摘要
利用1987、2000、2006、2013年4景Landsat TM/ETM+/TIRS遥感影像反演了格尔木绿洲的地表温度,并对地表温度进行归一化处理,得到地表温度等级图,构建了相邻两景地温等级区转移矩阵。结合归一化植被指数(NDVI)和土地利用/覆盖变化(LUCC),分析了格尔木绿洲冷岛效应的时空变化及影响因素。结果表明:近27年来,格尔木绿洲地表温度呈以高覆盖植被区为低温中心,沿低温中心逐渐向四周升高的空间分布格局,绿洲冷岛效应显著,且具有强-弱-强的变化特征;归一化地表温度(NLST)与NDVI存在显著负相关,并与冷岛效应的强弱变化相吻合,表明植被是影响绿洲冷岛效应的重要因素;2000年林地与耕地增加,2006年林地减少、水体萎缩,2013年城市绿化及耕地、林地增加,反映了LUCC对冷岛效应的影响显著;林地、耕地、草地和水体处于地温等级最低的3个等级中,其低温作用对维护绿洲冷岛效应的稳定性意义重大。
Four remote sensing images from Landsat TM / ETM+/ TIRS in 1987,2000,2006 and2013 were used to invert the land surface temperature( LST) of Golmud oasis. The LST values were normalized and the grade figures of LST were obtained. Then the transition matrix of the grade figures of LST from two adjacent dates was constructed. Furthermore,the normalized vegetation index( NDVI) and land use / land coverage change( LUCC) were used to analyze the spatiotemporal variation and the causes of cold island effect in Golmud oasis. Overall,in the last 27 years,the LST in the oasis presented a spatial pattern of lower temperature center in high vegetation coverage and it ascended from the center to surrounding areas. Oasis cold island effect was significant,and had a strong-weak-strong variation. There was a remarkable negative correlation between normalized surface temperature( NLST) and NDVI that coincided with the changes of the cold island effect,indicating that the vegetation played an important role in the cold island effect. The area increased for farmland and woodland in 2000,decreased for farmland and water bodies in 2006,and increased for farmland,woodland and urban greening in 2013. Woodland,farmland,grassland and water bodies had the priority position in the lowest three grades of LST,and their temperature reduction roles were of significance in maintaining the stability of oasis cold island effect.
出处
《生态学杂志》
CAS
CSCD
北大核心
2014年第12期3339-3348,共10页
Chinese Journal of Ecology
基金
国家自然科学基金项目(40961035)
甘肃省自然科学基金项目(0803RJZA094)
甘肃省级重点学科自然地理学项目资助
关键词
冷岛效应
时空特征
NDVI
LUCC
遥感数据
格尔木绿洲
cold island effect
spatiotemporal characteristics
NDVI
LUCC
remote sensing data
Golmud oasis.