Atmosphere doi: 10.3390/atmos17111128

Authors: Effrosyni Baxevanaki Panagiotis Kosmopoulos Lukas Prenner Duc Tran Dimitris G. Kaskaoutis Andreas Halbmayr Ioannis Papoutsis

Rooftop photovoltaic (PV) systems within energy communities require accurate monitoring and modelling approaches capable of representing their performance under real operating conditions. This study develops a Digital Twin framework for distributed solar energy production across four energy communities in Lower Austria, by integrating Earth Observation data, theoretical PV modelling, geospatial shading analysis, and high-temporal-resolution smart meter measurements. The analysis comprises 27 locations and 262 rooftop PV installations with 15 min generation records for 2025. Theoretical PV energy production was estimated and was evaluated against measured smart meter production. The theoretical estimates closely reproduced the variability of PV generation (R2 = 0.901), despite a systematic positive bias (slope = 1.137). The analysis of aerosol (AMF), cloud (CMF), and shadow (SMF) modification factors further provided a physically interpretable assessment of atmospheric and site-specific (topographic) effects associated with PV performance, with cloud influence exhibiting the strongest seasonal variability. Finally, an additional equivalent-coverage scenario was developed to investigate the potential implications of incomplete PV production monitoring for the energy communities. Under this scenario, estimated annual PV generation increased from 4262 to 8505 MWh, the renewable energy coverage increased from 80.2% to 160.0%, and the annual energy balance shifted from a 1054 MWh deficit to a 3190 MWh surplus. The resulting 4243 MWh difference between monitored and scenario-estimated PV generation corresponded to an indicative annual economic value of €0.42–1.31 million and approximately 225 tCO2eq of avoided emissions.

Atmosphere doi: 10.3390/atmos17111127

Authors: Sanghee Chae Kyu Rang Kim

This study examined the association between dead fuel moisture content (DFMC) and wildfire occurrence in Gangwon Province, South Korea, during 2020–2024. Hourly 10-h DFMC, estimated with a Random Forest model at the nearest automatic weather stations, was used to compare four daily statistics—the daily mean (DFMC_Mean), maximum (DFMC_Max), minimum (DFMC_Min), and the value at the hour of ignition (DFMC_HH)—between 232 wildfire events and 711 non-wildfire events matched by station, month, and hour. Wildfires were concentrated in spring (March–May), when 79% of events occurred at DFMC_HH of 7–11%. All four statistics discriminated wildfire events better than chance, but discrimination was low: DFMC_Mean and DFMC_HH showed the highest out-of-fold AUC (0.643 each), and combining all four statistics improved it only marginally (0.648). Only DFMC_Mean and DFMC_HH contributed independently in the full model, and a 1 SD increase in either was associated with roughly halving the odds of wildfire occurrence (odds ratios = 0.534 and 0.517). Wildfire occurrence decreased gradually with DFMC without a distinct breakpoint, but DFMC of about 11% best separated wildfire from non-wildfire events (Youden cut-offs of 11.7% and 11.2%), close to values reported in previous studies. These results suggest that both the daily mean and the ignition-hour value of DFMC carry information on wildfire occurrence. Because discrimination was low and the associations, estimated from a case–control sample, cannot be separated from other weather conditions, seasonality, or human activity, DFMC should be used as a candidate variable in multivariate wildfire risk models rather than as a standalone forecasting indicator.

Atmosphere doi: 10.3390/atmos17111126

Authors: Xinying Fan Long Bai

Utilizing meteorological data from 1980 to 2020 provided by the China Meteorological Data Center, the spatial and temporal distribution characteristics of outdoor climate comprehensive information perceived by humans were presented using spatial interpolation methods. Five climate types were obtained through cluster analysis, with Type I, Type II, Type III, Type IV, and Type V climates distributed in the southwest plateau region, northwest Xinjiang region, Loess Plateau and middle and lower reaches of the Yellow River, northeast plain region, and central and eastern coastal region, respectively. The effects of various climatic factors on human perception were quantitatively analyzed through sensitivity analysis methods, and the dominant factors affecting human perception at different time scales were identified. The annual time scale results showed that the dominant meteorological factor in the northeast region was water vapour pressure, the dominant meteorological factor in the southwest region was wind speed, and the dominant meteorological factor in other regions was temperature. Finally, taking natural ventilation buildings in typical regions of China as examples, building construction strategies were presented and compared with traditional building construction strategies based on single parameters.

Atmosphere doi: 10.3390/atmos17111125

Authors: Yan Zhou Yi Yao Zhaonan Wang Jiayi Zhang Yueru Zhao Lu Liu Dayang Li

Extreme wind analysis commonly assumes stationarity, although climate change may alter the statistical properties of wind extremes. For a coastal region with substantial offshore wind energy development, assessing the spatial applicability of this assumption is important for robust extreme wind characterization. This study examines the spatial variability of extreme wind stationarity using annual maximum 100-m wind speeds derived from hourly ERA5 reanalysis data for 1980–2025. Four generalized extreme value (GEV) models with stationary and nonstationary formulations were compared using the corrected Akaike information criterion (AICc), with the Bayesian information criterion (BIC) and parametric bootstrap likelihood ratio tests (LRTs) as complementary assessments. A parametric bootstrap approach was used to quantify uncertainty in the estimated return levels. Results show that stationary GEV models were selected at 68.8% of locations, with nonstationary models selected at the remaining 31.2%, primarily in the northern offshore area. Among all locations, 12.6% were classified as criterion-consistent nonstationary, based on agreement among AICc, BIC, and nominally significant bootstrap LRT results, whereas 18.6% were classified as AICc-only nonstationary. However, none of the bootstrap LRT results remained significant after Benjamini–Yekutieli correction for multiple testing. A leave-one-year-out analysis further showed a mean stationary/nonstationary classification stability of 95.7% across locations, indicating limited sensitivity of the AICc-based classifications to the exclusion of individual years. The agreement between AICc-based classifications and those based on the BIC and bootstrap LRTs was 83.7% and 81.8%, respectively. Relative differences between AICc-selected and stationary return levels increased from 0.20% at 10 years to 1.41% at 100 years, while the median relative half-width of the 95% bootstrap confidence interval was 15.2% at 50 years and 19.6% at 100 years. These findings support the use of stationary GEV models across much of the study area while underscoring the need to account for spatial differences in the statistical support for nonstationarity in extreme wind risk assessment.

Atmosphere doi: 10.3390/atmos17111124

Authors: Yu-Jin Chae Sung-Hwa Park Wonbae Bang GyuWon Lee Kyu Rang Kim Yoonjae Kim

Jeju Island has complex local winds shaped by its island terrain and sea–land breezes, and point observations cannot resolve their three-dimensional structure. This study developed the Multi-scale Analysis and Prognosis System (MAPS), which combines Doppler lidar radial velocity, wind profiler (WPF) data, and a background field in a variational framework to produce hourly three-dimensional wind fields at 100 m horizontal and vertical resolutions around Jeju International Airport. Compared with Automatic Weather Station (AWS) observations for April 2026, the MAPS analysis reduced the wind speed root-mean-square error (RMSE) from 3.58, 4.01, and 3.33 m s−1 in the background field to 1.51, 1.46, and 1.98 m s−1 at the Jeju, Oedo, and Odeung stations, respectively. A case study under weak synoptic forcing (5 April 2026) used WPF, Low-Level Wind Shear Alert System (LLWAS), and Aircraft Meteorological Data Relay (AMDAR) observations. The MAPS analysis captured the wind direction shift observed around 09 UTC and a local circulation near LLWAS station RS004, and its wind speed errors were smaller than those of the background field. These results indicate that MAPS reduces background-field wind speed errors and can reflect the temporal, spatial, and vertical structure of local winds around Jeju.

Atmosphere doi: 10.3390/atmos17111122

Authors: Elizabeth A. Silber

Ground-based infrasound provides a passive means of detecting and characterizing elevated hypersonic sources, but parameter recovery is limited by covariance between source and propagation effects. Controlled spacecraft reentries are rare, well-constrained hypersonic sources for testing weak-shock models because vehicle geometry and trajectory are known independently. The 24 September, 2023, OSIRIS-REx Sample Return Capsule reentry was recorded by 39 ground-based infrasound stations, with direct arrivals associated by eigenray tracing with emission points at 44–62 km altitude and Mach 8–37. Using the ReVelle weak-shock treatment, this study examines whether the weak-shock transition (WST) coefficient C can be constrained independently of the blast-radius normalization k. Joint analysis shows strong k–C covariance: the apparent optimum in C shifts across the reference range as k varies, while the profiled residual remains weakly dependent on C. For the 23 stations within the weak-shock-continuation limit at the reference normalization k = 0.399, the nominal 95% range of the coefficient covers the full 1–100 domain once the period uncertainties are inflated to the model-data misfit. These ranges are indicative under the adopted error model rather than calibrated confidence intervals. This behavior reflects the low logarithmic period sensitivity, SC ≈ 0.034, such that doubling C produces the same local-period response as a 3.2% change in k. WST sensitivity varies with transition geometry, and geometries favorable for signal reception differ from those most sensitive to C. Period therefore constrains source scale more effectively than the WST coefficient. As an example, an Orion-sized body on the same trajectory would yield periods 3.9 times the capsule values (near the three-quarter power of the diameter ratio) with equally weak dependence on C.

Atmosphere doi: 10.3390/atmos17111123

Authors: Chuang Du Yuhe Li Wenhua Piao Yue Quan Ying Li Jingdan Cao Bin Jin

Whether meteorology adds information for emerging fine particulate matter (PM2.5) pollution may depend on the baseline and evaluation population. We assessed next-day crossings above 75 µg m−3 among currently non-exceeding days in 12 development cities and six additional Chinese cities. Additive logistic models were trained on 2016–2021 data, selected using 2022 data and frozen for retrospective evaluation through 16 September 2026. In 2023–2024, same-day ERA5 surface meteorology increased average precision (AP) from 0.307 to 0.358 across 7651 city-days and 268 onsets. The unrounded paired increment was 0.050409 (95% calendar-block interval 0.020–0.088). Exploratory co-pollutant enrichment reduced the increment to 0.029; city-specific seasonal adjustment left it uncertain. Additional daily ERA5 boundary-layer height did not establish a further ranking benefit. Consistent transfer to the six additional cities was not established. In the six added cities, pooled and descriptive event-weighted within-city increments had opposite signs in partial 2026. That panel contained only 21 onsets in partial 2026, with uncertainty sensitive to calendar-block choices. These retrospective reanalysis results support informative baselines and local validation with predictors available at issuance.

Atmosphere doi: 10.3390/atmos17111121

Authors: Moshe Mello Tumisang Seodigeng Itumeleng Kohitlhetse Robert Makomere

The efficient recovery of methanol is essential for improving the overall performance and economic viability of integrated carbon dioxide utilization processes. This work presents the simulation of the process, separation performance and techno-economic evaluation of the methanol recovery section of an integrated CO2/CH4 co-valorization process for dimethyl ether (DME) production using ChemCAD. In an attempt to assess the recovery of methanol (MeOH) from the methanol-rich stream obtained after dimethyl ether separation, a rigorous equilibrium-based distillation model was developed following a preliminary shortcut column design. The separation train was composed of flash separation and rigorous staged distillation. The operating conditions were centered on maximizing methanol recovery with stable column operation. The developed model achieved a MeOH recovery of 98.39% (overhead = 99.60 wt.% MeOH), specifically 420.73 kg/h in the distillate stream and efficient rejection of water to the bottom stream. Under the chosen operating conditions, the reboiler duty and condenser duty were 900.86 MJ/h and 1050.86 MJ/h, respectively. The specific energy consumption was 4.64 MJ/kg of recovered methanol. The equipment costing analysis calculated the total purchase cost of US$ 114,086 and installed cost of US$ 312,818 for the recovery column, condenser and reboiler. Screening calculations, performed one at a time, showed that a change in the price of steam has a bigger influence on annual utility spending than an equal proportional rise in the price of cooling water which correlates to solely the methanol recovery section.

Atmosphere doi: 10.3390/atmos17111120

Authors: Michelle Simões Reboita João Gabriel Martins Ribeiro Carlos Augusto Morales Rodriguez Raul Nicolas Maciel Chaves Flávio Augusto dos Santos Vitor Lucas dos Santos Rosa Tenório Rosmeri Porfírio da Rocha Pedro Lucas Lopes da Silveira Silva Enrique Vieira Mattos

The northern coast of São Paulo State, Brazil, experienced the most extreme 24-h rainfall event recorded in the country between the afternoon of 18 February and the morning of 19 February 2023, with accumulated precipitation exceeding 600 mm at several rain-gauge stations. This study investigates the atmospheric processes associated with the event, with emphasis on the potential contribution of a seeder–feeder mechanism, and evaluates the performance of convection-permitting (CP) WRF and RegCM5 models in simulating the event. A multiscale analysis was conducted using rain-gauge observations, ERA5 reanalysis, GOES-16 satellite imagery, polarimetric radar observations, and CP models. The event developed under a synoptic configuration commonly observed along the southeastern Brazilian coast, in which persistent southeasterly flow associated with an oceanic frontal system sustained moisture transport toward the coast. The exceptional rainfall intensity could not be explained by the large-scale environment alone, highlighting the importance of local topographic and cloud processes. Satellite and radar observations suggested that orographic precipitating clouds triggered by the Serra do Mar interacted with oceanic deeper convective bands that propagated inland. This interaction enhanced precipitation along the coastline and over the Serra do Mar through a seeder–feeder-like process. Radar observations captured signatures consistent with the main features of this mechanism: melted ice particles produced by the deeper convective band seeded the underlying orographic precipitation system, potentially enhancing collision–coalescence processes and, consequently, increasing precipitation. Both CP models reproduced the broad coastal precipitation band but displaced it inland. The WRF-3 km simulation exhibited the highest spatial correlation with radar-derived precipitation (r = 0.68), but also the largest positive bias (108.0 mm). Increasing the horizontal resolution to 1 km reduced the bias to 48.4 mm but did not improve the simulated location of the rainfall band. The RegCM5-CP experiments using different physical parameterizations highlighted the sensitivity of the simulated precipitation to model configuration. These findings indicate that the extreme rainfall resulted from the combined influence of sustained synoptic moisture supply, orographic lifting, and interactions between cloud systems, while also highlighting the challenges of accurately simulating the location and magnitude of extreme coastal precipitation.

Atmosphere doi: 10.3390/atmos17111118

Authors: Tae-Jun Kim Jin-Uk Kim Jae-Kwan Shim Myung-Ju Choi Kyung-On Boo Seung-Ki Min Dong-Hyun Cha Eun-Chul Chang

In this study, we comprehensively evaluate the simulation performance of a 12 km high-resolution multi-physics ensemble of the Weather Research and Forecasting (WRF) model, utilizing the CORDEX-East Asia (CORDEX-EA) Phase 2 grid domain. An eight-member ensemble was constructed by combining two schemes each for microphysics (MPS), planetary boundary layer (PBL), and cumulus parameterization (CPS). Although continuous simulations were conducted for a 46-year period (1979–2024), this performance evaluation focuses on various variables during a recent 20-year period (2005–2024). Validated against observational and reanalysis datasets, the ensemble members demonstrated highly capable overall performance for the surface and low tropospheric variables, with Taylor skill scores exceeding 0.7. However, simulation performance varied substantially across variables, seasons, and physics scheme combinations, with mean bias remaining a crucial factor to be considered. They successfully captured the spatiotemporal patterns of temperature and precipitation, albeit with distinct sensitivities to the chosen physics combinations. In particular, the CPS–PBL combination played the critical role in shaping inter-model variabilities through modulating precipitation over the Western North Pacific. The ensemble mean approach mitigated the sensitivities associated with individual physics configurations, thereby demonstrating robust overall performance. Further analysis for the Korean Peninsula reveals that the influence of physics combinations depends on the target analysis region. These results offer critical scientific insights into optimal physics selection and ensemble configuration for climate change scenarios over East Asia, while also providing a solid foundation for interpreting future regional climate projections based on the upcoming AR7 scenarios.

Atmosphere doi: 10.3390/atmos17111119

Authors: Hana Chaloupecká Václav Nevrlý Jan Suchánek Milan Mamula Michal Dostál Vít Klečka Jan Wild

While global decarbonization and low-temperature engines curb nitrogen oxides, they risk significant N2O emissions, which is also one of the greenhouse gases. This review evaluates N2O monitoring strategies across maritime and industrial combustion environments. Furthermore, this paper provides a rigorous comparative assessment of analytical monitoring methodologies. We evaluate the adaptability of established agricultural techniques. In parallel, advanced spectroscopic and non-spectroscopic sensor technologies (DAS, NDIR, TDLAS, PAS, CRDS, CEAS, FTIR, GC, and electrochemical/chemiresistive sensors) are critically analyzed regarding their sensitivity, response time, cross-interference vulnerabilities, and suitability for harsh industrial stacks and maritime settings. Finally, the levels of N2O above background from individual sources are estimated using wind-tunnel modeling studies. Emissions from ships, truck lines, and industrial turbines typically cause urban ambient concentrations ranging from hundreds down to tens of ppb, whereas individual trucks and rural ship levels are significantly lower, from below 1 ppb up to about 70 ppb depending on engine optimization. For industrial and combustion sources, real-time N2O analysis can be performed using FTIR, which allows for multi-gas mixture measurement, or TDLAS for targeted N2O detection. For distributed screening and monitoring around industrial plants, NDIR is a technically viable option under specific conditions. For mobile sources, TDLAS, FTIR, and highly sensitive cavity-enhanced methods (CRDS, CEAS) are suitable, provided they are optimized for harsh conditions. NDIR is applicable here only in dense urban areas. In all scenarios, gas chromatography can be used, though it requires physical sample collection and does not offer real-time analysis. As we will show, electrochemical and chemiresistive sensors cannot currently be used. Similarly, PAS is unsuitable for ambient enhancement detection, despite its applicability to stack sampling.

Atmosphere doi: 10.3390/atmos17111117

Authors: Qingjie Liu Shanshan Li Zhian Pan Guan Li

Regional solar irradiance nowcasting requires uncertainty estimates for both radiation magnitude and spatial structure. Advect-Spectral Flow separates stochastic regional mean and multiscale spectral budgets from spatial shape generation in advection-aligned coordinates. A matched comparison with an endpoint-constrained generator evaluates the effect of the complete generation scheme under shared budgets and equal-sized networks. Using hourly satellite fields over China, the prespecified 2014 in-domain test reduces 3 h empirical CRPS from 95.299 to 94.448 W m−2 (paired difference −0.852 W m−2; seven-day block-bootstrap 95% interval [−0.978, −0.725]). A frozen model exploratory evaluation on 1237 origins across six regions in a different 2023 satellite product retains overall Endpoint-relative CRPS reductions of 1.41%, 0.60%, and 0.26% at 1–3 h. The benefit is therefore modest and lead dependent. Static gating has lower 3 h CRPS. SolarSTEPS has lower external CRPS in every region and lead, and final-field spectral errors do not consistently favor Sphere. The framework and matched evaluations provide evidence on the relation between representation-level constraints and forecast quality, identifying the accuracy, structural fidelity, and computational trade-offs that matter when selecting constrained probabilistic nowcasting methods.

Atmosphere doi: 10.3390/atmos17111116

Authors: Jia Sun Wenyuan Wu Ziwei Wang Shuaijun Huang Yanqin Wang Liugang Che Dongling Xu Tangao Hu

Land surface temperature (LST) is a key metric of surface thermal state, governing heat fluxes while supporting climate and environmental applications. Using multi-temporal Landsat 8 Level-2 daytime LST and downscaled MODIS 1 km products (2017) over Zhejiang Province, this study examined the relationship between LST and natural terrain factors across scales, seasons, and day–night cycles. The main findings are as follows: (1) The thermal environment is shaped by multiple driving factors in concert. Their influences on LST show clear spatial and temporal heterogeneity across seasons and times of day. Quantitatively, the models explained 53.6–75.8% of LST variance (R2 = 0.536–0.758). NDBI was positive in spring and summer (β = 0.338 and 0.221) but negative in autumn and winter (β = −0.214 and −0.551), while NDVI and MNDWI were consistently negative (e.g., β = −1.394 and −1.517 in spring). Stable factors were significant at p < 0.05 or p < 0.01. (2) Seasonally, the elevation-induced temperature reduction is most consistent in daytime datasets during autumn, when radiation is moderate and vegetation withers. Vegetation cooling peaks in this season. In summer, strong solar radiation intensifies heat absorption by impervious surfaces, producing the most pronounced urban heat island. In winter, cold air accumulation causes a slope temperature inversion. (3) Diurnally, daytime LST is governed mainly by shortwave radiation, with slope, shaded relief, and evaporative cooling all playing significant roles. Data of different resolution consistently capture a stable negative correlation pattern. At night, LST is controlled primarily by longwave cooling and anthropogenic heat storage; cooling accelerates on mountaintops, and the thermal signals of water and vegetation reverse. These results highlight the utility of multi-scale, day–night LST data for understanding thermal environments in subtropical monsoon regions.

Atmosphere doi: 10.3390/atmos17111115

Authors: Jessica Chica Carmona Enrique Flores-Juca Diana Moscoso-Vanegas

Understanding how historic urban morphology and atmospheric conditions shape the spatial and vertical distribution of traffic-related carbon monoxide (CO) remains challenging in intermediate cities of the Global South. This study examined CO distribution across three spatial scales in the Historic Center of Cuenca, Ecuador, a UNESCO World Heritage Site, integrating urban functional analysis, morphology, traffic-flow assessment, sensor monitoring (2025), and dispersion modeling (May 2025). The historic center concentrated 30% of non-residential land uses and 20% of traffic congestion; CO at a high-traffic station (9 m) was 2.4 times that at a lower-traffic station (20 m). Mean modeled CO at 1.5 m peaked at 0.52 mg/m3 during 07:00–08:00 h; for similar flows, it was five times higher than at 08:00–09:00 h. Modeled concentrations decreased by up to 66.7% between 1.5 and 9 m, with the largest decreases under stable morning and evening conditions; the model met urban acceptance criteria against multi-height measurements. These results suggest that traffic intensity alone does not explain CO concentrations in historic urban canyons: morphology and atmospheric stability are likely to jointly regulate pollutant accumulation and dispersion, underscoring the need to incorporate vertical air-pollution gradients into mobility planning and sensor siting in historic centers.

Atmosphere doi: 10.3390/atmos17111114

Authors: Bo Zhang Xiao Xiao Yaohui Li Junchi Zhao Haoya Liu

Daily drought propagation across contrasting landforms remains insufficiently characterized. We examined meteorological drought propagation to surface soil drought in Sichuan Province, China, using daily ERA5 meteorological data at 0.25° resolution and CSMX surface soil moisture data (0–10 cm) at 1 km resolution for 2000–2024. The CSMX data were aggregated by area averaging to the ERA5 0.25° grid. Standardized precipitation evapotranspiration and soil moisture indices (SPEI and SSMI) were combined with run theory, one-to-one event matching, and copula models. The 10-day SPEI correlated most strongly with provincial SSMI (r = 0.643); gridwise correlations ranged from 0.243 to 0.712. Annual surface soil drought days correlated strongly with moderate or more severe meteorological drought days (r = 0.71). The Sichuan Basin had the highest meteorological drought propagation and surface soil drought matching rates (50.24% and 73.57%, respectively). Most matched surface soil droughts began during meteorological droughts, with regional median propagation times of 6–8 days. Copula dependence varied by region and event attribute. At conditional quantiles of 0.50 and 0.70, meteorological drought duration and severity thresholds for the same surface soil drought attribute class were highest in the Southwestern Sichuan Mountains, followed by the Northwestern Sichuan Plateau and Sichuan Basin; rankings varied at 0.90. These findings highlight short propagation times and regional differences relevant to surface soil drought monitoring.

Atmosphere doi: 10.3390/atmos17111113

Authors: Miklós Vincze Mihály Márkó Cserpák Gellért Déri Bálint Vass Mátyás Herein

Large-scale, slowly propagating wave-like structures have been observed in the atmosphere of Venus, but their dynamical origin remains uncertain. In laboratory experiments we investigated whether analogous structures can develop through baroclinic instability under conditions imitating the Venusian lower-to-middle cloud region. A differentially heated rotating annulus set-up was designed to conceptually imitate the planetary setting in terms of the relevant geometrical and dynamical similarity parameters, applying meridional and zonal temperature contrast as thermal forcing. Temperature time series analysis, infrared thermography, and feature tracking velocimetry were used to characterize the flow in the system. Slowly propagating, billow-like patterns were observed, similar in shape to the vortices of Venus. Wave activity was found to emerge above a critical meridional temperature contrast, but became substantially reduced when the ratio of the imposed zonal and meridional temperature contrasts exceeded approximately 1.6. The experiments indicate that the relative magnitude of meridional and day-night thermal forcing may play an important role in determining the occurrence of such wave-like atmospheric structures.

Atmosphere doi: 10.3390/atmos17111111

Authors: Ruisha Ma Jiang Li Yu Zhang Xinchen Gu Liangxiong Wang Xiaoying Lai

Remote sensing-based evapotranspiration (ET) provides spatially continuous information on basin water consumption, but its use in water resources management requires careful consideration of scale, attribution assumptions, and uncertainty. This narrative review synthesizes ET models and products, climatic and human controls, attribution approaches, and hydrological evaluation. Product performance varies across regions and scales, and agreement among products sharing inputs cannot replace independent evaluation. Statistical associations, model-conditioned contributions, and intervention effects answer different questions; climate–human interactions require explicit reference states and decomposition rules. We propose three checks for ET-change claims: whether the uncertainty interval supports a defined decision threshold, whether the attribution method supports the intended interpretation, and whether independent hydrological evidence constrains the conclusion. A workflow links these checks to irrigation, water-rights, groundwater, drought-restriction, and ecological-restoration applications. In a worked example from a Kansas groundwater management area, a published pumping-reduction interval meets the program threshold, and a hypothetical calculation shows that error correlation between periods can change whether an ET-based indicator meets a threshold. The framework offers a transparent basis for screening and conditional assessment, while accounting and causal claims require evidence matched to the quantity, domain, and period.

Atmosphere doi: 10.3390/atmos17111112

Authors: Junyu Cai Huafeng Luo Jiajia Han Ronghao Liu Jingliao Sun Zhen Li Xin Sun Ying Yao

Global ozone forecasts commonly depend on an initialized chemical state. We present O3CastNet, a diagnostic neural network that estimates ozone from two consecutive atmospheric states and their timestamps, without ozone input. The diagnostic operates on weather model outputs and can in principle be adapted to different medium range forecast systems. Here, FuXi supplies atmospheric predictors from ECMWF operational analyses and remains fixed while a single diagnostic model is trained for all leads. The ozone output covers nine pressure levels from 250 to 1000 hPa. Training spans 2003 to 2022, validation uses 2023, and testing covers all initializations in the 2024 test set at leads from 6 to 120 h. Verification against EAC4 shows lower root mean square error than CAMS throughout the forecast range at 500 to 1000 hPa. At 250 and 300 hPa, CAMS retains an advantage at longer leads. Spatial diagnostics reveal broad regional improvements and positive temporal anomaly correlation, with substantial geographical variation. These results demonstrate useful ozone prediction from forecast meteorology and time information without chemical initialization. They establish the performance of the FuXi implementation, while transfer to other weather models remains to be evaluated.

Atmosphere doi: 10.3390/atmos17111110

Authors: Chunxia Zhang

Aligning climate exposures with crop calendars is essential for credible climate-risk assessment in double-cropping systems. We quantified crop-season-matched precipitation and temperature effects on wheat and maize yields across 18 cities in Henan Province, China’s leading grain province, using a balanced panel of 522 city-year observations (1996–2024) and two-way fixed-effects models. Because winter wheat is harvested in early June and summer maize is sown in late June, wheat is modelled only against spring (March–May) climate and maize only against summer (June–August) climate. Spring precipitation significantly reduced wheat yield (elasticity −0.0299), and spring warming was associated with lower yield by 3.4% per °C once trends and province-wide shocks were removed; the apparent positive raw temperature correlation (+0.283) proved artefactual. Summer climate effects on maize were insignificant at the seasonal scale, consistent with irrigation buffering, but July heat and August rainfall revealed emerging vulnerabilities. Three documented extremes—2003 summer flood, July 2021 extreme rainfall, and 2023 harvest rain—were consistent with the panel mechanisms, with city-level yields falling 52–70%, up to 64%, and 11–16% below trend, respectively. K-means clustering identified three agro-climatic zones (robust to excluding yield variables), supporting spatially differentiated adaptation. Sensitivity analyses confirm that results hold when the wheat window is extended to March–June or when growing degree days replace mean temperature. Climate-risk assessment must be crop-, season-, and zone-specific.

Atmosphere doi: 10.3390/atmos17111109

Authors: Yongjun Zhu Qian Shen

To investigate the temporal responses of vegetation to climate extremes and the drivers of vegetation change across coastal mainland China, this study analyzed monthly NDVI, meteorological, and land-cover data from 2000 to 2024. Spatiotemporal NDVI changes and the lagged and cumulative response windows for 15 climate indices were examined. Partial least squares regression and modified residual analysis were then used to estimate the contributions of climate change and non-climatic factors to NDVI trends. The results showed the following: (1) The multiyear mean NDVI was 0.772, and 83.15% of valid pixels exhibited an increasing trend, including 64.31% with a significant increase, with localized declines occurring in densely urbanized areas and some coastal plains. (2) Cumulative effects dominated the NDVI responses to climate indices, with a mean lag of 0.41 months and a mean cumulative window of 2.64 months. Response windows were generally shorter along the northern coast than along the southeastern and southern coasts. (3) Mean air temperature, monthly maximum of daily minimum air temperature, and diurnal temperature range had relatively high importance in the partial least squares model; the proportions of samples with VIP ≥ 1 were 80.28%, 72.77%, and 62.91%, respectively. (4) From 2000–2024, climatic factors contributed 0.001333 year−1 (33.78%) to the increasing annual-mean NDVI trend, while the non-climatic residual—not a direct estimate of human activity—contributed 0.002613 year−1 (66.22%). Temperature was the principal component of the regional climatic contribution, but the composition of contributions varied substantially among vegetation types and across latitude and longitude bands.

Atmosphere doi: 10.3390/atmos17111108

Authors: Qingjie Liu Yutong Chang Shanshan Li Zhian Pan Xiaolong Wei Guan Li

The benefit of motion-based solar irradiance forecasts can vary with spatial texture even when their pooled skill is favorable. We compare a fixed history-only Farnebäck extrapolation with solar-geometry-adjusted persistence on the hourly Tang/TPDC product. A pilot-fixed texture split was applied to 21 geographically selected cores in 2013–2014. At 2 h, the equal-year high-minus-low contrast in advection-minus-persistence mean absolute error was −7.99 W m−2 (exploratory 95% interval: −9.96 to −6.02). Relative error reduction was 5.43% in high texture and 1.45% in low texture; their difference was 3.98 percentage points (2.97–5.09). Thus, the contrast was not confined to absolute error units. A screened morning disadvantage recurred on 322 previously unscored 2012 dates at one selected core, and retrospective diagnostics found recurring spring losses. Fixed blending reduced the local loss without demonstrating superiority over persistence. SETORS comparisons remained exploratory because satellite pixel timing remains unresolved, station records are ten-minute means, QC coverage is incomplete, and the 2013–2014 pooled method-difference interval includes zero. These findings support reporting the evaluated operator’s skill by texture, location and origin time, with ground validation before operational switching. They do not establish a causal cloud-motion mechanism or general performance limits of advection.

Atmosphere doi: 10.3390/atmos17111107

Authors: Chunlin Wang Ye Tian Weixiang Pu Rui Zhang Xi Zhang Jinyan Sun Yanfeng Huo Binfang He Maihui Zhang Di Wei Jianhe Peng Shangpei Xun

Mineral dust aerosols profoundly influence regional climate and air quality, yet large uncertainties persist in numerical dust modeling, particularly in the dust emission process, dust mass distribution analysis, and the quantification of source-specific contributions. This study focuses on an early-arriving East Asian dust storm that occurred during 18–23 February 2026 and extended to the Yangtze River Basin. Using a modified WRF-Chem model coupled with three dust emission schemes (GOCART, Shao2001, and KOK), GOT_EF0.5 performs best in capturing the spatial pattern and intensity of the observed AOD, attaining the highest R of 0.74 and the lowest RMSE of 0.16. Size-resolved analysis based on GOT_EF0.5 reveals that the event was heavily dominated by coarse dust particles. Dust size bins 5–8 contribute over 90% of the total dust mass to surface PM10, with bins 6 and 7 dominating long-range transport and bin 8 contributing mainly near source regions. Dust mass distribution analysis shows that regional dust emissions peaked on 20 February at over 4.0 Tg, whereas the area-averaged dust loading peaked 24 h later on 21 February at approximately 90 mg/m2, highlighting a distinct temporal hysteresis during downstream transport. Source attribution further demonstrates that the Gobi Desert is the predominant contributor, accounting for approximately two-thirds of the total dust loading during the event, whereas the Taklimakan Desert and other peripheral sources play only secondary, localized roles. These findings highlight the importance of the dust emission scheme in dust simulation, the quantitative basis for improving dust size-resolved forecasting, and the identification of the different desert contributions to a specific dust storm.

Atmosphere doi: 10.3390/atmos17111106

Authors: Weixin Zhu Xunjian Xu Yu Ye Yi Liu Ruizhe Zhang Kai Zhou Pei Zhang

Rime accretion on transmission lines in mountainous terrain is strongly influenced by local airflow and the availability of supercooled liquid water. This study combined nested Weather Research and Forecasting (WRF) simulations, tower observations, and the Makkonen ice-accretion model to investigate a rime-icing event at Tower No. 117 in southern Hunan, China, during 1–3 January 2026. Regional analysis revealed persistent low-level cold-air advection, warm and moist airflow aloft, and a pronounced temperature inversion near the tower, indicating a cold and stably stratified environment favourable for rime formation. At the local scale, simulated airflow and model-derived liquid water content (LWC) showed patterns consistent with terrain channelling and windward-slope ascent. During both active ice-growth stages, wind speed, LWC, and a diagnostic liquid-water transport metric were higher at the tower than in a nearby reference area. During the subsequent weakening phase, air temperature remained below freezing and wind speed remained relatively high, whereas model-derived LWC and the transport metric decreased to 18.8% and 16.9% of their first-stage values, respectively. This contrast indicates that weakened rime growth was more closely associated with reduced liquid-water availability than with temperature or wind speed changes alone. The 300 m WRF configuration yielded end-of-stage ice thicknesses closer to observations than the 1.5 km configuration. Sensitivity experiments showed that quantitative ice-accretion estimates remained sensitive to model initialisation, meteorological forcing, and icing-model assumptions. Overall, the results highlight the importance of representing terrain-modulated airflow and liquid-water variability in tower-scale icing assessments over mountainous terrain.

Atmosphere doi: 10.3390/atmos17111105

Authors: Mujahid Ali Khan Saif Said Fahdah Falah Ben Hasher Dimitris G. Kaskaoutis Mohamed Zhran

This study assesses Land Use and Land Cover (LULC) and Land Surface Temperature (LST) variability in the semi-arid Aligarh District, Uttar Pradesh, India, using two Landsat 9 images acquired on 30 May 2022 (summer) and 24 December 2022 (winter). This research employs a semi-automated hybrid classification approach integrating Maximum Likelihood Classification and Object-Based Image Analysis with Landsat imagery to map seasonal LULC for summer and winter 2022. Five classes, namely, cropland, built-up areas, barren land, plantations, and water bodies, were identified with overall accuracies of 94.4% (Kappa = 0.93) in summer and 91.2% (Kappa = 0.89) in winter. On the summer date LST ranged from 26.48 °C to 46.40 °C and on the winter date from 13.11 °C to 23.85 °C. Inter-predictor Pearson correlation analysis revealed severe multicollinearity among NDVI, SAVI, NDBI, and NDWI, with inter-predictor correlations ranging from |r| = 0.855 to 0.994, forming a tightly redundant spectral cluster. NDVI and SAVI were consequently excluded from the final OLS model. The reduced model using NDBI, BSI, and NDWI explains 54.3% of LST variability in summer and 57.8% in winter. NDBaI was excluded from the regression models because it is computed from the thermal band that is also used to retrieve LST, whereas BSI uses reflective bands only. NDBI is the dominant predictor in both seasons, followed by BSI as a significant secondary predictor, confirming built-up surface intensity and bare soil exposure as the primary thermal drivers. In contrast, GWR substantially improved model fit (Summer: R2 = 0.895, Winter: R2 = 0.918) and reduced the residual spatial autocorrelation (Moran’s I from 0.404 to 0.028 in summer and from 0.299 to 0.033 in winter), showing its effectiveness in capturing spatial heterogeneity. LST patterns differed markedly between the two acquisition dates in association with LULC differences, confirming the utility of spatially adaptive modeling for climate-responsive planning. The findings provide evidence for the scientific basis of sustainable urban management and climate adaptation strategies, contributing to the implementation of Sustainable Development Goal 13 (Climate Action).

Atmosphere doi: 10.3390/atmos17111104

Authors: Xueying Zeng Langfeng Zhu Hongyu Zhuang Li Ma Jian Guo Yan Liu

Hourly VOC observations collected in Shannan from September 2024 to September 2025 were analyzed using maximum incremental reactivity (MIR), positive matrix factorization (PMF), and backward trajectories. During the year-long analysis period, monthly median TVOC concentrations were lowest in May (52.65 µg m−3) and highest in October (208.52 µg m−3), whereas median ozone formation potential (OFP) was lowest in May (81.87 µg O3 m−3) and highest in January (231.46 µg O3 m−3). Isopentane had the highest median concentration, and acetaldehyde the highest species-level OFP. PMF of 8189 valid hours and 42 species resolved four factors accounting for 28.8%, 19.8%, 23.3%, and 28.1% of summed reconstructed mixing-ratio contributions. The factors were interpreted as traffic-related combustion (F1), mixed solvent-use and fuel emissions (F2), a halocarbon- and light-alkane-rich mixed factor (F3), and a fuel-related and OVOC-rich mixed factor (F4). During the study period, 7630 of 7949 PMF-valid hours had an assigned trajectory class; 319 remained unclassified. W trajectories comprised 356 h, 235 in winter. Season-adjusted W comparisons with pooled SSW and W/WNW hours showed positive conditional associations for F1 and F4, but not stable positive enhancement for F2 or F3. The 37-species PMF-reconstructed OFP showed weak in-sample agreement with observations (R2 = 0.0283), limiting source-specific reactivity interpretation; the trajectory associations do not quantify regional emissions.

Atmosphere doi: 10.3390/atmos17111103

Authors: Saliha Çelikcan Bilgin Osman Turan Turgay Tugay Bilgin Hüseyin Toros

Wind energy forecasting in complex terrain is a critical challenge for reliable power grid integration. This PRISMA 2020-guided structured review examines 134 eligible studies (2018–2026) that apply deep learning (DL) and other neural-network methods to wind speed forecasting, wind power prediction, and wind field downscaling, with particular emphasis on complex terrain. Four methodological pillars are analyzed: automated deep learning (AutoDL/NAS), multi-source data fusion, uncertainty quantification with emphasis on conformal prediction, and explainable AI (XAI). Only 24 studies (18%) explicitly evaluate mountainous, valley, coastal, or urban settings. CNN-based models are the most frequent architecture family, whereas AutoDL/NAS and automated hyperparameter optimization appear in only 11 studies, and no NAS framework has been evaluated in complex terrain. Reported fusion gains are study-specific (from about 13% RMSE reduction to over 50% MAE reduction) and are not directly comparable across datasets, metrics, and baselines. Conformal prediction offers finite-sample coverage guarantees under exchangeability, yet its empirical reliability under terrain-induced non-stationarity has not been evaluated. XAI appears in only five DL studies and is mostly applied post hoc. No study combines three or more pillars. We identify ten challenges and propose ten future directions toward integrated, terrain-aware, uncertainty-calibrated, and interpretable forecasting frameworks.

Atmosphere doi: 10.3390/atmos17111102

Authors: Xuefei Wang Hua Zeng Yongliang Luo Buncha Wattana Xiaohua Peng

The climate-relevant emission burden of internationally attended sporting events is rarely quantified. The data that determine it, namely where spectators come from and how they travel, are also rarely collected. We develop a transparent activity-based framework for event-scale greenhouse gas and co-pollutant inventories. The framework grades every parameter by provenance, separates person-times from unique attendees, propagates uncertainty by Monte Carlo simulation and attributes mitigation with exact Shapley values. We also list the event-specific data that an application requires. We illustrate the framework with the 2024 Shanghai Formula 1 Chinese Grand Prix (about 200,000 spectator person-times), combining official attendance data with documented assumptions. Under these assumptions, the event emitted 24,306 t CO2e (uncertainty interval 20,656–32,980 t), of which aviation contributed 78.7% and race-car fuel 0.4%. Aviation also supplied 94% of the 96.5 t of NOX; only 5–6 t of this was released in landing-and-take-off cycles. Ambitious 2030 measures cut gross emissions by 27.6–43.9%, leaving 56% of a frozen-technology baseline. The results depend on unverified spectator travel assumptions, which dominate the uncertainty. A spectator origin-and-mode survey would therefore be the most valuable addition to future event inventories.

Atmosphere doi: 10.3390/atmos17111101

Authors: Yun Gao Xusheng Zhang Xueqing Hou Jiangming Chai

Snow cover fraction (SCF) reflects the regional snow cover changes in arid mountainous areas. We developed a terrain-aware graph convolutional network–long short-term memory network (GCN-LSTM), where 9 meteorological stations in Xinjiang serve as graph nodes and 15 monthly meteorological variables as dynamic node features, while altitude, slope, and aspect define a fixed terrain information graph. The overall study period spanned March 2000–June 2022; the model uses a 12-month input window and predicts regional snow cover fraction in Xinjiang one month in advance under strict time series partitioning. On the 30-month test set (January 2020–June 2022), the model’s mean seed-level root mean square error (RMSE) was the lowest point estimate (0.0386); paired block-bootstrap intervals for ensemble-prediction differences from monthly climatology and ridge included zero. The model followed the pronounced seasonal cycle, but its five-seed ensemble had a deseasonalized anomaly R2 of −0.009, providing limited evidence of predictive skill beyond seasonality. The terrain-weighted graph had lower test RMSE than the geographic graph but similar error to the terrain-binary graph; intervals for most other graph comparisons crossed zero. Radiation accounted for 31.28% of total grouped SHAP (Shapley Additive Explanations) attribution and Lag-1 meteorological inputs for 36.24% of lag attribution. Removing the Moisture group during retrained ablation increased ensemble test RMSE by 0.0033. Different rankings indicate that attribution analysis and ablation analysis capture different aspects of model dependencies. Across seven retrospective annual outer Test folds (84 months), the ensemble did not show a stable improvement over monthly climatology (pooled RMSE 0.03896 vs.0.03763); terrain information remains a spatial prior whose broader predictive value requires further evaluation.

Atmosphere doi: 10.3390/atmos17111100

Authors: Qingfeng Li Mingyang Sun Zihan Song Kun Han Wei Zuo Aide Fang Xing Fan

Urban trees mitigate air pollution but release reactive biogenic volatile organic compounds (BVOCs), posing an inherent regulatory trade-off. Simultaneous field quantification of multiple pollutant deposition fluxes and BVOC emissions across co-occurring species remains limited. Here, we integrated micrometeorological gradient monitoring, foliar elution, dynamic branch enclosures, and canopy upscaling to synchronously quantify tree-level net foliar particulate accumulation, ozone (O3) dry deposition, and standardized isoprene emissions across five temperate tree species in Jinan, China. Significant interspecific divergence occurred across all metrics (p < 0.001). Platycladus orientalis exhibited the highest mean particulate accumulation rate (1.22 g day−1 tree−1 for PM2.5 and 3.02 g day−1 tree−1 for PM10), whereas Platanus acerifolia and Fraxinus chinensis achieved comparable tree-level PM2.5 accumulation rates. P. acerifolia delivered the highest O3 dry deposition rate (0.60 m3 s−1 tree−1), approximately 4.5 times that of the lowest-performing species. Conversely, standardized isoprene emissions spanned over three orders of magnitude, with Styphnolobium japonicum exhibiting the highest emission rate (595.98 mg C h−1 tree−1) while P. acerifolia emitted 220.33 mg C h−1 tree−1, and the remaining three taxa showed much lower emissions. Integrating positive air purification against volatile emission penalties via a comparative, emission-aware screening framework based on the Atmospheric Multifunctionality Index (AMItree) under equal and expert-based weighting regimes revealed clear interspecific differences in composite performance. F. chinensis, P. acerifolia, and P. orientalis formed the high-scoring group across the two weighting schemes (AMItree = 0.75–0.78), balancing pollutant removal against photochemical burdens. Coniferous P. orientalis ranked high due to particulate interception (0.70–0.76), whereas severe isoprene penalties suppressed S. japonicum to near zero (0.02–0.04). These findings highlight that urban greening programs must transcend single-trait selection toward an organismal-scale screening framework to optimize net urban air quality benefits.

Atmosphere doi: 10.3390/atmos17111099

Authors: Andrei Mikhailov Sergey Lukashenko Ekaterina Fedunova Andrei Thomson Murat Kassymzhanov Symbat Baigazy Nurlan Mukhamediarov Medet Aktaev Aray Temirzhanova Almagul Mukhamediarova Arafat Mamyrbekov Dinara Turgaliyeva Zhanat Baigazinov

This study investigated tritium distribution in river water, soil moisture, vegetation, and atmospheric and soil-air water vapor in a Shagan River section approximately 5 km downstream of Atomic Lake at the Semipalatinsk Test Site. Samples were collected from sites on both riverbanks, including soil depth profiles and diurnal water-vapor observations. Tritium activity was measured by liquid scintillation spectrometry. River water activity differed between the left and right banks (13,700 and 2300 Bq/L, respectively). At a left-bank site 30 m from the river, soil-water tritium increased with depth from 2600 to 53,300 Bq/L, whereas lower activities were found at more distant sites. Tritium in plant tissue free water varied considerably by sampling location and plant part, reaching 69,600 Bq/L in leaves and 76,100 Bq/L in stems at the most contaminated site. Diurnal patterns were observed in atmospheric and soil-air water vapor. These results indicate spatially heterogeneous tritium transfer within the riparian ecosystem and suggest that groundwater discharge, soil moisture dynamics, and plant-water interactions contribute to its redistribution.

Atmosphere doi: 10.3390/atmos17111098

Authors: Chen Han Hong Wang Yue Peng Wenjie Zhang Yang Zhao Ke Gui Huizheng Che Xiaoye Zhang

Low visibility (VIS) involves haze, fog, rain, snow, or their combination, while VIS prediction commonly treats aerosol and hydrometeor extinction separately. Observations and the coupled chemistry–weather model are used to quantify their contributions to low VIS in four subregions over eastern China in November 2024. Haze, fog, rain, and snow can all cause low VIS (≤10 km). For 3 < VIS ≤ 10 km and 1 < VIS ≤ 3 km, haze is the most frequent weather record in Harbin and Beijing, whereas fog prevails in Shanghai and Wuhan; for VIS ≤ 1 km, fog dominates in four cities. For 3 < VIS ≤ 10 km, aerosols contribute 87.1–97.3% of the model-estimated total extinction and hydrometeors 2.7–12.8%; for 1 < VIS ≤ 3 km, aerosols contribute 33.9–68.1% and hydrometeors 31.9–66.1%; for VIS ≤ 1 km, hydrometeors contribute 71.6–87.4%, mainly from fog, and fog and snow dominate Northeast China, while aerosols contribute 12.3–28.4%. Jointly considering aerosol and hydrometeor extinction generally produces the lowest root-mean-square error across all VIS grades and the highest threat scores for all VIS levels, especially for VIS ≤ 3 km. These results demonstrate that including both aerosol and hydrometeor extinction is essential for reliable low-VIS prediction.

Atmosphere doi: 10.3390/atmos17111097

Authors: Yu Zhang Jinqi Li

In recent decades, rapid Arctic warming has intensified Arctic amplification (AA), and its potential impacts on mid-latitude extreme weather events have received increasing attention. Using meteorological station observations and reanalysis datasets from 1979 to 2024, this study investigates the variations of regional extreme cold events (RECEs) over Northern China and their associated circulation characteristics under different Arctic amplification periods. The Arctic amplification index (AAI850), Siberian High Index (SHI), Ural Ridge Index (URI), and multi-scale atmospheric circulation fields are employed to examine the relationships between Arctic thermal conditions and Eurasian atmospheric circulation anomalies. The results show that: (1) During the study period, AAI850 showed a significant upward trend, while the frequency of RECEs in northern China decreased significantly. At the same time, the cold intensity of the event minimum temperature and the composite intensity of RECEs were significantly enhanced, reflecting a shift from a high-frequency/weak-anomaly mode to a low-frequency/high-intensity mode. (2) In the later Arctic amplification period, the distribution of RECEs shifted toward lower minimum temperatures, the event center moved northward, and the duration structure changed modestly, while the affected area showed no significant change. These findings indicate that the intensification of RECEs is mainly associated with enhanced cold-anomaly intensity, rather than prolonged duration or expanded spatial coverage. (3) AAI850 shows significant correlations with Siberian High and Ural ridge anomalies, suggesting a close association between Arctic amplification and high-latitude circulation anomalies in Eurasia. (4) The circulation analysis reveals widespread increases in 500 hPa geopotential height over the Arctic and Eurasia, regional differences in sea-level pressure changes, and warmer conditions at 850 hPa during the later Arctic amplification period. Nevertheless, northwesterly or northerly wind components persist over parts of northern China, suggesting that southward cold-air transport remains possible despite the warmer thermal background.

Atmosphere doi: 10.3390/atmos17111096

Authors: Zi Hao Moo Xiangming Sun Yuuka Xiao Ying Foo Keng Yeow Chung

A risk analysis framework is proposed to evaluate radiation exposure risks in an urban environment using long-term meteorological data. An in-house Gaussian plume dispersion model has been developed to incorporate the high-resolution subzone and population information in Singapore. This model enables the presentation of radiation risks across subzones. Risk index (RI), defined as the product of activity concentration and population, is used to quantify the radiation risks. Meteorological data utilised in the study includes ten years’ worth of surface wind observations collected by MSS across five different source locations, and the atmospheric stability statistics derived from the ERA5 reanalysis data. A series of dispersion simulations are performed using these source locations for different meteorological conditions. The computed RIs are then summed across the subzones for each month. Subsequently, the radiation exposure risks for all the source locations are ranked. The ranking derived from using the “atmospheric pollution coefficient” approach is also examined and compared with the proposed approach. The disparities between the two underscore the necessity of considering the population distribution when evaluating radiation exposure risks. Leveraging long-term meteorological data and high-resolution information on subzones and population distribution, the proposed approach can provide more reliable results for radiation exposure risk analysis.

Atmosphere doi: 10.3390/atmos17111095

Authors: Wanwan Wei Jian Li

Rapid intensification (RI) in the western North Pacific (WNP) occurs in broadly warm sea surface temperature (SST) conditions, making pre-storm ocean structure potentially informative. Using 1993–2024 best tracks, SST, the ERA5 atmospheric reanalysis, and the GLORYS12V1 global ocean reanalysis, we evaluate upper-ocean heat content above 26 °C (OHC26), 26 °C isotherm depth (D26), and atmospheric environmental conditions in relation to 24 h intensity increases of 15 and 20 m s−1 (RI15 and RI20). At fixed locations 24 h before arrival, RI15 environments had 33.91 kJ cm−2 higher OHC26 and 26.57 m deeper D26 than all non-RI records. Restricting controls to other intensifying records reduced, but did not remove, these contrasts: +18.28 kJ cm−2 and +15.37 m for RI15. Storm-grouped validation showed persistent gains from pre-storm OHC26 across pre- and post-2005 periods. RI frequencies were highest where weak vertical wind shear coincided with substantial warm-layer support; however, formal cluster-robust ocean–shear interaction terms were not statistically significant. The findings identify robust pre-existing warm-layer associations with RI while limiting inference about nonlinear environmental gating, ocean-response causality, or operational forecast skill.

Atmosphere doi: 10.3390/atmos17111094

Authors: Oliver Fischer Christina Schmidt Denis Pöhler Sven Riedner Johannes Lampel Katja Bigge Ulrich Platt

The spatial and temporal variability of nitrogen dioxide (NO2) concentrations in cities, along with the resulting personal exposure of pedestrians, cyclists and car passengers, remains poorly characterized due to the reliance of official monitoring on a small number of fixed stations and passive samplers that provide only limited spatial coverage, and in the case of passive samplers, only long averaging periods. This study aims to demonstrate the use of a mobile iterative cavity-enhanced differential optical absorption spectroscopy (DOAS) NO2 instrument for (i) city-wide spatial mapping and (ii) personal-exposure assessment across transport modes, then to (iii) develop and validate a method to extrapolate annual mean NO2 concentrations from short-term mobile measurements and (iv) evaluate activated-charcoal cabin air filters as a practical exposure reduction measure for car occupants. NO2 concentrations were measured in several German cities between 2015 and 2019 using a mobile iterative cavity-enhanced DOAS (ICAD) instrument (accuracy better than 1.7μmathvariantsans-serifg/m3 at 2 s integration time) mounted on a bicycle trailer or inside a car. Short-term measurements were combined with concurrent data from nearby reference stations to extrapolate local annual means. The approach was validated using long-term station data from Berlin and independently checked against passive sampler data in Frankfurt and stationary ICAD measurements in Walldorf. In-cabin exposure and cabin air filter performance were assessed by alternating sampling between the vehicle interior and the ambient air. For a single continuous 32 h measurement, 80% of the extrapolated annual means deviate from the measured annual mean by between −30 and +40% (10–90th percentile) (up to −43 to +74% for 1 h). This reduces to about −7 to + 12 % when 32 one-hour measurement periods are distributed across the year instead. Mobile bicycle mapping in Walldorf and in Mannheim/Ludwigshafen showed pronounced spatial and temporal variability, with extrapolated annual means ranging from about 16μmathvariantsans-serifg/m3 in residential areas/parks to 44μmathvariantsans-serifg/m3 on main roads without a separated cycle path. In-cabin NO2 concentrations reached above 1000μmathvariantsans-serifg/m3 in tunnels with a standard pollen filter, while activated charcoal cabin air filters reduced in-cabin concentrations by 92–96% (new filters) and 66–78% (filters with 30,000 km of use). Mobile ICAD measurements combined with the proposed extrapolation approach provide a cost-effective screening tool for identifying NO2 hotspots and characterizing personal exposure patterns that are frequently underestimated by fixed monitoring networks.

Atmosphere doi: 10.3390/atmos17111093

Authors: Martin Satrio Aimee Matland Joshua Aikins Amie Dobracki Cameron Wessel Matthew Decker Kaitlyn Suski

Cloud seeding can enhance precipitation, but attributing precipitation to individual seeding events remains difficult in operational programs. We analyze 71 radar-observed seeding signatures from 15 intensive observing periods using a conservative attribution protocol. We ask whether uncrewed aircraft system (UAS) silver iodide flare releases produce attributable radar signatures, how consistently they behave across a season, and how much radar-estimated precipitation is associated with them. To our knowledge, this is the largest single-season catalog of UAS seeding signatures and the first large enough to characterize timing, magnitude, geometry, and persistence statistically. Operational UAS seeded three U.S. programs during the 2025–2026 Rainmaker season. Median initial detection occurred 18 min after seeding, followed by peak reflectivity at 33 min, peak precipitation rate at 36 min, and maximum area at 42 min; median radar-detectable lifetime was 51 min. A 27-member reflectivity–snowfall ensemble adapted from SNOWIE estimated 10.42–23.02 acre-feet for PDT-025, while summed ensemble means across 15 IOPs yielded 430.44 acre-feet of signature-associated, radar-estimated liquid-equivalent precipitation. Ensemble spread represents uncertainty among selected reflectivity–snowfall relations, not between the lowest usable radar beam and the surface. The convergence of timing, location, motion, morphology, repetition, and supporting observations provides strong evidence that UAS seeding initiated these enhancements.

Atmosphere doi: 10.3390/atmos17111092

Authors: Dongting Wei Zhihao Zhang Xiao Liu Rongshuo Zhang Jingnan Hu

Ozone (O3) pollution exhibits complex, scale-dependent variability across short-term (ST), seasonal (SN), and long-term (LT) temporal scales. However, most existing studies examine individual scales separately and focus on static within-scale relationships, leaving cross-scale response relationships insufficiently understood. This study focuses on eastern and central China (105° to 125° E, 20° to 42.5° N) and combines daily gridded fields of O3, NO2, HCHO, and isoprene from the ECMWF EAC4 reanalysis with ERA5 meteorological data for 2017–2025. Treating O3 and its major precursors as a coupled system, a dual-layer Gaussian low-pass filter was applied to standardized pollutant fields to separate ST, SN, and LT components. Joint Empirical Orthogonal Function (EOF) decomposition was then used to identify spatial modes and principal component time series characterizing the co-evolution of ozone and its precursors at each scale. A Physics-Informed Neural Network (PINN) was incorporated to explore potential physical drivers underlying each mode, while rolling multivariate Ordinary Least Squares (OLS) regression and a Convolutional Neural Network–Bidirectional Long Short-Term Memory (CNN–BiLSTM) model were combined to quantify cross-scale response intensity (β) and its temporal variability. Results indicate that (1) modal coupling relationships across temporal scales are continuously evolving. When primary modes (M1) serve as targets, responses generally weaken (→: slope = −0.053 yr−1, →: slope = −2.82 yr−1). When secondary modes (M2) serve as targets, responses strengthen (→: slope = +4.285 yr−1, →: slope = +0.252 yr−1). (2) Even if the long-term upward trend in concentrations is curbed, the continued strengthening of the negative modulation of (−0.053 yr−1) may still cause short-term extreme event risks to increase rather than decrease. (3) Only a marginally significant upward trend is shown (Theil-Sen slope = +10.23 yr−1, p = 0.048).

Atmosphere doi: 10.3390/atmos17111090

Authors: Amnah Elaji

Dust storms frequently affect Saudi Arabia, disrupting transportation and agriculture, threatening public health, and incurring high economic costs. This study assessed whether elevated aerosol loading during dust storms was accompanied by consistent changes in precipitation. Data sources included Terra MODIS Deep Blue aerosol optical depth (AOD), GPM IMERG precipitation, METAR observations, and ERA5 mean sea-level pressure and wind data. AOD and precipitation were analyzed across Before, During, and After phases within the event footprints and nationwide, with the phase changes compared to earlier and later low-dust control periods. AOD consistently increased from the Before to During phase, exceeding the corresponding changes in the control periods. Precipitation responses varied across events and spatial scales, frequently occurring outside regions of enhanced aerosol loading, and pressure and wind patterns also differed. Therefore, elevated AOD was not accompanied by consistent precipitation enhancement or suppression. These findings suggest that precipitation responses to dust storms are event-specific and spatially varying.

Atmosphere doi: 10.3390/atmos17111091

Authors: Jaber H. Almutairi Issa F. Almesri Hamad H. Almutairi Mirza Pašić Amra Hasečić

Estimating Global Solar Radiation (GSR) is crucial for various engineering applications, including agriculture, residential planning, and energy systems. Reliable ground-based GSR measurements are limited in desert climates due to sensor degradation, sparse coverage and harsh environmental conditions. This study evaluates four ensemble-based machine learning (ML) models—Support Vector Machine (SVM), Random Forest (RF), Gradient Boosting (G-Boost), and Extreme Gradient Boosting (XG-Boost)—alongside two deep learning (DL) models, Multi-Layer Perceptron (MLP) and hybrid Long Short-Term Memory–MLP (LSTM-MLP)—for predicting daily GSR in Kuwait, one of the world’s hottest regions. The models incorporate 13 daily meteorological and time-related input parameters, including temperature (average, maximum, and minimum), potential sunshine duration, cloud cover, humidity, wind speed, dew point temperature, sea-level pressure, and time-based factors (day of month, day of year, month, and year). Data were collected from a ground weather station between March 2015 and December 2020. Performance was assessed using five metrics: Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and coefficient of determination (R2). Results showed strong predictive capabilities across all models, with G-Boost and XG-Boost demonstrating the best predictive performance (R2 = 0.996 and 0.994, respectively), while SVM achieved the highest held-out test R2 (0.9729) among the ensemble-based ML models. An additional hourly validation using the XG-Boost model demonstrated excellent agreement between observed and predicted hourly solar radiation, confirming that the proposed framework accurately captures short-term variability and is applicable for both daily solar resource assessment and short-term photovoltaic forecasting under extreme desert climate conditions. Among DL models, MLP outperformed LSTM-MLP. The models maintained consistently high accuracy across seasonal variations, including peak summer periods characterized by extremely high ambient temperatures. Feature importance analysis identified potential sunshine duration as the most influential factor, followed by maximum temperature. These findings demonstrate that the proposed framework provides accurate and reliable prediction of global solar radiation across multiple temporal resolutions, making it suitable for both long-term solar resource assessment and short-term photovoltaic forecasting in extreme desert climates.

Atmosphere doi: 10.3390/atmos17111089

Authors: Feng Zhang Haifeng Wang Lei Wang Yining Zhao

To reveal the spatiotemporal differentiation of precipitation, temperature and meteorological drought characteristics in the Shiyang River Basin, continuously reconstructed precipitation and temperature fields from 2004 to 2023 were used to analyze the spatial distribution, seasonal variation, and interannual variation in precipitation and mean temperature. The Standardized Precipitation Index (SPI) was used to characterize basin-scale meteorological dry–wet conditions, and Innovative Trend Analysis (ITA), the Mann–Kendall test, and Sen’s slope were combined to analyze trends in precipitation, temperature and meteorological drought. The results showed that the mean annual precipitation and mean temperature were 238.00 mm and 7.56 °C, respectively. Precipitation was higher in the southern mountainous areas and decreased toward the north, while temperature decreased with increasing elevation. Summer precipitation accounted for 54.70% of annual precipitation. Annual precipitation showed a non-significant upward trend (SITA = 1.5932, Mann–Kendall p = 0.721), whereas annual mean temperature showed a significant upward trend (Mann–Kendall p = 0.041, Sen’s slope = 0.046 °C yr−1). However, this warming signal is inherited from the gridded forcing and is not validated against station trends. Both SPI-3 and SPI-6 exhibited non-significant upward tendencies during 2004–2023 (Hamed–Rao modified MK p = 0.485 and 0.475, respectively); the statistical significance of these trends was evaluated using the Hamed–Rao modified Mann–Kendall test. SPI analysis indicated pronounced temporal variation and spatial heterogeneity in meteorological drought. Based on annual drought statistics, 2013 and 2023 were identified as the most severe drought years, with 2013 featuring the highest cumulative severity and drought area ratio, and 2023 the longest duration. Spatially, the reconstructed fields indicated potential drought signals in the central basin, while the southwestern mountains experienced milder conditions. These findings highlight the need for region-specific drought monitoring and adaptive agricultural water-resource allocation in the Shiyang River Basin.

Atmosphere doi: 10.3390/atmos17111088

Authors: Siriprapa Somboon Tarika Thumvijit Sompong Sriburee Worawat Poltubtim Chayanit Jumpee Kewalin Raktinnakorn Jeeranan Akkarachaicharukorn Nuttha Eiamjui Shinji Tokonami Chutima Kranrod

The combined influence of fault proximity, soil radioactivity, and house characteristics on residential exposure has rarely been examined in tropical Southeast Asia. This study provides an integrated baseline of soil radioactivity and indoor radon concentrations in the San Pong sub-district, located near the Mae Tha Fault Zone, northern Thailand. Fifty soil samples were analyzed using high-purity germanium gamma spectrometry, and indoor radon concentrations were measured in 50 houses using RADUET detectors over a three-month period. The mean activity concentrations of 226Ra, 232Th, and 40K (73.99 ± 7.73 Bq kg−1, 70.53 ± 6.34 Bq kg−1, and 610.54 ± 65.74 Bq kg−1, respectively) exceeded the UNSCEAR (2024) worldwide and Thai national averages; consequently, the outdoor absorbed dose rate in air (102.24 ± 8.96 nGy h−1), annual effective dose (0.13 ± 0.01 mSv y−1), and excess lifetime cancer risk (0.52 × 10−3) were approximately twice the worldwide values. In contrast, indoor radon concentration was low (21.27 to 47.02 Bq m−3; mean 33.38 ± 5.38 Bq m−3), and inhalation dose (0.84 mSv y−1) remained low. Neither soil radionuclides nor indoor radon increased toward the fault. Natural ventilation of the tropical houses may keep indoor radon low, whereas outdoor gamma exposure warrants continued monitoring.

Atmosphere doi: 10.3390/atmos17111087

Authors: Cheng Zhou Chengfang Yang Junjian Guo Liang Wang Haiwen Wei

Accurate weather forecasting is essential for decision-making in agriculture, transportation, disaster prevention, and energy management. In recent years, data-driven artificial intelligence (AI) models have shown considerable potential for medium-range weather forecasting by substantially improving computational efficiency while maintaining competitive forecast skill. This study evaluates the regional performance of Pangu-Weather initialized with three global numerical weather prediction (NWP) systems—ECMWF-IFS, CMA-GFS, and NCEP-GFS—over Shandong Province, China, from August to December 2024. Surface forecasts are verified against observations from 123 national meteorological stations, and upper-air forecasts are evaluated using observations from three radiosonde stations. Of the three configurations, Pangu-Weather initialized with ECMWF fields (EC_PG) provides the most robust overall performance. Relative to ECMWF operational forecasts (EC), EC_PG reduces the mean 2 m temperature RMSE by approximately 13.2%, with improvement at 38 of 40 lead times and a maximum improvement of 28.3%. Its 10 m wind-speed RMSE is lower at every lead time, with a mean improvement of 5.7% and a maximum of 10.6%. For upper-air variables, all three con-figurations generally improve specific-humidity forecasts, with mean improvements of 8.6–12.8% across three stations and three pressure levels and a maximum improvement of 33.7% in an individual comparison. Improvements in upper-air temperature and wind speed vary by pressure level, station, and lead time. EC_PG performs best overall, whereas CMA_PG and NCEP_PG have larger errors than EC at several longer lead times over coastal stations and at 500–700 hPa. These results support the potential use of Pangu-Weather as a regional forecasting aid while emphasizing that performance de-pends on the operational initial fields, variable, location, and lead time.

Atmosphere doi: 10.3390/atmos17111086

Authors: Aqsa Muhammadi Sławomir Sulik Piero Lionello Nouman Iqbal

Storm-relative helicity (SRH) and composite severe-weather diagnostics depend on the storm-motion vector assumed in their calculation. This study evaluates storm-motion sensitivity across 3317 reprocessed EF2+ tornado environments in the contiguous United States from 2000 to 2024. Using Bunkers right-moving motion as the baseline, storm motion was displaced by ±2.50 and ±5.00 m s−1 along the axis connecting the Bunkers right- and left-moving vectors. An analytical two-dimensional formulation was also used to determine the maximum absolute SRH response over all horizontal perturbation directions at each prescribed displacement magnitude. Median maximum absolute changes in 0–3 km SRH along the right–left-moving axis were 50.70 and 101.40 m2 s−2 for the ±2.50 and ±5.00 m s−1 perturbations, respectively. The baseline median 0–3 km SRH was 290.92 m2 s−2, and the median fixed-layer supercell composite parameter (SCP) was 7.83. Along the right–left-moving axis, 6.57% and 27.46% of environments crossed the illustrative SCP thresholds of 1 and 4, respectively, within the ±5.00 m s−1 perturbation envelope. These fractions represent conditional responses to prescribed storm-motion perturbations rather than probabilities associated with observed storm-motion errors. The SCP findings apply specifically to the fixed-layer formulation examined here and should not be extrapolated to operational effective-inflow-layer versions without further evaluation.

Atmosphere doi: 10.3390/atmos17111085

Authors: Maura González Robles Jesús López de la Cruz Juan Carlos Tejeda González Teresa Evangelina Martínez Díaz

Drought is one of the most complex extreme hydrometeorological phenomena to study and affects extensive regions worldwide. This study aimed to characterize droughts in the state of San Luis Potosí, Mexico. The analysis employed the parametric Standardized Precipitation Index (SPI), based on the two-parameter Gamma distribution. Drought intensity, magnitude, and duration were estimated at a 12-month accumulation time scale (SPI-12) based on records from 194 climatological stations covering the period 1955–2017. Multivariate statistical techniques were applied to identify the main precipitation patterns and the dominant modes of SPI-12 variability. The first two EOF modes explained 67% of the total SPI-12 variance (46% and 21%, respectively), and 14 and 11 drought events were identified for these modes. Significant spatially contrasting associations between the El Niño–Southern Oscillation (ENSO) and drought occurrence and spatial extent were identified. Drought conditions in the Altiplano and Centro regions were more strongly associated with La Niña, whereas the Media and Huasteca regions showed stronger associations with El Niño. Furthermore, the potential utility of SPI-12 as an explanatory covariate for changes in monthly runoff volume records was demonstrated for the Verde and Gallinas river systems, with significant Kendall’s tau correlations of 0.330 and 0.380, respectively. The findings provide a basis for improving drought characterization and understanding of its spatiotemporal variability in the region, while offering valuable information to support the development of effective planning, adaptation, and mitigation strategies.

Atmosphere doi: 10.3390/atmos17111084

Authors: Michael J. Rycroft

The concept of cyclotron resonance near the equatorial plane of the Earth’s magnetosphere between whistlers and energetic electrons originated in the 1960s. Since then, the subject has advanced substantially, with experimental, theoretical and computer modelling studies. A chronological review is undertaken not only of the basic physics of gyro resonance but also of very-low-frequency (VLF) radio observations made on the ground and measurements made by various rocket and satellite instruments of the fluxes of precipitating energetic electrons and whistler-mode waves. Lightning-induced electron precipitation (LEP) contributes appreciably to the dynamics (i.e., the acceleration, transport and loss) of trapped Van Allen radiation belt electrons. However, plasmaspheric hiss (at L < 4.5, for quiet to moderate activity conditions) and chorus (generally at L > 4.5, beyond or outside the plasmasphere and often associated with auroral phenomena) are believed to be more important overall than LEP in causing the precipitation of energetic electrons from the magnetosphere at L > 3.8. The gyro resonant interaction between the VLF radio transmitter signals near the equatorial plane and energetic electrons also contributes appreciably to their loss. With LEP being the major cause, each plays a role in the formation of the slot in the radiation belts at L ~ 2.5. The mechanisms whereby precipitating energetic electrons perturb the physics and chemistry of the mesosphere and D-region ionosphere are also briefly reviewed.

Atmosphere doi: 10.3390/atmos17111083

Authors: Mengmeng Liu Bin Jiang Enze Tian Jiaqi Fan Fangyu Wang Dong Li

Accurate prediction of PM2.5 concentrations is essential for public health protection and environmental management. However, existing forecasting models face critical challenges in capturing complex temporal dependencies and providing interpretability for decision-making. This study proposes a novel hybrid framework integrating one-dimensional convolutional neural networks (1D CNN), extended Long Short-Term Memory (xLSTM), and attention mechanism (CNN-xLSTM-AM), combined with SHapley Additive exPlanations (SHAP) for interpretable PM2.5 forecasting. The 1D CNN unit captures local temporal relationships of multivariate sequences. The xLSTM design, based on sLSTM-mLSTM-sLSTM blocks, addresses limitations of traditional LSTM architectures through exponential gating and dual memory structures: sLSTM captures the state-tracking dynamics of scalar memory mixing, and mLSTM captures multivariate dependencies using matrix memory with larger capacity. The attention mechanism assigns dynamic weights to temporal features according to the importance of each time step. Extensive experiments on Beijing hourly data (2020–2021) show that CNN-xLSTM-AM achieves an RMSE of 9.845 μg/m3 and an R2 of 0.977 for single-step prediction. Compared with CNN-xLSTM, it reduces RMSE by approximately 8.5% in single-step prediction and by 8.9% and 12.3% at the 2 h and 10 h forecasting horizons, respectively. The SHAP interpretability analysis identifies atmospheric pollutants as the highest contributors to predictive power (57.09%), with carbon monoxide as the top feature (32.791%). SHAP dependence analyses further suggest nonlinear associations between the input factors and model predictions, with potential turning points identified for selected variables, including carbon monoxide, relative humidity, wind speed, and wind direction. These model-derived patterns indicate potential changes in predictive relationships but should not be interpreted as definitive physical thresholds or universal regime shifts in PM2.5 behavior. These results demonstrate the potential of combining predictive modeling with post hoc interpretation to support environmental decision-making and may inform the development of targeted emission-control strategies and threshold-informed early-warning approaches.

Atmosphere doi: 10.3390/atmos17111082

Authors: Ornella Salimbene Nicola Leoce Ivano Salimbene María Teresa Baeza-Romero Deborah Panepinto Gregor Čok

Urban air pollution is a key driver of adverse respiratory outcomes, yet its precise impact on lung function trajectories in non-hospitalized patients with mild chronic obstructive pulmonary disease (COPD) has not been fully characterized. Objectives: This study investigated the complex interplay between ambient particulate matter exposure, residential microenvironmental factors, sociodemographic characteristics, and 1-year longitudinal lung function in mild COPD patients. Methods: A 1-year prospective cohort study was conducted on 50 non-hospitalized male volunteers residing in Turin (Italy) for a minimum of 5 consecutive years. Longitudinal lung function was assessed monthly via spirometric parameters, including forced expiratory volume in 1 s (FEV1), forced vital capacity (FVC), and the FEV1/FVC ratio. Ambient PM concentrations assigned according to residential location (PM2.5 and PM10) were integrated with a comprehensive residential microenvironmental and lifestyle profiling. To assess longitudinal changes in lung function while controlling for covariates, a Multivariate Analysis of Covariance (MANCOVA) was applied to estimate the simultaneous effects of these predictors on the joint lung function parameters over time. To fully account for the nested, repeated-measures structure of the dataset, linear mixed-effects models (LMMs) were deployed to calculate specific rates of functional decline over the 12 monthly intervals. This model assessed multiple exposure variables, including cooking appliances (gas vs. electric), domestic heating systems (internal vs. external boilers), smoking status, traffic exposure, and pets. Statistical associations were modeled using Principal Component Analysis (PCA) and Agglomerative Hierarchical Clustering (AHC). Results: The MANCOVA revealed a highly significant overall multivariate effect of the combined predictors on the joint lung function parameters (Wilks’ λ = 0.125, p < 0.001). Specifically, monthly PM2.5 concentrations (p = 0.001), residential proximity to primary roads (p = 0.004), and the utilization of gas-fueled cooking appliances (p = 0.002) exerted statistically significant independent effects on the decline in the spirometric profile, whereas PM10 and internal boilers did not (p > 0.05). PCA then identified four significant components explaining 88.8% of the total variance, demonstrating collinearity between ambient PM2.5 and PM10 concentrations and revealing that spatial residential distribution significantly accounts for the longitudinal variability in FEV1 and FVC. Furthermore, AHC successfully identified distinct patient clusters, confirming that indoor microenvironmental factors—predominantly the utilization of gas-fueled cooking stoves and internal boilers—could act as critical compounding risk factors alongside outdoor traffic pollution. Conclusions: These preliminary findings suggest how intersecting urban outdoor pollutants and specific household factors may potentially contribute to the trajectories of mild COPD. These data underscore the potential benefit of integrated environmental-indoor risk assessment methodologies and may inform future targeted public health interventions and preventive strategies to mitigate exacerbations in vulnerable urban populations.

Atmosphere doi: 10.3390/atmos17111081

Authors: Arisa Jiratampradab Thidaporn Supapakorn

Severe seasonal particulate matter pollution poses persistent environmental and public health challenges across upper northern Thailand. This study investigates the statistical distribution of daily average PM2.5 concentrations across five key monitoring stations (Chiang Rai, Mae Hong Son, Chiang Mai, Nan, and Lampang) using an extensive multi-year dataset from June 2023 to May 2026 (1096 daily observations per station). Candidate probability models—specifically Log-Normal, Gamma, and Weibull distributions—were fitted using maximum likelihood estimation. Their goodness of fit was rigorously evaluated through the Kolmogorov–Smirnov test, Anderson–Darling test, Akaike Information Criterion, and Bayesian Information Criterion. The empirical findings reveal that the Log-Normal distribution consistently outperformed the alternative models, yielding the lowest test statistics and information criteria values across all stations. Consequently, the Log-Normal model is established as the most robust framework for characterizing right-skewed particulate matter concentrations and capturing upper-tail extremes associated with regional biomass burning episodes. Furthermore, threshold exceedance evaluations relative to the Thai national 24-h standard (37.5 µg/m3) highlight high regional vulnerability, with Nan recording the highest exceedance frequency at 25.12%, followed by Lampang (21.70%) and Chiang Rai (20.87%). These validated probabilistic models provide a reliable statistical foundation for assessing long-term air quality risks, evaluating regulatory thresholds, and supporting targeted mitigation policies in regions susceptible to transboundary haze.

Atmosphere doi: 10.3390/atmos17111080

Authors: Huanhuan Zhu Brian Odhiambo Ayugi

This study investigated future changes in extreme climate events over China under carbon neutrality using an improved reliability ensemble averaging (REA) method, based on 12 Coupled Model Intercomparison Project Phase 6 (CMIP6) models under the SSP1-1.9 and SSP1-2.6 scenarios. Results indicated that carbon neutrality is reached in 2042/2063 under SSP1-1.9/SSP1-2.6. Relative to 1995–2014, the areal-mean increases in annual mean temperature (Tav), hottest day temperature (TXx), coldest night temperature (TNn), total precipitation (Prcptot), and heavy precipitation (R95p) are 1.27/1.61 °C, 1.31/1.69 °C, 1.45/1.84 °C, 9.5/13.3%, and 20.8/29.6% under SSP1-1.9/SSP1-2.6, respectively. Larger increases in Tav and TNn occur in northeastern (NE) and northern China, while those in TXx occur in northwestern (NW) China. Temperature projections are most (least) robust in southeastern China (NE). Prcptot increases largely in the Tibetan Plateau (TB) and NE, while R95p increases largely in TB and south of the Yangtze River. The smallest but most robust precipitation increase is in southwestern China, while the least robust is in NW. The areal-mean Tav, TXx, and TNn increase by 0.34 °C, 0.38 °C, and 0.39 °C under SSP1-2.6 (late carbon neutrality) compared with SSP1-1.9 (early carbon neutrality), while those for Prcptot and R95p increase by 3.8% and 8.8%, respectively. Achieving carbon neutrality earlier may contribute to mitigating future climate change and reducing the worsening of extreme temperature and precipitation extremes over parts of China.

Atmosphere doi: 10.3390/atmos17111079

Authors: Mengya Bu Gangfeng Zhang Yifeng Lu

Fugitive emissions from storage-tank accessories are important sources of volatile organic compounds (VOCs) in chemical industrial parks, but the applicability of estimation methods requires field evaluation. Optical gas imaging (OGI) was used to screen about 200 selected volatile organic liquid storage tanks in a Chinese chemical park. Evident emissions were detected at 46 tanks; breathing valves (14) and foam generators (13) accounted for about 59% of detections. Twelve of the 13 foam generators were inaccessible to flame ionization detector (FID) measurements, illustrating the complementary coverage of OGI and probe-based screening. FID readings for 14 breathing valves were 2125–8423 µmol mol−1, above the numerical criterion of 2000 µmol mol−1 in amendments to GB 31570/31571/31572. Regulatory interpretation requires the applicable operating-pressure condition and facility requirements. For 30 FID-accessible accessories, annualized estimates from the OGI emission-factor and correlation-equation methods were 26.3 and 1.86 t a−1, respectively, assuming continuous release for 8760 h. The approximately 14-fold difference reflects the methods’ different category mappings and dependence on screening readings; it does not establish their accuracy against an independent mass-emission reference. The findings support attention to inaccessible accessories and further field validation of accessory-specific emission estimates.

Atmosphere doi: 10.3390/atmos17111078

Authors: Junming Li Kemeng Yin Yanlei Zheng Yaxin Liang Yujuan Fan

Regional assessments of atmospheric ammonia must distinguish changes in annual abundance from changes in seasonality and hotspot organization. We analyzed monthly Infrared Atmospheric Sounding Interferometer (IASI) total columns over Liaoning, Jilin and Heilongjiang during 2008–2024 on a 0.2° grid. We constructed a Unified Spatiotemporal Bayesian Additive Regression Trees model (UST-BART) for 2008–2022 using nonnegative complete observations, combining location means and within-location departures with geographical and calendar backgrounds. Regional columns increased at 4.52 × 10−5 mol m−2 decade−1, and positive trends covered 93.6% of represented area. The seasonal amplitude of atmospheric NH3 columns increased by 119.7% between 2008–2012 and 2019–2024; relative hotspots remained concentrated in Liaoning, with southwestward centroid trends despite start-year-sensitive endpoint distances (477.4 vs. 46.8 km). Liaoning retained the largest absolute seasonal cycle, whereas Heilongjiang showed the largest proportional provincial amplification. UST-BART identified a positive temperature association, a U-shaped soil-moisture association and a broadly inverse-U association with boundary layer height. Temperature contrasts varied jointly with moisture and boundary layer conditions, with the largest provincial temperature response in Liaoning. Temporal and spatial holdouts yielded R2 values of 0.414 and 0.286. These findings reveal strengthening seasonality and spatial reorganization of NH3, with context-dependent environmental associations that support regionally focused monitoring.

Atmosphere doi: 10.3390/atmos17111077

Authors: Su-Youn Cho Hee-Tae Roh

Whether particulate matter (PM) exposure modifies the brain-related benefits of exercise remains unresolved. This focused narrative review examined exercise-related neurotrophic responses, centred on brain-derived neurotrophic factor (BDNF), alongside cognitive and brain outcomes under particulate or traffic-related pollution. PubMed and Web of Science Core Collection were searched, and 37 reports, comprising 33 main and four indirect reports, were synthesised by biological outcome level and response pattern. Circulating BDNF responses were heterogeneous; vascular endothelial growth factor (VEGF) and insulin-like growth factor 1 (IGF-1) provided complementary peripheral evidence. Human experiments and population studies reported attenuation in selected cognitive, dementia and imaging contrasts, alongside retained favourable activity associations, uncertain modification and potential activity-related buffering. Central molecular evidence from inhalation experiments was limited; selected animal improvements arose after intravenous particle administration or in an airway-instillation model with cerebral ischaemia/reperfusion, constraining translation to exercise during ambient inhalation. Differences in comparators, interaction scales and exposure measurement, together with potentially overlapping cohorts, limited inference. Neurotrophic and functional findings should be interpreted in relation to the measured outcome and study comparison: a change in peripheral BDNF alone does not establish brain benefit under PM exposure. The evidence supports neither a uniform loss of exercise benefit nor a general concentration threshold for reversal to harm. Studies measuring neurotrophic markers and central or functional outcomes together, with activity-specific exposure and exercise-intensity assessment and appropriate exposure-by-exercise comparisons, are needed to test how these responses relate.

Atmosphere doi: 10.3390/atmos17111076

Authors: Banglong Pan Longfei Ye Zhixing Chen Qihang Liang Juan Du Jiulin Li Rongxian Zhao Ruyu Yan Chunjing Fang Zhuo Diao

The partial pressure of carbon dioxide in water (pCO2) is an important indicator of CO2 accumulation and potential air–water exchange in lakes. However, remote-sensing retrieval of pCO2 remains challenging in large, shallow eutrophic lakes due to the interacting biological, physicochemical, and optical processes. This study combined in situ pCO2 measurements, Landsat imagery, and aquatic environmental variables to develop a heterogeneous stacking ensemble comprising random forest (RF), gradient boosting decision tree (GBDT), and extreme gradient boosting (XGBoost) models for Lake Chaohu. Model comparison showed that the stacking model outperformed the individual models, with an R2 of 0.80, RMSE of 104.22 μatm, RPD of 2.24 and SMAPE of 17.50%. During the four investigated years, lake-wide pCO2 increased from 2013 to 2017 and was lower in 2021 and 2025; the maximum value in 2017 exceeded that in 2025 by 209.74 μatm. Autumn generally exhibited higher pCO2 than winter. Persistent hotspots occurred in the western basin, nearshore waters in the northwest, and the central connecting zone, while lower values characterized the eastern basin and open waters in the southern part. SHAP analysis identified chlorophyll-a (Chl-a) as the leading predictor. However, its moderate Pearson correlation with pCO2 (r = 0.34) indicates that its predictive importance may reflect nonlinear multivariate associations rather than a direct causal relationship. Water temperature (WT), dissolved oxygen (DO), pH, salinity (Sal), total dissolved solids (TDS), and electrical conductivity (EC) also contributed to the variability of pCO2. These findings provide a remote sensing approach to monitor lake pCO2 and characterize carbon dynamics in shallow eutrophic lakes.

Atmosphere doi: 10.3390/atmos17111075

Authors: Zhiyuan Fang Shu Li Hao Yang Zhiqiang Kuang

Due to variations in surface characteristics, terrain, and geographical conditions, the performance of planetary boundary layer (PBL) schemes in the Weather Research and Forecasting (WRF) model can vary substantially among coastal regions. In this study, 11 PBL schemes were evaluated for a representative sea–land breeze (SLB) event over Ningbo, China. The simulated 2 m temperature, 10 m wind speed, and 10 m wind direction were compared with observations from three surface meteorological stations, while radiosonde observations from Hangzhou were used to assess the simulated vertical temperature and relative humidity structures. For the 2 m temperature, YSU and SH produced the lowest nominal RMSE values of 1.588 and 1.595 °C, respectively, followed by GBM, UW, ACM2, and MYJ. However, paired moving-block bootstrap analysis showed that the RMSE differences between YSU and SH, GBM, UW, ACM2, and MYJ were not statistically distinguishable at the 95% confidence level. For 10 m winds, UW produced the lowest wind speed RMSE, BouLac showed the smallest absolute wind speed bias and the highest correlation coefficient, and MYNN2.5 yielded the lowest circular mean absolute error for the wind direction. The simulations also showed scheme-dependent differences in sea breeze evolution and vertical atmospheric structure. Overall, no single PBL scheme consistently outperformed the others across all evaluated variables and metrics. These results highlight the importance of evaluating PBL schemes using multiple observational constraints when simulating sea–land breeze circulation in complex coastal environments.

Atmosphere doi: 10.3390/atmos17111073

Authors: Chenxi Liu Rong Fang Zihang Chen

Ozone (O3) pollution in coastal urban regions is strongly influenced by sea–land breeze (SLB) circulation. However, linking its long-term surface effects with vertically resolved observations remains an important research need in the western Pearl River Delta (PRD). This study combines meteorological and surface O3 observations from 2014–2023 with lidar measurements and ERA5 reanalysis for a case study on 25–26 July 2023. We examine three complementary aspects: SLB frequency and temporal characteristics, surface O3 concentrations and peak timing, and vertical O3 redistribution. SLB days accounted for 13.1–19.4% of days annually, with mean sea-breeze and land-breeze durations of 8.79 and 6.63 h, respectively. Surface O3 concentrations were generally higher on SLB days in all seasons, with larger increases at inland sites, particularly in spring and autumn. Changes in peak timing varied by season and site. The case study suggests a vertical recirculation pathway in which nocturnal return flow transports O3-rich air inland aloft, followed by downward redistribution during daytime boundary-layer development. By integrating a decade of surface observations with vertically resolved evidence from an individual event, this study connects regional statistical patterns with a physical interpretation of coupled horizontal recirculation and vertical redistribution. These findings highlight the importance of considering breeze timing, elevated O3 layers, and boundary-layer mixing in coastal air-quality forecasting, chemical transport modeling, and regional pollution mitigation.

Atmosphere doi: 10.3390/atmos17111074

Authors: Qiwen Dai Tiyong Zuo Yawen Guo Ke Zhang Wenlong Zhao Zhantao Zhang Shuang Wang Lili Fu Yaming Sun Chenchen Song Bin Li Yue Zhang Lijun He

Organic aerosol constitutes a major fraction of atmospheric particulate matter. Secondary organic aerosol (SOA), which is formed through the atmospheric aging of volatile organic compounds (VOCs), semi-volatile organic compounds (SVOCs), and intermediate-volatility organic compounds (IVOCs), has important impacts on air quality, climate, and human health. Biomass burning emits a chemically complex mixture of gaseous and particulate organic compounds, which serve as a major source of SOA precursors. These compounds undergo coupled aging processes involving gas–particle partitioning, re-volatilization, multiphase reactions, and radical-initiated oxidation, thereby contributing to SOA formation. However, the formation pathways of SOA from biomass-burning precursors and the relative contributions of different precursor classes remain poorly understood. Oxidation flow reactors (OFRs), characterized by high oxidant exposure, short residence times, and the ability to simulate several days or even weeks of equivalent atmospheric aging within minutes, have become an important experimental platform for investigating SOA formation from biomass-burning emissions. This review systematically summarizes the structural configurations, construction materials, and oxidant systems of OFRs, with particular emphasis on their applications in studying SOA formation from the aging of VOCs, SVOCs, and IVOCs emitted from biomass burning. Briefly described the characteristics of OFRs and smog chambers in atmospheric simulation. Finally, current challenges in applying OFRs to biomass-burning SOA research are discussed, and future research priorities are proposed to improve their atmospheric representativeness and mechanistic interpretation.

Atmosphere doi: 10.3390/atmos17111072

Authors: Xuefei Wang Yongliang Luo Buncha Wattana Hua Zeng Xiaohua Peng

Mega sporting events report carbon footprints, but rarely describe the air pollutants that accompany them. This study links the verified greenhouse-gas inventory of the Beijing 2022 Winter Olympics (713,788 t CO2e, 2016–2022) to co-emitted NOx, SO2, PM2.5, CO and NMVOC in a screening-level, emission-side assessment using eight emission carriers with Chinese pollutant-to-CO2 ratios, Shapley attribution, and Monte Carlo and Sobol analyses. Construction accounted for 63% of the CO2e, an estimated 70% of the 1335 t of NOx, and 76% of the 279 t of PM2.5, remaining the largest NOx source in 99.9–100% of uncertainty draws and under alternative attributions. The measures taken, chiefly the employment of green electricity, avoided 31% of no-measure emissions and 18% of NOx under market-based accounting, but only 8% of NOx if green certificates were not used. Under literature-based assumptions, eight additional levers could have cut CO2e by 46% (41–53%) and NOx by 49%; zero-emission construction machinery avoids the most NOx per tonne of CO2. A residual of 384,375 t would remain; the offsets retired (2.4 times the emissions) reduced no air pollutants. For construction-dominated events, such as this pandemic-era edition, climate- and air-quality co-benefits are decided largely at the design stage, through materials and machinery, rather than by offsetting.

Atmosphere doi: 10.3390/atmos17111071

Authors: Ahmad Muhammad Fatih Külahcı

The Source–Environment–Response (SER) framework introduced a decomposition of radon transport into source-driven and environmental components, but treated the environment as a passive correction rather than an active memory accumulator. We present QSER (Generalized Source–Environment–Response), which extends the SER framework by making memory a central property of the radon transport phenomenon. The framework separates the governing transport operator into conservative (advective) and dissipative (diffusive and decay) parts, revealing the observed concentration as the difference between a conservative source field S and an environmental field E that accumulates the history of dissipative interactions. The continuity equation quantifies how the two fields interact to produce the observed concentration, and dissipation is the mechanism that writes history into E. For the forward case, the QSER steady plume is consistent with the Gaussian-plume structure of industry-standard atmospheric dispersion models (e.g., RESRAD-OFFSITE), exactly at the level of the Gaussian and sector structure and structurally at the level of the model-specific closures. For the inverse case, three routes are formulated and compared. The direct route exactly recovers the traditional fitting method, and the flexibility of QSER is demonstrated. QSER allows for hybridization with different methods and adds a physical layer to current forward and inverse approaches, giving additional insight into how radon transport processes evolve. The framework supports parameter estimation for radon monitoring applications in lithosphere–atmosphere coupling, earthquake precursor studies, uranium exploration, and atmospheric transport.

Atmosphere doi: 10.3390/atmos17111070

Authors: Kun Zhang Yanshi Huang Liping Lü Quanyang Hao Pingbing Zuo Fengsi Wei

The May 2024 super geomagnetic storm provided a rare opportunity to investigate the full-chain impact of extreme space weather on the thermosphere and Low Earth Orbit (LEO) satellites. This study presents, for the first time, an end-to-end quantitative analysis that links high-latitude energy input, thermospheric density response, multi-model performance evaluation, and satellite orbital decay. High-latitude Joule heating increased from ~2 mW/m2 during the quiet period to 112 mW/m2 during the main phase. The hemispherically integrated heating peaked at 1822 GW, and the enhanced heating persisted for approximately 18–20 h. Swarm-A observations show that the orbital-mean neutral density at ~470 km increased by a factor of 4.5 relative to quiet-time levels. However, DTM2020, JB2008, NRLMSIS 2.0, and TIEGCM2.0 all underestimated the magnitude of this enhancement, with mean absolute percentage error (MAPE) values of 32.6%, 39.0%, 60.8%, and 40.6%, respectively. Coupling these models to the High Precision Orbit Propagator (HPOP) shows that the China Space Station (CSS) decay rates are 0.5–0.7 km/day during the main phase, about 2–3 times the quiet-time values. These predictions agree with Two-Line Element (TLE) derived semi-major axis decay trends, but significant magnitude discrepancies persist among different models, indicating that thermospheric model selection is a crucial uncertainty source in orbit prediction. A statistical analysis of Space-Track TLE data for LEO population (<1000 km) quantifies decay distribution across altitudes, providing direct references for space situational awareness and orbit maintenance strategies.

Atmosphere doi: 10.3390/atmos17111068

Authors: Yi Qu Xiuyong Zhao Jiawei Wan Jingxin Xu Wenxin Tian

Urbanization significantly affects regional climate, and accurately representing urban canopy processes is crucial for simulating heatwaves in highly urbanized regions such as North China. To evaluate the impacts of different urban canopy schemes on surface parameter simulation, this study compares the No Urban Canopy Model (NOUCM), Single-Layer Urban Canopy Model (SLUCM), and Multi-Layer Urban Canopy Model (MLUCM) during a heatwave in North China from 5 to 7 July 2023. The simulations use WRFv4.6.1 and are verified against observations from 141 urban stations. NOUCM shows a positive bias in 10 m wind speed. SLUCM introduces a building drag effect, but its single-layer representation cannot resolve the vertical distribution of drag, leading to a 15.8% increase in RMSE relative to NOUCM. MLUCM explicitly represents the building drag effect and reduces RMSE by 29.5%. For 2 m temperature, all schemes show positive biases. SLUCM produces cooler daytime and warmer nighttime temperatures, with a large nighttime bias. MLUCM more realistically simulates shading and radiation trapping, reducing the daytime RMSE by 18.2% and yielding nighttime biases close to zero. For 2 m specific humidity, SLUCM and MLUCM reduce RMSE by 1.8% and 0.7%, respectively. For urban heat island (UHI) intensity, MLUCM achieves an overall RMSE of 0.721 °C, which is 30.2% and 33.2% lower than NOUCM and SLUCM, respectively. It also produces the smallest daytime and nighttime biases, with good agreement with observations during 12–20 h. SLUCM shows a nighttime UHI bias of +1.094 °C, while NOUCM shows a daytime bias of +1.176 °C. Overall, MLUCM demonstrates superior performance in simulating wind speed, temperature, and UHI intensity, though its advantage in 2 m specific humidity is limited. These findings provide a reference for selecting urban canopy schemes in operational numerical weather prediction over highly urbanized regions.

Atmosphere doi: 10.3390/atmos17111069

Authors: Hanxiao He Yuchun Wang Shanze Li Jie Wen Yufei Bao Yong Qin Meng Sun Nanping Wu Qiudi Dan

Reservoir water-level fluctuation zones (WLFZs) are important in carbon budgets, but opaque chambers omit daytime photosynthesis. We measured CO2 and CH4 fluxes at the air–water and air–soil interfaces of Hanfeng Reservoir (water-level amplitude 4.72 m), with three-chamber partitioning in the WLFZ during the growing season. The air–water interface was a weak CO2 source (2.1 ± 40.9 mg m−2 h−1) and a persistent CH4 source (0.71 ± 1.36 mg m−2 h−1). In the WLFZ, daytime gross primary production (1665 mg m−2 h−1) exceeded ecosystem respiration (1280 ± 757 mg m−2 h−1; daytime net ecosystem exchange NEE, −385 ± 1180 mg m−2 h−1), whereas the estimated 24 h NEE indicated a weak net CO2 source (+153 to +343 mg m−2 h−1). Air–soil CH4 emissions averaged 0.70 ± 1.51 mg m−2 h−1, with no significant net plant-mediated transport. At this site, and under modeled night-time scenarios, the opaque chamber overestimated daily net CO2 emissions 3.7 to 7.1-fold. In 24 h measurements of two dominant species at one site in August 2024, 09:00–13:00 sampling represented the daily mean CH4 flux within 15%, but not the CO2 flux. These site-specific results indicate that accounting for vegetation and diel variability can reduce bias in WLFZ carbon budgets.

Atmosphere doi: 10.3390/atmos17111067

Authors: Jean Carlos Ruiz-Hernández Steven Ramos Romero Arquímides Haro Velasteguí Iván Chávez Velasco Mariela Moreno Palacios Jorge Lara-Sinaluisa Nelly Perugachi Cahueñas

Precipitation variability in the tropical Andes is strongly influenced by complex topography and regional hydroclimatic processes, yet the spatial organization and temporal evolution of precipitation regimes remain insufficiently characterized. This study identifies dominant precipitation regimes in the Upper Pastaza River Basin, Ecuador, and examines their variability during 1981–2025 using daily precipitation data from the Climate Hazards Group InfraRed Precipitation with Stations (CHIRPS) dataset. CHIRPS estimates were evaluated against observations from three meteorological stations, revealing spatial differences in precipitation estimation accuracy and a general underestimation of monthly precipitation totals. Empirical Orthogonal Function (EOF) analysis and K-means clustering were applied to identify four recurrent precipitation Weather Types (WTs), ranging from a dominant low-intensity regime (WT1), representing approximately 70% of classified days, to a relatively infrequent but intense precipitation regime (WT2), accounting for approximately 1.7%. Seasonal and interannual variability was characterized through regime occurrence frequencies and temporal recurrence patterns. Monotonic trends were assessed using Kendall’s rank correlation coefficient and Sen’s slope estimator, while the burstiness index was used to examine temporal irregularity. A statistically significant positive trend was identified in WT2 occurrence (Kendall’s τ = 0.3830, p = 0.0002), whereas no significant trends were detected for the remaining regimes. Comparison with the recent period (2015–2025), using a consistent classification framework, revealed differences in regime occurrence while preserving broadly similar spatial precipitation patterns. These findings highlight changes in the occurrence of specific precipitation states rather than a generalized redistribution of all regimes, providing insights into hydroclimatic variability across the Andes–Amazon transition zone.

Atmosphere doi: 10.3390/atmos17111066

Authors: Ali Alhazmi Talal Daghriri

Accurate daily solar radiation estimates are needed to plan solar energy systems in arid regions, where long ground records are unavailable at most candidate sites. This study evaluates metaheuristic-optimized Random Forest models for daily global horizontal irradiance (GHI) at six climatically diverse Saudi sites (Riyadh, Tabuk, Najran, Dammam, Jeddah, and Turaif), using routinely available meteorological variables from the NASA POWER database for 2001–2024 (training 2001–2019, testing 2020–2024). Five metaheuristic optimization algorithms—the Genetic Algorithm (GA), Particle Swarm Optimization (PSO), the Firefly Algorithm (FFA), Elephant Herding Optimization (EHO), and Grey Wolf Optimization (GWO)—were used to tune Random Forest hyperparameters under a fairness-controlled experimental design in which all algorithms received an identical number of fitness evaluations, search bounds and fitness function, with every configuration repeated over ten random seeds. Across the six sites, test MAPE ranged from 4.75% to 6.97% and R2 from 0.772 to 0.945, while the five algorithms differed by at most 0.023 percentage points of MAPE within any site. Neither a Friedman test across sites (MAPE: χ2 = 5.20, p = 0.267; R2: χ2 = 2.67, p = 0.615) nor within-site Kruskal–Wallis tests found a statistically significant difference among the algorithms. Bootstrap confidence intervals, leave-one-year-out validation and bias analysis by sky condition show that the remaining error is concentrated on cloudy and dust-affected days, and that Najran’s lower R2 reflects its small seasonal variance rather than larger errors. An untuned Random Forest performed almost identically to the tuned models. The choice of metaheuristic optimizer is therefore far less consequential for accuracy than site climate and input data quality.

Atmosphere doi: 10.3390/atmos17111065

Authors: Xiong Song Huawei Xiang Lanqiang Gong Wenyue Xue Zijun Luo Menglong Wang Chunling Zhou Hong Zhang Zhe Cai Qiang Liu

High-altitude wind energy offers a vast and comparatively stable renewable resource, yet its development in China lacks systematic assessment that integrates generation system characteristics. This study establishes a multi-scale evaluation framework for China’s 1000 m altitude wind resources by coupling 45 years of ERA5 (ECMWF Reanalysis v5) data for 1980–2024, four nested WRF (Weather Research and Forecasting) simulations at 27, 9, 3 and 1 km for 2000–2024, and a standardized airborne wind energy power curve. The framework proceeds from reanalysis-based coarse screening, through WRF-based refinement, to a site-specific assessment, and it quantifies the effective wind speed, the wind power density and the annual equivalent full-load hours on a common basis. The results reveal pronounced regional and altitude-dependent distribution patterns. Core resource-rich zones, including the northern Qinghai–Tibet Plateau, central-eastern Inner Mongolia, and northeastern coastal areas, exhibit annual equivalent full-load hours exceeding 3000 h, with specific regions surpassing 4000 h, and 7.33% of mainland China exceeds the 4000 h screening threshold while 32.96% exceeds 3000 h. A two-stage siting procedure, combining reanalysis-based coarse screening with meteorological-model refinement, identifies 19 candidate grid cells in Inner Mongolia, Qinghai, Tibet and Northeast China and ranks them by resource quality. Detailed assessment of a representative coastal site in Jilin Province validates the framework’s reliability, and it reproduces the observed wind speed, temperature and relative humidity at 1000 m above ground level, characterizing the prevailing southwesterly–westerly regime, its seasonal cycle and its diurnal variability. The generation-potential values reported here are idealized, resource-based estimates rather than predictions for a specific device. This study addresses critical gaps in China’s high-altitude wind resource quantification, providing scientific foundations and technical guidance for strategic planning, engineering feasibility studies, and commercial deployment of high-altitude wind power systems.

Atmosphere doi: 10.3390/atmos17111064

Authors: Murat Durak

This study examines the spatial and temporal variability of Tourism Climate Index (TCI) values across selected meteorological stations in Türkiye during 2000–2025. Using observations from 29 meteorological stations, the analysis investigates regional differences in TCI-derived climatic suitability and evaluates temporal trends at selected coastal locations. The study applies TCI as an established tourism climatology assessment framework; it does not constitute an empirical validation of the index against tourist preferences or physiological measurements. The Tourism Climate Index (TCI) is a valuable tool for assessing climatic suitability in tourism destinations in Türkiye, particularly in the Aegean and Mediterranean regions. In this study, data from 29 meteorological stations and several major coastal tourism destinations, including Antakya, Mersin, Antalya, Marmaris, Bodrum, Çeşme, İzmir, and Çanakkale, were analyzed to calculate TCI values and evaluate tourism climate conditions. Biometeorological research related to tourism and recreation has received increasing attention because tourism activities are strongly influenced by climatic conditions. Consequently, assessing the potential impacts of climate change on tourism and recreation has become an important research priority. This study contributes to the biometeorology literature by investigating the spatial variability of tourism-related bioclimatic conditions in Türkiye using the Tourism Climate Index (TCI). The analysis is based on meteorological data obtained from stations located along the coastal corridor extending from Çanakkale to Antakya. The results provide insights into the spatial distribution of climatic suitability for tourism across Türkiye’s coastal regions and highlight the role of climate as a key factor influencing tourism development. The findings are expected to support tourism planning and management by improving understanding of regional tourism climate conditions and raising awareness of the importance of climatic factors in shaping tourism activities. The Tourism Climate Index (TCI) values indicate that the Aegean coastal zone generally exhibits more favorable tourism climate conditions than the Mediterranean coastal zone. This difference is primarily attributable to the more moderate thermal conditions and stronger wind regimes characteristic of the Aegean region, which contribute to improved outdoor comfort, particularly during the summer months. In contrast, the Mediterranean coast is more frequently affected by higher temperatures and greater heat stress, which can reduce climatic suitability despite its abundant sunshine and limited precipitation. As a result, the Aegean coast tends to achieve higher TCI scores and provides more comfortable conditions for a wider range of tourism activities, including beach tourism, urban sightseeing, nature-based recreation, and shoulder-season tourism. Trend analysis using the Mann–Kendall test and Sen’s slope estimator revealed statistically significant decreasing trends (p < 0.001) in the Tourism Climate Index (TCI) at seven of the eight coastal stations. Sen’s slope values ranged from −0.18 to −0.28 TCI units year−1, with the greatest declines observed at Antalya and Antakya. Marmaris was the only station that showed no statistically significant trend (p = 0.73). Overall, the results indicate a consistent deterioration in climatic suitability for tourism across the majority of the study area during the study period.

Atmosphere doi: 10.3390/atmos17111063

Authors: Carlos Blanco-Alegre Estela Domingos Vicente Ana Isabel Calvo Fernanda Oduber Célia Alves Roberto Fraile

This study investigates particle number concentrations (PNCs), particle number size distributions (PNSDs), and equivalent black carbon (eBC) concentrations during ironing with a typical steam iron and a steam iron with a portable boiler under different ventilation conditions. Measurements were conducted in a residential environment under low- and higher-ventilation conditions. Ironing caused a rapid and substantial increase in PNC, with concentrations increasing by up to approximately two orders of magnitude during the activity, particularly for particles between 20 and 80 nm. The maximum PNC reached 7.12 × 105 particles cm−3 for the typical steam iron under low-ventilation conditions. Particle concentrations decreased after ironing, although the decay was slower for particles larger than approximately 80 nm. Estimated particle number emission rates ranged from 8.1 × 1011 to 1.5 × 1012 particles min−1, with the highest emission rate obtained for the steam iron with a portable boiler under low-ventilation conditions. PNSDs were fitted using one or two lognormal distributions, depending on the shape of the measured distributions. Ironing produced a shift in the modal particle diameters towards larger sizes, particularly during the post-ironing period. Higher ventilation reduced particle concentrations and accelerated particle removal from the indoor environment. eBC concentrations showed only modest changes during ironing, with a slight increase under low-ventilation conditions. These results demonstrate that ironing can be an important indoor source of ultrafine and fine particles and that ventilation plays an important role in controlling their indoor concentration and persistence.

Atmosphere doi: 10.3390/atmos17111062

Authors: Charles Johnson Paul Miller Neil Debbage

Aerosol influences on local thunderstorms are complex and often contradictory. Within aerosol-laden city environments, these interactions are further complicated by additional thermodynamic and kinematic effects related to the urban landscape. Observational studies are limited in their ability to distinguish the different influences of urban land cover and aerosols on thunderstorms, but numerical modeling can alleviate the problem by systematically accounting for each variable separately. This study used the NASA Unified-Weather Research and Forecasting (NU-WRF) model to simulate frontal, scattered thunderstorms during a summer day in coastal Louisiana. The study period was simulated by iteratively including and excluding atmospheric chemistry and the physical attributes of the urban land cover, producing a total of four simulations. Three sub-zones, distinguished by their thunderstorm activity, level of urbanization, and aerosol concentrations, were selected for more detailed analysis to examine the sensitivity of precipitation and thunderstorm coverage to the experimental conditions (i.e., atmospheric chemistry and urban landscape). The aerosol and urban landscape simulations demonstrated a minimal impact on storm initiation timing, suggesting that these processes were secondary to the synoptic atmospheric conditions driving the initial development of convection. However, the quantity of precipitation and reflectivity coverage was more sensitive to the experimental modeling scenarios, with the individual simulations (i.e., aerosol or urban effects only) frequently overproducing precipitation compared to observations. Whether the aerosol or urban effects were the more influential factor was regionally dependent and varied between sub-zones. When both aerosol and urban effects were activated, the precipitation and reflectivity coverage aligned more closely with NASA’s Integrated Multi-SatellitE Retrievals for GPM (IMERG) estimated rainfall across much of the study area. Over the heavily urbanized coastal region, the combined model run with both aerosol and urban effects overproduced the rainfall and reflectivity coverage, whereas on a regional scale it modulated rainfall quantity, creating a non-linear precipitation reduction in most regions. These results suggest that urban and aerosol properties weakly influence the meso-beta meteorological pattern but can drive meso-gamma-scale changes that accumulate over larger areas.

Atmosphere doi: 10.3390/atmos17111061

Authors: Züleyha Reçber Mikail Safa Zehra Işık Yasin Paşa Mehmet Ali Çelik

This study evaluated meteorological drought conditions in Tuzluca District (Iğdır, Türkiye) over the period 1960–2024 using four complementary drought indices: the Standardized Precipitation Index (SPI), Standardized Precipitation–Evapotranspiration Index (SPEI), Percent of Normal Index (PNI), and Rainfall Normalized Index (RNI), calculated at 1-, 3-, 6-, and 12-month time scales. Daily precipitation, temperature, and evaporation data obtained from the 16th Regional Directorate of the Turkish State Meteorological Service (MGM) were quality-controlled and used for index computation. The results revealed substantial temporal variability in drought and wetness conditions across short-, medium-, and long-term accumulation periods. Short-term indices (1- and 3-month) effectively captured rapid monthly and seasonal fluctuations, whereas longer accumulation periods (6- and 12-month) highlighted the persistence of extended dry and wet episodes. The longer accumulation scales revealed pronounced negative anomalies around 1999–2003 and renewed dry conditions during parts of the early 2020s. Trend analysis using the modified Mann–Kendall test identified statistically significant increasing trends in the annual mean SPI-12, PNI-12, and RNI-12 series, with Z statistics of 2.0191, 1.9997, and 2.0299, respectively (p < 0.05). In contrast, SPEI-12 exhibited a decreasing but statistically non-significant tendency (Z = −1.1451, p = 0.2522). The corresponding Sen’s slope estimates were 0.0147 index units per year for SPI-12, −0.0063 index units per year for SPEI-12, 0.3527 index units per year for PNI-12, and 0.0142 index units per year for RNI-12. These findings demonstrate that the direction and statistical significance of the detected trends differed among the indices. Overall, this study highlights the importance of combining precipitation-based and water-balance-based drought indices across multiple accumulation periods to characterize temporal drought variability and support regional water-resources management, agricultural planning, and meteorological drought early-warning systems in semi-arid regions.

Atmosphere doi: 10.3390/atmos17111060

Authors: Hiroshi Okochi Yasuhiro Ishihara Masaki Takeuchi Mizuo Kajino

Micro- and nanoplastics are now recognized as emerging components of atmospheric particulate matter [...]

Atmosphere doi: 10.3390/atmos17111059

Authors: Chaobiao Meng Liyan Zhang Yifan Xie Yumeng Gao Jinghao He Keren Yang Feiyan Xu Weihan Wang Yanbing Chi Si Chen

Rice paddies are a major agricultural source of greenhouse gas emissions. However, balancing emission reduction with food security remains a core challenge for climate-smart rice production management. This review synthesizes current evidence on water-saving irrigation (WSI), biochar, and their combined effects on methane (CH4), nitrous oxide (N2O), global warming potential (GWP), and greenhouse gas intensity (GHGI) in paddy fields, emphasizing interaction mechanisms and environmental and management constraints. WSI generally suppresses CH4 emissions but can increases N2O emissions during drying–rewetting transitions. Mean reductions of approximately 43–52% in CH4 emissions and 37–47% in GWP have been reported in studies of alternate wetting and drying relative to flooded controls. Biochar can modify these responses through changes in soil aeration, pH, carbon availability, and nutrient retention. Under WSI, reported N2O decreases following biochar addition ranging from 19.3% to 62% relative to no-biochar controls under the same irrigation regime, although these decreases are not uniformly significant, and nonsignificant responses also occur. Additional CH4 mitigation is likewise context-dependent, with increases reported under some conditions. When CH4 reductions outweigh N2O increases in CO2-equivalent terms while yield remains stable, combined management can lower field-emission GWP and GHGI. A net benefit across the full production system additionally depends on carbon storage and upstream emissions. Excessive drying, poorly coordinated nitrogen inputs and unsuitable biochar–soil combinations can weaken these benefits. Uncertain aging effects, limited long-term multi-site evidence and inconsistent accounting boundaries constrain the assessment of persistence and transferability. Integrated WSI–biochar management therefore offers conditional mitigation potential, requiring locally evaluated coordination of water, nitrogen, and biochar inputs rather than uniform prescriptions.

Atmosphere doi: 10.3390/atmos17111058

Authors: Zhanat Baigazinov Yelena Polivkina Nurlan Mukhamediyarov Merey Jambaev Valeriy Monaenko Aliya Zhadyranova Sayan Salmenbayev Laura Kenzhina

Atmospheric metal pollution in urban ecosystems is a critical global issue. Feral pigeons (Columba livia) serve as effective bioindicators of ambient air quality. This pilot study evaluates their applicability across three contrasting regions in Kazakhstan: Ust-Kamenogorsk (non-ferrous metallurgy), Pavlodar (coal-fired power generation), and Bigash village (rural background). Concentrations of 14 metal(loid)s and 10 rare earth elements (REEs) in feathers were determined via ICP-MS/ICP-OES and analyzed using principal component analysis, hierarchical cluster analysis, and pollution indices. In Ust-Kamenogorsk, high contamination factor (Cf) values for Cd (26.0), Pb (41.2), and As (19.0) indicated ‘very high contamination’ from non-ferrous metallurgy. In Pavlodar, Cf values for Cd (5.0), As (5.0), and Cu (7.3) showed ‘considerable contamination’ linked to coal combustion and oil refining. The total Metal Pollution Index for Ust-Kamenogorsk (11.1) was 4.8 times higher than for Pavlodar (2.3). REE profiles showed a ‘dilution effect’ driven by industrial dust, except for gadolinium, for which a pronounced concentration peak was recorded in Pavlodar. Multivariate statistics confirmed the effectiveness of using pigeon feathers as an integrative tool to differentiate zones of high technogenic pressure. Overall, this pilot study lays the foundation for long-term biomonitoring of the urban environment in industrial centers of Kazakhstan with heavy pollution loads.

Atmosphere doi: 10.3390/atmos17111057

Authors: Zhe Wei Caixia Wen

Fine particulate matter (PM2.5) and ozone (O3) are the two dominant air pollutants in China, yet their coupling relationship at the daily urban scale remains debated because observational correlations are often confounded by meteorology. This study investigates the nonlinear PM2.5–O3 relationship in Puyang, a typical oil and gas industrial city in the “2 + 26” region, using daily observations from 2015 to 2025. A set of progressively nested generalized additive models were constructed, adding meteorological factors and NO2 sequentially, followed by humidity and season interaction terms. Stratified correlation analyses revealed a concentration-stratified reversal: positive in the low-concentration segment (<35 μg·m−3; r = 0.271) and negative in the mid (35–75 μg·m−3; r = −0.208) and high (>75 μg·m−3; r = −0.313) segments. After controlling for meteorology and NO2, however, the conditional partial effect increased over most of the PM2.5 range, indicating that the uncontrolled negative correlation largely disappeared after controlling for meteorology. Humidity exerted a significant conditional modulating effect (p < 0.001), whereas seasonal modulation was concentrated in autumn (p < 0.001). The positive correlation in the low-concentration segment showed a suggestive increase over time (0.190 to 0.306), while the negative correlation in the high-concentration segment weakened (−0.380 to −0.198), indicating an asymmetric redistribution of the PM2.5–MDA8 O3 coupling during the clean-air transition. These findings suggest that coordinated control of PM2.5 and MDA8 O3 should be designed by concentration interval rather than as a single-direction linear reduction, with precursor-ratio regulation needed in the low-concentration range.

Atmosphere doi: 10.3390/atmos17111056

Authors: Siyi Chen Xiaohui Du Jie Zhu Yun Shu Hong Sun Han Wang Linlin Ma Weiyan Lin Ting Niu Zhigang Liu

Rapid urbanization and industrial expansion in energy-intensive regions continue to hinder the simultaneous mitigation of air pollutants and CO2 emissions, yet effective synergistic mitigation pathways remain poorly understood because highly industrialized border regions exhibit distinct emission structures and spatial heterogeneity. Here, we combined a newly developed georeferenced emission inventory comprising approximately 3600 industrial point sources with the Greenhouse Gas–Air Pollution Interactions and Synergies model to assess future air pollutant and CO2 emission trajectories in the Urumqi–Changji–Shihezi (UCS) region during 2021–2035 under alternative policy scenarios, including a policy scenario representing the implementation of China’s Air Quality Continuous Improvement Action Plan. We found that emission co-hotspots were concentrated within the central UCS corridor, particularly around central Urumqi, where power and industrial activities jointly dominated emissions. Under the assumptions specified for the policy scenario, annual emissions of PM2.5, SO2, NOx, VOCs and CO2 were projected to decrease by 43%, 38%, 40%, 41%, and 34%, respectively, by 2035 relative to 2021, whereas NH3 emissions were projected to decline by only 5%. Owing to its high baseline emissions, Changji Prefecture contributed more than half of the regional emission reductions. Unlike the national transition, where power-sector emissions have largely been controlled, the greatest mitigation potential in this region remained concentrated in the power and industrial sectors. Expanding renewable electricity achieved the strongest air pollution–carbon co-benefits while avoiding the emission penalties associated with electrification under a coal-dominated electricity system. In addition, controlling industrial NH3 slip emerged as a critical local strategy complementing agricultural ammonia mitigation. These findings demonstrate that accelerating clean power development together with region-specific industrial emission controls is essential for maximizing synergies between air-pollutant emission reductions and climate mitigation.

Atmosphere doi: 10.3390/atmos17111055

Authors: Saleh A. Aloraini Abdullah S. Alnasser

Backward-trajectory statistics locate arid-region dust sources on the assumption that trajectories separate emission from transport, an assumption seldom tested where the local contribution is known. We build one from ten years (2016–2025) of routine METAR observations at Riyadh, Saudi Arabia, where present-weather codes distinguish locally lifted dust (BLDU/BLSA) from dust in suspension (DU/SA). Suspension-dominated days outnumbered locally lifted days 345 to 179 (1.9:1), and ERA5-driven models reproduced the locally lifted class substantially better (held-out AUC 0.914 versus 0.765), indicating distinct meteorological signatures. Dust activity then fell 85.9% below the 2016–2022 baseline in 2023–2024, recovering only to 56.0% below it in 2025. The model did not anticipate this: observed dust days in 2023–2024 were 13–27% of the calibrated expectation in every class while local surfaces in 2024 stayed favorable for emission; upwind dust availability and transport remain hypotheses. We then computed PSCF and CWT fields from 14,612 HYSPLIT trajectories, running the locally lifted class, for which a distant source is not required, as an imperfect negative control, tested against permutation nulls. Under unmatched and season-matched backgrounds, the control was more enriched than the suspension-dominated class over the eight a priori candidate regions (p ≤ 0.007), and wind matching removed most regional structure from both. Both fields are shaped largely by the shared Shamal transport climatology and the wind selection of dust days, limiting their ability to identify emitting terrain. At wind-selected, single-regime receptors, PSCF and CWT source fields should be interpreted alongside such a control.

Atmosphere doi: 10.3390/atmos17111054

Authors: Chae-Ho Kim Dong-Chul Shin

This study developed a machine-learning-based detection-alarm algorithm using hydrogen sulfide (H2S) and ammonia (NH3) as core chemical inputs to determine the analytical detection status of 20 odor compounds and evaluated its external applicability at an independent facility. Data collected at 20 min intervals from three odor-emitting facilities were used, with detection defined as a concentration at or above the compound-specific detection limit. Data from a livestock farm and a wastewater treatment facility were used to develop a meteorological–temporal baseline model (M0b) and an alarm model additionally incorporating H2S and NH3 (M4). Model selection, probability calibration, and compound-specific thresholds were determined using nested site–week cross-validation, and the finalized models were applied unchanged to Site C. In the development dataset, mean average precision (AP) increased from 0.362 for M0b to 0.463 for M4 (ΔAP = +0.101), with improvements for 17 of 20 compounds. At Site C, mean AP increased from 0.424 to 0.443 (ΔAP = +0.019), with improvements for 13 compounds. Excluding exploratory methyl isobutyl ketone, mean ΔAP was +0.106 and +0.020, respectively. External evaluation included only four calendar-week groups, yielding a minimum attainable two-sided permutation p-value of 0.125. No compound-specific improvement reached Benjamini–Hochberg-adjusted significance (q < 0.05) in either dataset; the external findings are therefore exploratory. Under fixed-threshold operation, dimethyl sulfide achieved a precision of 0.921 and a false-positive rate of 0.057, but recall was 0.182. These results suggest that H2S and NH3 may provide useful additional information for screening multiple odor compounds, while emphasizing the need for site-specific threshold assessment and further independent validation.

Atmosphere doi: 10.3390/atmos17111053

Authors: Dan Zhang Qiang Ma

Whether advances in global weather forecasting translate into local forecast skill requires regional verification. Using observations at 117 stations in Inner Mongolia during the June–August 2026 flood season, we compared three publicly distributed forecast fields at 6–84 h lead times: the Integrated Forecasting System control forecast (IFS-Control), the Artificial Intelligence Forecasting System (AIFS), and WeatherNext 2’s 64-member ensemble mean (WN2). Because WN2 is an ensemble mean and the others are single deterministic forecasts, this compares operational products, not the underlying models. Verification of 2 m temperature and precipitation used Murphy error decomposition and block-bootstrap sensitivity tests. Temperature mean absolute error was approximately 15.7% lower for AIFS and 13.1% lower for WN2 than for IFS-Control, mainly through smaller correlation-error terms, though IFS-Control had the smallest systematic bias. Both AI products outperformed IFS-Control for 6 h moderate rain (≥4 mm). IFS-Control robustly outperformed AIFS for 6 h heavy rain (≥13 mm), but its advantage over WN2 did not survive the stricter multiple-comparison correction. Both AI products under-forecast heavy-rain frequency and extremes; a limited member experiment suggests ensemble averaging may contribute, though not separably from model differences. Findings apply only to the product versions, region, and season evaluated.

Atmosphere doi: 10.3390/atmos17111052

Authors: Ola Abu Elezz Krishnendu Sekhar Paul Blessy Varghese Ankitha Nandakumar Haris Haralambous Tiju Joseph Mathew

Spread sporadic-E (Spread-Es) appears in ionograms as diffuse and range-spread Es echoes associated with spatial structuring of the reflecting layer. In this study, five years (2020–2024) of Digisonde ionograms from three European mid-latitude stations—Juliusruh (54.6° N, 13.4° E), Dourbes (50.1° N, 4.6° E), and Roquetes (40.8° N, 0.5° E)—were visually inspected to investigate the seasonal, diurnal, and interannual variability of Spread-Es during the ascending phase of Solar Cycle 25. The Daily sunspot number is included only to provide solar-activity context for the 2020–2024 observational interval; no statistical assessment of solar-activity dependence is attempted. Spread-Es occurrence exhibited a pronounced summer and nighttime enhancement at all three stations. The normalised summer occurrence was 16.9% at Juliusruh, 8.6% at Dourbes, and 12.5% at Roquetes, compared with winter values of 5.6%, 2.6%, and 4.8%, respectively. The maximum monthly–hourly occurrence rates were approximately 0.48, 0.35, and 0.40, respectively, while Roquetes showed a broader post-midnight distribution. A representative multi-station interval on 12–13 June 2022 was further examined using ionograms, Height–Time–Intensity (HTI) representations, and regional detrended total electron content (dTEC) observations. Spread-Es and spread-F signatures coexisted at all three stations, with Spread-Es simultaneously present from 21:25 to 22:13 UT. The corresponding dTEC maps showed predominantly southwestward-propagating regional perturbations with NW–SE-oriented phase fronts. The derived phase speeds were 180–210 m s−1, with periods of 48–64 min and horizontal wavelengths of approximately 518–806 km, characteristics consistent with nighttime MSTID activity over the European sector. Thus, the selected event shows the temporal coexistence of Spread-Es, spread-F, and regional MSTID-scale activity, while the available observations do not establish causal E–F electrodynamic coupling.

Atmosphere doi: 10.3390/atmos17111051

Authors: Yuxin Sun Zhiyuan Cai Weifang Man Tao Li Zheng Mao Weidi Huang Shenqi Zhang Jie Wu Chang Zhai

Methanol is a promising low-carbon fuel for large marine engines, but reliable simulation of methanol–diesel combustion depends on selecting diesel-surrogate chemistry that remains credible for both ignition and emissions. Four diesel-surrogate mechanisms were therefore assessed through engine-scale CFD validation and controlled homogeneous-reactor calculations before application to a low-speed two-stroke marine engine. At the diesel-only full-load validation point, all four mechanisms reproduced the measured pressure history, although their combustion-phasing, fuel-consumption, and NOx predictions differed. LLNL 3.0 gave the closest peak pressure and CA50, whereas the Pei mechanism provided the most balanced agreement across pressure, phasing, fuel consumption, and NOx. Closed adiabatic constant-volume calculations at 960 K, 160 bar, and φ = 1.0 revealed substantial differences in ignition delay and radical evolution, confirming that similar engine pressure traces can conceal distinct intrinsic kinetic responses. Pei was consequently applied to cases with 0%, 50%, 75%, and 90% methanol energy substitution at the baseline full-load boundary conditions. As only the diesel-only condition was experimentally validated, the substitution results represent relative model predictions rather than validated absolute values. Relative to diesel-only operation, 50% substitution advanced CA50 by 5.02 °CA, increased indicated thermal efficiency from 48.4% to 52.3%, and increased predicted NOx by 92.5%. At higher substitution, this response weakened: the 90% case returned CA50 and NOx close to the diesel baseline and strongly reduced CO, whereas predicted unburned methanol and formaldehyde reached their maxima at 75% substitution. These results show that mechanism screening is essential before interpreting methanol-substitution effects and reveal an efficiency–NOx–incomplete-oxidation trade-off under the modeled full-load condition.

Atmosphere doi: 10.3390/atmos17111050

Authors: Hui Zhang Hantao Wang Ye Yin Nanshan Zhao Cuihua Chen Chenghua Xie

Low-level wind shear (LLWS) poses a significant hazard to aircraft operating at airports in complex terrain, where wind fields may vary rapidly across the airport vicinity. This study examines the spatial evolution of LLWS-related wind structures at Xining Caojiapu International Airport (ZLXN) using 3°-elevation Doppler wind lidar (DWL) plan position indicator (PPI) observations and a combined shear identification algorithm (CSIA). The CSIA combines range- and tangential-direction gradients of radial velocity to identify enhanced-gradient regions. A convective event on 29 May 2021 exhibited a rapidly developing divergent outflow, whereas a cold-front-induced event on 1 November 2021 featured a propagating shear boundary associated with cold-air intrusion. In both cases, the identified regions were consistent with radial velocity discontinuities and the approximate locations and times of aircraft-reported LLWS encounters. Of 66 pilot reports collected during 2019–2023, 61 were retained for verification, of which 49 were detected, yielding a report-based probability of detection PODPIREP of 80.3% (95% cluster-bootstrap confidence interval: 69.2–89.8%). The observed PODPIREP values were 93.9%, 83.3%, 66.7%, and 28.6% for convective-related, cold-front-induced, downward-momentum-transfer-related, and other events, respectively. These results support PPI-based identification of LLWS-related radial velocity structures, but do not establish overall operational alerting performance.

Atmosphere doi: 10.3390/atmos17111049

Authors: Tingting Shu Qinglin Zhu Xiang Dong Houcai Chen Shoubao Zhang Qidong Chen Changsheng Lu Fang Sun

In this paper, a new method for the remote sensing and monitoring of atmospheric ducts using K-band ground-based microwave radiometers is presented. It explains the principle of using K-band ground-based microwave radiometers to monitor atmospheric ducts and elucidates the mechanisms governing the formation of marine atmospheric ducts, as well as the associated atmospheric ducting phenomena for electromagnetic waves. According to the detection characteristics of K-band ground-based microwave radiometers, the emissivity of K-band ground-based microwave radiometers on the sea surface is calculated, and the obtained emissivity is used to judge whether atmospheric ducting occurs at sea. When the sea emissivity (η) in the microwave range is not between 0.3 and 0.9, it is judged that atmospheric ducting is occurring at sea. Otherwise, it is judged that there is no atmospheric ducting over the sea. In this paper, the feasibility and effectiveness, in the relevant areas, of using K-band ground-based microwave radiometers for the real-time monitoring of atmospheric ducts over the sea are verified. It solves the problem wherein the existing atmospheric-duct-monitoring equipment easily exposes targets and struggles to monitor all day in real time.

Atmosphere doi: 10.3390/atmos17111048

Authors: Chan-Yeong Song Keunchang Jang Inhye Kim Yongchan Kim

This study evaluates a Weather Research and Forecasting (WRF) system with a 500 m horizontal grid spacing developed to provide mountain-focused rainfall guidance for forest disaster preparedness in South Korea. The system uses operational 1.5 km Local Data Assimilation and Prediction System (LDAPS) fields provided by the Korea Meteorological Administration as background and lateral boundary conditions. Quality-controlled Automatic Mountain Meteorology Observation System (AMOS) observations, excluding rainfall, are assimilated using three-dimensional variational (3D-Var) data assimilation. The resulting forecasts (DA_AMOS) are compared with LDAPS for the 16–20 July 2025 heavy rainfall event using 500 m gridded and in situ AMOS rainfall observations. Compared with LDAPS, DA_AMOS better reproduced observed accumulated and maximum rainfall magnitudes, with smaller regional station-mean accumulation errors in 13 of 15 administrative divisions with eligible AMOS stations. DA_AMOS generally showed smaller forecast errors at observed rainfall intensities ≥ 10 mm h−1 but larger errors under near-zero-rainfall conditions. During 45 observation-selected extreme-rainfall hours, DA_AMOS reduced mean absolute event-hour bias from 46.14 to 41.74 mm h−1 (9.5%) relative to LDAPS, although substantial underestimation remains. These findings support complementary guidance on rainfall magnitude, though reliable prediction of extreme-rainfall location and occurrence remains unresolved. Evaluation across additional events is required before broader generalization.

Atmosphere doi: 10.3390/atmos17111047

Authors: Hao Yang Yujie Yan Shaohan Li Min Chen

Near-surface wind prediction requires models to represent broad spatial patterns and localized variations. We introduce FDMNet, a frequency-aware dual-branch network for multi-step wind-vector prediction. A variable-wise encoder and multi-head variable attention combine meteorological fields. Latent channels are partitioned between attentive state-space and depthwise-convolution branches, whose outputs are fused using squeeze-and-excitation weights. This partition is motivated by spatial-scale differences without imposing an explicit spectral cutoff. We evaluate the corrected implementation using WeatherBench 2 data from the fifth-generation reanalysis (ERA5) of the European Centre for Medium-Range Weather Forecasts, with 1979–2015 for training, 2016 for validation, and 2017–2018 for testing. The model uses twelve six-hourly input fields to predict the next twelve fields, covering a 72 h horizon. The pooled area-weighted vector root mean square error is 2.7125 m s−1 for FDMNet, 2.7710 for the Temporal Attention Unit (TAU) baseline, and 5.6567 for Persistence. FDMNet reduces the pooled error relative to TAU by 2.11%, with lower errors at ten of twelve leads but higher errors at 12 and 18 h. The two models were trained once each, using different learning rates. Archived regional runs contain an input-grouping error and are not used to validate the corrected model. Repeated training, component ablations, and regional evaluation are needed to assess the mechanism and generality of the observed improvement.

Atmosphere doi: 10.3390/atmos17111046

Authors: Rui Hu Bo Wang Zongpeng Song Huang Ding

Wind power forecast errors grow markedly under complex weather and constrain secure grid integration. Earlier reviews are organised by method lineage, ramp events, uncertainties or systems, not by the error chain. This review follows the error through four steps, from the meteorological driver field and the station wind field to the conversion relationship and the turbine operating state, and separates meteorological errors from power forecast errors. Along this chain, cold fronts, deep convection, low-level jets and tropical cyclones give the error four distinct temporal forms. A timing shift, a random large deviation, a systematic diurnal bias, and a step change in operating state require different techniques. Where samples are scarce and the statistical relationship is unstable, physical constraints and data-driven methods must be fused. The error that remains divides into a part that technical means can reduce and a part set by a predictability ceiling, which does not fall with better resolution or data. Beyond the error itself, the power ramp follows the weather process rather than the forecast error, and its warning reaches dispatch separately. The gaps lie in the availability of observations and paired samples, and dispatch risk metrics cannot yet be converted into accuracy metrics.

Atmosphere doi: 10.3390/atmos17111045

Authors: Erdemalp Ozden Didem Guleryuz

Satellite nitrogen dioxide (NO2) over busy straits mixes ship exhaust with regional pollution. This study tests how much daily shipping activity helps predict extreme-NO2 days at 24 maritime chokepoints once regional NO2 is known, using TROPOMI NO2, ERA5 meteorology, and PortWatch transit and trade-volume estimates. A day was extreme when corridor NO2 exceeded the site’s own 90th percentile. Logistic regression and gradient-boosted trees were developed on 2019–2022 data and tested on 2023–2024, two years set aside in advance. In this test, regional NO2 was by far the strongest predictor, raising average precision (random guessing ≈ 0.11) from 0.17 to 0.39 (logistic) and from 0.25 to 0.67 (gradient-boosted). Previous-day shipping activity added much less; after accounting for regional NO2, it gave a small gain only in the logistic model, mainly at Bab el-Mandeb and the Cape of Good Hope, where the Red Sea crisis sharply changed traffic. Reference areas 10–50 km from the corridor gave the same pattern. Since the models use regional NO2 from the same satellite overpass and concurrent meteorology, they are conditional predictions of same-day extremes, not advance forecasts or early warnings. Interpreting extreme NO2 at maritime chokepoints therefore requires explicit account of the regional atmospheric state.

Atmosphere doi: 10.3390/atmos17111044

Authors: Jingkang Li Jianping Wang Chao Yue Yangan Ren

Topographic slope, aspect, terrain obstruction, and cloud variability jointly produce pronounced spatiotemporal heterogeneity in surface solar radiation (SSR) over complex terrain, while sparse ground observations in mountainous areas limit high-resolution estimation and independent reliability assessment. We therefore developed and tested a terrain-aware semi-empirical SSR framework for data-sparse complex terrain. Its methodological novelty involves the use of one precomputed directional horizon to consistently constrain terrain shadowing, sky-view factor, and terrain-view factor, combined with time-step clear-sky irradiance; daily cloud forcing; and explicit direct, diffuse, and first-order terrain-reflected components. At multiple time scales, the framework was applied to Liancheng (LC), Wuyishan (WYS), and Jinggangshan (JGS), three regions with contrasting terrain and climatic settings. Evaluation combined analytical terrain tests, benchmarks against HORAYZON and r.sun, ground-station observations, leave-one-station-out validation, module ablation, pixel-wise error-frequency analysis, correlation tests, and uncertainty analysis. The directional-horizon scheme represented sky obstruction and terrain shading, with close agreement with HORAYZON results. Annual SSR was 1166.10, 1000.85, and 922.95 kWh/m2 in LC, WYS, and JGS, respectively, yielding the order LC > WYS > JGS; WYS and JGS were strongly dominated by diffuse radiation. Independent station validation produced R2 values of 0.588–0.704 and RMSE values of 40.4–46.6 W/m2, while absolute-error frequency analysis showed that 61.4–70.2% of held-out daily estimates had absolute errors below 40 W/m2. The framework reproduced the principal temporal variability of SSR across the three mountain regions, although performance remained site-dependent and showed systematic underestimation. Uncertainty analysis identified cloud-cover input and its empirical parameters as the main sources of simulation uncertainty. Overall, the framework provides a physically interpretable and computationally tractable approach for regional SSR estimation in observation-sparse mountain environments.

Atmosphere doi: 10.3390/atmos17111043

Authors: Wanjun Du Qing Li Cheng Cao Yibo Ding Xiaowen Wang Li Zhou

Hydroclimatic extremes pose increasing challenges to water security and ecosystem stability, particularly in climate-sensitive transitional regions. Taking the Weihe River Basin in China as a representative monsoon-influenced basin, this study investigated the temporal evolution, spatial heterogeneity, severity, and transition characteristics of flood and drought events. An integrated framework combining the Soil and Water Assessment Tool (SWAT), the Standardized Runoff Index (SRI), the Standardized Soil Moisture Index (SSMI), run theory, three-dimensional drought-event identification, and drought–flood abrupt alternation (DFAA) analysis was developed to reconstruct and characterize historical hydroclimatic extremes. The results showed pronounced spatial heterogeneity in flood characteristics. Floods occurred more frequently in the downstream reaches of the Weihe River mainstream but were generally shorter and less severe, whereas the upstream mainstream and the middle reaches of the Jinghe and Beiluohe Rivers experienced fewer but longer-lasting and more severe flood events. Flood characteristics also varied markedly among representative years, with the 2003 event exhibiting particularly strong spatial differentiation in duration and severity. The mean frequencies of mild, moderate, severe, and extreme droughts were 0.16, 0.11, 0.05, and 0.05, respectively. Droughts occurred more frequently in the northern and western parts of the basin, while severe drought-event centers were mainly concentrated in the central region. The representative 2008–2009 drought originated in the western basin, expanded and intensified toward the central region, subsequently shifted westward while weakening, and finally dissipated in the southern basin. In addition, flood-to-drought events occurred more frequently than drought-to-flood events, whereas drought-to-flood events exhibited slightly greater transition intensity; the southwestern mountainous areas and downstream reaches were identified as major DFAA hotspots. These findings reveal pronounced spatial heterogeneity and directional asymmetry in hydroclimatic extremes and provide a scientific basis for integrated flood–drought monitoring and water-resources management in monsoon-influenced transitional basins.

Atmosphere doi: 10.3390/atmos17111042

Authors: Joong-Bo Kwon Shin-Young Park Ji-Yun Chung Do-Hyeong Kim Cheol-Min Lee Sung-Chul Seo Sung-Won Yoon

Seasonal observations of particulate composition are needed to characterize rural air quality near livestock facilities without assuming that nearby emissions determine ambient aerosol chemistry. This study examined PM2.5 and eight water-soluble inorganic ions in weekly 24 h filter samples collected in Duseo-myeon, Ulsan, South Korea, from June 2023 to November 2025. Seasonal concentrations and nitrate mass fractions were evaluated alongside particulate ammonium-to-sulfate ratios and a stoichiometric neutralization index. PM2.5 was highest in spring, whereas nitrate was highest in winter and lowest in summer, with seasonal means of 3.00 and 0.26 µg m−3, respectively. Spring and winter both frequently exhibited elevated ammonium-to-sulfate ratios, but nitrate exceeded sulfate by mass much more often in winter. The ratio intervals summarize observed composition and do not define thermodynamic thresholds or precursor-sensitivity regimes. The study provides a multi-year composition baseline for this rural site and shows that the seasonal PM2.5 maximum and the strongest nitrate contribution did not coincide. Weekly sampling, incomplete gas-particle time matching, and the absence of concurrent thermodynamic constraints limit attribution of the observed nitrate abundance to specific sources or formation processes.

Atmosphere doi: 10.3390/atmos17111041

Authors: Yujie Liu Zhe Wang Yun Wang Yihang Zou Qiaoyan Wu

Tropical cyclone (TC) size determines the spatial extent of hazardous winds, heavy rainfall, and storm surge, yet the environmental processes governing TC size remain poorly understood. In this study, we examine the environmental controls on TC initial size and subsequent size evolution during intensification using global TC observations from 2001 to 2023. Large TCs at formation tend to occur in environments characterized by stronger vertical wind shear (VWS) and higher mid-level relative humidity, whereas sea surface temperature (SST) and relative SST are less important. The influence of environmental factors on size growth shows a clear latitudinal dependence. VWS is more strongly linked to TC size growth at lower latitudes, while SST and relative SST become increasingly important at higher latitudes. Environmental relative humidity consistently favors TC size expansion across both latitudes. Generalized additive models and random forest analyses reveal that initial size and intensity together explain more than half of the variance in TC size, highlighting a strong structural memory during intensification. After accounting for these storm characteristics, relative SST emerges as the most important environmental factor on TC size expansion. Positive relative SST, combined with abundant environmental moisture, might enhance outer core deep convection and promote radial wind field expansion, providing new insight into the environmental regulation of TC structural evolution.

Atmosphere doi: 10.3390/atmos17111040

Authors: Hongchao Liu Huina Lin Junjie Ma Ren Li Jie Yang Meng Chang

The Mongolian Plateau is a critical region for land–atmosphere interactions in East Asia. Sensible heat flux (H), a key component of surface energy balance, significantly influences the Asian summer monsoon and atmospheric circulation. Using ERA5-Land reanalysis data (1980–2025), this study applies Empirical Orthogonal Function (EOF) and the Sequential Mann–Kendall test to diagnose the spatiotemporal patterns of H over the Mongolian Plateau and its dominant driving factors. Results indicate that H over the Mongolian Plateau exhibits a distinct spatial pattern, with higher in the west and south and lower in the east and north. With 105° E as the dividing line, the mean flux is 31.5 W m−2 in the western region and 25.0 W m−2 in the eastern region; with 46° N as the boundary, the corresponding values are 32.2 W m−2 in the south and 23.1 W m−2 in the north (reanalysis-relative, not observation-validated absolute values). EOF analysis identifies that the first mode captures the east–west gradient, while the second and third modes reflect opposing interannual thermal variations in zonal and meridional directions, respectively. Together, these three modes account for more than half of the total variance in surface thermal anomalies. Seasonally, H peaks in summer (46.8 W m−2), followed by spring (42.8 W m−2) and autumn (17.4 W m−2), and reaches its minimum in winter (3.9 W m−2). Over the 1980–2025 period, the annual mean H presents a statistically significant increasing trend of 0.068 W m−2 yr−1. Attribution analysis further reveals that the surface–air temperature difference dominates the spatiotemporal variations, whereas wind speed plays a comparatively minor role. These findings provide a scientific foundation for understanding ecological and environmental evolution on the Mongolian Plateau in the context of climate change.

Atmosphere doi: 10.3390/atmos17111039

Authors: Abidemi Kuye Prashant Kumar

Ultrafine particles (UFPs; <100 nm) can penetrate deep into the respiratory tract, yet real-world estimates of inhaled and deposited UFP doses in homes with wood-burning appliances remain scarce. In this small, real-world exploratory study, particle number concentrations were measured at 1 min resolution for 11–12 winter days in each of four UK homes heated by an Ecodesign stove, a multifuel Ecodesign stove (two homes), or an open fireplace. The measurements were combined with age-, sex-, and activity-specific ventilation rates and the International Commission on Radiological Protection (ICRP) deposition model to estimate body-mass-normalised daily inhaled doses and absolute respiratory deposition dose (RDD) rates for modelled adult, young adult, and child scenarios. The mean daily inhaled doses ranged from 0.45 × 109 to 16.0 × 109 particles kg−1 day−1 during sitting and from 2.37 × 109 to 73.2 × 109 particles kg−1 day−1 during moderate activity, which increased the dose 4.5–5.3-fold for each home and scenario. Child scenarios had the highest body-mass-normalised inhaled doses (2.7–3.8 times those of adults), whereas adult males had the highest absolute RDD (mean 4.69 × 108 particles min−1). The mean household RDD ranged from 1.50 × 108 to 5.99 × 108 particles min−1. These results demonstrate substantial variability in the modelled UFP dose across household, demographic, and activity scenarios; however, because the four homes differed simultaneously in appliance type, fuel use, room characteristics, ventilation, and occupant behaviour, between-home differences were interpreted as exploratory associations rather than independent stove- or fuel-specific effects.

Atmosphere doi: 10.3390/atmos17111038

Authors: Jiayu Li Guohao Ruan Zening Cheng Yanji Yao Yunfeng Pan

Tire additives are released during tire wear and can enter roadside air with fresh or resuspended particles and, for some compounds, through volatilization. Once airborne, they may partition between gas and particle phases, undergo atmospheric transformation, and reach terrestrial surfaces through dry and wet deposition. However, how much atmospheric deposition contributes to the transfer of tire additives from roadside air to soils and plants remains unclear. This review traces major tire additives and their transformation products from their occurrence, phase partitioning, and transformation in the atmosphere, through deposition, to their subsequent fate at the soil–plant interface. Available evidence shows that these processes are strongly compound-specific. Some additives remain mainly particle-associated, whereas others have substantial gas-phase fractions, and environmental aging can further change both parent compounds and transformation products before deposition. Field studies confirm the deposition of tire and road wear particles and selected tire additives, while controlled soil and plant experiments show retention, transformation, release, and plant uptake after environmental entry. However, the contribution of atmospheric deposition to additive burdens in field soils and crops is still difficult to separate from runoff, irrigation, road splash, and other rubber-related sources. Acute toxicity of N-(1,3-dimethylbutyl)-N′-phenyl-p-phenylenediamine quinone (6PPD-Q) is well established in highly sensitive aquatic species, whereas risks associated with atmospheric deposition in terrestrial systems and humans remain poorly characterized. Overall, atmospheric deposition represents a possible pathway by which tire additives are transferred from roadside air to terrestrial environments. Quantifying its contribution will require phase-resolved measurements, robust source attribution, and coordinated monitoring across air, deposition, soil, and plants.

Atmosphere doi: 10.3390/atmos17111037

Authors: Hye-Rim Jeon Sang Myeong Oh Subin Kim Joo Wan Cha Jeong Hoon Cho Yoonjae Kim Hee Choon Lee

To downscale artificial intelligence (AI) medium-range forecasts over the Korean Peninsula, we developed a 1.5 km CorrDiff system trained with ERA5 inputs and UM-LDAPS analyses as targets. For 1.5 m temperature and 10 m wind, the regression stage reduced root-mean-square error (RMSE) and root-mean-square vector error (RMSVE) relative to ERA5 and increased spatial correlation. CorrDiff had smaller absolute biases and greater spatial variability than Regression but higher RMSE and RMSVE and lower spatial correlation. The most consistent improvement relative to Regression occurred for high-temperature events at or above the target 99th percentile, with higher probability of detection (POD) and critical success index (CSI) and lower conditional temperature error. Without additional training or correction, AI Forecasting System (AIFS) forecasts were downscaled through +288 h. In a July 2026 heatwave case at 47 stations, CorrDiff reduced temperature errors and alleviated the underprediction of high-temperature occurrence relative to Regression. CorrDiff had higher POD and CSI than the AIFS for temperatures ≥33 °C but higher RMSE and lower spatial correlation. Generative downscaling should therefore be evaluated for spatial variability, spatial agreement, and event-specific performance beyond deterministic errors; producing high-resolution data does not necessarily imply improved forecast performance against observations.

Atmosphere doi: 10.3390/atmos17101036

Authors: Kang Gong Yusheng Liu Jie Zhang

Coarse-particle deposition over steep terrain is shaped by the mismatch between rapid flow deflection and slower particle motion. We apply an existing Eulerian–Lagrangian numerical model to simulate the transport of 100 µm spherical alumina over a flat floor and a 0.20 m high triangular ridge at an incoming wind speed of 10 m s−1. Wind-tunnel measurements are used to evaluate the simulated flow, particle motion, and surface deposition. For centerline deposition flux, root-mean-square errors normalized by the corresponding mean measured flux are 8.6% over the flat floor and 9.8% over the ridge. Transverse profiles show larger normalized errors of 20.1–40.8%. Over the flat floor, centerline deposition peaks about 3 m downstream of release. With the ridge present, floor deposition is enhanced on the approach and reduced at the sampled lee-side stations, followed by partial downstream recovery. Separate aperture measurements show greater collection at the sampled windward than leeward locations. The approach-region maximum overlaps the flat-floor landing zone; the measured approach-to-lee contrast provides clearer evidence of terrain-associated redistribution. Flow deflection, particle inertia, upstream removal, and wake exchange provide a physical interpretation of this sequence. The model captures the main streamwise redistribution under the tested conditions, although substantial discrepancies remain in some transverse deposition profiles.

Atmosphere doi: 10.3390/atmos17101035

Authors: Afnan Abdullah Alturki Islam M. Hamdi Wojciech Rykała Dominika Dąbrowska Youssef M. Youssef Monika J. Fabiańska

Fine particulate matter (PM2.5) over the Arabian Peninsula is poorly monitored at the surface. This study emulates the relative change in summer (June–August) PM2.5 between 2020 and 2025 in the NASA GEOS-CF composition system with interpretable ensemble machine learning, using thirteen predictors from Sentinel-5P TROPOMI, ERA5-Land, CHIRPS, and MODIS Aqua on the native 0.25° grid (3933 cells), and tests how far the resulting skill transfers in space. Under 10 × 10 spatial blocking, tree-based ensembles reached cross-validated R2 = 0.82–0.84, against 0.55 for linear regression. An identically tuned Random Forest using only latitude and longitude performed better (R2 = 0.91) and remained better under larger blocks, exclusion buffers, and four of five held-out sectors; combined with the coordinates, the predictors added only modest skill. The predictor model lost all skill once training cells within 300 km were excluded, failed on an unseen eastern sector containing the single contiguous cluster of changes above +40% (r = 0.11), and did not transfer to other summers, and interpolation of the reference filled simulated gaps more accurately than every emulator. The leading predictor, the total-column O3 change, is a latitudinal gradient (r = 0.94). Split-conformal intervals under a separated spatial design covered 82–98% of test cells at the 90% nominal level but 60% in the eastern sector. The spatially blocked skill of such emulators is therefore interpolative and should be reported together with transfer diagnostics.

Atmosphere doi: 10.3390/atmos17101034

Authors: Tomeu Rigo

The research consists of the analysis of the detection of overshooting tops (OT) over a period of 10 years using water vapour imagery from a satellite. OT is usually observed in thunderstorms producing severe hail. The results were compared with other remote sensing data (weather radar and lightning) to validate the algorithm and to characterise those severe thunderstorms correctly identified. First, it was observed how the algorithm detects false OT associated with non-convective structures, mainly during the non-warm season. Because the differences between the satellite parameters in well- and badly-corrected overshooting tops are small, with an average minimum brightness temperature difference of less than 2 °C, it is difficult to distinguish correct cases from incorrect ones using satellite imagery alone. For this reason, it is highly recommended to combine it with other data types, as in the present analysis. In warm-season cases, the results show that OT signatures were well identified in most cases, but there are cases with elevated false alarm ratios or low probabilities of detection. Finally, the characterisation of severe thunderstorms combining multiple systems led to the quantification of different parameters from different points of view, showing that TOP45 and electrification provide better results. Events with high electrification occur mainly in the warm season, while those without flashes were detected during the cold season. These results are relevant to improving the forecast of severe events in Catalonia.

Atmosphere doi: 10.3390/atmos17101033

Authors: Ermelinda Bloise Marimuthu Rajendran EzhilKumar Singaram Karthikeyan Adelaide Dinoi Daniele Contini Angelo Riccio Elena Chianese

This study examines PM10 and water-soluble carbon and nitrogen at three Chennai street environments during campaigns ST2–ST4 and at nominal heights of 3, 16 and 37 m. The measurements comprise 76 PM10 observations, 75 determinations each of water-soluble organic carbon (WSOC) and water-soluble total nitrogen (WSTN), and 74 records of water-soluble inorganic nitrogen (WSIN). Mean PM10 was 65.32 µg m−3, WSOC was 2.53 µg C m−3, WSTN was 1.28 µg N m−3, and WSIN was 1.26 µg N m−3. WSOC decreased from 4.01 µg C m−3 in ST3 to 0.62 in ST4, while WSIN increased from 1.12 to 1.86 µg N m−3. The site contrast was pronounced for PM10, with the highest concentrations at Velachery, and for WSIN, which was highest on average at Arumbakkam. Campaign contrasts for WSOC and WSIN remained significant when site-days were treated as the sampling units, whereas vertical differences were not significant. Positive water-soluble nitrogen residuals were quantified as WSON+=max(0,WSTN−WSIN), which is distinguished from the signed nitrogen balance. Campaign meteorology differed in humidity, wind speed and boundary-layer height, consistent with contrasting dilution and aerosol-processing conditions. The combined observations reveal pronounced campaign and urban-site variability in soluble PM10 composition.

Atmosphere doi: 10.3390/atmos17101032

Authors: Ni Liu Xiaoyu Ju Guangzhou Chen Hao Wang Yajun Lu

Mesoscale convective systems (MCSs) are organized convective precipitation regions that produce severe weather hazards, including heavy rainfall, flooding, and strong winds. Objective identification and tracking remain essential for MCS statistical analysis and operational forecasting. Conventional threshold-based methods using radar reflectivity or satellite brightness temperature are limited by terrain blockage, detection range constraints, and poor performance under nighttime or low-cloud conditions. This study developed a deep learning framework for MCS identification and tracking using hourly multi-source merged precipitation products from 18 summer seasons (2008–2025). A semi-supervised labeling strategy combining threshold-based pre-annotation and manual quality control was used to construct the dataset. Four architectures (DNN, U-Net, Attention U-Net, and Swin-Unet) were evaluated for pixel-level semantic segmentation, coupled with a conventional area overlap tracker and a proposed hybrid-criterion tracking method. Swin-Unet achieved the highest identification accuracy, with an Intersection over Union (IoU) of 0.90, and outperformed the other models in critical success index (CSI) and probability of detection (POD). The hybrid-criterion tracker, integrating area overlap, centroid distance, and deep feature similarity, attained a tracking accuracy of 0.94, with a 7.2% improvement over the area overlap method and a 34% lower break rate. A representative Meiyu rainfall event confirmed that the proposed approach captures the complete MCS life cycle. These results demonstrate that merged precipitation products offer a viable data source for operational MCS monitoring, particularly in radar-sparse regions.

Atmosphere doi: 10.3390/atmos17101030

Authors: Alexander Kirsanov Mikhail Varentsov Alexander Baklanov

The interaction between the urban heat island (UHI) and urban air pollution (UAP) is widely studied in temperate climates, but it remains underexplored in high latitudes. Using the COSMO-ART atmospheric model with the TERRA_URB urban parameterization, we analyzed feedback during winter and summer pollution episodes in Saint Petersburg, the world’s northernmost megacity. We show that UHI-induced enhancement of turbulent mixing and boundary layer height significantly reduces near-surface primary pollutants (NO, CO, PM10): by 15–54% on average in the winter case and by 5–33% in the summer case, with peak PM10 reduction exceeding 44%. In contrast, ozone concentrations increase. In turn, the UAP effect on UHI, i.e., urban aerosol radiative feedback, is weak (hundredths of °C on average, peak magnitudes < 0.3 °C), and it exhibits a seasonal asymmetry. In the summer case, aerosols attenuate incoming shortwave radiation, producing weak cooling, while in the winter case, high snow albedo reverses the sign, causing a weak warming. Overall, this aerosol feedback is substantially weaker than urban canopy forcing, amounting to less than a fifth of the average UHI intensity. Our findings underscore the importance of considering urban canopy effects and their influence on the boundary layer in air pollution modeling in cold-climate cities.

Atmosphere doi: 10.3390/atmos17101031

Authors: Abdullah Kaviani Rad Mohammad Javad Nematollahi Sasan Faridi

Meteorological observations are fundamental to environmental modeling, yet data gaps frequently compromise their utility. This study benchmarked five machine-learning algorithms (Random Forest, Extra Trees, Gradient Boosting, Histogram-based Gradient Boosting, and Ridge regression) for hourly temporal imputation of five key meteorological variables—temperature, humidity, dew point, atmospheric pressure, and wind speed—across 17 airport-based synoptic stations in Iran. A comprehensive feature-engineering approach generated 43 predictors capturing calendar, cyclical, continuous composite, binary regime, and cross-variable patterns. Performance was assessed using R2, RMSE, and MAE with time-series cross-validation, bootstrap resampling, split and Mondrian conformal prediction, three robustness stress tests, and comparison against persistence and climatological baselines. Results revealed a striking performance dichotomy. Thermodynamically coupled variables achieved exceptional accuracy, with mean cross-validated R2 values of 0.994 ± 0.003 (temperature), 0.997 ± 0.001 (humidity), and 0.988 ± 0.009 (dew point), with correspondingly low RMSEs (0.74 ± 0.22 °C, 1.10 ± 0.21%, and 0.60 ± 0.19 °C). Feature-ablation confirmed that this skill was driven primarily by inter-variable coupling, with the four cross-variable predictors accounting for 96.3–100% of importance; removing them caused the dew point R2 to collapse to 0.375. In contrast, dynamically controlled variables exhibited substantially lower skill: atmospheric pressure achieved a mean R2 of 0.44 ± 0.25 (RMSE 3.87 ± 0.86 hPa), while wind speed performed poorest (mean R2 0.13 ± 0.22, RMSE 2.20 ± 0.50 m/s), with negative R2 at several stations and no significant improvement over persistence. Their diffuse feature importance profiles and wider conformal prediction intervals indicated that temporal and cross-variable features alone cannot capture advective and synoptic-scale influences. Pressure proved less reconstructable than the thermodynamically coupled variables, reflecting its dependence on large-scale atmospheric dynamics rather than on local site conditions. This study demonstrates that ensemble tree-based models excel at reconstructing variables with strong inter-variable coupling but require spatial and physics-informed integration for dynamic parameters in regions with complex synoptic-scale dynamics.

Atmosphere doi: 10.3390/atmos17101029

Authors: José Francisco León-Cruz Alejandro Jaramillo-Moreno Jorge García-Franco

Thunderstorms in Mexico have recently attracted increased study. However, researchers have not systematically evaluated the relationship between thunderstorm activity and the El Niño–Southern Oscillation (ENSO) at the national scale. This study examines the association between ENSO and thunderstorm-day frequency across Mexico from 1991 to 2020, utilizing data from the Global Surface Summary of the Day and the Oceanic Niño Index (ONI). Standardized monthly thunderstorm-day anomalies were correlated with ONI at lags of 0 to 3 months and across four climatological seasons. Thunderstorm activity increases beginning in May, peaks during boreal summer, and declines in autumn, with considerable spatial variation among stations. At the national scale, thunderstorm-day anomalies show a weak but statistically significant positive correlation with ONI (r = 0.24, p < 0.001), i.e., higher ONI values were generally associated with positive standardized anomalies in thunderstorm-day frequency. Significant station-level relationships are limited and predominantly positive, with 18 of 48 stations exhibiting significant correlations, 17 of which are positive. No systematic strengthening of the relationship is observed when ONI leads thunderstorm activity by 1 to 3 months. Seasonal analyses suggest pronounced geographic variability and sign changes in correlations. Overall, ENSO imparts a detectable but limited statistical signal in Mexican thunderstorm variability, with the magnitude and direction of this relationship strongly dependent on location and season.