Rapid and accurate identification and tracking of lightning clusters from massive lightning detection data are crucial for real-time thunderstorm nowcasting and climatological analyses of thunderstorm activity. Although density-based clustering algorithms can identify clusters of arbitrary shapes at fine scales, their performance is often hindered by large data volumes and significant variations in lightning density. To address these challenges, we propose a multi-scale spatiotemporal lightning clustering framework, termed CC3D-CSCAP. It consists of two main components. First, the 3-D connected component algorithm (CC3D) performs coarse-scale segmentation by dividing the lightning dataset into spatiotemporally disconnected subsets using 26-connectivity. Then, the cylinder-based scan clustering algorithm with adaptive parameters (CSCAP) is applied to each subset for fine-scale identification of lightning clusters. Since the lightning subset may still contain multiple thunderstorms with varying lightning densities, CSCAP adaptively determines clustering parameters based on the statistical characteristics (time difference and spatial distance) of subset. Compared with fixed-parameter methods, CC3D-CSCAP identifies more clusters (771,033) while retaining a high percentage of usable lightning strokes (98.988%). The clustering results align well with the theoretical criteria for optimal clustering and are promising for global applications in lightning data analysis, nowcasting, and climatological studies of convective systems.
Interferometric synthetic aperture radar (InSAR) plays a significant role in monitoring permafrost deformation. However, owing to environmental constraints in permafrost regions, some regions exhibit temporal incoherence, which results in deformation with fewer measurement points and difficulties with deformation automatic detection. In this study, a full-coverage deformation rate map of the 10 km buffer of the Qinghai–Tibet Engineering Corridor (QTEC) was generated by combining nine driving factors and the deformation rate of the 5 km buffer along the QTEC based on three machine learning methods. The importance of the factors contributing to ground deformation was explored. The experimental results show that support vector regression (SVR) yielded the best performance (R2 = 0.98, RMSE = 0.76 mm/year, MAE = 0.74 mm/year). The 10 km buffer of deformation data obtained not only preserved the original deformation data well, but it also filled the blank areas in the deformation map. Subsequently, we trained the Faster R-CNN model on the deformation rate map simulated by SVR and used it for the automatic detection of permafrost thaw settlement areas. The results showed that the Faster R-CNN could identify the permafrost thawing slump quickly and accurately. More than 300 deformation areas along the QTEC were detected through our proposed method, with some of these areas located near thaw slump and thermokarst lake regions. This study confirms the significant potential of combining InSAR and deep learning techniques for permafrost degradation monitoring applications.