Microplastics in farmlands: A model of conditional fragmentation and source estimation

<jats:p>Microplastic (MP) pollution in agricultural systems has emerged as a critical environmental concern, yet quantitative models capturing the complex dynamics of MP fragmentation and accumulation remain underdeveloped. Here we present a cross-scale modeling framework that solves fragmentation mechanisms and input sources at the same time, and simulates the temporal evolution of MP size distributions in agricultural soils. The model captures the size-dependency of MP fragmentation rates and estimates the input intensity of different pollution sources. We synthesized literature data from 98 studies, covering 438 agricultural sites and 3235 datapoints across 16 countries. Parameter optimization using Huber loss (robust 𝑅2 = 0.412) showed higher performance than regression-based approaches, and showed that fragmentation rates are dependent on particle sizes, and influenced by climate types and cropping cycles. In general, MP fragmentation rates decrease as they break into smaller sizes. Comparing different site conditions, large MP particles (&amp;gt; 40 μm) experience faster fragmentation under high-UV climates due to more intense photoaging, which is further accelerated by mechanical disturbance depending on cropping cycles. On the other hand, small MP particles (&amp;lt; 40 μm) break down faster under low-UV climates, likely due to biological aging from soil microbes and hydrological cyclic micro-stress. Of the long-term MP input sources including mulch films, plastic greenhouses, organic fertilizers, wastewater irrigation, atmospheric deposition, and unmanaged littering, plastic mulch films were the primary contributor to the mass of soil MPs (around 5.8 mg/kg after 5 years) across all climate types and cropping cycles. If measured by particle abundance, organic fertilization introduces the highest MP abundance in soil at the first 5 years (1.1×103 items/kg) as they contain numerous small-sized MPs. A case study on China’s mulching history showed strong regional heterogeneity in MP accumulation, with legacy plastics continuing to generate secondary MPs even under zero-new-input scenarios. This work provides a theoretically grounded and empirically calibrated tool for assessing MP pollution dynamics and management strategies in agricultural systems from local to global scale.</jats:p>