Landslides can cause severe damage to property and human life. Identifying their locations and characteristics is crucial for emergency rescue and disaster risk assessment. However, existing methods need help in accurately detecting landslides because of their diverse characteristics and scales, as well as the differences in spectral features and spatial heterogeneity of remote sensing images. To overcome these challenges, a multiscale feature fusion landslide-detection network (MFLD-Net) is proposed. This network utilizes reflectance difference images from pre- and post-landslide Sentinel-2A images, along with digital elevation model (DEM) data. Moreover, a multichannel differential landslide dataset was constructed through spectral analysis of Sentinel-2A images, which facilitates network training and enables differentiation between landslides and other objects with similar spectral features, such as bare soil and buildings. The proposed MFLD-Net was tested in Shuzheng Valley and Detuo town in Sichuan, China, where earthquakes have occurred. The experimental results revealed that compared with advanced deep learning models, MFLD-Net has promising landslide detection performance. This study provides suggestions for selecting optimal deep learning methods and spectral band combinations for landslide detection and offers a publicly available landslide dataset for further research.

Urban cover-collapse sinkholes pose a significant global challenge due to their destructive impacts. Previous studies have identified groundwater fluctuations, subsurface soil conditions, pipeline leakage, precipitation, and subterranean construction activities as key contributors to these phenomena. However, unique geological settings across different urban environments lead to variations in the primary factors influencing sinkhole formation. This study focuses on Shanghai, a city notable for its extensive urbanization and rich historical context, to explore the dynamics of sinkholes within urbanized areas worldwide. We employ spatial analysis and statistical methods to examine data on sinkholes recorded in the past two decades in Shanghai, correlating these events with the city's shallow sand layer, ground elevation, and proximity to surface water. Our goal is to identify the dominant factors governing sinkhole occurrence in Shanghai and to lay the groundwork for their effective scientific management and prevention. Key findings indicate that most sinkholes in the area are associated with a thin shallow sand layer, low to moderate ground elevations, and the absence of nearby rivers. Additionally, many sinkholes correlate with subterranean voids within the confined aquifer beneath the cohesive soil layer. The lack of historical river channels, obscured by urban development, also indirectly contributes to sinkhole formation. We recommend enhancing urban river management and drainage systems to mitigate potential damage from water accumulation.

Assessing biodiversity in arctic-alpine ecosystems is a costly task. We test in the current study whether we can map the spatial patterns of spider alpha and beta diversity using remotely-sensed surface reflectance and topography in a heterogeneous alpine environment in Central Norway. This proof-of-concept study may provide a tool for an assessment of arthropod communities in remote study areas. Data on arthropod species distribution and richness were collected through pitfall trapping and subjected to a detrended correspondence analysis (DCA) to extract the main species composition gradients. The DCA axis scores as indicators of species composition as well as trap species richness were regressed against a combined data set of surface reflectance as measured by the Sentinel-2 satellite and topographical parameters extracted from a digital elevation model. The models were subsequently applied to the spatial data set to achieve a pixel-wise prediction of both species richness and position in the DCA space. The spatial variation in the modelled DCA scores was used to draw conclusions regarding spider beta-diversity. The species composition was described with two DCA axes that were characterized by post hoc-defined indicator species, which showed a typical annidation in the arctic-alpine environment under study. The fits of the regression models for the DCA axes and species richness ranged from R-2 = 0.25 up to R-2 = 0.62. The resulting maps show strong gradients in alpha and beta diversity across the study area. Our results indicate that the diversity patterns of spiders can at least partially be explained by means of remotely sensed data. Our approach would likely benefit from the additional use of high resolution aerial photography and LiDAR data and may help to improve conservation strategies in arctic-alpine ecosystems.