Remote sensing dynamic monitoring of vegetation ecological water based on NDVI spatio-temporal data fusion model

Current remote sensing dynamic monitoring methods typically lead to insufficient inversion accuracy due to limited data source or data fusion errors. To address this challenge, this research proposes a remote sensing dynamic monitoring method of vegetation ecological water based on normalized difference vegetation index (NDVI) spatio-temporal data fusion model. More specifically, by collecting the remote sensing data of Landsat and MODIS, this research implements radiometric calibration and atmospheric correction on the Landsat data, as well as projection and Savitzky-Golay (S-G) filtering processing on the MODIS data. Furthermore, this research utilizes the constructed fusion model to realize the spatio-temporal fusion of the above two types of data. Corresponding spatio-temporal fusion results are employed to calculate the NDVI value of the survey region. Meanwhile, the calculation results of NDVI are adopted to conduct remote sensing inversion for vegetation ecological water. The results indicate that the vegetation canopy foliar moisture content (FMC) is at the highest level in summer. Among diverse vegetation types, broad-leaved forest presents the highest FMC and the lowest wildfire risk level, whereas meadow exhibits the lowest FMC and the highest wildfire risk level. Moreover, the FMC inversion results derived from the proposed method are consistent with the actual situation, which validates the high inversion accuracy of the proposed method.