Sea surface temperature (SST) is the major factor that affects the ocean-atmosphere interaction, and in turn the accurate prediction of SST is the key to ocean dynamic prediction. In this paper, an SST-predicting method based on empirical mode decomposition (EMD) algorithms and back-propagation neural network (BPNN) is proposed. Two different EMD algorithms have been applied extensively for analyzing time-series SST data and some nonlinear stochastic signals. The ensemble empirical mode decomposition (EEMD) algorithm and complementary ensemble empirical mode decomposition (CEEMD) algorithm are two improved algorithms of EMD, which can effectively handle the mode-mixing problem and decompose the original data into more stationary signals with different frequencies. Each intrinsic mode function (IMF) has been taken as input data to the back-propagation neural network model. The final predicted SST data are obtained by aggregating the predicted data of individual series of IMFs (IMFi). A case study of the monthly mean SST anomaly (SSTA) in the northeastern region of the North Pacific shows that the proposed hybrid CEEMD-BPNN model is much more accurate than the hybrid EEMD-BPNN model, and the prediction accuracy based on a BP neural network is improved by the CEEMD method. Statistical analysis of the case study demonstrates that applying the proposed hybrid CEEMD-BPNN model is effective for the SST prediction. Highlights include the following: Highlights.-An SST-predicting method based on the hybrid EMD algorithms and BP neural network method is proposed in this paper.-SST prediction results based on the hybrid EEMD-BPNN and CEEMD-BPNN models are compared and discussed.-A case study of SST in the North Pacific shows that the proposed hybrid CEEMD-BPNN model can effectively predict the time-series SST.
Abstract. The variability of the sea surface temperature (SST) in the northwest Pacific has been studied on seasonal, annual and interannual scales based on the monthly datasets of extended reconstructed sea surface temperature (ERSST) 3b (1854–2017, 164 years) and optimum interpolation sea surface temperature version 2 (OISST V2 (1988–2017, 30 years). The overall trends, spatial–temporal distribution characteristics, regional differences in seasonal trends and seasonal differences of SST in the northwest Pacific have been calculated over the past 164 years based on these datasets. In the past 164 years, the SST in the northwest Pacific has been increasing linearly year by year, with a trend of 0.033 ∘C/10 years. The SST during the period from 1870 to 1910 is slowly decreasing and staying in the range between 25.2 and 26.0 ∘C. During the period of 1910–1930, the SST as a whole maintained a low value, which is at the minimum of 164 years. After 1930, SST continued to increase until now. The increasing trend in the past 30 years has reached 0.132 ∘C/10 years, and the increasing trend in the past 10 years is 0.306 ∘C/10 years, which is around 10 times that of the past 164 years. The SST in most regions of the northwest Pacific showed a linear increasing trend year by year, and the increasing trend in the offshore region was stronger than that in the ocean and deep-sea region. The change in trend of the SST in the northwest Pacific shows a large seasonal difference, and the increasing trend in autumn and winter is larger than that in spring and summer. There are some correlations between the SST and some climate indices and atmospheric parameters; the correlations between the SST and some atmospheric parameters have been discussed, such as those of the North Atlantic Oscillation (NAO), Pacific Decadal Oscillation (PDO), Southern Oscillation Index (SOI) anomaly, total column water (TCW), NINO3.4 index, sea level pressure (SLP), precipitation, temperature at 2 m (T2) and wind speed. The lowest SST in China offshore basically occurred in February and the highest in August. The SST fluctuation in the Bohai Sea and Yellow Sea (BYS) is the largest, with a range from 5 to 22 ∘C; the SST in the East China Sea (ECS) is from 18 to 27 ∘C; the smallest fluctuations occur in the South China Sea (SCS), maintained at range of 26 to 29 ∘C. There are large differences between the mean and standard deviation in different sea regions.
The Northwest Pacific and the South China Sea region are the birthplaces of most monsoon disturbances and tropical cyclones and are an important channel for the generation and transmission of water vapor. The Northwest Pacific plays a major role in regulating interdecadal and long-term changes in climate. China experiences the largest number of typhoon landfalls and the most destructive power affected by typhoons in the world. The hidden dangers of typhoon disasters are accelerating with the acceleration of urbanization, the rapid development of economic construction and global warming. The coastal cities are the most dynamic and affluent areas of China's economic development. They are the strong magnetic field that attracts international capital in China, and are also the most densely populated areas and important port groups in China. Although these regions are highly developed, they are vulnerable to disasters. When typhoons hit, the economic losses and casualties caused by gale, heavy rain and storm surges were particularly serious. This chapter reviews the response of coastal ocean to tropical cyclones, included sea surface temperature, sea surface salinity, storm surge simulation and extreme rainfall under the influence of tropical cyclones.
Abstract. The variability of the sea surface temperature (SST) in the Northwest Pacific has been studied on seasonal, annual and interannual scales based on the monthly datasets of ERSST 3b (1854–2017, 164 years) and OISST V2 (1988–2017, 30 years). The overall trends, spatial-temporal distribution characteristics, regional differences in seasonal trends, and seasonal differences of SST in the Northwest Pacific have been calculated over the past 164 years based on these datasets. In the past 164 years, the SST in the Northwest Pacific has been increasing linearly year by year with a trend of 0.033 °C/10 yr. The period from 1880 to 1910 is a slow decreasing trend period in the past 164 years and the SST during the 1910–1930 period was a trough of the past 164 years. After 1930, SST has continued to increase until now. The increasing trend in the past 30 years has reached 0.132 °C/10 yr and the increasing trend in the past 10 years is 0.306 °C/10 yr, which is around ten times in the past 164 years. The SST in most regions of the Northwest Pacific showed a linear increasing trend year by year, and the increasing trend in the offshore region was stronger than that in the ocean and deep-sea region. The change trend of the SST in the Northwest Pacific shows a large seasonal difference, and the increasing trend in autumn and winter is larger than that in spring and summer. There are some correlations between the SST and some climate indexes and atmospheric parameters, the correlation between the SST and some atmospheric parameters have been discussed, such as NAO, PDO, SOI anomaly, TCW, Nino 3.4, SLP, Precipitation, T2 and wind speed. The lowest SST in the Near China Sea basically occurred in February and the highest in August. The SST fluctuation in the Bohai Sea and Yellow Sea (BYS) is the largest with a range from 5 °C to 22 °C, the SST in the East China Sea (ECS) is from 18 °C to 27 °C, the smallest fluctuations occurs in the South China Sea (SCS) maintained at range of 26 °C to 29 °C. There are large differences between the mean and standard deviation in different sea regions.
Abstract. Sea surface temperature (SST) is the major factor that affects the ocean-atmosphere interaction, and in turn the accurate prediction of SST is the key to ocean dynamic prediction. In this paper, an SST predicting method based on improved empirical mode decomposition (EMD) algorithms and back-propagation neural network (BPNN) is proposed. Two different EMD algorithms have been applied extensively for analyzing time-series SST data and some nonlinear stochastic signals. Ensemble empirical mode decomposition (EEMD) algorithm and Complementary Ensemble Empirical Mode Decomposition (CEEMD) algorithm are two improved algorithms of EMD, which can effectively handle the mode-mixing problem and decompose the original data into more stationary signals with different frequencies. Each Intrinsic Mode Function (IMF) has been taken as an input data to the back-propagation neural network model. The final predicted SST data is obtained by aggregating the predicted data of individual IMF. A case study, of the monthly mean sea surface temperature anomaly (SSTA) in the northeastern region of the North Pacific, shows that the proposed hybrid CEEMD-BPNN model is much more accurate than the hybrid EEMD-BPNN model, and the prediction accuracy based on BP neural network is improved by the CEEMD method. Statistical analysis of the case study demonstrates that applying the proposed hybrid CEEMD-BPNN model is effective for the SST prediction.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.