Journal of Hydrologic Engineering

Extreme flood simulation with synthetic extreme precipitation events raises unavoidable questions about the choice of initial conditions. State-of-the-art extreme flood estimation frameworks propose to address these questions with the help of semicontinuous modeling and reanalysis of simulated state variables. In this context, the present work proposes a new method for the selection of initial conditions for extreme flood simulation. The method is based on generating sets of initial conditions from the matrix of state variables corresponding to a long simulation run of the selected hydrological model. Two sets of initial conditions are obtained: a deterministic set composed of selected state variable quantiles and a stochastic set composed of state variable vectors randomly drawn from the complete state variable matrix. The extreme flood simulations corresponding to both sets are compared in detail, and the stochastic simulations are used in a sensitivity analysis to identify the dominant state variables and possible interactions. The aim hereby is to provide a tool to analyze the role of initial conditions and the importance to account for state variable interactions in extreme flood estimation. The proposed method is applied to probable maximum flood estimation for the Swiss Mattmark Dam catchment with a semilumped hydrological model. The obtained results for this case study show that for high flood peak quantiles, the initial soil saturation is dominating other state variables, and deterministic initial conditions are sufficient to generate extreme floods.

A proposed quantification of the fundamental concepts in the Natural Resources Conservation Service (NRCS) rainfall-runoff relation is examined to determine changes relevant to peak discharge estimation and drainage design. Changes to the NRCS curve number, storage, and initial abstraction relations result in different estimates of the runoff volume, timing of runoff, and peak discharge produced by long-established empirical equations that quantify event-based runoff. The ratio of existing-to-proposed runoff depth estimates is determined across a range of curve numbers and storm depths. Similarly, the ratio of existing to proposed peak discharge is determined as a function of curve number, storm depth, and design storm within the NRCS TR-55 model. Using an infrastructure design based on the existing equations as the baseline, that infrastructure is underdesigned (overdesigned) if the proposed NRCS equations produce larger (smaller) estimates of volumes or discharges than the existing equations for the same watershed conditions. This study shows that the proposed NRCS equations more likely identify existing infrastructure as being underdesigned for smaller storm events and lower curve numbers. Conditions and design storm produce large variation in the design ratios. Storms with return periods on the order of 2–10 years generally result in peak discharge underdesign ratios that span from 1 to as high as 2 or 3 depending on the watershed conditions, storm size, curve number, and design element in question. Existing infrastructure overdesign is more likely the case when return periods approach 25, 50, or 100 years, with typical overdesign ratios ranging downward from 1.0 to as small as 0.8. In most cases, required storage volumes, as prescribed by the NRCS TR-55 approach, are found to be smaller using the proposed equations, leading to both cost savings and reduced pond residence times. Three common design problems are provided only to illustrate typical underdesign and overdesign findings. The central focus is on the potentially negative consequences of proposed changes to a limited aspect of curve number hydrology.

Encountering uncertainty in engineering design problems is inevitable. Probability theory is a traditional method for uncertainty analysis despite the nature of uncertainties (i.e., aleatory or epistemic). Because of some limitations of probability theory in analyzing epistemic uncertainty, new methods such as evidence theory have been developed. This paper describes the analysis of uncertainty by means of both probability and evidence theories. Monte Carlo simulation is used along with the two theories to propagate the uncertainty. A computational strategy based on random sampling is used to analyze and represent the epistemic and aleatory uncertainty. Despite its advantage, the application of evidence theory in civil engineering is limited because of its mathematical complexities. This paper attempts to make this theory more applicable by using practical design problems, as well as by highlighting its advantages compared with probability theory. Three hypothetical flood-control-related problems are considered. The uncertainty of a design storm intensity resulting from an experimental equation and a design discharge resulting from the rational method are analyzed assuming storm intensity as a stochastic variable with Level 2 (i.e., both aleatory and epistemic) uncertainties. The design diameter of the pipes in a stormwater collection system is estimated. It is assumed that the input parameters have only epistemic uncertainty. The results show that if the desired risk changes, some diameters may stay fixed and some may change. The design of a flood control levee is considered. The levee height is an output variable, and it is assumed that the design discharge has Level 2 uncertainty, whereas all other inputs have epistemic uncertainty. The results of uncertainty analysis with the two theories are represented appropriately, discussed, and compared. Compared with probability theory, evidence theory offers more insight for decision makers when they encounter epistemic and/ or aleatory uncertainties in design problems.

The first-order second-moment approximation (FOSM) has been proposed to estimate the annual runoff ($Q$) distribution from the statistical information of annual precipitation ($P$), annual potential evapotranspiration ($Ep$), and interannual water storage change ($\mathrm{\Delta }S$). However, FOSM can only estimate two statistical parameters for theoretical distributions, thus is unable to estimate the annual runoff distribution that is fitted by three-parameter probability distribution. Hence, this paper extends FOSM to second-order third-moment approximation (SOTM). This study first establishes an annual rainfall-runoff model. Then the moments of $Q$ are derived from the data of $P$, $Ep$, and $\mathrm{\Delta }S$ in the annual rainfall-runoff model by FOSM and SOTM. Because the observed data of $\mathrm{\Delta }S$ are difficult to obtain, $\mathrm{\Delta }S$ is simulated by two monthly hydrological models. Both FOSM and SOTM are applied to the Gan River basin (GRB) and the Tongtian River basin (TRB). The results show that the determination coefficients $R_{QQ}^2$ of curve fitting by FOSM and SOTM are both larger than 0.80, indicating their capability to estimate the annual runoff distribution. The availability of $\mathrm{\Delta }S$ has a certain impact on the performance of FOSM and SOTM, particularly in the TRB, where $R_{QQ}^2$ with the use of $\mathrm{\Delta }S$ information can increase by 2–4% compared with that without using $\mathrm{\Delta }S$. In practice, however, the impact of $\mathrm{\Delta }S$ can be neglected in the ungauged basins. It is also found that although the $R_{QQ}^2$ values estimated by SOTM for both Pearson Type III and three-parameter lognormal distributions are not greater than the $R_{QQ}^2$ values of gamma distribution estimated by FOSM, SOTM is capable of presenting more statistical information with the $R_{QQ}^2$ values all larger than 0.80.

This study demonstrates the influence of climate change-induced sea level rise on multiobjective saltwater intrusion management strategies in coastal aquifers. Three metamodels were developed from the solutions of a numerical simulation model of the coupled flow and salt transport processes in a coastal aquifer system. Results revealed that the proposed metamodels are capable of predicting density-dependent coupled flow and salt transport patterns quite accurately. Based on a comparison of the three methods, the best metamodel was selected as a computationally inexpensive substitute for the simulation model in the coupled simulation-optimization-based saltwater intrusion management model. To achieve computational efficiency, the optimization routine of the proposed management model was performed on a parallel computing platform. The performance of the proposed methodology was evaluated for an illustrative multilayered coastal aquifer system in which the effect of climate change-induced sea level rise was incorporated for the specified management period. Results show that the proposed saltwater intrusion management model provides acceptable, accurate, and reliable solutions while significantly improving computational efficiency in a coupled simulation-optimization methodology.

Water resources management requires an integrated approach between sustainability and catchment management to provide an enhanced understanding of usage options and environmental impacts. Discharge contributes significantly to the lake water and hence its accurate assessment would help in designing the management policies. This study simulates river discharge contributing to a lake using different physical and data-based modeling approaches. Models such as support vector machine (SVM); extreme learning machine (ELM); MIKE SHE; and a combination of MIKE SHE and the Soil and Water Assessment Tool (SWAT), the hybrid MIKE SHE-SWAT model, are used to simulate discharge for three subcatchments of Loktak Lake. Discharge simulation at different sections of a river using physically based models is cumbersome because it requires data such as river geometry, hydrologic time series, channel roughness coefficients, and hydraulics of existing control structures. In developing countries such as India, quality and quantity of data become the limitation because of poor monitoring and record keeping. Therefore, a data-based modeling approach can also be used as an alternative tool in hydrologic simulation studies. The selected methods are evaluated using statistical indicators, and results show that both physical and data-based methods satisfactorily simulate the observed hydrograph at the outlet of each subcatchments. This study concludes that using the hybrid MIKE SHE-SWAT model simulates the river discharge with the highest Nash-Sutcliffe model efficiency (NSE), 0.854 for the Nambul subcatchment, 0.890 for the Iril subcatchment, and 0.845 for the Thoubal subcatchment. Subsequently, hydrological models for the other six subcatchments are developed using the hybrid MIKE SHE-SWAT model. Furthermore, water balance components of the Loktak Lake are also analyzed, which suggest that of the total inflows to the lake, discharge from the nine subcatchments accounts for about 91%, and the remainder is contributed by direct rainfall onto the lake area. Among outflows, Ithai barrage (69%) accounts for the largest fraction, whereas abstractions for hydropower generation, evaporation from the lake surface not covered by phumdis, evapotranspiration from the area of phumdis on the lake, irrigation, and domestic consumption account for 22, 2, 5, 1, and 1%, respectively.

The American River basin, a snow-dominated alpine watershed in Northern California, was simulated using the Soil and Water Assessment Tool (SWAT) model to evaluate change in future streamflow characteristics in response to potential climate change. Seven global climate model (GCM) outputs from Phase Five of the Coupled Model Intercomparison Project (CMIP5) were used to generate 50-year (2015–2064) future climate scenarios. GCM outputs predict an increase in winter precipitation in the watershed and the maximum and minimum daily temperatures to rise by more than 1°C in the fall and spring. Compared to the baseline scenario (1964–2014), the results showed that increased winter air temperature reduced the amount of snowfall, which eventually decreased the basin average snowmelt runoff by more than 70%. With the increase in winter rainfall, the mean annual streamflow in the basin increases by 6.7%, but peak streamflow shifts from mid-March to early February. Winter flood frequency as well as summer extreme low flow frequency are projected to increase. Therefore, future challenges in the watershed will be to balance the seasonal effects of climate change.

The likelihood of catastrophic floods in the Mississippi River correlates with a direct measure of variability, the magnitude of the day-to-day variations in river stage. At several sites along the Middle Mississippi River, the annual average daily stage change has increased twofold, from $\sim 0.12\mathrm{m}/\mathrm{day}$ in the mid to late 1800s to $0.2–0.25\mathrm{m}/\mathrm{day}$ at the present time; this increase is much higher than the increases we document at most sites on the Upper and Lower Mississippi River. Recent record floods have occurred in the highly channelized Middle Mississippi River where channel constriction is most extreme and the daily stage changes and annual amplitude of stage variations are largest. The long-term difference in the magnitude of daily stage change roughly correlates with the reduction in width of the river channel that was done to facilitate navigation, an effect that is consistent with theoretical predictions. Daily stage changes are well constrained compared to catastrophic floods, which are rare and have poorly defined frequency.

This paper studied the temporal and spatial variation of groundwater levels that characterized the evolution of the groundwater flow field in Xi’an, China, from 1965 to 2013, by using the wavelet analysis method. Then under the consideration of different hydrogeological conditions on a large scale, the hydrogeological areas were taken as research units, and the correlation between groundwater level and its driving forces were analyzed. Finally, the projection pursuit model (PPM) based on the particle swarm optimization (PSO) algorithm was adopted to identify the contributions of the driving forces in each hydrogeological unit. The results indicated that in the background of climate change and overexploitation in the last few decades, the groundwater level in Xi’an dropped sharply. There was a strong correlation between precipitation and groundwater level, especially in the flood seasons. In different hydrogeological areas, the sensitivities of groundwater level to precipitation were different. In addition, the groundwater level shows an exponential-function declining trend with the increasing exploitation, whereas in the groundwater sources, there was a negative linear correlation between groundwater level and exploitation. Likewise, the streamflow had an effect on the variation of groundwater level. The relative contributions of driving forces varied with the hydrogeological conditions. In conclusion, exploitation was the leading factor for the variation of the groundwater level, precipitation came second, and streamflow had an influence on groundwater levels in the regions adjacent to the rivers. The results indicated that in human-disturbed environments, hydrogeological conditions still play an important role in groundwater dynamic variation. Additionally, they suggested that the exploitation of groundwater should be adjusted to make it consistent with climate change; thus, the abnormal variation of groundwater flow field will be contained.