An Artificial Neural Network Emulator of the Rangeland Hydrology and Erosion Model

This is a Preprint and has not been peer reviewed. The published version of this Preprint is available: This is version 1 of this Preprint.

Add a Comment

You must log in to post a comment.


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


Download Preprint


Mahmoud Saeedimoghaddam, Grey Nearing , Mariano Hernandez, Mark Nearing, David C. Goodrich, Loretta J. Metz


Machine learning (ML) is becoming an ever more important tool in hydrologic modeling. Many studies have shown the higher prediction accuracy of the ML models over traditional process-based ones. However, there is another advantage of ML which is its lower computer time of execution. This is important for the applications such as hydraulic soil erosion estimation over a large area and at a finer spatial scale. Using traditional models like Rangeland Hydrology and Erosion Model (RHEM) requires too much computation time and resources. In this study, we designed an Artificial Neural Network that is able to recreate the RHEM outputs (runoff, soil loss, and sediment yield) with high accuracy (Nash-Sutcliffe Efficiency $\approx$ 1.0) and a very low computational time (13 billion times faster on average). We ran the RHEM for more than a million synthetic scenarios and train the Emulator with them. We also, fine-tuned the trained Emulator with the RHEM runs of the real-world scenarios (more than 32,000) so the Emulator remains comprehensive while it works specifically accurately for the real-world cases. We also showed that the sensitivity of the Emulator to the input variables is similar to the RHEM and it can effectively capture the changes in the RHEM outputs when an input variable varies. Finally, the dynamic prediction behavior of the Emulator is statistically similar to the RHEM with a 95\% confidence interval.



Artificial Intelligence and Robotics, Dynamic Systems, Hydrology, Natural Resources and Conservation



Published: 2022-06-17 19:22

Last Updated: 2022-06-17 23:22


CC BY Attribution 4.0 International

Additional Metadata

Data Availability (Reason not available):
The NRI and synthetic RHEM scenarios and outputs as well as the RHEM Em- 610 ulator, our training pipeline, all of the Python scripts we used for cross validation and 611 sensitivity analysis, and our model’s pre-trained weights are open source and publicly 612 available via (DOI: 10.5281/zenodo.6599824)