Copertina

Deep Learning Emulator Towards Both Forward and Adjoint Modes of Atmospheric Gas-Phase Chemical Process

Salvato in:
Pubblicato in:Atmosphere vol. 16, no. 9 (2025), p. 1109-1126
Autore principale: Liu, Yulong
Altri autori: Liao Meicheng, Liu, Jiacheng, Cheng, Zhen
Pubblicazione:
MDPI AG
Soggetti:
China
Air quality
Nitric oxide
Datasets
Sensitivity analysis
Nitrogen dioxide
Artificial neural networks
Neural networks
Chemistry
Computer applications
Photochemicals
Distance learning
Simulation
Ozone
Emulators
Numerical models
Isoprene
Graphics processing units
Chemical transport
Chemical bonds
Hydroxyl radicals
Central processing units > CPUs
Accuracy
Deep learning
Atmospheric chemistry
Data assimilation
Photochemistry
Machine learning
Graphics
Gases
Mathematical models
Air quality models
Ordinary differential equations
Embedding
Accesso online:Citation/Abstract
Full Text + Graphics
Full Text - PDF
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
Abstract:Gas-phase chemistry has been identified as a major computational bottleneck in both the forward and adjoint modes of chemical transport models (CTMs). Although previous studies have demonstrated the potential of deep learning models to simulate and accelerate this process, few studies have examined the applicability and performance of these models in adjoint sensitivity analysis. In this study, a deep learning emulator for gas-phase chemistry is developed and trained on a diverse set of forward-mode simulations from the Community Multiscale Air Quality (CMAQ) model. The emulator employs a residual neural network (ResNet) architecture referred to as FiLM-ResNet, which integrates Feature-wise Linear Modulation (FiLM) layers to explicitly account for photochemical and non-photochemical conditions. Validation within a single timestep indicates that the emulator accurately predicts concentration changes for 74% of gas-phase species with coefficient of determination (R2) exceeding 0.999. After embedding the emulator into the CTM, multi-timestep simulation over one week shows close agreement with the numerical model. For the adjoint mode, we compute the sensitivities of ozone (O3) with respect to O3, nitric oxide (NO), nitrogen dioxide (NO2), hydroxyl radical (OH) and isoprene (ISOP) using automatic differentiation, with the emulator-based adjoint results achieving a maximum R2 of 0.995 in single timestep evaluations compared to the numerical adjoint sensitivities. A 24 h adjoint simulation reveals that the emulator maintains spatially consistent adjoint sensitivity distributions compared to the numerical model across most grid cells. In terms of computational efficiency, the emulator achieves speed-ups of 80×–130× in the forward mode and 45×–102× in the adjoint mode, depending on whether inference is executed on Central Processing Unit (CPU) or Graphics Processing Unit (GPU). These findings demonstrate that, once the emulator is accurately trained to reproduce forward-mode gas-phase chemistry, it can be effectively applied in adjoint sensitivity analysis. This approach offers a promising alternative approach to numerical adjoint frameworks in CTMs.
ISSN:2073-4433
DOI:10.3390/atmos16091109
Fonte:Publicly Available Content Database