【摘要】 针对锡林郭勒草原的严重退化,近年来启动了草原生态恢复工程,但工程实施的效果由于范围广、时间短而很难予评价。利用锡林郭勒草原地区1999～2004年的SPOT-VGT10天最大值合成NDVI时间序列数据,在求得各年NDVI年最大值(NDVImax)的基础上,通过搜寻各像元邻域内NDVImax的最大值以及与之相应的气候因子,建立起了像元尺度上的气候—植被生长基准响应模型,并以此为依据运用相对残差趋势法识别出了处于恢复阶段的草原区域。最后,将遥感监测结果与羊单位统计资料进行对比分析,结果发现,在县级行政区域尺度上的平均恢复趋势与基于羊单位和NPP构建的放牧压力趋势具有较好的反比例关系,初步证明了该方法在时间序列较短情况下的有效性。更多还原

【Abstract】 Xilin Gol grassland has been experiencing severe degradation since the livestock kept on increasing in the past decades. Recent ’Grassland Ecological Restoration Project’ is being developed for improvement of the degraded grassland and sustainable development. However,due to the short-time of the project implementation,it is difficult to monitor the restored area and assess the effect of the project. In this study,SPOT-VGT maximum value composite (MVC) NDVI temporal series data from 1999 to 2004 of Xilin Gol was used to model climate-vegetation response relationships in pixel scale. The model was based on the maximum NDVI in the neighborhood and their climate factors including accumulated temperature and accumulated precipitation. Then the restored areas were identified by positive trend of normalized residuals between model predicted NDVI and actual values. The result was validated by grazing pressure index (GPI),which was defined as the sheep unit divided by NPP. Negative relationship was found between the trend of residual and GPI at county level,which proved the effectiveness of the method for short-term temporal data.更多还原