Significant progress has been made in numerical forecasting of typhoon tracks and intensity, yet meeting the needs of operational forecasting remains challenging. Based on the high-resolution Weather Research and Forecasting (WRF) model, eight sets of combination experiments are conducted with different model initial conditions, initialization times, and microphysical parameterization schemes. The results indicate that the model initial condition exhibits the highest sensitivity in forecasting the track of typhoon Lekima (1909). Representative experiments are selected based on track errors, and sensitivity experiments are conducted to further explore the mechanisms responsible for significant differences in typhoon track forecasts arising from different model initial conditions. The results indicate that accurate forecasting of the strength and extent of the western Pacific subtropical high (WPSH) plays a crucial role in determining the quality of typhoon track predictions. An overestimation of the WPSH in the initial field leads to a subsequent overestimation of WPSH, which blocks the typhoon that moves westward and northward along the WPSH and its moving speed is slowed. This results in significant errors in the typhoon track forecasting. Additionally, forecast errors of the typhoon inner core structure at different stages are related to the predicted typhoon track. Vertical wind shear in the large-scale circulation field may be a significant factor contributing to this error. Moreover, an overestimation of the WPSH extent and intensity leads to a deeper asymmetric distribution of horizontal wind speed near the typhoon, which contributes to the occurrence and development of deep convection. This is an important reason for the slower moving speed of the typhoon.

In 2023, the central part of the Hexi Corridor experienced an extreme high temperature and drought that would occur once every 60 years, which adversely affected local agricultural production and ecological environment, causing serious economic losses. This study uses various observational datasets to discuss the characteristics and effects of the high-temperature and drought in the central Hexi Corridor in 2023 from perspectives of meteorology, hydrology, ecology and agriculture. We further examine the responses of summer precipitation in the central Hexi Corridor to anomalous atmospheric circulation and sea surface temperature (SST). The datasets used in this study include monthly precipitation and temperature data during 1951—2023 from six national basic stations in the central Hexi Corridor, the Heihe river runoff data from Yingluoxia hydrological station in the upper reaches of the Heihe river, the FY-3D/MERSI satellite data, the GF-1 satellite data, and the NCEP/NCAR monthly mean reanalysis data, and 88 atmospheric circulation indexes and 26 SST indexes provided by National Climate Centre. The results show that precipitation and average temperature in the central part of the Hexi Corridor from May to September 2023 both exceeded historical extremes, with precipitation reached a new historical low and temperature reached a new historical high. The combined effect of high temperature and low precipitation significantly exacerbated the intensity of drought. In the middle Qilian mountains, the average snow cover area decreased by 45.8%, the annual runoff decreased by 15%, the main reservoir area decreased by 25.8%—66.0%. The characteristics of drought spread are meteorological drought—hydrological drought—ecological and agricultural drought. We find that precipitation in the central Hexi Corridor from May to September had significant multi-scale oscillations in 5, 11, 17 and 32 a, and multiple time scales showed that 2023 was in a period of less oscillation. When the drought in the central Hexi Corridor was observed, the atmospheric circulation anomalies included a southward shifted ridge line of the western Pacific subtropical high, an eastward shifted ridge point of the western extension, a westward shifted ridge point of the South Asian high pressure, a weaker than normal Indian low, a smaller polar vortex area in the northern hemisphere and Asia, a stronger blocking high in the middle and high latitudes of Asia, and increased longitudinal coverage of the westerly circulation in Eurasia. Meanwhile, positive SST anomalies occurred in the Nino3 area and the west wind drift region, while negative SST anomalies occurred in the Kuroshio region.

To obtain high precision wind profile and turbulence products and fully understand the application potential of wind lidar, quality control of Doppler wind lidar with five-beam swing (DBS5) mode is investigated in this work using radial velocity and signal-to-noise ratio (SNR). Three-dimensional wind and turbulence measurements by the DBS5 mode of wind lidar are systematically evaluated based on measurements of a three-dimensional ultrasonic anemometer mounted on a tower. The results show that the wind lidar exhibits excellent observational accuracy with root mean square errors (RMSEs) as low as 0.4 m/s, 0.1 m/s, 0.1 m/s, 0.1 m/s, and 0.5 m²/s² for horizontal wind speed (WS), vertical velocity (w), standard deviation of vertical velocity (σ_w), friction velocity (u_*), and turbulent kinetic energy (TKE), respectively at the height of 140 m. Moreover, the impacts of time scale, elevation angle, and spatial scale on observational accuracy of wind lidar are investigated. The results indicate that variations in time scale have little impact on observational accuracy, while elevation angle may affect the accuracy of TKE. Additionally, the RMSEs of WS and TKE gradually increase with increasing spatial scale, while the accuracy of w, σ_w, and u_* remain relatively stable. Further investigation of profiles of turbulence and vertical velocity under clear-sky conditions in Beijing indicates that the power spectra of vertical velocity is consistent with the classical −5/3 scaling law at different periods and heights, whereas white noise appears in the high-frequency region and intensifies with increasing height, especially above the boundary layer. Finally, the vertical turbulence of low-level jets (LLJs) observed by Doppler wind lidar is stronger below the jet height and weaker in and above the jet height.

To meet the need for weather forecasting, strong convection monitoring and warning services, an hourly 1 km wind gust product across China has been produced using hourly wind gust observations and model forecast data. These data are adopted after the procedure of quality control, space-time matching and fusion analysis. This product is updated at a 5 min lag with the overall RMSEs of 1.9 m/s for independent test and 0.68 m/s for the non-independent test. The results indicate that as the wind speed increases, the number of samples continuously declines, there is a relatively pronounced tendency for the product's error to rise, and the accuracy gradually decreases. However, the quality of this product improves effectively compared to model predictions of high speed of wind gusts. Specifically, the accuracy of wind speed above magnitude 9 has improved by 89.3%, and the relative error has dropped significantly with a reduction ratio of about 27.4%. The development of this product can support disaster mitigation and decision-making related to catastrophic gales and typhoons.

The strong cold pool is pivotal in the genesis of severe gales associated with squall lines, and its intensity simulation is closely related to parameter settings of cloud microphysical and boundary layer processes in the model. Despite parameter uncertainties, it remains challenging to apply reasonable parameter perturbations to squall line systems. To improve the performance of convective-scale numerical models in the forecast of squall line systems, based on the WRF (The Weather Research and Forecasting Model) model, five key parameters are selected from the cloud microphysical process and the boundary layer process to carry out sensitivity tests for the weak simulation of the cold pool associated with squall lines. Subsequently, the joint perturbation of the sensitive parameters is carried out, and the influence of this method on the simulation of a squall line process in Jiangsu is discussed. The results indicate that adjusting parameters that influence evaporation can significantly affect the estimation of the cold pool. Specifically, the parameter CONSTB, which reflects the impact of raindrop size on its terminal velocity, and the parameter VF1R, which accounts for the influence of surrounding airflow on raindrop behavior, exhibit the highest sensitivity to the cold pool dynamics. In the single-parameter and multi-parameter combined perturbation experiments, the simulated 2 m temperature in the cold zone of the squall line is 1—2℃ lower than that of the control experiment, which effectively overcomes the problem of weak simulation of the cold pool. In addition, the joint perturbation of CONSTB and VF1R parameters has a notable positive impact on forecast accuracy, with the simulated 10 m maximum wind speed being the most accurate in comparison to actual observations. Results show that the multi-parameter joint perturbation method for squall line cold pools effectively captures the uncertainty of parameters within physical parameterization schemes and improves cold pool simulation, thereby enhancing the accuracy of squall line gale predictions.

Based on the monthly Sea Surface Temperature (SST) data from the Met Office Hadley Centre, the Global Precipitation Climatology Project monthly precipitation data, and the historical simulations from Coupled Model Intercomparison Project Phase 6 (CMIP6) climate models, the present work investigates the seasonality of the SST-precipitation relationship over the tropical North Atlantic and possible role of ENSO (El Niño-Southern Oscillation). It is found that the relationship of SST anomalies in the tropical North Atlantic with local precipitation exhibits a remarkable seasonality. During spring and summer, there is a significant positive correlation between SST and precipitation in this region, indicating a strong local ocean-atmosphere coupling. In contrast, in autumn and winter, the ocean-atmosphere coupling weakens significantly, and almost no significant precipitation response to SST is detected. Further analysis reveals that this seasonality is mainly associated with the seasonal cycle of the background SST and local SST variability in the tropical North Atlantic. Despite cooler background SST in spring, the strong SST variability during this season makes SST easy to exceed the convection threshold and thus induces precipitation anomalies. In summer, the warm background SST favors the enhanced local ocean-atmosphere coupling. The relatively weak SST variability in autumn weakens the local precipitation response, despite a relatively warm SST background. The cooler background SST in winter results in a weak ocean-atmosphere coupling. ENSO has a significant influence on spring SST and precipitation anomalies in the tropical North Atlantic. As a result, the strong local SST anomalies in spring are more likely to actively trigger local convective responses under ENSO forcing. However, in other seasons, the impact of ENSO on SST anomaly in the tropical North Atlantic is relatively small, and thus there is almost no difference in local ocean-atmosphere coupling in the tropical North Atlantic with or without ENSO SST forcing. These findings emphasize the critical role of spring and summer tropical North Atlantic SST anomalies in local convection and associated climate impacts, which is important for short-term climate prediction related to the tropical North Atlantic SST.

An unprecedented persistent heavy precipitation occurred in Henan province during 17—22 July 2021, causing huge economic losses. Currently, extreme precipitation forecasting is still a hotspot and a difficult issue in sub-seasonal climate prediction research. Regional climate models provide a new way to further improve sub-seasonal precipitation forecasting in China with finer spatial resolution and better parameterization of physical processes compared to that of the global models. This study uses the regional Climate-Weather Research and Forecasting model (CWRF) nested with the China Meteorological Administration Climate Prediction System version 3 (CMA_CPSv3) to improve prediction capabilities for this persistent heavy precipitation event. It is shown that the spatial distribution, magnitude, and forecast accuracy of precipitation predicted by CWRF are improved compared to that predicted by CMA_CPSv3. Although both models underestimate the amount of precipitation, the CWRF forecasts larger accumulated precipitation and spatial distribution of precipitation is more consistent with observation. CWRF forecasts initialized on 26 June and 29 June are better than that of CMA_CPSv3 on the same initial dates. The CWRF significantly improves the forecast of low-level wind fields and low-level jets in East Asia compared with the CMA_CPSv3. The CWRF is particularly effective in improving the simulation of directions of low-level jets and water vapor fluxes, allowing water vapor to converge on the windward slopes of mountain ranges and providing favorable water vapor conditions for precipitation. The CWRF better forecasts the water vapor flux convergence and ascending motions over Zhengzhou, and all these improvements lead to higher precipitation forecasting skill of CWRF.

Back-building MCSs (Mesoscale Convective Systems) are highly conductive to sudden, localized short-duration heavy rainfall. In order to reveal the characteristics of this type of MCS and its association with heavy rainfall, this study systematically studies spatial and temporal distribution of back-building MCSs that triggered short-duration heavy rainfall during the warm seasons from 2015 to 2021 in Zhejiang province. Different organizational forms and environmental thermodynamic factors of different types are also explored. The results show that back-building MCSs in Zhejiang province exhibit significant monthly and diurnal variation patterns, i.e., MCSs mainly occur in June and July with peak hourly rainfall intensities of 30 and 50 mm in these two months, respectively. The MCS primarily form between 11:00 and 14:00 BT, with the highest frequency of formation occurring between 12:00 and 13:00 BT. The majority of MCSs have a duration of 12 h or less, with 10 h duration being the most common. The start time of backward propagation shows a quasi-bimodal pattern, which is 2—3 h later than the main formation time of the MCS. For 90% of the cases, the time of maximum hourly rainfall intensity occur within 0—2.5 h after the onset of the backward propagation. Based on the evolution characteristics of convective system organization, the back-building MCSs with short-duration heavy rainfall in Zhejiang province can be categorized into four types: Advective, quasi-stationary, turning, and propagating MCS, with about 42% occuring under the forcing of weak synoptic-scale system. The MCS usually occurs in an environment with medium convective available potential energy (CAPE), high humidity and appropriate vertical wind shear, but with different environmental factors for different types. The quasi-stationary MCSs account for the largest proportion (44.7%) and are characterized by significant environmental dynamic features, including strong atmospheric instability, large steering flow, and mid-to-lower-level vertical wind shear. They result in relatively weak maximum hourly rainfall intensity (the median is 50 mm/h). Propagating MCSs (accounting for about 17%) exhibit more distinct environmental thermodynamic characteristics with large CAPE and precipitable water (PW), and lead to the strongest maximum hourly rainfall intensity (the median is 70 mm/h).

To deepen understanding of mesoscale convective systems under the special topography of Hunan province, this paper uses high spatiotemporal resolution data obtained from the Variational Doppler Radar Analysis System to study the extreme precipitation process that occurred in Hunan province from 20:00 BT 29 May to 06:00 BT 30 May 2022. Results indicate that this extremely heavy precipitation event occurred in the convergence zone between dry, cold northerly airflow and warm, moist southwesterly jet at 700 hPa. In the initial stage, clustered convections were sporadically triggered and gradually organized into a banded mesoscale convection system. During the development of the banded mesoscale convection system, positive vorticity formed due to environmental vertical wind shear and negative vorticity generated by the cold pool gradually approached equilibrium, which, coupled with the enhancement of the southwesterly jet that transported a large amount of moisture, resulted in rapid development of the banded mesoscale convection system with extreme precipitation reaching 103 mm in one hour. In the maintenance stage, the compensatory downdraft for the updraft within the banded mesoscale convection system suppressed convection generation in the central part of the system. Additionally, the downdraft enhanced raindrops evaporation in the middle and lower levels and transferred horizontal westward momentum to the near-surface, intensifying convection in the eastern part of the banded mesoscale convection system and resulting in splitting of the convection system into a clustered mesoscale convective system. As the convection moved southward, the low-level southwesterly was blocked by Mingyang and Xuefeng mountains. As a result, new convections were mostly triggered on the west side of the Xiangjiang river valley, resulting in larger total precipitation there.

Atmosphere-wave interaction is a crucial dynamic process at the air-sea interface, with the sea surface momentum roughness length being a key variable in the coupled atmosphere-wave modelling system. The Global-Regional Integrated Forecast System (GRIST), a next-generation unstructured-grid unified weather and climate modelling system, has been independently designed and developed in China in recent years. By employing the ESMF/NUOPC framework, GRIST has been integrated with the WW3 model to create the coupled atmosphere-wave modelling system (GRIST-WW3). In this system, the atmospheric model provides 10 m wind fields over the sea surface to drive the wave model, while the sea surface roughness, derived from a wave parameterization scheme, is fed back into the atmospheric model. Preliminary results show that the GRIST-WW3 system accurately captures spatial distribution of sea surface wind field and significant wave height, both of which agree well with observations. However, in regions such as the Southern Hemisphere's westerlies and areas near typhoons, where the wind speeds are notably high, the model tends to overestimate 10 m wind speed and significant wave height. The two-way coupling process increases the average and dispersion of sea surface roughness, reducing wind speed biases in areas of high wind speed. In terms of typhoon simulation, improvements in the simulation of typhoon trajectory and 10 m maximum wind speed are evident with the atmosphere-wave coupled modelling system, although the minimum sea level pressure remains unaffected. In the coupled atmosphere-wave modelling system, the wave parameterization scheme of sea surface roughness is essential for accurately simulating high wind speed areas. The optimization of the scheme should be guided by the atmospheric model's bias characteristics, with the primary goal of reducing bias in high wind speed regions.