Ruimeng Zhou

As the world’s second-largest hydropower station, the reservoir drawdown zone of Baihetan Hydropower Station faces severe ecological disturbance from periodic water level fluctuations; exploring its ecological impact patterns is critical for global reservoir drawdown zone ecological restoration and assessment. Taking the drawdown zone of Baihetan Hydropower Station as the study area, an expert-guided PCA-based assessment framework is proposed: 13 key ecological indicators were screened through expert consultation to ensure ecological relevance and data reliability, followed by objective Principal Component Analysis (PCA) for indicator decorrelation, dimensionality reduction, and comprehensive scoring. Combined with spatiotemporal hot–cold spot and overlay analysis, spatially explicit ecological assessment maps were generated using multi-temporal remote sensing images (covering low-water exposure (July flood season), post-flood impoundment (November), and high-water stabilization (December) periods), spanning a continuous decade from 2016 to 2025 to construct a complete 10-year spatiotemporal cube, which include a 5 km buffer along the mainstream of the Jinsha River. Main results: (1) After the 2021 commissioning, comprehensive ecological quality declined overall, with weakened seasonal ecosystem responses, significant homogenization, and an oscillating cold spot-dominated pattern without stable high-quality hot spots. (2) Marked longitudinal differentiation existed: downstream had better, more stable ecological quality, while upstream showed declining quality and greater hot–cold spot variation associated with reservoir operation and anthropogenic disturbances. (3) Spatiotemporal active points only aggregated briefly in November, indicating autumn–winter impoundment inhibited ecological activity. (4) The first principal component explained 85.52% of variance, with VIIRS nighttime lights, wetness and NDVI as the dominant contributing indicators. The temporal changes reported here are associations with the commissioning of the hydropower station and are not presented as demonstrated causal effects. This study provides a transferable framework for the ecological assessment of hydropower-driven vulnerable areas worldwide.

Hyperspectral image classification (HSIC) plays a crucial role in fine-grained Earth observation tasks. However, balancing efficient long-range dependency modeling with the extraction of fine-grained local features remains a significant challenge, primarily due to the inherent high-dimensional spectral redundancy and complex spatial variability of hyperspectral data. Existing modeling paradigms exhibit distinct limitations: Convolutional Neural Networks (CNNs) are constrained by localized receptive fields, while Vision Transformers (ViTs), despite their global receptive capabilities, incur prohibitive quadratic computational complexity. Meanwhile, the emerging Mamba architecture has demonstrated remarkable effectiveness in sequence modeling with linear complexity, but it often lacks sufficient sensitivity to local textures when directly applied to non-causal 2D images. To address these limitations, this paper proposes a novel parallel hybrid architecture termed DualMambaFormer. Deviating from the traditional serial stacking paradigm, the proposed network utilizes a dual-stream design to achieve the complementary fusion of global static attention and dynamic sequence reasoning. Specifically, the model first employs an SS-ResNet for spectral dimensionality reduction and local feature embedding. Subsequently, the architecture bifurcates into a parallel encoding stage: one branch leverages Multi-Head Self-Attention (MHSA) to capture global spatial correlations, while the other introduces a Local Enhanced Mamba (LEM) branch. By integrating State Space Models (SSM) with depthwise separable convolutions, the LEM branch simultaneously captures long-range causal dependencies and local spatial context. Finally, a dual class token fusion strategy is designed to integrate heterogeneous representations at the decision level. Extensive experiments on four benchmark datasets—Indian Pines, Pavia University, Salinas, and WHU-HongHu—show that DualMambaFormer achieves OA values of 96.56%, 98.95%, 97.60%, and 96.09%, respectively, with consistently high AA and Kappa coefficients. These results demonstrate the effectiveness, robustness, and generalization capability of the proposed method for hyperspectral image classification. Compared with the second-best competing methods, DualMambaFormer improves OA by 5.55, 2.30, 1.68, and 4.30 percentage points on the Pavia University, Indian Pines, Salinas, and WHU-HongHu datasets, respectively.

As China’s largest clean energy group and the world’s largest hydropower development company, China Three Gorges Corporation (CTG) accounts for 3/4 of hydropower installed capacity, where most of which are distributed in the Yangtze River Basin. Since 2017, CTG has formally entered the field of water pollution control, taking the “water” as the core and the sewage treatment of cities/towns along the river as an entry point. From a technical point of view, with the advantages of hydropower dispatching and operation, the integration of water resources protection, water pollution control, as well as water ecological restoration will be gradually completed to achieve the “one water” control goal. From a management point of view, through the means of government-enterprise cooperation model, the establishment of a jointly sharing mechanism by relevant stakeholders, as well as the market-oriented commercial operation, the implementation and supervision of the watershed administration will be strengthened. CTG has started a new exploration of enterprises participating in the comprehensive governance of watershed management, which will provide ecological priority and sustainable development experience for the watershed management of the world. En tant que plus grand groupe d’énergie propre de Chine et plus grande société de développement hydroélectrique au monde, China Three Gorges Corporation (CTG) représente les 3/4 de la capacité hydroélectrique installée, dont la plupart est distribuée dans le bassin du Yangtsé. Depuis 2017, CTG est officiellement entré dans le domaine de la lutte contre la pollution de l’eau, en prenant « l’eau » comme noyau et le traitement des eaux usées des villes le long du fleuve comme point d’entrée. D’un point de vue technique, avec les avantages de la répartition et de l’exploitation de l’hydroélectricité, l’intégration de la protection des ressources en eau, du contrôle de la pollution de l’eau, ainsi que de la restauration écologique de l’eau sera progressivement complétée pour atteindre l’objectif de contrôle d’une seule eau. Du point de vue de la gestion, à travers les moyens du modèle de coopération gouvernement-entreprise, la mise en place d’un mécanisme de partage conjoint entre les parties prenantes concernées, ainsi que l’opération commerciale axée sur le marché, la mise en oeuvre et la supervision de l’administration du bassin versant seront renforcées. Le CTG a commencé une nouvelle exploration des entreprises participant à la gouvernance globale de la gestion des bassins versants, ce qui fournira une priorité écologique et une expérience de développement durable pour la gestion des bassins versants du monde.