A Deep Learning Network for Cloud-to-Ground Lightning Nowcasting with Multisource Data

Abstract

Precise and timely lightning nowcasting is still a great challenge for meteorologists. In this study, a new semantic segmentation deep learning network for cloud-to-ground (CG) lightning nowcasting, named LightningNet, has been developed. This network is based on multisource observation data, including data from a geostationary meteorological satellite, Doppler weather radar network, and CG lightning location system. LightningNet, with an encoder–decoder architecture, consists of 20 three-dimensional convolutional layers, pooling and upsampling layers, normalization layers, and a softmax classifier. The central–eastern and southern China was selected as the study area, with considerations given to the topography and spatial coverage of the weather radar and lightning observation networks. Brightness temperatures (TB) of six infrared bands from the Himawari-8 satellite, composite reflectivity mosaic, and CG lightning densities were used as the predictors because of their close relationships with lightning activity. The multisource data were first interpolated into a uniform spatial/temporal resolution of 0.05° × 0.05°/10 min, and then training and test datasets were constructed, respectively. LightningNet was trained to extract the features of lightning initiation, development, and dissipation. The evaluation results demonstrated that LightningNet was able to achieve good performance of 0–1-h lightning nowcasts using the multisource data. The probability of detection, the false alarm ratio, the area under relative operating characteristic curve, and the threat score (TS) of LightningNet with all three types of data reached 0.633, 0.386, 0.931, and 0.453, respectively. Because geostationary meteorological satellite and radar both possess the capability of capturing lightning initiation (LI) features, LightningNet also showed good performance in LI nowcasting. When all three types of data were used, more than 50% LI was predicted accurately and the TS exceeded 0.36. LightningNet’s nowcast performance using triple-source data was clearly superior to that using only single-source or dual-source data, and these findings indicate that LightningNet has good capability of combining multisource data effectively to produce more reliable lightning nowcasts.

Improving Nowcasting of Convective Development by Incorporating Polarimetric Radar Variables Into a Deep‐Learning Model

“…Compared with the ConvLSTM‐based models, the U‐Net has more flexible architecture, which can be easily adjusted for exploiting multiple input information. The U‐Net is increasingly used by spatiotemporal forecasting studies, such as nowcasting of lightnings (Zhou et al., 2020) and global weather forecasting (Weyn et al., 2020). As such, we choose U‐Net as our backbone model in this study.…”

A deep learning framework for lightning forecasting with multi‐source spatiotemporal data

“…Several investigations (Geng et al ., 2019; Wang et al ., 2019; Zhou et al ., 2020) have verified that integrating different data sources contributes to lightning forecasting. For example, (Wang et al ., 2019) designed a DNN model combining past observations and NWP to predict temperature, relative humidity and wind at an AWS.…”

“…They demonstrated that information fusion brings a significant performance improvement compared with the case using a single data source. (Zhou et al ., 2020) proposed a new semantic segmentation deep learning network for 1 hr cloud‐to‐ground lightning nowcasting. Their model displays good performance by introducing multi‐source data.…”

Lightning Strike Location Identification Based on 3D Weather Radar Data

“…The recognition performances of the learning models on thunderstorms and gales were compared. Zhou et al (2020) proposed a new semantic segmentation-based deep learning network for cloud-to-ground lightning nowcasting named LightningNet. This model conducts reliable lightning nowcasting by using multisource data.…”