Machine Learning Aided Global Meta‐Analysis of Microplastic Polymer Composition in Agriculture Soil, Urban Soil, and Wetland Sediments

Abstract

Microplastic (MP) pollution is a global environmental problem that harms aquatic and terrestrial ecosystems. The present study performs a critical meta‐analysis of available literature on the polymer composition of MP polymers in soils in agricultural lands, urban soil, and wetland sediments globally. The study investigated the pattern of MP polymers in multiple environments using a systematic literature review and data mining methods, such as the Apriori algorithm and Random Forest analysis. Polypropylene (PP) and polyethylene (PE) are the most widespread polymers within all known compartments, with each region possessing a distinct polymer profile due to human activities. The study highlights the importance of wetlands in MP assemblages as sinks and transitional zones where polymer degradation and land‐to‐water transport evidence exists. The study also recommends policies to reduce MP pollution in terrestrial ecosystems. These results highlight the ecological threats the prevailing polymers pose and confirm the necessity of implementing monitoring protocols and policy measures to reduce environmental MP contamination.