Climate change and human activities have caused extreme climate events to be non-stationary， and traditional stationarity analysis is unable to correctly reveal the true spatiotemporal variability. The daily temperature and precipitation data of 1960-2018 from 70 stations is used to calculate climate extremes commonly used in the Huaihe River Basin. Additionally， the non-stationary generalized extreme value distribution （GEV） and generalized Pareto distribution （GPD） model are used to interpret spatial distribution of return level. The main results are as follows： ① The low temperature extremes （Tmingpd and Tnn） shows a significant increasing trend， with a change magnitudes of 0.16 d/a and 0.04 ℃/a， respectively. Other climate extremes （Tmaxgpd， Txx， Pregpd and Rx1day） all show no significant changes. ② Different stationarity test will lead to a certain difference in stationarity results for a time series. In general， Tmingpd and Tnn are non-stationary， Preggpd and Rx1day are stationary， and Tmaxgpd and Txx are relatively complex with stationary and non-stationarity exist at the same time in the Huaihe River Basin； ③ The 30-year return level of low temperature extremes （Tnn， Tmingpd） change from south to north， reflecting the influence of latitude zonality. The high temperature extremes （Txx， Tmaxgpd） decreases from the east to the west， reflecting the thermal regulation of the ocean. The precipitation extremes （Rx1day， Pregpd） gradually decrease from southeast to northwest， reflecting the gradual weakening of the East Asian monsoon. The research can further improve the scientific understanding of the spatial and temporal distribution of extreme climate events， and provide scientific basis and theoretical support for water resources management and regional disaster prevention and mitigation in the Huaihe River Basin.

Different CMIP5 GCMs have different abilities to simulate precipitation， and the uncertainties are bigger. In this paper， Vine Copula is used to construct a multivariate joint probability distribution model between different GCMs and observed precipitation， and the conditional distribution function of the observed precipitation under the condition of model precipitation is deduced， according to which， the multi-model precipitation is synthesized to improve the accuracy of simulation and prediction. An example study is carried out by taking the area precipitation during 1961-2005 in the basin above Wuzhou Station as an example. The Taylor diagram is utilized to optimally select 6 GCMs with good comprehensive simulation ability from 12 GCMs， and then the joint distribution of observed precipitation and simulated precipitation of 6 GCMs is established by C-Vine and D-Vine Copula respectively. C-Vine Copula is selected as the synthesis model of multi-model precipitation based on AIC， BIC and log-likelihood value index. The results show that the accuracy of multi-model synthesis precipitation obtained by C-Vine Copula is higher than any single model， and also better than multi-model ensemble mean.

Drought is a natural disaster with complex formation mechanism， and its serious harm to human society promotes the development of drought prediction technology. In this paper， to improve the level of drought forecast in Huaihe River Basin， the standardized precipitation evapotranspiration index （SPEI） is used to monitor the drought condition， and the joint distribution of multi-dimensional variables is established by employing the regular vine （R-vine） Copula model. El Nino-Southern Oscillation （ENSO）， Central ENSO （ENSO Modoki）， North Atlantic Oscillation （NAO）， Indian Ocean Dipole （IOD） and Arctic Oscillation （AO） are taken as predictors to enhance the prediction accuracy of drought level transfer based on SPEI. The results show that the R-vine Copula model performs well in evaluating the impact of climatic indices on drought transfer probability， and that multi-dimensional indices can better improve the prediction accuracy compared with single-dimensional indices， which is consistent with the fact that regional meteorological drought is affected by various climatic factors. This paper provides new insights for research in drought transfer under the influence of multi-dimensional climatic indices.

In the traditional reservoir multi-objective optimal operation scheme decision-making， the influence of uncertainty inflow runoff is often ignored. To solve the problem， the interval number was selected as the evaluation index， and the upper and lower limits of the interval number were the maximum and minimum values obtained by the optimal calculation when the simulated inflow runoff considering the prediction error of inflow runoff was taken as the input. Based on the subjective weight of analysis method and objective weight of critical method， the game theory combination model was used to combine the comprehensive weights. The VIKOR model was modified to improve the reliability and rationality of evaluation index weight. Then this method was applied to the multi-objective operation scheme decision-making of Three Gorges reservoir. The result shows that the decision-making result is basically consistent with the actual runoff process. That indicates the rationality and practicability of the method， and provides a new way for the multi-objective operation scheme decision-making of reservoir when the inflow runoff is uncertainty.

Multi-source remote sensing precipitation estimates with high spatial and temporal resolution have provided hydrologists a potential alternative source for hydrological simulations in ungauged or poorly gauged basins， especially for the medium- and small-size rivers lack of consistent and long-term records. This study evaluates the accuracy and performance of the five remote sensing precipitation products （NLDAS2， Stage IV， TMPARP， TMPART， and IMERG） against CPC-U and then estimates their hydrological performance with the distributed hydrologic model DRIVE （Dominant river tracing-Routing Integrated with Variable Infiltration Capacity Environment） during 2002-2013 over the Iowa-Cedar River basin， a mid-size basin in Iowa. Results indicate that： ① NLDAS2 and Stage IV perform better than the other three datasets against CPC-U data at daily scale， and IMERG correlates slightly better than TMPARP； these four precipitation products can well capture the precipitation events with the CC （correlation coefficient） values 0.63~0.8， the BIAS （relative bias） values are -0.21%~15.37%， and the POD （probability of detection） values 0.61~0.74， while TMPART performs the worst. ② Using the same calibrated parameter sets from the Iowa-Cedar River basin， NLDAS2， CPC-U， Stage IV， and TMPARP-driven distributed hydrologic model performs good with the NSCE （Nash-Sutcliffe coefficient of efficiency） values of 0.82， 0.80， 0.79， and 0.63， respectively； while IMERG and 3B42RT performs poorly overall with the NSCE value of 0.46 and -1.09. Comparison shows that NLDAS2 and Stage IV precipitation products demonstrate the good hydrologic utility in the Iowa-Cedar River basin， followed by TMPARP and IMERG， indicating that multi-source remote sensing-based precipitation estimates is more suitable for flood modeling.

Rainfall is an important factor affecting hillslope runoff， different rainfall characteristics will change the runoff process. In order to explore the coupling influence of rainfall patterns and peak amount on the hillslope runoff process， two rainfall patterns with unique rain intensity processes that fall in the early and late periods of the entire rainfall process are designed. The simulation experiment of rainfall runoff is carried out. A mathematical model of hillslope runoff based on motion wave equation is constructed， the virtual working condition simulation of the rainfall runoff process is carried out. The variation law of runoff characteristic value with rainfall characteristic is analyzed. It is concluded that with the increase in rainfall peak， the runoff volume and runoff peak value generated by the two rain patterns both increase exponentially， and showing a linear correlation under same rainfall pattern. For different rain types， when the rainfall amount and duration are the same， the posterior peak rainfall produces about 1/4 more runoff than the anterior peak rainfall.

In recent years， with the aggravation of human influence and the rapid development of social economy， the surface water coverage has been decreasing gradually. It is difficult to accurately analyze the variation characters of surface water area in Hubei province due to the influence of wide research scope and statistical difficulty. In this paper， the Google Earth Engine platform （GEE） is used to calculate and analyze the variation characters of surface water cover in Hubei province and its administrative regions. The results showed that between 2000 and 2018， the decreasing of surface water coverage in Hubei province was significant， with a area decreasing of 1 886.5 km². Among the administrative regions， the surface water area of Jingzhou， Wuhan and Xiangyang decreased significantly， the area is 450， 364.8 and 210.8 km²， respectively. The seasonal characteristics of surface water cover in Hubei province are obvious and the surface area covered for 12 months is 46.4% of the total surface area. The analysis of the variation characters of land cover types in Hubei province shows that the permanent surface area of Hubei province is relatively small， and the seasonal surface area is larger. It is significantly affected by flood and waterlogging disasters， which involves a large range.

In order to study the hydrological response of land use change in the typical Karst Basin， Daotian River SWAT model is constructed. The model is used to discuss the law of flow variation， the change rule of subbasin runoff yield and regulation and storage capacity under different land use situations. Results show that the SWAT model after parameter calibration performs well in simulating the monthly flow of the Daotian River. The conversion of grass land to cultivated land will lead to the increase in flow and the decrease in regulation and storage capacity； the conversion of forest land and grass land to cultivated land can easily change the water yield of subbasin， while the conversion of grass land to forest land cannot easily change the water yield of subbasin； the conversion of forest land to cultivated land， grassland to cultivated land and grassland to forest land will weaken the regulation and storage capacity of subbasin， especially the conversion of grassland to cultivated land. The research results are of great significance for reducing drought and flood disasters and preventing rock desertification in Karst areas.

The spatial distribution characteristics of temperature and precipitation combination have significant influences on agricultural production and resource utilization. Based on the set pair analysis method and daily precipitation and temperature data during 1960-2019 at 40 meteorological stations in North China， the spatial distribution of temperature and precipitation combination is studied. The SPI （Standardize Precipitation Index） is used to determine the different precipitation year types （flood， normal， and dry）. The accumulated temperature anomaly is used to determine different temperature year types （high， normal， and low）. The results show that the stations with more high temperature years also have more low temperature years， in general， the frequency of high temperature years is lower in the central of the study region. The frequency of flood years is higher in the north and east of the study area， the spatial distribution has no obvious rule. The connection degree between temperature and precipitation is higher in southwest regions and gradually decreases to the north and east costal region， this can be caused by the frequency of temperature and precipitation opposite year types （high temperature and drought year， low temperature and flood year） is higher in the north and east costal region. The results are of some significance for understanding the distribution of water and heat in the research area， guiding disaster prevention and agricultural planning layout.

As the successor of TMPA， the Global Precipitation Mission （GPM） IMERG （Integrated Multi-satellitE Retrievals for GPM） products provide a source of precipitation data with greater coverage， longer time series， and higher accuracy and spatial and temporal resolution. Based on gauging daily rainfall from 824 meteorological stations， this paper adopts five extreme precipitation indices （R95p， R95pTOT， Rx1day， Rx5day and R25mm） and three statistical metrics （CC， Bias and rRMSE） to evaluate the extreme precipitation detection capability of IMERG Final product （hereafter IMERG） and compare with the results of TMPA 3B42V7 product （hereafter TMPA 3B42）. The result shows that： ① The accuracy of IMERG generally outperforms TMPA 3B42 product with higher CC values and lower rRMSE values of each extreme rainfall indices. ② Except for the underestimation of IMERG in R95p， both IMERG and TMPA 3B42 estimates show overestimation on all five extreme rainfall indices. ③ In extreme rainfall detection， the accuracy of the two satellite products has decreased under complex topography conditions， but IMERG exhibits smaller range and lower level of bias than that of TMPA 3B42. ④ For analyzing the frequency distribution of precipitation with different intensities， the distribution of IMERG is more consistent with gauge precipitation compared to that of TMPA 3B42， but there is underestimation at higher rainfall intensity（>150 mm/d） in both IMERG and TMPA 3B42.

An analysis of multi-time-scale periodic changes of water and sediment in small watersheds can provide a scientific basis for medium and long-term water and sediment prediction， and provide an engineering reference for the multi-scale characteristics of water and sediment sequence in the middle reaches of the Xijiang River. The annual runoff and annual sediment transport data of the Qianjiang Station of the Hongshui River from 1957 to 2016 is selected， and the Mann-Kendall trend test， wavelet analysis and coupling relationship model are used to analyze the main water-sediment series cycle and its multiple time at Qianjiang Station Scale change characteristics. The results show that the annual water-sediment sequence at Qianjiang Station has a downward trend， the multi-scale characteristics are basically the same and the main cycle exists. Human activities， climate fluctuations and water and soil conservation measures have a direct impact on the changes of the water-sediment sequence. The first major cycle of the runoff sequence is 25~30 a， the second major cycle is 13~24 a， and the third major cycle is 3~12 a； the first major cycle of the sediment transport sequence is 20~30 a， and the second major cycle is 9~20 a， The third main period is 3~9 a. On the time scales of 5， 16 and 28 a. Due to the influence of the ecological environment and human activities， the water-sediment sequence has become more turbulent， and the final sediment transport sequence tends to be stable compared to the runoff sequence， and its change is also lagging behind the runoff sequence. In the coupling relationship， R ² of the runoff sequence is 0.43， and the sediment transport sequence is 0.71. The significance of the sediment transport sequence is higher than that of the runoff sequence； and human activities use science and technology to alleviate the contradictions of the original water and sediment system by affecting the sediment transport system. Even the solution makes the water-sediment system in the middle reaches of the Xijiang River enter the regeneration period.

Urban waterfront is an important medium connecting cities and nature. It has irreplaceable functions in other fields such as flood storage and drought relief， soil and water conservation， improvement of urban ecological environment and improvement of regional biodiversity. In recent years， there are many cases of planning and utilizing urban waterfront， but most of the forms are still relatively single. The research on supporting urban theme function， fully embodying ecology， landscape， humanism and culture needs to be studied further. Combined with the current situation at home and abroad， this paper systematically summarizes the current situation of urban waterfront construction and renovation technologies and ideas， and discusses in depth the entirety， ecology， humanistic and diversified urban waterfront areas， combined with the regional culture and other aspects of the study， and puts forward the future direction of waterfront construction and rebuilding.

Jinan spring catchment has four spring groups： Baotu Springs， Heihu Springs， Zhenzhu Springs and Wulongtan Springs. Since the 1950s， the average flow of springs has been decreasing from 36.6×10⁴ m³/d to the current level of 8×10⁴ to 12×10⁴ m³/d. Springs base flow is a component of the relatively stable runoff， which is an important source of downstream river， and it is an important guarantee to maintain the ecosystem health of river course and lake. The paper calculates the base flow of Baotu Springs group in different frequency years， determines the stable base flow index， and calculates the spring base flow in different level years， and the ecological water demand of downstream river is calculated by using the inflection point method. This paper analyzes the relationship between the ecological base flow of spring water and river water demand， reveals the influence of karst springs on the health of rivers， lakes and other ecosystems in Jinan， and provides an important technical support for the protection of springs water in Jinan.

In order to analyze the value of roughness and influencing factors of grass riverbed， this paper adopts the experimental method of combining the indoor physical model with the natural plant， and selects the common vegetation festuca arundinacea in the river beach of Northeast China， and carries out an experimental analysis of three different flows， two different plant densities， three different plant heights， a total of 21 working conditions. Tests show that the approximate range of natural grass river bed roughness values is 0.033 2 ~ 0.064 1， grass riverbed roughness reduces with the increase in duration of erosion and flows， and increases with the increase in plant density and plant height. Under the same conditions， the resistance of plant height to water flow is more than that of plant density.

As economic and technological levels are underdeveloped in rural areas， finding a viable wastewater treatment solution is a challenge. The construction， operation and maintenance costs of urban aerobic wastewater treatment plants are high and not suitable for rural areas. Anaerobic baffled reactor （ABR） has the characteristic of simple structural design， strong solid biological retention capacity， low sludge yield， high impact resistance， good organic matter treatment and low operating costs，very suitable for rural decentralized wastewater treatment. The development and improvement of ABR reactor in recent years are introduced. The influence of temperature， pH， influent organic matter concentration， hydraulic retention time， reactor start-up， shock load and other factors on the ABR treatment efficiency are summarized. The suitable sewage concentration range of the ABR reactor and the feasibility of its application in rural areas are analyzed， and the combined process based on ABR is introduced.

The effluent from the anaerobic/aerobic stage of the sewage treatment system has a low ammonia nitrogen concentration and it is difficult to biodegrade. Aming at this characteristic pollutant from this type of wastewater the study is obtained through single factor experiment and response surface design experiment. The optimal reaction parameters of the three-dimensional electrode are as follows： the waste water initial pH is 5.0； the voltage is 9.86 V； the plate spacing is 6.47 cm； the electrolyte concentration is 0.20 mol/L.The final effluent ammonia nitrogen concentration from 140.0 to 34.5 mg/L when it comes to treat the raw water by these parameters. It is lower than the Wastewater Discharge Standard of Poyang Lake Eco-economic District in Jiangxi Province （DB36/852-2015）. X-ray diffraction method is used to detect that when the sintering temperature is 200 ℃， the surface of activated carbon is loaded with α-MnO₂， β-MnO₂ and_{Mn3O4. Then the reaction mechanism of three-dimensional electrode system is analyzed by electrochemical reaction characteristics.}

The diatom index is used to evaluate the water ecological health of Longjiang and Diaojiang， and 16 sample points are set up in the Longjiang and Diaojiang basins for investigation and research. The identification results show that 30 genera and 141 species of diatoms are identified in the Longjiang River Basin， and 138 species of diatoms in 24 genera are identified in the Diaojiang River Basin. The common dominant genera are Navicula， Trichothecium， Nitzschia， Hydracea Heterophaga. For the currently more mature sixteen diatom indexes， the applicability of the diatom indexes is studied through factor analysis， cluster analysis， box diagram analysis and other statistical analysis methods. The results show that most of the diatom indexes are significantly correlated with TP， TN， Cond， pH and extremely significantly correlated with V， and only the WAT index is significantly correlated with NO₃-N. Three indexes of IPS， IDG and CEE are selected as reference biological indicators for water quality evaluation in this area. Through a comprehensive judgment of the diatom index， it is concluded that the Longjiang samples are of good and medium water quality， a small part of the Diaojiang samples are of good grade water quality， 73% of the samples are of medium or poor water quality， and the overall water quality of Longjiang is better than that of Diaojiang.

Aiming at the problem that the river slope in saline soil area is not conducive to plant growth and is easy to corrode concrete and reinforcement， this paper proposes a protection technique using basalt fiber gabion. Taking a soda saline-alkali soil river slope protection as an example， the particle discrete element program PFC2D is used to calibrate the micro-parameters of the soil and basalt fiber gabion， and a discrete element calculation model is established. Considering the effect of seepage， the particle flow strength reduction method is used to analyze the stability of the gabion slope protection under the design flood level and compare it with the original slope. In addition， the stress characteristics of the composite reinforcement that are concerned in the project are studied. The results show that the stability factor of the slope has increased by 40.9% after gabion support is used， the soil in the slope is not damaged and slipped， and the failure mode of the gabion slope protection is shown by the separation between the gabions and the bending and separation of the composite bars. The tensile and compressive stress of the composite bar is much lower than its allowable strength value.

In order to investigate the applicability of the R-2 cross method in ecological flow calculations of small rivers in southern China， the following optimization suggestions are made after analyzing the applicable conditions of the wetted perimeter ratio， the selection of typical cross section， the reasonable selection of other hydraulic parameters and the composite constraints. Considering the assumptions made in the R-2 cross method for the critical condition of wide and shallowrivers， it is appropriate for the wetted perimeter ratio parameter to be considered as an evaluation factor rather than a computational factor in stream cross-sections that have been heavily influenced by human modification. Based on an analysis of the relationship between different hydraulic parameters， an optimization method of hydraulic parameters in R-2 cross method is proposed. By calculating the quantitative relationship between water depth and velocity， the h^{\frac{\mathrm{2}}{\mathrm{3}}} ~ u curve is established. With the composite constrained interval， it is convenient and fast to determine the appropriate values of average-depth and average-velocity parameters in R-2 cross method. The typical section suitable for the hydraulic conditions of ecosystem can also be selected from multiple sections by using the optimization method.