Skip to content Where Legends Are Made
Cooperative Institute for Research to Operations in Hydrology

Using TEEHR Dashboards to Explore Hydrologic Model Performance

Using TEEHR Dashboards to Explore Hydrologic Model Performance

Day 3 Session 1 (10:30AM)

Presenters:

Matt Denno (RTI)
Sam Lamont (RTI)
Katie van Werkhoven (RTI)
Sam Landsteiner (RTI)

This hands-on workshop introduces participants to TEEHR (Tools for Exploratory Evaluation in Hydrologic Research) as an evaluation framework aligned with the emerging Community Hydrologic Prediction Testbed (CHPT).

CHPT defines standardized experimental design and evaluation protocols for comparing hydrologic models and forecasts, while TEEHR provides the computational, visualization, and data management tools needed to implement these protocols at scale.

The workshop is organized into two complementary parts.

Part 1:
Participants are introduced to TEEHR-Cloud and a set of application-specific dashboards. Instructors will walk through dashboard structure and the types of analyses they support, demonstrating how users can explore model and forecast performance, compare results against benchmark or alternative models, and investigate evaluation outcomes.

Part 2:
Participants will work directly with TEEHR Python tools in a JupyterHub environment. Attendees will download data from the TEEHR data warehouse, load example (synthetic) model simulation data, and execute evaluation calculations and analyses in a Jupyter notebook. This portion illustrates an example evaluation protocol and shows how TEEHR supports transparent, reproducible workflows that connect computational analysis to dashboard-based interpretation.

The workshop leverages existing TEEHR Cloud infrastructure and emphasizes hands-on interaction with both dashboards and notebook-based workflows. It highlights how evaluation choices relate to different application goals, such as flood forecasting versus water supply forecasting.

Workshop Slides

Learning Outcomes:

  • Use application-specific TEEHR dashboards to compare model outputs against benchmarks
  • Interpret evaluation results in the context of application needs (e.g., flood peaks, timing, volumes)
  • Explain the relationship between CHPT evaluation protocols and TEEHR’s evaluation capabilities
  • Load data from the TEEHR data warehouse for comparison with their own research outcomes
  • Generate standardized evaluation metrics using the TEEHR Python package
  • Explore evaluation results to answer targeted research or operational questions

Prerequisites:

  • Prior knowledge:
    • Familiarity with hydrologic modeling or forecasting concepts
    • Basic understanding of model or forecast evaluation is helpful but not required (key concepts will be reviewed)
  • Software & Data
    • Modern web browser (Chrome or Firefox recommended)
    • No local software installation required
    • All computation and dashboards will be accessed via existing TEEHR Cloud infrastructure (and/or 2i2c)
    • Accounts Required
      • GitHub
      • CIROH-2i2c JupyterHub (if applicable)

Additional Notes:

Prior TEEHR workshops have focused on the python package. Recent development is focused on building a platform to streamline intercomparison of results across researchers via the TEEHR evaluation dashboards. This workshop will make others aware of these tools to streamline their research efforts.

The greater buy-in we get from the CIROH community in uploading model outcomes, the easier it will be to compare alternative approaches and the more useful the TEEHR dashboards will become.