Machine learning based bias correction for numerical chemical transport models

Xu, Min; Jin, Jianbing; Wang, Guoqiang; Segers, Arjo; Deng, Tuo; Lin, Hai

doi:10.1016/j.atmosenv.2020.118022

Cited by 16 publications

(5 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It could cause a high probability that the duration of future heatwaves in southwestern (northwestern) North China would be overestimated (underestimated). Therefore, it is necessary to reduce these CMIP model biases through various methods, such as bias correction (Yu et al, 2018; Zhang et al, 2020a), machine learning (Xu et al, 2021), dynamical downscaling (Liang et al, 2008) and global climate model upgrading.…”

Section: Discussionmentioning

confidence: 99%

Increasing heatwave with associated population and GDP exposure in North China

Zhang

Chen

et al. 2023

Intl Journal of Climatology

View full text Add to dashboard Cite

Twenty‐seven CMIP6 models were grouped into three ensembles based on the simulated performance of heatwaves in North China during present‐day (1995–2014), and future changes in the duration and intensity of heatwaves were projected under SSP1‐2.6, SSP2‐4.5 and SSP5‐8.5. The selected three ensembles showed consistent projections: both the duration and intensity of heatwaves would increase significantly, with the greatest under SSP5‐8.5. Besides, the heatwave growth in 2081–2100 would double in 2041–2060, except for SSP1‐2.6, where heatwaves would be similar in both periods. For the spatial distribution, the duration (intensity) would increase more in southern (western) parts of North China. Combining heatwaves with population and GDP, future heat exposures would concentrate on urban areas and the tertiary industry. For example, the 2041–2060 population exposure would reach 3.2–5.6 times the current level, with contributions from the urban population ranging from 55% to 60%. The GDP exposure would hit tens to hundreds of times the current level, with the tertiary sector replacing the secondary sector as the leading industry in North China, producing the major contribution and facing significant heat‐related risks. Overall, there will be significant heat‐related impacts under SSP5‐8.5, about 1.5–3.0 fold of those under SSP1‐2.6 and SSP2‐4.5. The urban and tertiary sectors would suffer greater risks relative to the rural and other industries. Our results reinforced the need to minimize global emissions and develop strategic plans to mitigate heat impacts under high‐emission scenarios, especially for urban areas and the tertiary industry, requiring great attention to climate adaptation.

show abstract

Section: Discussionmentioning

confidence: 99%

Increasing heatwave with associated population and GDP exposure in North China

Zhang

Chen

et al. 2023

Intl Journal of Climatology

View full text Add to dashboard Cite

show abstract

“…These methods are increasingly being used due to the progress in the development of ML hardware and technology. ML bias correction techniques are mainly used for chemistry transport model simulations of trace gases, e.g., [15], or used in situ surface data for aerosols, e.g., [16,17], or for post-processing weather forecasts, e.g., [18,19]. ML is used in the context of the monthly mean AOD estimation as well [20].…”

Section: Introductionmentioning

confidence: 99%

Machine Learning-Based Improvement of Aerosol Optical Depth from CHIMERE Simulations Using MODIS Satellite Observations

et al. 2023

View full text Add to dashboard Cite

We present a supervised machine learning (ML) approach to improve the accuracy of the regional horizontal distribution of the aerosol optical depth (AOD) simulated by the CHIMERE chemistry transport model over North Africa and the Arabian Peninsula using Moderate Resolution Imaging Spectroradiometer (MODIS) AOD satellite observations. Our method produces daily AOD maps with enhanced precision and full spatial domain coverage, which is particularly relevant for regions with a high aerosol abundance, such as the Sahara Desert, where there is a dramatic lack of ground-based measurements for validating chemistry transport simulations. We use satellite observations and some geophysical variables to train four popular regression models, namely multiple linear regression (MLR), random forests (RF), gradient boosting (XGB), and artificial neural networks (NN). We evaluate their performances against satellite and independent ground-based AOD observations. The results indicate that all models perform similarly, with RF exhibiting fewer spatial artifacts. While the regression slightly overcorrects extreme AODs, it remarkably reduces biases and absolute errors and significantly improves linear correlations with respect to the independent observations. We analyze a case study to illustrate the importance of the geophysical input variables and demonstrate the regional significance of some of them.

show abstract

“…These methods are increasingly being used thanks to the advancement in the development of ML hardware and technology. Until now, these ML bias correction techniques are mainly used for chemistry-transport model simulations of trace gases e.g., [19] or using in situ surface data for aerosols e.g., [20,21], or for post-processing forecasts e.g., [22,23].…”

Section: Introductionmentioning

confidence: 99%

Machine Learning-Based Improvement of Aerosol Optical Depth from CHIMERE Simulations using MODIS Satellite Observations

Lemmouchi¹,

Cuesta²,

Lachâtre³

et al. 2022

Preprint

View full text Add to dashboard Cite

We develop a machine learning (ML) approach for improving the accuracy of the horizontal dis-tribution of the aerosol optical depth (AOD) simulated by the CHIMERE chemistry-transport model over Northern Africa using Moderate-Resolution Imaging Spectroradiometer (MODIS) AOD satellite observations. These observations are used during the training phase of the ML method for deriving a regional bias correction of AODs simulated by CHIMERE. The results are daily maps of regional bias corrected AODs with full horizontal coverage over Northern Africa. We test four types of ML models: multiple linear regression (MLR), random forests (RF), gradi-ent boosting (XGB), and multiple layer perceptron networks (NN). We perform comparisons with satellite and independent ground-based observations of AOD that are not used in the training phase. They suggest that all models have overall comparable performances with a slight advantage of the RF model which expresses less spatial artifacts. While the method slightly underestimates the very high AODs, it significantly reduces biases and absolute errors, and clearly enhances linear correlations with respect to independent observations. This im-provement for deriving the AOD is particularly relevant for high dust pollution regions like the Sahara Desert, which dramatically lack ground-based measurements for validations of chemis-try-transport modeling which currently remains challenging and imprecise.

show abstract

Machine learning based bias correction for numerical chemical transport models

Cited by 16 publications

References 40 publications

Increasing heatwave with associated population and GDP exposure in North China

Increasing heatwave with associated population and GDP exposure in North China

Machine Learning-Based Improvement of Aerosol Optical Depth from CHIMERE Simulations Using MODIS Satellite Observations

Machine Learning-Based Improvement of Aerosol Optical Depth from CHIMERE Simulations using MODIS Satellite Observations

Contact Info

Product

Resources

About