Seasonal Prediction of Summer Precipitation in the Middle and Lower Reaches of the Yangtze River Valley: Comparison of Machine Learning and Climate Model Predictions

The middle and lower reaches of the Yangtze River valley (YRV), which are among the most densely populated regions in China, are subject to frequent flooding. In this study, the predictor importance analysis model was used to sort and select predictors, and five methods (multiple linear regression (...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Chentao He, Jiangfeng Wei, Yuanyuan Song, Jing-Jia Luo
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	Yangtze River valley seasonal prediction random forest machine learning Hydraulic engineering TC1-978 Water supply for domestic and industrial purposes TD201-500
Acceso en línea:	https://doaj.org/article/9821b115db424b10b09bd9fe58c98b66
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:9821b115db424b10b09bd9fe58c98b66
record_format	dspace
spelling	oai:doaj.org-article:9821b115db424b10b09bd9fe58c98b662021-11-25T19:16:31ZSeasonal Prediction of Summer Precipitation in the Middle and Lower Reaches of the Yangtze River Valley: Comparison of Machine Learning and Climate Model Predictions10.3390/w132232942073-4441https://doaj.org/article/9821b115db424b10b09bd9fe58c98b662021-11-01T00:00:00Zhttps://www.mdpi.com/2073-4441/13/22/3294https://doaj.org/toc/2073-4441The middle and lower reaches of the Yangtze River valley (YRV), which are among the most densely populated regions in China, are subject to frequent flooding. In this study, the predictor importance analysis model was used to sort and select predictors, and five methods (multiple linear regression (MLR), decision tree (DT), random forest (RF), backpropagation neural network (BPNN), and convolutional neural network (CNN)) were used to predict the interannual variation of summer precipitation over the middle and lower reaches of the YRV. Predictions from eight climate models were used for comparison. Of the five tested methods, RF demonstrated the best predictive skill. Starting the RF prediction in December, when its prediction skill was highest, the 70-year correlation coefficient from cross validation of average predictions was 0.473. Using the same five predictors in December 2019, the RF model successfully predicted the YRV wet anomaly in summer 2020, although it had weaker amplitude. It was found that the enhanced warm pool area in the Indian Ocean was the most important causal factor. The BPNN and CNN methods demonstrated the poorest performance. The RF, DT, and climate models all showed higher prediction skills when the predictions start in winter than in early spring, and the RF, DT, and MLR methods all showed better prediction skills than the numerical climate models. Lack of training data was a factor that limited the performance of the machine learning methods. Future studies should use deep learning methods to take full advantage of the potential of ocean, land, sea ice, and other factors for more accurate climate predictions.Chentao HeJiangfeng WeiYuanyuan SongJing-Jia LuoMDPI AGarticleYangtze River valleyseasonal predictionrandom forestmachine learningHydraulic engineeringTC1-978Water supply for domestic and industrial purposesTD201-500ENWater, Vol 13, Iss 3294, p 3294 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Yangtze River valley seasonal prediction random forest machine learning Hydraulic engineering TC1-978 Water supply for domestic and industrial purposes TD201-500
spellingShingle	Yangtze River valley seasonal prediction random forest machine learning Hydraulic engineering TC1-978 Water supply for domestic and industrial purposes TD201-500 Chentao He Jiangfeng Wei Yuanyuan Song Jing-Jia Luo Seasonal Prediction of Summer Precipitation in the Middle and Lower Reaches of the Yangtze River Valley: Comparison of Machine Learning and Climate Model Predictions
description	The middle and lower reaches of the Yangtze River valley (YRV), which are among the most densely populated regions in China, are subject to frequent flooding. In this study, the predictor importance analysis model was used to sort and select predictors, and five methods (multiple linear regression (MLR), decision tree (DT), random forest (RF), backpropagation neural network (BPNN), and convolutional neural network (CNN)) were used to predict the interannual variation of summer precipitation over the middle and lower reaches of the YRV. Predictions from eight climate models were used for comparison. Of the five tested methods, RF demonstrated the best predictive skill. Starting the RF prediction in December, when its prediction skill was highest, the 70-year correlation coefficient from cross validation of average predictions was 0.473. Using the same five predictors in December 2019, the RF model successfully predicted the YRV wet anomaly in summer 2020, although it had weaker amplitude. It was found that the enhanced warm pool area in the Indian Ocean was the most important causal factor. The BPNN and CNN methods demonstrated the poorest performance. The RF, DT, and climate models all showed higher prediction skills when the predictions start in winter than in early spring, and the RF, DT, and MLR methods all showed better prediction skills than the numerical climate models. Lack of training data was a factor that limited the performance of the machine learning methods. Future studies should use deep learning methods to take full advantage of the potential of ocean, land, sea ice, and other factors for more accurate climate predictions.
format	article
author	Chentao He Jiangfeng Wei Yuanyuan Song Jing-Jia Luo
author_facet	Chentao He Jiangfeng Wei Yuanyuan Song Jing-Jia Luo
author_sort	Chentao He
title	Seasonal Prediction of Summer Precipitation in the Middle and Lower Reaches of the Yangtze River Valley: Comparison of Machine Learning and Climate Model Predictions
title_short	Seasonal Prediction of Summer Precipitation in the Middle and Lower Reaches of the Yangtze River Valley: Comparison of Machine Learning and Climate Model Predictions
title_full	Seasonal Prediction of Summer Precipitation in the Middle and Lower Reaches of the Yangtze River Valley: Comparison of Machine Learning and Climate Model Predictions
title_fullStr	Seasonal Prediction of Summer Precipitation in the Middle and Lower Reaches of the Yangtze River Valley: Comparison of Machine Learning and Climate Model Predictions
title_full_unstemmed	Seasonal Prediction of Summer Precipitation in the Middle and Lower Reaches of the Yangtze River Valley: Comparison of Machine Learning and Climate Model Predictions
title_sort	seasonal prediction of summer precipitation in the middle and lower reaches of the yangtze river valley: comparison of machine learning and climate model predictions
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/9821b115db424b10b09bd9fe58c98b66
work_keys_str_mv	AT chentaohe seasonalpredictionofsummerprecipitationinthemiddleandlowerreachesoftheyangtzerivervalleycomparisonofmachinelearningandclimatemodelpredictions AT jiangfengwei seasonalpredictionofsummerprecipitationinthemiddleandlowerreachesoftheyangtzerivervalleycomparisonofmachinelearningandclimatemodelpredictions AT yuanyuansong seasonalpredictionofsummerprecipitationinthemiddleandlowerreachesoftheyangtzerivervalleycomparisonofmachinelearningandclimatemodelpredictions AT jingjialuo seasonalpredictionofsummerprecipitationinthemiddleandlowerreachesoftheyangtzerivervalleycomparisonofmachinelearningandclimatemodelpredictions
_version_	1718410108087566336

Seasonal Prediction of Summer Precipitation in the Middle and Lower Reaches of the Yangtze River Valley: Comparison of Machine Learning and Climate Model Predictions

Ejemplares similares