Geospatial Analytics Ph.D. students will share current research in a series of lightning talks:

> Jenna Abrahamson & Owen Smith: Modeling and Understanding using Temporal Analysis of Transient Earth Data (MUTATED) (advised by Josh Gray)

Abstract: Monitoring land cover change is crucial for timely detection of deforestation and other anthropogenic activities relevant to climate change and intelligence. Bayesian statistical methods are powerful for change detection because they can detect change without relying on long, stable periods of data to train on. However, applying such algorithms in near-real time across the globe while maintaining high levels of accuracy and computational efficiency is challenging. Our work highlights the application of roboBayes, a novel change detection algorithm that enables the broad-area search of multi-source satellite imagery to detect, monitor, and characterize the progression of heavy construction events across Earth.

> Margaret Lawrimore: Creating Spatially Complete Zoning Maps Using Machine Learning (co-advised by Georgina Sanchez and Ross Meentemeyer)

Abstract: Zoning regulates land use and intensity of urban development at the county and municipal level in the United States, promoting economic growth, community health, and environmental preservation. However, the limited availability of zoning data at scale hinders regional assessments of regulations and coordinated resilience planning efforts. Therefore, we developed an open-source, replicable, and transferable framework to predict spatially complete zoning. We applied the Random Forest algorithm statewide in North Carolina, USA. We found ~98% accuracy filling gaps within county zoning, but representative training data is required.

> Andrew Shannon: Fire & Drought: Projecting the Impacts of Future Forest Disturbances in the Southwestern United States (advised by Robert Scheller)

Abstract: Dry forests in the southwestern United States have experienced increasing disturbances from wildlfire and drought in recent decades, and it is uncertain how these disturbances will impact forest regeneration and mortality patterns under hotter and potentially drier climate futures. In this research, we examine the interactions between drought-driven and wildfire-driven tree mortality across public lands in Northern Arizona using a dynamic, spatially explicit simulation of forest succession and disturbance. This work highlights the role of future climate and vegetation structure in altering the disturbance regimes of these forests in the absence of changes to present day management patterns.

Recording: https://youtu.be/HAhohpVSDNE