To investigate the secondary formation of chloroform under the influence of colloidal effect during managed aquifer recharge， the formation of chloroform under different hydro-chemical conditions with and without colloids is investigated by batch experiments. Besides， the synergistic mode between chloroform and silica colloids is analyzed by tangential flow ultrafiltration technique combined with atomic force electron microscopy（AFM）. The results indicate that the formation of chloroform increases with the increase in contact time. High chloride/TOC ratios， high pH， and low ionic strength can promote the formation of chloroform in a specific contact time. The presence of silica colloid would affect the adsorption of the reaction precursor， which inhibits the formation of chloroform at all times and has a strong adsorption effect on the chloroform in water. The results can provide a scientific understanding of the secondary formation of chloroform during artificial recharging.

Aiming at the non-linear， time-varying and large delay characteristics of the phosphorus removal process from phosphate rock and the problems that the single PID control is prone to overshoot， this paper proposes a research method that combines fuzzy control technology with PID control， and designs a set of automatic control system for phosphorus removal from phosphate mine wastewater based on the adaptive fuzzy PID algorithm. Through an analysis of the past data， the membership function in the fuzzy PID controller is set， and the parameters in the fuzzy rules are modified in the real time so as to achieve the effect of self-adaptive optimization of control performance. MATLAB simulation analysis and experimental testing show that the control system has good stability and self-adaptability， and can achieve the effect of efficient use of phosphorus removal agents and economic efficiency.

The land use mode directly affects the health of the ecosystem. The ecological health evaluation of land use is of practical significance to promoting the sustainable land resources utilization.This paper applies the Markov theory to construct the land use conversion matrix. 19 indexes are selected to construct evaluation index system for land use ecological health. A dynamic evaluation method is put forward. Taking Weihui County， Henan Province as an example， the evaluated results show that from 2010 to 2020， land use and transformation is mainly concentrated in 7 types， i.e. agricultural， residential， industrial， fruit and forest land use. The health ecological types accounted for 40.51% and 36.20% of the total evaluated area both in 2010 and 2020 respectively，which decreased by 6.31% with the comparison of two periods. The results have certain reference value for improving land ecological health and promoting local sustainable economic and social development.

In order to clarify the influence of the operation of the sluice gate and the diversion pump station on the navigation flow conditions in the entrance area of the ship lock approach channel after the completion of the Madian Water Control Project， the SMS（Surface grater Modeling System）hydrodynamic software was used to numerically analyze the flow field characteristics in the entrance area of the upstream and downstream approach channels of the ship lock under three different operating conditions. The results show that in the Madian hydro-junction project， the flow conditions in the upper approach channel entrance area of the lock are good under the check condition of the flood discharge of the lock， and there is a large lateral velocity in the local area of the downstream approach channel entrance area. In the check condition of lock diversion， the lateral velocity in the local area of the upstream approach channel entrance does not meet the requirements， and the flow condition in the downstream approach channel entrance is good. In the water diversion condition of the pump station， there are some oblique flow near the end of the water partition wall near the side of the pump station in the upstream and downstream approach channel of the lock. It is suggested to increase the arc vertical separation facilities at the end of the water partition wall of the upstream and downstream approach channel. The conclusions provide a certain reference for the operation of Madian hub project.

The channel morphology and flood control capacity of Shenzhen River have changed significantly in the past decades owing to the Shenzhen River Regulation Project and river basin urbanization. Based on measured water， sediment， and bathymetric data， this paper systematically analyzes the recent erosion and deposition patterns of Shenzhen River and the changes of flood control capacity in Shenzhen River and its back-siltation characteristics after the implementation of different channel dredging schemes is explored by using a two-dimensional numerical model of flow and sediment. The results show that significant back-siltation occurred in Shenzhen River during 2007-2017， with an accumulated deposition volume of 74.6×10⁴ m³. The morphological changes featured alternate channel erosion and deposition and consistent shoal deposition， with the magnitude of erosion and deposition gradually decreased from the upstream to the downstream. The reduced sediment transport capacity after the implementation of the Shenzhen River Regulation Project was the main reason of recent rapid deposition in Shenzhen River， and the deposition rate was associated with the flood tide discharge at the river mouth and the upstream sediment load. The flood control capacity of Shenzhen River decreased owing to channel deposition， the proportion of river section on the sides of Shenzhen and Hong Kong that currently meet the 50-year return period of flood control capacity are 63.2% and 56.8% respectively. Modeling results show that the improvement extent of flood control capacity of different channel dredging schemes is positively correlated with their back-siltation amount. The optimization suggestion for the improvement scheme of the flood control capacity of Shenzhen River is therefore given from the view of benefits of flood control and economic costs.

The performance of the bioretention system is significantly affected by the soil media characteristics. In order to standardize the gradation of packing soil media and then improve the construction level of bioretention facilities in Chongqing urban area， two types of traditional bioretention soil media （BSM1 and BSM2） are mixed with loamy sand and fine sand in a mass ratio of 1∶9， and sandy soil and fine sand in a mass ratio of 2∶8， respectively， combining the results of the soil background survey in Chongqing. And on this basis， vermiculite （Ver.）， biochar （BC） and perlite （Per.） are added as modifiers into BSM1 and BSM2 in different mass ratios， which finally configurates 6 types of modified soil media. Furthermore， the hydraulic permeability and pollutants removal performance of bioretention system with different soil media are investigated. Results show that different soil media shows different initial hydraulic permeability coefficient and attenuation performance， among which， the hydraulic permeability of BSM2+8%Ver. soil medium changes in a stable range of 41~53 mm/h， while the permeability of BSM2+8%BC soil medium has the highest attenuated rate. All soil media show an effective removal performance in runoff pollutants， especially in suspended solids （SS）， and the effluent turbidity can reach 10~16 NTU. The removal capacity of the bioretention columns packed with BSM1 and its modified groups in organic pollutants and nutrients are superior to the other columns. Finally， this paper considers the factors of soil medium decontamination capacity， hydraulic permeability and price， the comprehensive ranking of modified soil media using the projection pursuit method： BSM1+2%Ver. （0.60）>BSM1 （0.52）>BSM1+2%BC （0.41）>BSM1+2%Per. （0.40）>BSM2+8%Ver. （0.33）>BSM2+8%BC（0.23）>BSM2 （0.15）>BSM2+8%Per. （0.14）.

The ecological environment in arid and semi-arid areas is relatively fragile， and the groundwater level decreases obviously due to the low efficiency of groundwater exploitation and irrigation. The research on the interaction mechanism between groundwater resources and land use types in these areas is still relatively limited， and a further exploration is needed. This paper takes the Kashgar Delta as the research area， combining Kriging interpolation method， land use dynamic degree model and transfer matrix to study the relationship between groundwater depth and the land use change and the impact mechanism of land use change on groundwater resources. The results show that ：① during 2010-2018， the area of cultivated land and construction land in Kashgar Delta is increasing， while the area of forest land， grassland and unused land is decreasing. ② The depth of groundwater is high in the northwest and low in the middle and south， and the depth ranges from 0.5 to 31 m. The depth of underground water in the north is between 10~31 m and that in the middle and south is shallow， generally between 0.5~10 m. ③ Compared with 2010， the groundwater depth of 5~10 m accounted for an increasing proportion of all land use types in the study area in 2018， while other land use depths， especially 0~5 m， accounted for an increasing proportion. The average groundwater depth range of different land use types was 5~10 m. Groundwater resource is the main water supply source and irrigation water source of cultivated land in the study area， the increase in groundwater depth corresponding to cultivated land area is more obvious. The results indicate that the groundwater depth in arid and semi-arid areas is sensitive to the changes in land use types caused by human activities， and the groundwater resources in these areas are closely related to land use types. This study provides a strong theoretical basis for protection and rational development of groundwater resources and the construction of regional ecological civilization in arid and semi-arid areas.

The water management of the Baiyangdian River Basin has entered a new era of implementation of the river chief system， and there are problems such as the incomplete implementation of policies and insufficient public participation. Based on the theory of rural governance by using 288 effective survey questionnaires of farmers in Baiyangdiandian District as samples， the binary logistic model is used to analyze the impact of farmers′ personal characteristics， psychological cognition， government means， and right protection on farmers′ willingness to participate in water environment governance. The results show that： ① In terms of personal characteristics： the individual characteristics of farmers have no significant influence on the willingness to participate in water environment governance； ② In terms of psychological perception： the willingness of farmers to pay for water governance has a significant negative impact on participation in water environment governance； the current situation of farmers in water environment governance cognition has no significant impact on it； ③ In terms of policy measures： farmers′ satisfaction with government water governance policies and responses to people′s demands has a significant positive impact on their willingness to participate in water environment governance； ④ In terms of right protection： government′s treatment of farmers， the protection of the right to livelihood of the water environment and the right to know has a significant positive impact on the willingness to participate in water environment governance， and the government′s protection of the water environment litigation right has no significant impact on it. Therefore， it is proposed that the government and farmers participate in the dual water environment co-governance progressive mechanism， as well as the horizontal and vertical progress of policies， and the evolution path of the river chief system in which farmers participate in the "Three Peoples" and gradually upgrade to a multi-faceted good governance type.

In order to study the influence of the tip clearance on the hydraulic performance and flow stability of the submersible sewage pump， a double-blade semi-open impeller submersible sewage pump is taken as the research object， taking the tip clearance as the design variable to perform numerical simulation of solid-liquid two-phase flow in the whole passage， and the influence of the clearance on the external characteristics of the pump and the distribution of the solid phase， pressure， velocity and leakage vortex in the impeller are analyzed. The results show that the clearance has a great influence on the flow characteristics in the pump. When the relative tip clearance expands from 0.58% to 1.74%， the head significantly decreases under the low flow rate condition， and the optimal efficiency point tends to shift to the low flow condition. The solid particles are concentrated in the impeller， resulting in a decrease in the performance of conveying solid particles. At the same time， the pressure distribution law in the flow passage of the impeller is characterized by a decrease in pressure gradient and deviation of the local low pressure area toward the outlet direction. With the increase in the relative tip clearance， the axial vortex in the impeller gradually intensifies， and it develops to be more severe when the relative tip clearance is 1.16%. As the size of the leakage vortex gradually increases， and a slender leakage vortex appears when it is 1.74%.

Under the background of climate change and rapid urbanization， urban waterlogging has become an “urban disease”. For identifying waterlogging risk area and prediction research， this paper explores the characteristics of waterlogging risk area， the main urban space factors affecting the waterlogging of flood season rainfall， runoff coefficient， highway system， drainage system and flood point distance and bridge from the perspectives of urban space factors， six areas as the study area， north of the capital collected waterlogging data information and the urban space factors， the binary logistic regression analysis， and waterlogging risk area prediction model is established by using the network map of water of water points for validation， results show that the prediction model is correct. In reality， rainstorm weather can be based on the forecast model in high-risk areas or areas with high probability of waterlogging prevention and control measures.

In fresh water， the difference of temperature and turbidity is an important factor for density difference and density flow. In order to find out the spatial-temporal differences and influencing factors of density flow in Xiangxi Bay， based on the monthly hydrodynamic monitoring data of Xiangxi Bay in 2018， hydrodynamic processes such as water temperature and turbidity are analyzed and the quantitative calculation is carried out. The results show that density flow is mainly caused by the temperature difference and turbidity difference in Xiangxi Bay. In the flood season， turbidity difference is the main reason for density flow in the flood season， and sediment concentration of CJXX and XX00 has a more significant effect on water density than temperature， the contribution rate， which is caused by sediment concentration on water density， is up to 96%. The contribution rate caused by water temperature difference to water density， is more than 95% in each month. When the ratio （The ratio of turbidity difference ΔZD to temperature difference ΔT） is between 167.27 and 357.27， the sediment concentration difference \mathrm{\Delta } {\rho }_S generated by turbidity difference is equal to the water density difference \mathrm{\Delta } {\rho }_T generated by temperature difference， both of them have the same effect on the water density. When the ratio is less than 167.27， the contribution caused by temperature difference to water density， is significant. When the ratio is more than 357.27， the contribution caused by sediment concentration difference to water density， is significant.

By taking Jiangsu Province as the research area， this paper collects 15-day weather forecast data from China Weather Network for 2019-2020 based on the actual temperature data and the ET₀ value calculated by the HS formula， and uses RMSE， MAE and other indicators to evaluate the ET ₀ forecast effect. The precipitation forecast is analyzed and evaluated by using indicators such as accuracy rate， false negative rate， false alarm rate and ROC curve. The results show that the accuracy of daily low temperature forecasts is higher than that of daily high temperature forecasts. The accuracy of forecasts for the same number of days differs by about 10%. The RMSE value of the forecast error of the future 1~7 weather in Jiangsu Province is below 4.0 ^oC， and the MAE value values below 2.0 ^oC can be used to calculate the ET ₀ forecast； the MAE values of the ET ₀ forecast errors in the next 1~7 days in Jiangsu Province are all within 1.5 mm/d， and the RMSE values are basically within 2 mm/d The forecast accuracy in the eastern part is higher than that in the central and western parts； the ROC curve of the precipitation forecast for the next 14 days in Jiangsu Province is above the diagonal， which is of forecast value. The future precipitation forecast in Jiangsu Province underreports the precipitation of moderate rain and above. The lowest rate and empty report rate are 0.08 and 0.34， respectively， and the province’s forecast accuracy rate for the next day reaches 0.9. The accuracy rate of precipitation forecasts for the next 14 days is not much different from north to south.

In order to avoid the diversion and backflow caused by the asymmetric layout of the pump-sluice pivot， the rationality of the whole plane layout of the original design pump-sluice is verified by the physical model. It is suggested that the asymmetric pump-sluice layout should be optimized by adding a secondary stilling basin by the side of the sluice and a diversion pier at the forebay of the pumping station to improve the water flow conditions of the pump and sluice， reduce the maximum flow velocity， eliminate backflow， reduce the cost of bottom protection and slope protection， and ensure the safe operation of the project. The calculation coefficients of weir flow and orifice flow are put forward to provide technical support for project operation.

The rock and soil body composition of the dam abutment is complex， and the structures are quite different. Under the action of hydrodynamic and hydrostatic pressure， the safety problem of multiple leakage in the joint area， the seepage investigation of a reservoir dam is taken as the research object， the spatial location information and causes of dam hidden dangers are explored by means of measurement， geophysical exploration，drilling and hydrogeological test. The survey results show that the UAV photography technology clearly shows the location relationship between the leakage point， the collapse area and the landslide body， so as to guide the geophysical prospecting work. The focus is the left dam section of the dam. Parallel electrical method technology has the ability to quickly obtain the resistivity distribution of the whole dam. Combined with transient electromagnetic technology， the location and area of the verification borehole can be further determined. Drilling in the section of the pile number K0-000-0+040 m reveal that the dam foundation is mainly composed of block stones， and the electrical conductivity relationship between drilling， collapse area and leakage point water is established through tracer test. It is deduced that leakage is caused by the failure of the inclined wall impervious body in front of the original landslide and the formation of the dam leakage channel in the block stone layer. According to the detection research results， the path and distribution range of the dam leakage channel are determined， which provides a target area for the directional seepage control of the reservoir dam leakage.

To explore the contribution of the forebay’s shape optimization on energy conservation and consumption in the pumping station and based on the numerical simulation method， the mechanism of energy conservation and consumption reduction of the reconstruction of the pumping station forebay is analyzed， the third pumping station forebay of Yanhuanding Water Station Water Hub is taken as the research object to compare and analyze the flow pattern character before and after reconstruction. The standard k-ε turbulence model equation and the segregated solver are used to solve the discrete equations， and then the flow pattern character of the pumping station forebay before and after transformation is numerically simulated. The results indicate that the changes in the shape and size of the pumping station forebay have impact on the loss of energy consumption. There is less eddy in the front inflow forebay， and the uniformity of the axial velocity distribution is 50.2%， which is better than that of the lateral inlet （46.9%）. By using comparative analysis method， both the turbulent kinetic energy and the turbulent energy dissipation rate in the front inflow forebay are better than that of the lateral inflow forebay. The energy dissipation rate in the forebay tends to be stable with the running time， and the energy consumption of the forebay gets close to 492.81 W/m²， which is lower than that of the lateral inflow forebay （608.13 W/m²）. The energy-saving effect of the front inflow forebay is better than that of the lateral inflow forebay， and the flow pattern of the front inflow forebay is better than that of the lateral inflow forebay.

To study the characteristics and influencing factors of rainfall-runoff in the rice irrigation districts， a typical polder area of Gaoyou Irrigation District was taken as the experimental area to carry out different scale drainage monitoring. The farmland waterlogging process model was used to simulate the influence mechanism of rainfall and water management of paddy fields on farmland rainfall runoff. The results showed that rainfall was the main factor affecting the drainage amount of the paddy field， and the total drainage amount increased with the increase of rainfall. The drainage amount of the paddy field was 163.80， 135.22 and 113.78 mm， respectively， in the recurrence period of 50， 20 and 10 years. Rainfall type has little effect on the total amount of paddy drainage but has a significant effect on the distribution of the paddy drainage process. When the initial water storage depth increases by 1cm， the increase of paddy drainage is about 10mm. Reasonable regulation of paddy water can reduce paddy drainage and increase flood control and drainage standards in irrigation districts. A scale effect in the drainage of plain rice irrigation areas was found in our study. In 2013 and 2014， the drainage of branch ditch scale decreased by 35.8% and 38.9%， respectively， compared with the paddy field scale.

Taking a shaft tubular pump station as the research object， four inlet channel schemes， three outlet channel schemes and five front guide vane schemes are designed according to the distance between the shaft tail and the impeller center， the upper and lower contraction profile， the unilateral diffusion angle of the outlet channel， the position and size parameters of the front guide vane. The schemes are calculated and analyzed by CFD numerical simulation method. The accuracy of the numerical results is verified by model tests. The results show that the distance between the shaft tail and the impeller center mainly affects the velocity uniformity of the outlet section of the inlet channel. The smooth upper and lower profile of the contraction section of the inlet channel can reduce hydraulic loss. When the unilateral diffusion angle of the outlet channel is reduced to 4.52°， the flow separation phenomenon in the channel is obviously weakened. The installation of the front guide vane will reduce the efficiency of the pump device by about 1%. With the decrease in the distance between the guide vane outlet and the impeller center， the flow velocity uniformity at the outlet section of the inlet channel will decrease slightly. The longer the guide vane， the greater the hydraulic loss.

Obtaining accurate monthly irrigation water consumption is of great significance for exploring the temporal and spatial variation of the irrigation water consumption in a river basin. A BPNN based on the TensorFlow architecture is constructed to calculate the monthly irrigation volume taking a case study over 24 irrigation districts in the Yellow River Basin. The data of crop types， rainfall， weather， and measured irrigation quotas in 24 irrigation districts are collected to train the model， verify the accuracy of the model， and analyze the temporal and spatial variation of the monthly irrigation water consumption in the Yellow River Basin in 2018. The results show that the irrigation water consumption of wheat and corn in the Yellow River Basin is the largest， accounting for 26.35% and 37.98% of the total irrigation water consumption. The irrigation water consumption of other crops is in the order of vegetable>oil crops>potatoes>rice>soybean>cotton. Irrigation water consumption shows an increasing trend first and then a decreasing one over time， reaching a peak in June. The spatial distribution of irrigation water shows a trend of gradual decrease from the northwest to the middle， and then to the east and west.

Waterlogging has always been a natural disaster that troubles the development of human society. The crop reduction caused by waterlogging has threatened food security. The research on risk index， risk level and risk evaluation of waterlogging has been one of the hot topics， which has important theoretical and practical significance. In this paper， the Luoshan Drainage Area in Four-lake Watershed is chosen as the study area. The waterlogging disaster index system is established， and the index weights are calculated by using the AHP and CRITIC method. Then the waterlogging risk assessment model is established， and the risk degree under different mitigation measures is analyzed. The results show that by improving the drainage capacity to the planning level， increasing the water level， and improving the water-drought ratio at the same time， the comprehensive waterlogging risk decreases the most in the study area. Compared with the current situation， the proportion of high-risk area decreases by 100%， 78.05% and 54.44% respectively when the rainfall intensity in the flood season occurs once in 5 years， 10 years and 20 years， and the proportion of the low-risk area decreases by 72.96% when the rainfall intensity in thr flood season occurs every 3 years.

In order to explore the dominant characteristic factors of farmland soil salinization during the growth period of sunflower under the condition of concealed pipe drainage in Hetao Irrigation District （HID）. Based on the field test data in 2020， the total amount and composition of salt are analyzed by Correlation Analysis and Principal Component Analysis （PCA）. The results show that during the growth period of the sunflower in the study area， the soil salt composition type of 0~1 m profile is mainly sodium salt， which experiences the transformation from soda type in the early growth stage to sulfate type in the late growth stage. In each layer of soil at the beginning and end of growth period， anions changes from HCO₃ ^- to SO₄ ^2-， and cations are always dominated by K⁺+Na⁺.With the increase in depth， the correlation between total salt content and soluble salt ions changes. The anion with the strongest correlation is SO₄ ^2-， and the cation is K⁺+Na⁺. Through a principal component analysis， the characteristic factors of soil salinization and alkalization are revealed. It is concluded that HCO₃ ^- can represent soil alkalization characteristics， SO₄ ^2-， Cl^-， K⁺+Na⁺ can represent soil salt composition and soil salinization trend， and the influence degree of salinization and alkalization on each layer of soil during the growth period is further calculated. The results can provide a theoretical basis for the rational layout of local underground drainage areas and the improvement of soil salinization.

The forecast error of runoff in multi-tributary rivers has complexity and multiple uncertainties. The research on the forecast error of runoff from multi-source rivers can provide a more reliable inflow basis for the formulation of water resources management scheme in the basin. Based on the t location-scale distribution with stability and robustness， a mixed distribution model of runoff forecast error with multiple sources was constructed to reflect the multiple uncertainties of runoff forecast error and alleviate the problem of incorrect distribution hypothesis.It is applied to 7 representative hydrological stations in the upper reaches of the Yangtze River， such as Pingshan Station and Zhutuo Station to study the error characteristics of runoff forecast in the flood season and the non-flood season. The results show that： \text{?} t location-scale distribution is basically suitable for the optimal marginal distribution of daily runoff forecast error series of hydrological stations in the flood season and non-flood season. The mixed distribution model of multi-source runoff forecast error at Zhutuo Station， Cuntan Station and Qingxiichang Station in the flood season has a good fitting effect， and it is second only to the optimal marginal distribution model in the non-flood season.The relative error of mean value and the coefficient of variation between the simulated value and the actual value is less than 11%， and the variation law of the skewness coefficient is similar， which can be used to modify the runoff forecast sequence and to provide a more accurate foundation for the development and utilization of water resources and optimal allocation in the basin.

The high-precision extraction of hydrological characteristics， such as water system route， small watershed boundary and the position of pour point， is the basis and premise of accurate analysis of various hydrological characteristics. In order to solve the problem of low accuracy extraction results of hydrological feature based on low resolution DEM data （SRTM DEM， ASTER GDEM）， this study first obtains relatively high accuracy DEM with spatial resolution and altimetry precision based on InSAR technology and Sentinel-1A/SLC data， and then compares and analyzes the extraction results of hydrological characteristics for small watershed based on different DEM data sources. Finally， the main conclusions of this study are as follows： ① Based on SRTM DEM， ASTER GDEM and InSAR DEM， the extraction errors of mean height of five sampling points are 8.1， 8.5 and 3.7 m respectively， and the InSAR DEM has the highest accuracy. ② Based on SRTM DEM， ASTER GDEM and InSAR DEM， the coincidence degrees of the six main water system routes are 74.12%， 85.50% and 88.36% respectively， and the coincidence degrees increases successively. ③ The mean error between the actual position of pour point and the extracted position of pour point from SRTM DEM， ASTER GDEM and InSAR DEM are 4.93， 4.69 and 4.63 km respectively， and the error decreases successively.

As the water source conservation area of Miyun Reservoir， Zhangcheng Area in the upper reaches of Miyun Reservoir bears the important task of providing drinking water to Beijing.The study of water supply service in Zhangcheng Water Source Conservation Area is of great significance to the high-quality development and ensuring the water ecological security of the capital. Based on the remote sensing the land use change in Zhangcheng Water Conservation Area were interpreted and analyzed from 2000 to 2019. Based on the InVEST water yield model the spatial distribution and change of water yield in the study area from 2000 to 2019 were quantitatively evaluated， and the relationship between water yield change and land use evolution was discussed by using scenario analysis.The results show that ： ① From 2000 to 2019， the overall land use in the study area was characterized by the decrease in plow land， and the increase in forest land， grassland and construction land. ②The total water yield of the study area in 2000， 2010 and 2019 was 2.61， 3.46 and 2.71 million m³.The water yield in different periods showed spatial heterogeneity， which changed gradually from high in the west and low in the east into high in the east and low in the west. ③ The average depth of water yield of different land use types in the study area from large to small is： construction land， bare land， water area， plow land， forest land and grass land. When the evapotranspiration coefficient remains unchanged， the change of land use from land type with low water yield capacity to land type with high water yield capacity will lead to an increase in regional water yields， and vice versa.

Taking Tuhai River in north Shandong plain as an example to carry out flood resources joint regulation of sluice and dam group. Storge runoff during flood withdrawal considers control level， storage opportunity， ecological flow， and source use rate constraints. Flood regulation includes opening and controlling two discharge stages， using one-dimensional hydrodynamic method to simulate flood routing， using gate outlet and weir formula by iteration to calculate the opening process of the sluice and rubber dam. The results show that regulation storage in dry， normal and rainy year is respectively 74.6， 80.4 and 84.1million m³， the full storage rate will be higher with flood magnitude increasing， and the storage is basically full in normal and rainy years， the discharge meets ecological base flow with no artificial flood peak.

The strong nonlinearity presented by the runoff process constrains the prediction performance of the existing hydrological models. The artificial intelligence methods such as deep learning with strong nonlinear fitting ability can break through the current bottleneck to a certain extent. To effectively extract the nonlinear time-varying feature information of runoff sequences and improve the accuracy of runoff simulation and the multi-step-head forecasting performance， the ForecastNet based runoff prediction model with time-varying structure has been established. The Yajiang River Basin in the upper reaches of the Yalong River is taken as a case study， and comparative analyses among the ForecastNet， traditional hydrological model of SWAT（Soil and Water Assessnent Teol）， and neural network models of RNN（Recurrent Neural Network）， LSTM（Long Short-Term Memory） and RNN-LSTM are carried out. The results show that the ForcastNet model has strong applicability in long term prediction， and can effectively improve the accuracy of runoff simulation and multi-step-ahead forecasting， thus providing technical support for high-precision real-time runoff prediction.

Based on the data of multi-regional input-output table and water resources bulletins in the year of 2007 and 2017， the relationships related to water transfers between different regions are identified. Then the characteristics of cross-regional water transfer network and its determinants can be revealed with social network analysis and ecological network analysis. The results show that relationships related to regional water transfers have developed into network topology with a “core-periphery structure”. However， the stability and transmission efficiency still need to be improved. Jiangsu and Guangdong occupy the center of the production-based and consumption-based regional water transfer networks. Xinjiang， Heilongjiang， Hunan and Jiangxi play a more producer role with a high out-degree； Guangdong， Chongqing， Zhejiang and Shaanxi play a more consumer role with a high in-degree； Jiangsu， Anhui and Henan with play a more transit role a high betweenness degree. QAP analysis shows that geographical location， close economic linkage and high openness are all beneficial to the formation and evolution of cross-regional water transfer network.

Social progress and economic development are based on the rational allocation of water resources. In this paper， linear trend analysis， R/S analysis， Lorentz curve， Gini coefficient and Lorentz asymmetry coefficient are used to analyze the water consumption in recent 13 years since the water consumption correction and the water consumption structure in recent 7 years since the implementation of the strictest water resources management system， the results serve as a foundation for Qinghai Province to optimize and adjust water use structures and improve water use efficiency. The results show that water consumption has shown a downward trend in recent years and the water consumption for agriculture and industry will continue to decrease. Meanwhile， the water consumption for living and ecological environment will continue to increase. There are differences in water use structure in Qinghai Province， among which the spatial difference of domestic water is the largest. The spatial difference of industrial water， ecological environment water and domestic water is gradually narrowing in recent years.There is a large gap in water consumption per capital among regions in Qinghai Province， we should improve the efficiency of water saving and optimize the industrial structure to adapt to the limited amount of water resources of key development zones.

In view of the complex dam break process caused by heavy rain and floods， rapid changes in dam break flow and breach， large disaster hazards， this paper takes the actual dam break flood of Sheyuegou Reservoir as the research object by using measured and survey data. This paper simulates and analyzes Sheyuegou Reservoir flood formation， dam break process， dam break flow and mechanism based on a spatio-temporal variable source hydrological model and two-dimensional hydrodynamic method. The results of the study show that the simulated value of peak discharge is 1 915 m³/s， and the error is only 3.5% compared with the survey results. The extreme heavy rainfall in a short duration caused the flood that far exceeded the flood control capacity of the reservoir， which caused the dam to burst. The second and third teams of Ergong Village were affected most seriously， and the submerged water depth was between 2~6 m. Two flood peaks in the downstream river course and the narrowing of the downstream river course of the dam are important reasons， which cause economic and property losses in the downstream villages. The method adopted in this paper can better solve the calculation problems of the storm flood， dam break process and flood inundation process in Sheyuegou Basin. It can provide support and technical reference for dam break mechanism analysis and disaster loss assessment.

Under the new development pattern， the sustainable development of water resources has become the bottleneck restricting the high-quality economic development of Shandong Province. How to investigate the evolution direction and internal mechanism of water resources from a systematic perspective is of great practical significance for coordinating the contradiction between economic growth and ecological construction and comprehensively practising the new development concept. In view of this， through a specific analysis of the water resources system in Shandong Province， this paper hypothesizes the evolution direction of the water resources system： the evolution of the dissipative structure， and constructs the positive entropy measurement model （require pressure effective RPE） and the negative entropy measurement model （provide cushion harmonic PCH） respectively， combined with the escaped “Brussels device” model for a quantitative analysis. The results show that the water resources system in Shandong Province evolved to dissipative structure from 2006 to 2019， but the evolution efficiency is low. Finally， relevant suggestions are put forward for the phenomenon that hinders the evolution of water resources system to dissipative structure. In order to alleviate the shortage of water resources， the construction of the sponge city is strengthened and the key period of industrial transformation is smoothly spent.

Rainstorm often trigger floods that exceed the standard， causing the dam body to bear greater unfavorable external loads and affecting the safe operation of the project. For cemented sand gravel dam （CSG dam）， the surge in the water level has a great impact on the stress of the dam body. In order to study the response of the dam body caused by the change of the water level and to obtain the change process of the reservoir water level rising to the overflow of the reservoir， fluent software is used to simulate the process of flood entering the reservoir， and the damage constitutive model is established based on the material properties of the cemented sand gravel， and the characteristics of the dam body’s mechanical damage are analyzed under different water level changes. The results show that： ① Under design conditions， the CSG dam bears the greatest tensile stress at the upstream dam heel， and bears the largest compressive stress at the downstream dam toe， and the dam has the largest displacement in the middle of the dam， the contact surface between the body and the dam foundation bears relatively large unfavorable stress， the dam has no characteristics of damage； ② During the rainstorm， the water level increases， the dam body is overloaded， the displacement of the dam body and the compressive stress at the toe of the dam gradually increase， excessive tensile stress at the heel of the dam leads to deterioration of material properties and damage； ③ After the rainstorm， the main damage mode of CSG dam is tensile failure， and the damage area is mainly concentrated at the heel of the dam.

In the hydraulic model test of Edie reservoir， it was found that the repelled downstream hydraulic jump occurred in the gradual expanding stilling basin， and the flow state was poor， which could not meet the requirements of energy dissipation and scouring prevention. In order to solve this problem， numerical analysis software FLOW-3D was used to explore the influence of the body type parameters of the gradual expanding stilling basin on the flow state， inundation degree， energy dissipation rate and so on in the stilling basin， The results show that： for the gradual expanding stilling basin with the repelled downstream hydraulic jump， increasing the height of the tail can significantly improve the inundation degree， the appropriate height of the tail sill can not only improve the flow state in the stilling basin， but also control the energy dissipation rate to a reasonable range； the increase of diffusion angle and pool length has little effect on the inundation degree and energy dissipation rate and is not beneficial to improving the flow state. Based on the results， a reasonable shape parameters was recommended and verified by physical model. The research results can provide a reference for design optimization in the same type of engineering.

In order to make the intelligent detection and identification of abnormal features of massive UAV images， such as water conservancy project inspection， river and lake shoreline， and the realization of river and lake environmental monitoring， the enhancement of management efficiency， the satisfaction of demands for intelligent water conservancy construction based on YOLO v3 algorithm framework， a high-precision YOLO v3-SE target detection algorithm is constructed by the introduction of the SE， the UAV image recognition technology system is formed and it is also successfully applied to the detection and identification of massive UAV images in multiple water conservancy projects. It is shown based on the results that are compared with the original YOLO v3. SKSet-YOLO v3 and CBAM-YOLO v3 algorithm are improved by a self-built 66 000 picture data training set and a test set of 35 514 picture data， for the algorithm， the detection accuracy of 6 types of targets such as water， landslide， carrier， landslide， aggregate garbage and decentralized garbage has been greatly improved. With regard to the average detection accuracy mAP， it has also increased from 59.83% to 90.17%， from 79% to 90.17%， and from 72% to 90.17%.

In order to analyze the alkali-aggregate reaction hazards of natural pebble aggregate and excavated sandstone aggregate to be used in a large-scale water conservancy and hydropower project in Tibet， the alkali activity of aggregates is detected by lithofacies method and accelerated mortar bar method， and the inhibition effect of mixing grade II fly ash is studied. And the comprehensive properties of pebble or sandstone aggregate concrete are also evaluated. The results show that， both pebbles and sandstones are all alkali active aggregates with potential harmful reactions， and contain varying amounts of alkali-silicon reactive components， microcrystalline or cryptocrystalline quartz. The minimum dosages of single-doped grade II fly ash to effectively inhibit the alkali activity of pebble and sandstone are 25% and 30%， respectively. The alkali-silica reaction of pebble or sandstone aggregate can be effectively inhibited by using the strategy of “P·O 42.5 cement + 20% grade II fly ash”. Both kinds of aggregates can be used to prepare C25W6F200 concrete， which meets the performance requirements.

Research and engineering practice show that pouring temperature control is as important as surface insulation and water cooling for cracking-preventing of the concrete dam. However， most of the strict pouring temperature control indicators need the air cooling and ice mixing of aggregates， which will greatly increase the cost of concrete production. Taking the RCC dam bank slope dam section of DG hydropower station located at high altitude as the research object， this paper systematically studies the temperature， stress and cracking characteristics of the bank slope dam section concrete during construction by using the three-dimensional finite element method. On this basis， the effects of pouring temperature on the cracking characteristics of different parts of the dam are compared through multi-condition simulation. The results show that for the project， the pouring temperature of machine mixed abnormal concrete on the upstream and downstream surfaces needs to be strictly controlled， but the pouring temperature of RCC in the dam body can appropriately increase by 2~5 ℃ based on the design index.

The drawdown water level before the flood season is the key water level connecting the dry and flood seasons of the cascade reservoir. With regard to the drawdown level before the flood， most studies have emphasized the benefit of power generation before the flood season， regardless of the risk of spilled water in the main flood season， and it is difficult to consider the complex constraints such as unit maintenance， ecological flow， leading to the difficult application of the research results in practice. Therefore， with the goal of the maximum power generation and minimum spilled water for cascade reservoir， a double-layer optimal operation model is proposed： the outer layer optimizes the drawdown level before the flood， the inner layer uses DDDP-SA algorithm to optimize the operation trajectory from the dry season to the main flood season， and the output constraint of the cascade power station is determined by the unit maintenance of cascade power station， and determine the drawdown water level range of the cascade reservoir. In Qingjiang Cascade Reservoir， the recommended drawdown water level range of Shubuya Reservoir and Geheyan Reservoir is 368~371 m and 185.5~187 m， respectively. The proposed method can provide technical support for the formulation of the drawdown level before the flood in cascade reservoirs.

With the development of our country’s rural society and economy， long-term compensation and resettlement methods for immigrants have received attention and have gradually been applied in practice. However， due to the short implementation period， the risk mechanism is still unclear. This paper selects the SPA-ITFN model based on comprehensive weight combination method for analysis and evaluation in view of the long-term compensation and resettlement methods of reservoir immigrants. The model couples the set pair analysis method with the interval triangular fuzzy number， and uses the entropy method and the over-standard multiple method for combined weighting， which can effectively avoid the subjectivity of weights and the ambiguity of the adjacent levels of indicators in the current reservoir resettlement risk assessment and the limitations of the evaluation results. This paper takes six sample villages that implement long-term compensation and resettlement in the Hanjiang Gaobei Reservoir in Guangdong Province as an example to conduct a risk assessment. The results show that the long-term compensation and resettlement risks of the six sample villages， from small to large， are communities such as Jiulong Village， Dangxi Village， Beipu Village， Gongxia Village， and Dutou Village. It is basically consistent with the actual survey data and can be used for resettlement risk assessment of similar projects.