The field of environmental and weather forecasting is rapidly advancing with the development of new deep learning models and techniques. Researchers are exploring the use of transformer-based models, such as the SSA-UNet and UNet with Axial Transformer, to improve the accuracy of precipitation nowcasting and weather forecasting. These models have shown significant promise in capturing complex patterns and dynamics in weather data, and have achieved state-of-the-art results in several benchmark datasets. Additionally, the use of latent diffusion models, such as Appa, is being investigated for global data assimilation and weather forecasting. These models have the potential to provide more accurate and efficient forecasts, and could have a significant impact on our ability to predict and prepare for severe weather events. Noteworthy papers include: STNet, which proposes a semi-transformer neural network for predicting underwater sound speed profiles, and Mjolnir, which presents a deep learning framework for global lightning flash density parameterization. RadioFormer is also noteworthy for its multiple-granularity transformer design for radio map estimation.
Advances in Environmental and Weather Forecasting
Sources
STNet: Prediction of Underwater Sound Speed Profiles with An Advanced Semi-Transformer Neural Network
SSA-UNet: Advanced Precipitation Nowcasting via Channel Shuffling
Discovering Governing Equations of Geomagnetic Storm Dynamics with Symbolic Regression
Appa: Bending Weather Dynamics with Latent Diffusion Models for Global Data Assimilation
Atlantes: A system of GPS transformers for global-scale real-time maritime intelligence
RadioFormer: A Multiple-Granularity Radio Map Estimation Transformer with 1\textpertenthousand Spatial Sampling
UNet with Axial Transformer : A Neural Weather Model for Precipitation Nowcasting
Mj\"olnir: A Deep Learning Parametrization Framework for Global Lightning Flash Density
Multidimensional precipitation index prediction based on CNN-LSTM hybrid framework
Autoencoder Models for Point Cloud Environmental Synthesis from WiFi Channel State Information: A Preliminary Study
Digital Shielding for Cross-Domain Wi-Fi Signal Adaptation using Relativistic Average Generative Adversarial Network
Temporal Attention Evolutional Graph Convolutional Network for Multivariate Time Series Forecasting
Enhancing Tropical Cyclone Path Forecasting with an Improved Transformer Network