Name
Integrating stream temperature simulation into a flexible hydrologic modelling framework
Date & Time
Wednesday, May 10, 2023, 3:30 PM - 3:45 PM
James Craig
Description
Stream temperature is an important hydrologic variable: it determines the viability of stream habitat for aquatic species, indicates the presence or absence of river ice, and can be used as an environmental tracer. Here, we discuss the recent integration of stream and lake temperature simulation capabilities into the Raven hydrologic modelling framework. The thermal model within Raven has a number of unique attributes. First, the in-stream model is solved as a semi-analytical solution to the governing heat transport equation posed in a Lagrangian coordinate system. Because the solution is analytical, it requires no in-reach spatial discretization and is unconditionally stable at any time step. Second, the model tracks water temperature in all soil, canopy, and snow storage compartments, and therefore provides estimates of runoff temperature entering the surface water network. Because the model is formulated in terms of enthalpy, it readily tracks phase change and ice content within soils, lakes, and streams, and properly simulates the zero curtain effect. Third, the model supports multiple user-specified model configurations, including several hydrologic models that can be emulated using Raven, including HBV-EC, the UBC watershed model, or GR4J. Lastly, the model extends and supports all of the existing functionality available through Raven, including support for gridded or interpolated meteorological inputs, temporally variable parameters, land cover and vegetation change, and simple integration with calibration and uncertainty analysis tools. The model implementation will be introduced in detail and demonstrated with application to simulation of lake and stream temperatures in Ontario and British Columbia.
Location Name
Lynx
Full Address
Banff Park Lodge Resort Hotel & Conference Centre
201 Lynx St
Banff AB T1L 1K5
Canada
Abstract
Stream temperature is an important hydrologic variable: it determines the viability of stream habitat for aquatic species, indicates the presence or absence of river ice, and can be used as an environmental tracer. Here, we discuss the recent integration of stream and lake temperature simulation capabilities into the Raven hydrologic modelling framework. The thermal model within Raven has a number of unique attributes. First, the in-stream model is solved as a semi-analytical solution to the governing heat transport equation posed in a Lagrangian coordinate system. Because the solution is analytical, it requires no in-reach spatial discretization and is unconditionally stable at any time step. Second, the model tracks water temperature in all soil, canopy, and snow storage compartments, and therefore provides estimates of runoff temperature entering the surface water network. Because the model is formulated in terms of enthalpy, it readily tracks phase change and ice content within soils, lakes, and streams, and properly simulates the zero curtain effect. Third, the model supports multiple user-specified model configurations, including several hydrologic models that can be emulated using Raven, including HBV-EC, the UBC watershed model, or GR4J. Lastly, the model extends and supports all of the existing functionality available through Raven, including support for gridded or interpolated meteorological inputs, temporally variable parameters, land cover and vegetation change, and simple integration with calibration and uncertainty analysis tools. The model implementation will be introduced in detail and demonstrated with application to simulation of lake and stream temperatures in Ontario and British Columbia.
Session Type
Breakout Session