-
- H32G-08: A Modular High-Performance Computing Framework for Forecast Skill Assessment of Cyanobacterial Harmful Algal Bloom Prediction Systems
-
NOLA CC
Author(s):Generic 'disconnected' Message
Noah Beckage, University of Vermont (First Author, Presenting Author)
Patrick Clemins, University of Vermont
Panagiotis Oikonomou, University of Vermont
Rakhshinda Bano, University of Vermont
Scott Turnbull, University of Vermont
Asim Zia, University of Vermont
Forecasts are an important tool for helping communities prepare for environmental risks, such as harmful algal blooms that threaten freshwater lakes. However, many forecasting systems for these events are complex and slow to run, making it difficult to explore how forecast accuracy depends on changing conditions or new sources of data. To address this, we developed a new harmful algal bloom forecasting workflow that uses high-performance computing to simulate thousands of different forecasting scenarios in parallel. We applied this workflow to predict harmful algal blooms in the Lake Champlain basin, which spans parts of the U.S. and Canada. Our system can test different combinations of weather and water quality data, as well as different scientific models of how nutrients flow into the lake, to understand how these factors influence bloom forecasts. This flexible approach allows researchers to explore “what-if” questions about future conditions, data sources, or model assumptions. By making it easier to test and improve forecasting systems, our workflow supports better understanding and decision-making around water quality and public health risks in freshwater systems.
Scientific DisciplineSuggested ItinerariesNeighborhoodType
Enter Note
Go to previous page in this tab
Session
