Ground surface evapotranspiration data downscaling method based on multi-source data and deep learning

Abstract

The invention provides an evapotranspiration data downscaling method based on multi-source data and deep learning. The method comprises the steps of obtaining low-spatial-resolution satellite surface evapotranspiration data, low-spatial-resolution atmosphere reanalysis data and high-spatial-resolution satellite remote sensing data; carrying out data preprocessing; based on a built deep learning regression network, establishing a surface evapotranspiration inversion model; and performing downscaling inversion on high-spatial-resolution surface evapotranspiration through the surface evapotranspiration inversion model established on the low-spatial-resolution surface evapotranspiration. According to the method, the earth surface evapotranspiration inversion precision is improved by comprehensively considering earth surface evapotranspiration related influence factors. The nonlinear complex relation between remote sensing earth surface parameters and atmospheric data and earth surface evapotranspiration is deeply analyzed based on deep learning. The relation between the remote sensing earth surface parameters and atmospheric data and earth surface evapotranspiration is learned by adopting BN and a dynamic learning rate. The BN processing avoids the gradient disappearance problem, the training speed is greatly increased, and the dynamic learning rate enables the network to converge to the optimal solution better.

Description

technical field [0001] This application relates to a method for obtaining surface evapotranspiration data, specifically, a method based on multi-source data and deep learning, using low spatial resolution evapotranspiration data to obtain high spatial resolution evapotranspiration data through inversion and downscaling and its storage media. Background technique [0002] Surface evapotranspiration (Evapotransspiration, ET) refers to the process of water entering the atmosphere in a gaseous state, mainly including surface soil water evaporation, vegetation transpiration, and vegetation canopy interception and evaporation of precipitation. It is the main factor for evaluating regional surface energy, climate change and water balance. Indicators are an important part of ecological environment and water resources assessment. The evapotranspiration acquisition methods are divided into actual observation and remote sensing inversion. Traditional observation can only measure the e...

AI Technical Summary

Problems solved by technology

The adaptive spatio-temporal fusion method and its improved method assume that the temporal variation of surface parameters is linear, which makes it difficult to apply to the fusion of long-term surface parameters in complex areas, and the model structure is complex, the input data is strict, and the input is at least two periods surface parameters, reducing the degree of model application

In other studies, based on the spatial scale-invariant effect of surface parameters, the inversion model of surface parameters is first established at coarse resolution, and then the high-resolution independent variable of surface parameters is used as the input of the model, and the output is the high-resolution target surface parameters. However, most of the traditional machine learning methods are currently used to construct models, which are difficult to reflect the complex nonlinear relationship between surface parameters and independent variables. In addition, the input parameters of most current machine learning methods are mostly single remote sensing satellite data, which fail to consider Climate variables affecting surface evapotranspiration, such as wetlands, surface radiation, etc.

Images