The Three Gorges hydropower station in China creates a large reservoir by diverting water from the Yangtze River, increasing the risk of geological disasters, especially massive landslides along the reservoir shoreline. To mitigate these risks, improving geological monitoring and early warning systems is crucial. Interferometric Synthetic Aperture Radar (InSAR) is widely used to monitor reservoir bank landslides. However, its potential in early warning systems is limited due to temporal resolution constraints, preventing timely warnings. To address this, we propose integrating daily hydrological data (precipitation and water level observations) with historical InSAR deformation sequences using our deep learning-based multivariate united state estimation network, “MUSEnet.” This approach generates customized daily landslide deformation products for high-risk areas, greatly enhancing early warning capabilities by providing timely and accurate information on landslide occurrence and magnitude. We validated our method using 161 Sentinel-1 A images of the Xinpu landslide in the Three Gorges Reservoir area. Through statistical analysis, we identified different degrees of influence from rainfall and reservoir water level on the deformation of the Xinpu landslide at various locations. Additionally, we observed distinct lag times between deformation and corresponding rainfall and reservoir water level events. By utilizing deep learning, our method estimates nonlinear states by considering hysteresis and intelligently accounts for the impact of rainfall and reservoir water level, resulting in more accurate estimations compared to traditional models.

Landslides are geological events that frequently cause major disasters. Research on landslides is essential, but current studies mostly use historical landslide data and do not reflect dynamic, real-time research results. In this study, landslide deformations and land-use changes were used to analyze the landslide distribution in Fengjie County and Wushan County in the Three Gorges Reservoir Area (TGRA) by using interferometric and polarimetric SAR. In this study, the mean annual rate of landslide deformations was obtained using the small baseline subset interferometric synthetic aperture radar (SBAS-InSAR) for the ALOS-2 (2014–2019) data. Land-use changes were based on the 2007 and 2017 land-use results from dual-polarization ALOS-1 and ALOS-2 data, respectively. To address the problem of classification accuracy reduction caused by geometric distortion in mountainous areas, we first used texture maps and pseudocolor maps synthesized with dual-polarization intensity maps to perform classification with random forest (RF), and then we used coherence and slope maps to run the K-Means algorithm (KMA). We named this the secondary classification method. It is an improvement on the single classification method, exhibiting a 94% classification accuracy, especially in rugged areas. Combined with land-use changes, GIS spatial analysis was used to analyze the spatial distribution of landslides, and it was found that the landslide rate was significantly correlated with the type after change, with a correlation coefficient of 0.7. In addition, land-use types associated with human activities, such as cultivated vegetation, were more likely to cause landslide deformation, which can be used to guide local land-use planning.

Previous studies have shown systematic bias in memory for visual information such as color, orientation and location. For instance, a study on visual working memory for color using method of adjustment found that responses drawn from working memory are significantly biased away from category boundaries and toward category centers (Bae, Olkkonen, Allred, & Flombaum, 2015). However, it is unclear whether systematic bias also occurs in memory for high-level visual features, such as the style of an image. Here we investigated long-term memory for image style using a method of adjustment. The style transfer algorithm based on deep learning (Gatys, Ecker, & Bethge, 2016) was used to convert scene photos into painting-like scene images. Based on painting styles of four well-known artworks (Rembrandt, Braque, Monet and Kandinsky), we created image series (72 image styles for each of 72 scenes) in which the style of painting changes continuously. The experiment consisted of a learning phase and a test phase, between which there was a 30-minute interval. In the learning phase, participants looked at 72 scene images, and each of the images was shown for 10s. Participants were asked to memorize the content and style of every scene image. In the test phase, the same set of scenes was shown in different styles from the learning phase, and participants adjusted the image styles to reproduce those appeared in the learning phase. Response frequency data show that more responses were made for near-prototypical stimuli, compared with mixed-style stimuli. The bias in the current study is different from that we found in a previous short-term memory experiment of image style, which showed fewer response for near-prototypical style, and more response for mixed-style stimuli. Taken together, the current study suggests that systematic bias also exists in memory for high-level features like image style.