In order to improve the accuracy of rainstorm forecast and reduce the disaster losses caused by rainstorm, based on the ground conventional meteorological observation data, TBB (Black Body Temperature) data from satellite images and reanalysis data from National Centers for Environmental Prediction (NCEP), the synoptic causes of a strong convective rainstorm in Yunnan in August 2017 were analyzed. The results show that the eastward movement of 500 hPa trough, the southward movement of 700 hPa shear line and the westward movement of surface cold front are the synoptic background of this precipitation process. The mesoscale convective systems (MCS) on the Meso-α and Meso-β scales directly trigger the convective rainstorm. The heavy rainfall generally happens in the region with a high gradient of TBB. The MCS is closely related to 700 hPa wind shear line where is located to the east of Mid-Yunnan. The MCS is elliptically-shaped, developing along the neighboring and backside of the wind shear line. After the wind shear line getting close and cross over the Ailao Mountain, the MCSs distribute in a belt from northwest to southeast, and develop in front of the wind shear line. The wind shear line moves fast during the daytime before crossing over the Ailao Mountain, mainly producing thunderstorm weather while it moves slowly at nighttime, and the rainfall is strong. The forecast of strong convective rainstorm should focus on the large value area of water vapor flux convergence and the area where the temperature difference between 800 hPa and 500 hPa is greater than 20 ℃. During the heavy rainfall, the whole layer atmosphere is ascending, and the heavy rainfall area maintains the dynamic pumping mechanism of convergence at lower level and divergence at middle and upper levels.

In recent years, Ningxia relies on unique natural resources to develop vigorously starry sky tourism industry, but the starry sky tourism is greatly affected by meteorological conditions. Based on the meteorological observation data at 25 weather stations of Ningxia, reanalysis data from European Centre for Medium-Range Weather Forecasts and radiance data of satellite from National Oceanic and Atmospheric Administration, the composite evaluation model of climate resources suitability of starry sky tourism in Ningxia is established from five aspects, such as cloud cover, light pollution, visibility, seeing and human body comfort degree, and the climate suitability of starry sky tourism in Ningxia is evaluated in different seasons. The results show that the starry sky tourism is suitable or very suitable in most areas of Ningxia all year round, while the suitability in urban area of cities (counties and districts) is relatively lower. The very suitable area to starry sky tourism is the largest in summer, and mostly in the plain area, followed by autumn and winter, and that is the smallest in spring.

In order to explore the characteristics and influence factors of atmospheric boundary layer inversion in Yinchuan City and their relationship with winter PM_2.5 pollution, based on the sounding and surface meteorological observation data at Yinchuan meteorological station and the air quality monitoring data in Yinchuan City from 2015 to 2020, the characteristics of atmospheric boundary layer inversion and surface meteorological elements are analyzed. And on this basis that the influence of inversion and surface meteorological elements on PM_2.5 pollution weathers in winter is explored. The results are as follows: (1) The inversions in atmospheric boundary layer are more easily to occur in the early morning than in the evening in Yinchuan City, and most of the inversions are ground inversions, they are stronger and thinner than suspension inversions. The inversion frequency and thickness are the largest in winter and the smallest in summer, while the inversion intensity is the strongest in autumn and the weakest in summer. (2) In winter, the inversions are easily to occur on a sunny day with average surface wind speed from 1.0 to 1.5 m·s^-1 and relative humidity from 30% to 60%. (3) Ground inversion is one of the main meteorological factors affecting PM_2.5 pollution weathers in winter. When the inversion thickness exceeds 596 m and the intensity exceeds 1.4 ℃·(100 m)^-1, PM_2.5 pollution weathers are easily to occur, and the pollution is aggravated with the increase of inversion thickness and intensity. (4) Under PM_2.5 pollution weathers in winter, the sky is mostly sunny in the early morning, the average surface wind speed is usually less than 1.3 m·s^-1, the relative humidity is greater than 54%, and the pollution is aggravated with the increase of humidity. (5) The boundary layer height is significantly negatively correlated with PM_2.5 mass concentration, and the lower boundary layer height is, the heavier PM_2.5 pollution is.

Based on ERA-Interim reanalysis data from ECMWF (European Centre for Medium-Range Weather Forecasts), precipitation retrieval products 3B42RT from tropical rainfall measuring mission (TRMM) satellite, measured data by Dual-frequency Precipitation Radar (DPR) on board global precipitation measurement (GPM) satellite, FY-2F cloud classification (CLC) and black body temperature (TBB) products, the southwest vortexes affecting precipitation in central-eastern China from April to October (the warm season) during 2010-2020 were analyzed. The results show that there were 108 cases of southwest vortexes moving out of the origin along different paths, and the eastward movement type, northeastward movement type and southeastward movement type accounted for 58.3%, 27.8% and 12.0%, respectively. The eastward movement type southwestern vortex mainly impacted the middle and lower reaches of the Yangtze, with the main rain-band distributing in an east-west orientation. In contrast, the major rain-band of the northeastward movement type southwestern vortex was located in the area from Huai River basin to North China, while the precipitation of the southeastward movement type southwestern vortex concentrated in south China and its coastal. In addition, the precipitation cloud characteristics of these three types of moving-out southwest vortex were distinctly different. The eastward movement type southwestern vortex affecting Jiangnan area tended to generate different precipitation types in different areas, dominated by cumulus clouds to the south of 30 °N, and by stratiform rainfall to the north. Meanwhile, the northeastward movement type southwestern vortex generated precipitation was a combination of convective and stratiform rainfall (i.e., mixed precipitation), with convective-stratiform mixed cloud precipitation. The southeastward movement type southwestern vortex precipitation was dominated by convective precipitation. The vortex precipitation cloud mainly consisted of cumulonimbus and dense cirrus clouds, with high extension height and low TBB. The eastward movement type and the northeastward movement type southwestern vortex both had wide cloud area, while the southeastward movement type southwestern vortex cloud was massive and dense.

Based on daily maximum and minimum temperature data of 99 national weather stations in Shaanxi, two methods of univariate linear regression and decaying average were used to correct the temperature forecasts of SCMOC (the data of the national meteorological center forecast), GRAPES_Meso (global/regional assimilation and prediction system) and ECMWF (European center for mediumrange weather forecasting). The result show that the prediction accuracy of daily minimum temperature was higher than that of daily maximum temperature for SCMOC, GRAPES_Meso and ECMWF. The accuracy of daily maximum and minimum temperature forecast of SCMOC was obviously highest among them, while that of GRAPES_Meso was lowest. The methods of univariate linear regression and decreasing average could significantly improve the accuracy of air temperature forecasts of GRAPES_Meso and ECMWF, but could not improve the accuracy of SCMOC. The accuracy of ECMWF’s daily maximum and minimum temperature forecast corrected from 2017 to 2019 was higher than that of SCMOC. The accuracy of GRAPES_Meso’s 24hour and 48hour daily maximum temperature prediction corrected in 2019 was higher than that of SCMOC, while the accuracy of daily minimum temperature prediction after correcting was still much lower than that of SCMOC. By comparison, the ability and stability of univariate linear regression method for rectifying air temperature forecast of numerical model were better than that of decreasing average method.

A provincial drought situation occurred in autumn 2019 in Zhejiang, especially in the central and western part of Zhejiang Province. Precipitation in autumn 2019 broke the least record in history, and the temperature anomaly was 1.5 ℃ at most of stations. Based on observation data of precipitation and temperature from 67 normal stations in Zhejiang Province and the NCEP-NCAR reanalysis data, the drought causes over Zhejiang Province in 2019 were analyzed systemically. Such severe autumn drought was directly caused by continuous less precipitation and high temperature. The main part of the polar vortex on 500 hPa was located in the Western Hemisphere, thus cold air intensity over Asia was relatively weak, furthermore, the East Asia trough position was eastward and northward, which was hard to cause the cold air to affect Zhejiang. The more intensive of western Pacific subtropical high (WPSH) and associated northward location of the ridge line and westward advance of the westward ridge point, the abnormal high pressure over South Asia, distinct downdraft over Zhejiang, the narrower-span and slower updraft of Hadley Cell and subsequent weaker northward transport of weaker vapor, stronger moisture divergence in the mid-lower troposphere worked together to bring about the typical drought climate in Zhejiang in autumn 2019.
Key words: drought; atmospheric circulation; western Pacific subtropical high; cause

 Using FY-2G satellite data, ERA5 reanalysis data, and the integrated hourly resolution precipitation data from automatic weather stations, as well as CMORPH precipitation products, this paper focused on a comparative analysis of cloud-image characteristics, circulation features, and the parcel buoyancy instability of a pseudo adiabatic process and a reversible adiabatic process in two particularly strong precipitation periods from 03:00 BST to 15:00 BST 28 (the period 1) and from 20:00 BST 28 to 08:00 BST 29 (the period 2) during a heavy rain process occurring at the boundary between Sichuan and Shaanxi from June 26 to 29, 2015. The results show that the maintenance of strong divergence on 200 hPa, a steady shear on 500 hPa, and maintenance of 850 hPa wind convergence at this boundary provided favorable conditions for the occurrence and development of heavy rain during the period 1. The heavy precipitation in the period 1 forming at the boundary between Sichuan and Shaanxi was affected by a MαCS (meso-α convective system) moving northward steered by the southwest flow in middle to upper levels. The 500 and 700 hPa shear and the maintenance of a southerly airflow in the lower layers were important factors leading to heavy rain in the period 2. The development and merging of convective cloud clusters near the shear contributed to the formation of heavy rain in the period 2 at the boundary between Sichuan and Shaanxi. The low-level water vapor content over the rainstorm center in the period 1 was higher than that in the period 2. The updraft area over the rainstorm area extended higher, the updraft was stronger, and the cloud layer in the rainstorm area was deeper and denser in the period 1 than in the period 2. In both periods, the instability of the parcel’s vertical displacement according to a reversible adiabatic process was higher than that according to a pseudo-adiabatic process over the center of the heavy rain. The high-value areas of CAPE over the rainstorm centers of both periods all distributed between 800 and 700 hPa, which indicated that the parcel lifted between 800 and 700 hPa had large buoyancy during the adiabatic process. The favorable instability condition and the water vapor condition reinforced one another in the lower layer, which was conducive to development of heavy rain.

基于粒子群优化（PSO）算法和遗传算法（GA）对支持向量机（SVM）的核函数及主要参数进行训练优化，分别建立PSO算法、GA的支持向量机模型（PSO_SVM、GA_SVM）。选用ECMWF及T639数值预报产品资料和乌东德水电站降水资料，普查最优预报因子，构建包含各种类型降水过程的训练样本和测试样本。比较分析SVM模型RBF和Sigmoid核函数优劣。尝试先分段寻找局部最优，再选择全局最优的参数优化方法。通过增大训练样本集、降低交叉验证准确率、迭代次数截断和控制惩罚系数范围的方法，提高模型的稳定性和泛化能力，防止过拟合和收敛缓慢现象。利用测试样本对SVM、PSO_SVM和GA_SVM三种方案进行对比检验，优化的GA_SVM预报效果较好且稳定。经2018年试报表明，GA_SVM逐3 h累计降水量预报TS评分在50%以上，漏报率在15%以下，与ECMWF和T639比较，该模型TS评分提高1.4%。

Based on the observation data of thunderstorm from 1998 to 2015 and the observation data from automatic weather stations at Changle Airport, daily reanalysis data from NCEP during 2010-2015, the temporal distribution characteristics of thunderstorm weather accompanying with wind shear at Changle Airport were analyzed, and the circulation types were summarized. Results show that the thunderstorm weathers occurred over Changle Airport all the year round, and it was most frequent in summer and its duration was longer. The daily change of thunderstorm weathers was obvious, it mainly occurred from the afternoon to the evening, and the duration was generally less than 2 hours, the majority of thunderstorms lasted less than an hour. The thunderstorm weathers occurred most easily in the westerly direction of Changle Airport. The probability of occurring wind shear from 3 hours before to 3 hours after the thunderstorm was 54%, and the probability of thunderstorms accompanying with wind shear in the westerly direction of Changle Airport was higher than that in other directions. The frequency of weak wind shear was much higher than that of moderate wind shear in Changle Airport, while the frequency of strong wind shear was the least. There were four main types of circulation situation for thunderstorm weathers accompanying with wind shear, they were the southern trough type, subtropical high control type, North China trough type and tropical cyclone type.

The temporal － spatial variation of reference crop evapotranspiration in northern Xinjiang was analyzed based on daily observations from 22 meteorological stations in the northern Xinjiang during 1962 － 2010 in this paper． Firstly，the reference crop evapotranspiration ( ET0)in northern Xinjiang from 1962 to 2010 was calculated with the Penman － Monteith formula，then the temporal －spatial variations of ET0 were analyzed by using the Mann － Kedall test and spatial analysis of ArcGIS． The results show that ET0 in the entire study area had an abrupt change in 1983 and presented a downward trend，which was mainly affected by relative humidity and wind speed there． The reference crop evapotranspiration gradually increased from the northeast and southwest to the center of the northern Xinjiang， and it was slightly higher in the southeast and the west region，which presented an obvious regional difference． The reference crop evapotranspiration from April to October significantly affected the annual distribution of reference evapotranspiration． The results should be useful for identifying an appropriate irrigation system that can be used in the region under environment conditions affected by global warming．

Abstract:By analysis of the climate characteristic of extreme temperature，the representative meteorological stations in Qingyang were
selected，and the correlation between the representative stations and other stations was analyzed．At first，the highest and lowest tem-
perature prediction model of the representative stations was established by using the method of the support vector machines(CMSVM)
and ECMWF product，and based on that the linear regression statistics relationship of the representative stations and other stations was
established．The forecast test indicated that the accuracy rate of highest and lowest temperature forecast was improved using this method
compared with that of the numerical forecasting release at every station in 2009，and forecast of highest temperature was good in effect
and the mean accuracy rate has enhanced 5%．So the model could be applied in the day－to－day business directly．

A contrast analysis of two heavy rainstorm occurred in Liaocheng of Shandong Province in different seasons was made from synoptic situation，physical quantity field and radar echoes characteristic．The results show that the summer rainstorm usually accompanied by strong convective weather，mid－cyclone and adverse wind areas．When the heavy rainstorm occurred in autumn，the intensity of radar echoes was not very strong but lasted longer，and some physical quantity indexes related to heavy rainstorm were lower in autumn than those of in  auummer．The forecast of heavy rainstorm in autumn is difficult．

By using the objective similaritymethod, A short - range forecastmodel of hail in Gannan Plateau was comp leted. The system comp iled by four parts, including a history database of hail, similar case and judgment, viewing history weathermap and comparing with intraday actually happening, adding new hail data. The result shows that the model has a good effect in short - range forecast of hail, and it is an effective method of short - range forecast.

By using the data of eightweather stations’records of thunderstorms over Gannan p lateau from the beginning of each station’s establishment to 2005, the temporal and spatial distribution features of thunderstormswere analyzed. The results show that thunderstorm occurred more frequently inMaqu, Luqu and Hezuo counties; The annual thunderstorm days p resented descending trend and the trend ismore obvious after the middle of the 1990 s, but the variation of the duration days between the beginning and the ending time of thunderstorm occurring in Gannan p lateau is not very distinct; Summer is thunderstorm season and July is the month in which thundstorms occurmost frequently, and the percentage of thundstorm times in July is 20. 2%. The late sp ring and early summer, the third ten - day of June, the second and third ten - day of July, the third ten - day ofAugust are the peak periods for thunderstorms occurring in Gannan; The daily thunderstorm occurring peak time is 14: 00 - 15: 00, and the percentage of thundstorm times is 30. 3% in this period.

An abrup t severe storm occurred on August 30 - 31 of 2006 in the Gannan Plateau was analyzed. Results show that the combination of convective clouds from the southeast of Qinghai and the local convective cloud ismain cause resulting in this p rocess, and the severe storm and hail occurred on the front of right side of the convective cloud. The development of airflow leaning east at the height of 600 - 700 hPa over the eastern part of Gannan p lateau is very important to the severe storm’s occurring and develop ing. The distributions ofθse high energy center at 700 hPa level,θse departure between 500 hPa and 700 hPa, the helicity and vapor flux divergency fields at 700 hPa level are meaningful to estimate the occurring region of this severe storm.

The variation characteritics of frost times and period in Ningxia region from 1961 to 2004 were analyzed. The results show that frost occurring in Ningxia wasmainly in Ap ril and October, and the frost times wasmost in the second ten - day of Ap ril; in the south and north of Ningxia region, the frost times was different significantly, and frost occurred frequently in mountainous region of South Ningxia; the annualmean frost times there was 8. 7 - 10. 8 times, and frost times was least in Tongxin for only 2. 1 times. The frost times p resented decreasing trend as a whole and itwas obvious in June and autumn. The abrup t of frost times occurred in 1984,and after 1984 the frost times reduced significantly. The first frost date was delayed and the last frost date advanced, as well as the frostless period was extended gradually in the three areas ofNingxia region.