Assessing the performance of a suite of machine learning models for daily river water temperature prediction
In this study, different versions of feedforward neural network (FFNN), Gaussian process regression (GPR), and decision tree (DT) models were developed to estimate daily river water temperature using air temperature (Ta), flow discharge (Q), and the day of year (DOY) as predictors. The proposed mode...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
PeerJ Inc.
2019-06-01
|
| Series: | PeerJ |
| Subjects: | |
| Online Access: | https://peerj.com/articles/7065.pdf |
| _version_ | 1827605872589668352 |
|---|---|
| author | Senlin Zhu Emmanuel Karlo Nyarko Marijana Hadzima-Nyarko Salim Heddam Shiqiang Wu |
| author_facet | Senlin Zhu Emmanuel Karlo Nyarko Marijana Hadzima-Nyarko Salim Heddam Shiqiang Wu |
| author_sort | Senlin Zhu |
| collection | DOAJ |
| description | In this study, different versions of feedforward neural network (FFNN), Gaussian process regression (GPR), and decision tree (DT) models were developed to estimate daily river water temperature using air temperature (Ta), flow discharge (Q), and the day of year (DOY) as predictors. The proposed models were assessed using observed data from eight river stations, and modelling results were compared with the air2stream model. Model performances were evaluated using four indicators in this study: the coefficient of correlation (R), the Willmott index of agreement (d), the root mean squared error (RMSE), and the mean absolute error (MAE). Results indicated that the three machine learning models had similar performance when only Ta was used as the predictor. When the day of year was included as model input, the performances of the three machine learning models dramatically improved. Including flow discharge instead of day of year, as an additional predictor, provided a lower gain in model accuracy, thereby showing the relatively minor role of flow discharge in river water temperature prediction. However, an increase in the relative importance of flow discharge was noticed for stations with high altitude catchments (Rhône, Dischmabach and Cedar) which are influenced by cold water releases from hydropower or snow melting, suggesting the dependence of the role of flow discharge on the hydrological characteristics of such rivers. The air2stream model outperformed the three machine learning models for most of the studied rivers except for the cases where including flow discharge as a predictor provided the highest benefits. The DT model outperformed the FFNN and GPR models in the calibration phase, however in the validation phase, its performance slightly decreased. In general, the FFNN model performed slightly better than GPR model. In summary, the overall modelling results showed that the three machine learning models performed well for river water temperature modelling. |
| first_indexed | 2024-03-09T06:29:04Z |
| format | Article |
| id | doaj.art-a034a5eee9fc4386a78cc60f8a4bf2bc |
| institution | Directory Open Access Journal |
| issn | 2167-8359 |
| language | English |
| last_indexed | 2024-03-09T06:29:04Z |
| publishDate | 2019-06-01 |
| publisher | PeerJ Inc. |
| record_format | Article |
| series | PeerJ |
| spelling | doaj.art-a034a5eee9fc4386a78cc60f8a4bf2bc2023-12-03T11:10:57ZengPeerJ Inc.PeerJ2167-83592019-06-017e706510.7717/peerj.7065Assessing the performance of a suite of machine learning models for daily river water temperature predictionSenlin Zhu0Emmanuel Karlo Nyarko1Marijana Hadzima-Nyarko2Salim Heddam3Shiqiang Wu4State Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering, Nanjing Hydraulic Research Institute, Nanjing, ChinaFaculty of Electrical Engineering, Computer Science and Information Technology Osijek, University J.J. Strossmayer in Osijek, Osijek, CroatiaFaculty of Civil Engineering Osijek, University J.J. Strossmayer in Osijek, Osijek, CroatiaFaculty of Science, Agronomy Department, Hydraulics Division, Laboratory of Research in Biodiversity Interaction Ecosystem and Biotechnology, University 20 Août 1955, Skikda, AlgeriaState Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering, Nanjing Hydraulic Research Institute, Nanjing, ChinaIn this study, different versions of feedforward neural network (FFNN), Gaussian process regression (GPR), and decision tree (DT) models were developed to estimate daily river water temperature using air temperature (Ta), flow discharge (Q), and the day of year (DOY) as predictors. The proposed models were assessed using observed data from eight river stations, and modelling results were compared with the air2stream model. Model performances were evaluated using four indicators in this study: the coefficient of correlation (R), the Willmott index of agreement (d), the root mean squared error (RMSE), and the mean absolute error (MAE). Results indicated that the three machine learning models had similar performance when only Ta was used as the predictor. When the day of year was included as model input, the performances of the three machine learning models dramatically improved. Including flow discharge instead of day of year, as an additional predictor, provided a lower gain in model accuracy, thereby showing the relatively minor role of flow discharge in river water temperature prediction. However, an increase in the relative importance of flow discharge was noticed for stations with high altitude catchments (Rhône, Dischmabach and Cedar) which are influenced by cold water releases from hydropower or snow melting, suggesting the dependence of the role of flow discharge on the hydrological characteristics of such rivers. The air2stream model outperformed the three machine learning models for most of the studied rivers except for the cases where including flow discharge as a predictor provided the highest benefits. The DT model outperformed the FFNN and GPR models in the calibration phase, however in the validation phase, its performance slightly decreased. In general, the FFNN model performed slightly better than GPR model. In summary, the overall modelling results showed that the three machine learning models performed well for river water temperature modelling.https://peerj.com/articles/7065.pdfRiver water temperatureArtificial neural networkFlow dischargeDecision treeAir temperatureGaussian process regression |
| spellingShingle | Senlin Zhu Emmanuel Karlo Nyarko Marijana Hadzima-Nyarko Salim Heddam Shiqiang Wu Assessing the performance of a suite of machine learning models for daily river water temperature prediction PeerJ River water temperature Artificial neural network Flow discharge Decision tree Air temperature Gaussian process regression |
| title | Assessing the performance of a suite of machine learning models for daily river water temperature prediction |
| title_full | Assessing the performance of a suite of machine learning models for daily river water temperature prediction |
| title_fullStr | Assessing the performance of a suite of machine learning models for daily river water temperature prediction |
| title_full_unstemmed | Assessing the performance of a suite of machine learning models for daily river water temperature prediction |
| title_short | Assessing the performance of a suite of machine learning models for daily river water temperature prediction |
| title_sort | assessing the performance of a suite of machine learning models for daily river water temperature prediction |
| topic | River water temperature Artificial neural network Flow discharge Decision tree Air temperature Gaussian process regression |
| url | https://peerj.com/articles/7065.pdf |
| work_keys_str_mv | AT senlinzhu assessingtheperformanceofasuiteofmachinelearningmodelsfordailyriverwatertemperatureprediction AT emmanuelkarlonyarko assessingtheperformanceofasuiteofmachinelearningmodelsfordailyriverwatertemperatureprediction AT marijanahadzimanyarko assessingtheperformanceofasuiteofmachinelearningmodelsfordailyriverwatertemperatureprediction AT salimheddam assessingtheperformanceofasuiteofmachinelearningmodelsfordailyriverwatertemperatureprediction AT shiqiangwu assessingtheperformanceofasuiteofmachinelearningmodelsfordailyriverwatertemperatureprediction |