西安市气象要素变化特征及可利用降雨量预测模型
Variation Characteristics of Meteorological Elements and Prediction Model of Available Precipitation of Xi’an City
  
DOI:
中文关键词:  马尔科夫链  频率曲线  可利用降雨量  特征分析  预测模型  西安市
英文关键词:Markov chain  frequency curve  available rainfall  feature analysis  forecasting model  Xi’an City
基金项目:长安大学博士生助研科技项目(CHD2011ZY022;CHD2011ZY025);中央高校基本科研业务费专项资金(CHD2011ZY020);国家自然科学基金项目(41172212)
作者单位
吴健华 长安大学 环境科学与工程学院 
李培月 长安大学 环境科学与工程学院 
钱 会 长安大学 环境科学与工程学院 
摘要点击次数: 1390
全文下载次数: 1267
中文摘要:
      利用西安气象站1951-2008年的气象资料,分析了西安市降雨量、蒸发量、可利用降雨量的变化特征,并建立了基于理论频率曲线修正的加权马尔科夫链年可利用降雨量的预测模型。分析结果表明:西安市可利用降雨量年内变化大,以9月最多,12月最少;四季中以秋季的可利用降雨量最多,冬季的最少,分别占全年总量的44.5%和2.1%;可利用降雨系数也以秋季最高,冬季最低,分别为0.36和0.13;近58年来,西安市平均可利用降雨量的总体趋势以每年0.99mm的线性倾向递减。经过分析,利用加权马尔科夫预测模型预测的西安市可利用降雨量的相对误差,与可利用降雨量实际值呈负相关关系;经过理论频率曲线修正后,其误差相应减少。因此经过理论频率曲线修正的加权马尔科夫预测模型,可以应用于西安市未来年可利用降水资源的预测。
英文摘要:
      The data from 1951 to 2008 from Xi’an station were used to analyze the variation characteristics of precipitation, evaporation and available precipitation in the area. The available precipitation forecasting model based on weighted Markov chain was established and the relationship between model forecasting errors and the observed value of available precipitation was also analyzed. The forecasting results were amended by the frequency curve. The results show that the variation of available rainfall is significant in a year, and it is most abundant in September while is least in December. The proportion of available precipitation in autumn and winter are 44.5% and 2.1%, respectively and the available rainfall coefficients are 0.36 and 011 respectively for the two seasons. During the past 58 years from 1951 to 2008, the average available rainfall reduces in a linear tendency at 0.99 mm per year. The forecasting of available precipitation in Xi’an was taken as an example. The forecasting errors and the real value of local available precipitation are inversely related. The relative errors verified by theoretic frequency curve decreases. It is concluded that the available precipitation forecasting model based on weighted Markov chain modified by theoretic frequency curve can be used to predict the available precipitation in Xi’an.
查看全文  查看/发表评论  下载PDF阅读器