Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Recurso digital |
| Language: | English |
| Published: |
Zenodo
2026
|
| Subjects: | |
| Online Access: | https://doi.org/10.5281/zenodo.20389201 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Table of Contents:
- <p>This archive contains the source code used to evaluate the reliability of learned basin attribute-parameter relationships in differentiable HBV parameter learning. The workflow compares deterministic, Monte Carlo dropout, and distributional parameter-learning formulations across CAMELS-US basins, loss functions, and random seeds. It includes scripts for processing learned parameters, computing streamflow and relationship-reliability diagnostics, analyzing environmental gradients, and generating figures and tables used in the manuscript. The code is archived to support reproducibility of the submitted HESS manuscript.</p>