ID: 2.2/08.C
Snow water equivalent is a critical variable for water resource management, and it tells us how much water is stored in the snowpack, which directly affects water availability downstream. Traditional methods to estimate SWE includes in-situ observations, remote sensing products and computational models. Since traditional physics-based models are computationally expensive, and hard to run at large scales, data-driven models offer an alternative approach for modeling those variables. Models like ConvLSTM can learn SWE dynamics directly from existing data, offering a faster and more flexible alternative. In this webinar you will learn how to build, train and apply your own data-driven model to predict snow water equivalent, step by step, from data collection to evaluation.
Learning Objectives
- Introduction of ConvLSTM models
- Data collection, preprocessing and training your model
- Post-processing: How to evaluate your model
Target Audience
- Decisions makers, modelers, scientists
Keywords
data-driven model, Snow Water Equivalent (SWE), ConvLSTM
Related Resources
- Deliverable 3.4: Data-driven modelling tools for the STARS4Water river basins
[Category: 2.2 / Level: 3]
Presenter:
Dr. Leandro Avila, Forschungszentrum Jülich (FZJ)
For further questions please contact us via this form https://stars4water.eu/contact/