REE-TTT: Highly Adaptive Radar Echo Extrapolation Based on Test-Time Training

TL;DR

This paper introduces REE-TTT, a novel radar echo extrapolation model with test-time training and attention mechanisms, significantly improving precipitation nowcasting accuracy and adaptability across diverse regions and extreme weather events.

Contribution

The paper presents a new adaptive model with a spatio-temporal test-time training block that enhances generalization in radar echo extrapolation for meteorology.

Findings

Outperforms state-of-the-art models in cross-regional scenarios

Demonstrates robustness to data distribution shifts

Significantly improves precipitation prediction accuracy

Abstract

Precipitation nowcasting is critically important for meteorological forecasting. Deep learning-based Radar Echo Extrapolation (REE) has become a predominant nowcasting approach, yet it suffers from poor generalization due to its reliance on high-quality local training data and static model parameters, limiting its applicability across diverse regions and extreme events. To overcome this, we propose REE-TTT, a novel model that incorporates an adaptive Test-Time Training (TTT) mechanism. The core of our model lies in the newly designed Spatio-temporal Test-Time Training (ST-TTT) block, which replaces the standard linear projections in TTT layers with task-specific attention mechanisms, enabling robust adaptation to non-stationary meteorological distributions and thereby significantly enhancing the feature representation of precipitation. Experiments under cross-regional extreme precipitation scenarios demonstrate that REE-TTT substantially outperforms state-of-the-art baseline models in prediction accuracy and generalization, exhibiting remarkable adaptability to data distribution shifts.