting.- Developed a segmentation model in TensorFlow for geological outcrop cavities from high-resolution drone imagery enhancing geological interpretation through integration with an interactive 3D texture mesh model.- Implemented an invasive vegetation model to 32,000 reclamation sites using time-series Sentinel-2 and Superdove imagery.

Edmonton, Alberta, Canada

- Applied time-series segmentation using estimated biophysical parameters to quantify disturbance patterns and vegetation recovery on oil & gas sites at the provincial scale.- Led iron-bearing rock detection in Peace River Alberta, applying spectral techniques using WorldView-2 imagery to identify high-grade iron locations.- Developed a serverless pipeline using Google Cloud Platform to ingest commercial satellite imagery to Cloud Buckets streamlining the image ingestion process.- Presented a technical demonstration of an Earth Engine application for 80 attendees on the application of InSAR and LiDAR time-series monitoring for visualizing vertical displacement for abandoned coal mines.

Calgary, Alberta, Canada

- Implemented a Random Forest classifier using MODIS time-series data to improve perennial crop classification and crop yield accuracy across the Canadian prairies.- Reduced manual crop progress report retrieval time by 80% by developing a Python-based web scraping script.- Quantified the impact of urban sprawl and land use changes on farmland from 1985 to 2021 by creating change detection maps using historical land use, crop classification, and census data.