부산대학교 도서관

A major computational barrier in global modeling of atmospheric chemistry is the numerical integration of the coupled kinetic equations describing the chemical mechanism. Machine-learned (ML) solvers can offer order of magnitude speedup relative to conventional implicit solvers but past implementations have suffered from fast error growth and only run for short simulation times (<1 month). A successful ML solver for global models must avoid error growth over yearlong simulations and allow for reinitialization of the chemical trajectory by transport at every time step. Here, we explore the capability of a neural network solver equipped with an autoencoder to achieve stable full-year simulations of tropospheric oxidant chemistry in the global 3-D Goddard Earth Observing System (GEOS)-Chem model, replacing its standard mechanism (228 species) by the Super-Fast mechanism (12 species) to avoid the curse of dimensionality. We find that online training of the ML solver within GEOS-Chem is important for accuracy, whereas offline training from archived GEOS-Chem inputs/outputs produces large errors. After online training, we achieve stable 1-year simulations with five-fold speedup compared to the standard implicit Rosenbrock solver with global tropospheric normalized mean biases of -0.3% for ozone, 1% for hydrogen oxide radicals, and -5% for nitrogen oxides. The ML solver captures the diurnal and synoptic variability of surface ozone at polluted and clean sites. There are however large regional biases for ozone and NOx under remote conditions where chemical aging leads to error accumulation. These regional biases remain a major limitation for practical application, and ML emulation would be more difficult in a more complex mechanism. Plain Language Summary Global models of atmospheric chemistry are computationally expensive. A bottleneck is the chemical solver that integrates the large-dimensional coupled systems of kinetic equations describing the chemical mechanism. Machine learning (ML) could be transformative for reducing the cost of an atmospheric chemistry simulation by replacing the chemical solver with a faster emulator. However, past work found that ML chemical solvers experience rapid error growth and become unstable over time. Here, we present results achieving for the first time a stable full-year global simulation of atmospheric chemistry with 3 months seasonal ML solvers and with five-fold speedup in computational performance over the reference simulation. We show that online training of the ML solver synchronously with the atmospheric chemistry model simulation produces considerably more stable results than offline training from a static data set of simulation results. Although our work represents an important step for using ML solvers in global atmospheric chemistry models, more work is needed to extend it to large chemical mechanisms and to reduce errors during long-term chemical aging. [ABSTRACT FROM AUTHOR]