摘要
Abstract
Existing single-sensor systems cannot provide long-term,high-spatial-resolution normalized difference vegetation index(NDVI)data products.Based on a data reconstruction framework,we adopted the flexible spatiotemporal data fusion(FSDAF)method to reconstruct NDVI datasets for Tumet Right Banner.Combined with the pixel dichotomy model,fractional vegetation cover(FVC)datasets were estimated,and vegetation changes were further analyzed using regression trend and correlation analyses.The results show that:1)The reconstructed NDVI data align more closely with those of actual NDVI observations.The R2 of NDVI was 0.637 under data-missing conditions,which increased to 0.934 after FSDAF reconstruction,representing an improvement of 0.297.2)Vegetation cover in Tumet Right Banner exhibits a spatial pattern of"low in the north,high in the south".While the overall vegetation improvement trend was weak,significant enhancement was observed north of the Beijing-Xizang Expressway.In contrast,the northern part of the study area showed no clear improvement,whereas significant degradation occurred in the northwestern region,plains with human settlements,and along the margins of the Yellow River.3)Precipitation showed a weak and spatially heterogeneous positive correlation with vegetation cover across the study area,with distinct differences between northern and southern regions.Temperature was generally positively correlated with vegetation cover,displaying a strong positive correlation in the north but a negative correlation in the south.Vegetation near the Yellow River exhibited negative and positive correlations with precipitation and temperature,respectively.The findings provide a novel perspective for data selection in vegetation studies,enabling research based on more accurate vegetation cover information.Furthermore,the results offer decision-making support for ecological conservation and sustainable development in Tumet Right Banner.关键词
土默特右旗区/时空融合/NDVI重建/高空间分辨率/归一化植被指数/植被覆盖/植被变化Key words
Tumet Right Banner/spatiotemporal fusion/NDVI reconstruction/high spatial resolution/normalized difference vegetation index(NDVI)/vegetation cover/vegetation change
