Elements: Aerosol Model Benchmarking Repository and Standards (AMBRS): A Community Cyberinfrastructure for Process-Level Aerosol Model Evaluation

Aerosols are tiny particles suspended in air. They include particles from natural sources, such as dust, sea spray, and wildfire smoke, as well as particles produced by human activities, such as combustion, industry, and transportation. Aerosols affect air quality, human health, clouds, and climate, but current computer models often represent aerosol processes in different and hard-to-compare ways. These differences make it difficult to understand why models disagree and whether a newer or more complex model is actually more reliable. This project creates the Aerosol Model Benchmarking Repository and Standards, or AMBRS, an open community resource for testing aerosol models under the same well-defined conditions. The project helps researchers compare models more fairly, improves confidence in predictions of air quality and climate, and trains students and early-career scientists in reproducible computational science. The resulting public tools, benchmark cases, and training materials support education, open science, and the national interest by strengthening the scientific basis for decisions related to air quality, public health, and climate resilience. This project develops and hardens a reusable cyberinfrastructure platform for process-level aerosol model evaluation. The Aerosol Model Benchmarking Repository and Standards (AMBRS) combines two open-source components: ambuilder, a curated build environment for supported aerosol box models, and the AMBRS Python package which generates benchmark cases, executes model ensembles, and analyzes results. Starting with a prototype that supports two aerosol box models, the particle-resolved Particle Monte Carlo model and the four-mode Modal Aerosol Module, the project adds support for two sectional aerosol models, the TwO-Moment Aerosol Sectional model and the Community Aerosol and Radiation Model for Atmospheres. The project establishes standardized input and output formats that can be used to initialize all supported models and implements benchmark cases for selected stages of the aerosol life cycle, including condensation, coagulation, cloud activation, and aerosol aging. The workflow uses harmonized chemistry through the Chemistry Across Multiple Phases framework where possible, and documented chemistry alignment protocols otherwise. The project also produces versioned benchmark datasets with provenance records, persistent identifiers, reproducible diagnostic bundles, documentation, notebooks, workshops, and contribution pathways for adding new models and cases. These activities aim to transform aerosol model evaluation from an ad hoc practice into an open, reproducible, transparent, and community-governed cyberinfrastructure service. This award by the Office of Advanced Cyberinfrastructure is jointly supported by the Research, Innovation, Synergies and Education section in the Directorate of Geosciences. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria. NSF Award ID: 2608597 | Program: 01002627DB NSF RESEARCH & RELATED ACTIVIT | Principal Investigator: Nicole Riemer | Institution: University of Illinois at Urbana-Champaign, URBANA, IL | Award Amount: $593,169 View on NSF Award Search: https://www.nsf.gov/awardsearch/show-award/?AWD_ID=2608597 View on Research.gov: https://www.research.gov/awardapi-service/v1/awards/2608597.html