NOAA Environmental Data Talks - Speaker Series
Learn more about the many applications of environmental data from NOAA and our partners, including how data can be transformed into visualizations to tell a story. Tune in every Tuesday at 2:00 PM EDT in September to hear experts discuss the importance of data in their daily work.
To see a presentation, join the Q&A session via Adobe Connect, click here and follow the prompts to "enter as a guest."
Talks scheduled for NOAA DataFest 2019
Visualization Research and Facilities at the University of Maryland, Baltimore County (UMBC)
September 3 at 2:00 pm EDT
A discussion of UMBC’s visualization research and related facilities, including spherical displays, VR/AR head mounted devices, 3D scanning, an immersive VR wall, and glasses-free 3D display. The presentation includes work taking place in UMBC’s Imaging Research Center (IRC) and the IRC-affiliated Assistive Visualization and Artificial Intelligence Lab (AVAIL).
Dr. Don Engel is a professor and the Assistant Vice President for Research at the University of Maryland, Baltimore County (UMBC). As AVPR, Don leads UMBC’s Office of Research Development. Don’s research is on the applications of visualization and artificial intelligence (esp. image processing and computational linguistics) to data-driven discovery (esp. in the physical and life sciences). After completing a Ph.D. in physics and master’s in computer science, Don spent several years working for Congress and executive branch agencies as a science and technology policy advisor. Don’s background also includes clinical experience as a resident in radiation oncology medical physics.
September 10 at 2:00 pm EDT
Every day, thousands of forecast maps are produced by NOAA and related agencies. All forecasts come with error but how can we best visualize uncertainty on a forecast map? Max Schneider, PhD student in Statistics and NCEP intern, says the effectiveness of different approaches towards visualizing this uncertainty can be carefully studied in user experiments. In this talk, he presents a human subjects experiment where three common visualization techniques go head-to-head, to see which one enables effective map-reading and judgments using the forecasts.
Max Schneider is a PhD student in Statistics at the University of Washington, Seattle. This summer, he is interning at the Environmental Modeling Center within NOAA's National Centers for Environmental Prediction in College Park, MD. In his dissertation work, Max builds spatiotemporal models to forecast earthquakes in the Pacific Northwest. He focuses on quantifying various sources of error in these models and how to visualize them to diverse audiences. He collaborates with cognitive psychologists to directly study the effect visualizations have on how people use forecasts. In his work at NOAA, Max quantifies the uncertainty within coupled numerical models of hurricane impacts, for the US COASTAL Act. His motivation is to improve operational forecasting with a statistical approach to uncertainty quantification and visualization.
How Can Artificial Intelligence Contribute to NOAA’s Mission?
September 17 at 2:00 pm EDT
Artificial intelligence has made a lot of headlines lately, but it is not clear to everyone how these tools can be applied to science. In this NEDTalk, we demonstrate how supervised deep-learning may be viewed as a new and powerful approach for developing software routines that were previously beyond our reach. With these tools, we can automate, accelerate, and improve upon existing applications and develop a host of new capabilities. After surveying many of the ways it may be applied to the Earth system, we’ll take a deeper dive into how AI can be used to detect and track tropical storms.
David Hall joined NVIDIA in January 2018, after working as an Assistant Professor of Research in Computer Science at CU Boulder. Dr. Hall has technical expertise in theoretical physics, numerical methods, computational fluid dynamics, and artificial intelligence. David spent the previous decade developing non-hydrostatic atmospheric models for high resolution climate modeling in HPC environments. As a solution architect at NVIDIA, Dr. Hall’s primary role is to help scientist and engineers understand and translate the latest breakthroughs in artificial intelligence into practical solutions in the areas of weather, climate, and space. Dr. Hall earned his PhD in Physics from the University of Santa Barbara, CA and a BA in physics from CU Boulder.
Computer Vision for Conservation
September 24 at 2:00 pm EDT
Motivated by recent developments in image recognition, we hosted a data science challenge on the crowdsourcing platform Kaggle to automate the identification of endangered North Atlantic right whales. The winning solution automatically identified individual whales with 87% accuracy with a series of convolutional neural networks to identify the region of interest on an image, rotate, crop, and create standardized photographs of uniform size and orientation and then identify the correct individual whale from these passport-like photographs. Recent advances in deep learning coupled with this fully automated workflow have yielded impressive results and have the potential to revolutionize traditional methods for the collection of data on the abundance and distribution of wild populations.
Christin Khan is a Fishery Biologist in the Protected Species Branch at the Northeast Fisheries Science Center in Woods Hole. She is an aerial survey observer and data manager of the North Atlantic Right Whale Sighting Survey which conducts aerial surveys to monitor right whale abundance and distribution from New Jersey to Canada. When not in the air, Christin also works on right whale social behavior, automated image recognition, right whale outreach signs, the Right Whale Sighting Advisory System, interactive Google map, and the Whale Alert app.
To see a presentation, join the Q&A session via Adobe Connect, click here and follow the prompts to "enter as a guest." If you have any trouble logging onto Adobe Connect, try switching browsers.