Multi-source data assimilation fusion high temporal-spatial resolution leaf area index estimation method

Abstract

The invention discloses a multi-source data assimilation fusion high temporal-spatial resolution leaf area index estimation method, which comprises the following steps of S1, performing leaf area index LAI inversion estimation based on machine learning; s2, constructing an LAI dynamic model based on a normalized difference vegetation index NDVI vegetation growth trend, and estimating an LAI background field in combination with the LAI at the initial moment; and S3, combining the LAI initial estimation of the background field with the LAI of remote sensing observation by adopting an ensemble Kalman filter ENKF model, and carrying out data assimilation to realize high-resolution time sequence LAI estimation. According to the high-temporal-spatial-resolution leaf area index estimation method based on multi-source data assimilation fusion, the vegetation growth trend is described through the high-temporal-spatial-resolution NDVI, the high-precision inversion capability of machine learning and the multi-source data fusion advantage of ENKF are combined, the temporal-spatial resolution and precision of an LAI product are effectively improved, and the method is suitable for large-scale popularization and application. And a technical support is provided for fine ecological system monitoring.

Images