Chemistry Across Multiple Phases (CAMP) version 1.0: An integrated multi-phase chemistry model

Matthew L. Dawson, Christian D. Guzmán, Jeffrey H. Curtis, Mario Acosta, Shupeng Zhu, Donald Dabdub, Andrew Conley, Matthew West, Nicole Riemer, Oriol Jorba

TL;DR

This paper introduces CAMP (Chemistry Across Multiple Phases), a novel, flexible framework for simulating complex atmospheric chemistry across gas and aerosol phases. CAMP decouples chemical mechanisms from aerosol representations, allowing for runtime configuration and easier integration into diverse atmospheric models, from simple box models to complex global systems.

What Problem Does This Paper Address?

Atmospheric chemistry models often struggle to keep pace with the increasing complexity of our understanding of atmospheric processes, particularly the intricate interactions between gas-phase and aerosol-phase chemistry. Traditionally, atmospheric models are built with specialized modules for different processes (gas chemistry, aerosol microphysics, etc.), which are often tightly coupled and difficult to modify or extend. This compartmentalization can lead to inaccuracies, especially when processes are closely linked, and makes it challenging to incorporate new chemical mechanisms or adapt models to different aerosol representations (like sectional, modal, or particle-resolved). The authors aim to solve this by creating a unified, flexible, and easily configurable framework that can handle multi-phase chemistry across various model scales and aerosol types without requiring extensive code modifications or recompilation.

Methods

The core of CAMP is an "abstracted aerosol representation" and an object-oriented design. This allows CAMP to treat aerosols as collections of condensed phases, independent of the host model's specific aerosol representation. Key features include:

  • Object-Oriented Design: CAMP uses generalized base classes (Process, Parameter, AerosolRepresentation) that allow for extensibility and abstraction. This means new chemical processes or aerosol types can be added without altering the core CAMP code.
  • Runtime Configuration (JSON): Chemical mechanisms are defined using JSON files, enabling users to modify chemistry without recompiling the model. This facilitates rapid testing and sensitivity analyses.
  • Decoupled Solver Interface: CAMP is designed to interface with various external ODE solvers, including those optimized for GPUs, separating the chemical system specification from the numerical solution.
  • Comprehensive Testing: Automated unit and integration tests are employed to ensure code stability as new features are added.

Datasets

Not discussed in the paper. The paper focuses on the framework's design and validation through simulations rather than training on specific datasets.

Key Insights and Ideas

  • Abstraction is Key: The paper highlights how abstracting the aerosol representation from the chemical mechanism is crucial for creating a truly flexible and portable multi-phase chemistry model.
  • Runtime Configurability Empowers Users: The use of JSON for defining chemical mechanisms significantly lowers the barrier to entry for modelers, experimentalists, and educators, allowing them to easily customize and test new chemistry without deep coding knowledge.
  • Unified System for Multi-phase Chemistry: CAMP treats gas and aerosol chemistry as a single, integrated system, which can lead to more accurate simulations, especially for processes like gas-aerosol partitioning.
  • Scalability Across Models: The framework's design allows it to be seamlessly integrated into models ranging from simple box models to complex regional and global atmospheric models, demonstrating its broad applicability.

Key Results

The paper demonstrates CAMP's functionality through box model simulations and a 3D Eulerian model (MONARCH).

  • Box Model Consistency: Three different aerosol representations (modes, bins, and particle-resolved) within CAMP yielded nearly identical results for gas-phase species (like ozone) when using the same chemistry mechanism. Differences observed in secondary organic aerosol (SOA) formation were attributed to the aerosol representation itself, not the chemistry.
  • MONARCH Integration: CAMP was successfully integrated into the MONARCH model, showing consistent results with box model simulations and reasonable agreement with observational data for ozone concentrations over Europe.
  • Performance: While not the primary focus of this initial phase, the paper mentions plans for GPU optimization, with preliminary results showing a significant speedup compared to CPU-based MPI simulations.

Limitations and Future Work

  • Phase 1 Focus: This paper presents the first phase of a three-phase development plan, focusing on a proof-of-concept library.
  • Optimization: While the framework is flexible, further optimization, particularly for large-scale models and GPU acceleration (Phase 2), is ongoing and crucial for computational efficiency.
  • Extensibility: Future work will involve extending CAMP to include new physicochemical processes, such as aerosol surface reactions and novel gas/condensed-phase reactions.
  • Refactoring: Phase 3 will focus on refactoring the code based on lessons learned to improve portability to various solvers and architectures.
  • Specific Aerosol Representations: While CAMP supports different aerosol representations, the paper notes that the specific implementation of aerosol growth or coagulation might differ between these representations, leading to some simulation differences.

Summary updated

Abstract

A flexible treatment for gas- and aerosol-phase chemical processes has been developed for models of diverse scale, from box models up to global models. At the core of this novel framework is an "abstracted aerosol representation" that allows a given chemical mechanism to be solved in atmospheric models with different aerosol representations (e.g., sectional, modal, or particle-resolved). This is accomplished by treating aerosols as a collection of condensed phases that are implemented according to the aerosol representation of the host model. The framework also allows multiple chemical processes (e.g., gas- and aerosol-phase chemical reactions, emissions, deposition, photolysis, and mass-transfer) to be solved simultaneously as a single system. The flexibility of the model is achieved by (1) using an object-oriented design that facilitates extensibility to new types of chemical processes and to new ways of representing aerosol systems; (2) runtime model configuration using JSON input files that permits making changes to any part of the chemical mechanism without recompiling the model; this widely used, human-readable format allows entire gas- and aerosol-phase chemical mechanisms to be described with as much complexity as necessary; and (3) automated comprehensive testing that ensures stability of the code as new functionality is introduced. Together, these design choices enable users to build a customized multiphase mechanism, without having to handle pre-processors, solvers or compilers. Removing these hurdles makes this type of modeling accessible to a much wider community, including modelers, experimentalists, and educators. This new treatment compiles as a stand-alone library and has been deployed in the particle-resolved PartMC model and in the MONARCH chemical weather prediction system for use at regional and global scales. Results from the initial deployment to box models of different complexity and MONARCH will be discussed, along with future extension to more complex gas--aerosol systems, and the integration of GPU-based solvers.
BibTeX
@misc{dawson2021chemistry,
  author = {Matthew L. Dawson and Christian D. Guzmán and Jeffrey H. Curtis and Mario Acosta and Shupeng Zhu and Donald Dabdub and Andrew Conley and Matthew West and Nicole Riemer and Oriol Jorba},
  title = {{Chemistry Across Multiple Phases (CAMP) version 1.0: An integrated multi-phase chemistry model}},
  year = {2021},
  eprint = {2111.07436},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2111.07436},
}
Figure