CN-121995542-A - Method for predicting abnormal rainfall of Jinjing in main summer flood season

CN121995542ACN 121995542 ACN121995542 ACN 121995542ACN-121995542-A

Abstract

The invention relates to the technical field of rainfall prediction, and particularly discloses a method for predicting abnormal rainfall in a main flood season in summer, which comprises the steps of constructing a rainfall circulation index in the main flood season in the lower half of July to the upper half of July in the JingJi region, calculating to obtain a rainfall circulation index value in a prediction period based on latitude wind field prediction data output in a main current numerical mode at home and abroad, quantitatively predicting the rainfall by the rainfall circulation index value in the prediction period, constructing a main flood season rainfall index in the lower half of July to the upper half of July in the JingJi region based on historical related rainfall data in the lower half of July to the upper half of July in the JingJi region, creating a unitary linear regression relation with global sea surface temperatures of 3-6 months each year, and obtaining a key sea area associated with the abnormal rainfall index in the main flood season in the current year, and carrying out qualitative rainfall prediction by taking the sea surface temperature of the key sea area as a key index.

Inventors

GAO KAILUN
CHEN LIJUAN
LUO SHANJUN
ZHAO SHAOSHUAI
LIANG NAN
WAN CHENGCHENG
Shang Yiwei
CHEN LIN
ZHANG ZHONGYUAN

Assignees

河南省科学院空天信息研究所
国家气候中心

Dates

Publication Date: 20260508
Application Date: 20251225

Claims (9)

1. The method for predicting the abnormal rainfall in the main summer flood season of Jinjing is characterized by comprising the following steps of: s100, constructing a rainfall circulation index in main flood period from the lower half of July to the upper half of July in the Jinjin Ji area; S110, carrying out standardized treatment on historical related precipitation data from the lower half month of July to the upper half month of July in the Jinjin Ji area; s120, performing experience orthogonal decomposition on the standardized precipitation data to obtain a first three space-time variation main mode after orthogonal decomposition, and calculating a Jing-dominant space-time mode by taking an interpretation variance of the first three space-time variation main mode as a weight; s130, constructing a unitary linear regression equation and performing unitary linear regression analysis by utilizing a time sequence of a Beijing Ji dominant space-time mode and a latitudinal wind field of a region between 100 degrees and 160 degrees of east longitude, and determining a region obviously related to precipitation in the lower half month of July to the upper half month of July in the Beijing Ji region; S140, determining the contribution weight of the latitudinal wind field of each significant relevant area to the precipitation circulation index by using an entropy weighting method, and creating a precipitation circulation index RSI formula from the lower half month of July to the upper half month of July in the Jingjin Ji area; s200, predicting abnormal rainfall intensity in main flood season from the lower half of July to the upper half of July in the Jinjin Ji area; S210, calculating to obtain a rainfall circulation index value in a prediction period based on weft wind field prediction data output in a main stream numerical mode at home and abroad, and quantitatively predicting rainfall by the rainfall circulation index value in the prediction period; S220, creating a unitary linear regression relation with global sea surface temperatures of 3-6 months in each year on the basis of precipitation indexes of the main flood season of the lower half-month to the upper half-month of July in the Jinjin Ji region constructed by historical related precipitation data of the lower half-month to the upper half-month of July in the Jinjin Ji region, acquiring a key sea area associated with abnormal precipitation indexes of the main flood season in the current year, and carrying out qualitative prediction on precipitation by taking the sea surface temperatures of the key sea area as key indexes; s230, comprehensively judging the precipitation conditions of the main flood season from the lower half month of July to the upper half month of July in the Jinjin Ji region according to the quantitative prediction and qualitative prediction results.
2. The method for predicting abnormal precipitation in main summer flood period of Jinjin Ji according to claim 1, wherein the historical related precipitation data in S110 comprises analysis data based on CMA station precipitation data and ERA5, and the historical related precipitation data is represented by formula And (3) performing standardization processing, wherein X i is the ith sample value in the historical related precipitation data sequence, mu is a sample mean value, sigma is a sample standard deviation, and X new is a standardized value.
3. The method for predicting precipitation anomalies in main summer flood season of Jinjin Ji as claimed in claim 1, wherein the empirical orthogonal decomposition in S120 is specifically to decompose a variable field containing p spatial points into different modes which change with time and space, namely to set a range-to-average observed value of any one spatial point i and any one time point j in the variable field From p spatial functions And a function of time The calculation formula of the Jing Ji dominant space-time mode is as follows: Wherein E0F 1 、E0F 2 、E0F 3 is the main mode of the first three space-time variation, w 1 、w 2 、w 3 is the weight coefficient corresponding to E0F 1 、E0F 2 、E0F 3 , and E0F m is the dominant space-time mode of the synthesized Jinjin.
4. The method for predicting the abnormal precipitation in the main summer season of Jinjin Ji as claimed in claim 1, wherein the air pressure layer in S130, which is obviously related to precipitation in the lower half of July to the upper half of July in Beijing and Tianjin Ji region, is determined to be 200hPa, the region in S130, which is obviously related to precipitation in the lower half of July to the upper half of July in Jinjin Ji region, comprises an R1 region, an R2 region and an R3 region, wherein the R1 region is 44 degrees to 59 degrees of north latitude, the R2 region is 27 degrees to 40 degrees of north latitude, the east longitude is 102 degrees to 140 degrees, the R3 region is 15 degrees to 25 degrees of north latitude, the east longitude is 102 degrees to 157 degrees, and the R4 region is 2 degrees to 10 degrees of north latitude, and the east longitude is 102 degrees to 157 degrees.
5. The method for predicting abnormal precipitation in main summer season of Jingjid according to claim 4, wherein the precipitation circulation index RSI in S140 is expressed by the following formula: , wherein avg R1 （U）、avg R2 （U）、avg R3 （U）、avg R4 (U) is the regional average value of the latitudinal wind field of the R1 region, the R2 region, the R3 region and the R4 region, and w R1 、w R2 、w R3 、w R4 is the weight coefficient of avg R1 （U）、avg R2 （U）、avg R3 （U）、avg R4 (U).
6. The method for predicting abnormal precipitation in main summer season of Jingjid according to claim 1, wherein the quantitative prediction in S210 comprises: S211, carrying out regional averaging on the predicted latitudinal wind data of each significant relevant region based on the 200hPa latitudinal wind field predicted data output by the main stream numerical modes at home and abroad to obtain a predicted period average latitudinal wind value of each significant relevant region; S212, substituting the average weft wind value of the prediction period of each significant relevant region into the constructed precipitation circulation index formula, and calculating to obtain a precipitation circulation index value RSI p of the prediction period; S213, incorporating RSI p into the RSI historical observation dataset, constructing a new sample set containing historical data and predicted values, carrying out standardization processing on the new sample set to obtain standardized predicted values RSI s , Wherein As an average value of the new sample set, Standard deviation for the new sample set; S214, taking the standard deviation of the new sample set As the determination threshold, the precipitation abnormality level is output according to the value range of RSI s .
7. The method for predicting abnormal precipitation in main summer season of Jingjid according to claim 6, wherein the specific judgment rule in S214 is as follows: If it is Predicting that the rainfall intensity in the main annual flood period is close to the average annual historical level; If it is The rainfall intensity in the main annual flood period is predicted to be slightly weaker than that in the history year after year; If it is The rainfall intensity in the main annual flood period is predicted to be abnormally weaker than that in the history perennial period; If it is The rainfall intensity in the main annual flood season is predicted to be slightly stronger than that in the history year by year; If it is The rainfall intensity in the main annual flood period is predicted to be stronger than that in the annual abnormal annual period.
8. The method for predicting abnormal precipitation in main summer season of Jingjid according to claim 1, wherein the qualitative prediction in S220 comprises: s221, performing grid division on global sea surface temperature; s222, carrying out standardization treatment on the time series of the sea surface temperature of 3-6 months of each grid, and synchronously carrying out standardization treatment on the corresponding time series of the precipitation circulation indexes; s223, for each grid of the global ocean, respectively constructing a unitary linear regression equation of a sea surface temperature standardized variable X std and a precipitation circulation index standardized variable Y std of each grid: where k is the grid number, As a constant term of the grid regression equation, Is a regression coefficient; s224, checking the regression coefficient of each grid, setting a significance level, and screening out grids with significant linear correlation between sea surface temperature and precipitation circulation indexes; s225, carrying out space aggregation on the grids with obvious correlation, and merging continuous high-correlation areas.
9. The method for predicting the abnormal precipitation in the main summer season of Jinjin Ji according to claim 8, wherein the method is characterized in that the method is obtained in S225 through space aggregation combination, wherein the obvious negative correlation exists between the sea surface temperature in the Indian region of 4-6 months and the precipitation circulation indexes from the lower half of July to the upper half of July in the Jinjin Ji region, and the obvious negative correlation exists between the sea surface temperature of Topacific in the equator of 3-5 months and the precipitation circulation indexes from the lower half of July to the upper half of July in the Jinjin Ji region.

Description

Method for predicting abnormal rainfall of Jinjing in main summer flood season Technical Field The invention relates to the technical field of rainfall prediction, in particular to a method for predicting abnormal rainfall in a main summer flood season of Jinjin Ji. Background From the climate point of view, the summer precipitation in Jingjin Ji area mainly appears in the first half of July to the first half of July, namely the so-called "seventh under eight top" main flood season, the significant annual change exists in the proportion of the precipitation amount to the total precipitation amount in summer, and the average value of the precipitation amount is about 50%. Meanwhile, the Jingjin Ji 'seven-eight-up' precipitation is accompanied with the characteristics of high process intensity, uneven space-time distribution, long duration and the like, thus being extremely easy to cause flood and induce secondary disasters such as mountain floods, landslides and the like and seriously threatening the life and property safety of people. The traditional North China rainy season monitoring and prediction is divided into two types, namely, the start date and the end date of the rainy season are monitored based on the station precipitation data statistics, and the start date of the North China rainy season is comprehensively judged and predicted according to key circulation factors such as the east Asia western wind speed shaft position, the western Pacific subsidiary tropical high-voltage ridge line position, the average 850hPa passing wind speed in North China region, the average 850hPa wetting in North China region and the like. The existing research and the index are focused on the judgment of the start and stop dates of the rainy season in the North China large area, and aiming at the defect of the rainfall index in the main flood season in the Ji region of Jinjin, the existing index is mainly used for monitoring, most of the current index cannot be combined with the numerical mode prediction output data to play a prediction effect, and although the existing research has revealed that the sea surface temperature abnormality in different regions of the world can be used for the prediction of the rainfall in different regions, the direct connection between the rainfall in the main flood season in the Ji region of Jinjin and the key region related to the global sea surface temperature is not obvious, and the sea surface temperature abnormality is rarely used as a basis for judging the main flood season abnormality in the Ji region of Jinjin. Disclosure of Invention The invention provides a method for predicting abnormal precipitation in main summer season of Jingjie, which aims at solving the problems in the related art. In order to achieve the technical purpose, the technical scheme of the invention is realized as follows: A method for predicting abnormal rainfall in the main summer season of Jinjing Ji comprises the following steps: s100, constructing a rainfall circulation index in main flood period from the lower half of July to the upper half of July in the Jinjin Ji area; S110, carrying out standardized treatment on historical related precipitation data from the lower half month of July to the upper half month of July in the Jinjin Ji area; s120, performing experience orthogonal decomposition on the standardized precipitation data to obtain a first three space-time variation main mode after orthogonal decomposition, and calculating a Jing-dominant space-time mode by taking an interpretation variance of the first three space-time variation main mode as a weight; s130, constructing a unitary linear regression equation and performing unitary linear regression analysis by utilizing a time sequence of a Beijing Ji dominant space-time mode and a latitudinal wind field of a region between 100 degrees and 160 degrees of east longitude, and determining a region obviously related to precipitation in the lower half month of July to the upper half month of July in the Beijing Ji region; S140, determining the contribution weight of the latitudinal wind field of each significant relevant area to the precipitation circulation index by using an entropy weighting method, and creating a precipitation circulation index RSI formula from the lower half month of July to the upper half month of July in the Jingjin Ji area; s200, predicting abnormal rainfall intensity in main flood season from the lower half of July to the upper half of July in the Jinjin Ji area; S210, calculating to obtain a rainfall circulation index value in a prediction period based on weft wind field prediction data output in a main stream numerical mode at home and abroad, and quantitatively predicting rainfall by the rainfall circulation index value in the prediction period; S220, creating a unitary linear regression relation with global sea surface temperatures of 3-6 months in each year on the basis of precipitation indexes of the main flood season of