Post-processing operational forecasts with LightGBM for improved flow prediction and water quality forecasts
Day 1 Session 2 (1:30 PM)
Presenters:
John T. Kemper, Utah State University
Mirce Morales-Velazquez, University of Vermont
In this workshop we will demonstrate the capabilities of and various use cases for post-processing National Water Model forecast outputs with the Light Gradient Boosting Machine (LightGBM) implementation of gradient-boosted decision trees. Within the Lake Champlain basin, we plan to demonstrate two different use cases for post-processed National Water Model forecasts. The first will be using field observations to improve operational flow forecasts; the second will be taking those improved flow forecasts and transforming them to forecasts of concentration for various water quality constituents (total phosphorus, nitrate, chloride).
Learning Outcomes:
- Gain understanding of how to work with NWM operational forecasts
- Learn how to perform variable selection to construct LightGBM models from operational data
- Learn how to structure data inputs to train and tune LightGBM models
- Learn how to robustly evaluate forecast performance
- After completion, participants will ideally have the tools to begin to apply the approaches demonstrated to their areas and use cases of interest.
Prerequisites:
Knowledge:
- Knowledge of R and Python
- Attendance at the last year’s workshop “Decision Tree Models for Post-Processing National Water Model Streamflow Outputs” could also be helpful.
Hardware/Software:
Accounts: