塔里木河下游植被恢复遥感模型建立  

A REMOTE SENSING MODEL FOR VEGETATION RESTORATIONIN THE LOWER VALLEY OF THE TARIM RIVER

作　　者：牛婷[1] 李霞[1] 

机构地区：[1]新疆农业大学林学院,乌鲁木齐830052

出　　处：《国土资源遥感》2008年第2期79-83,共5页Remote Sensing for Land & Resources

基　　金：国家自然科学基金(30460010);教育部科技研究重点项目(205183);新疆维吾尔自治区高校科研计划项目(XJEDU2004119)联合资助

摘　　要：以2000年-2006年塔里木河下游MODIS、ETM及ASTER数据为信息源，从不同时间和空间分辨率角度分析塔里木河下游喀尔达依断面输水后NDVI时空变化规律，并在此基础上建立了研究区不同离河距离、不同时间MODIS、ETM及ASTER数据NDVI的预测模型；根据模型预测的NDVI值，使用像元二分法反演植被覆盖度，并根据当年实地调查数据对反演植被盖度进行精度验证，其平均精度达82．88％以上。以上研究结果为监测和预测塔里木河下游植被恢复提供方法借鉴和恢复趋势预测参考。Taking MODIS, ETM and ASTER data obtained from the lower valley of the Tarim River in the period of 2000 -2006 as the information source, the authors studied NDVI spatial and temporal variation regularity after water conveyance of the Kardayi section in the lower valley of the Tarim River from different time and space resolution angles. On such a basis, a NDVI prediction model of MODIS, ETM AND ASTER data obtained from different distances from the river and different periods was established. According to NDVI values predicted by the model, pixel dichotomy was used to perform inversion of vegetation coverage, whose precision was then verified with the investigation data obtained in the same year. It is shown that the average precision reached 82.88% or even higher. These achievements provide valuable reference data for the monitoring and prediction of the vegetation restoration in the lower valley of the Tarim River.

关 键 词：遥感模型 植被恢复 盖度反演 不同分辨率遥感数据 NDVI 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置]