Inverse identification of groundwater contamination source involves the reconstruction of pollution source features using limited and discrete observation data. However， when the simulation-optimization method is used to perform the identification task， the numerical simulation model must be invoked repeatedly， inevitably leading a large computational load. Moreover， when optimization model is solved， traditional particle swarm optimization algorithm commonly falls into the local minimum， which hinders precise identification. In the present study， a hypothesis case is designed to evaluate the performance of proposed framework. The numeric simulation model is embedded as equation constraint of optimization model where the objective function is the bias between simulation outputs and observations and the decision variables denotes features of contamination source. In particular， the optimization model is established to simultaneously estimate the release history of three potential contamination sources and hydraulic conductivity. To reduce the huge calculated burden， the BP neural network is introduced to substitute the simulation model. Furthermore， to alleviate being trapped into local minimum， adaptive weighted strategy is proposed to improve the particle swarm algorithm. The identification results indicate that： ① The BP neural network surrogate model can approximate the input-output relationship of the simulation model with desired accuracy of R square of 0.99， and the running speed of surrogate is evidently swifter than that of the numerical simulation model. ② Compared with the traditional particle swarm optimization algorithm， the adaptive weighted particle swarm optimization algorithm can substantially improve the convergence speed and optimal efficiency by adjusting the parameters and iteration termination conditions of the optimization algorithm， and the relative error of the optimal solution is less than 5%.

The experiment simulated the erosion development process of slopes under natural conditions through multiple precipitation intervals on the slope surface， which fully represented the process from rill to shallow rill. Three-dimensional scanning data of slope erosion gully morphology at different erosion stages were obtained by sectional observation of slope erosion process. According to the change of width and depth of gully at different erosion stages on slope， the development process of gully on slope from rill to shallow gully was studied. The results showed that at the beginning of rainfall， the width of erosion gully was obviously more than the depth， and the cross-section of erosion gully at the beginning was shallow and wide. With the increase in rainfall duration， the average depth of erosion ditch increased rapidly. At the end of the test， the average depth and average depth growth rate of the erosion ditch were more than the average width. During the development of trench to shallow trench， trench with width exceeding the upper limit of trench （20 cm） first appeared， and its proportion increased with the increase in rainfall duration. Shallow ditches with a depth more than 20 cm appeared later than those with a width more than 20 cm. The erosion rate of gully wall was relatively stable during the development of gully. As the soil bulk density increased with the increase in soil depth， the cutting erosion under the ditch slowed down with the increase in ditch depth. This led to the transition period channel in which the width exceeded the upper limit of the rill and the depth does not reach the lower limit of the shallow ditch depth. With the development of erosion process， the number and frequency of ditches in the transition period increased gradually. As the runoff was mainly concentrated in a few erosion ditches， the vast majority of rills would not develop into shallow ditches. These rills were either swallowed or maintained in the slow rill development stage for a long time. In the process of erosion， only a few rills could eventually develop into shallow gullies， and the maximum proportion of shallow gullies was only 2.3%. After the emergence of shallow ditches on the slope， the width and depth of erosion ditches with ditch depth less than 20 cm in the section of erosion ditches on the upper， middle and lower slopes of the slope still changed obviously. The development of erosion ditches mainly focused on the deepening and widening of the original erosion ditches. After the emergence of shallow ditches， the runoff on the slope was more concentrated in the shallow ditches， and more erosion was concentrated in the side wall erosion and undercut erosion of the shallow ditches. At the same time， rill erosion in other parts of the slope was weakened， resulting in the decline of the overall erosion speed and sediment yield of the slope.

To evaluate the variation of the relationship between rainfall and sediment yield， the trend of the sediment discharge pouring into Yellow River is of significance for the layout of flood control and sedimentation reduction system， as well as the determination of the Yellow River governance strategy. Four watersheds including the Jialu River， Qiushui River， Yanhe River and Malian River in the middle Yellow River are selected， and long series of hydrological data are used to analyze the variation characteristics of the relationship between rainfall and sediment discharge of the watersheds in different periods. The sediment discharges of hydrologic control station of each watershed are predicted on extreme rainfall conditions affected by the current underlying surface. The results show that the annual sediment discharge in the four river basins all present a decreasing trend. The highest decreasing magnitude reduction rate of the sediment discharge takes place in the Yanhe River. The annual effective rainfall in four watersheds is characterized by obvious stages， and the recent effective rainfalls are larger than the average values of accumulated year. For the three watersheds in the Hekouzhen-Longmen Reach， the relationship between the effective rainfall and the sediment discharge in the current period has changed significantly compared with the reference period， and the sediment-producing capacity has reduced evidently. The relationship between sediment discharge and graded rainfall is established， which can reflect the influence characteristics of different levels of rainfall on the sediment discharge in reference and current period. The effects of precipitation grades on sediment transport of the four watersheds in the current period are significantly lower than those in the reference period. The proportions of heavy rain and storm rain in Jialu River， Qiushui River and Yanhe River have increased from 38%~46% in the reference period to over 68% in the current period， and the high intensity rain has performed still as the main driving force for the sediment transport. The contribution of rainfall below moderate to the total sediment discharge has decreased significantly or even disappeared. The relationship between rainfall and sediment discharge in the Malian River watershed has changed very little， and the reduction in the impact of precipitation grades on the sediment discharge is less than those of three watersheds in the Hekouzhen-Longmen Reach， indicating that the risk of rainfall erosion intensity remains serious. Based on an analysis of hydrological frequency and the combination of historically measured extreme rainfall， the designed extreme rainfall conditions are proposed. Our calculation indicates that the possible incoming sediment discharges of Jialu River， Qiushui River and Yanhe River would be 27 million tons， 25 million tons， and 0.30 million tons， respectively， and the sediment discharges have reduced by 83%~91% compared with the reference period on the condition of the current underlying surface. Since the increased sediment yield induced by the riverside-bank or the gully-bank and the limited water-and-soil conservation measures of Malian River， the sediment reduction ratio of Malian River is 48% and the possible incoming sediment discharge would reach above 150 million tons.

Mangrove forests are unique and essential ecosystems which have exhibited pronounced spatiotemporal variations owing to human activities in recent years. By integrating long-term remote sensing data with measured bathymetric data， the current study investigates the evolution process of mangrove forests in Shenzhen Bay in the recent 30 years， and analyzes the impact of human activities on mangrove forests. The results show that the mangrove forests in Shenzhen Bay increased rapidly from 259.6 hm² in 1988 to 527.1 hm² in 2017. However， significant spatial and temporal variations were found in the area extent change of mangrove forests， of which the mangrove area in the Mai Po reserve steadily increased， accounting for 81.8% of the total mangroves gain. While the area of mangrove forests in the Futian reserve first decreased and then subsequently increased under the impacts of deforestation and reforestation. The implementation of the Shenzhen Bay Reclamation Project favored shoal deposition， which contributed to mangrove forests seaward expansion by providing additional suitable accommodation space. In addition， the morphological adjustments in the Shenzhen Inner Bay caused by altered water and sediment regimes after the implementation of the Shenzhen River Regulation Project have further influence on the mangroves distribution. Correlation analysis showed that the Shenzhen Bay Reclamation Project and Shenzhen River Regulation Project explained 78.4% and 8.5% of total mangroves gain respectively.

The ecological compensation of inter-basin water transfer project is an effective way to solve the coordinated development of water resources ecological environment protection and economic development between the project water source area and the water receiving area. Based on the evolutionary game theory， this paper constructs an evolutionary game model of ecological compensation for inter-basin water transfer projects， and analyzes the interest demands and compensation behaviors among the local governments in the water source area， the water receiving area， and the higher-level governments. An empirical research is done on the ecological compensation of the eastern route of South-to-North Water Transfer Project. The research shows that it is difficult to achieve the optimal stable equilibrium strategy （water source area protection， water receiving area compensation） in the water source area and water receiving area of the inter-basin water transfer project without the supervision of the higher government. The optimal stable equilibrium strategy is realized and the best social benefits are achieved. In addition， the weighted moving average method is introduced to measure the emission rights of the water source area of the eastern route of South-to-North Water Transfer Project from 2014 to 2019， which clarifies the changes of pollution discharge in the water source area. Based on the emission trading method， the ecological compensation standards of the whole water source area from 2014 to 2019 were determined to be 26， 113， 398， 507， 421， and 328 million yuan respectively， and the compensation standards changed with the change of water transfer； The scope of ecological compensation constraints and incentives for the East Route Project to achieve the optimal stable equilibrium strategy， clarifies that the higher the government imposes on the water source area or the water receiving area unilaterally fails to fulfill the obligation， the greater the penalty， the easier it is to achieve the optimal stable equilibrium strategy and restrict the incentive funds not less than twice the ecological and environmental protection investment in the water source area and the ecological compensation fee in the water receiving area are necessary conditions to achieve the optimal stable and balanced strategy， and provide a decision-making basis and thinking methods for the construction of the ecological compensation mechanism for the eastern route of South-to-North Water Transfer Project.

Taking Pinghutang River in Jiaxing City and its surrounding river waters as the research area， taking ammonia nitrogen and BOD5 content in water as water quality indexes， the coupling model of hydrodynamic and water quality of plain river network is established by using MIKE21， and the migration and diffusion process of pollutants in the river network model is simulated. By analyzing the accident discharge process under different working conditions， it is concluded that under certain conditions， the diffusion area of pollutants per unit time is obviously linear with the diffusion time. When other conditions are the same， the diffusion area of pollutants at the same time has a four-time correlation with the concentration of accident discharge. Under the condition of keeping the total emission constant， the diffusion area of pollutants in a certain period of time has a great relationship with the emission pattern. The results show that the model has good adaptability and high accuracy in this study area. The results can provide reference for simulation analysis of pollutant diffusion in plain river network.

Vegetative filter strips （VFS） are areas of permanent vegetation established between agricultural field borders and waterways， which have been recommended as one of the best management practices for retaining nutrients and sediments from surface runoff. Grass filter strip is a common type of VFS for its tall， erect， stiff-stemmed and native perennial grasses. Many laboratory and field studies indicate that grass filter strips can significantly reduce sediment and nutrient loss in runoff， while grass filter strip functions as a natural terrace， where sediment and other pollutants are removed from runoff by infiltration， deposition， absorption， adsorption， decomposition and volatilization. Natural rainfall， simulated runoff and simulated rainfall are the main experimental ways to investigate pollutants interception in VFS. The advantage of simulated rainfall experiments is the ability to control VFS scale， rainfall parameters and underlying surface conditions， thus the data obtained from simulated rainfall tests is generally less variable and easier to interpret. The study is conducted at an experimental plot next to Kunyu River in Beijing. The plot is an area of 4 m by 14 m with a slope of 5.0%. The plot is divided into two parts， including one simulated rainfall run-off generation area （4 m by 2 m） as the pollutant source area and one runoff transport area （4 m by 12 m） as the treatment area. There are four treatment strips in runoff transport area. One strip is the control system without vegetation， while the other three strips are parallel grass filter strips with the same vegetation composed of bermuda， tall fescue and white clover. A series of simulated rainfall experiments are designed to compare the transport of pollutants from strips with and without vegetation. This study aimsat evaluating the interception effects of the grass filter strip on suspended solids， nitrogen and phosphorus in the surface runoff formed by rainfall hydraulic erosion so as to provide a reference for pollutants trapping in the grass filter strip. The results show that：① In three simulated rainfall experiments， the outflow concentrations of SS， NH₄ ⁺-N， TN， TDP， TP and COD_Mn from grass filter strips are less than those from the control system significantly， except for NO₃ ^--N. Grass filter strips have no significant effect on NO₃ ^--N concentration reduction. ② Compared to the control system， the average outflow concentrations of SS， NH₄ ⁺-N， TN， TDP， TP and COD_Mn from grass filter strips has decreased by 80.80%， 12.23%， 39.33%， 12.46%， 61.44% and 83.34% respectively. ③There existed particulate and dissolved forms of pollutants in the surface runoff. Grass filter strips have more efficiency in the concentration reduction of particulate pollutants such as SS and COD_Mn， compared with dissolved pollutants such as NH₄ ⁺-N， TDP and NO₃ ^--N. The contents in TP and TN are sediment bounds， thus their concentration reductions are higher than dissolved pollutants. In dissolved pollutants， the concentration reductions of NH₄ ⁺-N and TDP with stronger absorptivity are higher than NO₃ ^--N with weaker absorptivity. The results indicate that the grass filter strip is an efficient measure for reducing contamination from non-point source pollutants deriving from soil erosion when rain falls.

The extraction of river regions from remote sensing images is of great theoretical and practical significance for grasping the hydrological characteristics， environmental protection and development of a certain region. The ground objects in remote sensing images are complex and diverse. In order to obtain high-precision river area information and make up for the limited applicability of single-characteristic river extraction， a river region extraction method that integrates texture， spectral and shape features is proposed. Firstly， the infrared channel of the image is selected as the input， the gray level co-occurrence matrix is calculated according to the texture feature， and the angular second-order moment map in the statistics of the gray level co-occurrence matrix is selected as the processing object. Then， maximum inter-class variance （OTSU） calculation is conducted according to the diagonal second moment map of the spectral characteristics to obtain a binary image. Then， the connected domain is marked， and a specific filter is constructed according to the geometric features to filter out the non-river noise， and finally a high-precision river basin map is generated. Based on the remote sensing images of Taishan City obtained by the GF-1 satellite， the intercepted images include different types of river basins such as forest land， urban， mountainous， and cultivated land. This method and other algorithms for river extraction is used and then the performances are compared. These include the OTSU algorithm commonly used in remote sensing image segmentation， the SWT algorithm proposed by some scholars in recent years， and the U-Net algorithm in the fully convolutional neural network segmentation method. The results show that the method in this paper has obvious advantages in terms of accuracy and completeness. It is not easily affected by changes in the surrounding environment of the river， and has strong robustness. In particular， it has high practicability for extracting the global river basin of a certain region. The method in this paper comprehensively takes into account the texture， grayscale and geometric characteristics of the river， and carries out a comprehensive and effective feature description， which can maintain the integrity of the river basin to the greatest extent and can effectively suppress the noise adhesion phenomenon that occurs in other algorithms. It can effectively improve the accuracy and completeness of the river basin extraction.

Obtaining the permeability and spatial distribution of sediments accurately can provide a scientific basis for calculating the water conversion between river and groundwater and scientific measures are taken to improve and maintain the ecological environment of the basin. The vertical and horizontal hydraulic conductivity of the sediments in the lower reaches of the Hotan River are obtained by the in-situ standpipe test， and the comprehensive hydraulic conductivity of the sediments are predicted by the artificial neural network model based on the soil transfer function. The spatial variation law of sediments hydraulic conductivity in the studied river reach is systematically studied. The results show that：①The average vertical hydraulic conductivity is between 0.81 and 2.77 m/d， the horizontal hydraulic conductivity is between 4.32 and 10.37 m/d， and the average comprehensive hydraulic conductivity is between 0.54 and 1.30 m/d. The average value of the anisotropy ratio by horizontal and vertical of the riverbed sediments at each measuring point is between 3 and 9， and the maximum value is 9.09， which reflects the significant anisotropy in the profile of the riverbed sediment.②Along the longitudinal section of the riverbed from upstream to downstream， the average value of K_v and K_h show an increasing trend on the whole， the average value of K_s shows a trend of first increasing and then decreasing， and the anisotropy ratio of riverbed sediments shows a decreasing trend as a whole. The difference between K_h and K_s is much more than that between K_v and K_s . On the whole， the values of K_v and K_s are equivalent， but compared with the state without mud film， the value of K_s is closer to K_v in the sediment state covered by mud film.③The hydrodynamic conditions of the river affect the permeability of the riverbed sediments by shaping the erosion and deposition state of the riverbed and the characteristics of the arrangement and combination of sediment particles. The influence of the particle arrangement and combination structure characteristics of the riverbed sediments on the hydraulic conductivity is more significant in the river reach. These dimentary mud layer on the surface of the riverbed significantly reduces the vertical permeability of the riverbed sediments， which is the main cause for the anisotropy. The conclusion of the research provides data support for the calculation of river leakage in this reach， and provide technical guidance for taking scientific and reasonable engineering measures to improve and maintain the ecological environment of Hotan River Basin.

The three major resource systems of water， energy and food are closely related to each other and have an important bond relationship， the coupling and coordinated development of the three is related to the security and sustainable development of the national resource strategy. In order to explore the level of water-energy-food （WEF） system coupling and coordinated development in high-energy-consuming areas， this paper designs a comprehensive evaluation system with 24 indicators in three layers. Combined with the entropy value method and the analytic hierarchy method to optimize the index weights， based on the coupling coordination model and spatial exploratory analysis， the variation trend and spatial distribution characteristics of the coupling coordination degree of the WEF system in Hebei Province are calculated. From the time dimension， the development characteristics of the system as a whole and subsystem are compared and analyzed. Spatial autocorrelation test of coupling coordination degree of 11 cities in Hebei Province is carried out from the spatial dimension， and the spatial agglomeration patterns are explored. The results show that the comprehensive evaluation index of the WEF system in Hebei Province is small in the range of 0.42~0.58， the growth is slow and the long-term development level is medium， the coupling degree value of the WEF system as a whole and two subsystems are close to 1， and the coupling between the subsystems is highly coupled. The coupling coordination degree of the system as a whole fluctuates horizontally in the range of 0.64~0.73， the WEF system is transitioning from the primary coordinated development stage to the intermediate coordinated development stage， the coupling and coordination level of the water and food system has entered the highly coordinated stage， and the level of coupling and coordinated development of the water energy system and the energy and food system is relatively backward. In terms of the spatial distribution of coupling coordination degree， the northeast region of Hebei Province is low and low agglomeration， the central region is low and high agglomeration or high agglomeration， and the coupling coordination degree of most other cities does not show significant spatial agglomeration characteristics， and its spatial distribution has randomness， indicating that in addition to geographical spatial distance， the coupling coordination degree of WEF system is also affected by the industrial structure characteristics of each city and its own resource advantages.

China’s agricultural water rights market has developed rapidly， but the number and scale of agricultural water rights transactions are small， and agricultural water rights transactions are limited to some areas. Revealing the driving mechanism of agricultural water rights trading （AWRT） is of great significance for building an active agricultural water rights market. The driving mechanism of AWRT still lacks a systematic classification method， therefore the type and characteristics of the driving mechanism of AWRT cannot be effectively judged. A classification framework of the driving mechanism of AWRT is constructed from the two perspectives of transaction demand and transaction incentives of transaction subjects. The transaction demand is divided into four categories： the water rights allocation difference demand， the irrigation water use difference demand， the ecological restoration demand and the economic development demand. The transaction incentive is divided into three categories： the water rights constraint incentive， the water use cost saving incentive and the water rights buy-back incentive. A combination of transaction demand and transaction incentives yields nine types of AWRT driving mechanisms. Based on this method， the driving characteristics of typical cases of AWRT in China are analyzed. In addition， the driving path of intra-agricultural water rights trading and inter-industry AWRT is sorted out. The results show that the administrative cost of the water rights constraint type AWRT is high， and this type is only applicable to areas where water ecology has seriously deteriorated. Water rights repurchase type AWRT is difficult to sustain due to big financial pressure. Cost-saving type AWRT can generate sufficient economic incentives， which has the biggest development potential. This study can deepen the understanding of China's AWRT， and provide a basis for further quantitative research on the driving mechanism and trading potential of AWRT. It can also provide a reference for the construction of suitable agricultural water rights markets in different regions.

The carrying capacity of water resources is one of the key factors affecting the sustainable development of the region. The evaluation of the carrying capacity of water resources can provide a data basis for the spatial balance of regional water resources. Based on the framework model of “Water Resources-Economy-Society-Ecological Environment”， an evaluation index system for water resources carrying capacity in the five northwestern provinces is constructed， and the entropy TOPSIS model and the coupling coordination model are used to quantitatively evaluate the water resources carrying capacity in the five northwestern provinces from 2004 to 2019. The spatial-temporal evolution process of force and the coupling and coordination between subsystems， the obstacle degree model is introduced to obtain the obstacles to the improvement of the water resources carrying capacity in the five northwestern provinces. The results show that from 2004 to 2019， the water resources carrying capacity of the five northwestern provinces have been slowly increasing， from a water shortage state to a reasonable state of water resources. There is a clear gap in the water resources carrying capacity between Qinghai and Shaanxi. The carrying capacity level is high， but the gap between provinces shows a good trend of shrinking year by year； the carrying capacity of water resources， economy， society， and ecological environment system shows a year-fter-year growth trend， and the development level of the carrying capacity of water resources and ecological environment subsystems lags behind that of the economy. And the social system， the water resources carrying capacity subsystems of the five northwestern provinces are in a stage of high coupling and good coordination. The development level of the five provinces and autonomous regions shows a relatively balanced situation in the inter-provincial space. The amount of water resources per capita， the modulus of water production， and the effective irrigation rate， soil erosion control area， and nature reserve area percentage table are the main obstacles hindering the improvement of water resources carrying capacity in the five northwestern provinces. The research shows that the level of water resources carrying capacity and the internal coordination of the system in the five northwestern provinces and regions are currently at a high level， and the overall direction is advancing towards a healthy， coordinated and balanced direction. At the same time， there is still a lot of room for improvement in water resources management and regional water use transformation.

By using the methods of ensemble empirical mode decomposition and Mann-Kendall mutation test， the measured hydrological data of the top flow node and its upstream control station in the Pearl River Delta from 1959 to 2020 are analyzed， the distributary characteristics， correlation and trend change of the top node under the influence of geomorphic evolution and runoff-tidal current dynamic change are studied. The results show that because of the change of delta topography and its impact on the adjustment of runoff dynamics， the sudden change year of annual average diversion ratio and flood season average diversion ratio of Sanshui Station is 1988 and 1991， no evidence shows that dry season average diversion ratio has significant abrupt change. However， due to the large scale topographic cutting in the Pearl River Delta， especially the dramatic geomorphic change at key nodes and the resulting adjustment of runoff-tidal current dynamics， the divergence at the peak of the Pearl River Delta in dry seasons from 2010 to 2020 was more unstable than that from 2000 to 2010. There is a strong correlation between Sanshui Station and Makou Station in flood seasons， and it began to show an increasing trend around 2005 with an annual increase of about 0.001 6. On the contrary， Sanshui Station， Shijiao Station and Wuzhou Station always showed a weakening trend， with an annual decrease of -0.004 7 and -0.003 3， Sanshui Station is increasingly affected by Makou Station year by year. The correlation coefficients of Sanshui Station， Wuzhou Station， Shijiao Station， and Makou Station all show high frequency variation cycles of less than 5 months and about 1 year， the diverting ratio of Sanshui Station showed more obvious short period variation than the other three stations and there were differences of flood and dry season and inter-annual variation inside. In the long run， the diversion ratio of Sanshui Station shows an increasing trend regardless of the annual average or flood season， and the diversion uniformity of Makou and Sanshui Stations keeps increasing. At present， the diversion at the top of the Pearl River Delta has changed from a simple runoff to a complex process of runoff and tidal interaction， and the influence of tidal dynamic factors on the diversion at the top of the delta cannot be ignored.

Jinghe River Basin is located in the middle of the Loess Plateau and is one of the birthplaces of China’s agricultural and animal husbandry civilization. Studying the change law of blue and green water in the basin provides an important idea for water resources allocation and water recycling in the Loess Plateau， and provides a guarantee for the final realization of the overall goal of “ecological protection and high-quality development of the Yellow River Basin.” Based on SWAT （Soil and Water Assessment Tool） hydrological model and SUFI-2 algorithm， firstly， the spatial distribution difference of blue-green water resources in Jinghe River Basin in typical years is analyzed. Secondly， the scenario setting method is adopted to set scenarios by fixing climate elements and land use elements respectively， so as to quantitatively analyze the impact of two driving factors such as climate change and land use change on blue-green water change in Jinghe River Basin. The simulation results show that： ① the typical years are determined by Pearson Type III frequency curve method： 1997 is a dry year， 2010 is a flat year and 2003 is a wet year. ② The amount of blue-green water in wet years is the most abundant： the amount of blue-green water in wet years is 5.4 times that in dry years and 1.8 times that in normal years.The amount of green water is 1.4 times that in dry years and 1.1 times that in normal years. The green water coefficient is the highest in dry years and the lowest in wet years. ③ The runoff of Zhangjiashan Hydrological Station from 1979 to 2019 is analyzed by Mann Kendall mutation test， sliding t test and cumulative anomaly method. The results show that 1996 is the mutation year of runoff of Zhangjiashan Hydrological station. ④ From 1979 to 2019， land use change had little impact on the temporal and spatial distribution of blue and green water in the Jinghe River Basin， and had a greater impact under climate change： the blue water volume and green water volume increased by 1.97 and 21.48 mm respectively， and the green water coefficient increased by 0.52%. The change trend of blue and green water under the joint change of the two was consistent with that under climate change， which showed that climate change had a stronger impact on the amount of blue and green water resources in the Jinghe River Basin.

Exploring the propagation pattern of meteorological and hydrological drought is important for early warning and disaster mitigation of hydrological drought. The standardized precipitation evapotranspiration index （SPEI）， standardized precipitation index （SPI） and standardized runoff index （SRI） are calculated for different time scales based on the month-by-month precipitation， temperature and runoff data from 1960 to 2018 in the Ganjiang River basin， respectively. The characteristics of meteorological drought and hydrological drought are compared and analyzed by using operational theory and the Copula function conditional probability method， as well as the propagation pattern of meteorological drought to hydrological drought. The results show that the changing process of meteorological drought and hydrological drought in the Ganjiang River Basin on annual and seasonal scales is similar， and the changing trends show the same， and the drought index reflects that spring and winter droughts in the basin show an aggravating trend， while summer and autumn droughts show a weakening trend， and the summer drought intensity described by the SPEI index is heavier than that described by the SPI and SRI indices， while the winter drought intensity reflected by the SPEI index is weaker than that reflected by the other two drought indices. 87 and 75 meteorological drought events assessed by SPEI and SPI， respectively， and 65 hydrological drought events assessed by SRI. Although there are more meteorological drought events than hydrological drought events， hydrological drought events have longer durations and higher intensities. Drought propagation time fluctuates greatly within the year， with propagation time between 1 and 2 months during the rainy season and up to 6 months during the dry season， and the effect of temperature on drought propagation time is most pronounced in November and December. The optimal joint distribution function of meteorological and hydrological drought is Frank-Copula， and the probability of triggering hydrological drought events under meteorological drought conditions increases with the worsening of meteorological drought， while meteorological drought under the influence of temperature is more difficult to trigger hydrological drought events， but more likely to trigger more severe hydrological drought events. The results of the study contribute to the early monitoring and warning of hydrological drought events in the basin.

Flood forecasting is an important non-engineering means to prevent and reduce flood disasters. The continuous improvement of automatic measuring and forecasting system of water and rain regime has laid a foundation for the acquisition of hydrological information in flood forecasting. Precipitation is one of the key factors of flood generation， the development and application of precipitation forecasting technology provide new impetus for the development of flood forecasting. The hydrological model for flood forecasting explains the process of flood generation from the perspective of rainfall-runoff and so on， and enriches the methods of flood forecasting， the development of land-atmosphere coupling technology points out a new direction for further improving the accuracy and forecast period of flood forecasting. With the continuous development and improvement of the above technologies， the accuracy of flood forecasting is constantly improved. In this paper， the research progress of flood forecasting precision is systematically reviewed from four aspects： the research progress of automatic rainfall forecasting system， the application progress of precipitation forecasting in flood forecasting， the research progress of hydrological model of flood forecasting， and the research progress of land-atmosphere coupling model， the development direction and research hotspot of flood forecasting in the future are prospected. The automatic forecasting system of water and rain conditions needs urgently to introduce new theoretical technologies and non-contact equipment to increase the density and intensity of information. The development of flood forecasting and the application of artificial intelligence in flood forecasting research and practical work have already contributed to the development of flood forecasting technology. The breakthrough provides a strong support. The land-atmosphere coupling technology has developed from one-way coupling to two-way coupling based on physical concepts， and single-station observation and simulated meteorological data are driven to develop in the direction of combining remote sensing and four-dimensional assimilation， providing a reference for future research on improving flood forecasting accuracy. The research results can provide technical reference for hydrology and reservoir dispatching scientists.

In order to achieve harmony between human and water and sustainable economic and social development， this paper designs a water resources carrying capacity evaluation model based on the system dynamics model （WRCC-SEM）， including economic and social modules， water resources pressure modules， water resources supporting capacity modules， water resources pressure analysis & carrying capacity calculation module. Taking the main urban area of Zunyi City as the study area， the WRCC-SEM model of the main urban area of Zunyi City is constructed， and the authenticity of the model is verified with the statistical data of the study area， and then the future water resources carrying capacity of the main urban area of Zunyi City is evaluated and analyzed. The results show that the model simulation values are basically close to the corresponding statistical values in 2015 to 2019， the absolute relative errors are less than 15% in each year， which proves the rationality of the model structure and the effectiveness of the simulation. From the perspective of the potential carrying capacity of water resources， the amount of water resources in the main urban area of Zunyi City can support economic and social development. However， from the perspective of engineering water supply， from 2020 to 2035， due to engineering water shortages， the pressure relationship of water resources is overload in some years， including the year 2035， the main urban area of Zunyi City will be limited in economic and social development. In order to increase the carrying capacity of regional water resources and coordinate the relationship between water resources and economic and social development， the managers need rationally arrange the scale of water source projects and the timing of their commissioning.

Engineering water shortage is one of the main types of water shortage. In order to solve the shortage of evaluation methods of water shortage in engineering water shortage areas， as well as the problems of imperfect indicators and adaptability in the evaluation system， the index of congenital conditions of engineering water supply based on GIS data and classification standard data analysis matrix are introduced， improved TOPSIS models are used to evaluate the engineering water shortage limitation on Hainan Island. At the same time， the obstacle degree model is constructed to analyze limiting factors. The results show that there is a weak evaluation grade in most cities and counties of Hainan Island， especially 1/3 of cities and counties such as Wenchang， Ding’an， Tunchang， Chengmai， Lingao and Changjiang are rated as “poor （II）.” The analysis of obstacle degree model indicates that the lowest obstacle degree is the water production modulus， which means the supporting capacity of water resources in Hainan Island is strong. The top factors of obstacle degree mainly include river slope， water consumption per 10 000 yuan of GDP， irrigation water usage per unit area of cultivated land， water consumption per 10 000 yuan of industrial value-added and river network density. Further spatial difference analysis shows that the cities and counties in the middle and south of Hainan Island have the highest obstacle degree of river slope and great difficulty in engineering water intake. The main limiting factors of cities and counties with “poor” evaluation grade are water consumption efficiency indicators such as water consumption of 10 000 yuan GDP， water consumption of 10 000 yuan industrial value-added and irrigation water usage per unit area of cultivated land. The restrictive evaluation method and index system of engineering water shortage studied and constructed can provide a supplement to the evaluation method of water resources shortage in engineering water shortage areas， and the obstacle degree model can provide a reference to solve the problem of engineering water shortage.

The blockage in the water distribution and supply pipe system will cause additional energy loss and water quality problems. The accurate simulation and detection of pipeline blockage are extremely important. The finite volume method Godunov scheme can effectively avoid numerical dissipation or dispersion problems caused by interpolation or wave speed adjustment of the method of characteristics. Firstly， the mathematical models of discrete and extended blockage are established， and the Godunov scheme is proposed. The accuracy of the models is verified by comparing model results with the calculation results of published papers and experimental data. The reasons for pressure changes of blockage are analyzed based on the wave propagation theory， and the corresponding blockage parameter identification method is given. The results of parameter sensitivity analysis show that the flow rate， coefficient of loss， and location of the blockage will affect the reflection pressure of the discrete blockage， and the size and length of the extended blockage will change the pressure fluctuation period.

In order to solve the problem of optimal operation for the tidal pumping station under the water supply mode， this paper proposes a real-time optimization strategy. First， the pump characteristic curves are converted into interpolable data sets， and a mathematical model of the pump is constructed by using a three-dimensional interpolation method. Then the optimization period is discretely processed to construct a state matrix， which expresses the operating conditions of the pumping station at the target period. Finally， the length of the operating period， the number of working pumps， and the blade angle are limited to the feasible solution set， and the genetic algorithm is used to find the optimal solution. The strategy supports the free selection of optimal time periods and the free definition of feasible solution sets， which fit the actual needs of engineering operation and scheduling. Calculation examples show that the method is fast， and the results are reasonable， which can meet the needs of tidal pumping stations’ real-time optimal operation.

In view of the characteristics of Dashijian Reservoir Project， such as large water level difference between upstream and downstream， wide dam site area and narrow valley in energy dissipation section， this paper adjusts various parameters of the spillway noseway according to numerical simulation analysis and verifies the whole physical model test， and determines that a new type of the spillway is designed with tongue flap-shaped noseway. The experiment shows that the new type of tongue damper can adjust the water drop point through three different radii of tongue， increase the distance of tongue， avoid the concentrated scouring of river bed by tongue， reduce the scouring of foundation of guard tank and scouring intensity of both sides， and achieve the best effect of energy dissipation and anti-scour.

Sandy soil is not conducive to the survival of vegetation as the poor water-holding capacity and coarse texture of sandy soil. The water-holding capacity of sandy soil would be improved if its grain size distribution could be adjusted reasonably， being suitable for the vegetation reconstruction. In order to find appropriate improving materials and mixing ratio to improve the water-holding capacity of sandy soil， the experiment is conducted to mix 0~50 μm bentonite and 150~200 μm sand particles. The bulk density， soil water content， total porosity， capillary porosity， non-capillary porosity and cumulative evaporation of the mixed soil are measured， and the effect of the addition of Bentonite on the water-holding capacity of sandy soil is analyzed. The results show that the addition of Bentonite significantly reduces the soil bulk density and non-capillary porosity， and increased the total porosity and capillary porosity of the soil， and improves the water holding capacity of the soil. With the increase in the proportion of Bentonite， the bulk density and non-capillary porosity of soil decrease gradually， and the total porosity and capillary porosity increase gradually. Compared with the control group， the addition of Bentonite increases the soil saturated water capacity and residual water， decreases the cumulative evaporation， significantly prolongs the evaporation time， and enhances the water retention capacity of the soil， which is more conducive to plants. With the increase in the proportion of Bentonite， the soil saturated water holding capacity and residual water shows a trend of increasing gradually， while the cumulative evaporation shows a trend of decreasing gradually. The addition of Bentonite can also make the loose aeolian sand grains on the soil surface stick together， make the soil surface become dense and hard， and restrain the evaporation of soil moisture to a certain extent， this slows the release of water from the soil， allowing the soil to retain more water over a longer period of time. The results can provide technical support for the ecological restoration of sandy soil in the arid area of northwest China.

Under the new background of ecological civilization construction and “dual carbon” goal， more and more attention has been paid to the study of carbon effect of land consolidation projects. Based on the theory of carbon effect， from engineering construction effect of carbon， the carbon effect of land use structure and farmland carbon effect on three aspects of the management construction of land reclamation carbon effect analysis model， with Kinghsien Renovation Project in Hebei Province as an example， by using material balance algorithm， ecosystem types， such as carbon effect of land reclamation project， carbon break-even analysis， the studied area， the time to reach the carbon break-even point is calculated. The results show that：①the total carbon emission of the project construction is 8 988 t， of which the shelterbelt carbon sink is 197 t， the net carbon emission is 8 791 t， the main source of carbon emission is lime and concrete. ②The carbon sink increment of land use structure is 1 611 t， mainly from the increase of cropland and garden area. ③In the carbon effect of farmland management， the carbon sink increment of farmland ecosystem is 2 662 t， mainly due to the improvement of crop economic yield. With the increase in cultivated land area， the corresponding agricultural input will increase， resulting in an increase of 111 t of carbon emissions generated by agricultural farming. ④Land consolidation has no negative impact on the ecological environment， and the time to reach the carbon break-even point in the study area is 2.06 years， that is， since the completion of the project， the carbon emissions generated by the construction can be offset after 2.06 years.

Affected by factors such as climate， soil and irrigation and drainage conditions， the Huaibei Plains of Anhui is a typical area where drought and flood occur alternately. The main drainage ditch is the main drainage project of farmland in the Huaibei Plains. Through the reasonable control of the drainage ditch， the water level and groundwater level of the ditch are properly raised， which is helpful to realize the comprehensive management of drought and flood. Combined with the MIKE model， this paper takes the main ditch control area in the Huaibei Plains as a unit to study the regulation effect of drainage control at the small watershed scale on farmland water resources and the impact on crop growth. Drainage and water storage control can effectively raise the ditch water level and groundwater level， but the magnitude of the uplift should be reasonably controlled according to rainfall and crop planting structure. Waterlogging and waterlogging reduction should be the main focus， the water level in the ditch should be more than 1.2 m， lower than the average ground elevation， and the maximum water storage level should be controlled according to the design storage level. During the non-flood season （October to May）， the water level of the ditch is mainly controlled by water storage. The maximum water storage level can exceed the design storage level by 0.5 m. When the water level in front of the gate is lower than the design storage level by 0.5 m， the gate can be closed for water storage.

Adjusting the opening of the control gate is significant for controlling the water level and achieving reasonable water delivery. In this paper， the flow characteristics of the control gate and rectangular diversion are tested at different flow and different gate openings in trapezoidal channels， hydraulic elements of the water diversion and the upstream and downstream are observed such as water depth， flow velocity， and the variation of water surface along the main channel is measured. In addition， the relationship between flow coefficient and gate opening and water depth in front of the gate is established based on dimensional analysis. The results show that at different discharges， the water depth along the trapezoidal main canal has the same variation pattern， and the overall trend is decreasing first， then increasing and decreasing， and a backwater is formed at the downstream of the control gate. Under the condition of free outflow， when the inflow flow is 25.35， 28.93， 32.38， 35.50， 39.70 L/s respectively. The critical value of gate flow and weir flow appear when the gate opening is 11， 11.6， 12， 13， 14 cm. The range of relative gate opening （e/h） is 0.792， 0.798， 0.808， 0.819， 0.823. It is different from the theoretical value （0.6）. The boundary between weir flow and gate flow is not a fixed value， and there is a certain gap between critical value and theoretical value. The critical value is affected by the channel shape， gate pier shape， flow rate and other factors. On the other hand， it is found that the flow coefficient （m） in the side channel flow formula is related to Froude number and relative gate opening （e/h） based on the dimensionless principle. The flow coefficient （m） decreases with the increase in the Froude number and relative gate opening （e/h） under different flow rates， but the relative gate opening has a greater influence on the flow coefficient than the Froude number. The flow coefficient is approximately inversely proportional to the Froude number under different discharges. The flow coefficient （m） relation with relative and gate opening （e） and water depth in front of the gate （h） is obtained， and the average error is 0.73%. The diversion flow ratio decreases with the increase in the relative gate opening at different discharges. In the case of large flow， the diverging ratio tends to change greatly， and the maximum decrease is 16.78%， which means that the gate opening has little influence on the split ratio at a small discharge. This study is of great significance to guide rational water allocation and control in irrigated areas.

In order to clarify the hydraulic characteristics of the bidirectional vertical axial flow pump device of the Weicun Hub Expansion and Reconstruction Project Pumping Station， the structural characteristics of bidirectional vertical pump device and bidirectional shaft tubular pump device are compared and analyzed from the aspects of pump device efficiency， unit equipment installation， maintenance and reliability， and the bidirectional vertical pump device is optimized. In order to further clarify the hydraulic characteristics of the bidirectional vertical axial flow pump device， the tests on the energy performance， cavitation performance and flight performance of the pump device model are carried out. The results show that when the blade placement angle is -8°~+4°， the maximum efficiency of the pump device is 74.26%， and the blade placement angle is -6° and the pump device head is 3.981 m； in the range of 2.72~5.25 m， the model pump device runs stably without obvious adverse noise and vibration. When the maximum head is 5.25 m and the blade placement angle is +2°， the submerged depth of the pump meets the requirement of the maximum required cavitation amount of 9.35 m， which meets the requirement of the installation of the submerged depth of the pump. When the blade placement angle is -8°， the prototype pump has a runaway speed of 1.705 times its rated speed. The hydraulic loss of bidirectional outlet conduit is the main factor affecting the hydraulic efficiency of such pump devices.

The mechanical vibration state of hydropower unit is an important embodiment of its stability. It is of great significance to carry out mechanical vibration state evaluation of hydropower units based on the start-up dynamic process. Firstly， according to the characteristics of rapid dynamic change of speed during startup， an amplitude frequency feature extraction method of variable speed dynamic signal based on hardware order tracking and order analysis is proposed， and two mechanical vibration state samples of vibration peak and peak trend and characteristic frequency amplitude trend are constructed. Secondly， the distance measurement method in data mining is introduced to construct the key performance index of mechanical vibration state evaluation based on the change trend of the unit peak value and multivariate characteristic frequency energy ratio. The data of a hydropower station unit after overhaul are verified and analyzed. The results show that this method can effectively track， monitor and evaluate the change trend of mechanical vibration state of the unit， find the possible safety hazards in the early stage of startup， and serve as a guide to the safe and stable operation of the unit.

The long-term operation of the turbine in the low load area will lead to large vibration and poor flow pattern of the hydropower unit， which will lead to cracks in the runner blades and cavitation of the unit. In order to explore the flow characteristics and pressure pulsation characteristics in different parts of the unit， this paper makes a three-dimensional modeling of a mixed flow hydropower unit in a power station， based on Reynolds time averaged N-S equation and standard SST k-ω Based on the turbulence model， the steady and unsteady solutions are calculated for the rated working condition and partial load working condition （45% rated working condition）. The results show that the flow in the guide vane is more turbulent under low load conditions， mainly because the opening of the guide vane does not allow the water to enter the runner fully， and the pressure change near the crown of the runner is chaotic and does not decrease according to the law， and the energy conversion is low， resulting in lower efficiency. From the pressure pulsation results can be seen： the maximum amplitude at the movable guide vane is more than 3 times the maximum amplitude at the fixed guide vane， indicating that the closer the runner， the more obvious the effect of static and dynamic interference； the maximum amplitude inside the runner is several times the maximum amplitude at the guide vanes； the amplitude of pressure pulsation under low load conditions is mostly much higher than the maximum amplitude under rated conditions， indicating that the stability under low load conditions is poor. The results of this paper provide a certain theoretical basis for setting the safe operating range of mixed-flow turbine unit at low load.

Electromagnetic interference is one of the important factors affecting the reliability of monitoring system of hydropower station， which will affect the safe and stable operation of hydropower station. In order to solve the above problems， this study analyzed the electromagnetic interference detection method of hydropower equipment in consideration of small signal stability. Firstly， the identification method of electromagnetic interference small signal stability characteristics is analyzed， and the UHF method and high frequency current method is used to find small signal characteristics for the collection of interference signals. Secondly， the scheme of collecting electromagnetic interference signals is studied based on the establishment of data acquisition cards and its correlation with small signals. Thirdly， the electromagnetic interference signals are classified， and the detection platform is simulated according to different levels of electromagnetic interference signals to match the corresponding detection simulation platform. Then， the small-signal stable data is accurately recorded. Finally， a comparative experiment is used to verify that the electromagnetic interference detection method based on small-signal stability has improved detection accuracy by about 3.5% compared with the traditional method， which is a hydroelectric power station. The results of this research serve as a more accurate solution to electromagnetic interference signals of equipment.

The arch dam is an important dam type among high dams with complex geometry， having different sections in the vertical direction and the horizontal direction. Tasks such as arch dam BIM modeling， pouring bin modeling and construction unit shape extraction involve many shape parameters， complex block conditions， and large amounts of data. Based on the secondary development of CATIA， this paper puts forward the method of parametric modeling of dam body， construction block modeling and shape parameter extraction. Based on an analysis of the parametric logic of the dam construction unit， this paper draws up the dam model sectioning plan and uses the CATIA Automation API interface to implement the arch dam parametric modeling， model sectioning and construction unit construction in the C# programming environment. As a result， attribute parameters such as the surface area， volume and corner space coordinates of the construction unit can be dynamically updated according to the needs of project management during the construction period， which can help predict the construction progress and strength of the dam and guide the construction in a scientific and orderly manner. It also provides an efficient and easy-to-use tool for sectioning complex arch dam model construction units and shape parameter analysis.

Based on the original design closure scheme， physical model test and numerical simulation of river closure for TB Hydropower Station in Lancang River are carried out， the particle size and intensity of dumping material required for closure under different test cases are determined， and the recommended scheme of closure is put forward. Due to the poor diversion conditions of the diversion tunnel during the actual closure construction， it can be seen from the numerical simulation analysis that the closure difficulty is greatly increased under different discharge flows of the upstream power station. TB hydropower station and upstream power station belong to the same basin cascade company. The shutdown and discharge control of the upper power station can reduce discharge， reduce the difficulty of the closure to the maximum extent， and ensure the smooth implementation of the closure. The research results show that the analysis method combining model test and numerical simulation can provide a scientific decision-making basis for the implementation of the closure project， and cooperate with the “control discharge and assist in closure” mode of the upstream and downstream power stations in the cascade development of the river basin， which serves as a reference for the design and implementation of similar projects.

The widely used data fusion strategy based on the COMTRADE model at present is affected by multiple heterogeneous redundant data， and the security of data fusion is difficult to guarantee. In order to solve this problem， this paper proposes a research on the security strategy and method of cross-zone data fusion of hydropower plants. Binary variables and covariance thresholds are introduced to construct interference event judgment matrix， and an event-driven filtering processing flow is designed to reduce the transmission of redundant events. A data fusion process based on improved rough set is designed by using two-way simultaneous search and mutual coordination. On this basis， the cross-region data security fusion system of hydropower plant is constructed， and the cross-region data security fusion of hydropower plant is realized according to the data synchronization mode of physical and chemical view and the data capture mode of trigger mode. The scheme of this research has perfect theoretical support and practical guiding significance.

The unevenness of the opening at the outlet edge of the hydro-turbine runner blade causes radial hydraulic force， known as hydraulic unbalance force， is not zero， which can lead to the strong vibration of the hydro-turbine generator shafting system （HGSS）. In this paper， the asymmetric airfoil NACA2415 is taken as the research object. First， in the Reynolds number range of 0.5×10⁵ to 3.5×10⁵， the S-A one equation turbulence model is applied to the numerical simulation of the unsteady flow over NACA2415 airfoil. And based on the numerical simulation data， the fitting formulas of the lift coefficient and lift-drag ratio of the airfoil varying with the angle of attack are obtained. Further， the nonlinear dynamic mathematical model of the hydraulic unbalance force of the asymmetric airfoil is established according to the Kutta-Joukowski theorem. Then， considering the chain reaction among the hydraulic， mechanical and electrical instability， a combined nonlinear mathematical model of the HGSS is established according to the Lagrange equation. Finally， by using numerical simulation， the bifurcation diagrams of the HGSS with the variation of the deviation of the blade exit flow angle under different Reynolds number are studied. Moreover， it is found that with the increase in the deviation of the blade exit flow angle， the axis coordinates of the HGSS may fall into bifurcation and chaos. In particular， the bifurcation and chaos phenomena are intensified when Reynolds number increases， which means that the vibration and operation instability are more likely to occur in the case of high Reynolds number. Besides， the stable operating ranges of the HGSS with the deviation of the blade exit flow angle are also obtained under different Reynolds numbers. But above all， these results can serve as a reference for the optimal design of hydro-turbine blades and the operation of hydropower stations.

In order to ensure the safety of hydropower plant network operation and aiming at the problems of poor detection function and operation performance of the existing hydropower plant network intrusion detection system， this paper uses Winute network threat perception technology to realize the optimal design of the system from three aspects of hardware， database and software function. In terms of system hardware， the optimization of system hardware equipment is completed by refitting network traffic collector and network data processor， and adding Winute network threat sensor as the operation environment of network threat perception technology. Then the system database tables are built from the four aspects of attack rule base， application identification rule base， URL filter base and virus base， and the connection between database tables is realized according to the internal relationship between data. The network intrusion detection standard of hydropower plant is set， and the collector equipment is used to capture the network traffic data of hydropower plant. The characteristics of density and gain in the network flow data of hydropower plants are extracted， and the network intrusion events are identified by using Winute network threat perception technology. Combined with the intrusion event identification results， the detection results including network intrusion status， intrusion type and other information are output through feature matching. The system test results show that compared with the traditional network intrusion detection system， the intrusion detection error and type detection error rate of the optimized design system are 2.8 and 2.3% lower， respectively， and the maximum number of concurrent connections and the throughput rate of the system are higher， that is， the intrusion detection function and operation performance of the designed system have more advantages.

At present， the alarm method of home-made computer monitoring system of hydropower station has such problems such as low alarm efficiency， high false alarm rate， and low alarm efficiency， which is difficult to ensure the normal operation of the system. In order to ensure the safe and stable operation of home-made computer monitoring system of hydropower station， this paper proposes a new abnormal behavior alarm method. Based on the time series， the time series data of the collection device is described by constructing a time series symbol conversion model. The morphological characteristics of the device are combined with feature recognition technology and abnormal behavior detection technology to accurately judge the operating state of the device， and the alarm line is preset to realize automatic alarm. Experimental results show that the alarm method has strong adaptability， with an average effective alarm rate of 94%， an average false alarm rate of 8%， and an alarm efficiency of about 7.8 units/s. It has high alarm performance and application value.