Study On Spring Frost Index And Spatial-Temporal Distribution Characteristics Of Tea Plants With Medium And Lobular Species In Tea Regions Over The North Of The Yangtze River And Southwest China

Tea regions over the north of the Yangtze River(NYR)and Southwest China(SWC)are suitable for tea planting,tea plants are vulnerable to suffer from spring frost in spring.In order to improve the meteorological service quality of spring frost injury of tea plants,it is necessary to study spring frost index and analyze the temporal and spatial characteristics of spring frost in NYR and SWC.This paper took tea plants of medium and lobular species in NYR and SWC as the research object.Based on meteorological data and geographical distribution data of tea trees,the maximum entropy model was used to extract area suitable for tea plants growing and determine the range of NYR and SWC.The historical frost samples in the research area were inverted,and the critical threshold of spring frost of tea trees was determined according to the Overall Accuracy and Youden index and then verified.Accumulated harmful chilling was constructed based on the minimum air temperature and duration days of frost,and K-Means clustering analysis method was used to classify the spring frost thresholds of different grades.Then the frost characteristics in NYR and SWC were calculated,and the spring frost characteristics in NYR and SWC were analyzed from the perspective of time and space.The main results are as following:(1)The accuracy of the initial maximum entropy model based on nine potential climate factors is good.After considering the contribution percentage,permutation importance and Jackknife method of potential climate factors,the multi-year average of extreme minimum temperature,frequency of spring frost,annual average temperature,annual precipitation and average relative humidity from March to September were determined as the leading climate factors affecting the distribution of tea tree planting.The maximum entropy model based on the reconstruction of dominant climate factors also behaved good and was suitable for the simulation of tea trees planting distribution.(2)The optimal critical threshold of spring frost for tea trees in NYR and SWC determined by Youden index was 3.7℃ and 3℃,respectively.The reserved validation samples and typical frost processes were used to test the critical threshold,with an accuracy of 84.3%,74% and 90.9%,96.7%,which confirmed the rationality of spring frost critical threshold in NYR and SWC.Based on minimum air temperature and duration days,accumulated harmful chilling sequence was constructed.By using KMeans clustering analysis tool,the results showed that the slight,moderate,severe and extra severe accumulated harmful chilling of tea plants spring frost were [0,1.2),[1.2,3.8),[3.8,12.6),≥12.6 and [0,1.3),[1.3,4.5),[4.5,21.9),≥21.9.(3)The temporal and spatial distribution characteristics of spring frost were analyzed based on the critical threshold and grade indexes of tea trees in NYR and SWC.The results showed that the spring frost severity in NYR was more in the north and less in the south.First frost date(FFD)is not obvious in spatial distribution pattern,and last frost date(LFD)is later in the northern region than in the southern region.FFD showed an insignificant postponement trend,while LFD showed a significant advance trend.The extent of frost in SWC decreased with the increase of frost grades,and the severity of spring frost was positively correlated with altitude,and the number of spring frost days and times showed a significant decreasing trend.FFD appeared earlier in the low altitude area,showing an insignificant trend of delay,while LFD appeared later in the high altitude area,showing an insignificant trend of advance.