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.
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.
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.
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.
Several studies have established a link between outdoor air quality and cardiovascular disease (CVD), with air pollution causing millions of premature deaths annually according to the WHO. However, a similar link has not been clearly established for indoor air quality (IAQ), despite the fact that people spend much of their time indoors. In this prospective observational study, we monitored PM2.5 and PM10 (particulate matter) levels for several weeks in the homes of 49 patients referred for coronary CT angiography (CCTA) at two outpatient cardiology clinics in Amsterdam and compared these IAQ measurements with patients’ CCTA results. The CCTA outcomes included CAD-RADS scoring, vessel involvement, and the presence of obstructive disease. Patients were recruited from two Amsterdam neighborhoods with contrasting socioeconomic profiles, Amsterdam Zuid (N = 24) and Amsterdam Nieuw-West (N = 25). We found no strong systematic difference in indoor air quality between the two recruitment sites, but observed substantial within-neighborhood variability, suggesting that household-level factors may play an important role in indoor air quality. No significant associations were found between residential indoor PM levels and CAD or obstructive CAD. We conclude that the small sample size (49 individuals) and short IAQ monitoring period (a few weeks) substantially limit our ability to assess the relationship between long-term exposure and coronary artery disease.
