基于MODIS NDVI时间序列数据的藏北草地类型识别

Grassland Types Identification based on Time-series MODIS NDVI Data in Northern Tibet

选取西藏自治区申扎县北部部分区域作为藏北草原的典型代表区进行研究。对区域内2009年全年23期MODIS NDVI数据收集处理,得到23个波段的时间序列数据。通过Savitzky-Golay滤波算法优化时间序列数据并对不同类型的时间序列曲线做对比分析,提取分类的特征波段。最后采用非监督分类法把研究区分为沼泽草甸、高寒草甸、中盖度草原、低盖度草原、裸地和水体6种类型,并结合地形数据分析了其中4种草地类型的空间分布特征。一系列的处理结果及分析内容证明了MODIS NDVI时间序列数据在藏北草地类型识别方面的可行性和适用性。

This paper selected the north part of Shenzha in Northern Tibet as study area.The author collected and processed 23 times MODIS NDVI data of 2009 in order to get the time-series NDVI data of study area.Then reconstruct the NDVI time-series data to improve the data quality by the Savitzky-Golay filter.After that,this paper analyzed the time-series curves of six types to get the different characteristics and extract the best bands for classification.Based on the work before,the author divvied the study area to six types by ISODATA method in ENVI.They are swamp meadow,alpine meadow,middle coverage grassland,low coverage grassland,bare land and water.The author analyzed the spatial distribution characteristics of four grassland types combine with the terrain data.Series of processing results and analysis proved that using the MODIS NDVI time-series data to identify the grassland type in Northern Tibet is feasibility and applicability.