Abstract:This study analyzed the temperature data in the course of 29 mixed fog, haze weather in 2015 from L-band sounding radar at the Zhangqiu sounding station and temperature data from the German 14-channel RPG-HATPRO-G3 microwave radiometer located at the Shandong Meteorological Bureau. The quality analysis of these observation data was performed to determine their accuracy and reliability. In particularly, the macro physical characteristics and diurnal variation of the physical parameters in haze weather were analyzed. The changes of PM_2.5, PM₁₀, SO₂, NO₂, O₃, CO in the atmosphere were analyzed for each case of fog, haze, and haze .The changes of relative humidity, liquid water path, and water vapor content were analyzed. The results indicate that data from the two observation instruments have good consistency, with a goodness of fit higher than 0.97 under fog and haze conditions. There are seasonal variations in the surface inversion layer and some differences in the physical parameters of the overhanging inversion layer. The changes in the temperature inversion layer and pollution parameters are closely related to the formation of fog and haze. Different weather backgrounds are determined to have considerable influence on the physical atmospheric parameters, with obvious fluctuations in the PM₁₀, AQI (air quality index), and CO values, which peak during the same time period. The CO peak in fog and haze weather is particularly obvious, CO peak with a gradual increase from the morning to night. The CO peak of the fog and haze days is smaller, and fog and haze weathers are obviously larger than the fog or haze.

Abstract:The accurate quantification of soil heat flux in the freezing process of the alpine wetland in the source area of the Yellow River has an important scientific significance for understanding the water and heat exchanges between alpine wetlands and the atmosphere. By using the field observed data collected from the Maduo climate and environment comprehensive observatory of the Chinese Academy of Sciences from May 2014 to May 2015, the characteristics of soil heat flux as the alpine wetlands froze were analyzed. The effect of the latent heat of fusion on soil heat flux was also discussed. Both the heat storage and latent heat of fusion loss from above the plate must be considered when calculating the soil heat flux at the alpine wetland using the simple measurement approach algorithm. If the latent heat of fusion is ignored, then large errors can be found. The main results are as follows. (1) After the freezing front appeared, soil heat flux at a depth below the freezing front decreases and approaches zero, the liquid water content of the soil at the depth of the freezing front decreases rapidly, and the soil below the freezing front froze. In addition, the freezing released latent heat travels upward through the soil layer where the soil heat flux plate is located and observed. As the precipitation infiltrates into the soil, thus releasing the freezing latent heat, the freezing latent heat causes the observed soil heat flux to approach zero at a depth of 5 cm. (2) During the seasonally freezing processes, upward soil heat flux at a depth of 5 cm approaches zero if there is a high temperature in the morning and at noon of the previous day. This phenomenon indicates the existence of a diurnal freezing-thawing cycle. The latent heat released by soil water can reduce the amplitude of soil temperature and keep the soil temperature near the freezing point. The diurnal freezing-thawing processes solely occur in a very shallow soil layer, making it difficult to ascertain whether the diurnal freezing-thawing cycle happened not just by using soil temperature data at a depth of 5 cm. (3) Considering the latent heat of fusion factor decreases the root mean square errors of soil heat flux between the observed and calculated values from 11.5 W m^-2 to 6.2 W m^-2. These findings can contribute towards a better understanding of the land surface processes in cold regions.

Abstract:The sudden change of typhoon track is a challenging issue in typhoon forecast. Typhoon “Meranti” (2010) experienced an abrupt northward turning when it moved westward to the south of Taiwan Island. Ensemble forecast based on the European Medium Range Weather Forecasts (ECMWF) presented a large spread for the northward turning. The good group that consisted of eight members and predicted the abrupt northward turning is compared to the poor group that didn’t predict the unusual track to study the main influence factors on the northward turning of typhoon Meranti. Results show that: (1) An upper-tropospheric cold low (UTCL) cut off from the tropical upper-tropospheric trough (TUTT) was an important factor impacting the typhoon track change. The abrupt northward turning of typhoon Meranti was related to the coupling between Meranti and the nearby UTCL in the north-south direction. (2) The UTCL could change the upper-tropospheric environmental flow around the typhoon, and influence the typhoon steering flow. When the UTCL moved to the north of Meranti, the southerly component of the steering flow was enhanced, which was helpful for the northward movement of Meranti. (3) The northward turning of Meranti was also associated with the vertical structure of the UTCL. The UTCL with wide and deep cyclonic circulation in vertical direction was more favorable for the track change. (4) The diagnostic analysis based on the potential vorticity tendency equation shows that, under the coupling between Meranti and the UTCL in the north-south direction, horizontal advection of potential vorticity to the north of the typhoon was strengthened, which was helpful for the northward movement of Meranti. And the asymmetric advection of symmetric potential vorticity made a major contribution to the northward track under the influence of the UTCL.

Abstract:In this paper, combined complex Empirical Orthogonal Function（CEOF）analysis is applied to the atmospheric wind and ocean flow fields in January from 1950 to 2001 to understand the coupled ocean-atmosphere circulation associated with two types of El Niño-Southern Oscillation（ENSO）in January. The results are as follows. The first and second modes respectively correspond to the Eastern Pacific type of ENSO (EP-ENSO) and the Central Pacific type of ENSO (CP-ENSO). The former type of ENSO corresponds to sea surface temperature (SST) anomalies in the eastern and central equatorial Pacific, and the eastern anomaly is stronger; the latter corresponds to SST anomaly only in the central equatorial Pacific. Their time coefficients are closely related to ENSO. Thereby, they can be named as EP-ENSO mode and CP-ENSO mode regarding wind-current fields, respectively. The EP-ENSO mode has distinct interannual and interdecadal variations with periods of 3 to 6 and 13 to 14 years. The CP-ENSO mode has distinct interannual and interdecadal variations with periods of 7, 12 and 17 years. The 13-year period is similar to the NPGO（North Pacific Gyre Oscillation）period of winter climate mode in the North Pacific. During the Eastern and Central Pacific type of El Niño, the ascending branch of the Walker Circulation shifts eastward from Indonesia to the western or central equatorial Pacific, and the intensity weakens; the southern and northern branches of Hadley Circulation are respectively located to the east of and near the international date line, and the intensity enhances, which causes the subtropical high to become stronger; along the coast of South America near 5°S, there are anomalies of updraft and downdraft. For the two types of ENSO modes, westerly anomalies and wave packet solutions of Kelvin wave and Rossby wave are coupled near the central equatorial Pacific. Sea temperature kinetic anomaly has a negative feedback on the atmospheric circulation, which is favorable for the maintenance and stability of ENSO.

Abstract:Based on the data obtained from the 2015 summer campaign in Beijing area, including the total lightning location data from Beijing Lightning Network (BLNET), S-band Doppler radar data, ground-based automatic weather stations observations and radiosonde data, the evolution of lightning activities during a severe squall line process that occurred over Beijing metropolitan region on 7 August 2015 was analyzed. Its relation to convection region and surface thermodynamic condition was also discussed. According to radar echoes and lightning occurrence frequency, the whole squall line process can be divided into three stages (developing, intensifying and weakening), and the intra-cloud (IC) lightning flashes predominated during all the three stages in general. In the developing stage, several isolated γ mesoscale convective cells rapidly developed. With the echo top of the storm cell over Beijing metropolitan region extending to -20℃ level, lightning activities significantly increased, and the lightning radiation sources gradually spread to upper altitudes, but lightning rate was still less than 80 flashes/min for the whole system. In the intensifying stage, the flash rate increased rapidly, which was associated with the merging process of the cells. When the squall line formed, the volume of strong radar echoes (＞40 dBZ) increased significantly for both above and below 0℃ levels, and the total flash and cloud-to-ground (CG) flash peaked with rates of 248 flashes/min and 18 flashes/min, respectively. Negative CG flashes accounted for 90% of the total CG flashes. The lightning radiation sources were mainly detected in the linear convection area, and the number of radiation sources peaked within the layer of 5-9 km. In the weakening stage, the core of the squall line dropped below 0℃ level and quickly decayed, with the radiation sources obviously sloping backward to the area of stratiform clouds. About 95% of total flashes occurred within 10 km of the convective line, namely the convection and transition region. During intensifying and weakening stages, radiation sources reached active period simultaneously in the convection and stratiform region, while during the weakening stage, radiation sources in the convection region declined abruptly in the number. Lightning flashes mainly occurred over regions with strong surface equivalent potential temperature gradient induced by the outflow of convective cold pool and the relatively warm moist airmass from the plain.

Abstract:This study statistically analyzes gravity waves activity in the middle and upper troposphere (6-12 km) based on radiosonde data obtained at four stations, i.e., Jiange, Jinchuan, Jiulong, Mingshan, in the Intensive Atmospheric Science Observational Experiment of Southwest China vortex during 2012-2015. The results reveal downward propagation of gravity waves energy at Jiulong station, suggesting that the gravity waves are generated in the upper troposphere at Jiulong station. The opposite is true at other three stations. Corresponding to different types of the southwest vortex, gravity waves activities are quite different. At the initial stage of the moving-out type of southwest vortex, gravity waves mainly propagate northeastward with larger probability of upward wave energy transport compared to downward transport. Meanwhile, kinetic and potential energy are large with significant variation. At the initial stage of source southwest vortex, however, gravity waves largely propagate along the northwest-southeast direction, while wave energy can be transported upward or downward with similar probability. Meanwhile, kinetic and potential energy are relatively small with weak variation. It is noted that the horizontal propagation direction of gravity waves are important for the moving direction of southwest vortex. Gravity waves are divided into nocturnal type and daytime type based on their generation time. It is found that more nocturnal-type gravity waves [2000 BT to 0800 BT (Beijing time)] are generated at the initial stage of nocturnal southwest vortex, indicating that there may exist a correlation between the generations of nocturnal gravity waves and southwest vortex.

Abstract:Historical simulation outputs from the Coupled Model Intercomparison Program Phase 5 (CMIP5) climate models with a categorization method were used in this study to evaluate the performance of coupled models in simulating the Pacific Decadal Oscillation (PDO). While several of the 40 models under examination failed to reproduce the characteristics of PDO pattern, the majority of the models can reproduce the interdecadal cycle of PDO well. The good performance of the category-1 models in simulating the PDO pattern can be attributed to the relationship of the SSTA (sea surface temperature anomaly) between tropical Pacific and North Pacific being captured by these models. Further, the impact of the decadal SST variation in the tropic Pacific on the SST variation in the North Pacific via atmospheric teleconnection was reproduced using these models. In contrast, the models with poor simulation for the PDO pattern failed to reproduce the tropics-extratropics linkage in the SST anomalies that was induced by atmospheric teleconnection. This result indicates the importance of the decadal SST variation in the Tropic Pacific for the formation of PDO. Under RCP4.5 (Representation Concentration Pathway Scenarios 4.5), it is suggested that the first EOF (empirical orthogonal function) of SST variability over the North Pacific for the 21st century is a uniform positive pattern with a corresponding time series indicating an upward trend. Meanwhile, the second leading pattern of the 21st-century EOF analyses shows the spatial variability of the PDO diploe pattern.

Abstract:The daily high-resolution gridded station-observed rainfall over China and NCEP-DOE AMIP-II atmospheric reanalysis datasets for the period 1979-2015 are used to investigate the impact of the 30-60-day atmospheric intraseasonal oscillation (ISO) over the Asian monsoon region on persistent anomalous rainfall over the middle and lower reaches of the Yangtze River (MLYR). The interaction between ISOs over two sub-monsoon areas of South Asia and East Asia as well as the physical mechanism for the ISOs to jointly influence persistent extreme rainfall over the MLYR are revealed. Composite analyses demonstrate that the ISO over the South Asian monsoon region interacts with its counterpart over the East Asian monsoon region through upper-level divergent flows. During phases 1‒3 (5‒7) of the ISO, active (inactive) convection anomalies propagate northward from the equatorial Indian Ocean to the Indian Subcontinent‒Bay of Bengal. The accompanied divergence (convergence) anomalies in the upper troposphere cause anomalous convergence (divergence) downstream over the South China Sea‒western North Pacific (SCS‒WNP) through compensation effect, leading to anomalous descents (ascents) locally over the SCS‒WNP, and therefore is favorable for the development and persistence of anomalous inactive (active) convection anomalies. Such inactive (active) convection anomalies are characterized by prominent baroclinic divergence structure and further trigger a meridional-vertical circulation cell between the SCS and the Yangtze Basin, resulting in anomalous ascents (descents) and low-level moisture convergence (divergence) over the MLYR. As a consequence, positive (negative) rainfall anomalies continuously arise over the MLYR with the occurrence probability of extreme rainfall significantly increasing (decreasing), favorable (unfavorable) for the occurrence of persistent extreme rainfall events. Moreover, the intensity of ISO over the Asian monsoon region exhibits strong interannual variations and further modulates the frequency and duration of persistent extreme rainfall events over the MLYR. In strong (weak) ISO years, more (fewer) persistent extreme rainfall events occur over the MLYR and the duration is longer (shorter).

Abstract:This paper comparatively analyzes two rainstorm processes in Sichuan Province affected by the Tibetan Plateau vortex and Southwest vortex based on the ECMWF ERA-Interim reanalysis dataset, the CMORPH data (which is derived from station observations and satellite retrievals of precipitation), the hourly cloud top TBB (black body temperature) data of FY-2E and results of the Lagrangian trajectory model HYSPLIT v4.9. Major results are as follows. (1) The intensity and distribution of precipitation are significantly different in the two heavy rain processes, and the characteristics of mesoscale convective cloud system showed that MCC (mesoscale convective complex) was triggered in process one, but not found in process two. (2) During process one, the Tibetan Plateau vortex gradually weakened into a plateau trough and spread over the Sichuan Basin at 500 hPa. At 850 hPa, the MCC was triggered near the wind convergence zone. At 200 hPa, the southwest vortex was located to the southeast of South Asia high in an area of strong divergence below the westerly jet, which contributes to the intensification of the Southwest vortex. For the second process, however, the Tibetan Plateau vortex was gradually coupled with the southwest vortex, resulting in a stable, deep system. At 200 hPa, a distinct pumping effect can be found over the Sichuan Basin since it was always located below the northwesterly flow of the South Asian high. (3) During process one, PV (potential vortex) gradually spread to the east and the area where PV increased corresponds to strong precipitation area and MCC outburst area, which reflects the basic consistency between the development of rainstorm and PV. Besides, in the second process, the high PV value zone in the mid-level moved eastward from the plateau and the high PV was transmitted down to the basin. The coupling of the two vortices makes the PV value lager than that of each single vortex. Also, this article has confirmed a good relationship between the distribution of abnormal second-order PV and heavy rain. (4) The main difference between the two processes is the generation of MCC, which requires an atmospheric condition with strong positive vorticity, strong convergence and strong ascending movement. And the dynamic conditions in the two rainstorm processes may be opposite before the formation of MCC. From the thermodynamic perspective, there is an invasion of dry and cold air in process one, and the large gradient of potential pseudo-equivalent temperature benefits the generation of mesoscale convective system and causes heavy rain. (5) According to the Lagrangian trajectory model and clustering analysis method, there are two water vapor channels in process one: the first channel is from the bottom layer of the Arabian Sea and the Bay of Bengal, while the second channel is from the south of Sichuan at 750 m above the ground. In the second process, there are three main vapor transport channels. Channel 1 is from the western Mediterranean, Black Sea and Liberia in the layer between 1500 m to 2500 m. Channel 2 is from the bottom of the Arabian Sea and the Indian Ocean. Water vapor of channel 3 is directly transported to the Sichuan Basin from the bottom of the Bay of Bengal, passing by the Yunnan-Guizhou Plateau.

Abstract:The Indian Ocean Dipole (IOD), which is one of the dominant interannual variability modes of SST (sea surface temperature) in the tropical Indian Ocean, has striking impacts on regional and global climate. Thus, finding ways to achieve accurate and short-term climate predictions of IOD is an important subject of research. Recently, a near-term climate prediction system called IAP-DecPreS was constructed by the Institute of Atmospheric Physics (IAP), Chinese Academy of Sciences. It is based on a newly developed initialization scheme (EnOI-IAU), which can assimilate the observed ocean temperature profile records. In this paper, the authors compare the differences in skills of the IAP-DecPreS for the IOD in fall (September-November) between the following two distinct initialization approaches: anomaly and full-field initializations. The results indicate that, for predictions starting from August, the hindcast runs based on the full-field initialization are more skilled at both deterministic and probabilistic predictions compared with those based on the anomaly initialization. For predictions starting from May, the predictive skill of the hindcasts based on the two initialization approaches are similar. Compared with the anomaly initialization, the full-field initialization is superior because it improves the predictive skill for the IOD events occurring together with ENSO. The wind-evaporation-SST positive feedback over the tropical eastern Indian Ocean, which is excited by the ENSO remote forcing, is key for the development and maintenance of the IOD. The hindcasts based on the full-field initialization can reproduce the spatial distributions of precipitation and wind anomalies associated with the ENSO during the IOD development stage. In contrast, for the hindcasts based on anomaly initialization, the biases of precipitation and wind anomalies are much larger. Full-field initialization can reduce the initial errors in the climatological precipitation over the tropical Indian Ocean, thus improving the accuracies in simulating the response of precipitation and wind anomalies over the tropical Indian Ocean to the ENSO remote forcing. In comparison, the anomaly initialization nearly does not change the model inherent climatology. Thus, the ENSO-related precipitation and wind anomalies over the tropical Indian Ocean simulated by the hindcasts based on anomaly initialization show biases similar to those of the model’s historical runs.

Abstract:The land surface provides underlying lower boundary conditions for atmospheric models. Topography plays a crucial role in the accuracy of the model results. Most of current land surface models use the same atmospheric forcing in subgrid units within the same model grid, and don’t not consider the influence of subgrid topography on atmospheric forcing that would impact the simulation of meteorological elements and land-atmosphere interaction. In this paper, a modified Land Surface Subgrid Topographic Parameterization (LSSTP) is proposed to revise the input atmospheric forcing according to its relationship with the subgrid terrain height in the NOAH land surface model . The LSSTP is then introduced into the WRF (Weather Research and Forecasting) model for numerical experiments. Three groups of numerical experiments have been conducted to investigate the effect of this improvement on the simulation results of the WRF model. It is found that the new LSSTP introduced in this paper has clearly improved the simulation of 2m air temperature in the surface over the Tianshan Mountains, the Kunlun Mountains and the southern Qinghai-Tibet Plateau. However, the new LSSTP shows little improvement on precipitation simulation. In contrast, the experiments with higher resolution can well simulate precipitation, which is attributed to the fact that the experiment with higher resolution has been refined in both the land and the atmosphere, while the new LSSTP only considers subgrid effect in the surface. Temperature simulations using the new LSSTP are improved by revising surface upward long-wave flux and surface sensible heat flux, while the experiment with higher resolution has simultaneously refined the grids of both the atmosphere and land, and the improvement of precipitation and temperature simulation is achieved by comprehensively changing the surface energy balance.

Abstract:In the autumn and winter, severe air pollution events occur frequently in Beijing-Tianjin-Hebei region. As is known, industrial emissions and residential coal burning are the major reasons for atmospheric pollution. As a typical industrial city focusing on glass manufacturing and processing, Shahe city is selected as the research object. An air diffusion model and potential source analysis are employed to study spatiotemporal characteristics and sources of major pollutants based on the atmospheric monitoring data in Shahe city from January to December, 2017. Major results are as follows. (1) There are obvious seasonal characteristics of primary pollutants in Shahe. The primary pollutant is PM₁₀ in the spring, O₃ in the winter, and PM_2.5 in the autumn and winter. Their seasonal contribution rates are 43.3%, 72.3% and 61.5% respectively. (2) Affected by changes in urban atmospheric boundary layer and pollution emissions, the concentrations of PM₁₀, PM_2.5, SO₂, NO₂ and CO have strong seasonal-diurnal changes. (3) When northeasterly winds prevail in the winter, correlation coefficients between PM_2.5, NO₂, SO₂ and their concentrations are high, which may be attributed to emissions from glass factories to the northeast of the city. Meanwhile, the observational site may also be affected by coal burning in surrounding villages. (4) The mean concentration of PM_2.5 in Shahe is 143 μg m^-3 in the winter. The analysis of a severe pollution case shows that the max values of SOR (sulfur oxidation ratio), NOR (nitrogen oxidation ratio) are 0.67 and 0.39, respectively. This indicates that the conversion from primary gaseous pollutants to secondary pollutants is severe. It also proves that high humidity does favor the formation of air-borne particulates. The mean value of C(NO₃^-)/C(SO₄^2-) is 1.89. It is concluded that NO₂ in Shahe mainly comes from emissions of heavy transport vehicles and factories. (5) Local area is a strong potential source for Shahe’s PM_2.5. The surrounding industrial cities also make some contributions. Not only local emission reduction, but also regional management regimes are needed to control the PM_2.5 pollution.

Abstract:The interaction of East Asian summer monsoon (EASM) members constitutes the “the coupling wheels” pattern for the circulations over the upper and lower layers of EASM. In this paper, the variation, mechanisms, temporal and spatial stability of the coupling wheels as well as their impacts on abnormal precipitation over China in the East Asian summer monsoon (EASM) region are diagnosed and analyzed using Multi-Variable Empirical Orthogonal Function (MV-EOF). On the basis of the above work, conceptual models are constructed for typical coupling wheels in the EASM and their impacts on summer rainfall. Our results show the coupling wheels are manifested in interannual variation (with the period of about 2-6 years) and forced by vertical temperature and pressure fields as well as the large terrain of the Tibet Plateau. The first two modes of the coupling wheels are stable and reflect typical linkages between the EASM members. In the first mode, the Northern Cyclone (NC), the South Asian high (SAH) and the western Pacific subtropical high (WPSH) are the major coupling components. As a barotropic system, the NC couples with the SAH through westerly winds on its southern side, while the SAH interacts with the WPSH by divergent descending air flow in the eastern part of middle-latitude area. When this mode enhances, summer rainfall distribution in China shows a pattern of “＋－＋－” from north to south. The second mode mainly reflects the coupling characteristics of the mid- and high-latitude cyclones, the East Asian subtropical westerly jet (EASWJ), the SAH, the Northwest Pacific anticyclone (NPA) and the WPSH. As barotropic systems, the mid- and high-latitude cyclones and NPA are linked by southeasterly winds. In addition, the cyclones are coupled with the SAH and the EASWJ through westerly winds on the southern side of the cyclones in upper layers. While the intensity and area of the WPSH are affected by easterly winds on the southern side of the NAP in lower layers. Enhancement of this coupling mode is favorable for more precipitation over areas to the north of the Yellow River and the HETAO region, but less rainfall would occur to the south of the Yellow River.

Abstract:The locally triggered or organizational development of mesoscale convective systems is the core and key for the accuracy of convective precipitation nowcasting. In this paper, combined with radar extrapolation prediction technology, expert system, and high-resolution numerical model system, a new adaptive convective precipitation ensemble nowcasting method combined with diagnosis from numerical weather prediction mode for Beijing area is developed Based on the precipitation extrapolation, APEN (adaptive precipitation ensemble nowcasting) uses convective diagnostic factors provided by the RMAPS-IN (Rapid-refresh Multi-scale Analysis and Prediction System — Integration) to calculate the probability of convection evolution (initiation, growth, and dissipation) by a fuzzy logic algorithm, and get an ensemble probability by disturbing diagnosis factor thresholds and weights. Then, based on the expert experience, the adjustment of convective precipitation based on the extrapolation of precipitation is carried out. APEN has been tested to make precipitation nowcasting with two cases (one is with heavy rainfall and the other is with weak rainfall) in Beijing. The results show that the ensemble convective evolution probability, which is based on the RMAPS-IN multi-diagnosis factors, can reflect the trend of the convective system in both strong and weak weather conditions. Based on the expert experience model, adjustments of convective precipitation under three states (convection initiation, growth, and dissipation) can reasonably represent the impact of precipitation by evolution of convective systems. Comparison of precipitation nowcasting by APEN and RMAPS-IN (an operational system of Beijing Meteorological Bureau) shows that, regardless of whether the convective system is stimulated by systemic convection activity or locally triggered, the precipitation area and precipitation intensity predicted by APEN are closer to observations. In particular, the precipitation intensity forecast after considering the effect of convective evolution is much better than RMAPS-IN. The experimental comparison demonstrates APEN’s advantages and application potential in convective precipitation forecasting.

Abstract:The initial perturbation amplitude and ensemble size are important for ensemble forecast. The present study explores the impacts of initial perturbation amplitude and ensemble size on the ensemble forecast skill using a new strategy that applies orthogonal conditional nonlinear optimal perturbations (CNOPs) to the Lorenz-96 model. It is found that due to the effect of nonlinearity, the CNOPs-based ensemble forecast can achieve higher skills when the initial perturbation amplitude is appropriately smaller than the amplitude of initial analysis errors, and the highest skill of the CNOPs-based ensemble forecasts is always higher than that of its linear counterpart [i.e., singular vectors (SVs)-based ensemble forecast]. The results also show that an appropriate ensemble size is helpful for achieving higher skills in ensemble forecast. A better spread-skill relationship and a much flatter Talagrand diagram are found in CNOPs-based ensemble forecast, which indicates the reliability of the corresponding ensemble forecast system and makes the above results much solid. It is therefore inferred that the highest skill of CNOPs-based ensemble forecast is mostly likely achieved when initial perturbation amplitudes are properly smaller than those of initial analysis errors and the ensemble size is appropriate.

Abstract:The evolution of dynamic and thermal structure and instability condition of an extreme precipitation system in Beijing-Tianjin-Hebei are analyzed using the NCEP/NCAR Global Forecast System (GFS) data combined with national automatic stations observations of precipitation provided by the Meteorological Information Center of China Meteorological Administration, the CMORPH satellite precipitation data and the fusion of precipitation data from the FY2 precipitation and radar quantitative estimation of precipitation. This study reveals the configuration of weather systems in different air pressure zones over the Beijing-Tianjin-Hebei region. Vertical motion, water vapor condition and unstable stratification evolution during the precipitation process are explored. The results are as follows. (1) The circulation at 500 hPa presented an east-high-west-low pattern, which was coordinated with a low-level vortex at 700 hPa and jet streams in low and high levels. The subtropical high blocked the eastward movement of the low-level vortex in North China, making it stagnant in the Beijing-Tianjin-Hebei region. (2) The development and eastward-moving of the low level vortex was important for the occurrence of the rainstorm. (3) The potential divergence analysis was applied to explore changes in convective instability. Results indicate that in the rear of the precipitation area, lower-level potential instability was mainly determined by vertical wind shear, which reflected the joint effects of vertical wind shear and moist baroclinicity. Lower-level potential divergence was negative in weak precipitation area and behind the precipitation area, which was conducive to regional potential instability. Potential divergence was positive in strong precipitation area and in front of precipitation area, inhibiting the development of potential instability. The change in potential divergence affected precipitation region through affecting the atmospheric stability. The high value region of potential divergence corresponded to high value region of precipitation, especially the 700 hPa potential divergence was a good indicator for precipitation region, which could be estimated by the change of potential divergence at 700 hPa.