Retrieval method and device for water vapor profile

Abstract

The invention discloses a retrieval method and device for a water vapor profile.The method comprises the steps of obtaining mode data from an NCEP and hyperspectral satellite data from a CrIS; according to the mode data and the hyperspectral satellite data, searching for an experience orthogonality regression coefficient matched with the hyperspectral satellite data in a database, wherein the database comprises one or more experience orthogonality regression coefficients; according to the hyperspectral satellite data and the experience regression coefficient, conducting retrieval, so that the water vapor profile is obtained, wherein the simulation radiation value of a training sample and an experience orthogonality expansion matrix of the water vapor sample profile are fitted through a least square method, calculation is conducted, and the experience orthogonality regression coefficient is obtained.By means of the retrieval method and device, the water vapor profile can be obtained stably and rapidly through hyperspectral data retrieval, an important technical means is provided for monitoring water vapor distribution, the method and device are applicable to service application, and data product support is further provided for more deeply understanding and quantifying the feedback effect of water vapor on a weather system and a climate system.

Description

technical field [0001] The invention relates to the field of water vapor profile inversion of satellite remote sensing data, in particular to a water vapor profile inversion method and device. Background technique [0002] Water vapor is a very important trace gas in the Earth's atmosphere, and its distribution varies greatly on time and space scales. Water vapor plays an important role in the global radiation balance and atmospheric energy transport mechanism as well as photochemical processes, and its ability to absorb and emit radiation throughout the long-wavelength range makes it an important greenhouse gas in the atmosphere that regulates the Earth's climate system. The water vapor content in the upper troposphere is relatively small. According to model studies, the water vapor in the stratosphere also plays an important role in the earth's radiation balance. It not only absorbs solar short-wave radiation, but also emits long-wave radiation outward, which is an importa...

AI Technical Summary

Problems solved by technology

[0011] 1. The physical inversion method directly solves the atmospheric radiation transfer equation, and the result is unstable. The inversion accuracy depends on the selected initial profile accuracy and the establishment method of the prior covariance matrix, and the inversion process is time-consuming and not suitable for business applications.

[0012] 2. The inversion results of the neural network method are very sensitive to the selection of training data

Among them, the number of neurons in the hidden layer is crucial to the normal work of the entire neural network. If there are too many, new models cannot be accepted, and it will consume a lot of time and memory due to sample training. This overfitting phenomenon will As a result, the adaptability of the entire network deteriorates and the inversion accuracy decreases

Images