Motivation:
Abnormal sea surface temperatures (SST) can have devastating effects on coastal regions, leading to rapid erosion, ecosystem degradation, and more frequent extreme weather events. These changes threaten fisheries, coastal city safety, and hinder sustainable development, particularly in tropical areas. Precise monitoring of SST is crucial for understanding and mitigating phenomena like coral bleaching and marine heat waves.
 

Goals:
This project aims to develop a high-accuracy, continuous high spatiotemporal resolution SST dataset using multi-source remote sensing data and deep learning methods. By creating an advanced SST fusion method, we will enhance our understanding of SST dynamics, ultimately promoting the sustainable management of marine resources and environments.
 

Tasks and Opportunities:

• Data Collection and Preprocessing: Collect and preprocess multi-source SST remote sensing data. (Primarily handled by mentors; students will have limited involvement in this task.)
• Fusion Model Design: Explore deep learning approaches and develop a framework tailored for SST data fusion.
• Model Training: Train and validate the fusion model using real SST data.
• Further Opportunity - Pattern Analysis: Interested and capable students will have the opportunity to join further research after closing this project, such as analyzing spatiotemporal patterns of SST using the high-resolution dataset obtained from the fusion process.

Requirements for Students:

• Background in remote sensing, machine learning, or related fields.
• Experience with data processing and a good understanding of deep learning frameworks.
• Strong programming skills in Python or similar languages.
• Interest in marine science and sustainable development.

Join us in this cutting-edge research project to make a tangible impact on coastal sustainability and marine resource management.

Apply to this project here