AAAR 37th Annual Conference October 14 - October 18, 2019 Oregon Convention Center Portland, Oregon, USA
Abstract View
A Scalable, Portable, Gas-Aerosol Chemistry Treatment for Atmospheric Models
MATTHEW DAWSON, Christian Guzman, Matthew West, Nicole Riemer, Mario Acosta, Oriol Jorba, Donald Dabdub, Barcelona Supercomputing Center
Abstract Number: 897 Working Group: Aerosol Chemistry
Abstract Decades of progress in the identification of increasingly complex atmospherically relevant mixed-phase physiochemical processes have resulted in an advanced understanding of the evolution of atmospheric systems, but have also introduced a level of complexity that few large-scale atmospheric models were originally designed to handle. Most regional and global models thus comprise a collection of physiochemical modules for the treatment of gas- and aerosol-phase chemistry and physics that have often been developed independently and with a focus on computational efficiency that leads to significant development efforts when modules are coupled for the first time, or new science is introduced. In addition, these modules are often tightly tied to the aerosol micro-physics scheme used by the host model, making the porting of chemical mechanisms to new models challenging.
A flexible treatment for gas- and aerosol-phase chemical processes has been developed for incorporation in models of diverse scale, from box models up to global models. A key feature of this novel framework is an abstracted aerosol representation that allows the same chemical mechanism to be solved on models with different aerosol representations (e.g., binned, modal, or particle-resolved). This is accomplished by treating aerosols as a collection of condensed phases that can be 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. In 3-D models, multiple grid cells can also be treated as a single chemical system and solved simultaneously, thus improving model performance. The flexibility of the model is achieved through a combination of JSON input files and run-time model configuration. JSON format is widely used for structured data, and allows entire gas- and aerosol-phase chemical mechanisms to be described in human-readable format, with as much complexity as is required to describe the system. Run-time model configuration allows changes to be made to any part of the chemical mechanism without recompiling the model. This new treatment has been developed using the PartMC modeling framework and deployed in the NMMB-MONARCHv2.0 chemical weather prediction system for use at global and regional scales. Results from the initial deployment to NMMB-MONARCHv2.0 will be discussed, along with extension to more complex gas–aerosol systems, and the use of GPU-based solvers.