PNNL

Abstract

Secondary organic aerosols (SOA) are fine particles in the atmosphere and, interact with clouds, radiation and affect the Earth’s energy budget. SOA formation involves chemistry in gas phase, aqueous aerosols, and clouds. Simulating these chemical processes and the resulting changes in the particle size distribution and chemical composition involve solving a stiff set of differential equations, which are computationally expensive steps for three-dimensional chemical transport models. Deep neural networks (DNNs) are universal function approximators that could be used to represent the complex nonlinear changes in aerosol physical and chemical processes; however, key challenges such as generalizability to extended time periods, preservation of mass balance in the gas and particle phase, simulating sparse model output, and maintaining physical constraints have limited their use in atmospheric chemistry. Here, we develop an approach of using a physics-informed DNN that overcomes many of the previous challenges and demonstrates its applicability to simulate the complex physical and chemical formation processes of isoprene epoxydiol SOA (IEPOX-SOA) over the Amazon rainforest. The DNN is trained over just 12 hours of simulated IEPOX-SOA produced using the Weather Research and Forecasting coupled with Chemistry (WRF-Chem). The trained DNN is then embedded within WRF-Chem to replace the computationally expensive default solver of IEPOX-SOA formation. The trained DNN generalizes well and agrees well with the default model simulation of the IEPOX-SOA formation and its size distribution over all 20 size bins over 6 days of simulations. The DNN also reduces the computational expense of WRF-Chem by a factor of 2. Our approach shows promise in terms of application to other computationally expensive chemistry solvers in climate models that could greatly speed up the models while maintaining complexity.