Marine Microplastics

Motivation

The pathways of marine microplastics in the global ocean are shaped by the complex interactions of general circulation and microplastic properties. While early models simplified transport by treating microplastics as passive surface tracers, real-world microplastics are distributed throughout the water column.  

We develop a global 3-D Eulerian transport framework integrated into the Massachusetts Institute of Technology General Circulation Model (MITgcm) and constrained by data-assimilated ocean state estimates (ECCO). By coupling hydrodynamics dependent on particle size and density with biogeochemical and mechanical processes (i.e., biofouling and fragmentation), our model reveals how microplastics are vertically redistributed and horizontally dispersed across the global ocean.

Size and Density-Dependent 3-D Transport

• The Sifting Effect: Low-density, buoyant microplastics (e.g., Polyethylene or PE, ρp ≈ 900 kg/m3) accumulate in the five subtropical gyres only if their particle sizes are sufficiently large (d ≥ 10 μm).

• Neutrally Buoyant Penetration: As particle size decreases below a critical threshold (d ≤ 1 μm for PE/PVC; d ≤ 10 μm for PS), microplastics behave as neutrally buoyant tracers. They are sifted downward by vertical mixing and penetrate up to 1 km deep into the ocean interior.

• Seasonal Variability: The concentration of floating microplastics changes seasonally, peaking in summer and dropping in winter. This seasonal variation is strongly correlated with the mixed layer depth (MLD). Because buoyant particles are uniformly distributed within the mixed layer, their surface concentration scales inversely with MLD—a physical mechanism that matches satellite observations from NASA's CYGNSS.

• The Subsurface Dispersal Route: Biofouled microplastics sink below the mixed layer, where they encounter reversed subsurface currents flowing in the opposite direction of the surface circulation. These currents carry the particles outward from the gyre centers. Once the biofilm decays in the dark subsurface ocean, the particles regain buoyancy and resurface elsewhere, smoothing out surface garbage patch concentrations.

• Better Alignment with Observation: Validating the biofouling model against global trawler datasets (Isobe et al., 2021) shows a significant improvement in predictive skill, reducing the fraction of model-observation outliers (differing by over 2 orders of magnitude) from 25% to 13%.

• Size Evolution: We developed the first global 3D Eulerian model that couples fragmentation kinetics with transport. Microplastic mass is discretized into N=10 logarithmically spaced bins corresponding to diameters from 1 μm to 60 μm.

• In-situ Generation of Small Particles: Previous models assumed small microplastics were only released along coastlines. Our model demonstrates that small, neutrally buoyant fragments (d < 1 μm) can be generated in situ due to the fragmentation of accumulated larger debris. Consequently, these neutrally buoyant fragments peak at the gyre centers rather than the coasts.

• Correlation Improvements: Resolving fragmentation reduces the simulated surface concentration variance by up to 7 times in major basins, matching observed patterns. It also improves the spatial correlation between the model and observed surface concentrations (AOMI Level-3 dataset) from 45% to 58%.