A Correction Method of Rainfall Deviation Based on Segmentation Thought

Abstract

The invention discloses a method for correcting rainfall deviation based on the idea of ​​segmentation, aiming at improving the rainfall simulation accuracy of a series of rainfall models such as climate model, downscale model and rainfall generation model. The method includes: 1) dividing the measured rainfall sequence into three parts: minimum value, normal value and maximum value based on the quantile of the measured rainfall sequence; 2) dividing the simulated rainfall sequence into three parts based on the quantile of the simulated rainfall sequence It is also divided into three parts: minimum value, normal value and maximum value; 3) based on the three-segment data of the measured rainfall sequence, the deviation correction is performed on the three-segment data of the simulated rainfall sequence; 4) a newly proposed evaluation index is adopted , to effectively evaluate the effect of rainfall bias correction. The beneficial effects of the present invention are: effectively eliminating the error between the simulated rainfall of the rainfall model and the actual measured rainfall, improving the rainfall simulation accuracy of the rainfall model; laying a solid foundation for the correction of the deviation of the rainfall model's future prediction of rainfall, and improving the reliability of its prediction sex.

Description

technical field [0001] The invention relates to the technical field of deviation correction of rainfall model simulation rainfall, in particular to a rainfall deviation correction method based on segmentation idea. Background technique [0002] Rainfall data is one of the most important factors in the hydrological cycle, and the accuracy of its simulation directly affects the accuracy of runoff simulation and engineering design. Rainfall data simulation plays a vital role in studying the impact of climate change on hydrology and water resources and solving data bottlenecks in areas with no data. In particular, in recent years, with the increase of global temperature and the intensification of climate change, the spatial and temporal distribution of rainfall will inevitably be affected. [0003] At present, the tools for simulating rainfall include a series of models such as global or regional climate models (such as GCM or RCM), downscaling models and weather generators. A...

AI Technical Summary

Problems solved by technology

[0005] In order to solve the deficiencies in the prior art, the object of the present invention is to provide a method for correcting rainfall deviation based on the concept of segmentation, which divides both the measured rainfall sequence and the simulated rainfall sequence into segments, and separates the rainfall data of different segments The deviation correction can not only correct the deviation from the aspects of rainfall mean value, variance, quantile point and distribution, but also improve the phenomenon that the correction effect of the rainfall extreme value deviation is not good, thereby improving the accuracy of the rainfall model to simulate rainfall

Images