Back to top

Introduction

Investigating the Impact of Marine Heatwave on Seagrass Species A Machine Learning Approach
  • Introduction
  • Methodology
  • Visulization
  • Findings
  • Further Possible Research

    • Introduction

    Background and Importance

    Seagrasses are vital components of marine ecosystems, offering crucial services such as carbon sequestration, biodiversity conservation, and acting as nurseries for numerous marine species. They play a significant role in coastal protection and the maintenance of water quality. However, marine heatwaves, driven by global climate change, pose a significant threat to these ecosystems. These extreme temperature events can lead to widespread seagrass die-offs, resulting in the loss of their valuable ecosystem services. Understanding how different seagrass species respond to such thermal stress is essential for predicting the resilience of these ecosystems to climate change and for devising effective conservation strategies.

    Objectives and Hypotheses

    This analysis aims to compare the resistance of two tropical seagrass species to marine heatwaves, namely Halophila beccarii and Halophila ovalis, focusing on identifying distinct indicators of resilience through physiological and ecological responses. By leveraging machine learning algorithms to analyze experimental data, the research seeks to uncover patterns and predictors of seagrass survival under thermal stress. We hypothesize that differences in genetic makeup, physiological adaptations, and ecological interactions will result in distinct resistance indicators between the species, which can be effectively identified using advanced data analysis techniques.

    Methodological Overview

    The methodology involves using comprehensive datasets from controlled experiments, encompassing various physiological and ecological parameters of the seagrass species under study. The datasets will undergo preprocessing to ensure quality and compatibility for analysis. Machine learning algorithms, including supervised learning models, will then be applied to identify and validate indicators of heatwave resistance. The choice of machine learning as a tool is predicated on its ability to handle complex, multidimensional data and to reveal insights that might not be apparent through traditional statistical methods.

    Acknowlegement

    Special thanks to BASS Alissa V. and FALKENBERG Laura J. for their pivotal contributions to the study “Two tropical seagrass species show differing indicators of resistance to a marine heatwave,” available at https://doi.org/10.48668/T2WBX8, CUHK Research Data Repository, V2, UNF:6:Aw0dPH+DXKAHtk/PeP1FZA== [fileUNF]. Their work has significantly enriched our understanding and provided a foundational dataset for our research.

    We would also like to extend our gratitude to our coach from the CUHK Library for their invaluable support and guidance throughout this project. Their expertise and encouragement were instrumental in helping us navigate the challenges of this research.

    Teammates

    • DAI Zihan, David (MIEG/3)
    • HUANG Chengbo, Simon (MIEG/3)
    • QI Haomin, Harmin (MIEG/3)