">
Challenges and opportunities of ML and explainable AI in large-sample hydrology

This is a Preprint and has not been peer reviewed. This is version 2 of this Preprint.

Add a Comment

You must log in to post a comment.


Comments

There are no comments or no comments have been made public for this article.

Downloads

Download Preprint

Authors

Louise J. Slater , Georgios Blougouras, Liangkun Deng, Qimin Deng, Emma Ford, Anne Hoek van Dijke, Feini Huang, Shijie Jiang, Yinxue Liu, Simon Moulds , Andrew Schepen, Jiabo Yin, Boen Zhang

Abstract

Machine learning (ML) is a powerful tool for hydrological modelling, prediction, dataset generation, model interpretation, and process discovery. As such, ML has become integral to the field of large-sample hydrology, where hundreds to thousands of river catchments are included within a single ML model to capture diverse hydrological behaviours and improve model generalisability. This manuscript outlines recent advances in ML for large-sample hydrology. We review new tools in explainable AI (XAI) and interpretability approaches, as well as challenges in these areas. Key research avenues for ML in large-sample hydrology include addressing inconsistencies in model interpretability, enhancing hydrological predictions in data-sparse and human-modified regions, addressing hydrology's "cascade of uncertainty", developing improved methods for multivariate prediction, and uncovering causality.

DOI

https://doi.org/10.31223/X5069W

Subjects

Hydrology

Keywords

hydrology, machine learning, xAI

Dates

Published: 2024-09-09 11:47

Last Updated: 2024-11-17 15:14

Older Versions
License

No Creative Commons license

Additional Metadata

Conflict of interest statement:
None

Data Availability (Reason not available):
Review article; no data.