TY - JOUR
T1 - River water temperature forecasting using a deep learning method
AU - Qiu, Rujian
AU - Wang, Yuankun
AU - Rhoads, Bruce
AU - Wang, Dong
AU - Qiu, Wenjie
AU - Tao, Yuwei
AU - Wu, Jichun
N1 - This study was supported by National Key Research and Development Program of China (2017YFC1502704), and National Natural Science Fund of China (No.51679118).
This study was supported by National Key Research and Development Program of China (2017YFC1502704), and National Natural Science Fund of China (No.51679118). Data on daily discharge and water temperature at Yichang, Cuntan and Datong hydrological stations were provided by the Ministry of Water Conservancy of China (CWRC) (http://www.cjw.gov.cn/). The daily inflow of discharge to the TGR from 2003-2013 were obtained from the China Three Gorges Corporation (https://www.ctg.com.cn/). Daily air temperature for Yichang, Cuntan and Datong gauges chosen from the nearest national meteorological stations were provided by Meteorological Data Services Center, China (https://data.cma.cn/). Daily water temperature and discharge for three Swiss rivers (Mentue, Rh?ne, Dischmabach) were provided by Swiss Federal Office of the Environment (FOEN) (http://www.bafu.admin.ch/wasser/13465/13483/14087/index.html?lang = en). Daily air temperature for these three Swiss rivers were provide by the Swiss Meteorological Institute (MeteoSwiss) (http://www.meteoswiss.admin.ch/home/measurement-and-forecasting-systems/land-based-stations/automatisches-messnetz.html), Data of daily water temperature and discharge of the three rivers in USA (Cedar, Fanno and Irondequoit) are available from the United States Geological Survey (USGS) (https://waterdata.usgs.gov/nwis/inventory), and daily meteorological data of the three rivers are available from National Centers for Environmental Information, National Oceanic and Atmospheric Administration (NOAA) (https://www.ncdc.noaa.gov/cdo-web/datasets).
PY - 2021/4
Y1 - 2021/4
N2 - Accurate water temperature forecasting is essential for understanding thermal regimes of rivers in the context of climate change and anthropogenic disturbances, such as dam construction. Machine-learning models proffer an empirically based approach to predicting water temperatures with a high degree of accuracy. This study explores the potential of long short-term neural network (LSTM), a type of deep learning method, to forecast daily river water temperatures and quantify temporal variations in thermal regime induced by changes in climate and by dam construction. The performance of LSTM is compared with that of several other models using daily water-temperature data for nine river gauges around the world. In a detailed analysis, the models are evaluated for the Yichang gauge on the Yangtze River to reconstruct the natural thermal conditions and help to assess daily water temperature variations induced by operation of the Three Gorges Reservoir (TGR). The collective results show that LSTM outperforms other methods for predicting mean daily water temperature in rivers, capturing accurately mean daily variations in thermal regime. The construction of the TGR strongly influenced water temperature variations at Yichang, producing the strongest cooling effect from mid-April to mid-May and the greatest warming effect in late December and early January. These marked effects are most prominent at the highest water levels in the TGR. The enhanced predictive capabilities of the LSTM model provide a powerful tool for water temperature forecasting and ecological management of rivers in the Anthropocene.
AB - Accurate water temperature forecasting is essential for understanding thermal regimes of rivers in the context of climate change and anthropogenic disturbances, such as dam construction. Machine-learning models proffer an empirically based approach to predicting water temperatures with a high degree of accuracy. This study explores the potential of long short-term neural network (LSTM), a type of deep learning method, to forecast daily river water temperatures and quantify temporal variations in thermal regime induced by changes in climate and by dam construction. The performance of LSTM is compared with that of several other models using daily water-temperature data for nine river gauges around the world. In a detailed analysis, the models are evaluated for the Yichang gauge on the Yangtze River to reconstruct the natural thermal conditions and help to assess daily water temperature variations induced by operation of the Three Gorges Reservoir (TGR). The collective results show that LSTM outperforms other methods for predicting mean daily water temperature in rivers, capturing accurately mean daily variations in thermal regime. The construction of the TGR strongly influenced water temperature variations at Yichang, producing the strongest cooling effect from mid-April to mid-May and the greatest warming effect in late December and early January. These marked effects are most prominent at the highest water levels in the TGR. The enhanced predictive capabilities of the LSTM model provide a powerful tool for water temperature forecasting and ecological management of rivers in the Anthropocene.
KW - Deep learning
KW - Long short-term memory neural network
KW - The Yangtze River
KW - Three Gorges Reservoir
KW - Water temperature forecasting
UR - http://www.scopus.com/inward/record.url?scp=85100590961&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85100590961&partnerID=8YFLogxK
U2 - 10.1016/j.jhydrol.2021.126016
DO - 10.1016/j.jhydrol.2021.126016
M3 - Article
AN - SCOPUS:85100590961
SN - 0022-1694
VL - 595
JO - Journal of Hydrology
JF - Journal of Hydrology
M1 - 126016
ER -