Abstract:As an important pastoral area in northwest Sichuan,the grassland on the Zoige Plateau has been affected by natural conditions and human activities,resulting in serious desertification in recent years.In order to understand the spatial distribution and dynamic variation of grassland desertification in detail,based on the Landsat images,the deep learning combined with visual interpretation methods was used to extract desertification land from 1991 to 2020 in grassland on the Zoige P lateau and analyze their spatiotemporal variation characteristics and influencing factors.The results showed that:(1)The area of grassland desertification was the largest in Zoige County,followed by Maqu County,while desertification area of Hongyuan County,Luqu County and Aba County was quite small.(2)The sandy land area in the study area increased first and then decreased.(3)The spatial variation characteristics of desertification land in the Zoige Plateau were mainly"development"in the north and"reversal"in the south.(4)The desertification of Zoige grassland was affected by both natural and human factors.The desertification area decreased with the increase of temperature and precipitation,and decreased with the decrease of actual livestock carrying capacity.

Abstract:Focusing on the damage to power facilities in Xiaogan City,Hubei Province,caused by the rainstorm on August 12,2021,this study analyzed and quantitatively identified the risk sources of rainstorm and flood disasters for these facilities from the perspectives of hazard factors,disaster-predispo-sing environment,vulnerable entities,and disaster situations,employing methods such as case analysis,WBS-RBS(Work Breakdown Structure-Risk Breakdown Structure),weighted comprehensive evaluation,and scenario simulations using an inundation model.The results are as follows.(1)The flood disasters suffered by Xiaogan power facilities are mainly caused by the combined effects of multiple disaster types such as floods in the upper reaches of the Fuhuan River,local heavy precipitation,and downstream water level uplift.The continuous high water level in the Xiaogan section hinders the drainage of waterlogging in the area,causing slow water withdrawal.(2)Identification of rainstorm hazards can directly indicate that Xiaogan power facilities are in the high-risk area of rainstorm and flood during this heavy rainfall process,which has reference significance in the layout planning of power facilities stations,though further refinement is needed.(3)Submersion simulation can finely and dynamically express the flooding time and water depth evolution of Xiaogan power facilities.Based on the accuracy testing and analysis of the model,the simulation results show that this precipitation process causes a maximum depth of 0.4-2.0 meters of water accumulation near the Yuncheng substation and the Xiaogan urban shutoff area.The water accumulation mainly occurs between 06:00 and 08:00 on August 12.

Abstract:The distribution of 32 global atmospheric background stations of WMO/GAW,the instru-ments and technical methods for on-line observation of CO2 mole fraction were investigated,and the char-acteristics of CO2 mole fraction observation data of 23 stations with more than 10 years were analyzed.The results are as follows.(1)From the distribution of stations,they are mainly distributed in Asia and Eu-rope,and most of them are near the sea.In terms of observation methods,China's global or regional background stations,like other global atmospheric background stations,all adopt observation techniques and methods that meet the requirements of WMO/GAW quality objectives.(2)The CO2 mole fraction of stations in the middle and high latitudes in the northern hemisphere fluctuates greatly.As far as the long-term trend is concerned,the CO2 mole fraction of stations in the southern and northern hemispheres shows an overall increasing trend,but the annual mean increment of CO2 mole fraction in the northern hemi-sphere is overall higher than that in the southern hemisphere.After removing the long-term trend,the seasonal maximum value of CO2 mole fraction in low latitudes in the northern hemisphere lags behind that in middle and high latitudes.The seasonal variation of CO2 mole fraction in middle and high latitudes in the southern hemisphere is opposite to that in low latitudes and northern latitudes,and the CO2 mole frac-tion in WLG is equivalent to that in middle and high latitudes in the northern hemisphere.(3)The aver-aged monthly amplitudes of stations above 2000 m and stations below 2000 m in recent five years are 18.1×10-6 mol/mol and 18.6×10-6 mol/mol,respectively,while those of ocean stations and land sta-tions are 12.5×10-6 mol/mol and 19.5×10-6 mol/mol,respectively.In comparison,stations above 2000 m and ocean stations exhibit smaller amplitudes.(4)The daily amplitude of seasonal CO2 mean mole fraction at high latitude representative stations in the southern and northern hemispheres is small,and the maximum is only 0.8×10-6 mol/mol,while that at middle latitude representative stations is 3.7×10-6 mol/mol,and that at low latitude representative stations is 1.5×10-6 mol/mol,and the maximum ampli-tude occurs in summer in the northern hemisphere.

Abstract:Based on the daily meteorological data of 113 stations in Henan Province during 1971-2020 and the daily soil relative humidity data at 10-50 cm depth from 86 automatic soil moisture observa-tion stations during 2011-2020,the percentage of precipitation anomaly(Pa),Z index,K index and ac-cumulated humidity index(Ma)were selected,and the correlation coefficients between soil relative hu-midity in each layer and four water indexes were compared and analyzed,and the waterlogging of summer maize in Henan Province was determined based on the accumulated wetness index.Furthermore,the spa-tial distribution of waterlogging risk was evaluated by integrating the average summer maize planting area proportion in 108 major grain-producing counties(cities)over the past five years and the accumulated moisture index.The results are as follows:(1)The correlation coefficient between Ma and soil relative humidity is higher than the other three water indexes,with the best correlation coefficient between Ma and soil relative humidity of 20 cm.According to the quantitative relationship between Ma and soil relative hu-midity,it is determined that 0.5<Ma≤1.3 is light waterlogging,1.3<Ma≤3.0 is medium waterlogging,and Ma>3.0 is heavy waterlogging.(2)The waterlogging risk of summer maize in Henan Province mainly appeared in jointing-tasseling stage,followed by the emergence-jointing stage;High-risk areas are mainly distributed in central and eastern Nanyang,Zhumadian,Zhoukou and Shangqiu,with a comprehensive risk index from 10 to 20;In addition,regions such as Xinxiang,Hebi,and western Anyang in the eastern Taihang Mountains exhibit relatively high waterlogging risk during the jointing-tasseling stage.(3)The risk of waterlogging in tasseling to milking stage and milking to ripening stage is generally low,and the risk index in most areas is between 5 and 10.

Abstract:Using millimeter wave cloud radar data from Shangqiu Station of the East Henan Plain from April 2022 to December 2023,the frequency and vertical structure characteristics of clouds in this area are analyzed.The results show that the proportion of clear sky is 60.2％and the frequency of non-precipitation cloud is 36.3％throughout the year.For non-precipitation clouds,the vertical structure analysis suggests that the single-layer cloud has the highest occurrence frequency,with a monthly average frequency of 60.1％.The cloud base height analysis reveals that the medium and high clouds appear more frequently,with average monthly frequency of 40.7％and 34.3％,respectively.The macro param-eters of non-precipitation clouds have significant variations in different seasons.The cloud base height is most frequently seen at the 6-7 km height in spring and summer,while in autumn and winter it appears most frequently at the height of 5-6 km.In terms of cloud top height,it often reaches the highest height at 9-10 km in spring,8-9 km in autumn,and 7-8 km in winter.The frequency of cloud thickness less than 4 km in different seasons is above 75.0％.By comparing the macro parameters of near-precipitation clouds and non-precipitation clouds,it is found that the frequency of 10-20 dBZ echo intensity exceeds 30.0％from the height below 5 km to cloud base.When the cloud base height is less than 2 km,the cloud top height is higher than 4 km,and the cloud thickness is more than 3 km,precipitation is prone to occur.

Abstract:Using daily rime and meteorological observation data from 1981 to 2020 in Lushan Scenic Area,the multi-scale climate change characteristics of rime in Lushan Mountain are statistically ana-lyzed.The correlation between rime and meteorological factors is analyzed and the rime landscape tourism meteorological index is established.Then,this index is verified by the rime observation data from Novem-ber 2022 to April 2023.The results are that:(1)The annual average number of rime days in Lushan Mountain is 37.2 d,with a maximum of 60 d in 1984.In 1999,it was the least,only 15 d.The overall annual rime days show a significant decreasing trend,with a distinct mutation point appearing in 1998.The number of rime days had a significant 2-4 year oscillation period in the late 1980s,and from late 1990s to the early 21st century,and a significant 12-14 year oscillation period from the mid-1990s to the early 21st century.(2)Lushan rime usually occurs from November to April of the following year,with the highest occurrence frequency in January and the lowest frequency in April.Winter is the main season for Lushan rime,accounting for more than 70％of the whole year.The first recorded date to see rime in Lus-han Mountain is November 8th,and the averaged initial date of Lushan rime is November 14th and the averaged latest date is April 16th.The averaged end day of Lushan rime is March 19th and the longest du-ration of rime is 39 days.The minimum temperature range for the occurrence of Lushan rime is-16.7 to 4.2 ℃,the relative humidity range is 8％to 100％,and the wind speed range is 0 to 24.4 m·s-1.(3)Adequate water vapor,temperature low enough,and relatively low wind speed are the key factors for the formation of the rime landscape.The average annual number of days for the rime landscape in Lushan Mountain is 16.3 d.In the past 40 years,the lowest temperature range for the rime landscape in Lushan Mountain has been-13.4 to-0.2℃,with an average temperature range of-9.7 to1.6 ℃,a relative humidity range of 40％to 100％,and a wind speed range of 0.3 to 12.3 m·s-1.Under such conditions,the probability of fog formation is almost 100％,and the probability of rainfall that occurs on the previous day is greater than that on the very day.(4)The established Rime Tourism Meteorological Index(RTMI)has been tested and can be applied to practical operations.It can be a guide for tourists to enjoy the rime landscape,and helps improve the quality of tourism meteorological services in Lushan Mountain.

Abstract:Based on air quality monitoring data,conventional meteorological observation data,and air pollution reanalysis data of 18 prefectural cities in Henan Province from 2017 to 2020,the O3 pollu-tion characteristics and impact factors of Henan are analyzed.The results show that the annual average O3 concentration in most cities of Henan exceeded the standard during 2017-2019,but the number of cities with O3 concentration exceeding the standard was significantly reduced in 2020.The most serious areas with O3 concentration exceeding the standard were in the northwest of Henan.The annual average con-centration of O3 pollution in Henan shows a gradually decreasing trend from northwest to southeast.There was the most severe ozone pollution in 2017,the most O3 pollution days in 2019,and the lightest O3 pol-lution and the least pollution days in 2020.The areas with higher daily O3 exceeding rates are distributed in the northern and central regions,and areas with the lowest O3 exceeding rates are Sanmenxia or Xin-yang.The highest ozone concentration in a day occurs at 15:00 or 16:00,and the lowest value mostly at 07:00 in summer(08:00 in winter).The diurnal variation range of O3 is the largest in summer and the smallest in winter.The EOF analysis shows that the O3 concentration during 2017-2019 in Henan pres-ents a ladder-pattern of high in the west and low in the east.The time coefficient is positive from April to September in 2017,and from April to early October in 2018 and 2019.The spatial distribution character-istics of O3 EOF first mode and the corresponding time coefficients are consistent with the actual situation.The temperature has the highest correlation with the O3 concentration with the correlation coefficient sur-passing 0.7.The O3 concentration is positively correlated to air temperature and sunshine hours,but neg-atively correlated to pressure,relative humidity,CO concentration,NO2 concentration and PM2 5 concen-tration.When the probability of O3 concentration exceeding the standard is 80％in June,the maximum temperature range is 30.7-37.0 ℃,the relative humidity range is 36.3％—70.7％,the sunshine hours range is 0.9-11.8 h,and the pressure range is 971.2-1000.5 hPa.

Abstract:In view of the two atypical snowfall events in Jiangsu Province at the beginning of 2022,the key factors of the circulation situation and water vapor,dynamic and thermal conditions for the occur-rence and maintenance of the two snowfall events are clearly identified.The results show that:(1)There were obvious differences in the falling areas and magnitudes of the two snowfall events.The No.1 snow-fall event mainly occurred in southern Jiangsu,with weak snowfall intensity and complex phase transfor-mation of rain and snow.During the No.2 snowfall,most parts of the province saw pure snow,the accu-mulated precipitation decreased from south to north,and the snow depth was thick.(2)During the two snowfall events,there existed a"cold pad"on the ground,the deep northeast cold vortex in the middle and high layers transported cold air continuously,the intensity of the southwest jet in the middle layer was relatively strong,and warm shear line appeared in the lower layer.(3)The water vapor condition in the early stage of the No.2 snowfall event was significantly better than that in the No.1 snowfall event,the south-west warm and wet airflow in the middle and lower layers was stronger.In terms of dynamic condi-tions,there was a deep and lasting frontogenesis in the lower layer,forcing the warm and humid air flow to rise along the"cold pad".For thermal conditions,there was a shallow inversion layer within 850-700 hPa,but the temperature of the whole layer was below 0 ℃.The cold air near the ground formed a"cold pad".Strong wind shear and symmetrical instability had a positive feedback mechanism.(4)Both of the two snowfall events were atypical snowfall events,featured with good water vapor conditions,warm shear in the lower troposphere,obvious frontogenesis,shallow inversion layer in the middle and lower tropo-sphere,thin near-surface"cold pad",and low temperature in the whole layer.

Abstract:The Phoenix Mountain Ski Resort in Yakeshi is one of the venues of the 14th National Winter Games.Based on the data of the automatic weather station in February 2019 in the venue,the characteristics of over-threshold wind process(OWP)and extreme wind process(EWP),diurnal varia-tion and their relationship with terrain are analyzed.The results show that the Yakeshi venue is located on the south-west slope,the numbers of OWP and EWP descend from the high part to the low part of the venue.The longest durations of OWP and EWP are 21 h and 6 h respectively,and the peak wind speed tends to occur in 12:00-15:00.The maximum wind speeds of EWP and the OWP lasting less than 12 h increase with time.Affected by the mountain terrain,the OWP at the entrance of the gorge on the west side of the competition venue has the longest duration.EWP and OWP are prone to occur at the high alti-tude on the north side of the venue,and the wind speed is the largest.Affected by the high terrain in the northwest of the venue,the dominant wind of EWP and OWP is southwest wind,while the dominant wind in the rest of the venue is west wind.

Abstract:To solve the problem of insufficient research on systematic deviation detection anomalies in existing automatic soil moisture quality control methods,a quality control algorithm based on spatio-temporal matching consistency test of observation and background field data is designed by means of sta-tistical principles,combined with the characteristics of CMA Land Data Assimilation System(CLDAS)soil volume water content analysis products.The automatic soil moisture observation data from May to September 2023 is used as validation data to test and analyze the quality control ability of the algorithm.The results show that this method can identify three types of problems,that is,sensor calibration parame-ter drift,equipment performance degradation,and equipment failure in soil moisture observation stations.The algorithm has a detection accuracy of 92％and can effectively detect the systematic deviation problem of automatic soil moisture.This algorithm has been applied to the Integrated Meteorological Observation Data Quality Control System for conducting quality control and evaluation of real-time automatic soil mois-ture hourly data nationwide.

Abstract:For the quantitative monitoring and evaluation of winter wheat stripe rust,a disease index of wheat stripe rust extracted from canopy digital photos is taken as the dependent variable.By screening the remote sensing features from DJI Phantom 4 multiple spectral unmanned aerial vehicle(UAV)inclu-ding band reflectance,vegetation indices and GLCM(Gray-Level Co-occurrence Matrix)texture indices,and also comparing the simulation accuracy of eight machine learning algorithms under different combina-tions of remote sensing features,the optimal monitoring model of winter wheat stripe rust by multispectral UAV is confirmed.The results reveal that red channel digital number and the normalized red channel dig-ital number are the best indices to extract infected and normal green leaves from digital images,respec-tively.Among the five band reflectances,the reflectances of Near Infrared,Red edge,and Red channel are extremely significantly correlated to disease index.There are 29 out of 30 vegetation indices correlated to disease index at significant level or above,with the top 10 vegetation indices highly correlated being OSAVI,RBNDVI,PVI,SAVI,EVI2,NDVI,MSR,RVI,MNVI,and MTVI;Six texture indices from GLCM including Variance,Homogeneity,Contrast,Dissimilarity,Entropy,and Second Moment,are correlated to disease index at extremely significant level.The simulation test of eight machine learning al-gorithms of the four feature combination schemes shows that the ExtraTrees algorithm,which takes the band reflectance,vegetation indices and texture indices as remote sensing features,has the highest accu-racy,with a root mean square error of 1.2955,and is the optimal monitoring model for winter wheat stripe rust.

Abstract:Based on daily maximum and minimum temperature data of 753 national stations in north-ern China from 1961 to 2022 and the ERA5 monthly reanalysis data,this paper comparatively analyzes the spatial and temporal distribution characteristics of the summer warm day,warm night and compound extreme high temperature events in the Bohai Rim region and their corresponding atmospheric circulations in the same period and the previous period.The results show that,in the Bohai Rim region in summer,warm day shows a decreasing trend,while the warm night and compound extreme high temperature events show an increasing trend.Moreover,the increase in compound extreme high temperature becomes more significant.Both of the warm night and compound extreme high temperature have obvious inter-decadal variation characteristics.Warm day,warm night and compound extreme high temperature appear mainly in the early to middle of June,late June to mid-August,and mid-to-late July,respectively.In particular,the intensities of compound extreme high temperature events have increased significantly.Through the cor-responding circulation field of the three extreme high temperature index regression,it is found that the compound extreme high temperature has an obvious wave-train structure in Eurasia,similar to the result of warm day regression.But what is different is that the positive anomaly center in western Europe and the Baikal Lake region is much stronger,and the negative anomaly center in West Siberia is much wea-ker.The regression of the previous circulation fields suggests that the three extreme high temperature in-dices all have obvious wave-train structures in the mid-high latitudes.In the middle and upper tropo-sphere,a positive anomaly center of height field appears near the Baikal Lake area,accompanied by an abnormal anticyclonic circulation and the abnormal sinking motion.This is responsible for the occurrence of extreme high temperature events around the Bohai Rim region.

Abstract:To investigate the variation characteristics of air pollutants and the influence of meteoro-logical elements on the pollutant concentrations in Xiaoshan District,Hangzhou,during the Asian Games period,this paper analyzes the relationships between pollutant variations and meteorological elements in Xiaoshan by using the observation data from Xiaoshan Meteorological Station and Environmental Monito-ring Center from September 20 to October 10 during 2015-2022 as well as the Origin and Spearman rank correlation statistical analysis methods.The results demonstrate that during the same period of Asian Games from 2015 to 2022,the air quality in Xiaoshan,Hangzhou,was mainly excellent or good,with an excellent and good rate of 81.3%.Ozone(O3)was the primary pollutant,and the daily variation of its concentration exhibited a single-peak pattern,with the peak value occurring between 13:00 and 15:00 and the trough value at 06:00.Moreover,the concentrations of O3 were significantly higher in the day-time than over the nighttime,indicating its close relationship with solar radiation.In addition,different from the variation characteristics of O3 concentration,the daily variation of NO2 and particulate matter(PM2.5 and PM10)concentrations showed a double-peak and single-valley pattern.Further analysis shows that the meteorological elements strongly correlated to O3 concentration are relative humidity,temperature and sunshine hours.When the temperature is between 28℃and 35℃,the relative humidity is between 40%and 60%,and the hourly sunshine duration reaches 0.8 to 1.0 hours for 2-6 consecutive hours,light pollution or worse air quality conditions are more likely to occur.These results can preliminarily serve as meteorological warning indicators for determining O3 pollution.

Abstract:In this study,the glass greenhouse production is chosen as the study object.Then,based on the meteorological data from 69 surface observation stations from 1981 to 2020,according to the re-quirements of suitable light and temperature conditions for greenhouse crop growth,this paper determines three energy consumption stages,namely,no energy consumption period,heating energy consumption period and cooling energy consumption period,from the perspective of annual greenhouse production en-ergy consumption.In accordance with the main limiting meteorological factors that affect the glass green-house energy consumption and the research methods of agricultural climate resources,this paper also de-termines ten greenhouse climate zoning indexes,such as the period length,the total solar radiation and the effective accumulated temperature for each of the three stages,and the negative accumulated tempera-ture during the heating energy consumption period.Moreover,the principal component analysis method,systematic clustering method and the geographical information system are used in delineating the advanta-geous areas of glass greenhouse production in Jiangsu Province based on energy consumption.The results show that the most suitable areas lie in the northeast of Jiangsu Province and Fengxian County of Xuzhou City,the suitable areas are mainly between the Yangtze River and the Huaihe River and in the northeast-ern regions of the Huaihe River,and the sub-suitable areas are mainly located in southern Jiangsu.

Abstract:Based on the observations collected from 2038 national weather stations in Chinese Main-land from 2018 to 2020,the accuracy and adaptability of the latest generation of ERA5-Land reanalysis data of hourly temperature dataset in Chinese mainland are evaluated.The results are as follows:(1)During the assessment period,the ERA5-Land temperature can better reflect the characteristics of the temperature change in Chinese mainland,but it is somewhat lower than that observed by meteorological stations.(2)The evaluation of daily temperature variation shows that the temperature error from 06:00 to 09:00(UTC)is slightly larger than that at other times.(3)The evaluation of monthly and seasonal tem-peratures shows that the temperature error decreases gradually from January to July and increases gradually from July to the next January,with greater errors in winter than in summer.(4)The assessment of differ-ent topographies shows that,with the rises in altitude and slope,the correlation between ERA5-Land temperature and station-observed temperature has a decreasing trend while their error is gradually increas-ing.Overall,ERA5-Land temperature datasets are of good applicability in Chinese mainland,and can provide data support for climate change,scientific research and sustainable development.

Abstract:To find out the climatic quality of loquat in Fujian Province,taking the loquat cultivar"Jiefangzhong"in Putian City of Fujian Province as an example,we use the meteorological data of 23 re-gional meteorological stations in Putian City over the years 2013-2021 and the detection data of physical and chemical components of loquat quality at 24 picking sites in 2021 to analyze the key period and key meteorological factors affecting loquat quality and determine the indexes and grading standards of loquat climatic quality.The weight of each characterization index is determined by analytic hierarchy process.The weighted index summation method is used to construct the evaluation model of loquat climatic quality,and the K-means clustering analysis method is used to divide the loquat climatic quality grades.Moreo-ver,the climate quality evaluation of Putian loquat in 2021 is carried out.The results show that the key climate indexes affecting loquat quality are 30 d sunlight hours before picking,20 d diurnal temperature range before picking,30 d effective accumulated temperature higher than or equal to 10.0℃before picking,and the number of rainy days in 20 days before picking.Based on these four climatic indexes,the climate quality evaluation index model of loquat is established as follows:In accordance with the cli-matic quality index,the climate quality of loquat is divided into the following levels:"Excellent"(0.75≤Ic<1),"Good"(0.45≤Ic<0.75),"Satisfactory"(0.20≤Ic<0.45),"Average"(0≤Ic<0.20).With the above standard levels,the climatic quality grade of loquat at 24 loquat production bases in Pu-tian City in 2021 is evaluated,and the results are that the"Excellent"grade accounts for 29.2%,and the"Good"grade accounts for 70.8%.The coincidence rate between the evaluation results of climatic quality grade and the actual quality grade difference of loquat≤1 is 91.7%.

Abstract:Under the background of global warming and urbanization,exploring the spatio-temporal evolution and driving factors of thermal environment is very significant for reducing urban heat island effect and building ecologically livable cities.Based on Landsat remote sensing images,the land surface temperature(LST)of Xi'an and Xianyang in spring from 1992 to 2022 is extracted.The spatio-temporal evolution of thermal environment and its driving factors are investigated by means of landscape pattern and multi-scale geographically weighted regression.The results show that:(1)The average LST in the study area was in an upward trend from 1992 to 2022.The highest and lowest LSTs have increased in different ranges,leading to a greater dispersion and temperature difference between them.(2)The thermal grades of urban area have been gradually differentiated so that the heat islands in urban area are divided into stronger heat islands,weaker heat islands and non-heat islands.Then,the urban heat island changes from agglomeration to dispersion.The stronger heat islands in rural area have shrunk,while the stronger cold islands have presented an aggregated distribution with the transition from dispersion to agglomera-tion.(3)The landscape metrics reveal that the diversity and evenness of the landscapes in the study area have increased.Fragmentation of landscape types has decreased,connectivity has strengthened,and shapes have become regular due to human disturbance.(4)NDVI and TCW which have negative effects on LST are the main driving factors of the LST evolution.The constant term and NTL have opposite effects on LST at different spatial positions whose interaction intensity with LST are less than those of ND-VI and TCW.

Abstract:Local climate zone(LCZ),as a new classification framework for studying urban heat islands(UHI),can quantify the complex relationship between urban morphology and thermal environ-ment.Based on the high-precision classification of underlying surfaces and the data of automatic weather stations(AWSs)from March 2019 to February 2020,the morphological compositions of four major cities in the Beijing-Tianjin-Hebei metropolitan area and their contribution to the UHI intensity(UHII)are an-alyzed in this study.The results show that:(1)The urban areas of Beijing,Tianjin,Tangshan and Baoding are respectively composed of 16,13,13 and 14 LCZ types,of which only 12,7,4 and 4 LCZ types have AWSs.In these four cities,the building underlying surfaces with the highest proportion of area are LCZ5,LCZ4,LCZ3,and LCZ3,respectively.(2)The UHII is not only related to the city scale,but also significantly influenced by geographical location.The annual average UHIIs of the four cities are ranked as Beijing>Baoding>Tianjin>Tangshan.(3)The UHII on the underlying surfaces with dense me-dium and high-rise buildings(such as LCZ1 and LCZ2)is generally strong,while the green underlying surfaces(such as LCZB)have relatively weak UHII.(4)The contribution rates of various LCZs to the UHII are relatively stable,with LCZ5(30.4%),LCZ5(34.8%),LCZ3(60.6%)and LCZ3(67.4%)contributing the most to the annual average UHII in Beijing,Tianjin,Tangshan and Baoding,respectively.

Abstract:In the quest to establish a monthly-scale forecasting methodology for solar radiation in China,this paper analyzes monthly Global Horizontal Irradiance(GHI)data from 53 national stations spanning 1979 to 2020,ERA5 reanalysis data,and historical simulation data from the MRI-CGCM model.An EOF decomposition is conducted on the mean GHI from 1979 to 2017 to identify principal spatial modes and corresponding temporal coefficients.These time coefficients and correlation analysis are used to comprehensively examine the atmospheric circulation within ERA5 reanalysis and MRI-CGCM simula-tion data.Significant regions,as determined by monthly segmentation and statistical testing,are selected as predictors.Besides,machine learning techniques are employed to develop a forecast model for the time coefficients.The time coefficients corresponding to the first three modes of GHI in each month of China are predicted,thereby enabling monthly-scale forecasting of GHI across China.The modeling phase involves a comparison between various machine learning algorithms and traditional stepwise regres-sion(SWR)methods.Evaluation results of mean bias and mean absolute deviation indicate that,in this study,the machine learning methods,specifically the Random Forest(RF),Gradient Boosting Decision Tree(GBDT),Decision Tree Regression(DTR)and K-Nearest Neighbors(KNN),outperform the con-ventional stepwise regression(SWR).RF exhibits superior performance in both mean bias and mean ab-solute deviation assessments.Further validation by the RF model for the prediction results in 2017 and 2018,including a comparison of forecast results and ACC scoring,demonstrates that the RF model has a certain forecasting capability in all months.Forecasts for the winter half-year surpass those for the summer half-year,with the forecasts in January,November,and December showing the best performance in the two-year forecasts.Regionally,forecasts for southern and eastern China prove to be more accurate than those for the northern and western regions of China,with the smallest deviations primarily located in East China and South China.

Abstract:The transmitter filament power supply mainly provides power for the radar klystron fila-ment,which features complex circuit and suffers a high failure rate.Based on the signal flow of the fila-ment power supply in the SA-type China New Generation Weather Radar(CINRAD/SA),this study con-ducted a layer-by-layer fault tree analysis to trace the causal relationship of faults and established a logical and standardized chip-level fault diagnosis procedure combined with the waveform characteristics of criti-cal test points in the filament power supply.Through this process,the malfunction where the filament power supply ceased operation during power-on process due to abnormal alarm thresholds was successfully repaired.Among them,the undervoltage alarm was accompanied by the activating of the filament power supply voltage fault indicator light on the transmitter panel,while the overvoltage alarm was accompanied by the filament power supply panel fault indicator light illuminated.The results of the above two success-ful fault repair cases show that the fault tree analysis method can effectively optimize the chip-level fault diagnosis process and effectively improve the efficiency of fault diagnosis.The combination of chip-level fault diagnosing technique and approach and the chip-level fault diagnosing process established based on critical-test-point waveform enables rapid localization of faults to the minimal cut set(at chip level)of the fault tree.Owing to its standardized steps and clear logic,this method is easy for technicians to mas-ter and can greatly enhance maintenance efficiency and cut repair costs.Based on the chip-level fault di-agnosing process,by developing test procedures and fault repair modeling software,automatic localization of faults at the chip level becomes feasible,which could provide a key technical support for automated chip-level fault diagnosis in meteorological radar maintenance platforms.