Timely precise metro ridership forecasting is helpful to reveal real-time traffic demand, which is a crucial but challenging task in modern traffic management. Given the complex spatial correlation and temporal variation of riding behaviour in a metro system, deep learning algorithms have been widely applied owing to their superior performance in capturing spatio-temporal features. However, current deep learning models utilize regular convolutional operations, which can barely provide satisfactory accuracy due to either the ignorance of realistic topology of a traffic network or insufficiency in capturing representative spatiotemporal patterns. To further improve the accuracy in metro ridership prediction, this study proposes a parallel-structured deep learning model that consists of a Graph Convolution Network and a stacked Bidirectional unidirectional Long short-term Memory network (GCN-SBULSTM). The GCN module regards a metro network as a structured graph, and a K-hop matrix, which integrates the travel distance, population flow, and adjacency, is introduced to capture the dynamic spatial correlation among metro stations. The SBULSTM module considers both backward and forward states of ridership time series and can learn complex temporal features with stacked recurrent layers. Experiments are conducted on three real-life metro ridership datasets to demonstrate the effectiveness of the proposed model. Compared with state-of-the-art prediction models, GCN-SBULSTM presents better performance in multiple scenarios and largely enhances the efficiencies of training processes.

Arctic sea ice motion information provides an important scientific basis for revealing the changing mechanism of Arctic sea ice and assessing the navigational safety of Arctic waterways. For now, many satellite derived Arctic sea ice motion products have been released but few studies have conducted comparisons of these products. In this study, eleven satellite sea ice motion products from the Ocean and Sea Ice Satellite Application Facility (OSI SAF), the National Snow and Ice Data Center (NSIDC), and the French Research Institute for the Exploitation of the Seas (Ifremer) were systematically evaluated and compared based on buoys from the International Arctic Buoy Program (IABP) and the Multidisciplinary drifting Observatory for the Study of Arctic Climate (MOSAiC) over 2018–2020. The results show that the mean absolute errors (MAEs) of ice speed for these products are 1.15–2.26 km/d and the MAEs of ice motion angle are 14.93–23.19°. Among all products, Ifremer_AMSR2 achieves the best accuracy in terms of speed error, NSIDC_Pathfinder shows the lowest angle error and OSI-405-c_Merged performs best in sea-ice drift trajectory reconstruction. Moreover, season, region, data source, ice drift tracking algorithm, and time interval all influence the accuracy of these products: (1) The sea ice motion bias in the freezing season (1.04–1.96 km/d and 11.93–22.41°) is smaller than that in the melting season (1.13–3.90 km/d and 14.41–27.41°) for most of the products. (2) Most products perform worst in East Greenland, where ice movements are fast and complex. (3) The accuracies of the products derived from AMSR-2 remotely sensed data are better than those from other data sources. (4) The continuous maximum cross-correlation (CMCC) algorithm outperforms the maximum cross-correlation (MCC) algorithm in sea ice drift retrieval. (5) The MAEs of sea ice motion with longer time interval are relatively smaller. Overall, the results indicate that the eleven remote sensing Arctic sea ice drift products are of practical use for data assimilation and model validation if uncertainties are appropriately considered. Furthermore, this study provides some improvement directions for sea ice drift retrieval from satellite data.

Extraction of urban building damage caused by earthquake disasters, from very-high-resolution (VHR) satellite imagery and related geospatial data, has been widely studied in the past decade. In this study, a multi-stage collapsed building detection method, using bi-temporal (pre- and post-earthquake) VHR images and post-earthquake airborne light detection and ranging (lidar) data, is proposed. Ground objects that are intact and significantly different from collapsed buildings, such as intact buildings, pavements, shadows, and vegetation, were first extracted using the post-event VHR image and lidar data and masked out. Collapsed buildings were then extracted by classifying the combined bi-temporal VHR images and texture images of the remaining area using a one-class classifier, the One-Class Support Vector Machine (OCSVM). A post-processing procedure was adopted to refine the obtained result. The proposed method was quantitatively evaluated and compared to two existing methods in Port au Prince, Haiti, which was heavily hit by an earthquake in January 2010. In the two comparative methods, data for the whole study area were directly used. In the first method, collapsed buildings were extracted directly using the OCSVM, while in the second method, buildings and pavements were removed from the extraction result of the first method. The results showed that the proposed method significantly outperformed the existing methods, with increases of 21% and 40%, respectively, in the kappa coefficient. The proposed method provides a fast and reliable method to detect collapsed urban buildings caused by earthquake disasters, and could also be applied to other study areas using similar data combinations.