Long-term variations in precipitation during the major rainy periods in Japan—the Baiu (June–July) and Akisame (September–October) seasons—are investigated using precipitation records from 44 weather stations in western to eastern Japan over the past 120 years (1901-2020). The total amount of Baiu precipitation has increased over the 1901-2020 period, mainly during the mid–late stages of the season (late June–July) over regions on the Sea of Japan side of the country. In contrast, the precipitation amount during the Akisame season has decreased, mainly during the mid-stage (late September–early October) over all regions. The frequency and intensity of heavy precipitation have generally increased in both seasons, but the trends are much stronger for the Baiu season compared to those for the Akisame season. A prominent positive trend, 23.5 % per 100 years (18.1 % per °C), which is much higher than the Clausius–Clapeyron rate (approximately 7 % per °C), is observed for the Sea of Japan side of western Japan for the seasonal maximum 1-day precipitation total during the Baiu season. It may be noteworthy that the observed long-term trends differ greatly between the Baiu and Akisame seasons even though the statistical significances of the trends are not so high, because similar differences between the two rainy seasons are found in the results of global warming simulations.

 The precipitation system and environment causing the heavy rainfall event in July 2020 over Kyushu Island, Japan are analyzed focusing on a hierarchical structure. The moisture budget analysis over Kyushu reveals the contribution of the free-tropospheric moisture flux convergence moistening the atmosphere before the rainfall event. Further analyses by dividing the flux convergence into moisture-advection and wind-convergence terms indicate that the moisture-advection controlled the moistening. Contributions of both the boundary-layer and free-tropospheric wind-convergence terms increase after the moistening. Wide areas with weak precipitation characterize the moistening phase, whereas concentrated intense precipitation areas develop after the moistening. A synoptic scale upper-tropospheric trough transports free-tropospheric moisture from the South China Sea to Kyushu via southern China. The free-tropospheric moisture converges in a sub-synoptic scale cloud system in front of the trough, providing a moist environment favorable for the precipitation systems bringing a large amount of precipitation. A mesoscale depression below the trough developed with active convection over central China enhances the free-tropospheric moisture transport. Cyclonic circulations associated with the mesoscale depression and the sub-synoptic scale cloud system enhance the baroclinicity around Kyushu. An active convective area develops to a mesoscale convective system covering Kyushu under such conditions. A line-shaped convective area is generated along the southern edge of the convective system, causing the heavy rainfall event. Two intense precipitation areas are embedded in the convective area along the inflow direction. At the same time, weak precipitation areas spread downstream of the intense precipitation areas. The vertical cross-sections of the intense precipitation areas show structures consistent with the organized precipitation systems with deep inflow layers and the moist absolutely unstable layers. These results suggest that the organized precipitation system develops under the moist environment prepared by the free-tropospheric moisture flux convergence associated with the hierarchical structure.

 This study focused on the total precipitable water (TPW) products of the Advanced Microwave Scanning Radiometer 2 (AMSR2) onboard the Global Change Observation Mission—Water (GCOM-W). The GCOM-W satellite has been flying in the Afternoon Constellation (A-train) orbit to synergize with other A-train satellites, such as Aqua. In this study, we compared two datasets of AMSR2 TPW from July 2012 to December 2020, independently produced by the Japan Aerospace Exploration Agency (JAXA) and Remote Sensing Systems (RSS). There were no significant differences in TPW anomaly trends between them. However, significant differences in the absolute values of TPW were found in the northwest Pacific and northwest Atlantic Oceans during the boreal summer season. We investigated the meteorological conditions that caused these differences using reanalysis, in-situ observation data, and visible and infrared data from the MODerate resolution Imaging Spectroradiometer (MODIS) on the Aqua. The results showed that the lower atmosphere had an inversion layer with relative humidity close to 100 %, and very low altitude clouds (i.e., fog) were often distributed in the areas where the TPW differences between JAXA and RSS are large. The temperature profiles represented in the JAXA and RSS algorithms were approximated by a simple model. The influence of the inversion layer and fog on the JAXA and RSS TPW algorithms was also investigated using a radiative transfer model. Sensitivity experiments suggested that the inversion layer was associated with the underestimated TPW for the JAXA algorithm, while it was associated with the overestimated TPW for the RSS algorithm.

 We performed kinematic precise point positioning (PPP) to determine the optimum analysis settings for precipitable water vapor (PWV) retrieval at sea using a ship-based Global Navigation Satellite System (GNSS). Three analysis parameters were varied: the standard deviation of random walk process noise (RWPN) of Zenith Total Delay (ZTD) time variation, the analysis time width, and the time interval of update of the Kalman filter state vector. A comparison with the Meso-scale Analysis (MA) of the Japan Meteorological Agency revealed that, depending on the update interval and the time width, a strengthened RWPN constraint suppresses the unnatural time variation of GNSS-derived PWV, reduces negative bias against MA, but decreases the regression coefficient.

 On the basis of the results of comparison of GNSS-derived PWV with MA, a setting combination of 3 × 10⁻⁵ m s^−1/2, 1.5 h, and 2 s for the RWPN, the time width, and the update interval, respectively, was selected to compare with other observations. Biases and root-mean-square differences between the ship-based GNSS-derived PWV and radiosonde observation, a nearby ground-fixed GNSS station, and a satellite-borne microwave radiometer were −0.48 and 1.75, 0.08-0.25 and 1.49-1.63, and 1.04-1.18 and 2.17-2.43 mm, respectively.

 The factors yielding the differences in the GNSS-derived PWV bias were discussed, especially the errors in the estimated GNSS antenna altitude. The error in the vertical coordinate in GNSS positioning was confirmed as negatively correlated with the error in the GNSS-derived PWV. We found that the kinematic PPP is more likely to overestimate the altitude with shorter update intervals and wider time widths. When the RWPN and the update interval were set to 3 × 10⁻⁵ m s^−1/2 and 2 s, respectively, the bias of the analyzed altitudes by the kinematic PPP changed from negative to positive at approximately 1 h width. The results suggest that precise GNSS positioning is necessary for accurate GNSS-derived PWV analysis.

 Based on the vertical atmospheric sounding system carried by the FY-3D meteorological satellite (FY-3D/VASS) and the new wind radar instrument carried by the FY-3E meteorological satellite (FY-3E/WindRAD), a study of the potential application of research on the changes of temperature, humidity, and ocean wind vector (OWV) during the onset of the South China Sea summer monsoon (SCSSM) was carried out. The applications of these satellite datasets in SCSSM monitoring was evaluated, and the SCSSM onset process in 2022 was analyzed. The results showed that the mean bias of the FY-3D/VASS temperature and specific humidity at 850 hPa, compared with that of the fifth-generation ECMWF reanalysis, were −0.6 K and −0.53 g kg⁻¹, respectively, and the pseudo-equivalent potential temperature (θ_se) was slightly lower, by 1-2 K; the distribution of θ_se was consistent with the seasonal advancement of the SCSSM. Compared with Metop-C/ASCAT, the mean bias of FY-3E/WindRAD zonal wind was positive and that of meridional wind was negative. The correlation coefficient, mean bias, mean absolute error, and root-mean-square error of the wind speed were 0.79, −0.45 m s⁻¹, 1.56 m s⁻¹, and 2.03 m s⁻¹, respectively. The distributions of OWV were consistent, and the region and intensity of strong wind speed were close to each other. The temperature, humidity, and wind reversal during the onset of the SCSSM in 2022 were well-monitored by the FY-3D/E-derived θ_se and OWV dual indices, which are consistent with the SCSSM onset date, the third pentad in May, issued officially by the National Climate Center, China Meteorological Administration. Before the SCSSM onset in 2022, the tropical storms' pumping effect in early May increased the westerly wind over the tropical ocean north of the equator. After the storm weakened, the southwesterly wind passed across the Indochina Peninsula and reached South China Sea, causing the SCSSM onset.

 In a global numerical weather prediction (NWP) modeling framework we study the implementation of Gaussian uncertainty of individual particles into the assimilation step of a localized adaptive particle filter (LAPF). We obtain a local representation of the prior distribution as a mixture of basis functions. In the assimilation step, the filter calculates the individual weight coefficients and new particle locations. It can be viewed as a combination of the LAPF and a localized version of a Gaussian mixture filter, i.e., a Localized Mixture Coefficients Particle Filter (LMCPF).

 Here, we investigate the feasibility of the LMCPF within a global operational framework and evaluate the relationship between prior and posterior distributions and observations. Our simulations are carried out in a standard pre-operational experimental set-up with the full global observing system, 52 km global resolution and 10⁶ model variables. Statistics of particle movement in the assimilation step are calculated. The mixture approach is able to deal with the discrepancy between prior distributions and observation location in a real-world framework and to pull the particles towards the observations in a much better way than the pure LAPF. This shows that using Gaussian uncertainty can be an important tool to improve the analysis and forecast quality in a particle filter framework.

 Mineral dust affects health, climate, and ecosystems in various ways. East Asia is one of the major sources of mineral dust in the world. This study examines the year-to-year variability of dust deposition over Japan in April from the perspective of large-scale atmospheric circulations using atmospheric and aerosol reanalysis datasets for the period from 2011 to 2017. The increased dust deposition in Japan is explained by the intensified dust transport from the Mongolian Plateau by the anomalous westerly winds associated with a deepened trough over the East Asian continent toward the northwest of the Japanese islands in the middle to lower troposphere. The enhanced dust emission over Gobi Desert and the intensified extratropical cyclone activity are consistent with the larger-than-normal amount of dust in East Asia. Comparing the dust depositions over western and northern Japan, it is suggested that the slightly different anomalous trough positions may determine whether or not a large amount of dust is carried. A further analysis using the long-term (1967-2022) observation data of dust in Japan supports the importance of the intensified trough over the East Asian continent. Dust flux decomposed into cyclonic and anticyclonic components showed that both vortices contribute to the eastward dust transport in East Asia. These results suggest that Japanese dust events and their variability are affected by the stationary circulation anomaly as well as the baroclinic instability waves including transient cyclones and anticyclones.

 In this study, we aimed to systematically and globally evaluate the monthly precipitation forecasts of JMA/MRI-CPS2, a dynamical seasonal climate forecast (Dyn-SCF) system operated by the Japan Meteorological Agency, by comparing its forecasts with those of a statistical SCF (St-SCF) system using climate indices. We developed a new global St-SCF system using 17 climate indices and compared the monthly precipitation of this system with those of JMA/MRI-CPS2. Consequently, the skill of JMA/MRI-CPS2 was determined to be globally higher than that of the St-SCF for zero-month lead forecasts. In contrast, for forecasts made with one month or longer lead times, the deterministic skill of JMA/MRI-CPS2 was comparable to that of the St-SCF and the probabilistic skill of JMA/MRI-CPS2 remained slightly higher. In addition to evaluating the skill of JMA/MRI-CPS2, we identified several regions and seasons, for which JMA/MRI-CPS2 showed a low forecast skill, compared with the St-SCF. This indicated that JMA/MRI-CPS2 cannot adequately reproduce certain dynamics. In conclusion, comparing Dyn-SCFs with St-SCFs can clarify the potential regions and seasons to improve the forecast skill of Dyn-SCFs.

 Typhoon Krosa (2019) formed in the eastern part of the Philippine Sea and ∼ 1400 km east of another typhoon Lekima on 6 August and made landfall in the western part of Japan's mainland on 15 August. The operational global model forecasts, which were initialized just after Krosa's formation, showed a very large uncertainty and totally failed to predict the actual track of Krosa. In this study, we investigated the causes of this large uncertainty through 101-member ensemble forecast experiments by using a 28-km mesh global nonhydrostatic model. The experiments initialized at 1200 UTC 6 August, showed a large uncertainty. An ensemble-based lagged correlation analysis indicated that the western North Pacific Subtropical High (WNPSH) retreated further east in the members with large track forecast errors than in the members with small errors. For the members with a large track forecast error for Krosa, Krosa and Lekima approached each other by 250 km and Krosa moved northward faster than the observation in 36 hours from the initialization time. For the members with a small track forecast error for Krosa, two typhoons approached each other by only 50 km, and the northward moving speed was comparable with that of the observation. The typhoon-center relative composite analysis exhibited that at the initialization time, the members with a large Krosa track forecast error had a larger horizontal size of Krosa and the difference in Krosa's size was kept during the forecast period. This difference in size led to a stronger interaction between the two typhoons and retreatment of the WNPSH, thus resulting in a fast northward moving speed for the members with a large Krosa track error.

 High-resolution atmosphere–ocean coupled models are the primary tool for sub-seasonal to seasonal-scale (S2S) prediction. Seasonal-scale sea surface temperature (SST) drift is, however, inevitable because of the imbalance between the model components, which may deteriorate the prediction skill. Here, we examine the performance of a simple flux adjustment method specifically designed to suppress seasonal-scale SST drift through case studies. The Nonhydrostatic Icosahedral Atmospheric Model (NICAM)–Center for Climate System Research Ocean Component Model (COCO) coupled weather/climate model, named as NICOCO, was employed for wintertime 40-day integrations with a horizontal resolution of 14 km for the atmosphere and 0.25° for the ocean components. The coupled model with no flux adjustment suffers SST drift of typically -1.5–2°C in 40 days over the tropical, subtropical, and Antarctic regions. It is found that simple flux adjustment sufficiently suppressed the SST drift. Nevertheless, the lead-lag correlation analysis suggests that air–sea interactions are likely to be appropriately represented under flux adjustment. Thus, high-resolution coupled models with flux adjustment can substantially improve S2S prediction.