A Deep Spatio-Temporal Forecasting Model for Multi-Site Weather Prediction Post-Processing

A Deep Spatio-Temporal Forecasting Model for Multi-Site Weather Prediction Post-Processing

Year:    2022

Author:    Wenjia Kong, Haochen Li, Chen Yu, Jiangjiang Xia, Yanyan Kang, Pingwen Zhang

Communications in Computational Physics, Vol. 31 (2022), Iss. 1 : pp. 131–153

Abstract

In this paper, we propose a deep spatio-temporal forecasting model (DeepSTF) for multi-site weather prediction post-processing by using both temporal and spatial information. In our proposed framework, the spatio-temporal information is modeled by a CNN (convolutional neural network) module and an encoder-decoder structure with the attention mechanism. The novelty of our work lies in that our model takes full account of temporal and spatial characteristics and obtain forecasts of multiple meteorological stations simultaneously by using the same framework. We apply the DeepSTF model to short-term weather prediction at 226 meteorological stations in Beijing. It significantly improves the short-term forecasts compared to other widely-used benchmark models including the Model Output Statistics method. In order to evaluate the uncertainty of the model parameters, we estimate the confidence intervals by bootstrapping. The results show that the prediction accuracy of the DeepSTF model has strong stability. Finally, we evaluate the impact of seasonal changes and topographical differences on the accuracy of the model predictions. The results indicate that our proposed model has high prediction accuracy.

You do not have full access to this article.

Already a Subscriber? Sign in as an individual or via your institution

Journal Article Details

Publisher Name:    Global Science Press

Language:    English

DOI:    https://doi.org/10.4208/cicp.OA-2020-0158

Communications in Computational Physics, Vol. 31 (2022), Iss. 1 : pp. 131–153

Published online:    2022-01

AMS Subject Headings:    Global Science Press

Copyright:    COPYRIGHT: © Global Science Press

Pages:    23

Keywords:    Weather forecasting post-processing spatio-temporal modeling deep learning.

Author Details

Wenjia Kong

Haochen Li

Chen Yu

Jiangjiang Xia

Yanyan Kang

Pingwen Zhang

  1. Forecasting daily extreme temperatures in Chinese representative cities using artificial intelligence models

    An, Hongyu | Li, Qinglan | Lv, Xinyan | Li, Guangxin | Qian, Qifeng | Zhou, Guanbo | Nie, Gaozhen | Zhang, Lijie | Zhu, Linwei

    Weather and Climate Extremes, Vol. 42 (2023), Iss. P.100621

    https://doi.org/10.1016/j.wace.2023.100621 [Citations: 3]
  2. Computational Science and Its Applications – ICCSA 2023

    Urban Resilience Key Metrics Thinking and Computing Using 3D Spatio-Temporal Forecasting Algorithms

    Agbossou, Igor

    2023

    https://doi.org/10.1007/978-3-031-36808-0_22 [Citations: 0]
  3. Urban hotspot forecasting via automated spatio-temporal information fusion

    Jin, Guangyin | Sha, Hengyu | Xi, Zhexu | Huang, Jincai

    Applied Soft Computing, Vol. 136 (2023), Iss. P.110087

    https://doi.org/10.1016/j.asoc.2023.110087 [Citations: 12]
  4. Multi-Site Prediction Method for Atmospheric Pollutants Based on an Improved MLP-Mixer

    Tao, Tao | Zhang, Sihan | Bao, Han

    2024 International Applied Computational Electromagnetics Society Symposium (ACES-China), (2024), P.1

    https://doi.org/10.1109/ACES-China62474.2024.10699876 [Citations: 0]
  5. Literature Review on Integrating Generalized Space-Time Autoregressive Integrated Moving Average (GSTARIMA) and Deep Neural Networks in Machine Learning for Climate Forecasting

    Munandar, Devi | Ruchjana, Budi Nurani | Abdullah, Atje Setiawan | Pardede, Hilman Ferdinandus

    Mathematics, Vol. 11 (2023), Iss. 13 P.2975

    https://doi.org/10.3390/math11132975 [Citations: 2]
  6. Short-term wind speed forecasting bias correction in the Hangzhou area of China based on a machine learning model

    Fang, Yi | Wu, Yunfei | Wu, Fengmin | Yan, Yan | Liu, Qi | Liu, Nian | Xia, Jiangjiang

    Atmospheric and Oceanic Science Letters, Vol. 16 (2023), Iss. 4 P.100339

    https://doi.org/10.1016/j.aosl.2023.100339 [Citations: 6]
  7. Data-driven techniques for temperature data prediction: big data analytics approach

    Oloyede, Adamson | Ozuomba, Simeon | Asuquo, Philip | Olatomiwa, Lanre | Longe, Omowunmi Mary

    Environmental Monitoring and Assessment, Vol. 195 (2023), Iss. 2

    https://doi.org/10.1007/s10661-023-10961-z [Citations: 2]
  8. Extreme Weather Prediction for Highways Based on LSTM-CNN

    Yang, Jun | Cui, Jiajia | Huang, Ming

    2023 5th International Conference on Applied Machine Learning (ICAML), (2023), P.191

    https://doi.org/10.1109/ICAML60083.2023.00045 [Citations: 0]
  9. Site-Specific Deterministic Temperature and Dew Point Forecasts with Explainable and Reliable Machine Learning

    Han, Mengmeng | Leeuwenburg, Tennessee | Murphy, Brad

    Applied Sciences, Vol. 14 (2024), Iss. 14 P.6314

    https://doi.org/10.3390/app14146314 [Citations: 0]
  10. A deep learning network for improving predictions of maximum and minimum temperatures over complex terrain

    Xu, Lin | Zhou, Li | Chen, He | Cai, Rong-Hui | Zhou, Yue

    Theoretical and Applied Climatology, Vol. 155 (2024), Iss. 7 P.6949

    https://doi.org/10.1007/s00704-024-04901-1 [Citations: 0]
  11. Spatiotemporal forecasting model based on hybrid convolution for local weather prediction post-processing

    Xiang, Li | Xiang, Jie | Guan, Jiping | Zhang, Lifeng | Cao, Zenghui | Xia, Jilu

    Frontiers in Earth Science, Vol. 10 (2022), Iss.

    https://doi.org/10.3389/feart.2022.978942 [Citations: 0]
  12. ECMWF short-term prediction accuracy improvement by deep learning

    Frnda, Jaroslav | Durica, Marek | Rozhon, Jan | Vojtekova, Maria | Nedoma, Jan | Martinek, Radek

    Scientific Reports, Vol. 12 (2022), Iss. 1

    https://doi.org/10.1038/s41598-022-11936-9 [Citations: 32]