Temperature forecasting by deep learning methods
<p>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 netw...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Copernicus Publications
2022-12-01
|
| Series: | Geoscientific Model Development |
| Online Access: | https://gmd.copernicus.org/articles/15/8931/2022/gmd-15-8931-2022.pdf |
| Summary: | <p>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 bespoke 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 <span class="inline-formula">m</span> temperature
for the next 12 <span class="inline-formula">h</span> over Europe. We make use of 13 years of data from the ERA5 reanalysis, of which 11 years are utilized for training and
1 year each is used for validating and testing. We choose the 2 <span class="inline-formula">m</span> temperature, total cloud cover, and the 850 <span class="inline-formula">hPa</span> 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 <span class="inline-formula">h</span> forecasts of SAVP attain a mean squared error (MSE) of about 2.3 <span class="inline-formula">K<sup>2</sup></span>, an
anomaly correlation coefficient (ACC) larger than 0.85, a structural similarity index (SSIM) of around 0.72, and a gradient ratio (<span class="inline-formula"><i>r</i><sub>G</sub></span>) of
about 0.82. The ConvLSTM yields a higher MSE (3.6 <span class="inline-formula">K<sup>2</sup></span>), a smaller ACC (0.80) and SSIM (0.65), and a slightly larger <span class="inline-formula"><i>r</i><sub>G</sub></span>
(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 generative adversarial network (GAN) component in the SAVP loss leads to even better preservation of spatial variability at the cost of a somewhat increased MSE
(2.5 <span class="inline-formula">K<sup>2</sup></span>). Including the 850 <span class="inline-formula">hPa</span> 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 8 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.</p> |
|---|---|
| ISSN: | 1991-959X 1991-9603 |