Scholarlink

<p>Sea surface temperature (SST) is a crucial factor in ocean analysis, playing a vital role in understanding oceanic dynamics. The changes in SST have significant implications for global warming or climate change, as seen in its potential for extreme weather events like droughts and floods. Tropical cyclone (TC) development is significantly influenced by SST, which serves as a key driver of their growth and strengthening, and its fluctuations are closely tied to the presence and intensity of these storms. Short and medium-range SST forecasting is vital since it can aid in detecting these extreme events and reduce the losses they generate through timely warnings and preventive measures. The usage of realistic forecasts of SST in high-resolution TC models will improve the estimation of its lifetime, intensity, and rainfall. Accuracy in SST forecasts has a tremendous economic and social impact. The forecasted SST can also help tackle the problem of gaps in satellite SST data. With the large-scale availability of high-resolution satellite data, data-driven techniques are gaining popularity and are being used for the short-range forecasting of SST. The study provides a detailed analysis of various machine learning and deep learning techniques for predicting SST with a lead time of 1–7 days in the Arabian Sea, Bay of Bengal, and South Indian Ocean regions. Furthermore, it recommends a suitable approach for operational use based on a thorough statistical evaluation of these forecasting methods.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Forecasting of sea surface temperature using machine learning and its applications

  • 期刊论文

  • 发表日期：

    2025年1月24日

查看全文
  • Vishnu Vytla,
  • Balaji Baduru,
  • Siva Srinivas Kolukula,
  • N Nithish Ragav,
  • J Pavan Kumar

摘要

Sea surface temperature (SST) is a crucial factor in ocean analysis, playing a vital role in understanding oceanic dynamics. The changes in SST have significant implications for global warming or climate change, as seen in its potential for extreme weather events like droughts and floods. Tropical cyclone (TC) development is significantly influenced by SST, which serves as a key driver of their growth and strengthening, and its fluctuations are closely tied to the presence and intensity of these storms. Short and medium-range SST forecasting is vital since it can aid in detecting these extreme events and reduce the losses they generate through timely warnings and preventive measures. The usage of realistic forecasts of SST in high-resolution TC models will improve the estimation of its lifetime, intensity, and rainfall. Accuracy in SST forecasts has a tremendous economic and social impact. The forecasted SST can also help tackle the problem of gaps in satellite SST data. With the large-scale availability of high-resolution satellite data, data-driven techniques are gaining popularity and are being used for the short-range forecasting of SST. The study provides a detailed analysis of various machine learning and deep learning techniques for predicting SST with a lead time of 1–7 days in the Arabian Sea, Bay of Bengal, and South Indian Ocean regions. Furthermore, it recommends a suitable approach for operational use based on a thorough statistical evaluation of these forecasting methods.