Journal of Hydrologic Engineering

Semianalytic formulas are obtained for flow to a well screened in a leaky aquifer that was overlain by an aquitard and phreatic aquifer and underlain by a second aquitard and leaky aquifer. Formulas are also obtained for the scenario in which the phreatic aquifer is hydraulically connected to a rectilinear stream. Adaptive mesh refinement is used to obtain highly resolved simulations for a meandering stream depletion problem that is both numerically challenging and has physically interesting features. The associated inverse problem is also considered, and a simple example is presented of synthetic pumping test data, which demonstrates that it can be very difficult to quantify the actual effects of pumping. The conclusion is that a framework for uncertainty quantification of aquifer parameters is necessary for objective determination of the parameters.

Using 92-year precipitation data from 400 rain gauge stations located in 48 states throughout the continental United States, the change in precipitation in space and time was investigated. The variability in precipitation was investigated using an entropy-based approach. The Mann-Kendall, Spearman’s rho, and Sen slope tests were applied to assess the existence of a trend in precipitation. Analysis showed that annual precipitation exhibited lower temporal variability than did its constituent seasonal series. The fall precipitation had the highest variability, and the spring precipitation the lowest. Not many stations exhibited trends, and they are concentrated in limited areas of the United States. The greatest part of those exhibiting statistically significant trends showed increasing trends both in precipitation and rainy days. Only in a few cases negative trends were found. Finally, analysis showed that no relationship can be established between trend and variability in precipitation.

A systematic and holistic watershed model comparison and selection process, integrating a full range of relevant criteria, is illustrated here. The process consists of screening for a set of candidate models on the basis of prerequisite model attributes; assessing hydrologic simulation performance using various conventional statistical metrics; assessing operational logistics performance, reflecting somewhat subjective but centrally important issues around relative feasibility and suitability of candidate models in the intended context of use; and integrating the hydrologic and operational performance results, which are each evaluated using a weighted-matrix approach, into a single coherent and comprehensive ranking system. The process was applied to evaluation of watershed models for operational hydroelectric inflow forecasting in British Columbia, Canada. An important feature of the study was its horse-race project management approach, involving a supervised competition between expert teams using different models but the same data sets. Four models passed the prescreening phase, and these were evaluated for three test basins representing contrasting geophysical regimes. The model selection process presented here is believed to be potentially very broadly useful.

In this paper, a revised advection-aridity evaporation model is presented to estimate actual evaporation over short time periods. The revised model is applied and validated against an existing data set. The time series comparisons in a soybean site and a corn site reveal that this revised model shows good performance in characterizing the diurnal variation in evapotranspiration. The coefficient of determination for these two sites is 0.92 and 0.87, respectively. To demonstrate the improvement of this model, comparisons between the authors’ newly developed model and the other two revised models are carried out. The quantitative measures show that this revised model represents an obvious improvement compared with the other two models. For all of the sites, the mean absolute difference is less than $65\mathrm{W}/{\mathrm{m}}^2$, the root-mean-square difference is less than $85\mathrm{W}/{\mathrm{m}}^2$, and the coefficient of determination is greater than 0.7. Overall, results from this analysis indicate significant improvements for actual evapotranspiration estimation with this newly developed method.

Roughness-induced wind flow is commonly encountered around water bodies. It is well known that wind velocity enhances evaporation. However, the way this enhancement takes place in roughness-induced flow has been less explored. In this paper, a set of 200 data of an experimental investigation of the evaporation rates of water under forced convection conditions created by roughness elements from a circular pan fitted in the bottom wall of a wind tunnel test section are analyzed. The range of parameters used in this work are as follows: pan water temperature 30–50°C, wind velocity $4.68–17.52\mathrm{m}/\mathrm{s}$, and relative roughness height 0.011–0.075. Evaporation rates are related as a function of the vapor pressure differential, wind velocity, and relative roughness height of the obstructions introduced in the upstream of the air flow. It is found that evaporation rates are a nonlinear function of the relative roughness height, and evaporation estimates can contain a large error if such roughness-induced effects are neglected.

This study investigates the shape characteristics of hydrograph components of the Wu-Tu watershed in Taiwan based on observations of rainfall and streamflow. Component hydrographs were modeled using a model of three serial tanks with one parallel tank. The block kriging method was used to calculate the hourly mean rainfall of events, and eight model parameters of 34 cases were derived from the shuffled complex evolution optimal algorithm. The remaining 18 events were used to verify the applicability of the calibrated parameters. Results show that (1) times to peak of hydrograph components are positively nonlinearly correlated to peak time of rainfall; (2) peak discharges of hydrograph components are linearly proportional to those of streamflow hydrograph; and (3) relationships of total discharges also have direct ratios between hydrograph components and observed streamflow. Using the procedures proposed in this study, three evaluated shape characteristics of component hydrographs can be easily used to rapidly determine shapes of simple hydrographs.

The eco-hydrological requirements of a river are related to the health of the river. In recent years, changing climate and intensifying human activities have further aggravated water problems in the Yellow River. Considering hydrological changes, this study addressed the ecological water requirements of the Yellow River basin, wherein the threshold ecological water requirement corresponded to the streamflow with the largest occurrence probability determined by the Pearson Type III distribution. Instream flow ± one standard deviation was adopted as the water resource management target and abrupt changes in streamflow were detected by using the Mann-Whitney $U$ test. The results of the analysis indicated that: (1) the annual flow of the Yellow River substantially decreased after abrupt changes and the probability significantly dropped that the amount of annual streamflow satisfied the ecological instream flow requirement; and (2) the water resource management target was satisfied well after th echange point only at the Tangnaihai station, located in the upper Yellow River basin. From the upper to the lower Yellow River basins, the probability has dramatically decreased that the ecological water requirement target is satisfied. Therefore, the Yellow River basin will likely face increasingly serious water problems, particularly in the middle and lower parts.

The increase in runoff from urbanization is well known; one extreme example comes from a 13-ha residential neighborhood in southeast Arizona where runoff was 26 times greater than in an adjacent grassland watershed over a 40-month period from 2005 to 2008. Rainfall-runoff modeling using the newly described KINEROS2 urban element, which simulates a contiguous row of houses and the adjoining street as a series of pervious and impervious overland flow planes, combined with tension infiltrometer measurements of saturated hydraulic conductivity ($K_s$), indicate that $17\pm 14\%$ of this increase in runoff is due to a 53% decrease in $K_s$ in constructed pervious areas as compared to the undeveloped grassland. $K_s$ in the urban watershed identified from calibrating the rainfall-runoff model to measured runoff is higher than measured $K_s$ but much lower than indicated by a soil texture–based KINEROS2 parameter look-up table. Tests using different levels of discretization found that watershed geometry could be represented in a simplified manner, although more detailed discretization led to better model performance.

The current study was undertaken to analyze the performance of three back-propagation training algorithms of artificial neural network (ANN) along with a multiple linear regression model (MLR) for transient simulation of short/midterm soil temperature (TS). Each ANN used a different type of learning algorithm (gradient descent, conjugate gradient, and Levenberg-Marquardt) as methods for daily and weekly TS prediction. The analysis was performed as a case study using three meteorological parameters [air temperature (TA), net radiation (NR), and relative humidity (RH)] and two hydrological variables [precipitation (P) and runoff (Q)] as input data for a region in Wyoming. Pearson and cross correlation analyses were applied to investigate the relationships between input and target values to determine several input combinations. The correlation analysis indicated that the meteorological parameters of TA and NR were reasonably correlated with TS. Based on the input combinations, 14 models were constructed for each of the daily/weekly MLR and ANNs. The accuracy of the predictions was evaluated by the RMS error (RMSE), mean absolute error (MAE), and Nash-Sutcliffe coefficients between the measured and predicted TS values. The Levenberg-Marquardt ANN method was found to provide a more accurate prediction than the other two types of ANNs and the MLR model. Furthermore, it was found that weekly structure of ANNs performance surpasses that of any other daily structures. Although results illustrated that the soil temperature is a function of meteorological variables, the measured TS is also shown to be positively influenced by hydrological parameters—a fact that has not been pointed out by correlation analyses.

Large-scale distributed hydrological modeling is nowadays more widely applied for water resource management. To ensure the spatial consistency of distributed parameters, an alternative calibration strategy is suggested: the global calibration. This approach aims to identify a single parameter set suitable for every stations located within the modeled domain. A global calibration procedure is applied and tested over a large portion of the St. Lawrence River basin ($\mathrm{387,000}{\mathrm{km}}^2$) using Hydrotel, a physically based semidistributed hydrological model. Despite limitations in estimating high flows on small catchments, global calibration is regarded as a fair trade-off between local performance and regional consistency of parametric information. In the near future, the global calibration strategy will act as a key methodological framework for large-scale hydrological assessments.

The critical flow concept has many applications in open-channel flow problems. Determination of critical depths for canals and natural streams with different shapes is a classic task, but important for efficient hydraulic design. Generally, critical depth is used to classify the flow into subcritical and supercritical and may be used as a control point for computing water surface profiles in steady and unsteady gradually varied flows. For a given compound channel, the kinetic energy correction coefficient varies quite rapidly above the overbank level. In such a case, specific energy may have more than local minimum or maximum (multiple critical depth) for some combination of discharge and geometry. The compound channels are well documented in the literature, but occurrence condition of the multiple critical depth has not yet been determined. The first step in modeling critical depth is to determine whether multiple occurrences exist. In current research, symmetrical prismatic channels with two different side slopes (two-stage cross sections) are considered, and the condition for which multiple critical depths exist is mathematically determined. For this, an analytical procedure is used to determine limiting side slope for the upper portion of the cross section. If the side slope of the upper portion is milder than the limiting side slope, the multiple critical depths will occur in the two-stage cross section.

This study proposed a methodology for quantifying the flood runoff reduction effect of storage facilities by curve number (CN). The runoff hydrographs for various rainfall durations and return periods were derived using the modified rational method. The effect of runoff reduction due to the introduction of storage facilities was then quantified by CN. The result derived was summarized in a graph showing the decrease in CN as an effect of the runoff reduction with respect to the rainfall duration, return period, runoff coefficient of the catchment, and size of the storage facility. The runoff reduction effect of a storage facility was able to be easily and directly estimated using the derived graph. The result derived was also found to be well matched with the previous study for three different sites.