偏相关系数在林冠截留影响因子分析中的应用  

Application of Partial Correlation Coefficient to Canopy Interception Impacting Factor Analysis

作　　者：赵明扬[1] 孙长忠[1] 康磊[1] 

机构地区：[1]中国林业科学研究院华北林业实验中心,北京102300

出　　处：《西南林业大学学报（自然科学）》2013年第2期61-65,共5页Journal of Southwest Forestry University：Natural Sciences

基　　金：国家自然科学基金项目(30170769)资助;华北林业实验中心科学试验示范林功能恢复与提升研究项目(CAFYBB2012004)资助

摘　　要：通过对简单相关系数和偏相关系数的对比分析,利用实测资料对林冠截留量与林外降雨量、降雨持续时间、降雨强度、2次降雨间隔时间、温度、相对湿度、风速和降雨期间水面蒸发量之间的关系进行分析。结果发现:利用偏相关分析对其简单相关分析的结果进行修正更符合实际情况,更有助于准确描述变量之间的关系;对偏相关分析的结果结合逐步回归分析,能有效改善回归分析的质量。By comparative analyses of the differences between the simple correlation coefficients and the partial correlation coefficients,the relationship between the canopy interception and the related factors,which included rainfall,time of rainfall duration,rainfall intensity,interval time between the rainfalls,temperature,relative humidity,wind speed and water surface evaporation during the rainfalls was analyzed with the field measured data.The results showed that modifying the results of simple correlation analyses with the results of the partial correlation analyses would be helpful to making the results more actual so as to describe the relations among the variables more exactly.The quality of the regression analyses could be effectively improved by integrating the results of the partial correlation analyses with stepwise regression analyses.

关 键 词：相关系数 偏相关系数 林冠截留 逐步回归 

分 类 号：S715[农业科学—林学]