The identification of fine particulate matter (PM_2.5) concentrations and its driving factors are crucial for air pollution prevention and control. The factors that influence PM_2.5 in different regions exhibit significant spatial heterogeneity. Current research has quantified the spatial heterogeneity of single factors but fails to discuss the interactions between factors. In this study, we first divided the study area into subregions based on the spatial heterogeneity of factors in a multi-scale geographically weighted regression model. We then investigated the interactions between different factors in the subregions using the geographical detector model. The results indicate that there was significant spatial heterogeneity in the interactions between the driving factors of PM_2.5. The interactions between natural factors have significant uncertainty, as do those between the normalized difference vegetation index (NDVI) and socioeconomic factors. The interactions between socioeconomic factors in the subregions were consistent with those in the whole region. Our findings are expected to deepen the understanding of the mechanisms at play among the aforementioned drivers and aid policymakers in adopting unique governance strategies across different regions.

Hydrology is an important environmental factor for the evolution of wetland landscape patterns. In the past 20 years, Poyang Lake, the largest freshwater lake in China, has experienced significant inundation shrinkage and water level decrease, posing a significant threat to the local vegetation community. To explore the potential relationship between aquatic vegetation and hydrological processes in the recent hydrological situation, in this study, the landscape patterns of aquatic vegetation communities in Poyang Lake were studied using time-series Landsat remote sensing images and a support vector machine classifier. The stepwise regression analysis method was adopted to analyze the relationship between the vegetation area and hydrological factors. The results indicated that the area of submerged and emergent vegetation in the entire lake decreased significantly from 2001 to 2017, whereas the area of moist vegetation showed a remarkably increasing trend. The average distribution elevation of the submerged vegetation increased by 0.06 m per year. The corresponding landscape patterns showed that the degree of fragmentation of aquatic vegetation communities in Poyang Lake increased. Several hydrological factors were selected to quantify the potential impact of water level fluctuations. The correlation analysis results indicated that hydrological conditions during the rising- and high-water periods may be the key factors affecting the area of aquatic vegetation. This study systematically investigated the evolution of vegetation communities in Poyang Lake wetlands over the past two decades, which contributes to the protection and management of this unique ecosystem.

Sediment movements on the riverbeds are crucial in fluvial processes. Therefore, studying incipient sediment motion is valuable for predicting changes in sediment behavior. In this study, the influence of aquatic vegetation is introduced into the momentum balance equation under the condition of a bare bed. The presented model for incipient sediment motion in open channels covered by vegetation is derived, in consideration of vegetation density and relative submergence. This study simplifies the model format. The model and its coefficients demonstrate good applicability to various working conditions without requiring numerous changes. The predicted velocity determined by the proposed model is validated against experimental data. Results show excellent agreement between the two in various scenarios, including combinations of rigid or flexible conditions and submerged or emergent states. Additionally, the velocity for incipient sediment motion is observed to decrease with increasing vegetation density. However, a further increase in vegetation density may amplify vegetation resistance, hindering incipient sediment motion. Meanwhile, sediment movement becomes challenging as water depth increases, assuming that flow discharge and vegetation conditions remain constant.

Submerged aquatic vegetation can provide a habitat and food for marine and river organisms, and it has the ecological effect of purifying water by absorbing harmful substances. Therefore, it plays an important role in the maintenance, restoration, and improvement of marine and river ecosystems. Hydrodynamic problems caused by submerged vegetation have been a matter of wide concern. According to the distribution of submerged vegetation, the flow can be divided into three layers in the vertical direction: uniform, mixing, and logarithmic layers. This paper proposes an analytical model for the vertical distribution of longitudinal velocity in open-channel flows with submerged vegetation. A concept of velocity superimposition is applied in mixing and logarithmic layers. The velocity inside the vegetated layer can be solved by the balance between the drag force and bed gradient. The velocity difference between the vegetated layer and the free surface layer results in the formation of a mixing layer near the top of the vegetation. Flow at the junction between the vegetation and free surface layers is mainly controlled by the vortices in the mixing layer. The velocity in the mixing layer is commonly described by a hyperbolic tangent formula. The logarithmic distribution formula is applied to the free surface layer, where the velocity without effect arising from vortices is similar to the open-channel flow. The concept of the wake function is introduced to modify the distribution of velocity in the free surface layer. The longitudinal velocities from the theoretical model are compared to the measured velocities in the literature. The theoretical velocities agree well with the measured values in the flows with submerged vegetation, proving that the theoretical model proposed here can successfully predict the vertical distribution of velocity and has extensive adaptability.

The early warning and prediction of saltwater intrusion are crucial for the protection and management of estuarine and marine ecosystems, and water supply safety. Aiming at providing a high-accuracy and stable salinity prediction, this study proposes an integrated deep learning method based on long short-term memory (LSTM) networks, gated recurrent units (GRUs), and convolutional neural networks (CNN). Taking the Modaomen Waterway as the research area, an hourly saltwater intrusion prediction model is constructed with a prediction period of 6, 12, and 24 h. Based on upstream flow data, downstream tide data, and antecedent salinity data from three monitor stations during 2020–2022, the saltwater intrusion prediction model is trained and validated. Results show that the proposed model can provide satisfactory results in all stations and prediction periods. Through the comparisons among the four models, it demonstrates that the integrated model performs better in saltwater intrusion prediction, achieving peak Nash–Sutcliffe efficiency improvements of 65.4% and error reductions up to 54.9%. As the prediction period extends, the accuracy of the predictions decreases. By enhancing the precision and reliability of salinity forecasts, this research aids in the development of effective mitigation strategies to counteract the adverse effects of saltwater intrusion.