Using Data-Driven Prediction of Downstream 1D River Flow to Overcome the Challenges of Hydrologic River Modeling
Methods for downstream river flow prediction can be categorized into physics-based and empirical approaches. Although based on well-studied physical relationships, physics-based models rely on numerous hydrologic variables characteristic of the specific river system that can be costly to acquire. Mo...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-11-01
|
| Series: | Water |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4441/15/21/3843 |
| _version_ | 1797631194282590208 |
|---|---|
| author | Jeremy Feinstein Quentin Ploussard Thomas Veselka Eugene Yan |
| author_facet | Jeremy Feinstein Quentin Ploussard Thomas Veselka Eugene Yan |
| author_sort | Jeremy Feinstein |
| collection | DOAJ |
| description | Methods for downstream river flow prediction can be categorized into physics-based and empirical approaches. Although based on well-studied physical relationships, physics-based models rely on numerous hydrologic variables characteristic of the specific river system that can be costly to acquire. Moreover, simulation is often computationally intensive. Conversely, empirical models require less information about the system being modeled and can capture a system’s interactions based on a smaller set of observed data. This article introduces two empirical methods to predict downstream hydraulic variables based on observed stream data: a linear programming (LP) model, and a convolutional neural network (CNN). We apply both empirical models within the Colorado River system to a site located on the Green River, downstream of the Yampa River confluence and Flaming Gorge Dam, and compare it to the physics-based model Streamflow Synthesis and Reservoir Regulation (SSARR) currently used by federal agencies. Results show that both proposed models significantly outperform the SSARR model. Moreover, the CNN model outperforms the LP model for hourly predictions whereas both perform similarly for daily predictions. Although less accurate than the CNN model at finer temporal resolution, the LP model is ideal for linear water scheduling tools. |
| first_indexed | 2024-03-11T11:18:30Z |
| format | Article |
| id | doaj.art-711cb9a16fb24a959979f83d0153d1e1 |
| institution | Directory Open Access Journal |
| issn | 2073-4441 |
| language | English |
| last_indexed | 2024-03-11T11:18:30Z |
| publishDate | 2023-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Water |
| spelling | doaj.art-711cb9a16fb24a959979f83d0153d1e12023-11-10T15:15:31ZengMDPI AGWater2073-44412023-11-011521384310.3390/w15213843Using Data-Driven Prediction of Downstream 1D River Flow to Overcome the Challenges of Hydrologic River ModelingJeremy Feinstein0Quentin Ploussard1Thomas Veselka2Eugene Yan3Argonne National Laboratory, Environmental Science Division, 9700 S. Cass Ave., Lemont, IL 60439, USAArgonne National Laboratory, Energy Systems and Infrastructure Analysis Division, 9700 S. Cass Ave., Lemont, IL 60439, USAArgonne National Laboratory, Energy Systems and Infrastructure Analysis Division, 9700 S. Cass Ave., Lemont, IL 60439, USAArgonne National Laboratory, Environmental Science Division, 9700 S. Cass Ave., Lemont, IL 60439, USAMethods for downstream river flow prediction can be categorized into physics-based and empirical approaches. Although based on well-studied physical relationships, physics-based models rely on numerous hydrologic variables characteristic of the specific river system that can be costly to acquire. Moreover, simulation is often computationally intensive. Conversely, empirical models require less information about the system being modeled and can capture a system’s interactions based on a smaller set of observed data. This article introduces two empirical methods to predict downstream hydraulic variables based on observed stream data: a linear programming (LP) model, and a convolutional neural network (CNN). We apply both empirical models within the Colorado River system to a site located on the Green River, downstream of the Yampa River confluence and Flaming Gorge Dam, and compare it to the physics-based model Streamflow Synthesis and Reservoir Regulation (SSARR) currently used by federal agencies. Results show that both proposed models significantly outperform the SSARR model. Moreover, the CNN model outperforms the LP model for hourly predictions whereas both perform similarly for daily predictions. Although less accurate than the CNN model at finer temporal resolution, the LP model is ideal for linear water scheduling tools.https://www.mdpi.com/2073-4441/15/21/3843linear programmingunit hydrographdeep learningconvolutional neural networkconvolution methods |
| spellingShingle | Jeremy Feinstein Quentin Ploussard Thomas Veselka Eugene Yan Using Data-Driven Prediction of Downstream 1D River Flow to Overcome the Challenges of Hydrologic River Modeling Water linear programming unit hydrograph deep learning convolutional neural network convolution methods |
| title | Using Data-Driven Prediction of Downstream 1D River Flow to Overcome the Challenges of Hydrologic River Modeling |
| title_full | Using Data-Driven Prediction of Downstream 1D River Flow to Overcome the Challenges of Hydrologic River Modeling |
| title_fullStr | Using Data-Driven Prediction of Downstream 1D River Flow to Overcome the Challenges of Hydrologic River Modeling |
| title_full_unstemmed | Using Data-Driven Prediction of Downstream 1D River Flow to Overcome the Challenges of Hydrologic River Modeling |
| title_short | Using Data-Driven Prediction of Downstream 1D River Flow to Overcome the Challenges of Hydrologic River Modeling |
| title_sort | using data driven prediction of downstream 1d river flow to overcome the challenges of hydrologic river modeling |
| topic | linear programming unit hydrograph deep learning convolutional neural network convolution methods |
| url | https://www.mdpi.com/2073-4441/15/21/3843 |
| work_keys_str_mv | AT jeremyfeinstein usingdatadrivenpredictionofdownstream1driverflowtoovercomethechallengesofhydrologicrivermodeling AT quentinploussard usingdatadrivenpredictionofdownstream1driverflowtoovercomethechallengesofhydrologicrivermodeling AT thomasveselka usingdatadrivenpredictionofdownstream1driverflowtoovercomethechallengesofhydrologicrivermodeling AT eugeneyan usingdatadrivenpredictionofdownstream1driverflowtoovercomethechallengesofhydrologicrivermodeling |