基于RS和GIS技术,利用黄土高原近30年不同分辨率的归一化植被指数(ND-VI),采用基于小波分析的多尺度空间统计学方法,研究不同时期黄土高原植被景观的多尺度变化特征及其与地形的响应关系。结果表明:(1)黄土高原植被景观1982—2011年期间发生了周期性变化。其中,1990年以前为植被恢复阶段,1990--2001年为植被退化阶段,2001年以后为植被恢复阶段。(2)植被景观的空间异质性与数据获取的时间、数据分辨率、空间位置和地形均有关。一方面,植被指数和数据分辨率越高,植被景观的空间异质性就越大,而且沿经度方向植被景观的空间差异性大于纬度方向;另一方面,NDVI的多尺度变化特征与地形因子有不同程度的相关性,表现为:高程〉地形湿度〉坡度〉坡向,这对于今后黄土高原植被景观的合理布局,提高生物多样性,控制水土流失,增强景观的连续性具有特别重要的意义,也可为指导生态环境建设提供基础资料。
Based on remote sensing (RS) and geographic information system (GIS)technology, and using the normalized difference vegetation index (NDVI) of different resolution in the Loess Plateau from 1982 to 2011, the multi-scale change characteristics of different periods of the Loess Plateau vegetation landscape in relation to the terrain were analyzed by multi-scale spatial statis- tics method of wavelet analysis. The results indicated that the vegetation landscape of Loess Pla- teau exhibited periodic changes from 1982 to 2011, mainly including three phases, namely the vegetation restoration phase before 1990, the vegetation degradation phase from 1990-2001 and vegetation recovery phase after 2001. Furthermore, the spatial heterogeneity, data acquisition time, data resolution and spatial position of vegetation landscape had good correlations with to- pography in the Loess Plateau. On one hand, higher vegetation index and data resolution caused greater spatial heterogeneity of vegetation landscape, and the spatial heterogeneity of vegetation landscape along the longitude direction was greater than that along the latitude direction. On the other hand, the multi-scale NDVI variation was correlated to topographical factors, and the corre- lation coefficients ranked as follows: DEM 〉 CTI 〉 slope 〉 aspect. These results will be of great significance for vegetation landscape rational adjustments, biodiversity enhancement, soil erosion control, and landscape continuity enhancement, and provide important basis for ecological envi- ronment construction in the Loess Plateau.