Articles | Volume 15, issue 22
https://doi.org/10.5194/gmd-15-8439-2022
© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/gmd-15-8439-2022
© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Model description paper
| 
21 Nov 2022
Model description paper |
| 21 Nov 2022
| 21 Nov 2022
Development of an LSTM broadcasting deep-learning framework for regional air pollution forecast improvement
Haochen Sun
Department of Mathematics, Hong Kong University of Science and
Technology, Clear Water Bay, Hong Kong SAR, China
Department of Computer Science and Engineering, Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong SAR, China
Jimmy C. H. Fung
Department of Mathematics, Hong Kong University of Science and
Technology, Clear Water Bay, Hong Kong SAR, China
Division of Environment and Sustainability, Hong Kong University of
Science and Technology, Clear Water Bay, Hong Kong SAR, China
Atmospheric Research Center, HKUST Fok Ying Tung Research Institute, Guangzhou, China
Yiang Chen
Division of Environment and Sustainability, Hong Kong University of
Science and Technology, Clear Water Bay, Hong Kong SAR, China
Zhenning Li
Division of Environment and Sustainability, Hong Kong University of
Science and Technology, Clear Water Bay, Hong Kong SAR, China
Dehao Yuan
Department of Computer Science, University of Maryland, College Park, Maryland, USA
Wanying Chen
Division of Environment and Sustainability, Hong Kong University of
Science and Technology, Clear Water Bay, Hong Kong SAR, China
Department of Geography and Resource Management, Chinese University of Hong Kong, Sha Tin, Hong Kong SAR, China
Related authors
No articles found.
Wanliang Zhang, Chao Ren, Edward Yan Yung Ng, Michael Mau Fung Wong, and Jimmy Chi Hung Fung
Geosci. Model Dev. Discuss., https://doi.org/10.5194/gmd-2024-205, https://doi.org/10.5194/gmd-2024-205, 2024
Revised manuscript under review for GMD
Short summary
Short summary
This study focuses on improving the accuracy of numerical weather prediction (NWP) model particularly in urbanized areas. We coupled a recently validated boundary layer model with a building effect model within an NWP. Validation has been performed under idealized atmospheric conditions by benchmarking the coupled model with a fine-scale numerical model. Subsequently, the improvements and limitations are investigated aided by observations in real case simulations.
Naveed Ahmad, Changqing Lin, Alexis K. H. Lau, Jhoon Kim, Tianshu Zhang, Fangqun Yu, Chengcai Li, Ying Li, Jimmy C. H. Fung, and Xiang Qian Lao
Atmos. Chem. Phys., 24, 9645–9665, https://doi.org/10.5194/acp-24-9645-2024, https://doi.org/10.5194/acp-24-9645-2024, 2024
Short summary
Short summary
This study developed a nested machine learning model to convert the GEMS NO2 column measurements into ground-level concentrations across China. The model directly incorporates the NO2 mixing height (NMH) into the methodological framework. The study underscores the importance of considering NMH when estimating ground-level NO2 from satellite column measurements and highlights the significant advantages of new-generation geostationary satellites in air quality monitoring.
Yiang Chen, Xingcheng Lu, and Jimmy C. H. Fung
Atmos. Chem. Phys., 24, 8847–8864, https://doi.org/10.5194/acp-24-8847-2024, https://doi.org/10.5194/acp-24-8847-2024, 2024
Short summary
Short summary
This study investigates the contribution of pollutants from different emitting periods to ozone episodes over the Greater Bay Area. The analysis reveals the variation in major spatiotemporal contributors to the O3 pollution under the influence of typhoons and subtropical high pressure. Through temporal contribution analysis, our work offers a new perspective on the evolution of O3 pollution and can aid in developing effective and timely control policies under unfavorable weather conditions.
Yun Fat Lam, Chi Chiu Cheung, Xuguo Zhang, Joshua S. Fu, and Jimmy Chi Hung Fung
Atmos. Chem. Phys., 21, 12895–12908, https://doi.org/10.5194/acp-21-12895-2021, https://doi.org/10.5194/acp-21-12895-2021, 2021
Short summary
Short summary
In recent years, air pollution forecasting has become an important municipal service of the government. In this study, a new spatial allocation method based on satellite remote sensing and GIS techniques was developed to address the spatial deficiency of industrial source emissions in China, providing a substantial improvement on NO2 and PM2.5 forecast for the Pearl River Delta/Greater Bay Area.
Lin Su and Jimmy C. H. Fung
Atmos. Chem. Phys., 18, 11529–11545, https://doi.org/10.5194/acp-18-11529-2018, https://doi.org/10.5194/acp-18-11529-2018, 2018
Short summary
Short summary
By applying an updated version of WRF-Chem, we aim to investigate the full effects of dust, including direct, semi-direct, and indirect effects, on the regional weather system over East Asia during a dust-intensive period. This is the first study to document the full effects of dust during a typical dust-intensive period over East Asia by applying an online coupled regional numerical model.
Lin Su and Jimmy C. H. Fung
Atmos. Chem. Phys., 18, 8707–8725, https://doi.org/10.5194/acp-18-8707-2018, https://doi.org/10.5194/acp-18-8707-2018, 2018
Short summary
Short summary
A new treatment was implemented into a microphysics scheme in WRF-Chem to quantify and evaluate the effect of dust on the ice nucleation process. The simulation of atmospheric IWC during a typical dust-intensive period over East Asia was substantially improved with the effect of dust considered in the model. In addition, sensitivity experiments were carried out to optimize the parameters used in the implemented treatment.
Related subject area
Atmospheric sciences
The MESSy DWARF (based on MESSy v2.55.2)
An enhanced emission module for the PALM model system 23.10 with application for PM10 emission from urban domestic heating
Identifying lightning processes in ERA5 soundings with deep learning
Sensitivity of predicted ultrafine particle size distributions in Europe to different nucleation rate parameterizations using PMCAMx-UF v2.2
Explaining neural networks for detection of tropical cyclones and atmospheric rivers in gridded atmospheric simulation data
Accurate space-based NOx emission estimates with the flux divergence approach require fine-scale model information on local oxidation chemistry and profile shapes
Exploring a high-level programming model for the NWP domain using ECMWF microphysics schemes
Quantifying uncertainties in satellite NO2 superobservations for data assimilation and model evaluation
ML-AMPSIT: Machine Learning-based Automated Multi-method Parameter Sensitivity and Importance analysis Tool
Coupling the urban canopy model TEB (SURFEXv9.0) with the radiation model SPARTACUS-Urbanv0.6.1 for more realistic urban radiative exchange calculation
Forecasting contrail climate forcing for flight planning and air traffic management applications: the CocipGrid model in pycontrails 0.51.0
Simulation of the heat mitigation potential of unsealing measures in cities by parameterizing grass grid pavers for urban microclimate modelling with ENVI-met (V5)
AI-NAOS: an AI-based nonspherical aerosol optical scheme for the chemical weather model GRAPES_Meso5.1/CUACE
Orbital-Radar v1.0.0: a tool to transform suborbital radar observations to synthetic EarthCARE cloud radar data
The Modular and Integrated Data Assimilation System at Environment and Climate Change Canada (MIDAS v3.9.1)
Modeling of polycyclic aromatic hydrocarbons (PAHs) from global to regional scales: model development (IAP-AACM_PAH v1.0) and investigation of health risks in 2013 and 2018 in China
LIMA (v2.0): A full two-moment cloud microphysical scheme for the mesoscale non-hydrostatic model Meso-NH v5-6
SLUCM+BEM (v1.0): a simple parameterisation for dynamic anthropogenic heat and electricity consumption in WRF-Urban (v4.3.2)
NAQPMS-PDAF v2.0: a novel hybrid nonlinear data assimilation system for improved simulation of PM2.5 chemical components
Source-specific bias correction of US background and anthropogenic ozone modeled in CMAQ
Observational operator for fair model evaluation with ground NO2 measurements
Valid time shifting ensemble Kalman filter (VTS-EnKF) for dust storm forecasting
An updated parameterization of the unstable atmospheric surface layer in the Weather Research and Forecasting (WRF) modeling system
The impact of cloud microphysics and ice nucleation on Southern Ocean clouds assessed with single-column modeling and instrument simulators
An updated aerosol simulation in the Community Earth System Model (v2.1.3): dust and marine aerosol emissions and secondary organic aerosol formation
Exploring ship track spreading rates with a physics-informed Langevin particle parameterization
Do data-driven models beat numerical models in forecasting weather extremes? A comparison of IFS HRES, Pangu-Weather, and GraphCast
Development of the MPAS-CMAQ coupled system (V1.0) for multiscale global air quality modeling
Assessment of object-based indices to identify convective organization
The Global Forest Fire Emissions Prediction System version 1.0
Sensitivity Studies of Four‐Dimensional Local Ensemble Transform Kalman Filter Coupled With WRF-Chem Version 3.9.1 for Improving Particulate Matter Simulation Accuracy
Development of A Fast Radiative Transfer Model for Ground-based Microwave Radiometers (ARMS-gb v1.0): Validation and Comparison to RTTOV-gb
NEIVAv1.0: Next-generation Emissions InVentory expansion of Akagi et al. (2011) version 1.0
FLEXPART version 11: improved accuracy, efficiency, and flexibility
Low-level jets in the North and Baltic Seas: Mesoscale Model Sensitivity and Climatology
Challenges of high-fidelity air quality modeling in urban environments – PALM sensitivity study during stable conditions
Knowledge-inspired fusion strategies for the inference of PM2.5 values with a Neural Network
Air quality modeling intercomparison and multiscale ensemble chain for Latin America
Recommended coupling to global meteorological fields for long-term tracer simulations with WRF-GHG
Estimation of aerosol and cloud radiative heating rate in tropical stratosphere using radiative kernel method
Selecting CMIP6 global climate models (GCMs) for Coordinated Regional Climate Downscaling Experiment (CORDEX) dynamical downscaling over Southeast Asia using a standardised benchmarking framework
Improved definition of prior uncertainties in CO2 and CO fossil fuel fluxes and its impact on multi-species inversion with GEOS-Chem (v12.5)
RASCAL v1.0: an open-source tool for climatological time series reconstruction and extension
Porting the Meso-NH atmospheric model on different GPU architectures for the next generation of supercomputers (version MESONH-v55-OpenACC)
Accounting for effects of coagulation and model uncertainties in particle number concentration estimates based on measurements from sampling lines – A Bayesian inversion approach with SLIC v1.0
Introducing graupel density prediction in Weather Research and Forecasting (WRF) double-moment 6-class (WDM6) microphysics and evaluation of the modified scheme during the ICE-POP field campaign
Enabling high-performance cloud computing for the Community Multiscale Air Quality Model (CMAQ) version 5.3.3: performance evaluation and benefits for the user community
Atmospheric-river-induced precipitation in California as simulated by the regionally refined Simple Convective Resolving E3SM Atmosphere Model (SCREAM) Version 0
Recent improvements and maximum covariance analysis of aerosol and cloud properties in the EC-Earth3-AerChem model
GPU-HADVPPM4HIP V1.0: using the heterogeneous-compute interface for portability (HIP) to speed up the piecewise parabolic method in the CAMx (v6.10) air quality model on China's domestic GPU-like accelerator
Astrid Kerkweg, Timo Kirfel, Duong H. Do, Sabine Griessbach, Patrick Jöckel, and Domenico Taraborrelli
Geosci. Model Dev., 18, 1265–1286, https://doi.org/10.5194/gmd-18-1265-2025, https://doi.org/10.5194/gmd-18-1265-2025, 2025
Short summary
Short summary
Normally, the Modular Earth Submodel System (MESSy) is linked to complete dynamic models to create chemical climate models. However, the modular concept of MESSy and the newly developed DWARF component presented here make it possible to create simplified models that contain only one or a few process descriptions. This is very useful for technical optimisation, such as porting to GPUs, and can be used to create less complex models, such as a chemical box model.
Edward C. Chan, Ilona J. Jäkel, Basit Khan, Martijn Schaap, Timothy M. Butler, Renate Forkel, and Sabine Banzhaf
Geosci. Model Dev., 18, 1119–1139, https://doi.org/10.5194/gmd-18-1119-2025, https://doi.org/10.5194/gmd-18-1119-2025, 2025
Short summary
Short summary
An enhanced emission module has been developed for the PALM model system, improving flexibility and scalability of emission source representation across different sectors. A model for parametrized domestic emissions has also been included, for which an idealized model run is conducted for particulate matter (PM10). The results show that, in addition to individual sources and diurnal variations in energy consumption, vertical transport and urban topology play a role in concentration distribution.
Gregor Ehrensperger, Thorsten Simon, Georg J. Mayr, and Tobias Hell
Geosci. Model Dev., 18, 1141–1153, https://doi.org/10.5194/gmd-18-1141-2025, https://doi.org/10.5194/gmd-18-1141-2025, 2025
Short summary
Short summary
As lightning is a brief and localized event, it is not explicitly resolved in atmospheric models. Instead, expert-based auxiliary descriptions are used to assess it. This study explores how AI can improve our understanding of lightning without relying on traditional expert knowledge. We reveal that AI independently identified the key factors known to experts as essential for lightning in the Alps region. This shows how knowledge discovery could be sped up in areas with limited expert knowledge.
David Patoulias, Kalliopi Florou, and Spyros N. Pandis
Geosci. Model Dev., 18, 1103–1118, https://doi.org/10.5194/gmd-18-1103-2025, https://doi.org/10.5194/gmd-18-1103-2025, 2025
Short summary
Short summary
The effect of the assumed atmospheric nucleation mechanism on particle number concentrations and size distribution was investigated. Two quite different mechanisms involving sulfuric acid and ammonia or a biogenic organic vapor gave quite similar results which were consistent with measurements at 26 measurement stations across Europe. The number of larger particles that serve as cloud condensation nuclei showed little sensitivity to the assumed nucleation mechanism.
Tim Radke, Susanne Fuchs, Christian Wilms, Iuliia Polkova, and Marc Rautenhaus
Geosci. Model Dev., 18, 1017–1039, https://doi.org/10.5194/gmd-18-1017-2025, https://doi.org/10.5194/gmd-18-1017-2025, 2025
Short summary
Short summary
In our study, we built upon previous work to investigate the patterns artificial intelligence (AI) learns to detect atmospheric features like tropical cyclones (TCs) and atmospheric rivers (ARs). As primary objective, we adopt a method to explain the AI used and investigate the plausibility of learned patterns. We find that plausible patterns are learned for both TCs and ARs. Hence, the chosen method is very useful for gaining confidence in the AI-based detection of atmospheric features.
Felipe Cifuentes, Henk Eskes, Enrico Dammers, Charlotte Bryan, and Folkert Boersma
Geosci. Model Dev., 18, 621–649, https://doi.org/10.5194/gmd-18-621-2025, https://doi.org/10.5194/gmd-18-621-2025, 2025
Short summary
Short summary
We tested the capability of the flux divergence approach (FDA) to reproduce known NOx emissions using synthetic NO2 satellite column retrievals from high-resolution model simulations. The FDA accurately reproduced NOx emissions when column observations were limited to the boundary layer and when the variability of the NO2 lifetime, the NOx : NO2 ratio, and NO2 profile shapes were correctly modeled. This introduces strong model dependency, reducing the simplicity of the original FDA formulation.
Stefano Ubbiali, Christian Kühnlein, Christoph Schär, Linda Schlemmer, Thomas C. Schulthess, Michael Staneker, and Heini Wernli
Geosci. Model Dev., 18, 529–546, https://doi.org/10.5194/gmd-18-529-2025, https://doi.org/10.5194/gmd-18-529-2025, 2025
Short summary
Short summary
We explore a high-level programming model for porting numerical weather prediction (NWP) model codes to graphics processing units (GPUs). We present a Python rewrite with the domain-specific library GT4Py (GridTools for Python) of two renowned cloud microphysics schemes and the associated tangent-linear and adjoint algorithms. We find excellent portability, competitive GPU performance, robust execution on diverse computing architectures, and enhanced code maintainability and user productivity.
Pieter Rijsdijk, Henk Eskes, Arlene Dingemans, K. Folkert Boersma, Takashi Sekiya, Kazuyuki Miyazaki, and Sander Houweling
Geosci. Model Dev., 18, 483–509, https://doi.org/10.5194/gmd-18-483-2025, https://doi.org/10.5194/gmd-18-483-2025, 2025
Short summary
Short summary
Clustering high-resolution satellite observations into superobservations improves model validation and data assimilation applications. In our paper, we derive quantitative uncertainties for satellite NO2 column observations based on knowledge of the retrievals, including a detailed analysis of spatial error correlations and representativity errors. The superobservations and uncertainty estimates are tested in a global chemical data assimilation system and are found to improve the forecasts.
Dario Di Santo, Cenlin He, Fei Chen, and Lorenzo Giovannini
Geosci. Model Dev., 18, 433–459, https://doi.org/10.5194/gmd-18-433-2025, https://doi.org/10.5194/gmd-18-433-2025, 2025
Short summary
Short summary
This paper presents the Machine Learning-based Automated Multi-method Parameter Sensitivity and Importance analysis Tool (ML-AMPSIT), a computationally efficient tool that uses machine learning algorithms for sensitivity analysis in atmospheric models. It is tested with the Weather Research and Forecasting (WRF) model coupled with the Noah-Multiparameterization (Noah-MP) land surface model to investigate sea breeze circulation sensitivity to vegetation-related parameters.
Robert Schoetter, Robin James Hogan, Cyril Caliot, and Valéry Masson
Geosci. Model Dev., 18, 405–431, https://doi.org/10.5194/gmd-18-405-2025, https://doi.org/10.5194/gmd-18-405-2025, 2025
Short summary
Short summary
Radiation is relevant to the atmospheric impact on people and infrastructure in cities as it can influence the urban heat island, building energy consumption, and human thermal comfort. A new urban radiation model, assuming a more realistic form of urban morphology, is coupled to the urban climate model Town Energy Balance (TEB). The new TEB is evaluated with a reference radiation model for a variety of urban morphologies, and an improvement in the simulated radiative observables is found.
Zebediah Engberg, Roger Teoh, Tristan Abbott, Thomas Dean, Marc E. J. Stettler, and Marc L. Shapiro
Geosci. Model Dev., 18, 253–286, https://doi.org/10.5194/gmd-18-253-2025, https://doi.org/10.5194/gmd-18-253-2025, 2025
Short summary
Short summary
Contrails forming in some atmospheric conditions may persist and become strongly warming cirrus, while in other conditions may be neutral or cooling. We develop a contrail forecast model to predict contrail climate forcing for any arbitrary point in space and time and explore integration into flight planning and air traffic management. This approach enables contrail interventions to target high-probability high-climate-impact regions and reduce unintended consequences of contrail management.
Nils Eingrüber, Alina Domm, Wolfgang Korres, and Karl Schneider
Geosci. Model Dev., 18, 141–160, https://doi.org/10.5194/gmd-18-141-2025, https://doi.org/10.5194/gmd-18-141-2025, 2025
Short summary
Short summary
Climate change adaptation measures like unsealings can reduce urban heat stress. As grass grid pavers have never been parameterized for microclimate model simulations with ENVI-met, a new parameterization was developed based on field measurements. To analyse the cooling potential, scenario analyses were performed for a densely developed area in Cologne. Statistically significant average cooling effects of up to −11.1 K were found for surface temperature and up to −2.9 K for 1 m air temperature.
Xuan Wang, Lei Bi, Hong Wang, Yaqiang Wang, Wei Han, Xueshun Shen, and Xiaoye Zhang
Geosci. Model Dev., 18, 117–139, https://doi.org/10.5194/gmd-18-117-2025, https://doi.org/10.5194/gmd-18-117-2025, 2025
Short summary
Short summary
The Artificial-Intelligence-based Nonspherical Aerosol Optical Scheme (AI-NAOS) was developed to improve the estimation of the aerosol direct radiation effect and was coupled online with a chemical weather model. The AI-NAOS scheme considers black carbon as fractal aggregates and soil dust as super-spheroids, encapsulated with hygroscopic aerosols. Real-case simulations emphasize the necessity of accurately representing nonspherical and inhomogeneous aerosols in chemical weather models.
Lukas Pfitzenmaier, Pavlos Kollias, Nils Risse, Imke Schirmacher, Bernat Puigdomenech Treserras, and Katia Lamer
Geosci. Model Dev., 18, 101–115, https://doi.org/10.5194/gmd-18-101-2025, https://doi.org/10.5194/gmd-18-101-2025, 2025
Short summary
Short summary
The Python tool Orbital-Radar transfers suborbital radar data (ground-based, airborne, and forward-simulated numerical weather prediction model) into synthetic spaceborne cloud profiling radar data, mimicking platform-specific instrument characteristics, e.g. EarthCARE or CloudSat. The tool's novelty lies in simulating characteristic errors and instrument noise. Thus, existing data sets are transferred into synthetic observations and can be used for satellite calibration–validation studies.
Mark Buehner, Jean-Francois Caron, Ervig Lapalme, Alain Caya, Ping Du, Yves Rochon, Sergey Skachko, Maziar Bani Shahabadi, Sylvain Heilliette, Martin Deshaies-Jacques, Weiguang Chang, and Michael Sitwell
Geosci. Model Dev., 18, 1–18, https://doi.org/10.5194/gmd-18-1-2025, https://doi.org/10.5194/gmd-18-1-2025, 2025
Short summary
Short summary
The Modular and Integrated Data Assimilation System (MIDAS) software is described. The flexible design of MIDAS enables both deterministic and ensemble prediction applications for the atmosphere and several other Earth system components. It is currently used for all main operational weather prediction systems in Canada and also for sea ice and sea surface temperature analysis. The use of MIDAS for multiple Earth system components will facilitate future research on coupled data assimilation.
Zichen Wu, Xueshun Chen, Zifa Wang, Huansheng Chen, Zhe Wang, Qing Mu, Lin Wu, Wending Wang, Xiao Tang, Jie Li, Ying Li, Qizhong Wu, Yang Wang, Zhiyin Zou, and Zijian Jiang
Geosci. Model Dev., 17, 8885–8907, https://doi.org/10.5194/gmd-17-8885-2024, https://doi.org/10.5194/gmd-17-8885-2024, 2024
Short summary
Short summary
We developed a model to simulate polycyclic aromatic hydrocarbons (PAHs) from global to regional scales. The model can reproduce PAH distribution well. The concentration of BaP (indicator species for PAHs) could exceed the target values of 1 ng m-3 over some areas (e.g., in central Europe, India, and eastern China). The change in BaP is lower than that in PM2.5 from 2013 to 2018. China still faces significant potential health risks posed by BaP although the Action Plan has been implemented.
Marie Taufour, Jean-Pierre Pinty, Christelle Barthe, Benoît Vié, and Chien Wang
Geosci. Model Dev., 17, 8773–8798, https://doi.org/10.5194/gmd-17-8773-2024, https://doi.org/10.5194/gmd-17-8773-2024, 2024
Short summary
Short summary
We have developed a complete two-moment version of the LIMA (Liquid Ice Multiple Aerosols) microphysics scheme. We have focused on collection processes, where the hydrometeor number transfer is often estimated in proportion to the mass transfer. The impact of these parameterizations on a convective system and the prospects for more realistic estimates of secondary parameters (reflectivity, hydrometeor size) are shown in a first test on an idealized case.
Yuya Takane, Yukihiro Kikegawa, Ko Nakajima, and Hiroyuki Kusaka
Geosci. Model Dev., 17, 8639–8664, https://doi.org/10.5194/gmd-17-8639-2024, https://doi.org/10.5194/gmd-17-8639-2024, 2024
Short summary
Short summary
A new parameterisation for dynamic anthropogenic heat and electricity consumption is described. The model reproduced the temporal variation in and spatial distributions of electricity consumption and temperature well in summer and winter. The partial air conditioning was the most critical factor, significantly affecting the value of anthropogenic heat emission.
Hongyi Li, Ting Yang, Lars Nerger, Dawei Zhang, Di Zhang, Guigang Tang, Haibo Wang, Yele Sun, Pingqing Fu, Hang Su, and Zifa Wang
Geosci. Model Dev., 17, 8495–8519, https://doi.org/10.5194/gmd-17-8495-2024, https://doi.org/10.5194/gmd-17-8495-2024, 2024
Short summary
Short summary
To accurately characterize the spatiotemporal distribution of particulate matter <2.5 µm chemical components, we developed the Nested Air Quality Prediction Model System with the Parallel Data Assimilation Framework (NAQPMS-PDAF) v2.0 for chemical components with non-Gaussian and nonlinear properties. NAQPMS-PDAF v2.0 has better computing efficiency, excels when used with a small ensemble size, and can significantly improve the simulation performance of chemical components.
T. Nash Skipper, Christian Hogrefe, Barron H. Henderson, Rohit Mathur, Kristen M. Foley, and Armistead G. Russell
Geosci. Model Dev., 17, 8373–8397, https://doi.org/10.5194/gmd-17-8373-2024, https://doi.org/10.5194/gmd-17-8373-2024, 2024
Short summary
Short summary
Chemical transport model simulations are combined with ozone observations to estimate the bias in ozone attributable to US anthropogenic sources and individual sources of US background ozone: natural sources, non-US anthropogenic sources, and stratospheric ozone. Results indicate a positive bias correlated with US anthropogenic emissions during summer in the eastern US and a negative bias correlated with stratospheric ozone during spring.
Li Fang, Jianbing Jin, Arjo Segers, Ke Li, Ji Xia, Wei Han, Baojie Li, Hai Xiang Lin, Lei Zhu, Song Liu, and Hong Liao
Geosci. Model Dev., 17, 8267–8282, https://doi.org/10.5194/gmd-17-8267-2024, https://doi.org/10.5194/gmd-17-8267-2024, 2024
Short summary
Short summary
Model evaluations against ground observations are usually unfair. The former simulates mean status over coarse grids and the latter the surrounding atmosphere. To solve this, we proposed the new land-use-based representative (LUBR) operator that considers intra-grid variance. The LUBR operator is validated to provide insights that align with satellite measurements. The results highlight the importance of considering fine-scale urban–rural differences when comparing models and observation.
Mijie Pang, Jianbing Jin, Arjo Segers, Huiya Jiang, Wei Han, Batjargal Buyantogtokh, Ji Xia, Li Fang, Jiandong Li, Hai Xiang Lin, and Hong Liao
Geosci. Model Dev., 17, 8223–8242, https://doi.org/10.5194/gmd-17-8223-2024, https://doi.org/10.5194/gmd-17-8223-2024, 2024
Short summary
Short summary
The ensemble Kalman filter (EnKF) improves dust storm forecasts but faces challenges with position errors. The valid time shifting EnKF (VTS-EnKF) addresses this by adjusting for position errors, enhancing accuracy in forecasting dust storms, as proven in tests on 2021 events, even with smaller ensembles and time intervals.
Prabhakar Namdev, Maithili Sharan, Piyush Srivastava, and Saroj Kanta Mishra
Geosci. Model Dev., 17, 8093–8114, https://doi.org/10.5194/gmd-17-8093-2024, https://doi.org/10.5194/gmd-17-8093-2024, 2024
Short summary
Short summary
Inadequate representation of surface–atmosphere interaction processes is a major source of uncertainty in numerical weather prediction models. Here, an effort has been made to improve the Weather Research and Forecasting (WRF) model version 4.2.2 by introducing a unique theoretical framework under convective conditions. In addition, to enhance the potential applicability of the WRF modeling system, various commonly used similarity functions under convective conditions have also been installed.
Andrew Gettelman, Richard Forbes, Roger Marchand, Chih-Chieh Chen, and Mark Fielding
Geosci. Model Dev., 17, 8069–8092, https://doi.org/10.5194/gmd-17-8069-2024, https://doi.org/10.5194/gmd-17-8069-2024, 2024
Short summary
Short summary
Supercooled liquid clouds (liquid clouds colder than 0°C) are common at higher latitudes (especially over the Southern Ocean) and are critical for constraining climate projections. We compare a single-column version of a weather model to observations with two different cloud schemes and find that both the dynamical environment and atmospheric aerosols are important for reproducing observations.
Yujuan Wang, Peng Zhang, Jie Li, Yaman Liu, Yanxu Zhang, Jiawei Li, and Zhiwei Han
Geosci. Model Dev., 17, 7995–8021, https://doi.org/10.5194/gmd-17-7995-2024, https://doi.org/10.5194/gmd-17-7995-2024, 2024
Short summary
Short summary
This study updates the CESM's aerosol schemes, focusing on dust, marine aerosol emissions, and secondary organic aerosol (SOA) . Dust emission modifications make deflation areas more continuous, improving results in North America and the sub-Arctic. Humidity correction to sea-salt emissions has a minor effect. Introducing marine organic aerosol emissions, coupled with ocean biogeochemical processes, and adding aqueous reactions for SOA formation advance the CESM's aerosol modelling results.
Lucas A. McMichael, Michael J. Schmidt, Robert Wood, Peter N. Blossey, and Lekha Patel
Geosci. Model Dev., 17, 7867–7888, https://doi.org/10.5194/gmd-17-7867-2024, https://doi.org/10.5194/gmd-17-7867-2024, 2024
Short summary
Short summary
Marine cloud brightening (MCB) is a climate intervention technique to potentially cool the climate. Climate models used to gauge regional climate impacts associated with MCB often assume large areas of the ocean are uniformly perturbed. However, a more realistic representation of MCB application would require information about how an injected particle plume spreads. This work aims to develop such a plume-spreading model.
Leonardo Olivetti and Gabriele Messori
Geosci. Model Dev., 17, 7915–7962, https://doi.org/10.5194/gmd-17-7915-2024, https://doi.org/10.5194/gmd-17-7915-2024, 2024
Short summary
Short summary
Data-driven models are becoming a viable alternative to physics-based models for weather forecasting up to 15 d into the future. However, it is unclear whether they are as reliable as physics-based models when forecasting weather extremes. We evaluate their performance in forecasting near-surface cold, hot, and windy extremes globally. We find that data-driven models can compete with physics-based models and that the choice of the best model mainly depends on the region and type of extreme.
David C. Wong, Jeff Willison, Jonathan E. Pleim, Golam Sarwar, James Beidler, Russ Bullock, Jerold A. Herwehe, Rob Gilliam, Daiwen Kang, Christian Hogrefe, George Pouliot, and Hosein Foroutan
Geosci. Model Dev., 17, 7855–7866, https://doi.org/10.5194/gmd-17-7855-2024, https://doi.org/10.5194/gmd-17-7855-2024, 2024
Short summary
Short summary
This work describe how we linked the meteorological Model for Prediction Across Scales – Atmosphere (MPAS-A) with the Community Multiscale Air Quality (CMAQ) air quality model to form a coupled modelling system. This could be used to study air quality or climate and air quality interaction at a global scale. This new model scales well in high-performance computing environments and performs well with respect to ground surface networks in terms of ozone and PM2.5.
Giulio Mandorli and Claudia J. Stubenrauch
Geosci. Model Dev., 17, 7795–7813, https://doi.org/10.5194/gmd-17-7795-2024, https://doi.org/10.5194/gmd-17-7795-2024, 2024
Short summary
Short summary
In recent years, several studies focused their attention on the disposition of convection. Lots of methods, called indices, have been developed to quantify the amount of convection clustering. These indices are evaluated in this study by defining criteria that must be satisfied and then evaluating the indices against these standards. None of the indices meet all criteria, with some only partially meeting them.
Kerry Anderson, Jack Chen, Peter Englefield, Debora Griffin, Paul A. Makar, and Dan Thompson
Geosci. Model Dev., 17, 7713–7749, https://doi.org/10.5194/gmd-17-7713-2024, https://doi.org/10.5194/gmd-17-7713-2024, 2024
Short summary
Short summary
The Global Forest Fire Emissions Prediction System (GFFEPS) is a model that predicts smoke and carbon emissions from wildland fires. The model calculates emissions from the ground up based on satellite-detected fires, modelled weather and fire characteristics. Unlike other global models, GFFEPS uses daily weather conditions to capture changing burning conditions on a day-to-day basis. GFFEPS produced lower carbon emissions due to the changing weather not captured by the other models.
Jianyu Lin, Tie Dai, Lifang Sheng, Weihang Zhang, Shangfei Hai, and Yawen Kong
EGUsphere, https://doi.org/10.5194/egusphere-2024-3321, https://doi.org/10.5194/egusphere-2024-3321, 2024
Short summary
Short summary
The effectiveness of assimilation system and its sensitivity to ensemble member size and length of assimilation window have been investigated. This study advances our understanding about the selection of basic parameters in the four-dimension local ensemble transform Kalman filter assimilation system and the performance of ensemble simulation in a particulate matter polluted environment.
Yi-Ning Shi, Jun Yang, Wei Han, Lujie Han, Jiajia Mao, Wanlin Kan, and Fuzhong Weng
EGUsphere, https://doi.org/10.5194/egusphere-2024-2884, https://doi.org/10.5194/egusphere-2024-2884, 2024
Short summary
Short summary
Assimilating Ground-based microwave radiometers' observations into numerical weather prediction models holds significant promise for enhancing forecast accuracy. Radiative transfer models (RTM) are crucial for direct data assimilation. We propose a new RTM capable of simulating brightness temperatures observed by GMRs and their Jacobians. Several improvements are introduced to achieve higher accuracy.The RTM align with RTTOV-gb well and can achieve smaller STD in water vapor absorption channels.
Samiha Binte Shahid, Forrest G. Lacey, Christine Wiedinmyer, Robert J. Yokelson, and Kelley C. Barsanti
Geosci. Model Dev., 17, 7679–7711, https://doi.org/10.5194/gmd-17-7679-2024, https://doi.org/10.5194/gmd-17-7679-2024, 2024
Short summary
Short summary
The Next-generation Emissions InVentory expansion of Akagi (NEIVA) v.1.0 is a comprehensive biomass burning emissions database that allows integration of new data and flexible querying. Data are stored in connected datasets, including recommended averages of ~1500 constituents for 14 globally relevant fire types. Individual compounds were mapped to common model species to allow better attribution of emissions in modeling studies that predict the effects of fires on air quality and climate.
Lucie Bakels, Daria Tatsii, Anne Tipka, Rona Thompson, Marina Dütsch, Michael Blaschek, Petra Seibert, Katharina Baier, Silvia Bucci, Massimo Cassiani, Sabine Eckhardt, Christine Groot Zwaaftink, Stephan Henne, Pirmin Kaufmann, Vincent Lechner, Christian Maurer, Marie D. Mulder, Ignacio Pisso, Andreas Plach, Rakesh Subramanian, Martin Vojta, and Andreas Stohl
Geosci. Model Dev., 17, 7595–7627, https://doi.org/10.5194/gmd-17-7595-2024, https://doi.org/10.5194/gmd-17-7595-2024, 2024
Short summary
Short summary
Computer models are essential for improving our understanding of how gases and particles move in the atmosphere. We present an update of the atmospheric transport model FLEXPART. FLEXPART 11 is more accurate due to a reduced number of interpolations and a new scheme for wet deposition. It can simulate non-spherical aerosols and includes linear chemical reactions. It is parallelised using OpenMP and includes new user options. A new user manual details how to use FLEXPART 11.
Bjarke Tobias Eisensøe Olsen, Andrea Noemi Hahmann, Nicolás González Alonso-de-Linaje, Mark Žagar, and Martin Dörenkämper
EGUsphere, https://doi.org/10.5194/egusphere-2024-3123, https://doi.org/10.5194/egusphere-2024-3123, 2024
Short summary
Short summary
Low-level jets (LLJs) are strong winds in the lower atmosphere, important for wind energy as turbines get taller. This study compares a weather model (WRF) with real data across five North and Baltic Sea sites. Adjusting the model improved accuracy over the widely-used ERA5. In key offshore regions, LLJs occur 10–15 % of the time and significantly boost wind power, especially in spring and summer, contributing up to 30 % of total capacity in some areas.
Jaroslav Resler, Petra Bauerová, Michal Belda, Martin Bureš, Kryštof Eben, Vladimír Fuka, Jan Geletič, Radek Jareš, Jan Karel, Josef Keder, Pavel Krč, William Patiño, Jelena Radović, Hynek Řezníček, Matthias Sühring, Adriana Šindelářová, and Ondřej Vlček
Geosci. Model Dev., 17, 7513–7537, https://doi.org/10.5194/gmd-17-7513-2024, https://doi.org/10.5194/gmd-17-7513-2024, 2024
Short summary
Short summary
Detailed modeling of urban air quality in stable conditions is a challenge. We show the unprecedented sensitivity of a large eddy simulation (LES) model to meteorological boundary conditions and model parameters in an urban environment under stable conditions. We demonstrate the crucial role of boundary conditions for the comparability of results with observations. The study reveals a strong sensitivity of the results to model parameters and model numerical instabilities during such conditions.
Matthieu Dabrowski, José Mennesson, Jérôme Riedi, Chaabane Djeraba, and Pierre Nabat
EGUsphere, https://doi.org/10.5194/egusphere-2024-2676, https://doi.org/10.5194/egusphere-2024-2676, 2024
Short summary
Short summary
This work focuses on the prediction of aerosol concentration values at ground level, which are a strong indicator of air quality, using Artificial Neural Networks. A study of different variables and their efficiency as inputs for these models is also proposed, and reveals that the best results are obtained when using all of them. Comparison of networks architectures and information fusion methods allows the extraction of knowledge on the most efficient methods in the context of this study.
Jorge E. Pachón, Mariel A. Opazo, Pablo Lichtig, Nicolas Huneeus, Idir Bouarar, Guy Brasseur, Cathy W. Y. Li, Johannes Flemming, Laurent Menut, Camilo Menares, Laura Gallardo, Michael Gauss, Mikhail Sofiev, Rostislav Kouznetsov, Julia Palamarchuk, Andreas Uppstu, Laura Dawidowski, Nestor Y. Rojas, María de Fátima Andrade, Mario E. Gavidia-Calderón, Alejandro H. Delgado Peralta, and Daniel Schuch
Geosci. Model Dev., 17, 7467–7512, https://doi.org/10.5194/gmd-17-7467-2024, https://doi.org/10.5194/gmd-17-7467-2024, 2024
Short summary
Short summary
Latin America (LAC) has some of the most populated urban areas in the world, with high levels of air pollution. Air quality management in LAC has been traditionally focused on surveillance and building emission inventories. This study performed the first intercomparison and model evaluation in LAC, with interesting and insightful findings for the region. A multiscale modeling ensemble chain was assembled as a first step towards an air quality forecasting system.
David Ho, Michał Gałkowski, Friedemann Reum, Santiago Botía, Julia Marshall, Kai Uwe Totsche, and Christoph Gerbig
Geosci. Model Dev., 17, 7401–7422, https://doi.org/10.5194/gmd-17-7401-2024, https://doi.org/10.5194/gmd-17-7401-2024, 2024
Short summary
Short summary
Atmospheric model users often overlook the impact of the land–atmosphere interaction. This study accessed various setups of WRF-GHG simulations that ensure consistency between the model and driving reanalysis fields. We found that a combination of nudging and frequent re-initialization allows certain improvement by constraining the soil moisture fields and, through its impact on atmospheric mixing, improves atmospheric transport.
Jie Gao, Yi Huang, Jonathon S. Wright, Ke Li, Tao Geng, and Qiurun Yu
EGUsphere, https://doi.org/10.5194/egusphere-2024-2815, https://doi.org/10.5194/egusphere-2024-2815, 2024
Short summary
Short summary
The aerosol in the upper troposphere and stratosphere is highly variable, and its radiative effect is poorly understood. To estimate that effect, the radiative kernel is constructed and applied. The results show that the kernels can reproduce aerosol radiative effects and are expected to simulate stratospheric aerosol radiative effects. This approach reduces computational expense and consists well with radiative model calculations and can be applied to atmospheric models with speed requirements.
Phuong Loan Nguyen, Lisa V. Alexander, Marcus J. Thatcher, Son C. H. Truong, Rachael N. Isphording, and John L. McGregor
Geosci. Model Dev., 17, 7285–7315, https://doi.org/10.5194/gmd-17-7285-2024, https://doi.org/10.5194/gmd-17-7285-2024, 2024
Short summary
Short summary
We use a comprehensive approach to select a subset of CMIP6 models for dynamical downscaling over Southeast Asia, taking into account model performance, model independence, data availability and the range of future climate projections. The standardised benchmarking framework is applied to assess model performance through both statistical and process-based metrics. Ultimately, we identify two independent model groups that are suitable for dynamical downscaling in the Southeast Asian region.
Ingrid Super, Tia Scarpelli, Arjan Droste, and Paul I. Palmer
Geosci. Model Dev., 17, 7263–7284, https://doi.org/10.5194/gmd-17-7263-2024, https://doi.org/10.5194/gmd-17-7263-2024, 2024
Short summary
Short summary
Monitoring greenhouse gas emission reductions requires a combination of models and observations, as well as an initial emission estimate. Each component provides information with a certain level of certainty and is weighted to yield the most reliable estimate of actual emissions. We describe efforts for estimating the uncertainty in the initial emission estimate, which significantly impacts the outcome. Hence, a good uncertainty estimate is key for obtaining reliable information on emissions.
Álvaro González-Cervera and Luis Durán
Geosci. Model Dev., 17, 7245–7261, https://doi.org/10.5194/gmd-17-7245-2024, https://doi.org/10.5194/gmd-17-7245-2024, 2024
Short summary
Short summary
RASCAL is an open-source Python tool designed for reconstructing daily climate observations, especially in regions with complex local phenomena. It merges large-scale weather patterns with local weather using the analog method. Evaluations in central Spain show that RASCAL outperforms ERA20C reanalysis in reconstructing precipitation and temperature. RASCAL offers opportunities for broad scientific applications, from short-term forecasts to local-scale climate change scenarios.
Juan Escobar, Philippe Wautelet, Joris Pianezze, Florian Pantillon, Thibaut Dauhut, Christelle Barthe, and Jean-Pierre Chaboureau
EGUsphere, https://doi.org/10.5194/egusphere-2024-2879, https://doi.org/10.5194/egusphere-2024-2879, 2024
Short summary
Short summary
The Meso-NH weather research code is adapted for GPUs using OpenACC, leading to significant performance and energy efficiency improvements. Called MESONH-v55-OpenACC, it includes enhanced memory management, communication optimizations, and a new solver. On the AMD MI250X Adastra platform, it achieved up to 6× speedup and 2.3× energy efficiency gain compared to CPUs. Storm simulations at 100 m resolution show successful results , positioning the code for future use on exascale supercomputers.
Matti Niskanen, Aku Seppänen, Henri Oikarinen, Miska Olin, Panu Karjalainen, Santtu Mikkonen, and Kari Lehtinen
EGUsphere, https://doi.org/10.5194/egusphere-2024-1898, https://doi.org/10.5194/egusphere-2024-1898, 2024
Short summary
Short summary
Particle size is a key factor determining the properties of aerosol particles, which have a major influence on the climate and on human health. When measuring the particle sizes, however, sometimes the sampling lines that transfer the aerosol to the measurement device distort the size distribution, making the measurement unreliable. We propose a method to correct for the distortions and estimate the true particle sizes, improving measurement accuracy.
Sun-Young Park, Kyo-Sun Sunny Lim, Kwonil Kim, Gyuwon Lee, and Jason A. Milbrandt
Geosci. Model Dev., 17, 7199–7218, https://doi.org/10.5194/gmd-17-7199-2024, https://doi.org/10.5194/gmd-17-7199-2024, 2024
Short summary
Short summary
We enhance the WDM6 scheme by incorporating predicted graupel density. The modification affects graupel characteristics, including fall velocity–diameter and mass–diameter relationships. Simulations highlight changes in graupel distribution and precipitation patterns, potentially influencing surface snow amounts. The study underscores the significance of integrating predicted graupel density for a more realistic portrayal of microphysical properties in weather models.
Christos I. Efstathiou, Elizabeth Adams, Carlie J. Coats, Robert Zelt, Mark Reed, John McGee, Kristen M. Foley, Fahim I. Sidi, David C. Wong, Steven Fine, and Saravanan Arunachalam
Geosci. Model Dev., 17, 7001–7027, https://doi.org/10.5194/gmd-17-7001-2024, https://doi.org/10.5194/gmd-17-7001-2024, 2024
Short summary
Short summary
We present a summary of enabling high-performance computing of the Community Multiscale Air Quality Model (CMAQ) – a state-of-the-science community multiscale air quality model – on two cloud computing platforms through documenting the technologies, model performance, scaling and relative merits. This may be a new paradigm for computationally intense future model applications. We initiated this work due to a need to leverage cloud computing advances and to ease the learning curve for new users.
Peter A. Bogenschutz, Jishi Zhang, Qi Tang, and Philip Cameron-Smith
Geosci. Model Dev., 17, 7029–7050, https://doi.org/10.5194/gmd-17-7029-2024, https://doi.org/10.5194/gmd-17-7029-2024, 2024
Short summary
Short summary
Using high-resolution and state-of-the-art modeling techniques we simulate five atmospheric river events for California to test the capability to represent precipitation for these events. We find that our model is able to capture the distribution of precipitation very well but suffers from overestimating the precipitation amounts over high elevation. Increasing the resolution further has no impact on reducing this bias, while increasing the domain size does have modest impacts.
Manu Anna Thomas, Klaus Wyser, Shiyu Wang, Marios Chatziparaschos, Paraskevi Georgakaki, Montserrat Costa-Surós, Maria Gonçalves Ageitos, Maria Kanakidou, Carlos Pérez García-Pando, Athanasios Nenes, Twan van Noije, Philippe Le Sager, and Abhay Devasthale
Geosci. Model Dev., 17, 6903–6927, https://doi.org/10.5194/gmd-17-6903-2024, https://doi.org/10.5194/gmd-17-6903-2024, 2024
Short summary
Short summary
Aerosol–cloud interactions occur at a range of spatio-temporal scales. While evaluating recent developments in EC-Earth3-AerChem, this study aims to understand the extent to which the Twomey effect manifests itself at larger scales. We find a reduction in the warm bias over the Southern Ocean due to model improvements. While we see footprints of the Twomey effect at larger scales, the negative relationship between cloud droplet number and liquid water drives the shortwave radiative effect.
Kai Cao, Qizhong Wu, Lingling Wang, Hengliang Guo, Nan Wang, Huaqiong Cheng, Xiao Tang, Dongxing Li, Lina Liu, Dongqing Li, Hao Wu, and Lanning Wang
Geosci. Model Dev., 17, 6887–6901, https://doi.org/10.5194/gmd-17-6887-2024, https://doi.org/10.5194/gmd-17-6887-2024, 2024
Short summary
Short summary
AMD’s heterogeneous-compute interface for portability was implemented to port the piecewise parabolic method solver from NVIDIA GPUs to China's GPU-like accelerators. The results show that the larger the model scale, the more acceleration effect on the GPU-like accelerator, up to 28.9 times. The multi-level parallelism achieves a speedup of 32.7 times on the heterogeneous cluster. By comparing the results, the GPU-like accelerators have more accuracy for the geoscience numerical models.
Cited articles
Ayturan, Y. A., Ayturan, Z. C., and Altun, H. O.: Air pollution modelling
with deep learning: a review, International Journal of Environmental
Pollution and Environmental Modelling, 1, 58–62, 2018.
Bi, J., Knowland, K. E., Keller, C. A., and Liu, Y.: Combining Machine
Learning and Numerical Simulation for High-Resolution PM2.5
Concentration Forecast, Environ. Sci. Technol., 56,
1544–1556, 2022.
Bui, T.-C., Le, V.-D., and Cha, S.-K.: A deep learning approach for
forecasting air pollution in South Korea using LSTM, arXiv [preprint],
https://doi.org/10.48550/arXiv.1804.07891, 2018.
Fan, C., Li, Y., Guang, J., Li, Z., Elnashar, A., Allam, M., and de Leeuw,
G.: The impact of the control measures during the COVID-19 outbreak on air
pollution in China, Remote Sensing, 12, 1613, https://doi.org/10.3390/rs12101613, 2020.
Gilliam, R. C., Hogrefe, C., Godowitch, J. M., Napelenok, S., Mathur, R.,
and Rao, S. T.: Impact of inherent meteorology uncertainty on air quality
model predictions, J. Geophys. Res.-Atmos., 120,
12259–12280, 2015.
Greff, K., Srivastava, R., Koutník, J., Steunebrink, B., and
Schmidhuber, J.: LSTM: A search space odyssey, IEEE Transactions on Neural
Networks Learning Systems, https://doi.org/10.1109/TNNLS.2016.2582924, 2017.
Hähnel, P., Mareček, J., Monteil, J., and O'Donncha, F.: Using deep
learning to extend the range of air pollution monitoring and forecasting,
J. Comput. Phys., 408, 109278, 2020.
Han, J., Liu, H., Zhu, H., Xiong, H., and Dou, D.: Joint air quality and
weather prediction based on multi-adversarial spatiotemporal networks,
Proceedings of the AAAI Conference on Artificial Intelligence, 2–9 February 2021, virtual conference, 4081–4089, https://doi.org/10.48550/arXiv.2012.15037,
2021.
Hochreiter, S. and Schmidhuber, J.: Long short-term memory,
Neural Comput., 9, 1735–1780, 1997.
Holnicki, P. and Nahorski, Z.: Emission data uncertainty in urban air
quality modeling–case study, Environ. Model. Assess., 20,
583–597, 2015.
Huang, C.-J. and Kuo, P.-H.: A deep CNN-LSTM model for particulate matter
(PM2.5) forecasting in smart cities, Sensors, 18, 2220, https://doi.org/10.3390/s18072220, 2018.
Ioffe, S. and Szegedy, C.: Batch normalization: Accelerating deep network
training by reducing internal covariate shift, International conference on
machine learning, 6–11 July 2015, Lille, France, 448–456, https://doi.org/10.48550/arXiv.1502.03167, 2015.
Janarthanan, R., Partheeban, P., Somasundaram, K., and Elamparithi, P. N.: A
deep learning approach for prediction of air quality index in a metropolitan
city, Sustain. Cities Soc., 67, 102720, 2021.
Karim, F., Majumdar, S., Darabi, H., and Chen, S.: LSTM fully convolutional
networks for time series classification, IEEE access, 6, 1662–1669, 2017.
Kim, H. S., Park, I., Song, C. H., Lee, K., Yun, J. W., Kim, H. K., Jeon, M., Lee, J., and Han, K. M.: Development of a daily PM10 and PM2.5 prediction system using a deep long short-term memory neural network model, Atmos. Chem. Phys., 19, 12935–12951, https://doi.org/10.5194/acp-19-12935-2019, 2019.
Kingma, D. P. and Ba, J.: Adam: A method for stochastic optimization, arXiv
[preprint],https://doi.org/10.48550/arXiv.1412.6980, 2014.
Lee, K., Yu, J., Lee, S., Park, M., Hong, H., Park, S. Y., Choi, M., Kim, J., Kim, Y., Woo, J.-H., Kim, S.-W., and Song, C. H.: Development of Korean Air Quality Prediction System version 1 (KAQPS v1) with focuses on practical issues, Geosci. Model Dev., 13, 1055–1073, https://doi.org/10.5194/gmd-13-1055-2020, 2020.
Li, T., Hua, M., and Wu, X.: A hybrid CNN-LSTM model for forecasting
particulate matter (PM2.5), Ieee Access, 8, 26933–26940, 2020.
Lu, H., Xie, M., Liu, X., Liu, B., Jiang, M., Gao, Y., and Zhao, X.:
Adjusting prediction of ozone concentration based on CMAQ model and machine
learning methods in Sichuan-Chongqing region, China, Atmos. Pollut.
Res., 12, 101066, https://doi.org/10.1016/j.apr.2021.101066, 2021.
Lu, X., Fung, J. C. H., and Wu, D.: Modeling wet deposition of acid
substances over the PRD region in China, Atmos. Environ., 122,
819–828, 2015.
Lu, X., Wang, Y., Li, J., Shen, L., and Fung, J. C.: Evidence of
heterogeneous HONO formation from aerosols and the regional photochemical
impact of this HONO source, Environ. Res. Lett., 13, 114002, https://doi.org/10.1088/1748-9326/aae492,
2018.
Lu, X., Zhang, S., Xing, J., Wang, Y., Chen, W., Ding, D., Wu, Y., Wang, S.,
Duan, L., and Hao, J.: Progress of air pollution control in China and its
challenges and opportunities in the ecological civilization era,
Engineering, 6, 1423–1431, 2020.
Lu, X., Sha, Y. H., Li, Z., Huang, Y., Chen, W., Chen, D., Shen, J., Chen,
Y., and Fung, J. C.: Development and application of a hybrid long-short term
memory–three dimensional variational technique for the improvement of
PM2.5 forecasting, Sci. Total Environ., 770, 144221, https://doi.org/10.1016/j.scitotenv.2020.144221,
2021.
Lyu, B., Hu, Y., Zhang, W., Du, Y., Luo, B., Sun, X., Sun, Z., Deng, Z.,
Wang, X., and Liu, J.: Fusion method combining ground-level observations
with chemical transport model predictions using an ensemble deep learning
framework: application in China to estimate spatiotemporally-resolved
PM2.5 exposure fields in 2014–2017, Environ. Sci.
Technol., 53, 7306–7315, 2019.
Ma, J., Ding, Y., Cheng, J. C., Jiang, F., and Wan, Z.: A temporal-spatial
interpolation and extrapolation method based on geographic Long Short-Term
Memory neural network for PM2.5, J. Clean. Prod., 237,
117729, https://doi.org/10.1016/j.jclepro.2019.117729, 2019.
Mao, W., Wang, W., Jiao, L., Zhao, S., and Liu, A.: Modeling air quality
prediction using a deep learning approach: Method optimization and
evaluation, Sustain. Cities Soc., 65, 102567, https://doi.org/10.1016/j.scs.2020.102567, 2021.
Pak, U., Ma, J., Ryu, U., Ryom, K., Juhyok, U., Pak, K., and Pak, C.: Deep
learning-based PM2.5 prediction considering the spatiotemporal
correlations: A case study of Beijing, China, Sci. Total
Environ., 699, 133561, https://doi.org/10.1016/j.scitotenv.2019.07.367, 2020.
Qi, Y., Li, Q., Karimian, H., and Liu, D.: A hybrid model for spatiotemporal
forecasting of PM2.5 based on graph convolutional neural network and
long short-term memory, Sci. Total Environ., 664, 1–10, 2019.
Qin, D., Yu, J., Zou, G., Yong, R., Zhao, Q., and Zhang, B.: A novel
combined prediction scheme based on CNN and LSTM for urban PM2.5 concentration, IEEE Access, 7, 20050–20059, 2019.
Samal, K. K. R., Panda, A. K., Babu, K. S., and Das, S. K.: An improved
pollution forecasting model with meteorological impact using multiple
imputation and fine-tuning approach, Sustain. Cities Soc., 70,
102923, https://doi.org/10.1016/j.scs.2021.102923, 2021.
Sayeed, A., Choi, Y., Eslami, E., Jung, J., Lops, Y., Salman, A. K., Lee,
J.-B., Park, H.-J., and Choi, M.-H.: A novel CMAQ-CNN hybrid model to
forecast hourly surface-ozone concentrations 14 days in advance, Sci.
Rep.-UK, 11, 1–8, 2021a.
Sayeed, A., Lops, Y., Choi, Y., Jung, J., and Salman, A. K.: Bias correcting
and extending the PM forecast by CMAQ up to 7 days using deep convolutional
neural networks, Atmos. Environ., 253, 118376, https://doi.org/10.1016/j.atmosenv.2021.118376, 2021b.
Schuster, M. and Paliwal Kuldip, K.: Bidirectional recurrent neural
networks, IEEE T. Signal Proces., 45, 2673–2681, 1997.
Siami-Namini, S., Tavakoli, N., and Namin, A. S.: A comparison of ARIMA and
LSTM in forecasting time series, 2018 17th IEEE international conference on
machine learning and applications (ICMLA), 17–20 December 2018,
Orlando, Florida, USA, 1394–1401, https://doi.org/10.1109/ICMLA.2018.00227, 2018.
Skamarock, W. C., Klemp, J. B., Dudhia, J., Gill, D. O., Barker, D., Duda, M. G., Huang, X. Y., Wang W., and Powers, J. G.: A description of the Advanced Research WRF version 3, NCAR Technical note-475+ STR, https://doi.org/10.5065/D68S4MVH, 2008 (data available at https://www2.mmm.ucar.edu/wrf/users/download/get_source.html, last access: 14 November 2022).
Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., and
Salakhutdinov, R.: Dropout: a simple way to prevent neural networks from
overfitting, J. Mach. Learn. Res., 15, 1929–1958, 2014.
Sun, H., Fung, J. C., Chen, Y., Chen, W., Li, Z., Huang, Y., Lin, C., Hu,
M., and Lu, X.: Improvement of PM2.5 and O3 forecasting by
integration of 3D numerical simulation with deep learning techniques,
Sustain. Cities Soc., 75, 103372, https://doi.org/10.1016/j.scs.2021.103372, 2021.
Sun, H., Fung, J. C. H., Chen, Y., Li, Z., Yuan, D., Chen, W., and Lu, X.: Ground obeservation data (meteorological factors and air pollution) in Greater Bay Area, 2015–2021, Zenodo [data set], https://doi.org/10.5281/zenodo.6598377, 2022a.
Sun, H., Fung, J. C. H., Chen, Y., Li, Z., Yuan, D., Chen, W., and Lu, X.: Prediction of the broadcasting model and various baselines, Zenodo [data set], https://doi.org/10.5281/zenodo.6833673, 2022b.
Sun, H., Fung, J. C. H., Chen, Y., Li, Z., Yuan, D., Chen, W., and Lu, X.: Deep learning models in the study “Development of an LSTM-Broadcasting deep-learning framework for regional air pollution forecast improvement”, Zenodo [data set], https://doi.org/10.5281/zenodo.6827585, 2022c.
Sun, H., Fung, J. C. H., Chen, Y., Li, Z., Yuan, D., Chen, W., and Lu, X.: Processed ground observation and WRF-CAMQ data for Greater Bay Area, 2015–2021, Zenodo [data set], https://doi.org/10.5281/zenodo.6601173, 2022d.
Sun, J. H.: jvhs0706/regional-forecast-new: GMD paper code, Zenodo [code], https://doi.org/10.5281/zenodo.7019243, 2022.
Sutskever, I., Vinyals, O., and Le, Q. V.: Sequence to sequence learning
with neural networks, Adv. Nur. In., 27,
3104–3112,
https://doi.org/10.48550/arXiv.1409.3215, 2014.
Tang, Y., Lee, P., Tsidulko, M., Huang, H.-C., McQueen, J. T., DiMego, G.
J., Emmons, L. K., Pierce, R. B., Thompson, A. M., and Lin, H.-M.: The
impact of chemical lateral boundary conditions on CMAQ predictions of
tropospheric ozone over the continental United States, Environ. Fluid
Mech., 9, 43–58, 2009.
Tsai, Y.-T., Zeng, Y.-R., and Chang, Y.-S.: Air pollution forecasting using
RNN with LSTM, 2018 IEEE 16th Intl Conf on Dependable, Autonomic and Secure
Computing, 16th Intl Conf on Pervasive Intelligence and Computing, 4th Intl
Conf on Big Data Intelligence and Computing and Cyber Science and Technology
Congress (DASC/PiCom/DataCom/CyberSciTech), 12–15 August 2018,
Athens, Greece, 1074–1079, https://doi.org/10.1109/DASC/PiCom/DataCom/CyberSciTec.2018.00178, 2018.
United States Environmental Protection Agency: CMAQ (Version 5.0.2), Zenodo [software], https://doi.org/10.5281/zenodo.1079898, 2014.
Wu, Q. and Lin, H.: Daily urban air quality index forecasting based on
variational mode decomposition, sample entropy and LSTM neural network,
Sustain. Cities Soc., 50, 101657, https://doi.org/10.1016/j.scs.2019.101657, 2019.
Zhang, Y., Bocquet, M., Mallet, V., Seigneur, C., and Baklanov, A.:
Real-time air quality forecasting, part II: State of the science, current
research needs, and future prospects, Atmos. Environ., 60, 656–676,
2012.
Zhao, J., Deng, F., Cai, Y., and Chen, J.: Long short-term memory-Fully
connected (LSTM-FC) neural network for PM2.5 concentration prediction,
Chemosphere, 220, 486–492, 2019.
Zhou, X., Tong, W., and Li, L.: Deep learning spatiotemporal air pollution
data in China using data fusion, Earth Sci. Inform., 13, 859–868,
2020.
Short summary
This study developed a novel deep-learning layer, the broadcasting layer, to build an end-to-end LSTM-based deep-learning model for regional air pollution forecast. By combining the ground observation, WRF-CMAQ simulation, and the broadcasting LSTM deep-learning model, forecast accuracy has been significantly improved when compared to other methods. The broadcasting layer and its variants can also be applied in other research areas to supersede the traditional numerical interpolation methods.
This study developed a novel deep-learning layer, the broadcasting layer, to build an end-to-end...
