Surface air temperature quality control method based on spatial correlation and surface fitting

Abstract

The invention discloses a surface air temperature quality control method based on spatial correlation and surface fitting. The method includes the steps of collecting air temperature data of referencestations near a target station within a certain period of time, conducting basic quality control, determining the relative positions of the reference stations and the target station through the Euclidean distances and azimuth angles between the reference stations and the target station, establishing a spatial correlation function by means of the spatial correlation of air temperature elements ofeach reference station and the target station, conducting surface fitting on a test set through sample strips B to obtain a quality control model combining the spatial correlation and the surface fitting of the sample strips B, predicting the air temperature data of the target station through the test set, comparing a predicted value with an actually-observed value, implanting artificial errors into original data of the target station, and observing the error detection rate of the model. By means of the method, the surface meteorological data in China is effectively used, and a better prediction effect and error detection effect are achieved compared with a traditional spatial quality control algorithm.

Description

technical field [0001] The invention relates to a method for controlling the quality of surface temperature data, in particular to a method for controlling the quality of surface temperature data based on spatial correlation and B-spline surface fitting. Background technique [0002] Under the premise of continuous development of global climate change research, numerical weather prediction technology, and data assimilation technology, the quality control of surface meteorological observation data is the key link to ensure the high efficiency of data. In recent years, the number of surface meteorological observation stations has increased, and the resulting meteorological data has also become larger and larger. The quality requirements for surface meteorological observation data must also be higher and higher. [0003] The quality control of surface meteorological observation data can be divided into two situations, one is single station quality control, that is, the quality ...

AI Technical Summary

Problems solved by technology

Common multi-station quality control methods include Spatial Regression Test and Inverse Distance Weighting. However, from the point of view of the principle of the method, the IDW algorithm only considers the distance between each reference station and the target station, and completely ignores the distance between each reference station and the target station. The spatial correlation between sites and the spatial autocorrelation of the site itself are eliminated, so the prediction accuracy is greatly reduced; the advantage of the SRT algorithm over the IDW algorithm is that SRT selects the site with the smallest RMSE for weighted prediction, but still does not take into account the spatial factors of relevance

Images