刘珊珊　王建雄　牛超杰

摘要：选取云南省为研究区域，利用2001-2015年MODIS中国植被指数合成产品Tiff遥感图像，通过Arcgis软件对遥感图像进行分区统计，得到研究区域各年每月的植被归一化指数（NDVI）。在此基础上进行统计分析，结合当地的生长季与种植制度，分析研究区域植被指数的年际、季节及月变化的特点。结果表明，在2001-2015年，云南省植被指数有增加的趋势，表明在2001-2015年云南省植被覆盖度基本保持稳定或略有增加的趋势；2001-2015年云南省春季和冬季的NDVI整体呈增加趋势，冬季的增速大于春季，夏季NDVI整体呈减少趋势，而秋季处于平稳状态，说明云南省2001-2015年春、冬两季植被覆盖度不断增加；云南省NDVI月变化差异较为明显，与当地生长季和种植制度密切相关，说明植被指数的月变化特征主要受种植作物的影响。

关键词：NDVI；变化特征；趋势分析；云南省

中图分类号：P207；Q948 文献标识码：A 文章编号：0439-8114（2017）11-2037-04

DOI：10.14088/j.cnki.issn0439-8114.2017.11.009

Abstract： Taking Yunnan as the study area，using the Tiff remote sensing image of MODIS vegetation index composite products from 2001 to 2015，the remote sensing images were divided into districts by the Arcgis software，and the monthly NDVI values of each year in the study area were obtained. Based on the analysis，the characteristics of annual，seasonal and monthly changes of regional vegetation index were analyzed combination with local growing season and cropping system. The results showed that，during 2001 to 2015，there was an increasing trend of the vegetation index in Yunnan，which indicated that the vegetation coverage in Yunnan remained stable or increased slightly from 2001 to 2015. During the period of 2001 to 2015，the NDVI values of spring and winter in Yunnan were increasing in total， and the growth rate of winter was higher than that of spring. The NDVI value was decreasing in summer and the autumn was stable， which meant that the vegetation coverage of spring and winter from 2001 to 2015 was increasing. The monthly variation of the vegetation index was relatively obvious， which was closely related to the local growing season and cropping system， which indicated that the monthly change of vegetation index was mainly affected by the crop.

Key words： NDVI； variation characteristics； trend analysis； Yunnan Province

植被是聯结土壤、大气和水分等要素的自然“纽带”，具有明显的年际变化和季节变化的特点，在一定程度上能代表土地覆盖的变化[1，2]。植被指数是将遥感地物光谱资料经数学方法处理，以反映植被状况的特征量[3]。在众多植被指数中最为常用的是归一化植被指数（NDVI），它能够精确地反映植被绿度、光合作用强度、植被代谢强度及其季节和年际变化，是表征植物生长、植被覆盖、生长状况、生物量等的重要指标[4]，是监测植被变化的有效参数[5，6]，因此，在各地大尺度的植被动态监测、作物长势监测、自然灾害监测和作物产量预测等方面得到广泛应用[7-10]。……