Forecasting High-Flow Discharges in a Flashy Catchment Using Multiple Precipitation Estimates as Predictors in Machine Learning Models
The use of machine learning (ML) for predicting high river flow events is gaining prominence and among its non-trivial design decisions is the definition of the quantitative precipitation estimate (QPE) product included in the input dataset. This study proposes and evaluates the use of multiple conc...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-11-01
|
| Series: | Hydrology |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2306-5338/9/12/216 |
| _version_ | 1827638467882909696 |
|---|---|
| author | Andre D. L. Zanchetta Paulin Coulibaly Vincent Fortin |
| author_facet | Andre D. L. Zanchetta Paulin Coulibaly Vincent Fortin |
| author_sort | Andre D. L. Zanchetta |
| collection | DOAJ |
| description | The use of machine learning (ML) for predicting high river flow events is gaining prominence and among its non-trivial design decisions is the definition of the quantitative precipitation estimate (QPE) product included in the input dataset. This study proposes and evaluates the use of multiple concurrent QPEs to improve the performance of a ML model towards the forecasting of the discharge in a flashy urban catchment. Multiple extreme learning machine (ELM) models were trained with distinct combinations of QPEs from radar, reanalysis, and gauge datasets. Their performance was then assessed in terms of goodness of fit and contingency analysis for the prediction of high flows. It was found that multi-QPEs models overperformed the best of its single-QPE counterparts, with gains in Kling-Gupta efficiency (KGE) values up to 4.76% and increase of precision in detecting high flows up to 7.27% for the lead times in which forecasts were considered “useful”. The novelty of these results suggests that the implementation of ML models could achieve better performance if the predictive features related to rainfall data were more diverse in terms of data sources when compared with the currently predominant use of a single QPE product. |
| first_indexed | 2024-03-09T16:20:42Z |
| format | Article |
| id | doaj.art-7149383f157d42f892005309f0e4f834 |
| institution | Directory Open Access Journal |
| issn | 2306-5338 |
| language | English |
| last_indexed | 2024-03-09T16:20:42Z |
| publishDate | 2022-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Hydrology |
| spelling | doaj.art-7149383f157d42f892005309f0e4f8342023-11-24T15:17:47ZengMDPI AGHydrology2306-53382022-11-0191221610.3390/hydrology9120216Forecasting High-Flow Discharges in a Flashy Catchment Using Multiple Precipitation Estimates as Predictors in Machine Learning ModelsAndre D. L. Zanchetta0Paulin Coulibaly1Vincent Fortin2Department of Civil Engineering, McMaster University, 1280 Main Street West, Hamilton, ON L8S 4L7, CanadaDepartment of Civil Engineering, McMaster University, 1280 Main Street West, Hamilton, ON L8S 4L7, CanadaMeteorological Research Division, Environment and Climate Change Canada, 2121 Route Transcanadienne, Dorval, QC H9P 1J3, CanadaThe use of machine learning (ML) for predicting high river flow events is gaining prominence and among its non-trivial design decisions is the definition of the quantitative precipitation estimate (QPE) product included in the input dataset. This study proposes and evaluates the use of multiple concurrent QPEs to improve the performance of a ML model towards the forecasting of the discharge in a flashy urban catchment. Multiple extreme learning machine (ELM) models were trained with distinct combinations of QPEs from radar, reanalysis, and gauge datasets. Their performance was then assessed in terms of goodness of fit and contingency analysis for the prediction of high flows. It was found that multi-QPEs models overperformed the best of its single-QPE counterparts, with gains in Kling-Gupta efficiency (KGE) values up to 4.76% and increase of precision in detecting high flows up to 7.27% for the lead times in which forecasts were considered “useful”. The novelty of these results suggests that the implementation of ML models could achieve better performance if the predictive features related to rainfall data were more diverse in terms of data sources when compared with the currently predominant use of a single QPE product.https://www.mdpi.com/2306-5338/9/12/216flash floodsmachine learningrainfall-runoffflood forecastingurban catchment |
| spellingShingle | Andre D. L. Zanchetta Paulin Coulibaly Vincent Fortin Forecasting High-Flow Discharges in a Flashy Catchment Using Multiple Precipitation Estimates as Predictors in Machine Learning Models Hydrology flash floods machine learning rainfall-runoff flood forecasting urban catchment |
| title | Forecasting High-Flow Discharges in a Flashy Catchment Using Multiple Precipitation Estimates as Predictors in Machine Learning Models |
| title_full | Forecasting High-Flow Discharges in a Flashy Catchment Using Multiple Precipitation Estimates as Predictors in Machine Learning Models |
| title_fullStr | Forecasting High-Flow Discharges in a Flashy Catchment Using Multiple Precipitation Estimates as Predictors in Machine Learning Models |
| title_full_unstemmed | Forecasting High-Flow Discharges in a Flashy Catchment Using Multiple Precipitation Estimates as Predictors in Machine Learning Models |
| title_short | Forecasting High-Flow Discharges in a Flashy Catchment Using Multiple Precipitation Estimates as Predictors in Machine Learning Models |
| title_sort | forecasting high flow discharges in a flashy catchment using multiple precipitation estimates as predictors in machine learning models |
| topic | flash floods machine learning rainfall-runoff flood forecasting urban catchment |
| url | https://www.mdpi.com/2306-5338/9/12/216 |
| work_keys_str_mv | AT andredlzanchetta forecastinghighflowdischargesinaflashycatchmentusingmultipleprecipitationestimatesaspredictorsinmachinelearningmodels AT paulincoulibaly forecastinghighflowdischargesinaflashycatchmentusingmultipleprecipitationestimatesaspredictorsinmachinelearningmodels AT vincentfortin forecastinghighflowdischargesinaflashycatchmentusingmultipleprecipitationestimatesaspredictorsinmachinelearningmodels |