About Olena
Olena is a Geospatial Developer at U-Spatial and with the GEMS Informatics Initiative at the University of Minnesota. She received her M.Sc. in Geography and GIS from South Dakota State University.
Prior to joining U-Spatial, she worked as a Remote Sensing Scientist, contributing to global food security and drought monitoring research. In her current role, she builds geospatial solutions for projects spanning multiple spatial scales, technologies, and domains — from the agri-food sector to natural resources conservation.
Key Projects and Initiatives
St. Louis River Estuary Land Cover and Habitat Mapping
The St. Louis River Estuary, running along the WI/MN boundary in the Duluth-Superior metro area, is ecologically, culturally, and economically vital, including the Lake Superior National Estuarine Research Reserve. Using high-resolution NAIP aerial imagery (2021 and 2022) and a deep learning model (U-Net) for semantic segmentation, this project supports the St. Louis River Habitat Workgroup in understanding current habitat conditions and prioritizing areas for restoration, with data kept freely available, reproducible, and transparent.
Water Quality Improvement in Guatemala
Lake Atitlán, vital to Mayan communities, faces threats from climate change, population growth, and economic development. This project combines citizen science, analytical chemistry, mobile data collection with ArcGIS Survey123, and geospatial mapping to help communities in the Tzalá River and Lake Atitlán basins monitor and protect water resources.
Development of Machine Learning Models to Improve Turfgrass Evaluation
The National Turfgrass Evaluation Program has a vast repository of visual turfgrass quality ratings, but the ordinal, rater-dependent data is difficult to use consistently. This project develops machine-learning models to improve the reliability and consistency of that data.
Areas of Expertise
- Training machine learning models for land cover classification, crop evaluation, and crop-type predictions.
- Advanced processing of multi-platform satellite imagery (MODIS, Landsat, Sentinel, NAIP, LiDAR) for large-scale environmental monitoring.
- Modeling evapotranspiration (ET), vegetation dynamics, and climate anomalies for large-area water budgets and global food security assessments.
- Developing automated spatial data pipelines using Python (TensorFlow, GeoPandas, Rasterio) and deploying raster datasets on High-Performance Computing (HPC) environments.
Work with Olena
Reach out, consultations are free for the U of M community.