Microplastics migration mechanisms in high-erosion watersheds under climate warming

Area of Science:

• Environmental Science
• Ecology
• Geoscience

Background:

• Microplastic (MP) pollution is a growing concern in aquatic ecosystems.
• Understanding MP migration is vital for effective pollution management.
• Limited long-term data and modeling approaches hinder progress in MP research.

Purpose of the Study:

• To develop and validate a novel machine learning model for assessing MP migration in small watersheds.
• To identify sources and pathways of MP migration under varying land use and sediment conditions.
• To project future MP migration trends influenced by climate change and emission scenarios.

Main Methods:

• Utilized 15 years of sediment data from three distinct watersheds on the Qinghai-Xizang Plateau.
• Developed a novel MPs migration model incorporating Random Forest (RF), SHAP, and Deep Neural Network (DNN) algorithms.
• Validated model accuracy for source tracing (R² = 0.93) and pathway analysis (R² = 0.97).

Main Results:

• Cropland is a primary MP source under low sediment conditions (<6.5 cm), with significant contributions from grassland as sediment thickness increases.
• Climate warming intensified extreme precipitation, shifting MP migration patterns towards higher sediment scenarios.
• Future projections indicate a substantial increase in MP migration and content under high-emission scenarios, while low-emission scenarios show decreased migration but slightly increased content.

Conclusions:

• The developed machine learning model provides a robust framework for understanding MP migration dynamics.
• Sediment thickness, land use, and climate change are critical factors influencing MP migration.
• Findings support the development of targeted strategies for sustainable watershed pollution management.