Sea Surface Temperature and Marine Heat Wave Predictions in the South China Sea: A 3D U-Net Deep Learning Model Integrating Multi-Source Data DOI Creative Commons
Bowen Xie, Jifeng Qi, Shuguo Yang

et al.

Atmosphere, Journal Year: 2024, Volume and Issue: 15(1), P. 86 - 86

Published: Jan. 9, 2024

Accurate sea surface temperature (SST) prediction is vital for disaster prevention, ocean circulation, and climate change. Traditional SST methods, predominantly reliant on time-intensive numerical models, face challenges in terms of speed efficiency. In this study, we developed a novel deep learning approach using 3D U-Net structure with multi-source data to forecast the South China Sea (SCS). SST, height anomaly (SSHA), wind (SSW) were used as input variables. Compared convolutional long short-term memory (ConvLSTM) model, model achieved more accurate predictions at all lead times (from 1 30 days) performed better different seasons. Spatially, model’s exhibited low errors (RMSE < 0.5 °C) high correlation (R > 0.9) across most SCS. The spatially averaged time series both predicted by observed 2021, showed remarkable consistency. A noteworthy application research was successful detection marine heat wave (MHW) events SCS 2021. accurately captured occurrence frequency, total duration, average cumulative intensity MHW events, aligning closely data. Sensitive experiments that SSHA SSW have significant impacts which can improve accuracy play roles periods. combination variables, not only rapidly but also presented method forecasting highlighting its potential advantages.

Language: Английский

The 2022 summer marine heatwaves and coral bleaching in China's Greater Bay Area DOI
Yu Zhao, Mingru Chen, Tzu Hao Chung

et al.

Marine Environmental Research, Journal Year: 2023, Volume and Issue: 189, P. 106044 - 106044

Published: June 5, 2023

Language: Английский

Citations

13
Spatiotemporal Distribution of Heatwave Hazards in the Chinese Mainland for the Period 1990–2019 DOI Open Access
Wei Wu, Qingsheng Liu, He Li

et al.

International Journal of Environmental Research and Public Health, Journal Year: 2023, Volume and Issue: 20(2), P. 1532 - 1532

Published: Jan. 14, 2023

Heatwaves occur frequently in summer, severely harming the natural environment and human society. While a few long-term spatiotemporal heatwave studies have been conducted China at grid scale, their shortcomings involve discrete distribution poor continuity. We used daily data from 691 meteorological stations to obtain torridity index (TI) (HWI) datasets (0.01°) order evaluate of heatwaves Chinese mainland for period 1990-2019. The results were as follows: (1) TI values rose but with fluctuations, largest increase occurring North July. areas hazard levels medium above accounted 22.16% total, mainly eastern southern provinces China, South Tibet, East Xinjiang, Chongqing. (2) study divided into four categories according hazards. "high rapidly increasing" "low continually 8.71% 41.33% respectively. (3) "ten furnaces" top provincial capitals Zhengzhou, Nanchang, Wuhan, Changsha, Shijiazhuang, Nanjing, Hangzhou, Haikou, Chongqing, Hefei. urbanization level population aging developed further increased, continuously increasing should be fully considered.

Language: Английский

Citations

12
Temperature sensitivity of marine macroalgae for aquaculture in China DOI
Yuyang Zhang, Shuangen Yu, Wenlei Wang

et al.

Aquaculture, Journal Year: 2023, Volume and Issue: 567, P. 739262 - 739262

Published: Jan. 18, 2023

Language: Английский

Citations

11
Increasing impacts of summer extreme precipitation and heatwaves in eastern China DOI

Yao Yulong,

Wei Zhang, Ben P. Kirtman

et al.

Climatic Change, Journal Year: 2023, Volume and Issue: 176(10)

Published: Sept. 19, 2023

Language: Английский

Citations

11
Sea Surface Temperature and Marine Heat Wave Predictions in the South China Sea: A 3D U-Net Deep Learning Model Integrating Multi-Source Data DOI Creative Commons
Bowen Xie, Jifeng Qi, Shuguo Yang

et al.

Atmosphere, Journal Year: 2024, Volume and Issue: 15(1), P. 86 - 86

Published: Jan. 9, 2024

Accurate sea surface temperature (SST) prediction is vital for disaster prevention, ocean circulation, and climate change. Traditional SST methods, predominantly reliant on time-intensive numerical models, face challenges in terms of speed efficiency. In this study, we developed a novel deep learning approach using 3D U-Net structure with multi-source data to forecast the South China Sea (SCS). SST, height anomaly (SSHA), wind (SSW) were used as input variables. Compared convolutional long short-term memory (ConvLSTM) model, model achieved more accurate predictions at all lead times (from 1 30 days) performed better different seasons. Spatially, model’s exhibited low errors (RMSE < 0.5 °C) high correlation (R > 0.9) across most SCS. The spatially averaged time series both predicted by observed 2021, showed remarkable consistency. A noteworthy application research was successful detection marine heat wave (MHW) events SCS 2021. accurately captured occurrence frequency, total duration, average cumulative intensity MHW events, aligning closely data. Sensitive experiments that SSHA SSW have significant impacts which can improve accuracy play roles periods. combination variables, not only rapidly but also presented method forecasting highlighting its potential advantages.

Language: Английский

Citations

4