Search

Different conclusions have been drawn in literature as using reanalysis rainfall products to drive hydrological models for the simulation of streamflow, which warrant a need of further investigation before a generalised conclusion can be drawn. This paper assesses the utility of two widely used reanalysis rainfall datasets (WFD (developed by the WATCH project) and APHRODITE (Asian Precipitation–Highly-Resolved Observational Data Integration Towards Evaluation of the water resources)) against the gauged rainfall in terms of flood simulation in lumped (Xinanjiang Model) and distributed (SWAT Model) hydrological models in a tributary basin of Yangtze River with 94,660 km 2 drainage area in humid region of southern China. Differences in terms of rainfall accumulation, number of rainy days, spatial patterns of the rainfall amount and frequency distribution of the rain rates are evaluated. The APHRODITE product shows high consistency with the gauged rainfall while WFD data gives large errors in various statistical indices in the study region. Simulated discharges from the gauged and reanalysis rainfall data, respectively, are analysed and compared with the observed discharge in basin outlet over the period 1991–2005. The APHRODITE data show relatively high ability in modelling hydrological responses while the WFD data based models give large error in simulating the discharge. For the assessment of the high flows, both datasets exhibit some skills in flood prediction, however, APHRODITE data perform better than the WFD data when forcing into the lumped Xinanjiang Model than into the distributed SWAT Model in terms of flood duration, Probability of Detection, False Alarm Rate, regressions of annual peaks and partial duration series. [ABSTRACT FROM AUTHOR]

Abstract: Long monthly streamflow series of three control hydrological stations of the Pearl River basin were analyzed by using the scanning t-test and the scanning F-test. Possible implications of the changing properties of streamflow variations for the water resource management of the Pearl River Delta are also discussed. The results indicated that: 1) more complicated changes were observed in terms of the second center moment when compared to the first original moment. More significant abrupt changes of the second center moment imply more sensitive response of streamflow stability to climate changes and human activities; 2) abrupt behaviors of the first (second) center moment of the streamflow variations tend to be more sensitive to climate changes and/or human activities in the larger river basin when compared to those in the smaller river basin. These phenomena are attributed to buffering functions of more storage space of longer river channel, and more complicated and longer runoff yield and concentration processes in the river basin of larger drainage area; 3) annual minimum streamflow of the Pearl River basin tends to be increasing. This will be helpful for better human mitigation of the salinity intrusion in dry seasons across the Pearl River Delta. Annual maximum streamflow, when compared to annual minimum streamflow, shows larger-magnitude variability reflected by larger standard deviation, implying unfavorable conditions for flood mitigation in the Pearl River Delta. The results of this paper are of scientific and practical merits for water resource management and sound human mitigation to water hazards across the Pearl River Delta, and also are a good case study for similar researches in other river deltas in the world under the changing environment. [ABSTRACT FROM AUTHOR]

The hydrologic design of nonstationary flood extremes is an emerging field that is essential for water resources management and hydrologic engineering design to cope with changing environment. This paper aims to investigate and compare the capability of four nonstationary hydrologic design strategies, including the expected number of exceedances (ENE), design life level (DLL), equivalent reliability (ER), and average design life level (ADLL), with the last three methods taking into consideration the design life of the project. The confidence intervals of the calculated design floods were also estimated using the nonstationary bootstrap approach. A comparison of these four methods was performed using the annual maximum flood series (AMFS) of the Weihe River basin, Jinghe River basin, and Assunpink Creek basin. The results indicated that ENE, ER and ADLL yielded the same or very similar design values and confidence intervals for both increasing and decreasing trends of AMFS considered. DLL also yields similar design values if the relationship between DLL and ER/ADLL return periods is considered. Both ER and ADLL are recommended for practical use as they have associated design floods with the design life period of projects and yield reasonable design quantiles and confidence intervals. Furthermore, by assuming that the design results using either a stationary or nonstationary hydrologic design strategy should have the same reliability, the ER method enables us to solve the nonstationary hydrologic design problems by adopting the stationary design reliability, thus bridging the gap between stationary and nonstationary design criteria. [ABSTRACT FROM AUTHOR]

• A method of regional non-stationary flood frequency analysis is proposed. • The index incorporates climatic and reservoir factors as covariates. • MRI is more suitable for non-stationary flood frequency analysis. • The prediction period of flood frequency analysis is extended. Global climate change and reservoir regulations can alter the natural flow of rivers. Influenced by these two drivers, flood sequences may no longer satisfy the assumption of stationary, thereby making it difficult to accurately analysis flood frequency and to design water conservancy projects. Therefore, it is of great significance to analyse the non-stationary frequency of flood sequences in a changing environment. In this study, we proposed a method for conducting nonstationary flood frequency analysis caused by cascade reservoirs as well as the low-frequency climate indices. The proposed non-stationary model 2, with the explanatory variables of climate indices and modified reservoir index (MRI), was compared with the traditional stationary model and the widely used non-stationary model 1 with time as the explanatory variable. The study was conducted at six hydrological stations in the main stream and tributaries of the upper reaches of the Yangtze River in China (considered as the Three Gorges Reservoir Area). The results of the generalized additive model for location, scale and shape (GAMLSS) showed that the Akaike information criterion and Bayesian information criterion values of the proposed non-stationary model method 2 are smaller than those of the two comparison models. When the low-frequency South Oscillation Index is high or the Arctic Oscillation and North Pacific Oscillation are low, the stationary model underestimates the design value of flood quantiles compared with the non-stationary model 2. Compared with the non-stationary model 1, the MRI and low-frequency climate indices as the explanatory variables in model 2 can better describe the non-stationary characteristics of flood frequency and amplitude. In addition, the non-stationary model considering external physical factors can provide better prediction of future design flood compared with two traditional models. [ABSTRACT FROM AUTHOR]

• Nonstationary models that incorporate climatic variables and reservoir index are constructed. • The nonstationary models provide a better fitting of nonstationary flood and low flow series. • Nonstationary frequency analysis of flood and low flow in upper of Huaihe River Basin are conducted. • Climatic and reservoir impacts are related to nonstationarity of flood and low flow respectively. Conventional water infrastructure designs for flood and low flows are usually based on the assumption of stationarity of extreme events. However, recent evidence suggests that the influences of climate variability and human activities have made the hypothesis of stationarity questionable. In this study, we used the generalized additive models for location, scale, and shape (GAMLSS) to construct a nonstationary model in which the parameters of the selected distributions were modelled as a function of climatic variables (i.e., climate indices and precipitation) and/or the reservoir index (RI). The nonstationary models were then used to analyse annual flood and low flow frequency at four hydrological stations in the upper reaches of the Huaihe River Basin, including Dapoling (DPL), Changtaiguan (CTG), Zhuganpu (ZGP), and Xixian (XX) stations. Annual floods were represented by the maximum daily streamflow in each year, and low flows were represented by the 95th quantile of the daily streamflow (Q 95) in each year. The change point and trend analysis revealed that the flood series of the ZGP station and the low flow series of the DPL and XX stations exhibited significant downward and upward trends (p < 0.1)), respectively. The low flow series of the ZGP station showed a significant change point in 1980 (p < 0.1). GAMLSS modelling results showed that, in comparison with stationary models, nonstationary models that included precipitation and the Arctic Oscillation climate index as covariates for the gamma distribution location parameter provided a superior description of the flood series at the four stations. Nonstationary models that incorporated precipitation and/or RI as covariates for the Weibull distribution parameters fit the low flow series better than stationary models at all stations. Furthermore, we found that nonstationary models outperformed stationary models in terms of flood frequency analysis, covering all flood observation points and capturing the generally decreasing trend in flood series, as well as a decrease in the scatter of estimated flood value magnitudes. For the low flow frequency analysis, the comparison results showed that the nonstationary and stationary models performed identically for the DPL, CTG, and XX stations, where no significant change point was detected. However, for the ZGP station, where a significant change point was detected, the nonstationary models performed better than the stationary models and could accurately capture the changes in the magnitude of the estimated low flow values before and after the change point. Overall, the proposed nonstationary model can serve as a tool for nonstationary frequency analysis of flood and low flow series under the influence of climate variability and reservoir regulations, thus providing a reference for regional water infrastructure design. [ABSTRACT FROM AUTHOR]

• Nonstationary models that incorporate climatic variables and reservoir index are constructed. • The nonstationary models provide a better fitting of nonstationary flood and low flow series. • Nonstationary frequency analysis of flood and low flow in upper of Huaihe River Basin are conducted. • Climatic and reservoir impacts are related to nonstationarity of flood and low flow respectively. Conventional water infrastructure designs for flood and low flows are usually based on the assumption of stationarity of extreme events. However, recent evidence suggests that the influences of climate variability and human activities have made the hypothesis of stationarity questionable. In this study, we used the generalized additive models for location, scale, and shape (GAMLSS) to construct a nonstationary model in which the parameters of the selected distributions were modelled as a function of climatic variables (i.e., climate indices and precipitation) and/or the reservoir index (RI). The nonstationary models were then used to analyse annual flood and low flow frequency at four hydrological stations in the upper reaches of the Huaihe River Basin, including Dapoling (DPL), Changtaiguan (CTG), Zhuganpu (ZGP), and Xixian (XX) stations. Annual floods were represented by the maximum daily streamflow in each year, and low flows were represented by the 95th quantile of the daily streamflow (Q 95) in each year. The change point and trend analysis revealed that the flood series of the ZGP station and the low flow series of the DPL and XX stations exhibited significant downward and upward trends (p < 0.1)), respectively. The low flow series of the ZGP station showed a significant change point in 1980 (p < 0.1). GAMLSS modelling results showed that, in comparison with stationary models, nonstationary models that included precipitation and the Arctic Oscillation climate index as covariates for the gamma distribution location parameter provided a superior description of the flood series at the four stations. Nonstationary models that incorporated precipitation and/or RI as covariates for the Weibull distribution parameters fit the low flow series better than stationary models at all stations. Furthermore, we found that nonstationary models outperformed stationary models in terms of flood frequency analysis, covering all flood observation points and capturing the generally decreasing trend in flood series, as well as a decrease in the scatter of estimated flood value magnitudes. For the low flow frequency analysis, the comparison results showed that the nonstationary and stationary models performed identically for the DPL, CTG, and XX stations, where no significant change point was detected. However, for the ZGP station, where a significant change point was detected, the nonstationary models performed better than the stationary models and could accurately capture the changes in the magnitude of the estimated low flow values before and after the change point. Overall, the proposed nonstationary model can serve as a tool for nonstationary frequency analysis of flood and low flow series under the influence of climate variability and reservoir regulations, thus providing a reference for regional water infrastructure design. [ABSTRACT FROM AUTHOR]

Summary The assumption of stationarity of annual peak flood (APF) records at 28 hydrological stations across the Pearl River basin, China, is tested. Abrupt changes in mean and variance are tested using the Pettitt technique and the Loess method. Trends of APFs are analyzed using the Mann–Kendall method and the Spearman technique. And then the stationarity of the APF series is further investigated by GAMLSS models and long-term persistence. Results indicate that: (1) abrupt changes in mean and variance have similar influences on the changing properties of APFs, such as stationarity. Abrupt changes in mean and variance are only field significant in the East River basin; (2) the change points have a considerable impact on the detection of trends, and these may be attributed to the fact that a abrupt increase or decrease in mean values will affect the trend variations. Besides, for the APF series being free of change points and trend, the GAMLSS models also corroborate stationarity of the APF series; (3) the nonstationarity in the Pearl River basin is mainly due to the existence of the change point. However, the APF series with change points in mean and/or variance are also characterized by long-term persistence, and thus it is infeasible to assert that the abrupt behaviors and/or trends of the APF series are the result of human activities or long-term persistence, especially in the East River basin. Results of this study will provide information for management of water resources and design of hydraulic facilities in the Pearl River basin in a changing environment. [ABSTRACT FROM AUTHOR]

• Machine learning assists hybrid model to promote flood forecasting and early warning. • Hybridizing MCQRNN with XAJ model for flood probability density forecasting. • XAJ-MCQRNN conquers overfitting and biased-prediction bottlenecks. • XAJ-MCQRNN improves accuracy and reliability of flood probability density forecasts. Making accurate and reliable probability density forecasts of flood processes is fundamentally challenging for machine learning techniques, especially when prediction targets are outside the range of training data. Conceptual hydrological models can reduce rainfall-runoff modelling errors with efficient quasi-physical mechanisms. The Monotone Composite Quantile Regression Neural Network (MCQRNN) is used for the first time to make probability density forecasts of flood processes and serves as a benchmark model, whereas it confronts the drawbacks of overfitting and biased-prediction. Here we propose an integrated model (i.e. XAJ-MCQRNN) that incorporates Xinanjiang conceptual model (XAJ) and MCQRNN to overcome the phenomena of error propagation and accumulation encountered in multi-step-ahead flood probability density forecasts. We consider flood forecasts as a function of rainfall factors and runoff data. The models are evaluated by long-term (2009–2015) 3-hour streamflow series of the Jianxi River catchment in China and rainfall products of the European Centre for Medium-Range Weather Forecasts. Results demonstrated that the proposed XAJ-MCQRNN model can not only outperform the MCQRNN model but also prominently enhance the accuracy and reliability of multi-step-ahead probability density forecasts of flood process. Regarding short-term forecasts in testing stages at four horizons, the XAJ-MCQRNN model achieved higher Nash-Sutcliffe Efficiency but lower Root Mean Square Error values, while improving Coverage Ratio and Relative Bandwidth values in comparison to the MCQRNN model. Consequently, the improvement can benefit the mitigation of the impacts associated with uncertainties of extreme flood and rainfall events as well as promote the accuracy and reliability of flood forecasting and early warning. [ABSTRACT FROM AUTHOR]