Asia-Pacific Journal of Atmospheric Sciences最新文献

The accuracy of PM_2.5 forecasts in Seoul from 2019 to 2023 was assessed using multiple methods. Daily short-term PM_2.5 forecasts, which were provided as four categories (good, moderate, bad, and very bad), were directly compared with the corresponding PM_2.5 observation data. Although the probability of detection for days with high PM_2.5 concentrations increased, a simultaneous rise in the false alarm rate resulted in no improvement in the total accuracy and F1-score. To analyze these trends in more detail, the forecast accuracy was further examined based on the PM_2.5 categories. The results showed an annual improvement of 3.65% in the accuracy for the bad category. An analysis based on the announcement time also indicated an increase of over 20% in the accuracy for the bad category for next-day and day-after forecasts. The confusion matrices of forecasted and observed PM_2.5 categories confirmed this improvement, which was primarily due to a reduction in the number of instances where the moderate category was forecasted as bad. However, the accuracy for the good category showed no significant change and that for the moderate category even declined. These findings highlight the importance of category-specific evaluation in air quality forecasting and improving the forecast accuracy, particularly for the good and moderate categories. The reliability of forecasts and their policy relevance may be improved by utilizing these insights and addressing temporal and spatial limitations.

This study investigates the influence of the Quasi-BiWeekly Oscillation (QBWO) on the onset, monsoon breaks, and extreme rainfall events in the Philippines during the southwest monsoon season from 1980 to 2019. Findings show that the occurrence of extreme rainfall events over the Philippines follows the northward to northwestward movement of QBWO-related convective anomalies. These events initially appear in Mindanao and Visayas (Phase 5), then shift toward Luzon (Phases 7–8), and may persist along Luzon’s western coast up to Phase 2. The contribution of tropical cyclones (TCs), low-pressure systems (LPSs), and other disturbances to extreme rainfall was also assessed. Although their percentage contribution to the total extreme rainfall varies substantially across stations, their impacts are more apparent in Phase 6, coinciding with northward/northwestward propagation of the convective anomalies of the QBWO towards the Philippines. Monsoon breaks, which is defined as the period when the average rainfall across the western coastal stations in Luzon Island decreases below 5 mm day^-1 for at least three consecutive days, also show distinct QBWO-phase preferences such that short to medium breaks are more frequent in Phases 3–5, coinciding with the movement of the anticyclonic anomaly of the QBWO from the South China Sea toward East Asia in these phases. In contrast, longer breaks occur more frequently in Phases 4–6 and may persist until Phase 7, when the northeasterlies at the northern flank of the cyclonic circulation anomaly of the QBWO weaken the southwest monsoon. The southwest monsoon onset in the Philippines typically begins during Phase 1 and Phases 6–8. Notably, about 82% of monsoon onset events occurred during these phases when the QBWO amplitude exceeded 0.5, suggesting that a real-time QBWO index could aid in monitoring the monsoon onset.

This study conducts an observation-constrained intercomparison of 23 CMIP6 global climate models (GCMs) for precipitation over Japan. We quantify relative agreement between models and a regional reference analysis from the JMA mesoscale system (GPV-MSM) for 2015–2023. Eight precipitation indices (mean, 90th percentile, max5d, CWD, etc.) and four statistical metrics (correlation, Pbias, NRMSE, Taylor skill score) are synthesized via entropy-weighted TOPSIS to obtain a proximity-to-reference coefficient and a similarity-based ordering. To limit interpolation artifacts, all fields are remapped to a 1.0° × 1.0° common grid. Because the CMIP6 Historical experiment ends in 2014, ScenarioMIP SSP585 simulations are used for 2015–2023. Axis-specific results show a stable latitudinal top-tier cohort comprising ACCESS-CM2, KACE-1-0-G, and UKESM1-0-LL, each with Top-3 inclusion probabilities ≥ 89%. In the longitudinal direction, a distinct three-model cohort—KACE-1-0-G, CNRM-ESM2-1, and CESM2-WACCM—emerges with Top-3 probabilities near 60%. Bootstrap resampling　indicates consistent rank structures (median Kendall’s Tau = 0.87 for latitude and 0.72 for longitude), with wider intervals in the longitudinal rankings indicating larger sampling variability. Across indices, the intercomparison indicates systematic model-reference differences: many models show lower amplitudes in extreme-intensity indices (90P, max5d) together with positive differences in wet-spell persistence (CWD, CWD10). The synthesis therefore identifies cohorts with comparatively higher similarity across multiple indicators (e.g., ACCESS-CM2, KACE-1-0-G, UKESM1-0-LL) and others with lower similarity (e.g., INM-CM4-8, IPSL-CM6A-LR). All findings are interpreted as relative agreement with an observation-constrained regional analysis over 2015–2023 and are intended to support bias-aware, purpose-specific use of CMIP6 precipitation in regional applications.

This study analyzed the characteristics and mechanisms of low-level wind shear at Incheon International Airport (IIA), which is located in a coastal region with gentle terrain, using 1-minute wind data from the Aerodrome Meteorological Observation System (AMOS) and hourly ERA5 reanalysis data. Wind measurements from 8 AMOS stations positioned along runways revealed a dominant channeling effect along the runway, with prevailing northwesterly and southeasterly winds, along with small-scale horizontal variability. The analysis of joint frequency distributions between near-surface and upper-level winds indicated that the northwesterly winds resulted from downward momentum transport, while the southeasterly winds occurred due to baroclinicity and pressure-driven channeling under stable stratification. These southeasterly winds were enhanced by the land breezes. Thermally-driven circulations, including land-sea breezes and slope winds, further contributed to the horizontal heterogeneity in wind patterns. On strong-synoptic forcing days, wind heterogeneity increased with wind speed, whereas on weak-synoptic forcing days, it peaked in the early afternoon due to local sea breeze development. Vertical wind shear events exceeding 7.7 m s^-1 (15 knots) between the surface and 975 hPa were more frequent at night, accounting for 3.0% (headwind) and 4.2% (crosswind) of study period. Strong shear events were often accompanied with strong synoptic winds, but some events were associated with moderate synoptic southeasterly and southwesterly winds due to interactions with land-breeze forcing. The results emphasize the importance of the interaction between synoptic forcing and land breeze forcing in low-level wind shear in coastal regions and provide insights for enhancing flight safety and small air mobility operations.

Non-landfalling tropical cyclones (TCs) traversing north and northeast of the Philippines during the southwest monsoon season (locally, Habagat) often enhance the prevailing moisture-laden southwest monsoon flow, inducing heavy rainfall and flooding events, particularly along the western coast of the country. One such event occurred last July 23–24, 2024 (Habagat 2024), when the Habagat was enhanced by TC GAEMI (locally, Super Typhoon CARINA) and some parts of the country received a month’s worth of rainfall. This study examines its unique features and compares it with recent Habagat events. Events with rainfall amounts above the 95th percentile in weather stations within Metro Manila in July and August for at least two days from 2012 to 2024 were selected for the analysis. Habagat 2024 produced exceptionally high rainfall, most of which occurred within 24 h. This record rainfall was accompanied by higher than normal magnitudes of the horizontal winds, vertically integrated moisture flux convergence (VIMFC), stream function, vertically integrated moisture flux (VIMF), and surface latent heat flux. The Moisture Conveyor Belt for the Habagat 2024 event, in particular, was noticeably more pronounced, which explains the higher rainfall rate relative to previous Habagat events. Furthermore, the VIMF anomalies were further separated into the basic flow (monsoon only) and the TC perturbation flow using spatial Fourier decomposition analysis to identify which contributes more to rainfall enhancement over Metro Manila. Habagat 2024 exhibited the strongest basic monsoon flow and TC perturbation flow among all events, with the latter contributing more to the rainfall enhancement.

This study examines the synoptic-scale environment for heavy rainfall over the Korean Peninsula (KP) during the Changma (late June through mid-July) and August using a 30-year dataset (1990–2019). The K-means clustering of geopotential height and equivalent potential temperature at 850 hPa has produced nine clusters of synoptic-scale environment, which are classified into 4 patterns. For the Changma, 1) the pattern 1 (C2, C4 and C6 clusters): a synoptic cyclone to the north, the western North Pacific Subtropical high (WNPSH) to the southeast of the KP, a significant trough to the southeast of the Tibetan plateau, and the strong southwesterly band from SW China to the KP, 2) the pattern 2 (C3 and C5) is similar to the pattern 1 except for the ridge over northeastern China, 3) the pattern 3 (C1 and C7) is similar to the pattern 2, except for the synoptic-scale trough over the South China Sea, and 4) the pattern 4 (C8 and C9) is not a typical Changma pattern. For August, 1) the pattern 1 (A3, A8 and A9) is similar to the Changma pattern 1, however, the Tibet trough and the southwesterly belt are relatively weak, 2) the pattern 2 (A7 and A2) shows heavy rainfall along the western edge of the WNPSH accompanied by relatively frequent tropical cyclones (TCs), 3) the pattern 3 (A5 and A6) shows heavy rainfall along the western edge of the WNPSH with a trough over the Yellow Sea, and 4) the pattern 4 (A4 and A1) shows heavy rainfall away from the WNPSH.

From 31 July to 2 August 2022, a successive three-day gust front (GF) event occurred over southeastern coastal China, as clearly observed in satellite and radar data. Such successive GF events were uncommon in this region of complex hilly terrain. This study employed multi-source observations and reanalysis data to conduct observational and mechanistic analyses of these events. The GFs originated primarily from multicellular storms, with surface stations recording characteristic abrupt temperature drops and relative humidity increases. Results showed that the GF event was closely linked to specific synoptic conditions during the northward movement of Typhoon “Trases”, which established an environment conducive to parent storm and GF development. Composite analysis of non-GF days preceding and following the event indicated that GF occurrence was inhibited regardless of whether the typhoon was positioned south or north of the region. Conversely, the typhoon’s remote enhancement of land–sea circulation played an essential role in GF formation by moistening the boundary layer—an effect notably absent on non-GF days. Objective classification of mid-to-low-level circulation patterns from 2018 to 2022 (July–August) using T-mode principal component analysis with oblique rotation and self-organizing map showed that GF days corresponded to a circulation pattern occurring on only 2.9% of the total days, confirming the rarity of such synoptic configurations.

Cold surges over South Korea are among the most extreme winter weather events, typically triggered by wave-train propagation or subarctic blocking in the upper troposphere. A widely used classification framework categorizes cold surges based on the upper-tropospheric circulation pattern on the day of onset. To overcome the limitation of this framework, cold surge indices are newly developed by considering the spatiotemporal evolution of circulation features characteristic of each cold surge type. During the pre-onset period (Day − 3 to 0), the daily geopotential height anomalies at 300 hPa are projected onto wave-train and blocking features in the dynamically defined domains, and new wave-train index (WI_n) and blocking index (BI_n) are developed based on this projection. Compared to the original scheme, the newly developed indices more accurately represent the dynamical evolution of upper-tropospheric circulation patterns, significantly reducing the number of Mixed-type cold surges (MT) from 115 to 77 and Unclassified cold surges from 64 to 18. Furthermore, thermodynamic analysis reveals that blocking-type cold surges (BT) exhibit longer duration, colder temperature anomalies, and greater intensity than wave-train-type cold surges (WT), consistent with their quasi-stationary blocking structure. By incorporating the evolution of circulation features before onset, the new indices offer clearer dynamical distinctions between cold surge types and provide a robust, physically interpretable foundation for their improved diagnosis and prediction.

The research reveals the origins and persistence of the extremely heavy rainstorm in North China in July 2023 (known as the “23.7” rainstorm). Unlike previous research, it applies scale separation and an advanced energy diagnostic equation to analyze energy transformations among multi-scale weather systems, highlighting their collective impact. Utilizing hourly rainfall data from over 70,000 national stations, along with daily global Final Analysis Data (FNL 1 º × 1 º) from NCEP/NCAR, the study employs Barnes bandpass filtering for atmospheric field separation and an energy equation to diagnose energy conversions. It finds that the “23.7” rainstorm is a complex, multi-scale phenomenon, with the synoptic to meso-α scales scales playing a pivotal role. The storm’s sustenance mechanism resembles the conditional instability of second kind (named R-CISK), focusing on upper tropospheric westerly jet divergence that triggers mid-tropospheric ascent, diverging from CISK’s emphasis on boundary layer friction. This initiates cumulus convection, followed by lower tropospheric jet convergence and typhoon-induced low-pressure trough interaction, uplifting lower layer air and coupling with mid-to-upper level convection. The energy diagnosis shows that large scale and synoptic scale systems are the primary energy sources. The kinetic energy interplay and conversion of buoyancy work and baroclinic energy are the main drivers of energy transformation, essential for storm development. These insights and methods enhance traditional rainstorm models and diagnostic techniques, offering valuable references for future regional heavy rainstorm forecasting and analysis.

The Korean Integrated Model (KIM) is a global numerical weather prediction model that uses the spectral element method on a (quasi-) uniform-resolution cubed-sphere grid. However, to investigate high-resolution prediction performance in the domain of interest, global high-resolution experiments must be considered. These experiments incur a significant computational burden owing to the increased number of grid points and the need for temporal integration at finer time steps. To address this issue, we implemented a stretched variable-resolution grid in the KIM, based on the Schmidt transformation. This approach allows for high-resolution effects in the region of interest while using a coarser grid elsewhere. With the stretched global grid, grid sizes are seamlessly scaled between high and low resolutions.

In this study, we evaluated the forecasting accuracy and computational efficiency of the variable-resolution KIM and compared these metrics with those of the corresponding uniform-resolution KIM. In the higher-resolution target region, the forecast skill of the variable-resolution grid was similar to that of the 8-km high-resolution uniform grid and even outperformed the 25-km reference-resolution uniform grid in the KIM. Furthermore, the variable-resolution configuration offers a significant improvement in computational efficiency, reducing the total 3-day forecast run time by approximately 88% compared with that of the high-resolution configuration with a uniform grid.