- ML/DL workflows including data acquisition, label creation, model training, deployment, and knowledge transfer. Post-disaster building damage detection (World Food Program), solar panel and cell detection (MWDSoCal), tree mortality (USFS FHAAST, https://doi.org/10.1371/journal.pone.0272360), land cover (Cal Water Services), and forest fuel type classification (Alchemy Analytics).- Developed script to detect when wildfires have impacted a specified number of building footprints to initiate emergency NGA satellite response tasking and alert FEMA personnel.

- Stabilization of open-source ML libraries in Raster Analytics, resulting in a 91% reduction of processing time and stable continental scale processing. Processed CONUS harmonic regressions of 5-year increments for OLI, ETM, and TM Landsat archives to create ML derived attribute maps (USFS FIA).- Sentinel 2 surface reflectance harmonization for atmospheric and topographic corrections through Sen2Cor for Microsoft Planetary Computer hosting. - Image service automation for NASA, ConocoPhillips, and Census Bureau. Automated web scraping of NOAA and Maxar for rapid post-disaster ingestion by Joint Artificial Intelligence Center/FEMA.

-Delineated a 250,000-acre soil complex across the state of Wisconsin with machine learning on phyto-edaphic and satellite imagery information with iterative landscape modeling. Reworked public data at multiple classification levels to create partition rules and led field and lab crews for multi-year sampling. (https://doi.org/10.1002/saj2.20234).-Established nutrient budget to inform harvest intensity policy for soils representing 1.6 million acres in the state of Wisconsin, requiring forecasting of input and output terms and climatic modeling, varying spatially and by scenario (https://doi.org/10.3390/f12111556).-Graduate projects: ROIs, time series analysis, modeling of MNIST, world bank and UCI machine learning data.

Assessed impacts of biochar and wood-ash amendments on soil and seedling productivity (https://doi.org/10.5849/jof.2016-060R2 ; https://academic.oup.com/jof/article/116/3/222/4967653?login=true).