喀斯特高原区多源遥感数据时空融合模型适用性分析

Applicability Analysis of Spatio-temporal Fusion Model of Multi-source Remote Sensing Data in Karst Plateau

喀斯特石漠化综合治理防治措施已落实到小流域等较小空间单元内,迫切需要高时空分辨率植被覆盖度等数据支撑相关研究。由于技术和预算的限制,单一传感器难以获取同时满足高空间、高时间分辨率的数据,时空融合技术是目前解决遥感数据缺失和“时空矛盾”的重要方法之一。以Landsat8 OLI数据与MODIS数据为数据源,以喀斯特高原20 km×20 km区域为实验区,采用时空自适应反射融合模型(spatial and temporal adaptive rcflectance fusion model,STARFM)、增强型时空自适应反射融合模型(enhanced spatial and temporal adaptive rcflectance fusion model,ESTARFM)、灵活的时空数据融合模型(flexible spatiotemporal data fusion,FSDAF)模型三种模型融合生成高时空分辨率数据,分析三种模型在喀斯特高原区的应用能力。结果表明:STARFM、ESTARFM、FSDAF三种模型的融合影像与真实影像的R均高于0.6,ESTARFM模型的融合影像与真实影像的相关性最高,空间细节最为清晰,层次性更明显;ESTARFM模型在地表破碎,异质性较高的喀斯特高原区具有较好的适用能力。

The prevention and control measures for the comprehensive control of Karst rocky desertification have been implemented in smaller spatial units such as small watersheds.Data support research such as high spatial-temporal resolution vegetation coverage is urgently needed.Due to technical and budgetary constraints,it is difficult for a single sensor acquire data with high spatial and high temporal resolution.Spatial-temporal fusion technology is one of the most important methods to solve the problem of remote sensing data loss and temporal and spatial contradiction.Landsat8 OLI data and MODIS data were used as data sources.The 20 km×20 km area of the Karst plateau was used as the experimental area.The three models of spatial and temporal adaptive rcflectance fusion model(STARFM),enhanced spatial and temporal adaptive rcflectance fusion model(ESTARFM),and flexible spatiotemporal data fusion(FSDAF)were used to generate 30 m MODIS temporal resolution data,and the application capacity of three models in the Karst plateau area were analyzed.The results show that the correlation between the fusion images of STARFM,ESTARFM and FSDAF is higher than 0.6,and the fusion image of ESTARFM model has the highest correlation with real images,the spatial details are the clearest and the level is more obvious.ESTARFM has good applicability in the Karst plateau where the surface is broken and the heterogeneity is high.