An Improved Deep Learning Model for High-Impact Weather Nowcasting
Shun Yao,Haonan Chen,Elizabeth J. Thompson,Robert Cifelli
DOI: https://doi.org/10.1109/jstars.2022.3203398
IF: 4.715
2022-09-14
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Abstract:Accurate nowcasting (short-term prediction, 0–6 h) of high-impact weather, such as landfalling hurricanes and extreme convective precipitation, plays a critical role in natural disaster monitoring and mitigation. A number of nowcasting approaches have been developed in the past few decades, such as optical flow and the tracking radar echoes by correlation system. Most of these mainstream operational techniques are based on radar echo map extrapolation, which determines the velocity and direction of precipitation systems using historical and current radar observations. However, the skill of the traditional extrapolation method decreases rapidly within the first hour. In order to improve nowcasting skill, recent studies have proposed using deep learning methods, such as convolutional recurrent neural network and trajectory gate recurrent unit. But none of these methods focuses on high-impact weather events, and the deep learning models trained based on general precipitation events cannot meet the demand of accurate warnings and decision-making at the scales required for high-impact weather events, such as hurricanes. Using multiradar observations, this article introduces the idea of self-attention and develops a self-attention-based gate recurrent unit (SaGRU) to enhance its generalization capability and scalability in predicting high-impact weather events. In particular, two types of high-impact weather systems, namely, landfalling hurricanes and extreme convective precipitation events, are investigated. Three models are trained based on hurricane events, heavy rainfall (i.e., nonhurricane) events, and all events combined in the southeast United States during 2015 and 2020. The impacts of different data sources on the nowcasting performance are quantified. The evaluation results of nowcasting products show that our SaGRU performs very well in predicting hurricane-induced rainfall. In the new methodology, the data from nonhurricane events are shown- to provide useful information in enhancing the nowcasting performance during hurricane events as the model trained by combining all the hurricane and nonhurricane events has the best performance. In addition, this article quantifies the impact of the sequence length of input radar observations on the nowcasting performance, which shows that five consecutive observations are sufficient to obtain a stable model, and even two consecutive observations can produce reasonable results.
imaging science & photographic technology,remote sensing,engineering, electrical & electronic,geography, physical