To track sea ice motion, four ice‐tethered buoys were deployed at 84.6°N and 144.3°W, 87.3°N and 172.3°W, 81.1°N and 157.4°W, and 82.8°N and 166.5°W in summers of 2008, 2010, 2014, and 2016, respectively. In addition, the remote sensed ice motion product provided by National Snow and Ice Data Center was used to reconstruct backward and forward ice drifting trajectories from the buoy deployment sites during 1979–2016. Sea ice in the central Arctic Ocean in late summer is trending to have travelled from lower latitudes, and to be advected to the region more involved in the Transpolar Drift Stream (TDS) during 1979–2016. The strengthened TDS has played a crucial role in Arctic sea ice loss from a dynamic perspective. The trajectory of ice is found to be significantly related to atmosphere circulation indices. The Central Arctic Index (CAI), defined as the difference in sea level pressure between 84°N, 90°W and 84°N, 90°E, can explain 34–40% of the meridional displacement along the backward trajectories, and it can explain 27–40% of the zonal displacement along the forward trajectories. The winter Beaufort High (BH) anomaly can explain 18–27% of the zonal displacement. Under high positive CAI values or high negative winter BH anomalies, floes from the central Arctic tended to be advected out of the Arctic Ocean through Fram Strait or other marginal gateways. Conversely, under high negative CAI values or high positive winter BH anomalies, ice tended to become trapped within a region close to the North Pole or it drifted into the Beaufort Gyre region. The long‐term trend and spatial change in Arctic surface air temperature were more remarkable during the freezing season than the melt season because most energy from the lower troposphere is used to melt sea ice and warm the upper ocean during summer.

Arctic sea ice kinematics and deformation play significant roles in heat and momentum exchange between the atmosphere and ocean, and at the same time they have profound impacts on biological processes and biogeochemical cycles. However, the mechanisms regulating their changes on seasonal scales and their spatial variability remain poorly understood. Using position data recorded by 32 buoys in the Pacific sector of the Arctic Ocean (PAO), we characterized the spatiotemporal variations in ice kinematics and deformation for autumn–winter 2018/19, during the transition from a melting sea ice regime to a nearly consolidated ice pack. In autumn, the response of the sea ice drift to wind and inertial forcing was stronger in the southern and western PAO compared to the northern and eastern PAO. These spatial heterogeneities gradually weakened from autumn to winter, in line with the seasonal increases in ice concentration and thickness. Correspondingly, ice deformation became much more localized as the sea ice mechanical strength increased, with the area proportion occupied by the strongest (15 %) ice deformation decreasing by about 50 % from autumn to winter. During the freezing season, ice deformation rate in the northern PAO was about 2.5 times higher than in the western PAO and probably related to the higher spatial heterogeneity of oceanic and atmospheric forcing in the north. North–south and east–west gradients in sea ice kinematics and deformation within the PAO, as observed especially during autumn in this study, are likely to become more pronounced in the future as a result of a longer melt season, especially in the western and southern parts.

Extracting representative images of tourist attractions from geotagged photos is beneficial to many fields in tourist management, such as applications in touristic information systems. This task usually begins with clustering to extract tourist attractions from raw coordinates in geotagged photos. However, most existing cluster methods are limited in the accuracy and granularity of the places of interest, as well as in detecting distinct tags, due to its primary consideration of spatial relationships. After clustering, the challenge still exists for the task of extracting representative images within the geotagged base image data, because of the existence of noisy photos occupied by a large area proportion of humans and unrelated objects. In this paper, we propose a framework containing an improved cluster method and multiple neural network models to extract representative images of tourist attractions. We first propose a novel time- and user-constrained density-joinable cluster method (TU-DJ-Cluster), specific to photos with similar geotags to detect place-relevant tags. Then we merge and extend the clusters according to the similarity between pairs of tag embeddings, as trained from Word2Vec. Based on the clustering result, we filter noise images with Multilayer Perceptron and a single-shot multibox detector model, and further select representative images with the deep ranking model. We select Beijing as the study area. The quantitative and qualitative analysis, as well as the questionnaire results obtained from real-life tourists, demonstrate the effectiveness of this framework.

Arctic sea ice kinematics and deformation play significant roles in heat and momentum exchange between atmosphere and ocean. However, mechanisms regulating their changes at seasonal scales remain poorly understood. Using position data of 32 buoys in the Pacific sector of the Arctic Ocean (PAO), we characterized spatiotemporal variations in ice kinematics and deformation for autumn–winter 2018/19. In autumn, sea ice drift response to wind forcing and inertia were stronger in the southern and western than in the northern and eastern parts of the PAO. These spatial heterogeneities decreased gradually from autumn to winter, in line with the seasonal evolution of ice concentration and thickness. Areal localization index decreased by about 50 % from autumn to winter, suggesting the enhanced localization of intense ice deformation as the increased ice mechanical strength. In winter 2018/19, a highly positive Arctic Dipole and a weakened high pressure system over the Beaufort Sea led to a distinct change in ice drift direction and an temporary increase in ice deformation. During the freezing season, ice deformation rate in the northern part of the PAO was about 2.5 times that in the western part due to the higher spatial heterogeneity of oceanic and atmospheric forcing in the north. North–south and east–west gradients in sea ice kinematics and deformation of the PAO observed in autumn 2018 are likely to become more pronounced in the future as sea ice losses at higher rates in the western and southern than in the northern and western parts.

Increasing amounts of evidence have proven Arctic sea ice is undergoing remarkable loss. On the bright side, the Arctic sea routes are becoming increasingly accessible. In this study, the NSIDC product of sea ice motion was applied to reconstruct the northward speed of sea ice to obtain the kinematic features of the sea ice in the Arctic outflow region which specially referred to the Fram Strait and to the north of the Northeast Passage (NEP).In the Arctic outflow region, the average southward displacement of sea ice in 2007–2014 (1511 km) was more than twice the average prior to 2007 (617 km), which indicated continuous enhancement of the Transpolar Drift Stream (TDS) in comparison with previous years. In the regions to the north of the NEP, the long-term trend of northward sea ice speed in the Kara sector was +0.04 cm s−1 year−1 in spring. A significant statistical relationship was found between the NEP open period and the northward speed of the sea ice to the north of the NEP. The offshore advection of sea ice could account for the opening of sea routes by 33% and 15% in the Kara and Laptev sectors, respectively.The atmospheric circulation indices across the TDS, i.e., the Central Arctic Index (CAI), presented more significant correlation than for the Arctic atmospheric Dipole Anomaly index with the open period of the NEP, and the CAI could explain the southward displacement of sea ice toward Fram Strait by more than 45%. The impact from the summer positive CAI reinforces the thinning and mechanical weakening of the sea ice in the NEP region, which promoted the navigability of the NEP.