Since I was a child, I've been fascinated with place — the unwieldy mixture of land and landscape, distance and terrain. When my family would go on road trips, I'd follow the route on a paper map every mile of the way. I had to have a window seat on any airplane ride, and my eyes would stay glued to the ground below. This fascination has led to a lifelong passion to better understand the world and the creatures which inhabit it "from above", using geospatial data and tools. This passion continues to run through all the work I do. I'd love to help you with your next project. Contact me today!
I've worked with a variety of private companies, non-profit and governmental organizations around the world. In every project, I strive to provide clear and actionable information and work steadfast until this goal is met.
I develop and execute workflows to collect, process and analyze varied spatial and not-so-spatial data, including satellite and aerial imagery and field and other ground-based data, both locally and in the cloud.
I put geospatial data to work to understand patterns of ecosystems, plants, animals and humans across the terrestrial landscape. In every project, I strive to apply the best predictive modeling and machine learning approaches and work with the most relevant satellite imagery and other spatial and non-spatial data.
I strive to use the best tools and languages for the problem at hand. I love learning new things but pride myself on knowing when when more mature, stable options are best.
I've worked on dozens of projects leveraging the technologies I listed above. I've selected a few with publicly available code and results which I hope you will enjoy learning more about.
As part of an international team, I lead development of a number of spatial products across 22 Pacific island countries and territories. Initial beta products include mapping coastlines over 2+ decades, mangroves, and surface water. I also contribute to the establishment and support of infrastructure and patterns and principals for developing products at scale.
I developed a deep learning model to estimate the degree of human impacts across the contiguous U.S. for agriculture, transportation, and urban factors. The results depict the intensity of each impact in a clear and consistent way and could be readily integrated with existing projects and workflows. Built on the Microsoft Planetary Computer, this project utilized Landsat 8 imagery and a training dataset based on various publicly available data sources and a UNet artificial neural-network.