Authors: Ryan Johnson – University of Utah
Title: Collaborations with the APRFC and field campaigns to develop data-driven streamflow and stream temperature models for Alaska
Presentation Type: Lightning Talk
Abstract: Collaborations with the APRFC and field campaigns to develop data-driven streamflow and stream temperature models for Alaska
Abstract: Communication and collaboration are essential in advancing hydrologic models in data-sparse regions. As we begin our CIROH Expanding Streamflow prediction in SE Alaska, we want to share this collaborative effort with the Alaska Pacific River Forecast Center (APRFC) and corresponding field campaign, focused on predicting streamflow and stream temperature in south-central and south Alaska, with the CIROH community. In close partnership with APRFC operations and forecasters, we align research objectives with operational forecasting needs, emphasizing the development of data-driven modeling tools to address operational blind spots and challenges with using the Sac-SMA/Snow-17 modeling platform for high latitude streamflow prediction. Our approach centers on applying Long Short-Term Memory (LSTM) machine learning models to capture the influences of glacial, snowmelt, and atmospheric river processes on regional hydrology. Key takeaway items from our collaborations have been the discovery of additional data sources to develop and refine models, field data collection, local stakeholders, and key points of contact to help refine our models.