Assessing Emulator Design and Training for Modal Aerosol Microphysics Parameterizations in E3SMv2

2026-04-22

AI Summary

This study addresses the limitations in computational efficiency and accuracy of aerosol microphysical process simulations in global atmospheric models by developing a feedforward neural network–based surrogate model within the MAM4 framework of E3SMv2. Systematic evaluation demonstrates that, under appropriate variable normalization and training convergence criteria, a moderately complex neural network can accurately reproduce key features of sub-cloud aerosol concentration evolution. The work further elucidates the synergistic influence of network architecture, scaling strategies, and convergence behavior on surrogate model performance, offering generalizable design principles for efficient machine learning–based modeling of multiscale atmospheric physical processes.

Technology Category

• Machine Learning: Deep Neural Architectures and Foundation Models
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Multiagent Systems: Multiagent Learning

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems

Abstract

Toward the goal of using Scientific Machine Learning (SciML) emulators to improve the numerical representation of aerosol processes in global atmospheric models, we explore the emulation of aerosol microphysics processes under cloud-free conditions in the 4-mode Modal Aerosol Module (MAM4) within the Energy Exascale Earth System Model version 2 (E3SMv2). To develop an in-depth understanding of the challenges and opportunities in applying SciML to aerosol processes, we begin with a simple feedforward neural network architecture that has been used in earlier studies, but we systematically examine key emulator design choices, including architecture complexity and variable normalization, while closely monitoring training convergence behavior. Our results show that optimization convergence, scaling strategy, and network complexity strongly influence emulation accuracy. When effective scaling is applied and convergence is achieved, the relatively simple architecture, used together with a moderate network size, can reproduce key features of the microphysics-induced aerosol concentration changes with promising accuracy. These findings provide practical clues for the next stages of emulator development; they also provide general insights that are likely applicable to the emulation of other aerosol processes, as well as other atmospheric physics involving multi-scale variability.

Problem

• emulator
• aerosol microphysics
• Scientific Machine Learning
• parameterization
• E3SMv2

Innovation

• Scientific Machine Learning
• emulator design
• aerosol microphysics
• neural network architecture
• E3SMv2