Transforming Weather Data from Pixel to Latent Space

The increasing impact of climate change and extreme weather events has spurred growing interest in deep learning for weather research. However, existing studies often rely on weather data in pixel space, which presents several challenges such as smooth outputs in model outputs, limited applicability to a single pressure-variable subset (PVS), and high data storage and computational costs. To address these challenges, we propose a novel Weather Latent Autoencoder (WLA) that transforms weather data from pixel space to latent space, enabling efficient data representation. By decoupling weather reconstruction from downstream tasks, WLA improves the accuracy and sharpness of weather task model results. The incorporated Pressure-Variable Unified Module transforms multiple PVS into a unified representation, enhancing the adaptability of the model in multiple weather scenarios. Furthermore, weather tasks can be performed in a low-storage latent space of WLA rather than a high-storage pixel space, thus significantly reducing data storage and computational costs. Through extensive experimentation, we demonstrate its superior compression and reconstruction performance, enabling the creation of the ERA5-Latent dataset with unified representations of multiple PVS from ERA5 data. The compressed full PVS in the ERA5-Latent dataset reduces the original 244.34 TB of data to 0.43 TB. The downstream task further demonstrates that task models can apply to multiple PVS with low data costs in latent space and achieve superior performance compared to models in pixel space.

As climate change leads to more extreme weather, scientists are increasingly using deep learning to improve weather forecasting. However, standard AI models process raw, high-resolution weather data directly, which creates significant hurdles. These models often produce blurry forecasts that miss localized extreme events, are limited to analyzing specific combinations of weather variables, and require massive data computing and storage resource. To address these challenges, we introduce a new framework that compresses vast, complex weather data into a highly efficient, condensed digital format. By handling data compression separately from forecasting tasks, our approach helps AI models generate sharper, more accurate predictions for severe weather. Furthermore, it standardizes the representation of different atmospheric variable subsets, making it adaptable to many different weather scenarios. Crucially, this method drastically reduces data computing and storage demands. We used it to shrink a massive global weather dataset from 244.34 terabytes to just 0.43 terabytes while preserving its essential accuracy. This reduction significantly lowers data costs, making large-scale AI weather modeling much more practical and accessible.