Articles | Volume 16, issue 14
© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
An optimized semi-empirical physical approach for satellite-based PM2.5 retrieval: embedding machine learning to simulate complex physical parameters
School of Geodesy and Geomatics, Wuhan University, Wuhan 430079, China
School of Geodesy and Geomatics, Wuhan University, Wuhan 430079, China
Collaborative Innovation Center of Geospatial Technology, Wuhan 430079, China
Key Laboratory of Geospace Environment and Geodesy (Ministry of Education), Wuhan University, Wuhan 430079, China
School of Geospatial Engineering and Science, Sun Yat-Sen University, Zhuhai 519082, China
School of Geodesy and Geomatics, Wuhan University, Wuhan 430079, China
Collaborative Innovation Center of Geospatial Technology, Wuhan 430079, China
State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan 430079, China
No articles found.
Yuan Wang, Qiangqiang Yuan, Tongwen Li, Yuanjian Yang, Siqin Zhou, and Liangpei Zhang
Earth Syst. Sci. Data, 15, 3597–3622,Short summary
We propose a novel spatiotemporally self-supervised fusion method to establish long-term daily seamless global XCO2 and XCH4 products. Results show that the proposed method achieves a satisfactory accuracy that distinctly exceeds that of CAMS-EGG4 and is superior or close to those of GOSAT and OCO-2. In particular, our fusion method can effectively correct the large biases in CAMS-EGG4 due to the issues from assimilation data, such as the unadjusted anthropogenic emission for COVID-19.
Qiang Zhang, Qiangqiang Yuan, Taoyong Jin, Meiping Song, and Fujun Sun
Earth Syst. Sci. Data, 14, 4473–4488,Short summary
Compared to previous seamless global daily soil moisture (SGD-SM 1.0) products, SGD-SM 2.0 enlarges the temporal scope from 2002 to 2022. By fusing auxiliary precipitation information with the long short-term memory convolutional neural network (LSTM-CNN) model, SGD-SM 2.0 can consider sudden extreme weather conditions for 1 d in global daily soil moisture products and is significant for full-coverage global daily hydrologic monitoring, rather than averaging monthly–quarterly–yearly results.
Xiaobin Guan, Huanfeng Shen, Yuchen Wang, Dong Chu, Xinghua Li, Linwei Yue, Xinxin Liu, and Liangpei Zhang
Earth Syst. Sci. Data Discuss.,
Preprint withdrawnShort summary
This study generated the first global 1-km continuous NDVI product (STFLNDVI) for 4-decades by fusing multi-source satellite products. Simulated and real-data assessments confirmed the satisfactory and stable accuracy of STFLNDVI regarding spatial details and temporal variations. STFLNDVI is an ideal solution to the trade-off between spatial resolution and time coverage in current NDVI products, which of great significance for long-term regional and global vegetation and climate change studies.
Qiang Zhang, Qiangqiang Yuan, Jie Li, Yuan Wang, Fujun Sun, and Liangpei Zhang
Earth Syst. Sci. Data, 13, 1385–1401,Short summary
Acquired daily soil moisture products are always incomplete globally (just about 30 %–80 % coverage ratio) due to the satellite orbit coverage and the limitations of soil moisture retrieval algorithms. To solve this inevitable problem, we generate long-term seamless global daily (SGD) AMSR2 soil moisture productions from 2013 to 2019. These productions are significant for full-coverage global daily hydrologic monitoring, rather than averaging as the monthly–quarter–yearly results.
Yuan Wang, Qiangqiang Yuan, Tongwen Li, Siyu Tan, and Liangpei Zhang
Atmos. Chem. Phys. Discuss.,
Revised manuscript not acceptedShort summary
Estimating ambient PM2.5 and PM10 considering their precursors and chemical compositions instead of AOD products; Both remote sensing (Sentinel-5P) and assimilated data (GEOS-FP) are adopted; Sample-based Cross-Validation R2s and RMSEs are 0.93 (0.9) and 8.982 (17.604) μg/m3 for PM2.5 (PM10), respectively; Achieving better performance compared to the baseline (AOD-based) in different cases (e.g., overall and seasonal).
C. Zhou, J. Li, H. Shen, and Q. Yuan
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., V-5-2020, 101–107,
Xinghua Li, Yinghong Jing, Huanfeng Shen, and Liangpei Zhang
Hydrol. Earth Syst. Sci., 23, 2401–2416,Short summary
This paper is a review article on the cloud removal methods of MODIS snow cover products.
Tongwen Li, Chengyue Zhang, Huanfeng Shen, Qiangqiang Yuan, and Liangpei Zhang
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., IV-3, 143–147,
Zhiwei Li, Huanfeng Shen, Yancong Wei, Qing Cheng, and Qiangqiang Yuan
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., IV-3, 149–152,
X. Meng, H. Shen, Q. Yuan, H. Li, and L. Zhang
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-2-W7, 831–835,
Hongyan Zhang, Han Zhai, Wenzhi Liao, Liqin Cao, Liangpei Zhang, and Aleksandra Pižurica
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLI-B3, 945–948,
Tianzhu Xiang, Gui-Song Xia, and Liangpei Zhang
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., III-3, 287–294,
Related subject area
Atmospheric sciencesA Python library for computing individual and merged non-CO2 algorithmic climate change functions: CLIMaCCF V1.0The three-dimensional structure of fronts in mid-latitude weather systems in numerical weather prediction modelsThe development and validation of the Inhomogeneous Wind Scheme for Urban Street (IWSUS-v1)GPU-HADVPPM V1.0: a high-efficiency parallel GPU design of the piecewise parabolic method (PPM) for horizontal advection in an air quality model (CAMx V6.10)Variability and combination as an ensemble of mineral dust forecasts during the 2021 CADDIWA experiment using the WRF 3.7.1 and CHIMERE v2020r3 modelsBreakups are complicated: an efficient representation of collisional breakup in the superdroplet methodSensitivity of tropospheric ozone to halogen chemistry in the chemistry–climate model LMDZ-INCA vNMHCSegmentation of XCO2 images with deep learning: application to synthetic plumes from cities and power plantsEvaluating precipitation distributions at regional scales: a benchmarking framework and application to CMIP5 and 6 modelsThe Fire Inventory from NCAR version 2.5: an updated global fire emissions model for climate and chemistry applicationsAn approach to refining the ground meteorological observation stations for improving PM2.5 forecasts in the Beijing–Tianjin–Hebei regionAssessment of WRF (v 4.2.1) dynamically downscaled precipitation on subdaily and daily timescales over CONUSConvective-gust nowcasting based on radar reflectivity and a deep learning algorithmSelf-nested large-eddy simulations in PALM model system v21.10 for offshore wind prediction under different atmospheric stability conditionsHow does cloud-radiative heating over the North Atlantic change with grid spacing, convective parameterization, and microphysics scheme in ICON version 2.1.00?Updated isoprene and terpene emission factors for the Interactive BVOC (iBVOC) emission scheme in the United Kingdom Earth System Model (UKESM1.0)Technical descriptions of the experimental dynamical downscaling simulations over North America by the CAM–MPAS variable-resolution modelIntercomparison of the weather and climate physics suites of a unified forecast–climate model system (GRIST-A22.7.28) based on single-column modelingHalogen chemistry in volcanic plumes: a 1D framework based on MOCAGE 1D (version R1.18.1) preparing 3D global chemistry modellingPyFLEXTRKR: a flexible feature tracking Python software for convective cloud analysisCLGAN: a generative adversarial network (GAN)-based video prediction model for precipitation nowcastingLong-term evaluation of surface air pollution in CAMSRA and MERRA-2 global reanalyses over Europe (2003–2020)Emulating aerosol optics with randomly generated neural networksDevelopment of an ecophysiology module in the GEOS-Chem chemical transport model version 12.2.0 to represent biosphere–atmosphere fluxes relevant for ozone air qualityComparison of ozone formation attribution techniques in the northeastern United StatesImproving trajectory calculations by FLEXPART 10.4+ using single-image super-resolutionData fusion uncertainty-enabled methods to map street-scale hourly NO2 in Barcelona: a case study with CALIOPE-Urban v1.0Forecasting tropical cyclone tracks in the northwestern Pacific based on a deep-learning modelCHEEREIO 1.0: a versatile and user-friendly ensemble-based chemical data assimilation and emissions inversion platform for the GEOS-Chem chemical transport modelAccelerating models for multiphase chemical kinetics through machine learning with polynomial chaos expansion and neural networksA machine learning emulator for Lagrangian particle dispersion model footprints: a case study using NAMEImproving the representation of shallow cumulus convection with the simplified-higher-order-closure–mass-flux (SHOC+MF v1.0) approachISAT v2.0: an integrated tool for nested-domain configurations and model-ready emission inventories for WRF-AQMEstimation of CH4 emission based on an advanced 4D-LETKF assimilation systemAccelerated estimation of sea-spray-mediated heat flux using Gaussian quadrature: case studies with a coupled CFSv2.0-WW3 systemAMORE-Isoprene v1.0: a new reduced mechanism for gas-phase isoprene oxidationA method for generating a quasi-linear convective system suitable for observing system simulation experimentsThe second Met Office Unified Model–JULES Regional Atmosphere and Land configuration, RAL2Visual analysis of model parameter sensitivities along warm conveyor belt trajectories using Met.3D (1.6.0-multivar0)A dynamic ammonia emission model and the online coupling with WRF–Chem (WRF–SoilN–Chem v1.0): development and regional evaluation in ChinaSCIATRAN software package (V4.6): update and further development of aerosol, clouds, surface reflectance databases and modelsDeep learning models for generation of precipitation maps based on numerical weather predictionAn inconsistency in aviation emissions between CMIP5 and CMIP6 and the implications for short-lived species and their radiative forcingOn the use of Infrared Atmospheric Sounding Interferometer (IASI) spectrally resolved radiances to test the EC-Earth climate model (v3.3.3) in clear-sky conditionsIncorporation of aerosol into the COSPv2 satellite lidar simulator for climate model evaluationThe impact of altering emission data precision on compression efficiency and accuracy of simulations of the community multiscale air quality modelAerSett v1.0: a simple and straightforward model for the settling speed of big spherical atmospheric aerosolsThe High-resolution Intermediate Complexity Atmospheric Research (HICAR v1.0) Model Enables Fast Dynamic Downscaling to the Hectometer ScaleEnKF-based fusion of site-available machine learning air quality predictions from RFSML v1.0 and gridded chemical transport model forecasts from GEOS-Chem v13.1.0Optimization of weather forecasting for cloud cover over the European domain using the meteorological component of the Ensemble for Stochastic Integration of Atmospheric Simulations version 1.0
Simone Dietmüller, Sigrun Matthes, Katrin Dahlmann, Hiroshi Yamashita, Abolfazl Simorgh, Manuel Soler, Florian Linke, Benjamin Lührs, Maximilian M. Meuser, Christian Weder, Volker Grewe, Feijia Yin, and Federica Castino
Geosci. Model Dev., 16, 4405–4425,Short summary
Climate-optimized aircraft trajectories avoid atmospheric regions with a large climate impact due to aviation emissions. This requires spatially and temporally resolved information on aviation's climate impact. We propose using algorithmic climate change functions (aCCFs) for CO2 and non-CO2 effects (ozone, methane, water vapor, contrail cirrus). Merged aCCFs combine individual aCCFs by assuming aircraft-specific parameters and climate metrics. Technically this is done with a Python library.
Andreas A. Beckert, Lea Eisenstein, Annika Oertel, Tim Hewson, George C. Craig, and Marc Rautenhaus
Geosci. Model Dev., 16, 4427–4450,Short summary
We investigate the benefit of objective 3-D front detection with modern interactive visual analysis techniques for case studies of extra-tropical cyclones and comparisons of frontal structures between different numerical weather prediction models. The 3-D frontal structures show agreement with 2-D fronts from surface analysis charts and augment them in the vertical dimension. We see great potential for more complex studies of atmospheric dynamics and for operational weather forecasting.
Zhenxin Liu, Yuanhao Chen, Yuhang Wang, Cheng Liu, Shuhua Liu, and Hong Liao
Geosci. Model Dev., 16, 4385–4403,Short summary
The heterogeneous layout of urban buildings leads to the complex wind field in and over the urban canopy. Large discrepancies between the observations and the current simulations result from misunderstanding the character of the wind field. The Inhomogeneous Wind Scheme in Urban Street (IWSUS) was developed to simulate the heterogeneity of the wind speed in a typical street and then improve the simulated energy budget in the lower atmospheric layer over the urban canopy.
Kai Cao, Qizhong Wu, Lingling Wang, Nan Wang, Huaqiong Cheng, Xiao Tang, Dongqing Li, and Lanning Wang
Geosci. Model Dev., 16, 4367–4383,Short summary
Offline performance experiment results show that the GPU-HADVPPM on a V100 GPU can achieve up to 1113.6 × speedups to its original version on an E5-2682 v4 CPU. A series of optimization measures are taken, and the CAMx-CUDA model improves the computing efficiency by 128.4 × on a single V100 GPU card. A parallel architecture with an MPI plus CUDA hybrid paradigm is presented, and it can achieve up to 4.5 × speedup when launching eight CPU cores and eight GPU cards.
Geosci. Model Dev., 16, 4265–4281,Short summary
This study analyzes forecasts that were made in 2021 to help trigger measurements during the CADDIWA experiment. The WRF and CHIMERE models were run each day, and the first goal is to quantify the variability of the forecast as a function of forecast leads and forecast location. The possibility of using the different leads as an ensemble is also tested. For some locations, the correlation scores are better with this approach. This could be tested on operational forecast chains in the future.
Emily de Jong, John Ben Mackay, Oleksii Bulenok, Anna Jaruga, and Sylwester Arabas
Geosci. Model Dev., 16, 4193–4211,Short summary
In clouds, collisional breakup occurs when two colliding droplets splinter into new, smaller fragments. Particle-based modeling approaches often do not represent breakup because of the computational demands of creating new droplets. We present a particle-based breakup method that preserves the computational efficiency of these methods. In a series of simple demonstrations, we show that this representation alters cloud processes in reasonable and expected ways.
Cyril Caram, Sophie Szopa, Anne Cozic, Slimane Bekki, Carlos A. Cuevas, and Alfonso Saiz-Lopez
Geosci. Model Dev., 16, 4041–4062,Short summary
We studied the role of halogenated compounds (containing chlorine, bromine and iodine), emitted by natural processes (mainly above the oceans), in the chemistry of the lower layers of the atmosphere. We introduced this relatively new chemistry in a three-dimensional climate–chemistry model and looked at how this chemistry will disrupt the ozone. We showed that the concentration of ozone decreases by 22 % worldwide and that of the atmospheric detergent, OH, by 8 %.
Joffrey Dumont Le Brazidec, Pierre Vanderbecken, Alban Farchi, Marc Bocquet, Jinghui Lian, Grégoire Broquet, Gerrit Kuhlmann, Alexandre Danjou, and Thomas Lauvaux
Geosci. Model Dev., 16, 3997–4016,Short summary
Monitoring of CO2 emissions is key to the development of reduction policies. Local emissions, from cities or power plants, may be estimated from CO2 plumes detected in satellite images. CO2 plumes generally have a weak signal and are partially concealed by highly variable background concentrations and instrument errors, which hampers their detection. To address this problem, we propose and apply deep learning methods to detect the contour of a plume in simulated CO2 satellite images.
Min-Seop Ahn, Paul A. Ullrich, Peter J. Gleckler, Jiwoo Lee, Ana C. Ordonez, and Angeline G. Pendergrass
Geosci. Model Dev., 16, 3927–3951,Short summary
We introduce a framework for regional-scale evaluation of simulated precipitation distributions with 62 climate reference regions and 10 metrics and apply it to evaluate CMIP5 and CMIP6 models against multiple satellite-based precipitation products. The common model biases identified in this study are mainly associated with the overestimated light precipitation and underestimated heavy precipitation. These biases persist from earlier-generation models and have been slightly improved in CMIP6.
Christine Wiedinmyer, Yosuke Kimura, Elena C. McDonald-Buller, Louisa K. Emmons, Rebecca R. Buchholz, Wenfu Tang, Keenan Seto, Maxwell B. Joseph, Kelley C. Barsanti, Annmarie G. Carlton, and Robert Yokelson
Geosci. Model Dev., 16, 3873–3891,Short summary
The Fire INventory from NCAR (FINN) provides daily global estimates of emissions from open fires based on satellite detections of hot spots. This version has been updated to apply MODIS and VIIRS satellite fire detection and better represents both large and small fires. FINNv2.5 generates more emissions than FINNv1 and is in general agreement with other fire emissions inventories. The new estimates are consistent with satellite observations, but uncertainties remain regionally and by pollutant.
Lichao Yang, Wansuo Duan, and Zifa Wang
Geosci. Model Dev., 16, 3827–3848,Short summary
An approach is proposed to refine a ground meteorological observation network to improve the PM2.5 forecasts in the Beijing–Tianjin–Hebei region. A cost-effective observation network is obtained and makes the relevant PM2.5 forecasts assimilate fewer observations but achieve the forecasting skill comparable to or higher than that obtained by assimilating all ground station observations, suggesting that many of the current ground stations can be greatly scattered to avoid much unnecessary work.
Abhishekh Kumar Srivastava, Paul Aaron Ullrich, Deeksha Rastogi, Pouya Vahmani, Andrew Jones, and Richard Grotjahn
Geosci. Model Dev., 16, 3699–3722,Short summary
Stakeholders need high-resolution regional climate data for applications such as assessing water availability and mountain snowpack. This study examines 3 h and 24 h historical precipitation over the contiguous United States in the 12 km WRF version 4.2.1-based dynamical downscaling of the ERA5 reanalysis. WRF improves precipitation characteristics such as the annual cycle and distribution of the precipitation maxima, but it also displays regionally and seasonally varying precipitation biases.
Haixia Xiao, Yaqiang Wang, Yu Zheng, Yuanyuan Zheng, Xiaoran Zhuang, Hongyan Wang, and Mei Gao
Geosci. Model Dev., 16, 3611–3628,Short summary
Due to the small-scale and nonstationary nature of convective wind gusts (CGs), reliable CG nowcasting has remained unattainable. Here, we developed a deep learning model — namely CGsNet — for 0—2 h of quantitative CG nowcasting, first achieving minute—kilometer-level forecasts. Based on the CGsNet model, the average surface wind speed (ASWS) and peak wind gust speed (PWGS) predictions are obtained. Experiments indicate that CGsNet exhibits higher accuracy than the traditional method.
Maria Krutova, Mostafa Bakhoday-Paskyabi, Joachim Reuder, and Finn Gunnar Nielsen
Geosci. Model Dev., 16, 3553–3564,Short summary
Local refinement of the grid is a powerful method allowing us to reduce the computational time while preserving the accuracy in the area of interest. Depending on the implementation, the local refinement may introduce unwanted numerical effects into the results. We study the wind speed common to the wind turbine operational speeds and confirm strong alteration of the result when the heat fluxes are present, except for the specific refinement scheme used.
Sylvia Sullivan, Behrooz Keshtgar, Nicole Albern, Elzina Bala, Christoph Braun, Anubhav Choudhary, Johannes Hörner, Hilke Lentink, Georgios Papavasileiou, and Aiko Voigt
Geosci. Model Dev., 16, 3535–3551,Short summary
Clouds absorb and re-emit infrared radiation from Earth's surface and absorb and reflect incoming solar radiation. As a result, they change atmospheric temperature gradients that drive large-scale circulation. To better simulate this circulation, we study how the radiative heating and cooling from clouds depends on model settings like grid spacing; whether we describe convection approximately or exactly; and the level of detail used to describe small-scale processes, or microphysics, in clouds.
James Weber, James A. King, Katerina Sindelarova, and Maria Val Martin
Geosci. Model Dev., 16, 3083–3101,Short summary
The emissions of volatile organic compounds from vegetation (BVOCs) influence atmospheric composition and contribute to certain gases and aerosols (tiny airborne particles) which play a role in climate change. BVOC emissions are likely to change in the future due to changes in climate and land use. Therefore, accurate simulation of BVOC emission is important, and this study describes an update to the simulation of BVOC emissions in the United Kingdom Earth System Model (UKESM).
Koichi Sakaguchi, L. Ruby Leung, Colin M. Zarzycki, Jihyeon Jang, Seth McGinnis, Bryce E. Harrop, William C. Skamarock, Andrew Gettelman, Chun Zhao, William J. Gutowski, Stephen Leak, and Linda Mearns
Geosci. Model Dev., 16, 3029–3081,Short summary
We document details of the regional climate downscaling dataset produced by a global variable-resolution model. The experiment is unique in that it follows a standard protocol designed for coordinated experiments of regional models. We found negligible influence of post-processing on statistical analysis, importance of simulation quality outside of the target region, and computational challenges that our model code faced due to rapidly changing super computer systems.
Xiaohan Li, Yi Zhang, Xindong Peng, Baiquan Zhou, Jian Li, and Yiming Wang
Geosci. Model Dev., 16, 2975–2993,Short summary
The weather and climate physics suites used in GRIST-A22.7.28 are compared using single-column modeling. The source of their discrepancies in terms of modeling cloud and precipitation is explored. Convective parameterization is found to be a key factor responsible for the differences. The two suites also have intrinsic differences in the interaction between microphysics and other processes, resulting in different cloud features and time step sensitivities.
Virginie Marécal, Ronan Voisin-Plessis, Tjarda Jane Roberts, Alessandro Aiuppa, Herizo Narivelo, Paul David Hamer, Béatrice Josse, Jonathan Guth, Luke Surl, and Lisa Grellier
Geosci. Model Dev., 16, 2873–2898,Short summary
We implemented a halogen volcanic chemistry scheme in a one-dimensional modelling framework preparing for further use in a three-dimensional global chemistry-transport model. The results of the simulations for an eruption of Mt Etna in 2008, including various sensitivity tests, show a good consistency with previous modelling studies.
Zhe Feng, Joseph Hardin, Hannah C. Barnes, Jianfeng Li, L. Ruby Leung, Adam Varble, and Zhixiao Zhang
Geosci. Model Dev., 16, 2753–2776,Short summary
PyFLEXTRKR is a flexible atmospheric feature tracking framework with specific capabilities to track convective clouds from a variety of observations and model simulations. The package has a collection of multi-object identification algorithms and has been optimized for large datasets. This paper describes the algorithms and demonstrates applications for tracking deep convective cells and mesoscale convective systems from observations and model simulations at a wide range of scales.
Yan Ji, Bing Gong, Michael Langguth, Amirpasha Mozaffari, and Xiefei Zhi
Geosci. Model Dev., 16, 2737–2752,Short summary
Formulating short-term precipitation forecasting as a video prediction task, a novel deep learning architecture (convolutional long short-term memory generative adversarial network, CLGAN) is proposed. A benchmark dataset is built on minute-level precipitation measurements. Results show that with the GAN component the model generates predictions sharing statistical properties with observations, resulting in it outperforming the baseline in dichotomous and spatial scores for heavy precipitation.
Aleksander Lacima, Hervé Petetin, Albert Soret, Dene Bowdalo, Oriol Jorba, Zhaoyue Chen, Raúl F. Méndez Turrubiates, Hicham Achebak, Joan Ballester, and Carlos Pérez García-Pando
Geosci. Model Dev., 16, 2689–2718,Short summary
Understanding how air pollution varies across space and time is of key importance for the safeguarding of human health. This work arose in the context of the project EARLY-ADAPT, for which the Barcelona Supercomputing Center developed an air pollution database covering all of Europe. Through different statistical methods, we compared two global pollution models against measurements from ground stations and found significant discrepancies between the observed and the modeled surface pollution.
Andrew Geiss, Po-Lun Ma, Balwinder Singh, and Joseph C. Hardin
Geosci. Model Dev., 16, 2355–2370,Short summary
Atmospheric aerosols play a critical role in Earth's climate, but it is too computationally expensive to directly model their interaction with radiation in climate simulations. This work develops a new neural-network-based parameterization of aerosol optical properties for use in the Energy Exascale Earth System Model that is much more accurate than the current one; it also introduces a unique model optimization method that involves randomly generating neural network architectures.
Joey C. Y. Lam, Amos P. K. Tai, Jason A. Ducker, and Christopher D. Holmes
Geosci. Model Dev., 16, 2323–2342,Short summary
We developed a new component within an atmospheric chemistry model to better simulate plant ecophysiological processes relevant for ozone air quality. We showed that it reduces simulated biases in plant uptake of ozone in prior models. The new model enables us to explore how future climatic changes affect air quality via affecting plants, examine ozone–vegetation interactions and feedbacks, and evaluate the impacts of changing atmospheric chemistry and climate on vegetation productivity.
Qian Shu, Sergey L. Napelenok, William T. Hutzell, Kirk R. Baker, Barron H. Henderson, Benjamin N. Murphy, and Christian Hogrefe
Geosci. Model Dev., 16, 2303–2322,Short summary
Source attribution methods are generally used to determine culpability of precursor emission sources to ambient pollutant concentrations. However, source attribution of secondarily formed pollutants such as ozone and its precursors cannot be explicitly measured, making evaluation of source apportionment methods challenging. In this study, multiple apportionment approach comparisons show common features but still reveal wide variations in predicted sector contribution and species dependency.
Rüdiger Brecht, Lucie Bakels, Alex Bihlo, and Andreas Stohl
Geosci. Model Dev., 16, 2181–2192,Short summary
We use neural-network-based single-image super-resolution to improve the upscaling of meteorological wind fields to be used for particle dispersion models. This deep-learning-based methodology improves the standard linear interpolation typically used in particle dispersion models. The improvement of wind fields leads to substantial improvement in the computed trajectories of the particles.
Alvaro Criado, Jan Mateu Armengol, Hervé Petetin, Daniel Rodriguez-Rey, Jaime Benavides, Marc Guevara, Carlos Pérez García-Pando, Albert Soret, and Oriol Jorba
Geosci. Model Dev., 16, 2193–2213,Short summary
This work aims to derive and evaluate a general statistical post-processing tool specifically designed for the street scale that can be applied to any urban air quality system. Our data fusion methodology corrects NO2 fields based on continuous hourly observations and experimental campaigns. This study enables us to obtain exceedance probability maps of air quality standards. In 2019, 13 % of the Barcelona area had a 70 % or higher probability of exceeding the annual legal NO2 limit of 40 µg/m3.
Liang Wang, Bingcheng Wan, Shaohui Zhou, Haofei Sun, and Zhiqiu Gao
Geosci. Model Dev., 16, 2167–2179,Short summary
The past 24 h TC trajectories and meteorological field data were used to forecast TC tracks in the northwestern Pacific from hours 6–72 based on GRU_CNN, which we proposed in this paper and which has better prediction results than traditional single deep-learning methods. The historical steering flow of cyclones has a significant effect on improving the accuracy of short-term forecasting, while, in long-term forecasting, the SST and geopotential height will have a particular impact.
Drew C. Pendergrass, Daniel J. Jacob, Hannah Nesser, Daniel J. Varon, Melissa Sulprizio, Kazuyuki Miyazaki, and Kevin W. Bowman
We have built a tool called CHEEREIO that allows scientists to use observations of pollutants or gases in the atmosphere, such as from satellites or surface stations, to update supercomputer models that simulate the Earth. CHEEREIO uses the difference between the model simulations of the atmosphere and real-world observations to come up with a good guess for the actual composition of our atmosphere, the true emissions of various pollutants, and whatever else they may want to study.
Thomas Berkemeier, Matteo Krüger, Aryeh Feinberg, Marcel Müller, Ulrich Pöschl, and Ulrich K. Krieger
Geosci. Model Dev., 16, 2037–2054,Short summary
Kinetic multi-layer models (KMs) successfully describe heterogeneous and multiphase atmospheric chemistry. In applications requiring repeated execution, however, these models can be too expensive. We trained machine learning surrogate models on output of the model KM-SUB and achieved high correlations. The surrogate models run orders of magnitude faster, which suggests potential applicability in global optimization tasks and as sub-modules in large-scale atmospheric models.
Elena Fillola, Raul Santos-Rodriguez, Alistair Manning, Simon O'Doherty, and Matt Rigby
Geosci. Model Dev., 16, 1997–2009,Short summary
Lagrangian particle dispersion models are used extensively for the estimation of greenhouse gas (GHG) fluxes using atmospheric observations. However, these models do not scale well as data volumes increase. Here, we develop a proof-of-concept machine learning emulator that can produce outputs similar to those of the dispersion model, but 50 000 times faster, using only meteorological inputs. This works demonstrates the potential of machine learning to accelerate GHG estimations across the globe.
Maria J. Chinita, Mikael Witte, Marcin J. Kurowski, Joao Teixeira, Kay Suselj, Georgios Matheou, and Peter Bogenschutz
Geosci. Model Dev., 16, 1909–1924,Short summary
Low clouds are one of the largest sources of uncertainty in climate prediction. In this paper, we introduce the first version of the unified turbulence and shallow convection parameterization named SHOC+MF developed to improve the representation of shallow cumulus clouds in the Simple Cloud-Resolving E3SM Atmosphere Model (SCREAM). Here, we also show promising preliminary results in a single-column model framework for two benchmark cases of shallow cumulus convection.
Kun Wang, Chao Gao, Kai Wu, Kaiyun Liu, Haofan Wang, Mo Dan, Xiaohui Ji, and Qingqing Tong
Geosci. Model Dev., 16, 1961–1973,Short summary
This study establishes an easy-to-use and integrated framework for a model-ready emission inventory for the Weather Research and Forecasting (WRF)–Air Quality Numerical Model (AQM). A free tool called the ISAT (Inventory Spatial Allocation Tool) was developed based on this framework. ISAT helps users complete the workflow from the WRF nested-domain configuration to a model-ready emission inventory for AQM with a regional emission inventory and a shapefile for the target region.
Jagat S. H. Bisht, Prabir K. Patra, Masayuki Takigawa, Takashi Sekiya, Yugo Kanaya, Naoko Saitoh, and Kazuyuki Miyazaki
Geosci. Model Dev., 16, 1823–1838,Short summary
In this study, we estimated CH4 fluxes using an advanced 4D-LETKF method. The system was tested and optimized using observation system simulation experiments (OSSEs), where a known surface emission distribution is retrieved from synthetic observations. The availability of satellite measurements has increased, and there are still many missions focused on greenhouse gas observations that have not yet launched. The technique being referred to has the potential to improve estimates of CH4 fluxes.
Ruizi Shi and Fanghua Xu
Geosci. Model Dev., 16, 1839–1856,Short summary
Based on the Gaussian quadrature method, a fast algorithm of sea-spray-mediated heat flux is developed. Compared with the widely used single-radius algorithm, the new fast algorithm shows a better agreement with the full spectrum integral of spray flux. The new fast algorithm is evaluated in a coupled modeling system, and the simulations of sea surface temperature, wind speed and wave height are improved. Thereby, the new fast algorithm has great potential to be used in coupled modeling systems.
Forwood Wiser, Bryan K. Place, Siddhartha Sen, Havala O. T. Pye, Benjamin Yang, Daniel M. Westervelt, Daven K. Henze, Arlene M. Fiore, and V. Faye McNeill
Geosci. Model Dev., 16, 1801–1821,Short summary
We developed a reduced model of atmospheric isoprene oxidation, AMORE-Isoprene 1.0. It was created using a new Automated Model Reduction (AMORE) method designed to simplify complex chemical mechanisms with minimal manual adjustments to the output. AMORE-Isoprene 1.0 has improved accuracy and similar size to other reduced isoprene mechanisms. When included in the CRACMM mechanism, it improved the accuracy of EPA’s CMAQ model predictions for the northeastern USA compared to observations.
Jonathan D. Labriola, Jeremy A. Gibbs, and Louis J. Wicker
Geosci. Model Dev., 16, 1779–1799,Short summary
Observing system simulation experiments (OSSEs) are simulated case studies used to understand how different assimilated weather observations impact forecast skill. This study introduces the methods used to create an OSSE for a tornadic quasi-linear convective system event. These steps provide an opportunity to simulate a realistic high-impact weather event and can be used to encourage a more diverse set of OSSEs.
Mike Bush, Ian Boutle, John Edwards, Anke Finnenkoetter, Charmaine Franklin, Kirsty Hanley, Aravindakshan Jayakumar, Huw Lewis, Adrian Lock, Marion Mittermaier, Saji Mohandas, Rachel North, Aurore Porson, Belinda Roux, Stuart Webster, and Mark Weeks
Geosci. Model Dev., 16, 1713–1734,Short summary
Building on the baseline of RAL1, the RAL2 science configuration is used for regional modelling around the UM partnership and in operations at the Met Office. RAL2 has been tested in different parts of the world including Australia, India and the UK. RAL2 increases medium and low cloud amounts in the mid-latitudes compared to RAL1, leading to improved cloud forecasts and a reduced diurnal cycle of screen temperature. There is also a reduction in the frequency of heavier precipitation rates.
Christoph Neuhauser, Maicon Hieronymus, Michael Kern, Marc Rautenhaus, Annika Oertel, and Rüdiger Westermann
Geosci. Model Dev. Discuss.,
Revised manuscript accepted for GMDShort summary
Numerical weather prediction models rely on parameterizations for sub-grid scale processes, which represent a source of uncertainty. We present novel visual analytics solutions to analyze interactively the sensitivities of a selected prognostic variable to multiple model parameters along trajectories wrt. similarities in temporal development and spatio-temporal relationships. The proposed workflow is applied to cloud microphysical sensitivities along coherent strongly ascending trajectories.
Chuanhua Ren, Xin Huang, Tengyu Liu, Yu Song, Zhang Wen, Xuejun Liu, Aijun Ding, and Tong Zhu
Geosci. Model Dev., 16, 1641–1659,Short summary
Ammonia in the atmosphere has wide impacts on the ecological environment and air quality, and its emission from soil volatilization is highly sensitive to meteorology, making it challenging to be well captured in models. We developed a dynamic emission model capable of calculating ammonia emission interactively with meteorological and soil conditions. Such a coupling of soil emission with meteorology provides a better understanding of ammonia emission and its contribution to atmospheric aerosol.
Linlu Mei, Vladimir Rozanov, Alexei Rozanov, and John P. Burrows
Geosci. Model Dev., 16, 1511–1536,Short summary
This paper summarizes recent developments of aerosol, cloud and surface reflectance databases and models in the framework of the software package SCIATRAN. These updates and developments extend the capabilities of the radiative transfer modeling, especially by accounting for different kinds of vertical inhomogeneties. Vertically inhomogeneous clouds and different aerosol types can be easily accounted for within SCIATRAN (V4.6). The widely used surface models and databases are now available.
Adrian Rojas-Campos, Michael Langguth, Martin Wittenbrink, and Gordon Pipa
Geosci. Model Dev., 16, 1467–1480,Short summary
Our paper presents an alternative approach for generating high-resolution precipitation maps based on the nonlinear combination of the complete set of variables of the numerical weather predictions. This process combines the super-resolution task with the bias correction in a single step, generating high-resolution corrected precipitation maps with a lead time of 3 h. We used using deep learning algorithms to combine the input information and increase the accuracy of the precipitation maps.
Robin N. Thor, Mariano Mertens, Sigrun Matthes, Mattia Righi, Johannes Hendricks, Sabine Brinkop, Phoebe Graf, Volker Grewe, Patrick Jöckel, and Steven Smith
Geosci. Model Dev., 16, 1459–1466,Short summary
We report on an inconsistency in the latitudinal distribution of aviation emissions between two versions of a data product which is widely used by researchers. From the available documentation, we do not expect such an inconsistency. We run a chemistry–climate model to compute the effect of the inconsistency in emissions on atmospheric chemistry and radiation and find that the radiative forcing associated with aviation ozone is 7.6 % higher when using the less recent version of the data.
Stefano Della Fera, Federico Fabiano, Piera Raspollini, Marco Ridolfi, Ugo Cortesi, Flavio Barbara, and Jost von Hardenberg
Geosci. Model Dev., 16, 1379–1394,Short summary
The long-term comparison between observed and simulated outgoing longwave radiances represents a strict test to evaluate climate model performance. In this work, 9 years of synthetic spectrally resolved radiances, simulated online on the basis of the atmospheric fields predicted by the EC-Earth global climate model (v3.3.3) in clear-sky conditions, are compared to IASI spectral radiance climatology in order to detect model biases in temperature and humidity at different atmospheric levels.
Marine Bonazzola, Hélène Chepfer, Po-Lun Ma, Johannes Quaas, David M. Winker, Artem Feofilov, and Nick Schutgens
Geosci. Model Dev., 16, 1359–1377,Short summary
Aerosol has a large impact on climate. Using a lidar aerosol simulator ensures consistent comparisons between modeled and observed aerosol. We present a lidar aerosol simulator that applies a cloud masking and an aerosol detection threshold. We estimate the lidar signals that would be observed at 532 nm by the Cloud-Aerosol Lidar with Orthogonal Polarization overflying the atmosphere predicted by a climate model. Our comparison at the seasonal timescale shows a discrepancy in the Southern Ocean.
Michael S. Walters and David C. Wong
Geosci. Model Dev., 16, 1179–1190,Short summary
A typical numerical simulation that associates with a large amount of input and output data, applying popular compression software, gzip or bzip2, on data is one good way to mitigate data storage burden. This article proposes a simple technique to alter input, output, or input and output by keeping a specific number of significant digits in data and demonstrates an enhancement in compression efficiency on the altered data but maintains similar statistical performance of the numerical simulation.
Sylvain Mailler, Laurent Menut, Arineh Cholakian, and Romain Pennel
Geosci. Model Dev., 16, 1119–1127,Short summary
Large or even
giantparticles of mineral dust exist in the atmosphere but, so far, solving an non-linear equation was needed to calculate the speed at which they fall in the atmosphere. The model we present, AerSett v1.0 (AERosol SETTling version 1.0), provides a new and simple way of calculating their free-fall velocity in the atmosphere, which will be useful to anyone trying to understand and represent adequately the transport of giant dust particles by the wind.
Dylan Stewart Reynolds, Ethan Gutmann, Bert Kruyt, Michael Haugeneder, Tobias Jonas, Franziska Gerber, Michael Lehning, and Rebecca Mott
Geosci. Model Dev. Discuss.,
Revised manuscript accepted for GMDShort summary
The challenge of running geophysical models is often compounded by a question of where to obtain appropriate data to give as input to a model. Here we present the HICAR model, which is a simplified atmospheric model capable of running at spatial resolutions of hectometers for long time series or over large domains. This makes physically consistent atmospheric data available at spatial and temporal scales needed for some terrestrial modeling applications, for example seasonal snow forecasting.
Li Fang, Jianbing Jin, Arjo Segers, Ke Li, Bufan Xu, Wei Han, Mijie Pang, Hai Xiang Lin, and Hong Liao
Geosci. Model Dev. Discuss.,
Revised manuscript accepted for GMDShort summary
Machine learning model have gained great popularity in air quality prediction. However, they are only available at the air quality monitoring stations. In contrast, chemical transport models (CTM) provide forecasts that are continuous in 3D field. Owing to complex error sources, they are typically biased. In this study, we proposed a gridded prediction with high accuracy by fusing predictions from our regional-feature-selection machine learning prediction (RFSML v1.0) and a CTM forecast.
Yen-Sen Lu, Garrett H. Good, and Hendrik Elbern
Geosci. Model Dev., 16, 1083–1104,Short summary
The Weather Forecasting and Research (WRF) model consists of many parameters and options that can be adapted to different conditions. This expansive sensitivity study uses a large-scale simulation system to determine the most suitable options for predicting cloud cover in Europe for deterministic and probabilistic weather predictions for day-ahead forecasting simulations.
Belgiu, M. and Drăguţ, L.: Random forest in remote sensing: A review of applications and future directions, ISPRS J. Photogramm., 114, 24–31, https://doi.org/10.1016/j.isprsjprs.2016.01.011, 2016.
Bowe, B., Xie, Y., Li, T., Yan, Y., Xian, H., and Al-Aly, Z.: The 2016 global and national burden of diabetes mellitus attributable to PM2.5 air pollution, Lancet Planet. Health, 2, e301–e312, https://doi.org/10.1016/S2542-5196(18)30140-2, 2018.
Chen, X., de Leeuw, G., Arola, A., Liu, S., Liu, Y., Li, Z., and Zhang, K.: Joint retrieval of the aerosol fine mode fraction and optical depth using MODIS spectral reflectance over northern and eastern China: Artificial neural network method, Remote Sens. Environ., 249, 112006, https://doi.org/10.1016/j.rse.2020.112006, 2020.
Friedman, J. H.: Greedy function approximation: a gradient boosting machine, Ann. Stat., 29, 1189–1232, 2001.
Gao, J., Zhou, Y., Wang, J., Wang, T., and Wang, W. X.: Inter-comparison of WPSTM-TEOMTM-MOUDITM and investigation on particle density, Huan Jing Ke Xue, 28, 1929–1934, https://doi.org/10.3321/j.issn:0250-3301.2007.09.005, 2007.
Gao, L., Li, J., Chen, L., Zhang, L., and Heidinger, A. K.: Retrieval and validation of atmospheric aerosol optical depth from AVHRR over China, IEEE T. Geosci. Remote, 54, 6280–6291, https://doi.org/10.1109/TGRS.2016.2574756, 2016.
Geng, G., Zhang, Q., Martin, R. V., van Donkelaar, A., Huo, H., Che, H., Lin, J., and He, K.: Estimating long-term PM2.5 concentrations in China using satellite-based aerosol optical depth and a chemical transport model, Remote Sens. Environ., 166, 262–270, https://doi.org/10.1016/j.rse.2015.05.016, 2015.
Geurts, P., Ernst, D., and Wehenkel, L.: Extremely randomized trees, Mach. Learn., 63, 3–42, https://doi.org/10.1007/s10994-006-6226-1, 2006.
Giles, D. M., Holben, B. N., Eck, T. F., Smirnov, A., Sinyuk, A., Schafer, J., Sorokin, M. G., and Slutsker, I.: Aerosol robotic network (AERONET) version 3 aerosol optical depth and inversion products, in: American Geophysical Union (AGU) 98th Fall Meeting Abstracts, New Orleans, America, 11–15 December 2017, A11O-01, 2017AGUFM.A11O..01G, 2017.
Giles, D. M., Sinyuk, A., Sorokin, M. G., Schafer, J. S., Smirnov, A., Slutsker, I., Eck, T. F., Holben, B. N., Lewis, J. R., Campbell, J. R., Welton, E. J., Korkin, S. V., and Lyapustin, A. I.: Advancements in the Aerosol Robotic Network (AERONET) Version 3 database – automated near-real-time quality control algorithm with improved cloud screening for Sun photometer aerosol optical depth (AOD) measurements, Atmos. Meas. Tech., 12, 169–209, https://doi.org/10.5194/amt-12-169-2019, 2019 (data available at: https://aeronet.gsfc.nasa.gov/, last access: 30 September 2022).
Gupta, P. and Christopher, S. A.: Particulate matter air quality assessment using integrated surface, satellite, and meteorological products: Multiple regression approach, J. Geophys. Res.-Atmos., 114, D14205, https://doi.org/10.1029/2008JD011496, 2009.
Hand, J. L. and Kreidenweis, S. M.: A new method for retrieving particle refractive index and effective density from aerosol size distribution data, Aerosol Sci. Technol., 36, 1012–1026, https://doi.org/10.1080/02786820290092276, 2002.
Hänel, G. and Thudium, J.: Mean bulk densities of samples of dry atmospheric aerosol particles: A summary of measured data, Pure Appl. Geophys., 115, 799–803, https://doi.org/10.1007/BF00881211, 1977.
He, J., Yuan, Q., Li, J., and Zhang, L.: PoNet: A universal physical optimization-based spectral super-resolution network for arbitrary multispectral images, Inform. Fusion, 80, 205–225, https://doi.org/10.1016/j.inffus.2021.10.016, 2022.
He, J., Li, J., Yuan, Q., Shen, H., and Zhang, L.: Spectral Response Function-Guided Deep Optimization-Driven Network for Spectral Super-Resolution, IEEE T. Neur. Net. Lear., 99, 1–15, https://doi.org/10.1109/TNNLS.2021.3056181, 2021.
Ho, T.: Random decision forests, in: Proceedings of 3rd International Conference on Document Analysis and Recognition, Montreal, QC, Canada, 14–16 August 1995, 278–282, https://doi.org/10.1109/ICDAR.1995.598994, 1995.
Hersbach, H., Bell, B., Berrisford, P., Biavati, G., Horányi, A., Muñoz Sabater, J., Nicolas, J., Peubey, C., Radu, R., Rozum, I., Schepers, D., Simmons, A., Soci, C., Dee, D., and Thépaut, J.-N.: ERA5 hourly data on single levels from 1979 to present, Copernicus Climate Change Service (C3S) Climate Data Store (CDS) [data set], https://doi.org/10.24381/cds.adbb2d47, 2018.
Holben, B. N., Eck, T. F., Slutsker, I., Tanré, D., Buis, J. P., Setzer, A., Vermote, E., Reagan, J. A., Kaufman, Y. J., Nakajima, T., Lavenu, F., Jankowiak, I., and Smirnov, A.: AERONET–A federated instrument network and data archive for aerosol characterization, Remote Sens. Environ., 66, 1–16, https://doi.org/10.1016/S0034-4257(98)00031-5, 1998.
Irrgang, C., Boers, N., Sonnewald, M., Barnes, E. A., Kadow, C., Staneva, J., and Saynisch-Wagner, J.: Towards neural Earth system modelling by integrating artificial intelligence in Earth system science, Nat. Mach. Intell., 3, 667–674, https://doi.org/10.1038/s42256-021-00374-3, 2021.
Jin, C.: An optimized semi-empirical physical approach for satellite-based PM2.5 retrieval: using random forest model to simulate the complex parameter, Zenodo [code], https://doi.org/10.5281/zenodo.7183822, 2022.
Koelemeijer, R. B. A., Homan, C. D., and Matthijsen, J.: Comparison of spatial and temporal variations of aerosol optical thickness and particulate matter over Europe, Atmos. Environ., 40, 5304–5315, https://doi.org/10.1016/j.atmosenv.2006.04.044, 2006.
Kokhanovsky, A. A., Prikhach, A. S., Katsev, I. L., and Zege, E. P.: Determination of particulate matter vertical columns using satellite observations, Atmos. Meas. Tech., 2, 327–335, https://doi.org/10.5194/amt-2-327-2009, 2009.
Lee, J.-B., Lee, J.-B., Koo, Y.-S., Kwon, H.-Y., Choi, M.-H., Park, H.-J., and Lee, D.-G.: Development of a deep neural network for predicting 6 h average PM2.5 concentrations up to 2 subsequent days using various training data, Geosci. Model Dev., 15, 3797–3813, https://doi.org/10.5194/gmd-15-3797-2022, 2022.
Li, T., Shen, H., Zeng, C., Yuan, Q., and Zhang, L.: Point-surface fusion of station measurements and satellite observations for mapping PM2.5 distribution in China: Methods and assessment, Atmos. Environ., 152, 477–489, https://doi.org/10.1016/j.atmosenv.2017.01.004, 2017.
Li, Z., Zhang, Y., Shao, J., Li, B., Hong, J., Liu, D., Li, D., Wei, P., Li, W., Li, L., Zhang, F., Guo, J., Deng, Q., Wang, B., Cui, C., Zhang, W., Wang, Z., Lv, Y., Xu, H., Chen, X., Li, L., and Qie, L.: Remote sensing of atmospheric particulate mass of dry PM2.5 near the ground: Method validation using ground-based measurements, Remote Sens. Environ., 173, 59–68, https://doi.org/10.1016/j.rse.2015.11.019, 2016.
Lyapustin, A., Wang, Y., Laszlo, I., Kahn, R., Korkin, S., Remer, L., Levy, R., and Reid, J. S.: Multiangle implementation of atmospheric correction (MAIAC): 2. Aerosol algorithm, J. Geophys. Res.-Atmos., 116, D03211, https://doi.org/10.1029/2010JD014986, 2011.
Lyapustin, A., Wang, Y., Xiong, X., Meister, G., Platnick, S., Levy, R., Franz, B., Korkin, S., Hilker, T., Tucker, J., Hall, F., Sellers, P., Wu, A., and Angal, A.: Scientific impact of MODIS C5 calibration degradation and C6+ improvements, Atmos. Meas. Tech., 7, 4353–4365, https://doi.org/10.5194/amt-7-4353-2014, 2014.
Lyapustin, A. and Wang, Y.: MCD19A2 MODIS/Terra+Aqua Aerosol Optical Thickness Daily L2G Global 1km SIN Grid, NASA LP DAAC [data set], https://doi.org/10.5067/MODIS/MCD19A2.006, 2015.
Lyu, B., Huang, R., Wang, X., Wang, W., and Hu, Y.: Deep-learning spatial principles from deterministic chemical transport models for chemical reanalysis: an application in China for PM2.5, Geosci. Model Dev., 15, 1583–1594, https://doi.org/10.5194/gmd-15-1583-2022, 2022.
Ma, Z., Hu, X., Huang, L., Bi, J., and Liu, Y.: Estimating ground-Level PM2.5 in China using satellite remote sensing, Environ. Sci. Technol., 48, 7436–7444, https://doi.org/10.1021/es5009399, 2014.
Pope III, C. A., Burnett, R. T., Thun, M. J., Calle, E. E., Krewski, D., Ito, K., and Thurston, G. D.: Lung cancer, cardiopulmonary mortality, and long-term exposure to fine particulate air pollution, JAMA, 287, 1132–1141, https://doi.org/10.1001/jama.287.9.1132, 2002.
Raut, J.-C. and Chazette, P.: Assessment of vertically-resolved PM10 from mobile lidar observations, Atmos. Chem. Phys., 9, 8617–8638, https://doi.org/10.5194/acp-9-8617-2009, 2009.
Rodriguez, J. D., Perez, A., and Lozano, J. A.: Sensitivity analysis of k-fold cross validation in prediction error estimation, IEEE T. Pattern Anal., 32, 569–575, https://doi.org/10.1109/TPAMI.2009.187, 2009.
Shi, X., Zhao, C., Jiang, J. H., Wang, C., Yang, X., and Yung, Y. L.: Spatial representativeness of PM2.5 concentrations obtained using observations from network stations, J. Geophys. Res.-Atmos., 123, 3145–3158, https://doi.org/10.1002/2017JD027913, 2018.
Simmons, A. J., Untch, A., Jakob, C., Kållberg, P., and Undén, P.: Stratospheric water vapour and tropical tropopause temperatures in ECMWF analyses and multi-year simulations, Q. J. Roy. Meteor. Soc., 125, 353–386, https://doi.org/10.1002/qj.49712555318, 1999.
Van Donkelaar, A., Martin, R. V., and Park, R. J.: Estimating ground-level PM2.5 using aerosol optical depth determined from satellite remote sensing, J. Geophys. Res.-Atmos., 111, D21201, https://doi.org/10.1029/2005JD006996, 2006.
Wang, Y., Yuan, Q., Li, T., Shen, H., Zheng, L., and Zhang, L.: Evaluation and comparison of MODIS Collection 6.1 aerosol optical depth against AERONET over regions in China with multifarious underlying surfaces, Atmos. Environ., 200, 280–301, https://doi.org/10.1016/j.atmosenv.2018.12.023, 2019.
Wu, X., Wang, Y., He, S., and Wu, Z.: ratio prediction based on a long short-term memory neural network in Wuhan, China, Geosci. Model Dev., 13, 1499–1511, https://doi.org/10.5194/gmd-13-1499-2020, 2020.
Xiao, Y., Wang, Y., Yuan, Q., He, J., and Zhang, L.: Generating a long-term (2003–2020) hourly 0.25∘ global PM2.5 dataset via spatiotemporal downscaling of CAMS with deep learning (DeepCAMS), Sci. Total Environ., 848, 157747, https://doi.org/10.1016/j.scitotenv.2022.157747, 2022.
Xu, P., Chen, Y., and Ye, X.: Haze, air pollution, and health in China, Lancet, 382, 2067, https://doi.org/10.1016/S0140-6736(13)62693-8, 2013.
Yan, X., Zang, Z., Li, Z., Luo, N., Zuo, C., Jiang, Y., Li, D., Guo, Y., Zhao, W., Shi, W., and Cribb, M.: A global land aerosol fine-mode fraction dataset (2001–2020) retrieved from MODIS using hybrid physical and deep learning approaches, Earth Syst. Sci. Data, 14, 1193–1213, https://doi.org/10.5194/essd-14-1193-2022, 2022.
Yan, X., Li, Z., Shi, W., Luo, N., Wu, T., and Zhao, W.: An improved algorithm for retrieving the fine-mode fraction of aerosol optical thickness, part 1: Algorithm development, Remote Sens. Environ., 192, 87–97, https://doi.org/10.1016/j.rse.2017.02.005, 2017.
Yan, X.: Physical and deep learning retrieved fine mode fraction (Phy-DL FMF), Zenodo [data set], https://doi.org/10.5281/zenodo.5105617, 2021.
Yang, Q., Yuan, Q., Li, T., and Yue, L.: Mapping PM2.5 concentration at high resolution using a cascade random forest based downscaling model: Evaluation and application, J. Clean. Prod., 277, 123887, https://doi.org/10.1016/j.jclepro.2020.123887, 2020.
Yuan, Q., Shen, H., Li, T., Li, Z., Li, S., Jiang, Y., Xu, H., Tan, W., Yang, Q., Wang, J., Gao, J., and Zhang, L.: Deep learning in environmental remote sensing: Achievements and challenges, Remote Sens. Environ., 241, 111716, https://doi.org/10.1016/j.rse.2020.111716, 2020.
Zhang, Y., Li, Z., Bai, K., Wei, Y., Xie, Y., Zhang, Y., Ou, Y., Cohen, J., Zhang, Y., Peng, Z., Zhang, X., Chen, C., Hong, J., Xu, H., Guang, J., Lv, Y., Li, K., and Li, D.: Satellite remote sensing of atmospheric particulate matter mass concentration: Advances, challenges, and perspectives, Fundamental Research, 1, 240–258, https://doi.org/10.1016/j.fmre.2021.04.007, 2021.
Zhang, Y., Li, Z., Chang, W., Zhang, Y., de Leeuw, G., and Schauer, J. J.: Satellite observations of PM2.5 changes and driving factors based forecasting over China 2000–2025, Remote Sens., 12, 2518, https://doi.org/10.3390/rs12162518, 2020.
Zhang, Y. and Li, Z.: Remote sensing of atmospheric fine particulate matter (PM2.5) mass concentration near the ground from satellite observation, Remote Sens. Environ., 160, 252–262, https://doi.org/10.1016/j.rse.2015.02.005, 2015.
The semi-empirical physical approach derives PM2.5 with strong physical significance. However, due to the complex optical characteristic, the physical parameters are difficult to express accurately. Thus, combining the atmospheric physical mechanism and machine learning, we propose an optimized model. It creatively embeds the random forest model into the physical PM2.5 remote sensing approach to simulate a physical parameter. Our method shows great optimized performance in the validations.
The semi-empirical physical approach derives PM2.5 with strong physical significance. However,...