Rapid urbanization has led to deteriorated wetland water quality, reduced biodiversity, and fragmented wetland landscapes, which seriously threaten the sustainable development of regional ecology. Based on land use data of Dongying City, Shandong Province, in 2020, this study selected the landscape disturbance degree and landscape fragility index to construct a landscape ecological risk evaluation model and to analyze the spatial distribution characteristics of landscape ecological risk in Dongying City in 2020. The MSPA-Conefor-MCR model was used to extract the ecological network of wetlands in Dongying City, and the topological structure indices were quantitatively analyzed. Combined with the actual situation within the study area, the source sites to be optimized were identified by risk zoning and source importance; the ecological resistance surface was modified using landscape ecological risk, and the ecological network was optimized by simulating edge increase in order to evaluate the robustness of the ecological network before and after optimization and to verify the edge increase effect. The results show that the ecological risk in Dongying is high, mainly distributed in the central region and extending to the northeast, southeast, southwest, and northwest. A total of 131 ecological source sites (6 core and 125 resting-stone source sites) and 180 ecological corridors were extracted, and the whole ecological network was found to be less stable and to have stronger network heterogeneity using a topological analysis. By simulating 11 additional edges, the robustness of the optimized ecological network was significantly improved. Optimizing the simulated-edge increase can enhance the smoothness of ecological energy flow, which can provide a scientific basis for the construction of the ecological security pattern of wetlands in Dongying City.

(1) Background: Eco−spatial networks play an important role in enhancing ecosystem services and landscape connectivity. It is necessary to study landscape structure optimization to achieve synergistic gains in network connectivity and ecosystem functionality. (2) Method: Based on remote sensing data, RS and GIS were used to evaluate the spatiotemporal changes in ecosystem services in China. Combined with complex network theory, the spatiotemporal evolution of China’s ecological spatial network and its topological structure from 2005 to 2020 is discussed. Network function–structure co−optimization was carried out using the edge augmentation strategy. (3) Result: The “three River resource” has high water conservation and high soil and water conservation in southeastern hilly areas. There is strong windbreak and sand fixation in southeastern Inner Mongolia. In the past 15 years, there have been about 8200 sources and about 14,000 corridors. The network has the characteristics of small−world and heterogeneity. After optimization, 18 sources and 3180 corridors are added, and the network connectivity and robustness are stronger. Finally, five regions are divided according to the network heterogeneity and corresponding protection and management countermeasures are proposed to provide scientific guidance for the country’s territorial space planning.

Forest and grass ecological space is the key component of the ecosystem and plays a vital role in regulating the carbon, water, and energy cycle. The long-term exploitation of forest and grass ecological space and huge population pressure have gradually degraded the function of China’s ecosystem. Therefore, forest and grass ecological space plays an important role in maintaining the stability of the ecosystem. The relationship between forest and grass ecospatial network structure and ecosystem service has been the focus of research. In this study, the forest and grass ecospatial network is constructed based on the minimum cumulative resistance (MCR) model. Then, the topological indicators (degree, weight clustering coefficient, node weight, unit weight, weight distribution difference, betweenness, PageRank) of the forest and grass ecospatial network were calculated by combining the complex network theory to analyze the relationship between these topological indicators and the three ecosystems (water retention, soil conservation, carbon storage). Based on the ecological significance of topological indicators, we identified ecologically fragile areas and proposed areas and directions for optimizing the ecospatial structure. Results show that the spatial distribution of the three ecosystem services in the southeast region of China is higher than that in the northwest region of China and shows a gradual decrease from the east to the west. The degree, node weight, unit weight, PageRank, and betweenness were highly significant and positively correlated with the three ecosystem services, among which PageRank had the highest correlation with water retention (p < 0.01, R² = 0.835). Based on the spatial distribution characteristics of the different topological indicators, the quantitative relationship between the structural characteristics of the forest and grass ecospatial network and ecosystem services is clarified, revealing the intrinsic connection between ecological processes and ecosystem services. Through rational optimization of the forest and grass ecospatial network, ecosystem services can be effectively improved and ecosystem stability can be enhanced.

Permafrost and alpine vegetation are widely distributed in Tibet, which is a sensitive area for global climate change. In this study, we inverted the surface deformation from 22 May 2018 to 9 October 2021 in a rectangular area within the city of Linzhi, Tibet, using the Sentinel1-A data and two time-series interferometric system aperture radar (InSAR) techniques. Then, the significant features of surface deformation were analyzed separately according to different vegetation types. Finally, multiple machine learning methods were used to predict future surface deformation, and the results were compared to obtain the model with the highest prediction accuracy. This study aims to provide a scientific reference and decision basis for global ecological security and sustainable development. The results showed that the surface deformation rate in the study area was basically between ±10 mm/a, and the cumulative surface deformation was basically between ±35 mm. The surface deformation of grassland, meadow, coniferous forest, and alpine vegetation were all significantly correlated with NDVI, and the effect of alpine vegetation, coniferous forest, and grassland on permafrost was stronger than that of the meadow. The prediction accuracy of the Holt–Winters model was higher than that of Holt′s model and the ARIMA model; it was expected that the ground surface would keep rising in the next two months, and the ground surface deformation of alpine vegetation and the coniferous forest was relatively small. The above studies indicated that the surface deformation in the Tibetan permafrost region was relatively stable under the conditions of alpine vegetation and coniferous forest. Future-related ecological construction needs to pay more attention to permafrost areas under grassland and meadow conditions, which are prone to surface deformation and affect the stability of ecosystems.

Automatic recognition of species is important for the conservation and management of biodiversity. However, since closely related species are visually similar, it is difficult to distinguish them by images alone. In addition, traditional species-recognition models are limited by the size of the dataset and face the problem of poor generalization ability. Visual-language models such as Contrastive Language-Image Pretraining (CLIP), obtained by training on large-scale datasets, have excellent visual representation learning ability and demonstrated promising few-shot transfer ability in a variety of few-shot species recognition tasks. However, limited by the dataset on which CLIP is trained, the performance of CLIP is poor when used directly for few-shot species recognition. To improve the performance of CLIP for few-shot species recognition, we proposed a few-shot species-recognition method incorporating geolocation information. First, we utilized the powerful feature extraction capability of CLIP to extract image features and text features. Second, a geographic feature extraction module was constructed to provide additional contextual information by converting structured geographic location information into geographic feature representations. Then, a multimodal feature fusion module was constructed to deeply interact geographic features with image features to obtain enhanced image features through residual connection. Finally, the similarity between the enhanced image features and text features was calculated and the species recognition results were obtained. Extensive experiments on the iNaturalist 2021 dataset show that our proposed method can significantly improve the performance of CLIP’s few-shot species recognition. Under ViT-L/14 and 16-shot training species samples, compared to Linear probe CLIP, our method achieved a performance improvement of 6.22% (mammals), 13.77% (reptiles), and 16.82% (amphibians). Our work provides powerful evidence for integrating geolocation information into species-recognition models based on visual-language models.