利用线性融合方法进行金花茶自然保护区植被覆盖度时空变化研究

Research on temporal and spatial variation of fractional vegetation cover in Golden Camellia National Nature Reserve using linear fusion method

在利用高空间分辨率影像研究小区域植被覆盖度(FVC)变化时,传感器、成像条件及云量的影响会导致连续的长时序影像数据缺失或数据质量较差,同时影像的低时间分辨率也限制了对小区域连续时序FVC变化的研究。针对该问题,本文采用线性融合方法融合出连续的长时序FVC影像,解决了在研究FVC时空变化时云量和条带影响导致的Landsat影像连续时序数据缺失和低时间分辨率问题;利用Sen+Mann-Kendall进行趋势分析发现,金花茶自然保护区的FVC在2000—2016年整体呈增加趋势,FVC显著增加的区域约占37.32%,不显著增加的区域约占58.56%。线性融合方法得到的FVC影像可以精细地表征地表FVC的变化,较好地解决了高空间分辨率影像FVC连续时序数据缺失的限制,有利于小区域FVC的长时序时空变化研究。

When using high spatial resolution images to study FVC changes in a small area,due to the influence of sensors,imaging conditions,and cloud cover,continuous long time series image data is missing or unavailable.At the same time,the low time resolution of the image also greatly limits the study of continuous time series FVC changes in small areas.In response to this problem,this paper uses a linear fusion method to fuse continuous long time series FVC images,which solves the problem of the lack of continuous time series data and low time resolution of Landsat images due to cloud cover and banding when studying FVC spatiotemporal changes.And using Sen+Mann-Kendall trend analysis,it is found that the FVC of the Golden Camellia Natural Reserve showes an overall increase from 2000 to 2016.The area with a significant increase in FVC accounted for about 37.32%,and the area with no significant increase accounted for about 58.56%.The FVC image obtained by the linear fusion method can finely characterize the changes of the surface FVC,and better solve the limitation of the lack of continuous time series data of the high spatial resolution FVC image,which is conducive to the study of long time series spatiotemporal changes of FVC in small areas.