An artificial neural network emulator of the rangeland hydrology and erosion model
Machine learning (ML) is becoming an ever more important tool in hydrologic modeling. Previous studies have shown the higher prediction accuracy of those ML models over traditional process-based ones. However, there is another advantage of ML which is its lower computational demand. This is importan...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
KeAi Communications Co., Ltd.
2024-06-01
|
| Series: | International Soil and Water Conservation Research |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2095633923000965 |
| _version_ | 1797226723008315392 |
|---|---|
| author | Mahmoud Saeedimoghaddam Grey Nearing Mariano Hernandez Mark A. Nearing David C. Goodrich Loretta J. Metz |
| author_facet | Mahmoud Saeedimoghaddam Grey Nearing Mariano Hernandez Mark A. Nearing David C. Goodrich Loretta J. Metz |
| author_sort | Mahmoud Saeedimoghaddam |
| collection | DOAJ |
| description | Machine learning (ML) is becoming an ever more important tool in hydrologic modeling. Previous studies have shown the higher prediction accuracy of those ML models over traditional process-based ones. However, there is another advantage of ML which is its lower computational demand. 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 (annual average runoff, soil loss, and sediment yield and not the daily storm event-based values) with high accuracy (Nash-Sutcliffe Efficiency ≈ 1.0) and a very low computational time (13 billion times faster on average using a GPU). 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. |
| first_indexed | 2024-04-24T14:29:26Z |
| format | Article |
| id | doaj.art-d883aa5d06a1404580ce5c563ee1486b |
| institution | Directory Open Access Journal |
| issn | 2095-6339 |
| language | English |
| last_indexed | 2024-04-24T14:29:26Z |
| publishDate | 2024-06-01 |
| publisher | KeAi Communications Co., Ltd. |
| record_format | Article |
| series | International Soil and Water Conservation Research |
| spelling | doaj.art-d883aa5d06a1404580ce5c563ee1486b2024-04-03T04:26:29ZengKeAi Communications Co., Ltd.International Soil and Water Conservation Research2095-63392024-06-01122241257An artificial neural network emulator of the rangeland hydrology and erosion modelMahmoud Saeedimoghaddam0Grey Nearing1Mariano Hernandez2Mark A. Nearing3David C. Goodrich4Loretta J. Metz5Department of Land, Air & Water Resources, University of California, Davis, CA, USA; Corresponding author.Google Research, Mountain View, CA, USAUSDA-Agricultural Research Service, Southwest Watershed Research Center, Tucson, AZ, USAUSDA-Agricultural Research Service, Southwest Watershed Research Center, Tucson, AZ, USAUSDA-Agricultural Research Service, Southwest Watershed Research Center, Tucson, AZ, USAUSDA-NRCS Resource Inventory and Assessment Divisionxo, CEAP-Grazing Lands, Tucson, AZ, USAMachine learning (ML) is becoming an ever more important tool in hydrologic modeling. Previous studies have shown the higher prediction accuracy of those ML models over traditional process-based ones. However, there is another advantage of ML which is its lower computational demand. 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 (annual average runoff, soil loss, and sediment yield and not the daily storm event-based values) with high accuracy (Nash-Sutcliffe Efficiency ≈ 1.0) and a very low computational time (13 billion times faster on average using a GPU). 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.http://www.sciencedirect.com/science/article/pii/S2095633923000965RHEMSediment yieldSoil lossRunoffDeep learning |
| spellingShingle | Mahmoud Saeedimoghaddam Grey Nearing Mariano Hernandez Mark A. Nearing David C. Goodrich Loretta J. Metz An artificial neural network emulator of the rangeland hydrology and erosion model International Soil and Water Conservation Research RHEM Sediment yield Soil loss Runoff Deep learning |
| title | An artificial neural network emulator of the rangeland hydrology and erosion model |
| title_full | An artificial neural network emulator of the rangeland hydrology and erosion model |
| title_fullStr | An artificial neural network emulator of the rangeland hydrology and erosion model |
| title_full_unstemmed | An artificial neural network emulator of the rangeland hydrology and erosion model |
| title_short | An artificial neural network emulator of the rangeland hydrology and erosion model |
| title_sort | artificial neural network emulator of the rangeland hydrology and erosion model |
| topic | RHEM Sediment yield Soil loss Runoff Deep learning |
| url | http://www.sciencedirect.com/science/article/pii/S2095633923000965 |
| work_keys_str_mv | AT mahmoudsaeedimoghaddam anartificialneuralnetworkemulatoroftherangelandhydrologyanderosionmodel AT greynearing anartificialneuralnetworkemulatoroftherangelandhydrologyanderosionmodel AT marianohernandez anartificialneuralnetworkemulatoroftherangelandhydrologyanderosionmodel AT markanearing anartificialneuralnetworkemulatoroftherangelandhydrologyanderosionmodel AT davidcgoodrich anartificialneuralnetworkemulatoroftherangelandhydrologyanderosionmodel AT lorettajmetz anartificialneuralnetworkemulatoroftherangelandhydrologyanderosionmodel AT mahmoudsaeedimoghaddam artificialneuralnetworkemulatoroftherangelandhydrologyanderosionmodel AT greynearing artificialneuralnetworkemulatoroftherangelandhydrologyanderosionmodel AT marianohernandez artificialneuralnetworkemulatoroftherangelandhydrologyanderosionmodel AT markanearing artificialneuralnetworkemulatoroftherangelandhydrologyanderosionmodel AT davidcgoodrich artificialneuralnetworkemulatoroftherangelandhydrologyanderosionmodel AT lorettajmetz artificialneuralnetworkemulatoroftherangelandhydrologyanderosionmodel |