Under the background of frequent extreme climate events, analyzing the spatio-temporal changes of surface water bodies in the Songhua River Basin affected by climate change over the past 30 years is of great significance for the region to take and optimize climate response measures in advance. Based on the spatio-temporal dynamic changes of surface water area in the Songhua River Basin from 1990 to 2020, seasonal water bodies were classified and analyzed. By using the EC JRC global surface water product dataset and combining with MATLAB software, the time series feature +K-means clustering and dynamic statistical threshold method were adopted to analyze the changing trend of surface water area in the Songhua River Basin in the past 30 years, and to classify the seasonal water body area as well as detect and analyze the recovery dynamics of extreme hydrological years. The results show that during the study period, the seasonal water body area in the Songhua River Basin increased by 114%, while the permanent water body area decreased by 49.46%, and the total surface water area showed an upward trend. The annual seasonal water body area was classified into three categories: fluctuating-wet year, low fluctuation - dry year, and stable - normal year, with a silhouette coefficient of 0.549, indicating a good clustering effect. Two significant high-value years, 1998 and 2013, were identified, and it was analyzed that neither had fully recovered. The research results provide a direction for the effective management and protection of water resources by referring to historical extreme hydrological events and making flood and drought prevention preparations in advance for the Songhua River Basin in the face of future extreme weather.

The Leizhou Peninsula exhibits one of the highest incidences of drought within the Pearl River Basin. Research on the characteristics and response mechanisms of meteorological and hydrological droughts in this area holds significant theoretical and practical importance for understanding drought patterns in regions with abundant rainfall yet fragile water resource systems. Based on the monthly meteorological and hydrological data from the representative station in Leizhou Peninsula, the standardized precipitation-evapotranspiration index (I_SPEI) and standard runoff index (I_SRI) were used to analyze the variation patterns of meteorological drought and hydrological drought in this region from 1970 to 2016. The Mann-Kendall trend test method, moving average method and Morlet wavelet analysis method were adopted to analyze the changing trends and cycles of meteorological drought and hydrological drought in this region at different time scales (3 and 12 months). Propagation characteristics of meteorological drought to hydrological drought were examined using the run-length theory. The results show that the meteorological drought in Leizhou Peninsula is increasing during the study period. Particularly, the 12-month I_SPEI obtained from Zhanjiang station and Xuwen station decreased significantly, at the rate of -0.014/month and -0.008/month, respectively. However, no obvious trend of hydrological drought was detected. Both drought types showed multi-scale periodicity, sharing a primary first-order period of 20 years in the north and 14 years in the south. The main cycles of meteorological drought and hydrological drought in the south and north are consistent, and the variation patterns are relatively synchronous. Hydrological drought in the southern peninsula demonstrated stronger responsiveness to meteorological drought than the northern region.

To solve the problem of decreasing prediction accuracy caused by nonlinear runoff sequence and instability of single prediction model, this paper proposes a "selection-combination-correction" modeling strategy based on the "decomposition-prediction" model. Firstly, five models including DNN, SVM, LSTM, TCN, and GBRT are used to establish 15 coupled models based on EMD, CEEMDAN, and VMD, and the models are selected. Then, the selected model is used as the base model, and the predicted results of each period of the base model are processed and input into a multi-layer perceptron to construct a new combination model. A residual correction equation is constructed for the test period of the combination model to further improve the prediction accuracy. Finally, the method is applied to the test studies of Huaxian Station in Weihe River Basin and Yangxian Station in Hanjiang River Basin. The results show that the combination model constructed by the multi-layer perceptron has higher prediction accuracy than the single model, and can integrate the advantages of other models to improve the model's generalization ability. The model with residual correction technology is superior to the combination model in all aspects, especially in the fitting of peak discharge, further improving the prediction accuracy.

To estimate the spatiotemporal characteristics of CO₂ partial pressure (p_CO2) and water-air interface CO₂ flux (F_CO2) along the Yangtze River, this study selected eight monitoring sections along the main channel of the river from upstream to downstream. Based on water quality data from 2020 to 2022, p_CO2 and F_CO2 were calculated using the CO₂ SYS software. The spatiotemporal variations of p_CO2 and F_CO2 were analyzed, and the relationships between p_CO2, F_CO2, and environmental factors were assessed using the Mantel test. The results show that from 2020 to 2022, p_CO2 in the Yangtze River ranged from 450.02 to 3 615.88 μatm, which was higher than the global atmospheric CO₂ average partial pressure of 414.78 μatm during the same period. Spatially, the distribution of p_CO2 was as follows: midstream > upstream > estuary. The average values of water-air F_CO2 during 2020-2022 were 40.25 mol/(m²·a), 74.74 mol/(m²·a), and 61.85 mol/(m²·a), respectively, with a spatial distribution of upstream > midstream > estuary. The Yangtze River is in a state of CO₂ oversaturation, and that the water-air F_CO2 flux showed an overall increasing trend from 2020 to 2022. p_CO2 exhibited a significant positive correlation with water temperature, dissolved oxygen, and pH, while F_CO2 was positively correlated with pH, dissolved oxygen, and conductivity.

In order to solve the real-time online flow monitoring under complex conditions of wide and shallow river sections, based on the applicability of conventional online flow measuring equipment, the combined application of two-way probe H-ADCP facing each other was proposed. Taking the Gaogang Water Conservancy Project at the source of water intake of the East Route of the South-to-North Water Transfer Project as an example, the whole process of numerical simulation of river sections, flow field analysis, equipment installation, comparison measurement setting and other aspects were carried out applied research. The results show that the change of flow field in the flow measuring reach is complicated under different operation conditions of the control hub gate pump. The dual probe H-ADCP can be installed at the same section and the same height, and the ultrasonic beam crossing will not affect the collection of effective unit velocity data. The relationship between the double index velocity and the average velocity of the section can be established to calculate the real-time flow, and the accuracy of the combined applied flow measurement is higher than that of the single H-ADCP flow measurement data. The research results form a set of real-time flow monitoring application methods under complex conditions of wide and shallow rivers, which can provide ideas for the combined use of online flow measuring equipment.

In view of the sudden water pollution events in the main stream of Minjiang River, taking the lower reaches of the main stream of the Minjiang River as the study area, a coupling model of hydrodynamic and water quality was established to accurately simulate the flow dynamics and water quality changes under different scheduling scenarios. By comparing four different scheduling strategies: background simulation (i.e. no special scheduling measures), single optimization of Qianwei Avionics Water Conservancy Project, single optimization of Longxikou Avionics Water Conservancy Project, and joint optimization of two projects, this study comprehensively evaluated the emergency treatment ability of each strategy for water pollution in the main stream of the Minjiang River. The results show that from the point of view of pollutant concentration peak and pollutant exceeding standard time, the joint scheduling strategy of increasing the sluice flow of two hub projects simultaneously shows the best pollution group disposal effect. This strategy not only significantly reduces the peak concentration of pollutants, but also effectively shortens the period of time when pollutants exceed the standard, thus minimizing the negative impact of water pollution on the environment and ecology. The research results can provide reference for the emergency treatment of the Minjiang River main stream water pollution incident and improve the emergency treatment efficiency.

To clarify the first flush characteristics of particulate and dissolved pollutants in urban roof and road runoff and to scientifically determine the first flush volume (V_FF), this study investigated the typical cement roof and asphalt road runoff in Beijing using the M (V) curve method. The approach involved calculating average pollutant concentrations in specified runoff increments during rainfall events to assess water quality variations, quantify V_FF, and evaluate corresponding runoff pollution control effectiveness. The results show that the pollution levels of particulate and dissolved chemical oxygen demand (COD) and phosphorus in runoff from asphalt road are significantly higher than those from cement roof (p<0.05), but no such significant difference is observed for particulate and dissolved nitrogen. Particulate pollutants in cement roof runoff exhibit a stronger first flush phenomenon, whereas the first flush phenomenon of dissolved pollutants is more pronounced in asphalt road runoff. Except for the pollution indicators with weaker first flush phenomenon, there are significant differences between V_FF of COD, total nitrogen (TN), and total phosphorus (TP) and their particulate or dissolved indicators. According to the current national standard, the first 3 mm of runoff shall be discarded, cement roof and asphalt road can achieve SS pollution control rates of (65.26±32.40)% and (65.43±23.81)%, respectively, and COD pollution control rates of (58.17±32.49)% and (59.06±31.48)%, respectively. Higher runoff pollution control results will be obtained if V_FF is determined based on actual monitoring data. The findings will provide reference for implementing targeted control strategies and methods for particulate and dissolved pollutants in urban stormwater runoff.

To explore the storage effect of functional zoning on different layout methods of LID facilities in industrial parks and identify the optimal LID facility combination scheme, a specific industrial park in Guangdong Province was selected as the subject of study. The software SWMM was employed to investigate the storage effect of individual LID facility layouts and the comprehensive storage benefits of various LID facility combination schemes. The results show that affected by the park's functional zoning, when the downstream greening positions in the study area are concentrated and the rainfall is relatively low, LID facilities installed in the middle of the concentration area demonstrate superior effectiveness compared to those installed downstream. As the rainfall increases, the downstream's storage advantages become fully apparent. When designing LID facility combinations, the scheme combining 5% green roofs and 15% sunken green spaces yields the best storage effect per unit layout area. However, due to the mutual influence among facilities, the effectiveness of this combination scheme is not as good as the arithmetic sum of the data for the individual facilities, failing to produce the desired synergistic effect of "1+1>2".

The construction of fish-friendly water passage has become an inevitable demand for habitat connectivity and protection. To further optimize the hydraulic characteristics of fish-friendly channels, based on a flume experiment, large-eddy simulation (LES) is used to construct a three-dimensional hydrodynamic model. The channel flow is simulated under the influence of regular triangular baffles and permeable triangular baffles. The results indicate that the permeable triangular baffle design effectively reduces the strength of the main recirculation zone and limits its development. The turbulent kinetic energy and Reynolds stress behind the baffle are reduced due to water inflow through the permeable apertures, leading to a 34.5% reduction in maximum turbulent kinetic energy along the vertical axis. Furthermore, at two different water depths, the permeable baffle design increases the proportion of resting zones and local low-velocity zones, while significantly reducing the proportion of negative flow velocity zones that are harmful to fish.

The stability of dam construction by tailings accumulation is closely related to the ultimate bearing capacity of geotextile used in dam construction. The fine tailings of Dayu Shilei Tungsten Mine in Ganzhou are used as the filling material. Through the uniaxial compressive strength test of the filling geotextile filled with fine tailings under different consolidation time and different filling degree, the variation law of the ultimate compressive strength of the geotextile and the failure mode of the geotextile and the strength reinforcement mechanism of the bag body are analyzed. The results show that when the geotextile reaches the ultimate compressive strength, the vertical strain is between 20% and 30%; The ultimate compressive strength of geotextile decreases with the increase of tailings filling degree (filling height), and increases with the increase of consolidation time. Under uniaxial compression, the failure position of the geotextile mainly occurs at the bottom of the intermediate geotextile, and the lower the filling degree, the greater the failure range is; The theoretical calculation value of the modified formula of ultimate compressive strength considering the influence of filling degree is compared with the experimental value of ultimate compressive strength, and the error between them is less than 4%. The conclusion can provide reference for similar fine-grained tailings geotextile dam construction.