Quantifying Black Carbon Mixing State Heterogeneity Using a Machine Learning Model
Open Access
- 9 September 2026
- journal article
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Atmospheres
- Vol. 131 (17)
Abstract
The climate impact of black carbon (BC) is strongly sensitive to the mass ratio of non‐BC coatings to BC (R BC ). However, current global climate models (GCMs) typically assume a uniform composition within individual BC‐containing particle populations, neglecting the significant particle‐to‐particle R BC heterogeneity. To address this, we train a machine learning (ML) emulator using high‐fidelity particle‐resolved model (PartMC‐MOSAIC) simulations, and then integrate the ML emulator with the Community Atmosphere Model version 6 (CAM6) to quantify global R BC distributions. By integrating k‐means clustering with multi‐metric cross‐validation on PartMC‐MOSAIC data, we identify three distinct patterns corresponding to progressive aging: long‐tailed (fresh BC), bimodal (transitional BC), and unimodal (aged BC), accurately fitted using Gamma, Bi‐Gaussian, and Gaussian functions, respectively. Our trained ML emulator predicts both the R BC pattern categories (81% overall accuracy) and their specific function parameters with coefficients of determination ( R 2 ) above 0.62 for all the parameters. We utilize offline CAM6 outputs to drive the ML emulator for global predictions, yielding R BC distributions highly consistent with multi‐site field observations ( R 2 > 0.74). Ultimately, global quantification reveals three distinct regimes with the following occurrence frequencies: the bimodal regime (16%) with the strongest heterogeneity (coefficient of variation, CV > 1), the unimodal distribution (70%) with the highest homogeneity (CV < 0.4), and the long‐tailed regime (14%, CV < 0.6). This study establishes a robust ML framework for BC mixing state heterogeneity, providing a critical tool for improving BC parameterizations and radiative forcing assessments within GCMs.Keywords
Funding Information
- National Natural Science Foundation of China (42505149)
- National Natural Science Foundation of China (U2342223)
- Frontiers Clinical and Translational Science Institute, University of Kansas (GeoX20260207)
- China Postdoctoral Science Foundation
This publication has 41 references indexed in Scilit:
- Learning to Simulate Aerosol Dynamics with Graph Neural NetworksACS ES&T Air, 2025
- Integrating Simulations and Observations: A Foundation Model for Estimating the Aerosol Mixing State IndexACS ES&T Air, 2025
- The Contribution of Black Carbon-Containing Particles to PM2.5: Variability, Drivers, and ImpactsEnvironmental Science & Technology, 2025
- Absorption Enhancement of Black Carbon Aerosols Constrained by Mixing-State HeterogeneityEnvironmental Science & Technology, 2022
- How China achieved its 11th Five-Year Plan emissions reduction target: A structural decomposition analysis of industrial SO2 and chemical oxygen demandScience of The Total Environment, 2017
- Rapid changes in biomass burning aerosols by atmospheric oxidationGeophysical Research Letters, 2014
- Bounding the role of black carbon in the climate system: A scientific assessmentJournal of Geophysical Research: Atmospheres, 2013
- Weighted Flow Algorithms (WFA) for stochastic particle coagulationJournal of Computational Physics, 2011
- Coupled Partitioning, Dilution, and Chemical Aging of Semivolatile OrganicsEnvironmental Science & Technology, 2006
- A new lumped structure photochemical mechanism for large‐scale applicationsJournal of Geophysical Research: Atmospheres, 1999