Search

• Out-of-sample prediction of extreme precipitation using nonstationary models. • Significant trends in data of extreme precipitation are insufficient to warrant the selection of a nonstationary model. • Large prediction uncertainty reduces the support for nonstationary models of extreme precipitation. • (Non)stationary model selection should explicitly account for model complexity. Knowledge of the recurrence intervals of precipitation extremes is vital for infrastructure design, risk assessment, and insurance planning. However, trends and shifts in rainfall patterns globally pose challenges to the application of extreme value analysis (EVA) which relies critically on the assumption of stationarity. In this paper, we explore: (1) the suitability of nonstationary (NS) models in the presence of statistically significant trends, and (2) their potential in modeling out-of-sample data to improve frequency analysis of extreme precipitation. We analyze the benefits of using a nonstationary Generalized Extreme Value (GEV) model for annual extreme precipitation records from Southern Brazil. The location of the GEV distribution is allowed to change with time. The unknown GEV model parameters are estimated using Bayesian techniques coupled with Markov chain Monte Carlo simulation. Next, we use GAME sampling to compute the evidence (and their ratios, the so-called Bayes factors) for stationary and nonstationary models of annual maximum precipitation. Our results show that the presence of a statistically significant trend in annual maximum precipitation alone does not justify the use of a NS model. The location parameter of the GEV distribution must also be well defined, otherwise, stationary models of annual maximum precipitation receive more support by the data. These findings reiterate the importance of accounting for GEV model parameters and predictive uncertainty in frequency analysis and hypothesis testing of annual maximum precipitation data records. Furthermore, a meaningful EVA demands detailed knowledge about the origin and persistence of observed changes. [ABSTRACT FROM AUTHOR]

Summary In the present paper multi-temporal Lidar (Light detection and ranging) data and Landsat images are used to assess the spatial variability of snow at the end of the accumulation season (April–May) in a glacierized catchment (167 km 2 ) in Tyrol, Austria. Snow cover characteristics in the Tyrolean Alps have been analysed using regular snow measurements and snow course data. Results are used for the conversion of basin-wide Lidar snow depth into snow water equivalent (SWE). When considering different possible error sources, uncertainties of the mean basin-wide SWE obtained from Lidar are between 12% and 16%. Available distributions of SWE and snow covered area (SCA) in the catchment are used for the calibration and validation of the fully distributed hydrological model SES. The study focuses especially on the simulation of snow accumulation and the corresponding variability of snow. Observed accumulation patterns are related to the topography (elevation, slope and curvature), and according parameter settings of the hydrological model are derived by means of Monte Carlo simulations. The majority of the model runs simulates SCA for various datasets with an accuracy of 85–95%. The paper demonstrates that using SWE data is superior to SCA for constraining model parameter ranges. Results at the watershed scale are in agreement with respect to the total water volume of the snow cover with deviations lower than 5% between SWE from Lidar or from the hydrological model. [ABSTRACT FROM AUTHOR]

• New Bethlahmy variant is proposed to estimate Probable Maximum Precipitation (PMP) • Potential of the proposed variant is demonstrated through Monte Carlo simulations. • A case study is presented on 37,872 stations from a global precipitation database. • Bethlahmy, Hershfield, and probabilistic approach-related variants are compared. The design and risk analysis of large-scale hydraulic structures (e.g., dams) and sensitive installations (e.g., nuclear facilities) downstream of those structures rely on design flood corresponding to probable maximum precipitation (PMP). In areas where there is a sparsity of information on hydrometeorological variables, practitioners use various statistical methods to arrive at a PMP estimate, assuming it to be the possible upper bound for precipitation. However, the assumption is violated in different parts of the world. Hence, there is a need to improve the existing statistical methods and develop their potential alternatives. Against this backdrop, this paper proposes a new variant of a non-parametric method (Bethlahmy) to facilitate the estimation of PMP at locations with sparse records of extreme precipitation. It involves mapping of datapoints in annual maximum series and their ranks to a non-dimensional space (NDS) and using the information on sample size and observed maximum precipitation in the NDS to arrive at a surrogate variable representing PMP, which is eventually mapped back to PMP in the original space. The effectiveness of the proposed Bethlahmy variant over various existing statistical techniques is illustrated through Monte Carlo Simulation experiments and a case study on 37,872 stations from a global precipitation database. The existing techniques include the original Bethlahmy and Hershfield methods, conventional probabilistic approach, and relevant variant(s). Insight is provided into the relative performance of these methods, as there is a dearth of such attempts in the literature. Results indicate that the proposed Bethlahmy variant exhibits better performance than other methods/variants across samples varying in size and extreme precipitation characteristics, making it a promising statistical alternative for PMP estimation. [ABSTRACT FROM AUTHOR]

• The transient storage model (TSM) is a popular solute transport model for streams. • Parameters of the TSM are of primary interest and are used to assess streams. • Parameters estimation is difficult due to the presence of uncertainties. • Approximate uncertainties are generally used and were found to be inadequate. • Flow variations impact parameter uncertainty and complicate parameter estimation. The transient storage model (TSM) is a popular solute transport model among hydrologists and ecologists for characterizing solute retention potential of small headwater streams. TSM parameters are most commonly estimated in an inverse modeling framework using tracer breakthrough data. A key challenge in interpreting TSM parameters is the inherent difficulty in the estimation process and relating these parameters with the physical characteristics of streams. While often recognized that multiple sources of uncertainty impact transient storage modeling, rigorous evaluation of TSM parameter uncertainty is rare. Stream flow magnitude has been identified as a key variable in characterizing stream solute transport dynamics, yet no study has shown the impact of flow magnitude on TSM parameter uncertainty. In this paper we study the impact of flow magnitude on TSM parameter uncertainty by using tracer data from repeated tracer studies with changing stream flow magnitudes, and introduce a software tool to perform Markov chain Monte Carlo based parameter estimation and uncertainty evaluation. The software tool developed is capable of performing thousands of model runs within a few hours, which is necessary for Monte Carlo based parameter estimation, and provides a potentially attractive framework for model testing and improvement. TSM parameters were found to be relatively less important during higher flows with large uncertainty bounds and clear evidence of parameter correlations. The importance of TSM parameters on solute transport seemed to increase as flow decreased with seasonal baseflow recession. Our findings highlight the need for improved field methodologies to reduce parameter uncertainty and the need for further improvement in the model structure for successful stream characterization. [ABSTRACT FROM AUTHOR]

• Relative impacts of human activities and climate change on river flow are assessed. • The proposed methodology is a stochastic version of the fixing-changing method. • The uncertainties of the models' parameters are analyzed using the DREAM algorithm. • To evaluate the methodology, it was applied to the Zayandehrud basin in Iran. • Results show the applicability and efficiency of the proposed methodology. Climate change and human activities are two important factors affecting surface runoff. In water resource planning and management, it is usually important to separate the contribution of the mentioned factors to the runoff change. In this paper, a new methodology is proposed to obtain the relative impact of human activities and climate change on streamflow under uncertainty. In this methodology, the breakpoint of the annual time series of the observed runoff is estimated and the time series is divided into two study periods; before and after the breakpoint, namely the "natural" and "impacted" periods. The Jazim monthly water balance model is used for monthly runoff simulation. To incorporate the effect of snow in simulating runoff, two Jazim-based monthly water balance models with Rao and Al Wagdany (1995) and McCabe and Markstrom (2007) snow modules are used. Then, the monthly water balance models are calibrated and verified for the natural and impacted periods. The main parameters of the water balance models are optimized using the Genetic Algorithms. The uncertainty of the parameters of the models is analyzed using differential evolution adaptive Metropolis (DREAM) algorithm, which is based on the Markov Chain Monte Carlo (MCMC) algorithm. This algorithm provides the posterior probability distribution functions (PDFs) of parameters of the runoff simulation models. Next, the PDFs of the contribution rate (CR) of the climate change and human activities to runoff are determined using the fixing-changing method by taking into account the PDFs of the models' parameters. To show the applicability and efficiency of the proposed methodology, it is applied to the Zayandehrud sub-basin in Iran. Using the Standard Normal Homogeneity Test (SNHT), Pettitt and Buishand range tests, one breakpoint at 2006 is detected in runoff time series. Through the monthly rainfall-runoff simulations and the fixing-changing method, it is estimated that decrease in the runoff after the breakpoint can be attributed to climate change by 57.8% and 45%, using the Jazim-Rao & Al Wagdany model and the Jazim-McCabe & Markstrom model, respectively. The results of the uncertainty analysis of the parameters of the water balance models show that the parameters of LSMX (maximum moisture capacity of the lower soil layer) and K 3 (deep percolation coefficient) are the most significant parameters in the uncertainty of runoff simulated by both the models. According to the PDFs of contribution rate (CR) of climate change and human activities which are obtained using each rainfall-runoff model, it can be inferred that the Jazim-McCabe & Markstrom model significantly decreased the uncertainty of estimations. [ABSTRACT FROM AUTHOR]

• Three surrogate models are assembled to enhance the optimization process robustness. • AM-MCMC with strong adaptability is adopted to assemble three surrogate models. • Chemical reaction is considered in identifying the groundwater contaminant sources. In identifying groundwater contaminant sources, given that the simulation model is computationally inefficient, an ensemble surrogate model is proposed to improve the accuracy and robustness of results. The proposed ensemble surrogate model in this paper consists of the following three individual surrogate models: Kriging, radial basis functions and least squares support vector machines. The Adaptive Metropolis-Markov Chain Monte Carlo method is used to assign weights to the three models. Accuracy and robustness of the ensemble surrogate model were tested on not only conservative contaminants but also contaminants containing chemical reaction. The results indicated that the proposed ensemble surrogate model is an effective method to solve the inverse contaminant source identification problems with a high degree of accuracy and short computation time. [ABSTRACT FROM AUTHOR]

Highlights • A framework to estimate landslide return period under climate change is proposed. • Method is based on the combined use of RCMs and Monte Carlo simulations. • A sequence of future landslide probabilities is estimated. • The sequence allows estimation of return period according to nonstationary formulas. • Application to the Peloritani Mts (Italy) with MED-Cordex projections is performed. Abstract Quantifying the potential influence of climate change on future landslide hazard requires methodologies that allow to properly take into account nonstationarities in the hydro-meteorological causes. In this paper we provide a methodology for estimating return period of landslide triggering under climate change. The methodology capitalizes on the combined use of a stochastic rainfall generator and a hydrological and slope stability model. The stochastic rainfall generator takes into account the statistical dependency between rainfall event duration and intensity through copulas. The hydrological model is based on an analytical solution of a simplified version of the Richards vertical infiltration equation and slope stability is assessed by the infinite slope model. The combined model enables to estimate landslide probability through Monte Carlo simulations. Climate change is then introduced by perturbing the parameters of the rainfall stochastic generator based on factors of change derived from the comparison of future scenarios and the baseline climate as simulated by Regional climate models (RCMs). The Monte Carlo simulations are conducted sequentially on a future moving time window, to derive a yearly series of future landslide triggering probability. This series is then used to compute landslide return period by formulas suitable under nonstationary conditions. An application to the landslide prone region of the Peloritani Mountains, Southern Italy, is carried out to demonstrate the proposed approach. For the application, climate change projections of three RCMs of the MED-CORDEX initiative are used, and a preliminary assessment of the impacts of intermediate- and high-emission Representative concentration pathways (RCPs) 4.5 and 8.5 is carried out. [ABSTRACT FROM AUTHOR]

Combined simulation-optimization (S/O) schemes have long been recognized as a valuable tool in coastal groundwater management (CGM). However, previous applications have mostly relied on deterministic seawater intrusion (SWI) simulations. This is a questionable simplification, knowing that SWI models are inevitably prone to epistemic and aleatory uncertainty, and hence a management strategy obtained through S/O without consideration of uncertainty may result in significantly different real-world outcomes than expected. However, two key issues have hindered the use of uncertainty-based S/O schemes in CGM, which are addressed in this paper. The first issue is how to solve the computational challenges resulting from the need to perform massive numbers of simulations. The second issue is how the management problem is formulated in presence of uncertainty. We propose the use of Gaussian process (GP) emulation as a valuable tool in solving the computational challenges of uncertainty-based S/O in CGM. We apply GP emulation to the case study of Kish Island (located in the Persian Gulf) using an uncertainty-based S/O algorithm which relies on continuous ant colony optimization and Monte Carlo simulation. In doing so, we show that GP emulation can provide an acceptable level of accuracy, with no bias and low statistical dispersion, while tremendously reducing the computational time. Moreover, five new formulations for uncertainty-based S/O are presented based on concepts such as energy distances, prediction intervals and probabilities of SWI occurrence. We analyze the proposed formulations with respect to their resulting optimized solutions, the sensitivity of the solutions to the intended reliability levels, and the variations resulting from repeated optimization runs. [ABSTRACT FROM AUTHOR]

Due to the uncertainty (i.e., fuzziness, stochasticity and imprecision) existed simultaneously during the process for groundwater remediation, the accuracy of ranking results obtained by the traditional methods has been limited. This paper proposes a cloud model based multi-attribute decision making framework (CM-MADM) with Monte Carlo for the contaminated-groundwater remediation strategies selection. The cloud model is used to handle imprecise numerical quantities, which can describe the fuzziness and stochasticity of the information fully and precisely. In the proposed approach, the contaminated concentrations are aggregated via the backward cloud generator and the weights of attributes are calculated by employing the weight cloud module. A case study on the remedial alternative selection for a contaminated site suffering from a 1,1,1-trichloroethylene leakage problem in Shanghai, China is conducted to illustrate the efficiency and applicability of the developed approach. Totally, an attribute system which consists of ten attributes were used for evaluating each alternative through the developed method under uncertainty, including daily total pumping rate, total cost and cloud model based health risk. Results indicated that A14 was evaluated to be the most preferred alternative for the 5-year, A5 for the 10-year, A4 for the 15-year and A6 for the 20-year remediation. [ABSTRACT FROM AUTHOR]

• Flood warning reliability is assessed for deterministic and probabilistic forecasts. • Forecast quality scenarios are considered to evaluate flood warning reliability. • A floodplain property-based criterion (FPC) for evaluating warning reliability is proposed. • Probabilistic warnings are robust to forecast biases. • FPC is a more demanding criterion than traditional flood warning criteria. Despite the significant progress in probabilistic forecasting science in the last two decades, particularly in the quantification of predictive uncertainty (PU), most operational flood early warning systems (FEWSs) continue to be based on deterministic forecasts. Thereupon, additional work is needed to demonstrate the advantages of using PU over deterministic forecasting to enhance the uptake of probabilistic forecasts in flood warning decision-making. In this paper, a Monte-Carlo (MC)-based sensitivity analysis is done to explore how the outcomes of flood peak water level-based deterministic and -probabilistic warning strategies behave when factors controlling the forecast quality are perturbed. The flood warning reliability is evaluated through the probability of detection (POD) and false alarm ratio (FAR) based on two criteria: a flooding threshold-based criterion (FTC), and a new floodplain property-based criterion (FPC) based on inundation level forecasting. The results of this work show that the advantage of a probabilistic strategy over a deterministic one is greater when PU is relatively high. The probabilistic strategy is robust to biases in the mean and variance of forecasts by maintaining POD and FAR stability, while this is not the case for a deterministic warning strategy. Likewise, it was concluded that if inundation level forecasting is undertaken (FPC), improved forecasts would be needed to achieve the same reliability level as for FTC, reflecting the more demanding FPC criterion. Also, the levels of correlation needed to achieve acceptable operational values of POD and FAR are shown. These results provide new insight into the advantages of a probabilistic forecasting strategy under several forecast quality scenarios and guidance for the design of operational FEWSs. [ABSTRACT FROM AUTHOR]

• Presentation of new distribution free method for change point detection. • Application of several confidence curve methods to hydrometeorological data. • Evaluation of the performance of several confidence curve methods for change point detection. In this paper, a method based on Approximate Empirical likelihood ratio and a Deviance function combined with bootstrapping (AED-BP) is proposed to construct a confidence curve for the location of a change point. The method is compared with a method based on parametric Profile Likelihood and a Deviance function combined with Monte Carlo simulation (PLD-MC). A confidence curve provides a representation of the uncertainty in the outcome of the change point analysis. To evaluate the practical usability of confidence curves constructed by AED-BP, its properties were examined and its performance was compared to that of PLD-MC. The methods were applied to both synthetic and real data. Synthetic data were generated from three parametric distributions: Fréchet with a constant shape parameter, log-normal, and gamma distributions. The real data are the hydrometeorological data analysed in other studies. The change points found in the original publications are used as a reference in this present paper. The results show that AED-BP has a performance that is similar to PLD-MC, but has an advantage in that it is not necessary to select a distribution family for the data. The AED-BP results on the Annual Maximum Runoff series for the stations Yichang and Hankou along the Yangtze river are among the first that show a possible effect of the presence of the Three Gorges dam. [ABSTRACT FROM AUTHOR]

The optimization of spatially complex groundwater management models over long time horizons requires the use of computationally efficient groundwater flow models. This paper presents a new stochastic multi-cell lumped-parameter aquifer model that explicitly considers uncertainty in groundwater recharge. To achieve this, the multi-cell model is combined with the constrained-state formulation method. In this method, the lower and upper bounds of groundwater heads are incorporated into the mass balance equation using indicator functions. This provides expressions for the means, variances and covariances of the groundwater heads, which can be included in the constraint set in an optimization model. This method was used to formulate two separate stochastic models: (i) groundwater flow in a two-cell aquifer model with normal and non-normal distributions of groundwater recharge; and (ii) groundwater management in a multiple cell aquifer in which the differences between groundwater abstractions and water demands are minimized. The comparison between the results obtained from the proposed modeling technique with those from Monte Carlo simulation demonstrates the capability of the proposed models to approximate the means, variances and covariances. Significantly, considering covariances between the heads of adjacent cells allows a more accurate estimate of the variances of the groundwater heads. Moreover, this modeling technique requires no discretization of state variables, thus offering an efficient alternative to computationally demanding methods. [ABSTRACT FROM AUTHOR]

The stochastic runoff-runon process models the volume of infiltration excess runoff from a hillslope via the overland flow path. Spatial variability is represented in the model by the spatial distribution of rainfall and infiltration, and their “correlation scale”, that is, the scale at which the spatial correlation of rainfall and infiltration become negligible. Notably, the process can produce runoff even when the mean rainfall rate is less than the mean infiltration rate, and it displays a gradual increase in net runoff as the rainfall rate increases. In this paper we present a number of contributions to the analysis of the stochastic runoff-runon process. Firstly we illustrate the suitability of the process by fitting it to experimental data. Next we extend previous asymptotic analyses to include the cases where the mean rainfall rate equals or exceeds the mean infiltration rate, and then use Monte Carlo simulation to explore the range of parameters for which the asymptotic limit gives a good approximation on finite hillslopes. Finally we use this to obtain an equation for the mean net runoff, consistent with our asymptotic results but providing an excellent approximation for finite hillslopes. Our function uses a single parameter to capture spatial variability, and varying this parameter gives us a family of curves which interpolate between known upper and lower bounds for the mean net runoff. [ABSTRACT FROM AUTHOR]

• A unit plot SWMM model was developed to evaluate the effect of LID control. • The uncertainty of vegetative swales is the highest, permeable pavement is the lowest. • Obtain the optimal combination of parameters based on the Pareto principle. • The three indicators of LID can be estimated through the regression equation. Low Impact Development (LID) facilities are the key to realizing the source control of stormwater. However, there are few studies on the control effects of different LID facilities on runoff volume and runoff quality. In this paper, based on the Storm Water Management Model (SWMM), a Monte Carlo method was used to simulate the runoff process for five LID facilities. The correlation analysis and scoring of the three indicators of runoff coefficient, peak runoff, and Suspended Solid (SS) wash-off in the simulation results were conducted to evaluate the control effects of different LID facilities. The simulation results showed an evident law of decreasing marginal effect in the deployed area of LID facilities. In addition, the uncertainty of the hydrological response of permeable pavement is the smallest among the five types of LID facilities, and the uncertainty of vegetative swales is the largest. Under the suitable combination of parameters, the area percentage of permeable pavement reaching the upper limit of the control effect of three indicators is the smallest; the control effect of three indicators of infiltration trench is the best. Correlation analysis showed that the three indicators of the LID facilities could be estimated from each other through the regression relationship equation. [ABSTRACT FROM AUTHOR]

• The confidence intervals of the KG efficiency do not have a closed-form description. • The uncertainty of KG efficiency is attributed to measurement errors of training data. • Statistical inference with an informal goodness-of-fit metric is coined diagnostic regression. • Parameter confidence regions of KG efficiency in agreement with linear regression theory. • Marginal parameter distributions of KG efficiency are approximately Gaussian. The Kling-Gupta efficiency, hereafter referred to as KG efficiency rather than its common abbreviation KGE, proposed by Gupta et al. (2009) has become a widely used metric for evaluating the goodness-of-fit of n -vectors of observations, y ̃ = [ y ̃ 1 y ̃ 2 ... y ̃ n ] ⊤ , and corresponding model simulations, y θ = y 1 θ y 2 θ ... y n θ ⊤ . This metric rectifies some of the shortcomings of the coefficient of determination, R 2 , also known to hydrologists as the efficiency of Nash and Sutcliffe (1970) , by using a Euclidean-distance based weighting of the correlation, bias and temporal variability of the observed, y ̃ , and simulated, y θ , data. But as the KG efficiency is not borne out of assumptions with respect to the statistical distribution of the residuals, e θ = y ̃ - y θ , we cannot formally characterize its uncertainty. The NS efficiency suffers a similar problem, yet, statistical theory postulates that its confidence intervals should follow a beta distribution in certain special cases. Without a formal description of the confidence intervals of the KG efficiency, we cannot (amongst others) quantify parameter uncertainty, compute confidence and prediction limits on simulated model responses, inform decision makers about critical modeling uncertainties, evaluate model adequacy and assess the information content of calibration data. More fundamentally, without confidence intervals we cannot establish whether the KG efficiency is a consistent, efficient and unbiased estimator. In this paper we present an empirical description of the confidence intervals of the KG efficiency. We relate the unknown probability distribution of the KG efficiency to the measurement errors of the training data record, y ̃ , and use the bootstrap method to carry out statistical inference. We illustrate our method by application to a simple linear regression function for which the least squares parameter confidence regions are exactly known and two hydrologic models of contrasting complexity. The empirical parameter confidence regions and/or intervals of the KG efficiency are compared to those derived from generalized least squares, objective function contouring and Bayesian analysis using Markov chain Monte Carlo simulation. The marginal parameter distributions of the KG efficiency are generally well described by a normal distribution. Results further confirm that the distribution of the KG efficiency is a complex function of data length and the magnitude, distribution and structure of the measurement errors. This prohibits an analytic description of the empirical confidence regions and/or intervals of the KG efficiency and reiterates the need for the bootstrap method. [ABSTRACT FROM AUTHOR]

Summary This paper describes the training, validation, testing and uncertainty analysis of general regression neural network (GRNN) models for the forecasting of dissolved oxygen (DO) in the Danube River. The main objectives of this work were to determine the optimum data normalization and input selection techniques, the determination of the relative importance of uncertainty in different input variables, as well as the uncertainty analysis of model results using the Monte Carlo Simulation (MCS) technique. Min–max, median, z -score, sigmoid and tanh were validated as normalization techniques, whilst the variance inflation factor, correlation analysis and genetic algorithm were tested as input selection techniques. As inputs, the GRNN models used 19 water quality variables, measured in the river water each month at 17 different sites over a period of 9 years. The best results were obtained using min–max normalized data and the input selection based on the correlation between DO and dependent variables, which provided the most accurate GRNN model, and in combination the smallest number of inputs: Temperature, pH, HCO 3 − , SO 4 2− , NO 3 -N, Hardness, Na, Cl − , Conductivity and Alkalinity. The results show that the correlation coefficient between measured and predicted DO values is 0.85. The inputs with the greatest effect on the GRNN model (arranged in descending order) were T , pH, HCO 3 − , SO 4 2− and NO 3 -N. Of all inputs, variability of temperature had the greatest influence on the variability of DO content in river body, with the DO decreasing at a rate similar to the theoretical DO decreasing rate relating to temperature. The uncertainty analysis of the model results demonstrate that the GRNN can effectively forecast the DO content, since the distribution of model results are very similar to the corresponding distribution of real data. [ABSTRACT FROM AUTHOR]

• A general uncertainty partition approach was developed for river flood routing. • The effects of three uncertainty sources are significantly different and dependent. • The simulated water stage is most influenced by riverbed roughness uncertainty. • The partition results are useful for flood routing uncertainty identification. The partition of flood routing uncertainties from multiple sources provides a foundation for managing flood risks. This paper develops a partition approach for the one-dimensional river flood routing uncertainty due to boundary conditions (upstream inflow hydrograph and downstream stage hydrograph) and riverbed roughness. A theoretical derivation based on the Monte Carlo method is presented using a one-dimensional hydrodynamic model. The developed approach is successfully applied to a real-world case of the Xunjiang River in China. The results indicate the following: (1) each uncertainty source leads to unique effects on flood routing, and the uncertainty contribution ratio varies significantly with both time and space, providing evidences for identifying the dominant uncertainty sources; (2) the uncertainty of the simulated discharge is most influenced by the boundary conditions whereas the simulated water stage is most influenced by the riverbed roughness; and (3) the highly coupled uncertainty propagation is hardly affected by the flood pattern, as well as the probability distribution of uncertainty source, but it is significantly affected by the river morphology. The developed approach can effectively partition the uncertainty effects of the boundary conditions and riverbed roughness without explicating their interactions. And the uncertainty partition of flood routing is useful for risky decision making in flood management. [ABSTRACT FROM AUTHOR]

Summary: In regional hydrologic regression analysis, model selection and validation are regarded as important steps. Here, the model selection is usually based on some measurements of goodness-of-fit between the model prediction and observed data. In Regional Flood Frequency Analysis (RFFA), leave-one-out (LOO) validation or a fixed percentage leave out validation (e.g., 10%) is commonly adopted to assess the predictive ability of regression-based prediction equations. This paper develops a Monte Carlo Cross Validation (MCCV) technique (which has widely been adopted in Chemometrics and Econometrics) in RFFA using Generalised Least Squares Regression (GLSR) and compares it with the most commonly adopted LOO validation approach. The study uses simulated and regional flood data from the state of New South Wales in Australia. It is found that when developing hydrologic regression models, application of the MCCV is likely to result in a more parsimonious model than the LOO. It has also been found that the MCCV can provide a more realistic estimate of a model’s predictive ability when compared with the LOO. [ABSTRACT FROM AUTHOR]

Summary: Hybrid data assimilation (DA) is greatly used in recent hydrology and water resources research. In this study, one newly introduced technique, the ensemble particle filter (EnPF), formed by coupling ensemble Kalman filter (EnKF) with particle filter (PF), is applied for a multi-layer soil moisture prediction in the Meilin watershed based on the support vector machines (SVMs). The data used in this paper includes six-layer soil moisture: 0–5cm, 30cm, 50cm, 100cm, 200cm and 300cm and five meteorological parameters: soil temperature at 5cm and 20cm, air temperature, relative humidity and solar radiation in the study area. In order to investigate this EnPF approach, another two filters, EnKF and PF are applied as another two data assimilation methods to conduct a comparison. In addition, the SVM model simulated data without updating with data assimilation technique is discussed as well to evaluate the data assimilation technique. Two experimental cases are explored here, one with 200 initial training ensemble members in the SVM training phase while the other with 1000 initial training ensemble members. Three main findings are obtained in this study: (1) the SVMs machine is a statistically sound and robust model for soil moisture prediction in both the surface and root zone layers, and the larger the initial training data ensemble, the more effective the operator derived; (2) data assimilation technique does improve the performance of SVM modeling; (3) EnPF outweighs the performance of other two filters as well as the SVM model; Moreover, the ability of EnPF and PF is not positively related to the resampling ensemble size, when the resampling size exceeds a certain amount, the performance of EnPF and PF would be degraded. Because the EnPF still performs well than EnKF, it can be used as a powerful data assimilation tool in the soil moisture prediction. [Copyright &y& Elsevier]

Summary: This paper evaluates three distinct approaches for disaggregating daily rainfall to sub-daily sequences: (1) random multiplicative cascades (microcanonical and canonical versions), (2) point process (randomized Bartlett–Lewis model – RBLM), and (3) resampling (method of fragments). These methods are used to perform disaggregation of daily rainfall to hourly rainfall at four point locations across Australia (Sydney, Perth, Cairns, and Hobart), which are associated with different climatic regimes. The methods are evaluated based on parameter estimation procedures applied (including introduction of the sequential Monte Carlo sampler in RBLM), the capability of the resulting sequences to reproduce standard validation statistics, and the representation of observed rainfall variability and intermittency, within-day wet spells, and extreme rainfall percentiles. The results generally indicate that the method of fragments outperforms the other models. While all the models are found to simulate reasonably well the commonly used statistical measures (e.g. mean and dry proportions) of rainfall at the hourly timestep, the microcanonical model is found to significantly overestimate the hourly rainfall variance. With respect to extreme value characteristics, the resampling approach is found to match well the observed intensity–frequency relationship at an hourly scale, with the cascade models underestimating (canonical) and overestimating (microcanonical) extreme rainfall. The point process model’s performance is poor in Cairns but reasonably good at other locations. An analysis of the empirical within-day wet- and dry-spell distributions further reveals that the cascade-based models are not robust for observed wet and dry spells. [Copyright &y& Elsevier]

Summary: Design floods for a given location at a stream can be estimated by a number of approaches including flood frequency statistics and the design storm method. If applied to the same catchment the two methods often yield quite different results. The aim of this paper is to contribute to understanding the reasons for these differences. A case study is performed for 10 alpine catchments in Tyrol, Austria, where the 100-year floods are estimated by (a) flood frequency statistics and (b) an event based runoff model. To identify the sources of the differences of the two methods, the 100-year floods are also estimated by (c) Monte Carlo simulations using a continuous runoff model. The results show that, in most catchments, the event based model gives larger flood estimates than flood frequency statistics. The reasons for the differences depend on the catchment characteristics and different rainfall inputs that were applied. For catchments with a high storage capacity the Monte Carlo simulations indicate a step change in the flood frequency curve when a storage threshold is exceeded which is not captured by flood frequency statistics. Flood frequency statistics therefore tends to underestimate the floods in these catchments. For catchments with a low storage capacity or significant surface runoff, no step change occurs, but in three catchments the design storms used were larger than those read from the IDF (intensity duration frequency) curve leading to an overestimation of the design floods. Finally, also the correct representation of flood dominating runoff components was shown to influence design flood results. Geologic information on the catchments was essential for identifying the reasons for the mismatch of the flood estimates. [ABSTRACT FROM AUTHOR]

Summary: The reliability of over-year storage reservoirs has been traditionally estimated by using Monte Carlo simulation (behavior analysis), numerical solution of the stochastic reservoir equation (difference or integral equations), or by using approximations and empirical relationships. The case of persistent inflow, which is the case for most natural streams, adds complexity to the problem. Apart from simulation, which is the most flexible approach, other methods have had partial success in this case, and are mainly limited to the case of Markovian autocorrelation. This paper suggests an alternative method for calculating the reliability of over-year storage reservoirs with persistent gaussian inflow using basic probability principles and the numerical integration of the standard multi-variate normal distribution. It is shown that the classical probabilities of the reservoir being empty and those of spilling can be directly calculated based on the distribution of the critical drawdown and filling periods. The results obtained by the application of the proposed method compare very well with those obtained by Monte Carlo simulation and those reported in previous studies. The proposed method thus offers another alternative that should be useful in handling complex problems. [Copyright &y& Elsevier]

Summary: In petroleum hydrocarbon contamination scenarios, assessment of the dimensions of contaminant plumes and prediction of their fate requires predictions of the rate of dissolution of contaminants from residual non-aqueous-phase-liquid (NAPLs) into the aquifer and the rate of contaminant removal through biodegradation. The central idea in this paper is to present an inverse modeling methodology for characterization of structural uncertainty in these parameters by tailoring their estimation to the distributions of source geometry and hydraulic conductivity field. For this purpose, a synthetic study site with two reference cases is established and a Monte Carlo type inverse modeling methodology is presented where dissolution and first-order biodegradation rates of the contaminants are estimated for joint realizations of source and hydraulic conductivity. The joint realizations are constructed by the distance-function (DF) and sequential self-calibration (SSC) approaches and a gradient-based optimization is adapted to solve the inversion problem. The results show larger uncertainty in the estimated dissolution rate and a moderate positive correlation between the two parameters. It is also observed that tailoring the estimation of the parameters to the constructed joint realizations can effectively reduce the uncertainty in the shape and size of the plume, and this uncertainty, as well as the uncertainty in the source size, can be further reduced by ranking and screening the conditional realizations based on the value of the objective function. The effects of measurement errors in head and concentration observations on uncertainty in the predicted parameters are shown to be as expected; measurement error translates to substantially greater uncertainty. [Copyright &y& Elsevier]

Summary: The validity of load estimates from intermittent, instantaneous grab sampling is dependent on adequate spatial coverage by monitoring networks and a sampling frequency that reflects the variability in the system under study. Catchments with a flashy hydrology due to surface runoff pose a particular challenge as intense short duration rainfall events may account for a significant portion of the total diffuse transfer of pollution from soil to water in any hydrological year. This can also be exacerbated by the presence of strong background pollution signals from point sources during low flows. In this paper, a range of sampling methodologies and load estimation techniques are applied to phosphorus data from such a surface water dominated river system, instrumented at three sub-catchments (ranging from 3 to 5km2 in area) with near-continuous monitoring stations. Systematic and Monte Carlo approaches were applied to simulate grab sampling using multiple strategies and to calculate an estimated load, Le based on established load estimation methods. Comparison with the actual load, Lt , revealed significant average underestimation, of up to 60%, and high variability for all feasible sampling approaches. Further analysis of the time series provides an insight into these observations; revealing peak frequencies and power-law scaling in the distributions of P concentration, discharge and load associated with surface runoff and background transfers. Results indicate that only near-continuous monitoring that reflects the rapid temporal changes in these river systems is adequate for comparative monitoring and evaluation purposes. While the implications of this analysis may be more tenable to small scale flashy systems, this represents an appropriate scale in terms of evaluating catchment mitigation strategies such as agri-environmental policies for managing diffuse P transfers in complex landscapes. [ABSTRACT FROM AUTHOR]

Summary: A recently developed unified model for partially-penetrating slug tests in unconfined aquifers () provides a semi-analytical solution for aquifer response at the wellbore in the presence of inertial effects and wellbore skin, and is able to model the full range of responses from overdamped/monotonic to underdamped/oscillatory. While the model provides a unifying framework for realistically analyzing slug tests in aquifers (with the ultimate goal of determining aquifer properties such as hydraulic conductivity K and specific storage S s ), it is currently unclear whether parameters of this model can be well-identified without significant prior information and, thus, what degree of information content can be expected from such slug tests. In this paper, we examine the information content of slug tests in realistic field scenarios with respect to estimating aquifer properties, through analysis of both numerical experiments and field datasets. First, through numerical experiments using Markov Chain Monte Carlo methods for gauging parameter uncertainty and identifiability, we find that: (1) as noted by previous researchers, estimation of aquifer storage parameters using slug test data is highly unreliable and subject to significant uncertainty; (2) joint estimation of aquifer and skin parameters contributes to significant uncertainty in both unless prior knowledge is available; and (3) similarly, without prior information joint estimation of both aquifer radial and vertical conductivity may be unreliable. These results have significant implications for the types of information that must be collected prior to slug test analysis in order to obtain reliable aquifer parameter estimates. For example, plausible estimates of aquifer anisotropy ratios and bounds on wellbore skin K should be obtained, if possible, a priori. Secondly, through analysis of field data – consisting of over 2500 records from partially-penetrating slug tests in a heterogeneous, highly conductive aquifer, we present some general findings that have applicability to slug testing. In particular, we find that aquifer hydraulic conductivity estimates obtained from larger slug heights tend to be lower on average (presumably due to non-linear wellbore losses) and tend to be less variable (presumably due to averaging over larger support volumes), supporting the notion that using the smallest slug heights possible to produce measurable water level changes is an important strategy when mapping aquifer heterogeneity. Finally, we present results specific to characterization of the aquifer at the Boise Hydrogeophysical Research Site. Specifically, we note that (1) K estimates obtained using a range of different slug heights give similar results, generally within ±20%; (2) correlations between estimated K profiles with depth at closely-spaced wells suggest that K values obtained from slug tests are representative of actual aquifer heterogeneity and not overly affected by near-well media disturbance (i.e., “skin”); (3) geostatistical analysis of K values obtained indicates reasonable correlation lengths for sediments of this type; and (4) overall, K values obtained do not appear to correlate well with porosity data from previous studies. [Copyright &y& Elsevier]

Summary: A major challenge confronting the scientific community is to understand both patterns of and controls over spatial and temporal variability of suspended sediment dynamics in rivers, as these sediment govern nutriment export, river morphology, siltation of downstream reservoirs and degradation of water quality. High-frequency suspended sediment monitoring programs are required to meet this goal, particularly research in highly erodible mountainous catchments which supply the sediment load of the entire downstream fluvial network. However, in this context, analysis of the data and their interpretation are generally limited by many sources of uncertainty in river monitoring. This paper proposes to estimate the global uncertainty of suspended sediment monitoring using turbidimeter in a small mountainous river catchment (22km2; Southern French Alps). We first conducted a detailed analysis of the main uncertainty components associated with the turbidity approach, i.e. a widely used method to continuously survey the suspended sediment concentration (SSC). These uncertainty components were then propagated with Monte Carlo simulations. For individual records, SSC uncertainties are found to be on average less than 10%, but they can reach 70%. At the flood scale, the mean and the maximum SSC uncertainties are on average 20% (range, 1–30%), whereas sediment yield uncertainty is a mean 30% (range, 20–50% depending on the flood considered; discharge error, 20%). Annual specific sediment yield (SSY*) was then 360±100tkm−2 year−1. Uncertainty components associated with the automatic pumping procedure, discharge measurement and turbidity fluctuation at the short time scale were found to be the greatest uncertainties. SSC and SSY uncertainties were found highly site- and time-dependent as they vary significantly with the hydro-sedimentary conditions. This study demonstrates that global uncertainty accounts for only a small part of inter-flood SSC and SSY variability. It outlines the controlling factors of land use, relief, geology and rainfall regime on suspended sediment yields. [Copyright &y& Elsevier]

Summary: The design of an effective groundwater remediation system involves the determination of the source zone characteristics and subsequent source zone removal. The work presented in this paper focuses on the three-dimensional extension and field application of a previously described source zone identification and delineation algorithm. The three-dimensional search algorithm defines how to achieve an acceptable level of accuracy regarding the strength, geographic location and depth of a dense non-aqueous phase liquid (DNAPL) source while using the least possible number of water quality samples. Target locations and depths of potential sources are identified and given initial importance measures or weights using a technique that exploits expert knowledge. The weights reflect the expert’s confidence that the particular source location is the correct one and they are updated as the investigation proceeds. The overall strategy uses stochastic groundwater flow and transport modeling assuming that hydraulic conductivity is known with uncertainty (Monte Carlo approach). Optimal water quality samples are selected according to the degree to which they contribute to the total concentration uncertainty reduction across all model layers and the proximity of the samples to the potential source locations. After a sample is taken, the contaminant concentration plume is updated using a Kalman filter. The set of optimal source strengths is determined using linear programming by minimizing the sum of the absolute differences between modeled and measured concentration values at sampling locations. The Monte Carlo generated suite of plumes emanating from each individual source is calculated and compared with the updated plume. The scores obtained from this comparison serve to update the weights initially assigned by the expert, and the above steps are repeated until the optimal source characteristics are determined. The algorithm’s effectiveness is demonstrated by performing a ‘blind test’ at a field site in the Anniston Army Depot (ANAD) in Alabama. [Copyright &y& Elsevier]

Summary: This paper proposes a method for using major flash flood events occurred at ungauged catchments to reduce the uncertainties in estimating regional flood quantiles. The approach is based on standard regionalization methods assuming that the flood peak distribution rescaled by a site-dependent index flood is uniform within a homogeneous region. A likelihood formulation and a Bayesian Markov Chain Monte Carlo (MCMC) algorithm are used to infer the parameter values of the regional distributions. This statistical inference technique has been selected for its rigorousness – various hypotheses are explicitly formulated in the likelihood function, its flexibility as for the type of data that can be treated, and its ability to compute accurate estimates of the confidence intervals for the adjusted parameters and for the corresponding flood quantiles. The proposed method is applied to two data sets from Slovakia and the South of France that consist of series of annual peak discharges at gauged sites and estimated peak discharges of extreme flash flood events that have occurred at ungauged sites. The results suggest that the confidence intervals of the quantiles can be significantly narrowed down provided that the set of ungauged extremes is the result of a comprehensive sampling over the selected region. This remains valid, even if the uncertainties in the estimated ungauged extreme discharges are considered. The flood quantiles estimated by the proposed method are also consistent with the results of site specific flood frequency studies based on historic and paleoflood information. [ABSTRACT FROM AUTHOR]

Summary: Australia is typical of many countries in that it has an extensive network of rainfall recording stations which record rainfall data in various forms ranging from a daily time step down to 6-min resolution. However, while the length of historical daily records is often large, there are very few 6-min (pluviograph) records available of significant length. Not only does this lack of significant short time scale data impose a major obstacle in the application of a Monte Carlo approach to risk estimation, it also inhibits the application of rainfall simulation models that use this data for direct calibration. While the advent of numerous stochastic rainfall models provides methods for extending historical rainfall records, without adequate historical rainfall data available for calibration their accuracy is questionable. This paper describes the development of a new technique which significantly extends the applicability of stochastic point rainfall models that require historical data for calibration. The technique uses a new ‘master–target’ scaling relationship. A model calibration is undertaken at a ‘master’ site with a long pluviograph record, which is then scaled to the ‘target’ site using the information from the target site in form of either a short pluviograph or a daily rainfall record. This approach removes the need for significant pluviograph data at the ‘target’ site and enables the stochastic rainfall model to be applied at sites with either short pluviograph or daily rainfall records. The master–target scaling technique is demonstrated using an existing high-resolution point rainfall model based on wet-dry alternating storm events. Extensive testing using numerous pairs of Australian sites demonstrates its validity. [ABSTRACT FROM AUTHOR]

Summary: According to the Chinese Flood Control Act, reservoir water levels generally are not allowed to exceed the flood limited water level (FLWL) during flood season in order to offer adequate storage for flood prevention. However, the operation rules based on the current FLWL have neglected meteorological and real-time flood forecasting information and give too much priority to low probability floods. For floodwater utilization, dynamic control of reservoir FLWL is a valuable and effective methodology to compromise between flood control and conservation for reservoir operation during the flood season. The dynamic control bound is a fundamental key element for implementing reservoir FLWL dynamic control operation. In this paper, a dynamic control operation model that considers inflow uncertainty, i.e. the inflow forecasting error and uncertainty of the flood hydrograph shape is proposed and developed. The model consists of three modules: the first one is a pre-release module, which is used to estimate the upper boundary of dynamic control bound on basis of inflow forecasting results; the second one is a refill operation module, which is used to retain recession flood, and the third one is a risk analysis module, which is used to assess flood risk. The acceptable flood control operation risk constraints and quantificational analysis methods are given, and the dynamic control bound of reservoir FLWL is estimated by using Monte Carlo simulation. The China’s three gorges reservoir (TGR) is selected as a case study. A multiple-input single-output linear systematic model is chosen for inflow forecasting of the TGR, and the future inflows are derived from gauged records by assuming that the inflow forecasting error follows a normal distribution. The application results show that the dynamic control of reservoir FLWL can effectively increase hydropower generation and the floodwater utilization rate without increasing flood control risk. [ABSTRACT FROM AUTHOR]

Summary: Reach-average hydraulic geometry (RHG) defines the relations between the average hydraulic variables estimated at river reach (water depth, velocity, channel width) and the discharge. Single log-linear models are classically used by fluvial geomorphologists, hydrologists and ecologists to represent these relations. In this paper, we show that these single log-linear models are not relevant in most of cases and so can lead to significant errors. RHG estimated on 15 alluvial river reaches (the catchment sizes vary from 10km2 to 1700km2) were found non-linear for two-thirds of the river reaches when considering discharges ranging from low to bankfull flows. So we propose using two power functions (or log-linear) to model these relations, i.e. the log piecewise linear model. This model can detect a discharge Qbreak, i.e. a physical transition or “break point” between both relations. From very low flows (about 30% of the median discharge) to Qbreak, a first relation is comparable to the RHG found in the literature for aquatic habitat analysis. The second relation is relevant for flows ranging from Qbreak to bankfull discharge, and therefore may be more useful for sediment transport and flood routing analysis. Averaging along a reach including several pool–riffle units makes it difficult to clearly ascribe the form of RHG to particular physical processes. The non-linearity of RHG may have multiple origins: the presence of lateral gravel bars, the hydraulic conditions of riffle cross-sections, a change in the resistance to flow with water depth (vegetation, sediment influences). Further research is needed to focus on the physical processes that govern the RHG non-linearity. [Copyright &y& Elsevier]

Summary: The evaluation of urban stormwater quality is of relevant importance for urban drainage, and mathematical models may be of great interest in this respect. To date, several detailed mathematical models are available to predict stormwater quantity–quality characteristics in urban drainage systems. However, only a few models take sewer sediments into account, considering their cohesive-like properties that influence the build-up process of the pollutant load. Furthermore, the model data requirements, especially for the quality aspects, are extensive, which limit their applicability and affect model results with large uncertainty. Uncertainty analysis provides a measure or index regarding the significance and the accuracy of the results obtained by mathematical modelling and is therefore of high interest. Nevertheless, only few studies have been carried out in the urban drainage field, and very few deal with water quality issues. One of the main reasons for this lack of research is the computational burden required by detailed models that preserve this analysis and generally require several Monte Carlo simulation runs. A possible to this problem may be the adoption of simplified parsimonious models that generally require shorter computational times. In this context, this paper presents a parsimonious conceptual model for the evaluation of the pollutant load in-sewers. The model contains two modules: a hydrological and hydraulic module that calculates the hydrographs at the inlet and at the outlet of the sewer system, and a solid transfer module that calculates the pollutographs. The cohesive properties of sewer sediments were carefully considered. Further, the effectiveness of the innovative sewer sediment modelling approach has been verified by taking into account the uncertainty assessed according to the GLUE methodology. The model has been tested using experimental quantity–quality data gathered in two Italian catchments, Fossolo (Bologna) and Parco d’Orlèans (Palermo). [Copyright &y& Elsevier]

Summary: This paper presents a method for regional frequency analysis and spatio-temporal pattern characterization of rainfall-extreme regimes (i.e. extremes, durations and timings) in the Pearl River Basin (PRB) using the well-known L-moments approach together with advanced statistical tests including stationarity test and serial correlation check, which are crucial to the valid use of L-moments for frequency analysis. Results indicate that: (1) the entire Pearl River Basin (40 sites) can be categorized into six regions by cluster analysis together with consideration of the topography and spatial patterns of mean precipitation in the basin. The results of goodness-of-fit measures indicate that the GNO, GLO, GEV, and PE3 distributions fit well for most of the basin for different HOM regions, but their performances are slightly different in term of curve fitting; (2) the estimated quantiles and their biases approximated by Monte Carlo simulation demonstrate that the results are reliable enough for the return periods of less than 100years; (3) excessive precipitation magnitude records are observed at Guilin region of Guangxi Province and Fogang region of Guangdong Province, which have sufficient climate conditions (e.g. precipitation and humidity) responsible for the frequently occurred flood disasters in the regions. In addition, the spatial variations of precipitation in different return periods (Return period=1, 10, 50years to 100years) increase from the upstream to downstream at the regional scale; (4) the seasonal patterns of precipitation extremes for different topographical regions are different. The major precipitation events of AM1R, AM3R, AM5R and AM7R in regions of low-elevation in lower (south-eastern) part of the basin occur mainly in May, June, July and August, while the main precipitation periods for the mountainous region upstream are June, July and August. Further analysis of the NCAR/NCEP reanalysis data indicates that the eastern Asian summer monsoon and typhoons (or hurricanes) are major metrological driving forces on the precipitation regimes. Additionally, topographical features (i.e. elevation, distance to the sea, and mountain’s influences) also exert different impacts on the spatial patterns of such regimes. To the best of our knowledge, this study is the first attempt to conduct a systematic regional frequency analysis on various annual precipitation extremes (based on consecutive 1-, 3-, 5-, 7-day averages) and to establish the possible links to climate pattern and topographical features in the Pearl River Basin and even in China. These findings are expected to contribute to exploring the complex spatio-temporal patterns of extreme rainfall in this basin in order to reveal the underlying linkages between precipitation and floods from a broad geographical perspective. [Copyright &y& Elsevier]

Summary: The main aim of frequency analysis of hydrological data is the definition of the quantiles with a given return period, also known as return levels, namely, the values of a hydrological variable X exceeded on average once in T time intervals. To make the analysis effective, these values have to be complemented with information about their uncertainty, which is usually quantified by confidence intervals (CIs). The paper discusses and compares some parametric, non-parametric, and simulation techniques to build CIs for quantiles with emphasis on the extreme ones. In particular, the work focuses on the usefulness of fractional order statistics to compute CIs for quantiles. This approach appears attractive as it involves simple analytical formulas and allows computing fully non-parametric CIs without requiring information about the parent distribution of the data. Since CIs for order statistics can be used to approximate CIs for quantiles based on order statistics, the relation between them is explored as well. The applicability of methods based on order statistics is assessed by studying the actual coverage probability through simulations, and applying these approaches to annual maximum peak discharge records from 22 streamgage stations in the continental United States and a streamgage station downtown Rome, Italy. For extreme quantiles, results show that these techniques are generally outperformed in term of coverage probability by other methods, such as Monte Carlo approach, when the parent distribution is well-specified or mis-specification involves distributions with tail behavior similar to the one used to generate the sample. On the other hand, the application to real world data indicates that the values of confidence limits based on order statistics are close to those obtained by Monte Carlo approach. In particular, the upper limits from non-parametric setups often result in cautionary values. [Copyright &y& Elsevier]

Summary: Contamination created by dense non-aqueous phase liquids (DNAPLs) is a serious threat to the quality of the groundwater supply; thus, delineating and removing the DNAPL source is an essential step in a successful remediation strategy. The goal of this work is to create an optimal search strategy that identifies a DNAPL source using the least number of water quality samples. The search strategy includes a Monte Carlo stochastic groundwater flow and transport model, a predetermined set of potential source locations and a Kalman filter that updates the simulated contaminant concentration field using contaminant concentration data. The updated plume is compared to the concentration fields that emanate from each individual potential source using a technique rooted in fuzzy set theory. The comparison provides weights that reflect the degree of truth regarding the location of the source at the selected potential source locations. The steps described above are repeated until the weights stabilize and the optimal source location is determined. This paper presents the various mathematical tools employed in the search algorithm and their modifications for incorporation into the proposed methodology. The algorithm’s effectiveness in identifying a DNAPL source is shown using a synthetic example. [Copyright &y& Elsevier]

Summary: Many water resource investigations require an estimate of evaporation from water supply reservoirs. Despite its recognised limitations, the pan coefficient method for estimating reservoir evaporation is widely used in industry due to its moderate data requirements. This paper presents a framework to quantify the uncertainties associated with estimates of reservoir evaporation generated using the pan coefficient method. The uncertainties associated with each of the inputs are quantified using a combination of statistical techniques, including Bayesian statistics and subjective judgement; before being combined using Monte Carlo simulation. The framework is applied to three reservoirs in the Werribee River catchment, Australia. The 95% probability intervals surrounding the estimates of reservoir evaporation are as large as ±40% of the best estimate. The largest contributing factor to uncertainty is shown to be the estimation of pan evaporation at reservoirs without a climate station, followed by the uncertainty surrounding the annual pan coefficient itself. The assessment shows that the largest reductions in uncertainty can be achieved by installing an evaporation pan at the reservoir. [Copyright &y& Elsevier]

Summary: This paper compares two Monte Carlo sequential data assimilation methods based on the Kalman filter, for estimating the effect of measurements on simulations of state error variance made by a one-dimensional hydrodynamic model. The first method used an ensemble Kalman filter (EnKF) to update state estimates, which were then used as initial conditions for further simulations. The second method used an ensemble transform Kalman filter (ETKF) to quickly estimate the effect of measurement error covariance on forecast error covariance without the need to re-run the simulation model. The ETKF gave an unbiased estimate of EnKF analysed error variance, although differences in the treatment of measurement errors meant the results were not identical. Estimates of forecast error variance could also be made, but their accuracy deteriorated as the time from measurements increased due in part to model non-linearity and the decreasing signal variance. The motivation behind the study was to assess the ability of the ETKF to target possible measurements, as part of an adaptive sampling framework, before they are assimilated by an EnKF-based forecasting model on the River Crouch, Essex, UK. The ETKF was found to be a useful tool for quickly estimating the error covariance expected after assimilating measurements into the hydrodynamic model. It, thus, provided a means of quantifying the ‘usefulness’ (in terms of error variance) of possible sampling schemes. [Copyright &y& Elsevier]

Summary: Monte-Carlo (MC) type inverse modelling techniques like the sequential self-calibration (SSC) method, can be used as a benchmark to test other inverse modelling procedures. In comparison studies MC type inverse modelling methods outperformed other inverse parameter estimation methods, but the large amount of CPU time needed can become prohibitive to use such methods. Therefore, an interest exists to develop alternative methods that perform nearly as good and need much less CPU time. In this paper Ensemble Kalman Filtering (EnKF) is promoted for the off-line calibration of transient groundwater flow models with many nodes. An augmented state vector approach is used to calibrate parameters together with the updating of the states. EnKF and SSC are compared in two calibration experiments, for a mildly heterogeneous case and a strongly heterogeneous case . For the mildly heterogeneous case, EnKF gives very similar results to SSC, also in terms of calibrated log-transmissivity fields. For the strongly heterogeneous case, EnKF gives still similar results as SSC, although the characterisation of the log T field improves less compared to SSC. On the other hand, EnKF needed in both cases around a factor of 80 less CPU time than SSC. In addition, the performance of EnKF and SSC was compared in two prediction experiments: (1) the prediction of groundwater flow in a different flow situation (without pumping and with a different time series of recharge rate), (2) the prediction of solute transport towards a pumping well. Both in the mildly and strongly heterogeneous cases the quality of the predictions with EnKF was as good as for SSC. Given the good performance of EnKF, the strength of EnKF to include multiple sources of uncertainty (e.g., related to external forcing) and the reduced CPU time needed compared to MC type inverse modelling, EnKF seems to be an interesting candidate for the stochastic calibration of large subsurface hydrological models. [Copyright &y& Elsevier]

• The modified Kumaraswamy (MK) and the reflected MK (RMK) distributions are presented. • The MK and RMK are alternatives to the beta and Kumaraswamy distributions. • The MK density assumes increasing-decreasing-increasing shapes. • Important statistical properties of MK and RMK distributions are presented. • An application to Brazilian hydro-environmental data is presented and discussed. In this paper, we propose two-parameter probability models referred to as modified Kumaraswamy (MK) and reflected modified Kumaraswamy (RMK) distributions for modeling hydro-environmental data in the unit interval. The MK distribution presents increasing-decreasing-increasing density shapes while the RMK distribution presents decreasing-increasing-decreasing shapes, which are not obtained in classical distributions for modeling double bounded data, such as the beta and Kumaraswamy. Some statistical properties are derived, such as moments and quantile function. We also consider the maximum likelihood approach for parameter estimation and perform a Monte Carlo simulation study to evaluate the performance of the estimators on synthetic hydro-environmental data. Finally, an empirical application to the percentage of useful volume of 37 Brazilian water reservoirs is presented and discussed. The modeled densities and goodness-of-fit measures evidence that the proposed distributions outperform the beta and Kumaraswamy models, being suitable alternatives for modeling hydro-environmental data. [ABSTRACT FROM AUTHOR]

Summary: Groundwater quality management relies more and more on models in recent years. These models are used to predict the risk of groundwater contamination for various land uses. This paper presents an assessment of uncertainties and sensitivities to input parameters for a regional model. The model had been set up to improve and facilitate the decision-making process between stakeholders and in a groundwater quality conflict. The stochastic uncertainty and sensitivity analysis comprised a Monte Carlo simulation technique in combination with a Latin hypercube sampling procedure. The uncertainty of the calculated concentrations of nitrate leached into groundwater was assessed for the various combinations of land use, soil type, and depth of the groundwater table in a vulnerable, sandy region in The Netherlands. The uncertainties in the shallow groundwater were used to assess the uncertainty of the nitrate concentration in the abstracted groundwater. The confidence intervals of the calculated nitrate concentrations in shallow groundwater for agricultural land use functions did not overlap with those of non-agricultural land use such as nature, indicating significantly different nitrate leaching in these areas. The model results were sensitive for almost all input parameters analyzed. However, the NSS is considered pretty robust because no shifts in uncertainty between factors occurred between factors towards systematic changes in fertilizer and manure inputs of the scenarios. In view of these results, there is no need to collect more data to allow science based decision-making in this planning process. [Copyright &y& Elsevier]

Summary: The homogeneity of the flood frequency regime for a given pooling-group of sites is a fundamental assumption for many regional flood frequency analysis techniques. Assessing regional homogeneity is a critical step, which may be complicated by the presence of cross-correlation among flood sequences. The scientific literature proposes a number of statistical homogeneity tests and documents that inter-site correlation of floods is normally not negligible, but does not specifically address the impact of cross-correlation on such statistical tests. This paper analyzes the effectiveness of a well-known homogeneity test proposed in the scientific literature in the presence of inter-site cross-correlation through a series of Monte Carlo experiments. The numerical experiments enable us to comment on a possible theoretical correction for the test and to identify an empirical tool that accounts for the impact of inter-site cross-correlation of floods. [Copyright &y& Elsevier]

Summary: Earlier research results have shown that the range of the ensemble flow simulations from a well-validated operational hydrologic model possesses a well-defined log-linear relationship to catchment area for which the ensembles are generated. The ensemble range is a function of both parametric and rainfall input uncertainty. In this paper the sensitivity of this scaling relationship for simulation uncertainty to the numerical discretization scale of the distributed model is quantified when the model is driven by WSR-88D (NEXRAD) radar data and for two catchments in the south-central United States. The distributed model used is subcatchment (rather than pixel) based, has components that are based on operational models employed in the US National Weather Service, and has been validated for the two application catchments of this work as part of the Distributed Model Intercomparison Project (DMIP). The approach taken develops a new parsimonious model for spatially correlated radar-rainfall errors with radar-pixel error variance that depends on the magnitude of the observed rainfall. Monte Carlo methods are used to translate parametric and radar-rainfall input uncertainty to ensemble flow simulations at a number of subcatchments of varying size. Selected events in the period from May 1993 to July 1999 were analyzed. The results confirm that the ensemble flow range is dependent on subcatchment scale with a well-defined log-linear relationship for both application catchments. This finding holds for all uncertainty scenarios examined and across all the scales of distributed model discretization. It is further found that a higher model discretization leads to a shorter range of simulation uncertainty for subcatchment spatial scales that are substantially larger than the average subcatchment size. Flow simulation uncertainty that is due to either parametric or input uncertainty alone scale in a way that is similar to the combined uncertainty scaling. [Copyright &y& Elsevier]

Abstract: Hydrological models generally contain parameters that cannot be measured directly, but can only be meaningfully inferred by calibration against a historical record of input–output data. While considerable progress has been made in the development and application of automatic procedures for model calibration, such methods have received criticism for their lack of rigor in treating uncertainty in the parameter estimates. In this paper, we apply the recently developed Shuffled Complex Evolution Metropolis algorithm (SCEM-UA) to stochastic calibration of the parameters in the Sacramento Soil Moisture Accounting model (SAC-SMA) model using historical data from the Leaf River in Mississippi. The SCEM-UA algorithm is a Markov Chain Monte Carlo sampler that provides an estimate of the most likely parameter set and underlying posterior distribution within a single optimization run. In particular, we explore the relationship between the length and variability of the streamflow data and the Bayesian uncertainty associated with the SAC-SMA model parameters and compare SCEM-UA derived parameter values with those obtained using deterministic SCE-UA calibrations. Most significantly, for the Leaf River catchments under study our results demonstrate that most of the 13 SAC-SMA parameters are well identified by calibration to daily streamflow data suggesting that this data contains more information than has previously been reported in the literature. [Copyright &y& Elsevier]

Uniqueness of the problem surrounding non-detects has been a concern alike to researchers and statisticians dealing with summary statistics while analyzing censored data. To incorporate non-detects in the estimation process, a simple substitution by the MDL (method detection limit) and the maximum likelihood estimation method are routinely implemented as standard methods by US-EPA laboratories.In situations where numerical standards are set at or near the MDL by regulatory agencies, it is prudent and important to closely investigate both the variability in test measurements and the estimation bias, because an inference based on biased estimates could entail significant liabilities. Variability is understood to be not only inevitable but also an inherent and integral part of any chemical analysis or test. In situations where regulatory agencies fail to account for the inherently present variability of test measurements, there is a need for regulated facilities to seek remedial action merely as a consequence of inadequate statistical procedure.This paper utilizes a mathematical approach to derive the bias functions and resulting bias curves are developed to investigate the censored samples from a variety of probability distributions such as normal, log-normal, gamma, and Gumbel distributions. Finally, the bias functions and bias curves are also compared to the results obtained by using Monte Carlo simulations. [Copyright &y& Elsevier]

In this paper, the sensitivities of stream–aquifer seepage flow and the state of connection or disconnection between stream and aquifer, to spatial variability in the aquifer-saturated hydraulic conductivity, are studied with the variably saturated modelling programme, SWMS_2D, using a Monte Carlo technique. A simple stream–aquifer flow system is used with a rectangular stream, with a clogging layer on its perimeter, partially penetrating an alluvial aquifer. A 2D nearest-neighbour method is used to generate fields of log normally distributed saturated conductivity for the alluvial aquifer for a specific set of degree of heterogeneity and the nearest-neighbour autoregressive parameters. Seven different cases are studied. Five of these investigate the effect of the nearest-neighbour parameters, and three cases investigate the effect of aquifer water table level. The results show that the nearest-neighbour parameters and the aquifer degree of heterogeneity significantly affect the modelling of stream seepage, pressure heads and the state of connection/disconnection between stream and aquifer, particularly when the water table is some distance below the streambed. Moreover, comparisons of these results with a purely deterministic model show good agreement when the stream and the aquifer are connected or disconnected but with a high water table. However, there are significant differences between the deterministic and stochastic models when the water table is well below the streambed. [Copyright &y& Elsevier]

In this paper, we investigate the importance of unsealed roads as suspended sediment sources in a predominantly agricultural catchment within a temperate zone in south-eastern Australia. A selection of sediment tracers and an improved multivariate numerical mixing model are used to quantify the source composition of suspended sediment samples collected during six high-frequency rainfall events. Relative contributions from gravel-surfaced roads, grouped lands (un-gravelled roads, pasturelands and cultivated lands on basalt-derived soils), cultivated lands on granite-derived soils, and forest to sediments in the falling limbs of event hydrographs were 0.41±0.17, 0.18±0.13 0.13±0.11 and 0.14±0.07, respectively. The relative contributions during a peak discharge event were 0.52±0.12, 0.30±0.17, 0.15±0.11 and 0.17±0.08, respectively. Relative contributions per unit area from the gravel-surfaced roads, grouped lands and cultivated lands on granite soils were as high as 500-, 100- and 10-times, respectively, the areal contributions from the forest. These results indicate that unsealed roads are moe important suspended sediment sources than agricultural lands for the suspended sediment samples investigated in the study catchment. The contributions from the gravel-surfaced roads increased with increasing sediment particle size class. Consequently, we believe that considerable emphasis should be placed on unsealed roads, when sediment control measures are planned for agricultural catchments. [Copyright &y& Elsevier]

When designing hydraulic structures, civil engineers have to evaluate design floods, i.e. events generally much rarer that the ones that have already been systematically recorded. To extrapolate towards extreme value events, taking advantage of further information such as historical data, has been an early concern among hydrologists. Most methods described in the hydrological literature are designed from a frequentist interpretation of probabilities, although such probabilities are commonly interpreted as subjective decisional bets by the end user. This paper adopts a Bayesian setting to deal with the classical Poisson–Pareto peak over treshold (POT) model when a sample of historical data is available. Direct probalistic statements can be made about the unknown parameters, thus improving communication with decision makers. On the Garonne case study, we point out that twelve historical events, however imprecise they might be, greatly reduce uncertainty. The 90% credible interval for the 1000 year flood becomes 40% smaller when taking into account historical data. Any kind of uncertainty (model uncertainty, imprecise range for historical events, missing data) can be incorporated into the decision analysis. Tractable and versatile data augmentation algorithms are implemented by Monte Carlo Markov Chain tools. Advantage is taken from a semi-conjugate prior, flexible enough to elicit expert knowledge about extreme behavior of the river flows. The data augmentation algorithm allows to deal with imprecise historical data in the POT model. A direct hydrological meaning is given to the latent variables, which are the Bayesian keytool to model unobserved past floods in the historical series. [Copyright &y& Elsevier]

Scientific evidence indicates that a global climate change due to human activity is possible within a century. It is expected that doubling of CO2 and increasing the amount of other greenhouse gases in the atmosphere would result in more severe weather, among other consequences. The purpose of the paper is to examine how flood frequencies and magnitudes in a mid-size subalpine watershed on the eastern slopes of the Rocky Mountains in Alberta, Canada, would change under 2×CO2 conditions. Given the poor spatial resolution of present general circulation models (GCM) and their uncertain performance at the regional scale, a first-order analysis is carried on, in which only rainfall intensity changes are considered to have the most significant impact on future floods. Estimates of storm rainfall increases are based on the literature survey, GCM projections for the study area, and transposition of southern climatic conditions. Two scenarios of likely most severe changes were selected: first, a 25% increase in the mean and standard deviation of Gumbel distribution of rainfall depth for storm durations from 6 to 48 h; second, a 50% increase in the standard deviation only. The HEC-1 watershed model and the soil conservation service runoff curve method for abstractions were used in Monte Carlo simulation. Comparison of Monte Carlo derived flood frequency curves for the two scenarios with the present day curve shows that scenario 1 is more critical in terms of flood flow increases than scenario 2. Under scenario 1, the mean annual flood on the study watershed would increase by almost 80% and the 100-year flood would increase by 41%. [Copyright &y& Elsevier]

• A probabilistic methodology that forecast the future storm overflows has been developed. • The methodology considers the precipitation events and the impervious area. • The precipitations simulations were carried out by Monte Carlo simulations. • A logistic regression simulates the overflow as result of precipitations on impervious area. This paper presents a probabilistic methodology that allows the study of the interactions between changes in rainfall dynamics and impervious areas in urban catchment on a long- and short-term basis. The proposed probabilistic model predict future storm overflows while taking into account the dynamics of changes in impervious areas and rainfall. In this model, a logistic regression method was used to simulate overflow resulting from precipitation events based on average rainfall intensity and impervious area. The adopted approach is universal (as it can be used in other urban catchments) and is a significant simplification of classic solutions; a hydrodynamic model is used to analyse the operation of the overflow. For the rainfall simulations, a rainfall generator based on the Monte Carlo method was used. In this method, a modification that allows the simulation of changes taking place in rainfall dynamics, including the effects of climate change, was introduced. This method provides the opportunity to expand and modify probabilistic models in which outflow from the catchment is modelled to predict the functioning of reservoirs and to design sewer networks that have the ability to deal with future rainfall dynamics, including moderate, strong, and violent downpours according to the Sumner scale. To verify the simulation results with a probabilistic model, an innovative concept using a hydrodynamic model was considered. This verification considers the changes in the impervious area in the period covered by the simulations and is limited using standard calculation procedures. In practice, the model presented in this work creates opportunities for defining the concept of sustainable development in urban catchments while taking into account the factors mentioned above. From the perspective of landscaping, this is important because it creates the opportunity to limit the impacts of climate change and area urbanization on the receiving waters. [ABSTRACT FROM AUTHOR]