STARFM is developed for blending Landsat and MODIS surface reflectance by fusing high-frequency temporal information from MODIS and high spatial resolution information from Landsat data.STARFM predicts pixel values based upon a spatially weighted difference computed between the Landsat and the MODIS data acquired at T1, and the Landsat T1-scene and one or more MODIS scenes of prediction day (T2), respectively . A moving window technique is used to minimize the effect of pixel outliers thereby predicting changes of the center pixel using the spatially and spectrally weighted mean difference of pixels with in the window area . The STARFM was also extended for blending NDVI data of different spatial and temporal resolutions to produce high temporal and spatial resolution NDVI dataset, and satisfactory results were achieved. Therefore,Landsat ETM+ NDVI data acquired on 18 August, 2005 which was scaled to 0–10,000 was assigned as the Landsat T1-scene data, andthe same scaled and spatial resampled MODIS NDVI data acquiredon 13 August, 2005 which was nearest to ETM+ data was assignedas the MODIS T1 data. The same scaled and spatial resampled MODIS NDVI data acquired at other date were used for prediction of Landsat-like NDVI data. Finally, the 30 m spatial resolution 16-day temporal resolution fused time series NDVI data was generated for further analysis.