Inter-catchment comparison of flow regime between the Hailiutu and Huangfuchuan rivers in the semi-arid Erdos Plateau,Northwest China

Abstract

Two river catchments, the Huangfuchuan and the Hailiutu, located in the same climate zone in the Erdos Plateau, China, have distinctly different flow regimes. This study systematically compared differences between the flow regimes of these two catchments using several statistical methods, and analysed the possible causes. The variations in yearly, monthly and daily mean discharges were found to be much greater in the Huangfuchuan catchment than in the Hailiutu catchment. Preliminary analysis indicated that these differences are not caused by changes in climate, but are instead attributable to differences in geology, geomorphology, hydrological processes and human interventions. In the Hailiutu catchment, the dominant groundwater contribution maintains stationary daily and monthly river discharges, while shifts in yearly mean discharges were closely associated with the expansion or reduction of crop area. In the Huangfuchuan catchment, the dominant direct rainfall–runoff process generates flashier daily and monthly river discharges, while the decrease of yearly mean discharges is caused mainly by the construction of check dams. These findings have significant implications for water resource management and the implementation of proper soil and water conservation measures in the middle reach of the Yellow River Basin of China.

Editor Z.W. Kundzewicz; Associate editor Y. Gyasi-Agyei     

Enhancing the Forecasting of Monthly Streamflow in the Main Key Stations of the River Nile Basin

Predicting the streamflow of rivers can have a significant economic impact, as this can help in agricultural water management and in providing protection from water shortages and possible flood damage. In this study, two statistical models have been used; Deseasonalized Autoregressive moving average model (DARMA) and Artificial Neural Network (ANN) to predict monthly streamflow which important for reservoir operation policy using different time scale, monthly and 1/3 monthly (ten-days) flow data for River Nile basin at five key stations. The streamflow series is deseasonalized at different time scale and then an appropriate nonseasonal stochastic DARMA (p, q) models are built by using the plots of Partial Auto Correlation Function (PACF) to determine the order (p) of DARMA model. Then the deseasonalized data for key stations are used as input to ANN models with lags equals to the order (p) of DARMA model. The performance of ANN and DARMA models are compared using statistical methods. The results show that the developed model (using 1/3 monthly (ten-days) and ANN) has the best performance to predict monthly streamflow at all key stations. The results also show that the relative error in the developed model result did not exceed 9% while in the traditional models reach to 68% in the flood months in the testing period. The result also indicates that ANN has considerable potential for river flow forecasting.     

Changes to flow regime on the Niger River at Koulikoro under a changing climate

Abstract

A significant decrease in mean river flow as well as shifts in flood regimes have been reported at several locations along the River Niger. These changes are the combined effect of persistent droughts, damming and increased consumption of water. Moreover, it is believed that climate change will impact on the hydrological regime of the river in the next decades and exacerbate existing problems. While decision makers and stakeholders are aware of these issues, it is hard for them to figure out what actions should be taken without a quantitative estimate of future changes. In this paper, a Soil and Water Assessment Tool (SWAT) model of the Niger River watershed at Koulikoro was successfully calibrated, then forced with the climate time series of variable length generated by nine regional climate models (RCMs) from the AMMA-ENSEMBLES experiment. The RCMs were run under the SRES A1B emissions scenario. A combination of quantile-quantile transformation and nearest-neighbour search was used to correct biases in the distributions of RCM outputs. Streamflow time series were generated for the 2026–2050 period (all nine RCMs), and for the 2051–2075 and 2076–2100 periods (three out of nine RCMs) based on the availability of RCM simulations. It was found that the quantile-quantile transformation improved the simulation of both precipitation extremes and ratio of monthly dry days/wet days. All RCMs predicted an increase in temperature and solar radiation, and a decrease in average annual relative humidity in all three future periods relative to the 1981–1989 period, but there was no consensus among them about the direction of change of annual average wind speed, precipitation and streamflow. When all model projections were averaged, mean annual precipitation was projected to decrease, while the total precipitation in the flood season (August, September, October) increased, driving the mean annual flow up by 6.9% (2026–2050), 0.9% (2051–2075) and 5.6% (2076–2100). A t-test showed that changes in multi-model annual mean flow and annual maximum monthly flow between all four periods were not statistically significant at the 95% confidence level.     

Simulation von Variablen der Fließgewässerqualität mittels Mehrfachregressionen

There are investigated the variables O₂, BOD₅, seston, NO₃-N, NH₄-N and o-PO₄ from at least five-year series of five stations along a river section of 50 km. After exclusion of a linear trend and substraction of the individual monthly mean values from the monthly mean of many years in order to eliminate the effect of the annual variation as well as testing for normal distribution, first the correlation coefficients of the variables to the flow rate Q and the temperatures of air and water are determined, which show directional changes just in the longitudinal profile of the river. The same holds for the correlation of the variables between the measuring points. From this the model structure is derived, according to which the concentration at one measuring station can be simulated by multiple regression to Q and T at the same level as well as the concentration at the upper level. The results are discussed in detail and evaluated with respect to their inclusion in longterm management models of water quantity management.     

Trend detection in river flow series: 1. Annual maximum flow / Détection de tendance dans des séries de débit fluvial: 1. Débit maximum annuel

Abstract

Results of a study on change detection in hydrological time series of annual maximum river flow are presented. Out of more than a thousand long time series made available by the Global Runoff Data Centre (GRDC) in Koblenz, Germany, a worldwide data set consisting of 195 long series of daily mean flow records was selected, based on such criteria as length of series, currency, lack of gaps and missing values, adequate geographical distribution, and priority to smaller catchments. The analysis of annual maximum flows does not support the hypothesis of ubiquitous growth of high flows. Although 27 cases of strong, statistically significant increase were identified by the Mann-Kendall test, there are 31 decreases as well, and most (137) time series do not show any significant changes (at the 10% level). Caution is advised in interpreting these results as flooding is a complex phenomenon, caused by a number of factors that can be associated with local, regional, and hemispheric climatic processes. Moreover, river flow has strong natural variability and exhibits long-term persistence which can confound the results of trend and significance tests.     

An improved time series model for monthly stream flows

This study aims to develop an improved time series model to overcome difficulties in modeling monthly short term stream flows. The periodic, serial dependent and independent components of the classical time series models are improved separately by information transfer from a surrounding long term gauging station to the considered flow section having short term records. Eventually, an improved model preserving the mathematical model structure of the classical time series model, while improving general and monthly statistics of the monthly stream flows, is derived by using the improved components instead of the short term model components in the time series modeling. The correlative relationships between the current short term and surrounding long term stations are used to improve periodic and serial dependent behaviors of monthly flows. Independent components (residuals) are improved via the parameters defining their theoretical probability distribution. The improved model approach is tested by using 50 year records of Göksu-Himmetli (1801) and Göksu-Gökdere (1805) flow monitoring stations located on the Ceyhan river basin, in south of Turkey. After 50 year records of the station 1801 are separated into five 10 year sub series, their improved and classical time series models are computed and compared with the real long-term (50 year) time series model of this station to reveal efficiencies of the improved models for each subseries (sub terms with 10 year observation). The comparisons are realized based on the model components, model estimates and general/monthly statistics of model estimates. Finally, some evaluations are made on the results compared to the regression method classically applied in the literature.     

Variability in the discharge of South American rivers and in climate / Variabilité des débits de rivières d'Amérique du Sud et du climat

Abstract

Changes in trend and quasi-periodicities are sought in the time series of river discharges in all major South American basins. The relationship between trends and quasi-periodicities found and climate variations on interannual and longer time scales are discussed. Consideration of multiple rivers gives insight into the geographical extent of hydrological signals and climate impacts. It is found that the streamflow of all major rivers of South America has experienced an increased trend since the early 1970s. It is suggested that this simultaneity may reflect the impact of a large-scale climate change. All the time series of river streamflows that were analysed show El Niño-like periodicities. Only for La Plata Basin do these explain a larger part of the total variance than the other quasi-periodicities. There are two other quasi-oscillations in the time series analysed: one of them with a longer period—around 17 years—and the other of about 9 years. Previous work has related these oscillations to sea-surface temperature anomalies in the Atlantic Ocean.     

Trend detection in river flow series: 2. Flood and low-flow index series / Détection de tendance dans des séries de débit fluvial: 2. Séries d'indices de crue et d'étiage

Abstract

Major floods in Europe and North America during the past decade have provoked the question of whether or not they are an effect of a changing climate. This study investigates changes in observational data, using up to 100-year-long daily mean river flow records at 21 stations worldwide. Trends in seven flood and low-flow index series are assessed using Mann-Kendall and linear regression methods. Emphasis was on the comparison of trends in these flow index series, particularly in peak-over-threshold (POT) series as opposed to annual maximum (AM) river flow series. There is a larger number of significant trends in the AM than in the POT flood magnitude series, probably relating to the way the series are constructed. Low flood peaks occurring at the beginning or end of a time series with trend may be too low to be selected for the POT analysis. However, one peak per year will always be selected for the AM series, making the slope steeper and/or the series longer, resulting in a more significant trend. There is no general pattern of increasing or decreasing numbers or magnitudes of floods, but there are significant increases in half of the low-flow series.     

A comparison of performance of several artificial intelligence methods for forecasting monthly discharge time series

Developing a hydrological forecasting model based on past records is crucial to effective hydropower reservoir management and scheduling. Traditionally, time series analysis and modeling is used for building mathematical models to generate hydrologic records in hydrology and water resources. Artificial intelligence (AI), as a branch of computer science, is capable of analyzing long-series and large-scale hydrological data. In recent years, it is one of front issues to apply AI technology to the hydrological forecasting modeling. In this paper, autoregressive moving-average (ARMA) models, artificial neural networks (ANNs) approaches, adaptive neural-based fuzzy inference system (ANFIS) techniques, genetic programming (GP) models and support vector machine (SVM) method are examined using the long-term observations of monthly river flow discharges. The four quantitative standard statistical performance evaluation measures, the coefficient of correlation (R), Nash–Sutcliffe efficiency coefficient (E), root mean squared error (RMSE), mean absolute percentage error (MAPE), are employed to evaluate the performances of various models developed. Two case study river sites are also provided to illustrate their respective performances. The results indicate that the best performance can be obtained by ANFIS, GP and SVM, in terms of different evaluation criteria during the training and validation phases.     

A hydrologic contribution to risk assessment for the Caspian Sea

The Caspian Sea (CS), the world's largest inland sea, may also be considered as large-scale limnic system. Due to strong fluctuations of its water level during the 20th century and the flooding of vast areas in a highly vulnerable coastal zone, economic and environmental risk potentials have to be considered. Since the major water input into the CS is attributed to the Volga river, the understanding of its long-term flow process is necessary for an appropriate risk assessment for the CS and its coastal area. Therefore, a top-down approach based on statistical analyses of long-term Volga flow series is pursued. For the series of annual mean flow (MQ) of the Volga river basin during the 20th century, a complex oscillation pattern was identified. Analyses for multiple gauges in the Volga river basin and Eurasian reference basins revealed that this oscillation pattern resulted from the superposition of oscillations with periods of ∼30 years (MQ) in the western part of the Volga river basin, and ∼14 years (flow volume of snowmelt events) and ∼20 years (flow volume of summer and autumn) in the eastern part of the Volga river basin (Kama river basin). Almost synchronous minima or maxima of these oscillations occurred just in the periods of substantial changes of the Caspian Sea level (CSL). It can thus be assumed that the described mechanism is fundamental for an understanding of the CSL development during the 20th century. Regarding the global climate change, it is still difficult to predict reliably the development of the CSL for the 21st century. Consequently, we suggest an ongoing, interdisciplinary research co-operation among climatology, hydrology, hydraulics, ecology and spatial data management.     

A combined rotated general regression neural network method for river flow forecasting

ABSTRACT

This study focused on the performance of the rotated general regression neural network (RGRNN), as an enhancement of the general regression neural network (GRNN), in monthly-mean river flow forecasting. The study of forecasting of monthly mean river flows in Heihe River, China, was divided into two steps: first, the performance of the RGRNN model was compared with the GRNN model, the feed-forward error back-propagation (FFBP) model and the soil moisture accounting and routing (SMAR) model in their initial model forms; then, by incorporating the corresponding outputs of the SMAR model as an extra input, the combined RGRNN model was compared with the combined FFBP and combined GRNN models. In terms of model efficiency index, R², and normalized root mean squared error, NRMSE, the performances of all three combined models were generally better than those of the four initial models, and the RGRNN model performed better than the GRNN model in both steps, while the FFBP and the SMAR were consistently the worst two models. The results indicate that the combined RGRNN model could be a useful river flow forecasting tool for the chosen arid and semi-arid region in China.

Editor D. Koutsoyiannis; Associate editor not assigned     

Einfluss korrelativer Abhängigkeiten auf simulierte Zeitreihen: Empfindlichkeit des Fiering-Modells gegenüber Veränderungen der Korrelationsstruktur

The simulation of time series is based on estimated statistical parameters of the empirical time series. The Fiering Model generating monthly sums of streamflows is used as an example for the simulation in order to account for the error of the model, theoretically and practically, caused by statistically inprecise parameter estimation. The sensitivity of this model, especially to the correlation coefficients, is analyzed by means of systematic variations of the correlation coefficients, since these are most affected by the error of estimation. No significant dependency could be found comparing the empirical and simulated parameters mean, standard deviation, and skewness. From this follows, that the importance of the correlation coefficients in the Fiering Model is generally overestimated. The results are given for monthly sums of streamflows at four stations with different hydrological characteristics.      

Evaluating piezometric trends using the Mann-Kendall test on the alluvial aquifers of the Elqui River basin,Chile

Abstract

Today, more than ever, there is a need to implement robust statistical methods to ensure the proper evaluation of water resources data to support decision makers in water resources planning and management. Graphing or mapping data for visualization is the easiest way to communicate trends, especially to a non-technical audience. This paper describes the use of an approach that combines the Mann-Kendall test, Sen slope test and principal component analysis to detect and map the monthly trends of piezometric time series and their magnitude in the period 1979–2008. The data were obtained in 23 shallow wells in the alluvial aquifers of the Elqui River basin in central Chile, an area characterized by scarce water resources and intense agricultural and mining activities. The results show significant downward trends at the majority of the wells. Because groundwater in these shallow wells is highly dependent on the water in the river and its tributaries, the reasons for these downward trends are mainly related to a decrease of streamflow observed in the Elqui River. The streamflow is derived from mountain snowmelt rather than from rainfall, which showed no flow trend during the same period.     

REDUCTION OF SERIALLY CORRELATED HYDROLOGIC DATA

Abstract

The need for specifying water availability in terms of its time sequence and distribution, rather than in terms of lumped quantities or flow duration, has given rise to data asquisition programs which use shorter sampling intervals. Shorter time intervals, in turn, has resulted in serially correlated observations. In order to determine the reliability of design variables derived from these observations, it is necessary to reduce these data to an equivalent series of independent observations, or find the effective lenght of the series.

A general parametric formula is developed for the effective number of observations for the second order autoregressive process, which has been found to apply to the series of mean daily river flows. Different values of the parameters were programmed in a digital computer to obtain tables for data reduction.     

El Niño and La Niña influence on mean river flows of southern South America in the 20th century

The influence of the El Niño Southern Oscillation (ENSO) phenomenon on monthly mean river flows of 12 rivers in the extreme south of South America in the 20th century is analysed. The original dataset of each river is divided into two subsets, i.e. warm ENSO events or El Niño, and cold ENSO events or La Niña. The elements of the subsets are composites of 24 consecutive months, from January of the year when the ENSO event begins to December of the following year. The ENSO signal is analysed by comparing the monthly mean value of each subset to the long-term monthly mean. The results reveal that, in general, monthly mean El Niño (La Niña) river flows are predominantly larger (smaller) than the long-term monthly mean in the rivers studied. The anomalies are more evident during the second half of the year in which the event starts and the first months of the following year.     

Numerical modelling of the impacts of water abstraction for hydraulic fracturing on groundwater–surface water interaction: a case study from northwestern Alberta,Canada

ABSTRACT

The temporal dynamics of groundwater–surface water interaction under the impacts of various water abstraction scenarios are presented for hydraulic fracturing in a shale gas and oil play area (23 984.9 km²), Alberta, Canada, using the MIKE-SHE and MIKE-11 models. Water-use data for hydraulic fracturing were obtained for 433 wells drilled in the study area in 2013 and 2014. Modelling results indicate that water abstraction for hydraulic fracturing has very small (<0.35%) negative impacts on mean monthly and annual river and groundwater levels and stream and groundwater flows in the study area, and small (1–4.17%) negative impacts on environmental flows near the water abstraction location during low-flow periods. The impacts on environmental flow depend on the amount of water abstraction and the daily flow over time at a specific river cross-section. The results also indicate a very small (<0.35%) positive impact on mean monthly and annual groundwater contributions to streamflow because of the large study area. The results provide useful information for planning long-term seasonal and annual water abstractions from the river and groundwater for hydraulic fracturing in a large study area.     

Multivariate hydrologic time series analysis

The objective of this study is to establish a multivariate watershed hydrologic system model involving meteorological data as the input and river flow as the output of the system. Monthly hydrological time series of runoff, temperature and precipitation were selected for analysis. A first-order autoregressive-moving average (ARMA) transfer function model was found adequate to describe the multivariate watershed hydrologic system for the monthly runoff and meteorological time series. The results also indicated that the coordinated use of the meteorological and hydrometric data would enhance the accuracy of estimation of runoff characteristics.     

A multi-model ensemble approach for the assessment of climatic and anthropogenic impacts on river flow change

ABSTRACT

Climate variability and human activities are considered to be the most likely reasons for negative trends in river inflow and the water level of some lakes and wetlands in the world. To quantify the uncertain impacts of climate variations and anthropogenic activities on Ajichay River flow in Iran, a multi-model ensemble approach based on the Bayesian model averaging (BMA) method is applied. Several statistical and simulation-based methods are used to distinguish the impacts of climatic and anthropogenic factors on river flow. The results show that almost all the methods identified human activities as the dominant impact on streamflow (about 73–85% of the change). The between-model and within-model uncertainty analyses using BMA showed that the 95% uncertainty intervals of the individual approaches have relatively large deviation ranges. The BMA mean prediction could reduce the range of between-model uncertainties to 14–27% for climate impacts and 74–80% for human impacts. This approach provides a way to better understand the contributions of climatic and anthropogenic impacts on river flow change.