Physically Explainable Deep Learning for Convective Initiation Nowcasting Using GOES-16 Satellite Observations

TL;DR

This paper develops object-based probabilistic deep learning models to improve convective initiation nowcasting using GOES-16 satellite data, outperforming classical models and providing scientific insights into cloud and moisture features.

Contribution

It introduces a novel deep learning approach for CI nowcasting that incorporates model explanations and baseline comparisons to enhance understanding and accuracy.

Findings

Deep learning models outperform classical logistic models up to 1 hour lead time.

Model explanations reveal the importance of moisture and cloud features at multiple levels.

Using different baselines enhances understanding of model decision-making.

Abstract

Convection initiation (CI) nowcasting remains a challenging problem for both numerical weather prediction models and existing nowcasting algorithms. In this study, object-based probabilistic deep learning models are developed to predict CI based on multichannel infrared GOES-R satellite observations. The data come from patches surrounding potential CI events identified in Multi-Radar Multi-Sensor Doppler weather radar products over the Great Plains region from June and July 2020 and June 2021. An objective radar-based approach is used to identify these events. The deep learning models significantly outperform the classical logistic model at lead times up to 1 hour, especially on the false alarm ratio. Through case studies, the deep learning model exhibits the dependence on the characteristics of clouds and moisture at multiple levels. Model explanation further reveals the model's decision-making process with different baselines. The explanation results highlight the importance of moisture and cloud features at different levels depending on the choice of baseline. Our study demonstrates the advantage of using different baselines in further understanding model behavior and gaining scientific insights.