Small watershed surface rainfall interpolation method based on mountain terrain and rainfall characteristic prediction

Abstract

The invention relates to a small watershed surface rainfall interpolation method based on mountain terrain and rainfall characteristic prediction. The method is based on national Doppler radar body scanning base data products and jigsaw products. Doppler radar strong echo block space-time vector characteristics and shape characteristics are extracted; establishing a space-time sequence predictionmodel by utilizing a 3D convolutional neural network; extracting rainfall block space-time distribution, a movement prediction result and a historical movement track; based on high-precision landformdata, a rainfall block prediction motion vector is taken as an advancing direction, and a fuzzy matching method is adopted to match mountainous area landform features with rainfall space-time distribution and a motion prediction result, so that multidimensional factors influencing rainfall are judged, and a surface rainfall interpolation factor library is established. And establishing an interpolation algorithm library based on a deterministic spatial interpolation method and a geostatistics method, linking the interpolation algorithm library to the interpolation factor library in real time, and dynamically selecting an optimal interpolation algorithm model and a parameter combination under a corresponding rainfall situation, thereby obtaining an optimal interpolation result of the small-watershed surface rainfall of the mountainous area throughout the country.

Description

technical field [0001] The invention relates to the fields of rainfall spatial interpolation in small watersheds in mountainous areas and area rainfall calculation in small watersheds, specifically a small watershed area rainfall interpolation method based on high-precision topography and rainfall characteristics prediction in mountainous areas, which is mainly used for high-precision in mountainous areas with complex terrain Acquisition of areal rainfall in small watersheds enables accurate early warning and forecasting of mountain torrent disasters in complex terrain areas. Background technique [0002] Affected by various factors such as special natural and geographical environment, extreme disastrous weather, and economic and social activities, mountain torrent disasters occur frequently in my country, and mass death and mass injury incidents occur from time to time. According to statistics, before the 1990s, the annual death toll from mountain torrent disasters in the c...

AI Technical Summary

Problems solved by technology

At present, the applicability of various single rainfall interpolation algorithms is not strong. In the face of different basic environmental conditions and rainfall spatio-temporal characteristics, variation extremes and errors often appear, the application range is not wide, and it is only applicable to a certain scene or fixed area. It cannot be popularized and applied on a large scale. In practical applications, it is difficult to provide assistance for early warning and decision-making of mountain torrent disasters

Images