• 3:30pm – Kellyn Montgomery – 3D Models of Crop Canopy Structure for Precision Agriculture
• 3:40pm – Umesh Gupta – Parsing the Earth with Deep Learning
• 3:50pm – Devon Gaydos – The Social Side of Forest Disease Modeling 
• 4:00pm – Jason Matney – Geospatial Data Management Recommendations for Land Managers

3D Models of Crop Canopy Structure for Precision Agriculture

Small unmanned aerial systems (UAS) have emerged in recent years as a versatile remote sensing tool used by agricultural producers to collect precisely-timed, fine-grained data for informing management responses to intra-field crop variability. In addition to spectral information, UAS can provide photogrammetric 3D geometry resulting in detailed canopy surface models (CSM). The complexity and scale of CSM make them distinct from terrain surface models and pose challenges with accuracy and analytical approaches. My research explores the use of canopy structure in combination with environmental and spectral data for maximizing crop productivity while minimizing natural resource degradation.

Parsing the Earth with Deep Learning

Satellite imagery is a rich and structured source of information containing more uniform data than everyday natural scene images. The computer vision and AI community is progressing in tackling challenging tasks on everyday image datasets using deep learning. In contrast to everyday image datasets, high-resolution satellite images are only recently gaining attention from the community for many important science applications including map composition, effective precision agriculture, and autonomous driving. Join me to learn how academic and industry together with government agencies are solving some of the biggest challenges in this field today.

The Social Side of Forest Disease Modeling 

Invasive pests and pathogens can cause tremendous damage to the health and functioning of forest ecosystems and the regional economies that rely on them. Geospatial models of disease spread allow stakeholders to explore different management scenarios, but these models are often underutilized by policy makers and land managers. My work combines these geospatial models with participatory methods to engage stakeholders battling a new, more aggressive strain of sudden oak death disease in Oregon. 

Geospatial Data Management Recommendations for Land Managers

Land managers increasingly leverage sophisticated decision support tools when assessing large-scale natural resource problems. Perennial topics such as accurately accounting for seasonal visitation trends and visitor behaviors are poised to be reevaluated through the use of robust analytical methods. Interdisciplinary researchers are contributing to this trend by facilitating the use of modern geospatial data management tools. However, these developments are not universal, and opportunities for strengthening support for the application of GIScience to land management are numerous. This presentation highlights recent research that seeks to address significant preexisting gaps between current and future geospatial data administration strategies for land management.