Drought risk prediction method and system

Abstract

The invention provides a drought risk prediction method and system, and the method comprises the steps: obtaining a historical meteorological record text and a historical drought record text of each year of each local region of a research region, and building a keyword dictionary; determining the frequency of different levels of drought in each local area by adopting a co-word analysis method; determining the periodicity of different levels of drought in each local area by adopting a topic model analysis method; and finally, calculating the risk value of different levels of drought occurring in the prediction year of each local area of the research area. According to the invention, statistics of drought regularity and periodicity is carried out based on a text data mining method; the method is advantaged in that drought risk prediction of the research area is realized, because the prediction range is only related to the acquired historical text data and is not limited by the area, large-area drought risk prediction is realized, the prediction result is accurate to each local area of the research area, and acquired data is more targeted.

Description

technical field [0001] The present invention relates to the technical field of drought risk prediction, in particular to a drought risk prediction method and system. Background technique [0002] Drought disaster is one of the natural disasters. It has the characteristics of high frequency, long duration and wide range of influence. It usually causes serious losses. It also caused great damage. Drought monitoring and drought loss assessment are the premise and foundation of drought relief and drought management, and also one of the core contents of drought risk assessment. Research on major drought disaster risk assessment technology is of great significance for improving disaster prevention and rapid response to major droughts. It can provide scientific and technological support for major drought disaster management and regional sustainable development, and improve the scientific and technological support capabilities of major drought disasters. [0003] Existing risk ass...

AI Technical Summary

Problems solved by technology

[0005] However, the method of using meteorological observation data has the problem of sparse ground observation points, and the conventional station monitoring data can only represent the data of 30 square kilometers near the station, and there are no stations in many places, so it is difficult to quickly and timely obtain large-scale drought conditions information, making large-scale drought risk assessments timeless and unrepresentative

The data of monitoring stations also has the problems of missing data and monitoring errors, and it is difficult to obtain data

[0006] Remote sensing technology only developed rapidly in the 1960s, and it was only used in drought monitoring research in the 1980s. It is impossible to obtain data from longer periods before this technology appeared, so it is also impossible to analyze drought disasters on a larger scale interannual and spatial variations, and it is also difficult to acquire and process data

[0007] It can be seen that the existing methods based on meteorological observation data and remote sensing technology can only realize drought monitoring in the current state, but cannot realize large-scale drought risk monitoring and drought risk prediction.

Images