Abstract:The development of high-precision,high-quality,and spatiotemporally continuous multi-source meteorological data fusion grid-based products using data assimilation and data fusion techniques as well as multi-source observation data has become a key focus of meteorological data research.Based on the LAPS model for multi-source data fusion,this paper establishes a provincial analysis system for Henan Province.By integrating ground,sounding,and radar data,grid-based observation products for Henan Province is produced,including precipitation fusion analysis products,three-dimensional grid-based observation products(temperature,wind,relative humidity),and refined grid-based soil moisture data products.Systematic evaluation is conducted on the grid-based observation products obtained from the national and provincial observatories,and the observation data at any time and in any region is evalu-ated based on conventional evaluation,characteristic evaluation,and refined evaluation schemes,which provides more accurate and refined evaluation conclusions.Based on the B/S architecture and adopting the technological route of service-oriented architecture,a multi-source meteorological data fusion system platform for Henan Province has been built,achieving the display and evaluation of national and provin-cial grid-based observation products and fully leveraging the application value of gridded observation and fusion data products.

Abstract:Using the hourly observation data of PM2.5 in Huanggang City from 2015 to 2020,the changes in PM2.5 concentration are analyzed.Combined with the data of NCEP global data assimilation system and MeteoInfo software,the transport paths and characteristics of PM2.5 are simulated by the meth-od of trajectory clustering analysis.In addition,the potential source contribution function(PSCF)and the concentration-weighted trajectory(CWT)analysis method are used to study the potential source areas of PM2.5 in Huanggang City.The results show that the overall PM2.5 concentration in Huanggang City has been at a relatively low level from 2015 to 2020,showing a trend of decreasing year by year,and the pro-portion of PM2.5 as the primary pollutant has decreased significantly.Thus,the air quality in Huanggang City is generally better.Combined with wind rose diagram and backward trajectory cluster analysis,Huanggang City is found to have three transport paths of air mass,respectively in the directions of north-west,northeast,and south.Among them,the southerly short-distance air mass trajectory accounts for the highest proportion,while the northwest long-distance air mass trajectory leads to the highest PM2.5 pollu-tion concentration.At the same time,the northern path can transport pollutants farther than the southern path,and the transport effect is significantly greater than that of the local accumulation on pollutant con-centration.The annual PSCF and CWT analysis results show that the main PM2.5 potential source areas and concentration-weighted areas in Huanggang City are relatively consistent,generally in two directions:one is the northwestern Hubei,Henan,Shanxi and Shaanxi in the northwest or north direction,and the other is the southeastern Hubei,Jiangxi and Hunan in the southeast direction.The PM2.5 potential sources in the northwest or north direction are more significant.The analysis results of potential sources also indicate that in recent years,the concentration of PM2.5 in Huanggang City has been decreasing,the range of potential source areas has been significantly shrinking,and the concentration weight has also de-creased,except for a slight rebound in 2019.

Abstract:Recently,the domestically developed open-source large language model DeepSeek has at-tracted significant global attention,with rapid adoption across vertical sectors——including telecommuni-cations,finance,healthcare,environmental protection,education,and public administration——driven by its technical capabilities and open-source approach.Within meteorological departments,DeepSeek de-ployment solidifies the foundation for self-reliant meteorological science while advancing operational sys-tem intelligence,facilitates precise supply-demand alignment for smart meteorological services,provides novel management tools,and drives industry-scale growth and efficiency gains.Concurrently,it poses challenges to specialized meteorological applications,governance frameworks,and risk management.Therefore,when deploying and applying DeepSeek,it is necessary to strengthen policy guidance and in-stitutional mechanisms,establish an independent,controllable,resilient,and scalable technical support system for AI meteorological applications.In light of the technical features of DeepSeek and the service demands of meteorological services,it should be promoted and applied in an orderly manner.We should flexibly apply various policy tools to strengthen the governance of meteorological applications using large AI models.Drawing on the successful experiences of the Hangzhou Municipal People's Government and DeepSeek Company,a complete meteorological science and technology innovation ecosystem should be created.

Abstract:Based on the ozone(O3)monitoring data and relevant meteorological element data in Kun-ming City from 2017 to 2021,the variation characteristics of ozone concentration and main meteorological influence factors in Kunming in recent five years are analyzed,and the weather system during non-optimal(below the"optimal"level)ozone days is subjectively classified.The results show that,in recent five years,the non-optimal days of ozone in Kunming have shown an increasing trend,and the average ozone concentration and the number of non-optimal ozone days in dry season are higher than those in rainy sea-son.The O3 non-optimal peak period of ozone in the year is from March to April(spring),and the second non-optimal peak of O3 in the year appears in August(summer).Ozone concentration is positively correla-ted to sunshine hours and daily maximum temperature,but negatively to relative humidity.In other words,when it is high temperature,low humidity and strong sunshine,ozone concentration is prone to rise.The weather patterns that cause the non-optimal days of ozone in Kunming at 500 hPa geopotential height mainly include six types,i.e.,the flat westerly type(X,accounting for 35.0%of the total,which is the most),the southwest airflow type in front of the dry south branch trough(N,which is the second most),the northwest airflow type behind the trough(B),the westerly airflow type outside the subtropical high(G),the easterly wind type outside the South China Sea low pressure(D)and the Qinghai-Tibet Plateau high pressure bottom type(Q,accounting for at least 5.7%of the total,but the ozone concentration during the period is the highest).In dry season,the weather types are mainly X,N,B,with westerly wind mainly blowing,and the weather is dry,warm and sunny.However,in rainy season,the weather types are mainly D and Q,and the weather is humid,hot and sunny,accompanied by easterly wind.

Abstract:Based on air quality data from 2017 to 2021 for Zaozhuang,Shandong Province,this article statistically analyzes the interannual and seasonal variation characteristics of the Air Quality Index(AQI)and primary pollutants.The HYSPLIT trajectory model is used to simulate the backward trajectories hav-ing arrived in Zaozhuang,and cluster analysis method is employed to evaluate the effects of different trans-port pathways on AQI,PM10,PM2.5 and O3 concentrations.Potential Source Contribution Factor(PSCF)and Concentration Weighted Trajectory(CWT)analyses are used to identify the potential source regions contributing to air pollution in Zaozhuang in each season.The results show that air quality in Zaozhuang is mainly categorized as excellent,good or slightly polluted,with excellent and good air quality days account-ing for 57%of the total during the five-year period.O3 and PM2.5 are the primary pollutants with the high-est proportions in summer and winter,accounting for 88%and 73%of the days with primary pollutants,respectively,while PM10 is the main pollutant in spring and autumn.The backward trajectories to Zaozhuang can reflect the monsoon characteristics of the region:in spring,autumn and winter,the trajecto-ries having arrived in Zaozhuang mainly originate from the northwest and northeast directions,while in summer they primarily come from southeast and southwest.Among them,the AQI and pollutant concentra-tions associated with the third and fifth trajectory clusters,which originate from the ocean,are lower than those of the other three clusters.The area of high-value potential source regions affecting AQI in Zaozhuang during winter is significantly larger than in spring,summer and autumn.The potential source regions affect-ing AQI in each season include most of Shandong Province except Qingdao,Yantai and Weihai,as well as eastern Henan,northern Anhui,and Jiangsu.In spring,areas with WCWT values exceeding 70 μg/m3 are concentrated in Taian,Jining,Shangqiu,Huaibei,Suzhou and Xuzhou;in summer,the areas include Jining,Huaibei,Xuzhou and Bengbu;in autumn,the areas are concentrated in Taian,Jining,Heze and Xuzhou;and in winter,these areas cover Ji'nan,Jining,Huaibei,Xuzhou,Suqian and Linyi.

Abstract:Two correction methods,moving average and Kalman filtering,are used to test and cor-rect the wind speed forecast products for Heilongjiang expressways.The results show that the forecast value of maximum wind speed with each forecast lead time is close to the observed wind speed,but gen-erally has a lower value.The revised wind speed value can reduce the systematic error of the model prod-ucts,and generally presents a positive correction effect.The forecast errors(mean error,mean absolute error and root mean square error)decrease at each forecast lead time.Among them,the effect of Kalman filtering method is better than that of the moving average method,and the effect of maximum wind speed correction is better than that of the corrected minimum wind speed.From the perspective of geographical distribution,the reduction range of root mean square error at most sites is 0.5-1.0 m/s.The correction effect in daytime is better than that at night,and the daytime forecast test score is more stable with de-creasing fluctuation after the correction.Further analysis shows that there are significant differences in the correction effects for different expressway sections.The correction effect in the plain area is better than that in the mountainous area,because the complex terrain in the mountainous area leads to large varia-tions in wind speed and the adaptability of the correction algorithm is poor.With the increase of actual wind speed,the mean error of forecast before and after correction decreases first and then increases,and the wind speed forecast error of the mountainous expressway sections increases more obviously with the in-crease of the actual wind speed.

Abstract:Based on the sounding data of Hubei Province from 2017 to 2021 and combined with the air quality monitoring data,four types of multi-layer inversion characteristics and their influence on the concentrations of primary air pollutants in Hubei Province are analyzed.It has been found that:(1)Multi-layer inversion occurs frequently in Hubei Province with frequency reaching 71.12％,which is much higher than that of single-layer inversion.(2)The inversion frequency and thickness at 08:00 are greater than at 20:00,and the 20:00 inversion intensity is greater than that at 08:00.(3)The monthly variation curves of total frequency,thickness and intensity all present a 'V' shape.The decreasing rate from January to July is smaller than the increasing rate from August to December.(4)The type Ⅳ inversion(SI+EI+LTI)has the most obvious effect on the growth of particulate matter,while the inhibitory effect of type Ⅲ inversion(EI+LTI)on O3 is the most obvious.(5)The particle concentration is positively correlated to the thickness and intensity,and the correlation at 08:00 is more obvious.O3 is negatively correlated to the thickness and intensity of inversion layer,and the correlation at 20:00 is more obvious.

Abstract:Improving the accuracy of short-term runoff forecast model is important for the decision-making of reservoir operation.Based on the meteorological and hydrological observation data of Zeya Res-ervoir watershed,four integrated learning models of random forest,GBDT,AdaBoost,and Stacking are used to simulate daily runoff forecasting in small watersheds,and the influence of meteorological factors such as temperature,pressure,and humidity on runoff prediction and simulation is comparatively ana-lyzed.The results show that:(1)The random forest,GBDT,and Stacking models based on multiple meteorological elements have good simulation and prediction capabilities for short-term inflow runoff in small watersheds.The compliance rates of these models during the verification period exceed 70％,meet-ing the accuracy requirements for issuing formal forecasts.Among them,GBDT is the best in terms of prediction and measurement and random forest is the best in terms of accuracy.Therefore,they can be comprehensively applied for decision-making in actual operations.(2)The introduction of meteorological elements(temperature,pressure,and humidity)can significantly improve the accuracy of short-term runoff prediction by each integrated learning model.In particular,the compliance rate of base and medi-um flow prediction by random forest,GBDT,and Stacking models can increase from about 40％to more than 80％,the efficiency coefficient increase to about 0.85,while the average absolute error can decrease by about 50％and the average relative error decrease by about 60％.These findings have high reference value for runoff prediction and simulation during non-flood periods.

Abstract:Using the daily meteorological data of flue-cured tobacco field period from 1981 to 2022 at 12 meteorological stations in western Henan Province and taking the Standardized Antecedent Precipi-tation Evapotranspiration Index(SAPEI)as the drought index of flue-cured tobacco,this paper analyzes the temporal and spatial distribution characteristics of drought in the flue-cured tobacco field period in western Henan and divides different grades of drought climate risks.Moreover,based on the flue-cured tobacco yield per unit area at two representative stations from 1981 to 2022,the impact of drought in dif-ferent periods on meteorological yield of flue-cured tobacco is analyzed.The results show that the occur-rence frequencies of light,moderate,and severe droughts in the root elongation period of flue-cured to-bacco in western Henan are 20％,16％,and 6％,respectively.The occurrence frequencies of light,moderate and severe droughts in its vigorous growth period are 13％,12％,and 11％,respectively.The occurrence frequencies of light,moderate,and severe droughts in its maturing period are 12％,7％,and 6％.respectively.Spatially,the drought frequency shows a distribution characteristic of high in the east and low in the west in the root elongation period,high in the west and low in the east in the vigorous growth period and maturing period.The drought climate risk indices are 44.4,44.1,and 27.8 respec-tively in the root elongation period,vigorous growth period and maturing period.In terms of inter-annual variation,the root elongation period and maturing period show an upward trend at a rate of 7.5/decade and 4.8/decade,respectively,but the vigorous growth period has a downward trend at a rate of-1.1/decade.The study area is divided into different levels of drought climate risk areas according to the drought climate risk indices.In the root elongation period,Xin'an and Mengjin are the primary high-risk areas,while Mianchi,south of Yiyang,Songxian and Ruyang are the sub-high-risk areas,with other places as medium-risk areas basically.In the vigorous growth period,Lingbao is the high-risk area,while Shaanxian,Ruyang and north of Lushi,northwest of Luoning are the sub-high-risk areas,and other places belong to medium-risk areas.In the maturing period,all these places become low-risk and sub-low-risk areas.SAPEI has a significantly positive correlation with meteorological yield of flue-cured to-bacco in mid-May and from late June to early July(P＜0.05),which means that drought condition in these two periods has the most significant impact on the flue-cured tobacco meteorological yield.

Abstract:Based on the hail data of 119 observation stations in Inner Mongolia Autonomous Region from 1959 to 2020 and the ERA5 reanalysis data,the objective hail sample labeling procedure is pro-posed.Hail recognition model is constructed by using Support Vector Machine(SVM),K Nearest Neigh-bor(KNN)algorithm,Logistic Regression Model and ensemble model for training,testing and optimiza-tion,and the guidance ability of the models is evaluated.The research results show that the models con-structed by SVM and KNN are more accurate in hail prediction.The probability of detection(POD)of hail based on independent test samples is more than 82%,and the TS score surpasses 0.70.The ensem-ble model constructed based on the fusion of the three models has a score of 0.74,and is capable of provi-ding better hail prediction.The comparison of the performance of the three optimized models and the en-semble model reveals that the SVM and KNN have higher recognition accuracy in identifying hail and non-hail,while the Logistic Regression Model has the lowest recognition accuracy.The ensemble model has a higher POD and a higher false alarm rate(FAR)than KNN algorithm,so it is more capable of giv-ing correct recognition results for hail samples.For the prediction of 136 hail processes with specific time records in 2021 and 2023,KNN has the highest POD and the Logistic Regression Model has the poorest prediction effect.

Abstract:Wildfire exerts great influences on the ecosystem and climate change.Analyzing its inter-annual variability and mechanisms could deepen the understanding of wildfire variability and provide sci-entific basis for fire prevention decision-making in China.Using the burned area data from GFED4 and MODIS satellite remote sensing and ERA5 re-analysis data,this study explores the inter-annual variability of wildfire burned area by means of power spectrum analysis.Then,the key meteorological factors,which dominate the inter-annual variation of wildfire area,and their circulation characteristics are identified by the random forest model and correlation/regression analysis.The results show that:(1)South China(January to March),Southwest China(February to April)and Northeast China(March to May)are the key wildfire regions with significant inter-annual variations in wildfire burned area in China and the inter-annual cycle of each significant area is different.There is an inter-annual cycle of 4 to 6 years for the wildfire burned area in South China and a 3-year cycle in Northeast China.(2)Relative humidity is a key meteorological factor for the inter-annual variation of wildfires in South China and Northeast China,and it shows a significant negative correlation with the area of wildfires.Wind speed is a key meteorological fac-tor in Southwest China and has a significant positive correlation with the area of wildfires.(3)Distinct at-mospheric circulations exist in different key regions.The westerly(southwesterly)wind on the south side of the abnormal low sea level pressure in Southwest China is positively abnormal,which increases the cli-matological wind speed and is conducive to the spread of fire.Local high-pressure anomaly appears in Northeast China with descending motion,and the sinking airflow reduces the near-surface relative humidity,benefiting the increase in wildfire burned area.However,no significant circulation anomaly is found in South China.

Abstract:Based on the temperature and precipitation data and climate annual index from the Na-tional Meteorological Observatory of Sichuan Province from 1990 to 2022,the variation characteristics of the climate annual index in Sichuan Province are studied by the methods such as percentile method,rota-tional empirical orthogonal function,and linear trend equation.The results show that:(1)There is a sig-nificant difference in the average state of climate annual index in various regions of Sichuan Province.Aba Prefecture and most of the basin have higher values of climate annual index,with the central part of the basin being the center of the high values,while Ganzi Prefecture and Panxi(Panzhihua and Xichang)region have relatively lower values.(2)The climate annual index of Sichuan Province has shown a significant increasing trend in the past 33 years,with a clear deviation trend in the annual evalu-ation level.The preferred annual sceneries all appeared before 2010,and the deviation annual scenes mostly appeared in the past 20 years.The climate annual index in the western part of the basin and the northern part of the western Sichuan Plateau shows a significant increasing trend,and the annual climate may become worse and worse.(3)The regional characteristics of the climate annual index changes in Si-chuan Province are mainly divided into southern basin type,western basin type,central basin type,and Panxi type.The climate annual index of the western basin and Panxi regions has shown a significant in-creasing trend,with frequent deviations since the beginning of the 21st century;the southern and central parts of the basin exhibit weak growth or inconspicuous changes.(4)The quantitative evaluation results of the annual climate of Sichuan Province are consistent with the observation evaluation.The climate an-nual index has a good representational ability for climate abnormal events,but it is negatively correlated to the magnitude of disaster loss rate and shows an opposite trend of change,which needs to be consid-ered comprehensively in actual operations.

Abstract:This article evaluates the detection performance of the Ganzhou SC Doppler weather radar in Jiangxi Province before and after upgrading to a dual polarization radar,and,based on previous re-search,proposes a clear and efficient algorithm for evaluating the reflectivity consistency of the two ra-dars.The results show that after completing the upgrade of the Ganzhou radar,the range resolution is in-creased from 1000 m to 250 m,and the minimum measurable echo intensity detected at 50 km is in-creased from-7 dBZ to-10 dBZ,indicating its capability of detecting in clear sky.After the Ganzhou radar is updated,it is found that under both light and heavy precipitation conditions,the reflectivity changes detected by the radars in Ganzhou,Ji'an,and Shaoguan are basically consistent,but the reflec-tivity values can be influenced by distance and mountain ranges.After eliminating the influence of mountain ranges,in the overlapping observation area,the correlation coefficient of reflectivity factors between Ganzhou and Ji'an radars is 0.93,and the correlation coefficient of Ganzhou radar with that of Shaoguan radar is 0.96.

Abstract:Northeast Guangxi suffered a warm-sector rainstorm process during 3-4 June 2022,fea-tured with intensive rainfall and large accumulated rainfall amount,causing flooding and landslide and other secondary disasters.To enrich the understanding of the mechanism of warm-sector rainstorm of Guangxi and provide reference for forecasting services,the dynamic and thermal processes of the warm-sector rainstorm are diagnosed and analyzed based on multi-source data as well as the kinetic energy equation and stratified stability tendency equation.The results show that the divergence in the east side of the South Asia high,the maintenance of mid-level trough and low-level vortex provided a favorable back-ground for the torrential rainfall.The strong warm-humid southwest monsoon provided sufficient water va-por for the occurrence and development of the torrential rainfall.The quasi-stationary meso-α-scale con-vective system and train effect caused the local accumulation of heavy precipitation,which was mainly lo-cated in the large TBB gradient region.The positive vorticity zone on the left of 850 hPa jet coincided with the exit zone of 925 hPa jet and the mountain terrain barrier provided dynamic conditions for the de-velopment of the severe precipitation.The increase of local kinetic energy caused by the decrease of geo-potential height from south to north along with southerly wind was the main cause for the development of low-level jet at night.On the one hand,the development of low-level jet enhanced the low-level conver-gence and promoted the development of ascending motion.On the other hand,the pattern of low-level jet with southerly flow and the θse isoline formed local frontogenesis forcing,conducive to triggering and strengthening the maintenance of torrential rain.The whole process was accompanied by significant low-level convective instability,and the unstable stratification in the early stage of precipitation was jointly maintained by horizontal advection term and potential divergence term.In the heaviest rainfall stage,the strong warm-humid air transport and the upward motion made the convective instability over 900-700 hPa develop continuously,and the potential divergence term promoted the instability near the 925 hPa bound-ary layer.

Abstract:Using a variety of data,the characteristics of cloud-to-ground(CG)lightning in various stages of 10 typical squall lines in the Beijing-Tianjin-Hebei Region from 2007 to 2013 and their relation-ships with disastrous weather was analyzed.The results showed that the CG lightning was mainly distribu-ted in the strong echoes of 35-55 dBZ at the front of the squall line.The temporal evolution of CG light-ning showed that the frequency evolution of positive CG lightning best reflected the generation-decline and strength changes of squall lines.In the cases where positive CG lightning was dominant,the difference in the number of positive and negative CG lightning was small.In the cases where negative CG lightning was dominant,the difference in the number of positive and negative CG lightning was large.In the dominant cases of negative CG lightning,the squall line had a larger average size and a longer duration,at the same time,the squall line produced more CG lightning,strong winds and short-term strong precipitation stations,and the rain intensity was also higher.In the cases of squall line where either positive or nega-tive CG lightning was dominant,the high occurrence of positive CG lightning indicated or accompanied frequent occurrence of disastrous weather.Considering the squall line system as a whole,the temporal evolution of CG lightning showed that the frequency evolution of positive CG best reflected the generation-decline and high frequency of strong winds and short-term strong precipitation,as well as the intensity of rain.The rising frequency of positive CG lightning was indicative of the formation and strengthening of squall lines,as well as the occurrence and frequency of disastrous weather.

Abstract:Based on the products of boundary layer wind profile radar in Pingdingshan,Henan Prov-ince,the variation characteristics of the radar data during the pollution episode from December 17 to 28,2019 were analyzed.Using correlation analysis and regression analysis,typical physical parameters were identified as indicators for pollution monitoring and early warning.The results show that:(1)Pollutant concentrations increased rapidly under conditions of low horizontal wind speed and prevailing downdrafts.The persistence of pollution was facilitated by weak local circulation and low boundary layer height,which hindered pollutant dispersion.(2)Horizontal wind speed,vertical wind shear,boundary layer ventilation,and atmospheric stability variation exhibited significant temporal and spatial correlations with the Air Quality Index(AQI).Temporally,the strongest correlation was observed when these parameters led the pollution by 4 hours.Spatially,boundary layer ventilation flux was negatively correlated with AQI;within the boundary layer,lower-level variables including horizontal wind speed,vertical wind shear,and atmospheric stability variations were negatively correlated with AQI,while the correlations were positive in the upper levels above the boundary layer.(3)Regression analysis and validation dem-onstrated that the wind profile radar variables at 4 hours ahead of pollution occurrence provided the best prediction for AQI.Among these,horizontal wind speed,boundary layer ventilation,atmospheric stability variation,and vertical wind shear could serve as effective indicators for pollution monitoring and early warning.

Abstract:Two dust weather episodes are studied in this paper based on hourly air quality data,LI-DAR monitoring data from Anyang National Climate Observatory,ERA5 reanalysis data,GDAS meteoro-logical data,and wind profiler radar data from Zhengzhou.The results indicate that the intensity of epi-sode 1(March 19-24,2023)is stronger than that of episode 2(April 9-13,2023).The concentra-tion of PM10 shows the characteristics of"rapid increase to rapid decline",and the time of serious air quality pollution is 14-28 hours in episode 1.The duration of serious air quality pollution in episode 2 is 29-58 hours,showing a bimodal characteristic.The two dust weather episodes are formed under the combined action of the Mongolian Cyclone and its rear cold high pressure.Compared with episode 1,the Mongolian cyclone and its rear cold and high pressure are weaker and the location of the Mongolian cy-clone is also slightly easterly in episode 2.The southeast air flow behind the maritime high pressure cau-ses the dust backflow southward to affect Henan twice in episode 2,which is the reason for the serious air pollution for a long time.In episode 1,the dust particles that caused the spike in PM10 concentration in Zhengzhou are mainly transported by the northeast path.In episodes 2,the dust particles that greatly af-fected the air quality in Zhengzhou are the northwest path and the southeast path.According to the com-prehensive analysis of the vertical evolution of Anyang LIDAR and the upstream topography,the dust par-ticles transported from the northwest path will be blocked by the Taihang Mountains or bypass the Taihang Mountains,enter from the lower terrain,resulting in the air quality of Xinxiang to reach serious pollution 5 hours later than that of Zhengzhou in episodes 2.There are high-altitude transmission of dust particles around 3600 m and 4200-4800 m over Anyang and the apparent increase of wind speed at different alti-tudes before two episodes,which indicates that the new observation data is essential to the prediction and early warning of dust weather in Henan.

Abstract:From December 28,2023 to January 2,2024,Zhejiang province experienced a moderate haze pollution process,which was the strongest and longest lasting haze process since 2018.Using the air quality data and meteorological data of the same period,the weather patterns,air mass transport charac-teristics and meteorological causes of this haze process were studied.The results showed that surface me-teorological conditions such as warming,humidifying,continuous low wind speed(<2 m/s),and no precipitation,as well as long duration and strong vertical inversion(inversion intensity reaching the peak value of 2.28℃/100m),and static weather conditions controlled by uniform pressure field or high pres-sure bottom,resulted in adverse meteorological conditions that led to local pollution accumulation.The contribution of local emissions in Zhejiang province reached as high as 47%,which was the main cause of this haze process.Owing to the implementation of the Action Plan for Air Pollution Prevention and Con-trol,the number of days with moderate or more severe pollution events in Zhejiang has significantly de-creased,but adverse meteorological conditions might still cause local particulate matter pollution,leading to prolonged haze weather processes in Zhejiang.

Abstract:According to the precipitation and loss data in Hebei Province from 1984 to 2020,307 regional rainfall events were accessed by constructing two indices.The hazard intensity index was con-structed with regional and short-term precipitation intensity.The disaster degree index was constructed based on the direct economic losses,dead-missing population,and affected population.Using the pro-posed catastrophe identification method,the"23·7"rainstorm event was determined as a catastrophe.Catastrophe characteristics were analyzed in this paper.The results are as follows.According to the super strong hazard intensity and super serious disaster degree,three catastrophic rainfall processes were deter-mined,namely"96·8","16·7"and July 11-13,1994.Based on the precipitation and disaster loss data of the"23·7"rainstorm process,the hazard intensity index was 0.99(super strong)and the dis-aster degree index was 0.164(super serious),both of which reached the highest degree.The catastrophe processes showed wide impact range,high regional and short-term intensity,and long duration.Mean-while,it might cause serious casualties and economic loss.Four types of secondary disasters and eight types of impact events would occur during catastrophe process.The higher the hazard intensity grades and disaster degrees,the probability of the secondary disasters and affected industries was higher.The occur-rence probability of secondary disasters from large to small showed that water logging>flood>mountain flood>geological disasters,and the impact events showed agricultural facilities>houses>road-bridges.

Abstract:Compared to hourly GPS PWV(GPS_PWV)retrieved from multi-station network,the single-station PWV retrieval algorithm based on BDS-3 Precise Point Positioning(PPP)enables real-time retrieval of BD_PWV with temporal resolutions of 5 min and 1 h at single station.Chosing sounding PWV(RS_PWV)as a reference to evaluate the accuracy of hourly BD_PWV from January to March 2022.Furthermore,a qualitative and quantitative analysis is conducted to compare the variation trends,increments and increase rates of hourly and 5 min BD_PWV with those of GPS_PWV during a rainstorm process in Haikou from February 16th-18th.The results show that,the bias between BD_PWV and RS_PWV is quite small and the RMSE is less than 3 mm,indicating the high accuracy of BD_PWV that is comparable to that of GPS_PWV.PWV usually rises significantly before a rainfall and decreases after the rainfall,and most of the maximum rainfall occurs near the PWV peak.Compared to GPS_PWV,the CSI scores of hourly BD_PWV under different precipitation grades and thresholds are usually higher and it increases with precipitation intensity,the max CSI score of heavy precipitation can improve by 20%.In the rainstorm case,5 min BD_PWV features a higher peak value,an earlier occurrence time ahead of precipitation and a larger increase rate,which can exceed 10 mm/h at most.