Atmosphere

25 pages, 14874 KB

Open AccessArticle

Objective Classification and Environmental Characteristics of Different High-Wind Types in the Mid- and Lower Reaches of the Yangtze River Basin

by Yuanwei Xie and Yujie Pan

Atmosphere 2026, 17(5), 493; https://doi.org/10.3390/atmos17050493 (registering DOI) - 11 May 2026

Abstract

This study develops an objective attribution framework, integrating a two-step K-means clustering procedure with a random forest algorithm, to classify the weather systems responsible for high winds over the mid- and lower reaches of the Yangtze River from 2020 to 2023. The analysis [...] Read more.

This study develops an objective attribution framework, integrating a two-step K-means clustering procedure with a random forest algorithm, to classify the weather systems responsible for high winds over the mid- and lower reaches of the Yangtze River from 2020 to 2023. The analysis utilizes hourly automatic weather station observations, ERA5 reanalysis, and merged precipitation data. Four dominant HW types are identified: cold-air (CAHWs, 40.3%), tropical-cyclone (TCHWs, 27.9%), convective-system (CSHWs, 22.2%), and a residual “other” category (9.6%). Three main types exhibit distinct spatiotemporal distributions and environmental characteristics. CAHWs occur mainly in spring, autumn, and winter, concentrated in three sub-regions within the terrain channel or above the lake surface. CAHWs are characterized by non-precipitating northerlies associated with deformation frontogenesis and modulated by boundary layer processes, including terrain channeling and surface friction. TCHWs are confined to coastal areas in July and September, primarily controlled by tropical cyclone motion and land-sea distribution. CSHWs peak in afternoons from March to October and can be further divided into precipitating (PCSHWs, 40.3%) and non-precipitating (NPSCHWs, 59.7%) types. NPCSHWs typically occur in precipitation-free zones within 50 km of convective systems producing moderate to heavy rainfall, whereas PCSHWs form in smaller convective systems along the periphery of precipitation regions rather than within heavy-rainfall cores. PCSHWs are associated with higher instability, stronger low-level shear, and weaker inhibition than NPCSHWs, indicating a more organized convective environment. Full article

(This article belongs to the Section Meteorology)

► Show Figures

Graphical abstract

21 pages, 1016 KB

Open AccessArticle

A Self-Attention U-Net for Cloud Detection from FY-4B/GIIRS Observations

by Qiumeng Xue, Pei Zhao, Yuxuan Wang, Xuanyuan Yang and Zhenxing Liu

Atmosphere 2026, 17(5), 492; https://doi.org/10.3390/atmos17050492 (registering DOI) - 11 May 2026

Abstract

Accurate cloud detection for geostationary infrared hyperspectral observations is important for the effective use of clear sky radiances in atmospheric retrieval and related applications. In this study, FY-4B/GIIRS observations were used to develop and evaluate three deep-learning cloud detection models, namely a conventional [...] Read more.

Accurate cloud detection for geostationary infrared hyperspectral observations is important for the effective use of clear sky radiances in atmospheric retrieval and related applications. In this study, FY-4B/GIIRS observations were used to develop and evaluate three deep-learning cloud detection models, namely a conventional 1D-CNN (GCD-1D), a standard U-Net (GCD-U1), and a self-attention U-Net (GCD-U2). Cloud labels were generated by time-space matching between AGRI Level 2 cloud mask pixels and GIIRS field of views, and model performance was assessed using overall accuracy (OA), probability of detection (POD), and false alarm ratio (FAR) under different seasons, day/night conditions, and surface types. The results show that GCD-U2 achieved the best overall performance, with an OA of 85.88%, a POD of 77.07%, and a FAR of 24.40%, outperforming both GCD-1D and GCD-U1. The learned channel attention pattern was also physically consistent, with high weights assigned to LWIR window channels and selected MWIR bands. In the comparison with the GIIRS L2 operational cloud mask product, GCD-U2 showed higher consistency with the AGRI reference, with an average recognition–performance difference of about 10%. These results demonstrate the potential of attention-enhanced deep learning for operational cloud detection from geostationary infrared hyperspectral sounders. Full article

(This article belongs to the Section Atmospheric Techniques, Instruments, and Modeling)

28 pages, 6859 KB

Open AccessArticle

Analysis of the COVID-19 Influence on Air Quality in Urban Areas of Japan Using Multiple Satellites and Ground-based Measurements

by Tamaki Fujinawa, Satoshi Inomata, Takafumi Sugita, Kohei Ikeda, Masahiro Yamaguchi and Hiroshi Tanimoto

Atmosphere 2026, 17(5), 491; https://doi.org/10.3390/atmos17050491 (registering DOI) - 11 May 2026

Abstract

We examined the effect of the coronavirus disease 2019 (COVID-19) pandemic on air quality in the Kanto region of Japan using multiple satellites and ground-based observations. The vertical column density (VCD) of nitrogen dioxide (NO₂) derived from the Ozone Monitoring Instrument [...] Read more.

We examined the effect of the coronavirus disease 2019 (COVID-19) pandemic on air quality in the Kanto region of Japan using multiple satellites and ground-based observations. The vertical column density (VCD) of nitrogen dioxide (NO₂) derived from the Ozone Monitoring Instrument (OMI) and the Tropospheric Monitoring Instrument (TROPOMI) showed decreases of 38% and 27%, on average, respectively, in March of 2020, compared with the same month in 2015–2019, for OMI and in 2019 for TROPOMI. Surface NO₂ concentrations measured by the Atmospheric Environmental Regional Observation System (AEROS) also declined by up to 22% relative to the 2015–2019 mean, which is consistent with previously reported reductions. To investigate interactions between ozone (O₃) and NO_x, we calculated the ratio of non-methane hydrocarbon (NMHC) and NO_x and potential ozone (PO) surface concentrations from the AEROS data. The results indicated that the ozone formation regime in the Kanto region remained within the NMHC-limited domain during the COVID-19 period and was unchanged from the previous five years. Nevertheless, the baseline O₃ concentration decreased by 2.5–8.5 ppbv, depending on site (urban vs. suburban) and year (2020 vs. 2021). Diurnal variations in PO concentrations (defined as O₃ + NO₂-0.1NO_x), which is the net O₃ concentration produced by photochemical reactions and/or transport excluding the NO titration effect, showed significant reductions of 6.3 ppbv in 2020 and 3.2 ppbv in 2021, suggesting that lower PO levels were mainly attributed to the reductions in baseline O₃ concentrations in 2020. These findings highlight how pandemic-related emission reductions affected chemical processes and dynamics related to both NO_x and O₃ in a major Japanese metropolitan region. Full article

(This article belongs to the Section Air Quality)

33 pages, 5689 KB

Open AccessArticle

A Paleoclimate-Compatible Framework for Modeling Lightning-Caused Ignition Probability in Alaska

by Charlotte Uden, Patrick J. Clemins and Brian Beckage

Atmosphere 2026, 17(5), 490; https://doi.org/10.3390/atmos17050490 (registering DOI) - 11 May 2026

Abstract

Understanding the role of historical lightning-driven fire regimes in shaping terrestrial ecosystems and carbon cycles requires reconstructing fire from data beyond the instrumental record. Previous efforts have relied on paleo proxies, such as charcoal records, but these approaches are limited by their coarse [...] Read more.

Understanding the role of historical lightning-driven fire regimes in shaping terrestrial ecosystems and carbon cycles requires reconstructing fire from data beyond the instrumental record. Previous efforts have relied on paleo proxies, such as charcoal records, but these approaches are limited by their coarse spatial extent. Alternatively, process-based modeling offers a spatially continuous pathway for simulating lightning-caused fire regimes. However, existing lightning prediction models use upper-atmospheric variables, such as convective available potential energy (CAPE), that are not available in paleoclimate reconstructions, limiting their use beyond the instrumental period. Here, we develop a probabilistic framework for simulating lightning-caused fire ignitions that (1) relies on variables available in paleo reconstructions (near-surface climate, fuel moisture, and land cover) and (2) decomposes lightning-caused fire occurrence into two components: lightning strike rate and lightning ignition efficiency. Both components were trained on modern observational data for Alaska during 2002–2011, and then combined in a Bernoulli model to estimate daily fire probability. Near-surface climate predictors captured spatial and temporal variability in lightning activity with performance comparable to CAPE-based models, and ignition efficiency models showed strong discrimination between fire-causing and non-fire-causing strikes. Despite overestimation under high-risk conditions, the Bernoulli model demonstrated strong discriminatory skill (ROC AUC = 0.894), effectively ranking fire risk across space and time. By explicitly separating lightning occurrence from ignition efficiency and relying on variables available in paleo reconstructions, this approach provides a transferable framework for simulations of historical lightning-fire regimes. Full article

(This article belongs to the Section Climatology)

► Show Figures

Figure 1

37 pages, 193191 KB

Open AccessArticle

Nonlinear Local Wisdom of Waterscape Form Design in Urban Renewal for Improving Microclimate Suitability: A Case Study of Suzhou Xinsheng District

by Chundong Ma, Yiyan Chen, Jiandong Hu, Jie Liang, Hongling Li and Binyi Liu

Atmosphere 2026, 17(5), 489; https://doi.org/10.3390/atmos17050489 (registering DOI) - 11 May 2026

Abstract

Urban design that improves microclimate can significantly enhance the ecological livability of human settlements, while the climate-adaptive wisdom of applying local water-net landscapes to modern urban renewal requires further validation. To investigate the optimization mechanism of waterscape on microclimate comfort, this study focuses [...] Read more.

Urban design that improves microclimate can significantly enhance the ecological livability of human settlements, while the climate-adaptive wisdom of applying local water-net landscapes to modern urban renewal requires further validation. To investigate the optimization mechanism of waterscape on microclimate comfort, this study focuses on the public space of Xinsheng District in the Suzhou water-net region. By integrating continuous incremental multi-scenario form design, computational fluid dynamics (CFD) multi-physics simulation, and climate sensation evaluation, we reproduce the spatial differentiation of microclimate and comfort gradients across multi-hour periods during hot summer daytime within the built-up environment involving waterbodies, vegetation, and buildings. Consequently, an indicator of comfort improvement efficiency (CIE) is proposed to measure the spatial effectiveness of per-unit-area water surface expansion on climate sensation. Results show that when controlling other morphological parameters and designing three incremental waterbody scenarios—no water surface, 50% water, and 100% waterscape—the relative comfort area expanded across all time periods as water increased. This implies that waterscape variations exert a positive effect on microclimate suitability. However, during the expansion of water area at each time, the CIE was higher in the 0–50% initial stage of water surface increase compared to the 50–100% later morphological stage. Therefore, this study reveals the stepwise nonlinear trend by which increased water area in the built-up environment improves the climate suitability of waterfront spaces. Furthermore, under constraints of equivalent area and other geometric forms, a more dispersed and networked waterscape was found to be a superior spatial strategy. This confirms the microclimate wisdom of the water-net landscape in the Jiangnan locality, providing form optimization guidance for ecologically oriented urban renewal design. Full article

(This article belongs to the Section Biometeorology and Bioclimatology)

► Show Figures

Figure 1

21 pages, 4375 KB

Open AccessArticle

Mobile Observations of Air Pollution in an Urban Area: Characteristics and Variability

by Hancheng Hu, Yidan Zhang, Jiabin Jia, Langfeng Zhu, Dongyang Pu, Chenyang Shu, Tao Du, Mengqi Liu and Hao Wu

Atmosphere 2026, 17(5), 488; https://doi.org/10.3390/atmos17050488 (registering DOI) - 11 May 2026

Abstract

Urban air pollution exhibits pronounced spatial heterogeneity, yet conventional fixed-site monitoring often cannot resolve fine-scale hotspot patterns. To address this issue, this study conducted a winter intensive observation campaign combining mobile measurements and synchronous fixed-site observations in Chengdu. The mobile observation was used [...] Read more.

Urban air pollution exhibits pronounced spatial heterogeneity, yet conventional fixed-site monitoring often cannot resolve fine-scale hotspot patterns. To address this issue, this study conducted a winter intensive observation campaign combining mobile measurements and synchronous fixed-site observations in Chengdu. The mobile observation was used to characterize the spatial distribution of particulate pollution, while fixed-site pollutant and meteorological data were used to provide temporal and background context. Three mobile observation sessions were performed each day at fixed local times (09:00–11:00, 14:00–16:00, and 19:00–21:00). Based on the PM₂.₅ concentration, the observation period was categorized into two episodes: polluted episodes (PM₂.₅ > 75 μg m⁻³) and clean episodes (<75 μg m⁻³). Polluted episodes were characterized by substantially elevated PM2.5, PM10, NOx, CO, and particle number concentrations, together with relatively weak wind speed, indicating enhanced accumulation under stagnant conditions. In contrast, clean episodes generally occurred under stronger ventilation and lower pollutant levels. The results revealed marked small-scale spatial variability and distinct temporal changes in particulate pollution. PCA was suitable for the dataset (Kaiser–Meyer–Olkin = 0.788; Bartlett’s test, p < 0.001), and the first three principal components explained 82.7% of the total variance. Cluster analysis further identified three pollution regimes among 224 samples: clean/ventilated (34.4%), intermediate accumulation (39.7%), and heavy accumulation (25.9%). These findings demonstrate that short-term intensive mobile monitoring can serve as a cost-effective supplement to conventional monitoring for hotspot identification and targeted urban air-pollution management. Full article

(This article belongs to the Section Air Pollution Control)

18 pages, 3793 KB

Open AccessArticle

A New Time-Based Real Driving Emission (RDE) Evaluation Method for Heavy-Duty Vehicles Focused on NO_x Emissions Using Remote Monitoring Data

by Shuojin Ren, Gang Li, Fengbin Wang, Xianglin Zhong, Jianfu Zhao, Hao Zhang, Dongzhi Gao and Quanshun Yu

Atmosphere 2026, 17(5), 487; https://doi.org/10.3390/atmos17050487 (registering DOI) - 11 May 2026

Abstract

The real driving emission (RDE) test is going to be a necessary and effective evaluation method in the next-stage heavy-duty vehicle (HDV) emission standards, the rulemaking of which is under way worldwide (e.g., EPA 2027, Euro 7 and China 7). In this work, [...] Read more.

The real driving emission (RDE) test is going to be a necessary and effective evaluation method in the next-stage heavy-duty vehicle (HDV) emission standards, the rulemaking of which is under way worldwide (e.g., EPA 2027, Euro 7 and China 7). In this work, a time-based method (TBM) was proposed for future HDV RDE calculation. In TBM, cold-start and hot-run emissions are evaluated separately with moving average windows, yet no type-approval test results are needed so that it can also be used as a remote monitoring algorithm. This study analyzes the emissions of NO_x. The value of 0.1 times maximum engine power is utilized to determine the cold-start window, while a 2-bin window structure is adopted for hot-run analysis. In order to further illustrate and validate this method, 16,629.4 h of remote monitoring data with a sampling rate of 1 Hz from 36 China 6 HDVs and 4 different months were analyzed for driving and NO_x emission characteristics with TBM. The average duration of the 21,466 trips analyzed in this work was found to be 0.68 h, and the average ratio of trip work to WHTC (world harmonized transient driving cycle) work was around 1.38, indicating that lower duration and work requirements are needed in future RDE test. Moreover, the average cold-start length was approximately 912.4 s (15.2 min), and long cold starts could be found in cases with low ambient temperatures, low driving speeds and frequent stops. As for hot-run analysis, the proportion of Bin 1 (low-load windows) and Bin 2 (high-load windows) is directly related to the driving scenarios. The calculation results of TBM are comparable to the 2-bin method in EPA 2027. In addition, the optimization of NO_x emissions under cold start and idle conditions are challenging for future HDV updates. Full article

(This article belongs to the Special Issue Traffic Related Emission (3rd Edition))

► Show Figures

Figure 1

19 pages, 14285 KB

Open AccessArticle

Seasonal Differences in the Local and Teleconnected Climate Responses to Vegetation Greening in China and India

by Min Xiao, Miao Yu and Shiyang Zhou

Atmosphere 2026, 17(5), 486; https://doi.org/10.3390/atmos17050486 (registering DOI) - 11 May 2026

Abstract

Based on leaf area index (LAI) and enhanced vegetation index (EVI) datasets, this study systematically analyzes the spatial distribution and temporal variation characteristics of vegetation index trends at the global scale, clarifying the overall pattern of global greening and the seasonal differences in [...] Read more.

Based on leaf area index (LAI) and enhanced vegetation index (EVI) datasets, this study systematically analyzes the spatial distribution and temporal variation characteristics of vegetation index trends at the global scale, clarifying the overall pattern of global greening and the seasonal differences in vegetation greening between eastern China and India. Regions with significant greening in China and India were selected as sensitivity zones, and a coupled land–atmosphere model was used to simulate seasonal differences in the climate response to greening. The findings reveal that: (1) Vegetation greening in eastern China is most pronounced in summer, whereas in India, the greening effect is most prominent in autumn; (2) The synergistic greening of both regions induces a year-round cooling effect in southeastern China, whereas northeastern China experiences summer warming and cooling in the other seasons. Furthermore, spring greening in China and India leads to a pronounced and widespread cooling across the mid-to-high latitudes of Eurasia. (3) In terms of precipitation, southwestern China shows an increasing trend in summer rainfall, while southeastern China shows a decreasing trend. In India, synergistic greening leads to spring and summer warming and autumn and winter cooling, with the cooling and increased precipitation effects being most significant in autumn. Full article

(This article belongs to the Section Biometeorology and Bioclimatology)

► Show Figures

Figure 1

18 pages, 5046 KB

Open AccessArticle

Homogenization of Air Temperature in Xinxiang, China, Using RHtestsV4 and Implications for Regional Climate Change Assessment

by Yaxuan Chen, Qingxiang Li and Boyang Jiao

Atmosphere 2026, 17(5), 485; https://doi.org/10.3390/atmos17050485 (registering DOI) - 11 May 2026

Abstract

Inhomogeneities in climate observations can bias assessments of regional and local climate change. Taking Xinxiang, a major grain-producing region in Henan Province, China, as the study area, this study compiled temperature observations from eight national basic meteorological stations for 1951–2024. The RHtestsV4 software [...] Read more.

Inhomogeneities in climate observations can bias assessments of regional and local climate change. Taking Xinxiang, a major grain-producing region in Henan Province, China, as the study area, this study compiled temperature observations from eight national basic meteorological stations for 1951–2024. The RHtestsV4 software package was employed, using the penalized maximal t test (PMT) and a mean-adjustment scheme for series homogeneity testing and correction. Climate change characteristics in Xinxiang were analyzed at annual, seasonal, and monthly time scales and across stations to describe spatial patterns. The results indicate that the temperature series in this region exhibit marked inhomogeneity, with minimum temperature (Tmin) being the most sensitive to inhomogeneous factors; station relocation is the primary cause (accounting for over 50%). Over the past ~60 years, the regional mean warming rate increased from 0.278 °C/decade before correction to 0.370 °C/decade after correction. At the seasonal scale, spring showed the strongest warming (0.433 °C/decade), and at the monthly scale, March had the highest warming rate (0.68 °C/decade). Homogenization reduces non-climatic noise such as station relocation, improves the representation of regional temperature contrasts, highlights the stronger warming response in rapidly urbanizing areas in the east, and better represents the spatial pattern and temporal evolution of temperature. Full article

(This article belongs to the Section Meteorology)

► Show Figures

Figure 1

25 pages, 1619 KB

Open AccessReview

Evolution of Understanding of COVID-19 Transmission

by Stephanie J. Dancer

Atmosphere 2026, 17(5), 484; https://doi.org/10.3390/atmos17050484 - 8 May 2026

Abstract

In early 2020, a respiratory virus swept across the world. The World Health Organization (WHO) confirmed pandemic status and the virus was identified as a coronavirus with superlative transmission properties. Using work from the 1950s, the WHO declared that the virus was transmitted [...] Read more.

In early 2020, a respiratory virus swept across the world. The World Health Organization (WHO) confirmed pandemic status and the virus was identified as a coronavirus with superlative transmission properties. Using work from the 1950s, the WHO declared that the virus was transmitted through respiratory ‘droplets’, which were expelled by infected persons through coughing/sneezing. These would fall to the ground within 1–2 m. Scientists investigating viral transmission questioned this premise because recent work had shown that viruses populate the smallest respiratory particles, remaining airborne for much longer than larger ‘droplets’ and capable of spreading throughout the indoor environment. Advice such as handwashing, surface disinfection and social distancing was not as important as face masks and adequate indoor ventilation. People needed to know that poor ventilation constituted the highest risk for contracting the virus. Instead, homes and surfaces were disinfected and social distancing was maintained in community settings. The scientists formed a consortium named Group 36 in order to contest the WHO over airborne transmission but they could not present definitive evidence in the short term to reverse initial guidance. This account details the evolution of understanding of COVID-19 transmission and the role of Group 36 and others in challenging WHO-based policies based on dated physical science. Full article

(This article belongs to the Special Issue Ventilation and Indoor Air Quality)

► Show Figures

Graphical abstract

15 pages, 1435 KB

Open AccessArticle

Preliminary Study of the Isotopic Characteristics of Atmospheric Ammonia at a Coal Coking Industrial Park in Taiyuan, China, Using OGAWA Sampling

by Tianyu Gao, Yang Cui, Wenbin Yan, Zeqian Liu, Lili Guo, Xiaojing Hu, Qiusheng He, Ruiping Chai, Jianjun Niu, Dongsheng Ji and Xinming Wang

Atmosphere 2026, 17(5), 483; https://doi.org/10.3390/atmos17050483 - 8 May 2026

Abstract

Ammonia (NH₃) is an important alkaline gas and a key precursor to secondary inorganic aerosol. In the Fen River valley, coking plants are concentrated due to transportation advantages, while NH₃ emissions from coking processes have received limited attention despite their [...] Read more.

Ammonia (NH₃) is an important alkaline gas and a key precursor to secondary inorganic aerosol. In the Fen River valley, coking plants are concentrated due to transportation advantages, while NH₃ emissions from coking processes have received limited attention despite their potential importance. In this study, atmospheric NH₃ was sampled by OGAWA samplers in a typical coal coking industrial park in Taiyuan during autumn and winter of 2024–2025, and its nitrogen isotopic composition was used for source apportionment. The results showed that the NH₃ concentration in the industrial park was 27.4 ± 3.8 μg m⁻³, significantly higher than that in the urban area (9.3 ± 4.2 μg m⁻³) and higher than winter levels reported for North China cities. The δ¹⁵N-NH₃ was −29.7 ± 1.6‰ and increased to −14.7 ± 1.6‰ after correcting for passive sampling bias. Source apportionment further indicated that NH₃ in the industrial park was dominated by non-agricultural sources (80.7%), with ammonia slip as the largest contributor (34.2 ± 20.1%), followed by coal combustion (25.8 ± 16.5%), traffic emissions (20.7 ± 11.6%) and agricultural sources (19.3 ± 11.6%). Therefore, some measures should be taken to reduce the NH₃ emissions from ammonia slip and traffic during autumn and winter. Full article

(This article belongs to the Special Issue Air Pollution: Emission Characteristics and Formation Mechanisms)

19 pages, 1157 KB

Open AccessArticle

Haze Events Enhance Water Solubility and Bioaccessibility of Fine-Particle-Bound Arsenic in Beijing: Size-Resolved Distribution and Inhalation Health Risk

by Xueming Zhou, Shaoxuan Shi, Naijia Zheng, Juanjuan Qin, Qingqing Wang, Jihua Tan and Xinguo Zhuang

Atmosphere 2026, 17(5), 482; https://doi.org/10.3390/atmos17050482 - 8 May 2026

Abstract

Atmospheric arsenic (As) poses significant health threats in heavily polluted urban environments. However, the size-resolved distribution of water-soluble arsenic (WSAs) in atmospheric particulate matter, as well as the size-dependent variation in As concentration and solubility under contrasting haze and non-haze conditions, remains insufficiently [...] Read more.

Atmospheric arsenic (As) poses significant health threats in heavily polluted urban environments. However, the size-resolved distribution of water-soluble arsenic (WSAs) in atmospheric particulate matter, as well as the size-dependent variation in As concentration and solubility under contrasting haze and non-haze conditions, remains insufficiently characterized. This study investigated the concentration, size distribution, water solubility, sources, and health risks of particulate-bound As and WSAs in Beijing from April 2014 to February 2015. The annual mean PM_0.1–18 concentration was 136.96 ± 54.21 μg·m⁻³, with significantly higher levels observed during haze episodes (179.61 ± 41.71 μg·m⁻³) compared to non-haze periods (118.00 ± 49.42 μg·m⁻³). The annual mean concentration of As was 6.42 ± 3.69 ng·m⁻³, exceeding both WHO guidelines and Chinese standards during haze periods, while WSAs averaged 4.54 ± 2.50 ng·m⁻³. Distinct size distribution patterns were observed: As displayed, a unimodal fine-mode peak (0.32–0.56 μm) was observed during haze periods and a bimodal distribution during non-haze conditions, whereas WSAs followed comparable size-dependent behavior, reflecting shifts in dominant emission sources and atmospheric processes. The average WSAs/As ratio (0.72 ± 0.07) indicated high As solubility and strong associations with secondary species and anthropogenic emissions. Size-resolved analysis revealed that As was preferentially enriched in fine particles, particularly during haze episodes, whereas coarse particles became more prominent under non-haze conditions, especially in spring, likely driven by regional dust transport and its interactions with anthropogenic emissions. Deposition modeling based on the ICRP framework showed that As and WSAs were primarily deposited in the headway (HA: 0.68 and 0.32 ng·h⁻¹, respectively), followed by the alveolar region (AR: 0.29 and 0.20 ng·h⁻¹, respectively). Fine particles enhanced deposition in deeper lung regions during haze episodes, whereas coarse particles contributed more to upper airway deposition under non-haze conditions. Although inhalation carcinogenic risks remained within acceptable limits (10⁻⁶–10⁻⁴), risks were 1.60 times higher during haze periods, with adults bearing the greatest exposure burden. These findings demonstrate that haze conditions substantially alter the size distribution, solubility, and health risks of atmospheric arsenic, and provide a scientific basis for developing size-resolved and haze-targeted heavy metal monitoring strategies in urban environments subject to significant anthropogenic pollution. Full article

(This article belongs to the Section Air Quality and Health)

29 pages, 3632 KB

Open AccessArticle

Estimating Local Air Pollutant Contribution Ratio Based on Concentration Variability Among Monitoring Stations

by Yixuan Wang, Jianghui Liu, Qiaoyu Ma, Xinxin Yang, Yadong Wang, Ying Zhou and Jianlei Lang

Atmosphere 2026, 17(5), 481; https://doi.org/10.3390/atmos17050481 - 8 May 2026

Abstract

Quantifying the relative contributions of local emissions and regional transport is critical for urban air quality management. Chemical transport models (CTMs) are widely applied for source apportionment, but they require detailed emission inventories, extensive input data, and substantial computational resources, which limit their [...] Read more.

Quantifying the relative contributions of local emissions and regional transport is critical for urban air quality management. Chemical transport models (CTMs) are widely applied for source apportionment, but they require detailed emission inventories, extensive input data, and substantial computational resources, which limit their operational use. In contrast, urban monitoring networks provide continuous and readily available observations. This study develops an observation-based framework that estimates regional contribution ratios (RCs) from inter-station concentration variability, quantified by the coefficient of variation (CV), using WRF–CAMx results as a reference. Using Linyi as the primary case, with Xi’an and Beijing for comparison, concentration-stratified regression was applied to establish CV–RC relationships. Results show a consistent nonlinear relationship between CV and RC, with coefficients of determination (R²) up to 0.86 for PM₁₀ (daily), 0.81 for NO₂ (hourly), and 0.78–0.79 for O₃. CV decreases markedly with increasing concentration; for PM_2.5, values decline from ~0.17–0.18 to 0.05–0.06 (≈65–70%), indicating enhanced spatial homogeneity under regional influence. The relationship is most stable within a 10–15 km spatial scale. Application-based evaluation for January 2022 shows moderate agreement between estimated and modeled RC (R = 0.55–0.65), reflecting pollutant-dependent uncertainties, partly associated with biases in the model-derived reference RC. These results demonstrate that inter-station concentration variability provides a first-order, computationally efficient indicator of the balance between local emissions and regional transport. Full article

(This article belongs to the Section Air Quality)

► Show Figures

Figure 1

21 pages, 7377 KB

Open AccessArticle

Research on Prediction Methods of Anthropogenic Economic Pollutant Emission Index by Coupling of Ensemble Machine Learning and Time-Series Models Under Multiple Features

by Mengzhen Li, Yang Cao and Jianlei Lang

Atmosphere 2026, 17(5), 480; https://doi.org/10.3390/atmos17050480 - 8 May 2026

Abstract

Predicting the anthropogenic economic pollutant emissions index helps balance economic growth and environmental protection. In this study, a set of constructed features was derived using provincial-level basic data from 1995 to 2023. By constructing a weighted ensemble strategy incorporating Extreme Gradient Boosting, light [...] Read more.

Predicting the anthropogenic economic pollutant emissions index helps balance economic growth and environmental protection. In this study, a set of constructed features was derived using provincial-level basic data from 1995 to 2023. By constructing a weighted ensemble strategy incorporating Extreme Gradient Boosting, light gradient boosting machine, Random Forest, and Multi-Layer Perceptron, and integrating it with the Autoregressive Integrated Moving Average model and Shapley Additive Explanations, an anthropogenic economic pollutant emission index (AS_GPI) forecasting model was finally established, with basic and constructed features employed as its inputs (AFEA-AG). The AS_GPI forecasting model (EA-AG) was developed without constructed features, with all other settings consistent with the AFEA-AG model for comparison. Results show that the proposed model achieves high forecasting accuracy for AS_GPI across four typical pollutants, with R² values exceeding 0.95 for the AS_GPI_NOx and AS_GPI_PM_2.5. The mean absolute percentage error was as low as 0.0744. The forecasting model with constructed features yielded lower errors and higher stability in its prediction results compared with the one without such features. Feature contribution analysis revealed differing key contributors, with AS_GPI lagged values and economic-related characteristics among underlying variables exhibiting strong predictive importance. The 2024 projection results indicated certain disparities in pollution control effectiveness between key and non-key regions. Further analysis of historical and predicted data revealed a nationwide decline in AS_GPI between 1995 and 2024. Beijing and Shanghai achieved notable environmental quality improvements through anthropogenic emission reductions. The coefficient of variation values of different AS_GPIs reveal spatial heterogeneity and differences. Nationwide efforts should prioritize the control of anthropogenic NMVOC and NOx emissions. This framework provides a prediction method that offers certain reference for the development of the economy and the environment. Full article

(This article belongs to the Section Air Quality)

► Show Figures

Figure 1

29 pages, 3397 KB

Open AccessArticle

PEDNet: A Predictive Encoder–Decoder Network with Multi-Scale Global–Local Modeling for Radar Precipitation Nowcasting

by Zhuo Wang, Chaorong Li, Wenjie Luo and Chuanhu Deng

Atmosphere 2026, 17(5), 479; https://doi.org/10.3390/atmos17050479 - 8 May 2026

Abstract

Radar precipitation nowcasting remains challenging because a model must not only represent the overall motion trends of large-scale precipitation systems, but also capture the fine-grained structural variations of localized strong echo regions while maintaining stable temporal evolution in multi-step forecasting. To address this [...] Read more.

Radar precipitation nowcasting remains challenging because a model must not only represent the overall motion trends of large-scale precipitation systems, but also capture the fine-grained structural variations of localized strong echo regions while maintaining stable temporal evolution in multi-step forecasting. To address this issue, this paper proposes PEDNet, a predictive encoder–decoder network for radar precipitation nowcasting, and evaluates its performance under a unified 6-to-12 frame forecasting setting. The proposed framework jointly models global contextual perception, local structural refinement, and temporal dependencies within a unified architecture. Specifically, the designed multi-scale global–local spatial modeling strategy is used to simultaneously capture large-scale precipitation organization patterns and local echo details, while the temporal modeling module introduced at the bottleneck stage enhances sequence representation across multiple future lead times. Experimental results on the KNMI radar dataset and the SEVIR VIL benchmark dataset show that PEDNet achieves competitive overall performance across multiple categorical and continuous metrics under the adopted evaluation protocol. Meanwhile, the model maintains a practical computational cost, with 14.84 M parameters, 15.63 G FLOPs, and an inference throughput of 2.12 complete forecast samples per second. These results indicate that PEDNet provides a competitive balance between predictive accuracy and computational efficiency for short-term radar precipitation nowcasting. Full article

(This article belongs to the Section Atmospheric Techniques, Instruments, and Modeling)

► Show Figures

Graphical abstract

19 pages, 17993 KB

Open AccessArticle

Analysis of Meteorological-to-Hydrological Drought Propagation and Influencing Factors Across Arid and Humid Climate Regions in China

by Jingjing Fan, Tongning Wang, Yaodong Feng, Shibo Wei and Wei Liu

Atmosphere 2026, 17(5), 478; https://doi.org/10.3390/atmos17050478 - 8 May 2026

Abstract

Drought events have become more frequent worldwide under ongoing climate change. However, how precipitation deficits evolve into runoff deficits across contrasting dry and humid climate regions, and which factors control this transition, remains insufficiently understood. This study takes five climate zones in China [...] Read more.

Drought events have become more frequent worldwide under ongoing climate change. However, how precipitation deficits evolve into runoff deficits across contrasting dry and humid climate regions, and which factors control this transition, remains insufficiently understood. This study takes five climate zones in China (arid, semi-arid, semi-humid, humid–semi-humid, and humid) from 1970 to 2020 as examples to explore the propagation process of drought and its key driving factors. We used the Standardized Precipitation Index (SPI), Standardized Runoff Index (SRI), maximum correlation coefficient method, Kendall’s test, and multiple linear regression to identify the drought propagation time (DPT), its dynamic changes, and its main influencing factors. The results indicate that DPT exhibits significant seasonal and regional variations: on a national scale, its peak occurs in winter (7.35 months) and its trough in summer (2.54 months); specifically, propagation times in humid regions are relatively short and stable, whereas those in semi-arid, semi-humid, and humid–semi-humid regions are relatively long and highly variable. Temperature (20.69% in spring; 16.67% in summer) and potential evaporation dominate in spring, autumn, and winter, while summer precipitation (9.38%) also has a significant impact. The El Niño–Southern Oscillation (ENSO) has the most significant impact on humid regions, increasing the model R² from 34.6–37.2% to 43.7–45.0%. These results improve the understanding of drought propagation mechanisms across climatic regions, highlight the significant influence of ENSO on seasonal and regional variations in DPT, and provide a basis for regional drought early warning and water-resource management. Full article

(This article belongs to the Special Issue Assessing Hydroclimate, Environmental, and Ecosystem Impacts of Anthropogenic and Climatic Changes)

► Show Figures

Figure 1

25 pages, 2343 KB

Open AccessArticle

VOC Characteristics, Sources, and O₃ Precursor Sensitivity During Severe Summer Photochemical Pollution in a Central China Megacity

by Hui Wang, Chaofang Xue, Beibei Wang, Jiahua Guo, Zongwei Wang, Hongyu Liu, Jiakun Bai, Zhaolin Yang, Shenao Wang and Shijie Yu

Atmosphere 2026, 17(5), 477; https://doi.org/10.3390/atmos17050477 - 7 May 2026

Abstract

Despite substantial reductions in precursor emissions, persistent summer ozone (O₃) pollution remains a critical environmental challenge in the North China Plain. This study integrated O₃ and volatile organic compound (VOC) data from the summers of 2014–2020 with an observation-based box [...] Read more.

Despite substantial reductions in precursor emissions, persistent summer ozone (O₃) pollution remains a critical environmental challenge in the North China Plain. This study integrated O₃ and volatile organic compound (VOC) data from the summers of 2014–2020 with an observation-based box model (OBM) to analyze O₃ pollution trends, VOC composition, sources, and sensitivity in Zhengzhou. The results indicated a continuous intensification of summer O₃ pollution, a progressive annual increase in polluted days, and an average annual concentration increase of 6.72 μg m⁻³ yr⁻¹. Further, the average VOC concentration on polluted days was 11.7% higher than that on non-polluted days, with alkanes dominating the component distribution, followed by aromatic hydrocarbons, alkenes, and alkynes. Subsequently, a source-apportionment model (positive matrix factorization) was used to identify six VOC sources: motor vehicle emissions (28.4%), industrial emissions (23.2%), solvent use (16.0%), liquefied petroleum gas/natural gas use (15.8%), fuel combustion (11.4%), and biological sources (5.4%). The photochemical age method corrected VOC loss during atmospheric transport, revealing that the traditional O₃-formation potential (OFP) method underestimated the contributions of alkenes and aromatic hydrocarbons, with isoprene, m/p-xylene, and ethylene as key species. Furthermore, multi-scenario simulations showed that solely reducing nitrogen oxides (NO_x) emissions caused an O₃ concentration rebound, while a 4:1 VOC to NO_x reduction ratio provided optimal control. By identifying the causal drivers of O₃ pollution in Zhengzhou, this study provides a scientific basis for designing precise emission-reduction strategies applicable to the North China Plain and analogous urban regions. Full article

(This article belongs to the Special Issue Chemical Composition, Source and Formation Mechanism of Atmospheric Pollutants)

► Show Figures

Figure 1

7 pages, 653 KB

Open AccessArticle

Forward Simulation of X-Ray Transmittance Profiles in the Martian Atmosphere

by Daochun Yu

Atmosphere 2026, 17(5), 476; https://doi.org/10.3390/atmos17050476 - 7 May 2026

Abstract

The X-ray occultation technique has emerged as a novel remote sensing method for probing planetary neutral atmospheres, complementing traditional radio and ultraviolet stellar occultations. This study evaluates the feasibility and effective altitude range of X-ray occultation for retrieving Martian atmospheric density. Using the [...] Read more.

The X-ray occultation technique has emerged as a novel remote sensing method for probing planetary neutral atmospheres, complementing traditional radio and ultraviolet stellar occultations. This study evaluates the feasibility and effective altitude range of X-ray occultation for retrieving Martian atmospheric density. Using the Mars Climate Database (MCD) for atmospheric number density profiles and the XrayDB database for photoabsorption cross-sections, we calculate the X-ray transmittance as a function of tangent altitude for photon energies ranging from 0.25 keV to 20 keV. An onion-peeling ray-tracing model is employed to simulate the line-of-sight optical depth. The results indicate that X-ray photons in the soft to hard X-ray band (0.25–20 keV) are sensitive to the Martian atmosphere at altitudes between approximately 50 km and 160 km, bridging the gap between accelerometer measurements (surface to ∼50 km) and extreme ultraviolet (EUV) remote sensing (>100 km). This forward modeling framework provides a theoretical baseline for future X-ray occultation-based density retrieval in the Martian mid-atmosphere. Full article

(This article belongs to the Section Planetary Atmospheres)

► Show Figures

Figure 1

27 pages, 1286 KB

Open AccessArticle

Measurement Technique Effects on Greenhouse Gas Emissions from Conventionally Tilled, Furrow-Irrigated Soybean on a Silt-Loam Soil

by Lucia Escalante Ortiz, Kristofor R. Brye, Diego Della Lunga, Jonathan B. Brye, Lauren Gwaltney, Chandler M. Arel, Trenton L. Roberts, Caio Canella Vieira, Michelle A. Evans-White, Michael B. Daniels and Beth H. Baker

Atmosphere 2026, 17(5), 475; https://doi.org/10.3390/atmos17050475 - 6 May 2026

Abstract

Agricultural greenhouse gas (GHG) fluxes are often measured using chamber-based methods. However, comparisons among chamber-based methods are limited. This study compared the effects of chamber type and GHG concentration measurement method on GHG fluxes, emissions, global warming potential (GWP), and reduced two-gas GWP [...] Read more.

Agricultural greenhouse gas (GHG) fluxes are often measured using chamber-based methods. However, comparisons among chamber-based methods are limited. This study compared the effects of chamber type and GHG concentration measurement method on GHG fluxes, emissions, global warming potential (GWP), and reduced two-gas GWP (GWP*) in conventionally tilled soybean (Glycine max L. [Merr]) on a silt-loam soil (Aeric Epiaqualfs) in southeast Arkansas. Carbon dioxide (CO₂) fluxes measured by optical feedback-cavity enhanced absorption spectroscopy (OF-CEAS, FT-LICOR) were greater (p < 0.01) than from the non-steady-state, non-flow-through, static, closed-chamber method analyzed by gas chromatography (NFT-GC), while methane (CH₄) and nitrous oxide (N₂O) fluxes were greater (p < 0.01) from the NFT-GC than the FT-LICOR method. Cumulative season-long CO₂ emissions were 36.2, 31.6, and 13.7 times greater (p < 0.01) from the FT-LICOR than the NFT-GC method for 0–15, 0–30, and 0–60 min gas sampling intervals, respectively. Methane N₂O emissions and GWP* did not differ (p > 0.05) between FT-LICOR and NFT-GC methods for the 0–15, 0–30, and 0–60 min intervals. Results suggest that differences in GHG flux, emissions, and GWP estimates reflect the combined influence of measurement methods rather than individual system components (i.e., chamber design, analyzer type, or closure time), which should be considered when designing and executing field studies. Full article

(This article belongs to the Section Biosphere/Hydrosphere/Land–Atmosphere Interactions)

► Show Figures

Figure 1

Journal Description

Atmosphere

Latest Articles

Journal Menu

Journal Browser

Highly Accessed Articles

Latest Books

E-Mail Alert

News

Topics

Conferences

Special Issues

Topical Collections

Further Information

Guidelines

MDPI Initiatives

Follow MDPI