Scholarlink

The Atlantic coastline of Morocco constitutes a vital component of the Easter Boundaries Upwelling Ecosystem, renowned for its copious marine biodiversity and the consistent occurrence of upwelling throughout the year. In this research endeavor, we have devised sophisticated deep learning methodologies to proficiently observe and assess the Moroccan upwelling patterns by leveraging satellite imagery. Our approach involves a state-of-the-art convolutional neural network (CNN) based on an encoder-decoder architecture called \(FusDeep_{up}\) -Net, specifically designed to identify and localize upwelling regions considering the input parameters the fusion results of 908 weekly Sea Surface Temperature (SST) and chlorophyll-a (Chl-a) images spanning from 2000 to 2019. To validate our methodology, we have applied it to a comprehensive database of 92 8-day SST images for the period 2021–2022. In order to investigate and analyze the upwelling activity in this particular region, we utilized a thermal index, denoted as \(I_{SST}\) , which was applied to the sea surface temperature (SST) dataset. The insights derived from this index significantly enhance our comprehension of this crucial phenomenon and its potential ramifications for the coastal environment.

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

Fusion and Deep Learning Methods for Upwelling Coast Extraction from Biological and Physical Satellite Images

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Z. El Abidi,
  • H. Belmajdoub,
  • K. Minaoui

摘要

The Atlantic coastline of Morocco constitutes a vital component of the Easter Boundaries Upwelling Ecosystem, renowned for its copious marine biodiversity and the consistent occurrence of upwelling throughout the year. In this research endeavor, we have devised sophisticated deep learning methodologies to proficiently observe and assess the Moroccan upwelling patterns by leveraging satellite imagery. Our approach involves a state-of-the-art convolutional neural network (CNN) based on an encoder-decoder architecture called \(FusDeep_{up}\) -Net, specifically designed to identify and localize upwelling regions considering the input parameters the fusion results of 908 weekly Sea Surface Temperature (SST) and chlorophyll-a (Chl-a) images spanning from 2000 to 2019. To validate our methodology, we have applied it to a comprehensive database of 92 8-day SST images for the period 2021–2022. In order to investigate and analyze the upwelling activity in this particular region, we utilized a thermal index, denoted as \(I_{SST}\) , which was applied to the sea surface temperature (SST) dataset. The insights derived from this index significantly enhance our comprehension of this crucial phenomenon and its potential ramifications for the coastal environment.