Dynamic statistics objective quantitative climate prediction method and system

Abstract

The invention provides a dynamic statistics objective quantitative climate prediction method and system. The method comprises the following steps: receiving a correlation test of historical climate data and climate factors, and establishing a prediction factor set of each region; judging whether early-stage factors in the prediction factor set are abnormal or not through a preset judgment index, if the early-stage factors in the prediction factor set are abnormal, carrying out an abnormal factor correction scheme, and otherwise, carrying out an optimal multi-factor combination correction scheme; performing error prediction: selecting a year similar to the year to be predicted and a similar error according to the correction scheme, and carrying out regional integration to form a national mode prediction error; obtaining a national climate prediction result according to the national mode prediction error and the coupled circulation mode original prediction result; and performing prediction inspection: according to the climate actual condition of the year to be predicted, inspecting the prediction result by calculating the PS score and the horizon correlation coefficient. According tothe method, different correction schemes are selected in a targeted manner by judging whether the prediction factor set of each region is abnormal or not, and finally the effect of improving the climate prediction accuracy is achieved.

Description

technical field [0001] The invention belongs to the technical field of climate prediction, and in particular relates to a dynamic-statistical objective and quantitative climate prediction method and system. Background technique [0002] The main factors affecting short-term climate prediction come from two aspects: external forcing and atmospheric interior. Then, the physical basis support for making short-term climate prediction should also come from these two aspects. In general research, it is known that external forcing factors include the ocean, the East Asian summer monsoon , sea temperature, sea ice, soil moisture, etc., and the movement and institutional changes of the atmospheric internal circulation include the North Atlantic Oscillation, North Pacific Oscillation, Southern Oscillation, Arctic Oscillation, Western Pacific Subtropical High, etc. [0003] Prediction of precipitation trend distribution in summer in my country is one of the most important business cont...

AI Technical Summary

Problems solved by technology

[0005] Climate numerical model prediction is the development direction of short-term climate prediction technology. However, the current climate model has problems such as insufficient resolution and the physical process described cannot fully reflect climate evolution. It has not yet reached the level of application in operational prediction.

Statistical methods still account for a large proportion of my country's short-term climate prediction business, but they have the defect of ignoring the evolution of physical processes

Images