Article Details

Abstract

Based on the daily meteorological observation data of the national-level meteorological stations in Xing'an League area in 2023, this paper conducts a comparative verification with the national-level intelligent grid forecast products issued by the Central Meteorological Observatory, and explores and summarizes the correction method by using Machine Learning method, then comes to conclusions. Combining the results of the CMA-GFS numerical forecast model and various ground observation facts, a temperature product correction model is established by an integrated learning method. The model can achieve excellent correction effects on both the highest and lowest temperatures, and the accuracy rate increases and the error is reduced significantly after correction. This correction method has good research value and practical application significance.

Keywords

intelligent grid; forecast verification; numerical forecast correction; Machine Learning

References

1. [1] Wei Qing, Journal, Lin Jian, et al. Evaluation and Verification of National Intelligent Grid Precipitation and Temperature Forecast from 2016 to 2018 [J]. Meteorology, 2020, 46 (10): 1272-1285
2. [2] Liu Haiyan, Zhou Xueying, Li Xianfeng, etc Preliminary exploration of intelligent grid meteorological forecasting business product service model based on data lake - taking Jiangxi Province as an example [J]. Meteorology and Disaster Reduction Research, 2022, 45 (1): 64-69
3. [3] Jia Xiaohong, Wei Wei, Shi Lan Comparative Study on Statistical Forecasting Methods for Winter Road Surface Minimum Temperature [J]. Inner Mongolia Meteorology, 2024 (1): 25-30
4. [4] Hang Yuehe, Sun Xin Application of Support Vector Machine Method in Objective Temperature Forecasting in Inner Mongolia [J]. Inner Mongolia Meteorology, 2020 (3): 16-19
5. [5] Jin Ronghua, Daikan, Zhao Ruixia, etc Progress and Challenges of Seamless Fine Grid Weather Forecasting Technology in China [J]. Meteorology, 2019, 5 (4): 445-457
6. [6] Sun Kanghui, Xiao An, Xiahou Jie Research on Short term Temperature Forecasting Model in Jiangxi Province Based on LightGBM Machine Learning Algorithm [J]. Plateau Meteorology, 2024, 43 (6): 1520-1535
7. [7] Liu Jie, Liu Gaoping, An Jingjing, etc Machine Learning based Model Temperature Prediction Correction Method [J]. Desert and Oasis Meteorology, 2024, 18 (3): 96-104
8. [8] Zhu Yuxiang, Liu Haiwen, Wan Wenlong, etc Progress and Future Trends of Artificial Intelligence Application in Aircraft Turbulence Prediction [J]. Journal of Atmospheric Sciences, 2023, 46 (6): 825-836
9. [9] Wu Lue, Jiao Ruili, Wang Yi, etc Research on the Correction Method of Perceived Temperature Error in Beijing Area [J]. Meteorological Science, 2022, 42 (2): 261-269
10. [10] Zeng Xiaoqing, Cao Yong, Wang Yu, etc A new refined interpolation method for model forecast fields [J]. Atmospheric Science, 2024, 48 (5): 1879-1890

Funding

Inner Mongolia Autonomous Region Meteorological Bureau Science and Technology Innovation Project (nmqxkjcx02412)

About the authors

Li Xuexue (February 1993-), female, Mongolian, from Tuquan, Inner Mongolia, engineer, master's degree in agriculture, research direction: weather forecasting and meteorological services research.