Frost Automatic Identification Model Based on Logistic Regression

doi:10.3969/j.issn.1000-6362.2019.08.006

Abstract

Abstract: Using the threshold-based methods of meteorological elements on frost-free days, the quality-controlled daily observation data were firstly established, using the observed data from 23 typical meteorological stations in Anhui province from 2003 to 2017. The automatic identification models of frost based on Logistic regression were constructed for each meteorological station in Anhui province, and the performance of the frost identification model was evaluated. The results showed that: (1) the daily frost-free phenomena could be correctly determined for each station in Anhui province by the thresholds of meteorological elements such as temperature, wind speed and precipitation amount. (2)Temperature, humidity and wind speed were selected as the identification elements in the frost identification models for all meteorological stations. The observation time of model elements mostly occurred at the stage from 4：00 to 8：00. (3)The accuracy rate, hit rate, miss rate, empty judgement rate and TS score of Logistic regression model based on independent sample test were 89.0%, 91.6%, 8.4%, 15.8% and 78.2%, respectively. This indicates that the frost recognition model established by Logistic regression has good ability to identify frost in Anhui province. (4) Compared with Bayes discriminant model, it was found that the frost identification model based on Logistic regression had higher accuracy rate and TS score, and lower empty judgement rate. Therefore, the proposed Logistic regression model can be applied to the automatic identification of frost in the future.

Key words: Frost, Logistic regression, Automatic identification

ZHU Hua-liang, WEN Hua-yang, HUA Lian-sheng, JIN Su-wen, CHEN Jing-jing. Frost Automatic Identification Model Based on Logistic Regression[J]. Chinese Journal of Agrometeorology, 2019, 40(08): 534-542.

References

[1] 中国气象局.地面气象观测规范[M].北京:气象出版社, 2003:23-26. China Meteorological Administration.Specification for ground meteorological observation[M].Beijing:China Meteorology Press, 2003:23-26.(in Chinese) [2] 李轩,张艳红,韩丽娟,等.2018年春季气象条件对农业生产的影响[J].中国农业气象,2018,39(7):490-492. Li X,Zhang H Y,Han L J,et al.The impact of meteorological conditions on agricultural production in the spring of 2018 [J].Chinese Journal of Agrometeorology,2018,39(7):490-492. (in Chinese) [3] 王景红,刘璐,高峰,等.陕西富士系苹果花期霜冻灾害气象指标的修订[J].中国农业气象,2015,36(1):50-56. Wang J H,Liu L,Gao F,et al.Revision on meteorological indices of florescence frost disaster for Fuji apple in Shaanxi Province[J].Chinese Journal of Agrometeorology,2015,36(1): 50-56.(in Chinese) [4] 张艳红,刘维,张蕾.2017年秋季气候对农业生产的影响[J].中国农业气象,2018,39(1):69-71. Zhang H Y,Liu W,Zhang L.The impact of climate on agricultural production in the autumn of 2017[J].Chinese Journal of Agrometeorology,2018,39(1):69-71.(in Chinese) [5] 刘红杰,武永峰,任德超,等.黄淮冬麦区气象因子与小麦晚霜冻害关系研究:以商丘市为例[J].中国农业气象,2017, 38(8):517-525. Liu H J,Wu Y F,Ren D C,et al.Correlation between meteorological factors and late frost damage in Huanghuai winter wheat region:a case study of Shangqiu in Henan province[J].Chinese Journal of Agrometeorology,2017,38(8): 517-525.(in Chinese) [6] 马舒庆,吴可军,陈冬冬,等.天气现象自动化观测系统设计[J].气象,2011,37(9):1166-1172. Ma S Q,Wu K J,Chen D D,et al.Automated present weather observing system and experiment[J].Meteorological Monthly, 2011,37(9):1166-1172.(in Chinese) [7] 宗晨临,朱海兵,王志成,等.电容式霜露传感器自动化观测对比分析[J].气象科技,2016,44(2):204-210. Zong C L,Zhu H B,Wang Z C,et al.Comparative analysis of observations from capacitive frost and dew sensor on frost and dew phenomenon[J].Meteorological Science and Technology, 2016,44(2):204-210.(in Chinese) [8] 茆佳佳,张雪芬,巩娜,等.基于近地面温湿度观测的露霜综合判别技术[J].气象科技,2016,44(2):197-203. Mao J J,Zhang X F,Gong N,et al.Comprehensive discrimination of dew and frost based on near surface temperature and humidity observations[J].Meteorological Science and Technology,2016,44(2):197-203.(in Chinese) [9] 华连生,温华洋,朱华亮,等.基于Bayes判别法的霜生自动化观测模型探讨[J].气象,2015,41(8):964-969. Hua L S,Wen H Y,Zhu H L,et al.Automated observation model for frost based on Bayes discriminant method [J].Meteorological Monthly,2015,41(8):964-969.(in Chinese) [10] 朱虹晖,武永峰,宋吉青,等.基于多因子关联的冬小麦晚霜冻害分析:以河南省为例[J].中国农业气象,2018,39(1): 59-68. Zhu H H,Wu Y F,Song J Q,et al.Analysis to late frost damage for winter wheat based on meteorological factors:taking Henan province as an example[J].Chinese Journal of Agrometeorology,2018,39(1):59-68.(in Chinese) [11] 刘红杰,倪永静,任德超,等.黄淮冬麦区晚霜冻易发时段冠层内最低气温分布及估算[J].中国农业气象,2018,39(12): 786-795. Liu H J,Ni Y J,Ren D C,et al.Estimation and distribution of minimum air temperature within winter wheat canopy in prone period of late frost[J].Chinese Journal of Agrometeorology, 2018,39(12):786-795.(in Chinese) [12] 郭小芹,钱莉.河西走廊东部一次霜冻成因分析[J].气象科技,2018,46(2):369-373. Guo X Q,Qian L.Causal analysis of a late frost event in eastern Hexi corridor[J].Meteorological Science and Technology,2018,46(2):369-373.(in Chinese) [13] Collins M,Schapire R E,Singer Y.Logistic regression, adaboost and bregman distances[J].Machine Learning,2002, 48(1-3):253-285. [14] Gareth J,Daniela W,Trevor H,等.统计学习导论:基于R应用[M].王星等译.北京:机械工业出版社,2015:92-94. Gareth J,Daniela W,Trevor H,et al.Introduction to statistical learning:based on R application[M].Translated by Wang X,et al.Beijing:Machinery Industry Press,2015:92-94.(in Chinese) [15] 盛裴轩,毛节泰,李建国,等.大气物理学[M].北京:北京大学出版社,2003:310-325. Sheng P X,Mao J T,Li J G,et al.Atmospheric physics [M].Beijing:Peking University Press,2003:310-325.(in Chinese) [16] 黄思源,傅伟忠.露?霜?结冰天气现象综合判别[J].气象科技,2014,42(3):359-363. Huang S Y,Fu W Z.A study of comprehensive discrimination of dew,frost and freezing[J].Meteorological Science and Technology,2014,42(3):359-363.(in Chinese) [17] 王松桂,陈敏,陈立萍.线性统计模型[M].北京:高等教育出版社,1999:127-134. Wang S G,Chen M,Chen L P.Linear statistical model[M]. Beijing:Higher Education Press,1999:127-134.(in Chinese) [18] 郑明,陈子毅,汪嘉冈.数理统计讲义[M].上海:复旦大学出版社,2006:148-151. Zheng M,Chen Z Y,Wang J G.Lectures on mathematical statistics[M].Shanghai:Fudan University Press,2006:148-151. (in Chinese)