Global mapping for the occurrence of all-sized microplastics in seafloor sediments

This is a series of codes for extracting state data on ocean surface currents and near-bed thermohaline currents, along with formulas for calculating ocean state parameters. These codes enable the acquisition of surface current conditions at any sampling point in the ocean at any longitude, latitude, and time, as well as the conditions of deep-sea currents. Data derived from these codes can be analyzed for driving forces in conjunction with the microplastic data presented in the main text of this study. By interpreting the state of global ocean currents, it is possible to deduce the mechanism by which ocean currents influence the burial of microplastics into seabed sediments, and to predict the total mass of microplastics buried in global seabed sediments. In the foreseeable future state of the oceans, predictions can be made about the pathways of microplastic migration in the ocean and the hotspots for their accumulation over the coming years. 1. Structure of the series code 1.1. Code The code for predicting the abundance of microplastics in global seabed sediments has been divided into five code folders [Sea_surface_data_extraction/](Sea_surface_data_extraction/README.md): This folder includes various raw NC data, extraction codes, and extracted data results, which are used to extract surface ocean current data corresponding to 155 sampling points in the 12 sea areas mentioned in this research article and 2,024 literature-derived observations. [Seabed_data_extraction_research/](Seabed_data_extraction_research/README.md): This folder includes various raw NC data, extraction codes, and extracted data results, which are used to extract near real-time state data of 155 sampling points in the 12 sea areas mentioned in this research article, as well as the near-bed thermohaline currents corresponding to 2,024 literature-derived observations. [Global_ocean_remote_sensing_data_extraction/](Global_ocean_remote_sensing_data_extraction/README.md): This folder consists of two parts: Global submarine current extraction and Global wave extraction. It describes how to extract surface wave and seabed current information corresponding to grid points through global 2D longitude and latitude grid files. [Seawater_density_calculation/](Seawater_density_calculation/README.md): This folder is used to calculate the density of seawater at this time when extracting information on seabed currents. The EOS80 seawater state equation was used to estimate the density of large-scale, non extreme water currents on the global seabed. Before use, it is necessary to convert the relative pressure at the bottom of the seabed into absolute pressure and then incorporate it into the calculation process. [Machine_learning_predictive_models/](Machine_learning_predictive_models/README.md): The folder integrates the aforementioned surface wave data and the state data of near-bed thermohaline currents. Using eight machine‑learning regression algorithms, we predict the abundance of MPs buried in global seafloor sediments and map global MP hotspots. We developed a leave‑one‑cluster‑out extrapolative cross‑validation (LOCO‑CV) framework that ensures the extrapolation span encompasses the separation between training and test sets, thereby reducing the risk of extrapolating to unrepresented environmental conditions during model training while maintaining consistency in the input feature space. It aims to determine the extent of microplastic content carried by deep-sea sediments, serving as a sink since the large-scale production of microplastics, and to identify the current hotspots of microplastic pollution. [Seafloor_MP_reservoir_calculation/](Seafloor_MP_reservoir_calculation/README.md): This folder consolidates the global data network of seafloor sediment microplastics predicted by the optimally extrapolating XGBoost ensemble derived from the machine-learning framework described above. We further quantified total microplastic mass within each 0.5° × 0.5° grid cell by implementing a Monte Carlo simulation (n equal to the number of models trained in bootstrapping and LOCO-CV), randomly sampling from the empirically constrained distribution of particle-to-mass (abundance-to-weight, mg) ratios measured in 155 samples and from habitat-specific sediment density distributions. This approach yields an integrated estimate of the global burial reservoir of microplastics in seafloor sediments. 1.2. Data sources For this article, we utilized a free ocean database to download global marine status data. You can choose the data source that suits your needs. For more details, please refer to the following link: `Sea_surface_wave_data/`: Global sea surface wave data is sourced from the [Global Ocean Physics Reanalysis](https://doi.org/10.48670/moi-00021) product, [Global ocean low and mid trophic levels biomass content hindcast](https://doi.org/10.48670/10.48670/moi-00020) product, [Global Ocean Waves Reanalysis](https://doi.org/10.48670/moi-00022) product and [Global Ocean Physics Analysis and Forecast](https://doi.org/10.48670/moi-00016) product within the Copernicus Marine Service. These can be downloaded using the toolbox provided on the website, eliminating the need for custom code development for downloading. `Seafloor_current_data/`: The global seabed current data originates from the [ECCO2](https://ecco.jpl.nasa.gov/drive/files/ECCO2/cube92_latlon_quart_90S90N/) database, developed by the Jet Propulsion Laboratory (JPL) under NASA (the National Aeronautics and Space Administration). 2. Using Code Ocean All ocean state data used in this analysis are derived from free and publicly accessible sources, available for download. However, the data concerning microplastics in this study are obtained from our own collected samples and from the abundance of microplastics recorded in selected literature. For a comprehensive understanding of the entire process, one can refer to this article. 3. Citation If you use this seawater density formula for research applications and publish papers, please cite: 1. Jingchun Feng, Canrong Li, Xiaonan Wu. Global mapping for the occurrence of all-sized microplastics in seafloor sediments, DOI: XXX. 2. Menemenlis, D., et al. (2005), NASA supercomputer improves prospects for ocean climate research, Eos Trans. AGU, 86(9), 89–96, doi:10.1029/2005EO090002. 3. Menemenlis, D., I. Fukumori, and T. Lee, 2005: Using Green's Functions to Calibrate an Ocean General Circulation Model. Mon. Wea. Rev., 133, 1224–1240, https://doi.org/10.1175/MWR2912.1. 4. Global Ocean Physics Analysis and Forecast, E.U. Copernicus Marine Service Information (CMEMS), Marine Data Store (MDS), DOI: 10.48670/moi-00016 (Accessed on 16 May 2023). 5. Global Ocean Waves Reanalysis, E.U. Copernicus Marine Service Information (CMEMS), Marine Data Store (MDS), DOI: 10.48670/moi-00022 (Accessed on 12 Jun 2023).