植被指数与气象因子在像元尺度的多元线性回归研究  

Multiple Linear Regression Study of Vegetation Index and Meteorological Factors at the Pixel Scale

作　　者：栾金凯 刘登峰[1] 刘慧[4] 林木[5] 黄强[1] LUAN Jinkai;LIU Dengfeng;LIU Hui;LIN Mu;HUANG Qiang(State Key Laboratory of Eco-hydraulics in Northwest Arid Region of China,Xi′an University of Technology,Xi′an 710048,China;Key Laboratory of Water Cycle and Related Land Surface Process,Institute of Geographic Sciences and Natural Resources Research,CAS,Beijing 100101,China;University of Chinese Academy of Sciences,Beijing 100049,China;China Institute of Water Resources and Hydropower Research,Beijing 100038,China;School of Statistics and Mathematics,Central University of Finance and Economics,Beijing 100081,China)

机构地区：[1]西安理工大学省部共建西北旱区生态水利国家重点实验室,陕西西安710048 [2]中国科学院地理科学与资源研究所,陆地水循环及地表过程重点实验室,北京100101 [3]中国科学院大学,北京100049 [4]中国水利水电科学研究院,北京100038 [5]中央财经大学统计与数学学院,北京100081

出　　处：《华北水利水电大学学报（自然科学版）》2020年第3期14-24,共11页Journal of North China University of Water Resources and Electric Power：Natural Science Edition

基　　金：国家自然科学基金项目(51779203,51609270);陕西省自然科学基础研究计划项目(2016JQ5105);榆林市科技计划项目(2016-21)。

摘　　要：遥感获取的归一化植被指数(NDVI)能很好地代表植被的区域特征和空间变异特征。分析区域植被指数的变化对监测和评价半干旱区植被保护效果具有重要价值。为了更好地在像元尺度上对NDVI进行模拟,本文基于陕西省榆林市2000—2017年MODIS/Terra NDVI遥感数据,结合该地区17个气象站点2000—2017年的气象数据,以2000—2014年为率定期,2015—2017年为验证期,应用多元线性回归分析方法进行逐像元分析。结果表明:6—7月平均气压、7月平均最高气温、7月平均气温和7月平均相对湿度与榆林市8月份NDVI的相关程度最大,是榆林市8月份NDVI的主要影响因子;对比榆林市2015—2017年NDVI的模拟值和实测值发现,2015年NDVI模拟值的残差为负值的区域主要位于靖边县、横山区、子洲县、神木市的北部及府谷县的南部,2016年残差为负值的区域主要分布在神木市的北部和府谷县的南部,2017年残差值为负值的区域主要位于靖边县的西部,佳县、米脂县和子洲县的部分区域。残差为正值的区域,说明人类活动对该区域植被的生长起到了促进作用,这些地区的封山育林、退耕还林、退牧还草等措施的实施效果较好。在现阶段人类活动变化幅度不大的情况下,应用本文方法可根据未来的气象条件预测未来植被理论上的生长状况。The normalized difference vegetation index(NDVI)obtained by remote sensing can well represent the regional characteristics and spatial variation characteristics of vegetation.It is of great value to analyze the change of regional vegetation index for monitoring and evaluating the effect of vegetation protection in semi-arid areas.In order to better simulate NDVI at the pixel scale,this article is based on MODIS/Terra NDVI remote sensing data from 2000 to 2017 in Yulin City,Shaanxi Province,and combining weather data from 2000 to 2017 at 17 weather stations in the region.This study takes 2000 to 2014 as the regular rate,2015 to 2017 as the validation period,and applies multiple linear regression analysis method to carry out pixel by pixel analysis.The results are as follows.Firstly,the average air pressure from June to July,the average maximum temperature in July,the average temperature in July and the average relative humidity in July have the greatest correlation with NDVI in August in Yulin City,and are the main factors affecting NDVI in August in Yulin City.Secondly,comparing the simulated and measured NDVI values of Yulin from 2015 to 2017,it is found that the areas with negative residual value of NDVI in 2015 are mainly located in Jingbian County,Hengshan District,Zizhou County,the north of Shenmu City and the south of Fugu County.In 2016,the areas with negative residual value are mainly distributed in the north of Shenmu City and the south of Fugu County.In 2017,the areas with negative residual value are mainly located in the west of Jingbian County,Jiaxian County,Mizhi County and Zizhou County.Thirdly,the areas with positive residuals indicate that human activities have played a role in promoting the growth of vegetation in these areas,and the implementation of measures such as closing mountains for forest cultivation,returning farmland to forest,returning grassland to grassland and other measures have achieved good results.Under the condition of little change of human activities at present,this method can

关 键 词：NDVI 多元线性回归 像元尺度 影响因素 榆林 

分 类 号：Q948[生物学—植物学]