Submitted as: model experiment description paper
08 Mar 2022
Submitted as: model experiment description paper | 08 Mar 2022
Status: this preprint is currently under review for the journal GMD.

Temperature forecasting by deep learning methods

Bing Gong, Michael Langguth, Yan Ji, Amirpasha Mozaffari, Scarlet Stadtler, Karim Mache, and Martin G. Schultz Bing Gong et al.
  • Jülich Supercomputing Centre, Forschungszentrum Jülich, 52425 Jülich, Germany

Abstract. Numerical weather prediction (NWP) models solve a system of partial differential equations based on physical laws to forecast the future state of the atmosphere. These models are deployed operationally, but they are computationally very expensive. Recently, the potential of deep neural networks to generate bespoken weather forecasts has been explored in a couple of scientific studies inspired by the success of video frame prediction models in computer vision. In this study, a simple recurrent neural network with convolutional filters, called ConvLSTM, and an advanced generative network, the Stochastic Adversarial Video Prediction (SAVP) model, are applied to create hourly forecasts of the 2 m temperature for the next 12 hours over Europe. We make use of 13 years of data from the ERA5 reanalysis, of which 11 years are utilized for training and one year each is used for validating and testing. We choose the 2 m temperature, total cloud cover and the 850 hPa temperature as predictors and show that both models attain predictive skill by outperforming persistence forecasts. SAVP is superior to ConvLSTM in terms of several evaluation metrics, confirming previous results from computer vision that larger, more complex networks are better suited to learn complex features and to generate better predictions. The 12-hour forecasts of SAVP attain a mean squared error (MSE) of about 2.3 K2, an anomaly correlation coefficient (ACC) larger than 0.85, a Structural Similarity Index (SSIM) of around 0.72, and a gradient ratio (rG) of about 0.82. The ConvLSTM yields a higher MSE (3.6 K2), a smaller ACC (0.80), and SSIM (0.65), but a slightly larger rG (0.84). The superior performance of SAVP in terms of MSE, ACC, and SSIM can be largely attributed to the generator. A sensitivity study shows that a larger weight of the GAN component in the SAVP loss leads to even better preservation of spatial variability at the cost of a somewhat increased MSE (2.5 K2). Including the 850 hPa temperature as an additional predictor enhances the forecast quality and the model also benefits from a larger spatial domain. By contrast, adding the total cloud cover as predictor or reducing the amount of training data to eight years has only small effects. Although the temperature forecasts obtained in this way are still less powerful than contemporary NWP models, this study demonstrates that sophisticated deep neural networks may achieve considerable forecast quality beyond the nowcasting range in a purely data-driven way.

Bing Gong et al.

Status: final response (author comments only)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on gmd-2021-430', Anonymous Referee #1, 05 Apr 2022
  • CEC1: 'Comment on gmd-2021-430', Juan Antonio Añel, 21 Apr 2022
    • AC1: 'Reply on CEC1', BING GONG, 21 Apr 2022
  • CEC2: 'Comment on gmd-2021-430', Juan Antonio Añel, 21 Apr 2022
  • RC2: 'Comment on gmd-2021-430', Anonymous Referee #2, 05 May 2022

Bing Gong et al.

Bing Gong et al.


Total article views: 709 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
517 172 20 709 6 5
  • HTML: 517
  • PDF: 172
  • XML: 20
  • Total: 709
  • BibTeX: 6
  • EndNote: 5
Views and downloads (calculated since 08 Mar 2022)
Cumulative views and downloads (calculated since 08 Mar 2022)

Viewed (geographical distribution)

Total article views: 613 (including HTML, PDF, and XML) Thereof 613 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
Latest update: 20 May 2022
Short summary
Weather forecasts are important for many aspects of modern society. We explore video prediction based on deep learning methods for hourly 2 m temperature forecasts over Europe. We made a comparison between a simple and an advanced generative model. We confirm that larger networks are better suited for learning complex features from meteorological data, which in turn generates better predictions. Notably, the generative model component can help to better capture the spatial variability.