Combining Synthetic and Observed Data to Enhance Machine Learning Model Performance for Streamflow Prediction

被引:3
|
作者
Lopez-Chacon, Sergio Ricardo [1 ,2 ]
Salazar, Fernando [1 ,3 ]
Blade, Ernest [3 ]
机构
[1] Int Ctr Numer Methods Engn CIMNE, Barcelona 08034, Spain
[2] Univ Politecn Cataluna, UPC BarcelonaTech, Barcelona 08034, Spain
[3] Univ Politecn Cataluna, Flumen Inst, Int Ctr Numer Methods Engn CIMNE, UPC BarcelonaTech, Barcelona 08034, Spain
关键词
machine learning; physically based; Iber; streamflow; high values; synthetic; floods; regression-enhanced random forest; CLIMATE-CHANGE; FLOOD RISK; CATCHMENT; SELECTION; EXTREMES;
D O I
10.3390/w15112020
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Machine learning (ML) models have been shown to be valuable tools employed for streamflow prediction, reporting considerable accuracy and demonstrating their potential to be part of early warning systems to mitigate flood impacts. However, one of the main drawbacks of these models is the low precision of high streamflow values and extrapolation, which are precisely the ones related to floods. Moreover, the great majority of these models are evaluated considering all the data to be equally relevant, regardless of the imbalanced nature of the streamflow records, where the proportion of high values is small but the most important. Consequently, this study tackles these issues by adding synthetic data to the observed training set of a regression-enhanced random forest model to increase the number of high streamflow values and introduce extrapolated cases. The synthetic data are generated with the physically based model Iber for synthetic precipitations of different return periods. To contrast the results, this model is compared to a model only fed with observed data. The performance evaluation is primarily focused on high streamflow values using scalar errors, graphically based errors and errors by event, taking into account precision, over- and underestimation, and cost-sensitivity analysis. The results show a considerable improvement in the performance of the model trained with the combination of observed and synthetic data with respect to the observed-data model regarding high streamflow values, where the root mean squared error and percentage bias decrease by 23.1% and 38.7%, respectively, for streamflow values larger than three years of return period. The utility of the model increases by 10.5%. The results suggest that the addition of synthetic precipitation events to existing records might lead to further improvements in the models.
引用
收藏
页数:25
相关论文
共 50 条
  • [1] Combining traditional hydrological models and machine learning for streamflow prediction
    Marcos Junior, Antonio Duarte
    Silveira, Cleiton da Silva
    da Costa, Jose Micael Ferreira
    Goncalves, Suellen Teixeira Nobre
    [J]. RBRH-REVISTA BRASILEIRA DE RECURSOS HIDRICOS, 2024, 29
  • [2] Monthly streamflow prediction and performance comparison of machine learning and deep learning methods
    Ayana, Omer
    Kanbak, Deniz Furkan
    Keles, Muemine Kaya
    Turhan, Evren
    [J]. ACTA GEOPHYSICA, 2023, 71 (06) : 2905 - 2922
  • [3] Monthly streamflow prediction and performance comparison of machine learning and deep learning methods
    Ömer Ayana
    Deniz Furkan Kanbak
    Mümine Kaya Keleş
    Evren Turhan
    [J]. Acta Geophysica, 2023, 71 : 2905 - 2922
  • [4] Machine Learning Model for Chest Radiographs: Using Local Data to Enhance Performance
    Mohn, Sarah F.
    Law, Marco
    Koleva, Maria
    Lee, Brian
    Berg, Adam
    Murray, Nicolas
    Nicolaou, Savvas
    Parker, William A.
    [J]. CANADIAN ASSOCIATION OF RADIOLOGISTS JOURNAL-JOURNAL DE L ASSOCIATION CANADIENNE DES RADIOLOGISTES, 2023, 74 (03): : 548 - 556
  • [5] High-performance prediction model combining minimum redundancy maximum relevance, circulant spectrum analysis, and machine learning methods for daily and peak streamflow
    Latifoglu, Levent
    Kaya, Esra
    [J]. THEORETICAL AND APPLIED CLIMATOLOGY, 2024, 155 (01) : 621 - 643
  • [6] High-performance prediction model combining minimum redundancy maximum relevance, circulant spectrum analysis, and machine learning methods for daily and peak streamflow
    Levent Latifoğlu
    Esra Kaya
    [J]. Theoretical and Applied Climatology, 2024, 155 : 621 - 643
  • [7] Optimizing Data Acquisition to Enhance Machine Learning Performance
    Wang, Tingting
    Huang, Shixun
    Bao, Zhifeng
    Culpepper, J. Shane
    Dedeoglu, Volkan
    Arablouei, Reza
    [J]. PROCEEDINGS OF THE VLDB ENDOWMENT, 2024, 17 (06): : 1310 - 1323
  • [8] A Hybrid Model Combining the Cama-Flood Model and Deep Learning Methods for Streamflow Prediction
    Ming Zhong
    Hongrui Zhang
    Tao Jiang
    Jun Guo
    Jinxin Zhu
    Dagang Wang
    Xiaohong Chen
    [J]. Water Resources Management, 2023, 37 : 4841 - 4859
  • [9] A Hybrid Model Combining the Cama-Flood Model and Deep Learning Methods for Streamflow Prediction
    Zhong, Ming
    Zhang, Hongrui
    Jiang, Tao
    Guo, Jun
    Zhu, Jinxin
    Wang, Dagang
    Chen, Xiaohong
    [J]. WATER RESOURCES MANAGEMENT, 2023, 37 (12) : 4841 - 4859
  • [10] Development of new machine learning model for streamflow prediction: case studies in Pakistan
    Rana Muhammad Adnan
    Reham R. Mostafa
    Ahmed Elbeltagi
    Zaher Mundher Yaseen
    Shamsuddin Shahid
    Ozgur Kisi
    [J]. Stochastic Environmental Research and Risk Assessment, 2022, 36 : 999 - 1033