Optional Kriging spatial interpolation rainfall estimation method

Abstract

The invention discloses an optional Kriging spatial interpolation rainfall estimation method which comprises the following steps: 1, calculating experimental variation function values of all rainfallsample point pairs through a discrete variation function formula; 2, fitting the experimental variation function values by adopting a two-scale wavelet least square support vector machine to obtain atheoretical variation function model; 3, establishing a Kriging spatial interpolation equation set, and solving a Kriging weight coefficient; 4, calculating a rainfall estimated value of the to-be-estimated position point according to the Kriging weight coefficient; and 5, quantitative precision evaluation. According to the estimation method disclosed by the invention, the defects that the shape of a traditional theoretical variation function model is fixed and the spatial variation trend and multi-scale characteristics are not considered in a common Kriging spatial interpolation rainfall estimation method are overcome, so that a rainfall estimation result conforms to the own spatial variation trend and reflects the multi-scale characteristics of spatial variation.

Description

technical field [0001] The invention belongs to the field of spatial interpolation rainfall estimation, in particular to an optional kriging spatial interpolation rainfall estimation method in this field. Background technique [0002] As an input item of regional water resources, rainfall is very important for the assessment of regional water resources. At present, the estimation of regional rainfall is mainly through the means of site observation, and then the spatial interpolation model is used to estimate the rainfall in the region. The spatial interpolation model mentioned here is to estimate the rainfall information of the location to be estimated based on the actual rainfall information of the known observation sample points. And spatial correlation, so as to estimate the rainfall of other arbitrary points or arbitrary divisions. [0003] Commonly used spatial interpolation methods can be classified into two categories, one is deterministic methods and the other is s...

AI Technical Summary

Problems solved by technology

In the traditional method, the interpolation model needs to use the theoretical variogram to fit the experimental variogram, and the theoretical variogram model is often selected based on human experience, which will lead to different theoretical variogram models on the quality of rainfall estimation results. greater impact

In addition, the shape of the theoretical variogram is fixed. When fitting the limited data of known rainfall stations, it cannot reflect the spatial correlation and difference of actual rainfall. The spatial variation trend of rainfall is submerged, and the submerged A spatial change trend often has multi-scale characteristics, which is also crucial for the selection of scales and the treatment of scale effects.

Images