Fig 5 - available via license: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International
Content may be subject to copyright.
![IEEE Big Data Weather4Cast 2021 Data Localization. The core challenge is shown in blue squares, while the regions in orange squares are for the transfer learning challenge [89].](publication/374721152/figure/fig2/AS:11431281198400384@1697293127971/IEEE-Big-Data-Weather4Cast-2021-Data-Localization-The-core-challenge-is-shown-in-blue.jpg)
IEEE Big Data Weather4Cast 2021 Data Localization. The core challenge is shown in blue squares, while the regions in orange squares are for the transfer learning challenge [89].
Source publication
Earth Observation is a growing research area that can capitalize on the powers of AI for short time forecasting, a now-casting scenario. In this work, we tackle the challenge of weather forecasting using a video transformer network. Vision transformer architectures have been explored in various applications, with major constraints being the computa...
Context in source publication
Context 1
... proposed system was tested on the challenging Traffic4cast 2021 [88], [89] weather dataset. The dataset used was part of the the IEEE BigData Conference competition for weather movie snippet forecasting. As shown in Fig. 5, the dataset covers 11 regions including: Fig. 4. Swin Transformer decoder block: A decoder stacking of two concurrent cross-attention blocks is shown, with shifted-window attention always coming after non-shifted ...
Similar publications
HIRAS-II is the hyperspectral detector carried on FengYun-3E which is the world’s first meteorological satellite in dawn–dusk orbit. It fills the observation gaps during the dawn and dusk periods of polar orbit meteorological satellites, enabling a 100% global data coverage and assimilation of polar orbit satellite data within each 6 h window for n...
Citations
... To capture evolutionary dynamics, temporal relations are encoded via a recurrent neural network (RNN) within each tier. Spatial patterns are extracted through a Swin Transformer, as in [49,50], to leverage sufficient spatial context for longer-term predictions. Details of the architecture are provided in the following subsections. ...
... Though convolution is usually coupled with pooling in a hierarchical structure (like pyramids), it is still challenging to model global and long-range information. Inspired by Liu [49,50], a Swin Transformerbased module is introduced in STPF-Net to capture both local patterns and global, longrange spatial dependencies within the radar data with great computational efficiency. The architecture of the Swin Transformer module is illustrated in Figure 3. Assuming that the resolution of the feature map is C × H × W, in which C is the channel, H is the height, and W is the width. ...
Accurate and timely precipitation forecasts are critical in modern society, influencing both economic activity and daily life. While deep learning methods leveraging remotely sensed radar data have become prevalent for precipitation nowcasting, longer-term forecasting remains challenging. This is due to accumulated errors in deep learning models and insufficient information about precipitation systems over longer time horizons. To address these challenges, we introduce the Short-Term Precipitation Forecast Network (STPF-Net), a recurrent neural network designed for longer-term precipitation prediction. STPF-Net uses a multi-tier structure with varying temporal resolutions to mitigate the accumulated errors during longer forecasts. Additionally, its transformer-based module incorporates larger spatial contexts, providing more complete information about precipitation systems. We evaluated STPF-Net on radar data from southeastern China, training separate models for 6 and 12 h forecasts. Quantitative results demonstrate STPF-Net achieved superior accuracy and lower errors compared to benchmark deep learning and numerical weather prediction models. Visualized case studies indicate reasonably coherent 6 h predictions from STPF-Net versus other methods. For 12 h forecasts, while STPF-Net outperformed other models, it still struggled with storm initiation over longer forecasting time.
