Journal of Hydrologic Engineering

Detecting trends in hydrometerological data through the commonly used Mann-Kendall test is misleading in the presence of data autocorrelation. Autocorrelation seriously interferes with type I errors and power of trend detection. To mitigate this effect, the authors introduce a variance correction prewhitening method. It addresses two important issues that lacked appropriate attention in the past application of trend-free prewhitening method: inflationary variance of slope estimator and deflationary serial variance. After serial and slope variances correction, the new method keeps a better balance between maintaining a low type I error and a relatively strong power of trend detection. In comparison, other methods for the same purpose only address one of these two characteristics. The new method bears some resemblance to the block-bootstrap method; however, it is superior in its simplicity for implementation. Case studies reveal that uncertainties arising from autocorrelation are substantial. Applying more than one test is helpful to interpret results with uncertainties information. The new method provides a robust choice to this strategy.

A sediment transport module added to the Hydrologic Engineering Center’s Hydrologic Modeling System (HEC–HMS) Version 4.0 computes surface erosion, reach routing, and reservoir trap efficiency. The HEC–HMS sediment transport module was applied to the Upper North Bosque River Watershed (UNBRW) in central Texas, a well-instrumented watershed with good parameterization and calibration data, to test model robustness. The UNBRW is 98% rural including rangeland, forage, and dairy waste application fields. Simulation results were compared to observed average daily total suspended solid (TSS) data at five gauge locations. The HEC–HMS results matched observed TSS across the UNBRW ($<1\%$ error at all gauges) during model calibration and maintained modest residuals ($-31$ to 12% error) during the validation period. The HEC–HMS also performed well compared to published results through two other commonly used models.

In this study, neuro-fuzzy-based group method of data handling (NF-GMDH) as an adaptive learning network is used to predict the flow discharge in straight compound channels. The NF-GMDH network is developed by using the particle swarm optimization (PSO) and gravitational search algorithm (GSA). The depth ratio (ratio of water depth in floodplain to that in main channel), coherence parameter, and the discharge ratio [ratio of flow discharge calculated from vertical divided channel method (VDCM) to the bank full discharge] are considered as input parameters to represent a functional relationship between input and output parameters. The performances of training and testing stages for NF-GMDH models were quantified in terms of statistical error parameters. Also, the results of performances were compared with those obtained by using linear genetic programming, nonlinear regression methods, and VDCM. Evaluation of the proposed model demonstrated that NF-GMDH-GSA network provides a more accurate prediction than the NF-GMDH-PSO network. Finally, statistical error parameters indicated that the NF-GMDH networks as a new soft-computing tool produced better prediction of flow discharge in comparison with linear genetic programming, nonlinear regression methods, and VDCM.

The design of urban drainage systems and the development of flood mitigation strategies commonly require knowledge of extreme rainfall for a given return period. In the context of climate variability that is no longer imputable to only natural forces, the estimation of this variable represents a critical issue. Climate change effects, particularly on precipitation, should be considered in the design and planning of hydraulics infrastructures. The aim of this study is to provide an assessment of the effects of statistically significant trends in extreme rainfall on the rainfall depth-duration-frequency (DDF) curves for the return periods typically used in design in order to obtain useful information for the definition of DDF curves in climate change scenarios. The study was applied in Sicily, in the center of the Mediterranean Area, because the study may provide interesting insights in a highly populated coastal region. The first step of the study consisted of the detection and quantification of trends in the annual maximum rainfall series of different durations (1, 3, 6, 12, and 24 h) observed in Sicily. The annual maximum rainfall series recorded using 65 rain gauges over the period 1950–2008 has been selected and analyzed. For each duration, the moving averages have been computed and then the Mann-Kendall test has been applied, considering different levels of confidence. The results showed that, for all the durations, increasing and decreasing trends occurred over the examined period. Once trends have been detected and quantified, the generalized extreme value distribution (GEV) has been employed to compute extreme rainfall with return periods equal to 5, 10, and 20 years. Finally, the parameters of the DDF curves have been updated for the Sicilian rain gauges, incorporating the information obtained from the trend analysis carried out on the local scale. The updating of the DDF parameters has been carried out by defining two climate scenarios on the basis of the trend analysis results. For each duration, results showed that, in the area of study, extreme rainfall series have been affected by both negative and positive statistically significant trends.

Overcoming the hydrologic shortcomings of conventional approaches to stormwater management requires the protection or restoration of flow regimes at small scales. A better understanding of how stormwater management strategies can achieve this aim is needed. This study modeled 28,800 design configurations of a typical stormwater management strategy at the scale of urban land parcels across a range of urban densities and climatic conditions. Realistic design configurations that achieved three hydrologic response targets were identified as part of this modeling. It was found that meeting the targets required a combination of stormwater harvesting (using tanks) and infiltration (using rain gardens). This was possible primarily because the amount of harvested impervious roof runoff made a large contribution to a hydrologic target, which measured the ability to restore volumetric losses. Management of flow regimes at small scales will require policy mechanisms that necessitate both stormwater harvesting and infiltration. Urban design challenges remain to ensure that such approaches can be incorporated into the urban landscape in a way that maximizes the benefits to humans and to the environment. Future research is also required to investigate how the use of small-scale stormwater management strategies can improve catchment-scale flow regimes.

Given the risk of intensive storms (the probability of exceeding certain storm intensity one or more times within the project life) of central Alberta is expected to change in the future, a new risk chart is proposed which represents the nonlinear relationship between storm intensity, design project life, and the risk of intensive storms being exceeded within the project life. First, a comparison between estimated risk charts of the past (1914–1995) and the present (1984–2010) for central Alberta shows that the risk of intensive storms occurring has increased for all storm durations in recent years, and the risk had been higher for storms of large return periods ($\ge 50\mathrm{year}$). Given a design project life of 50 years, the average increase in risk is 9%. Second, the uncertainty associated with projecting the risk of intensive storms occurring in 2011–2100 was assessed by considering three special reports on emissions scenarios (SRESs) of four global climate models (GCMs) dynamically downscaled by a regional climate model (RCM), i.e., MM5. Based on storms simulated by MM5 for central Alberta forced by SRES climate scenarios of Intergovernmental Panel on Climate Change (IPCC) for 2011–2100, the median risk for short-duration storms ($\le 1\mathrm{h}$) of a design project life of 25 and 50 years is projected to increase up to 37 and 38%, respectively. In other words, climate change impact could increase the vulnerability of central Alberta to the hazards of flooding by intensive storms in future. The proposed risk chart presents the risk in a straightforward and meaningful way which will be useful for the long-term planning and engineering design of municipal infrastructure.

Transforming a weather forecast into a flood hydrograph for a watershed requires a series of steps: (1) gathering and pre-processing data about the current conditions of the watershed, such as soil moisture, snow cover, etc.; (2) creating or modifying a hydrologic model; (3) running the model; and (4) assessing the results. Adequately performing all these steps in a short period of time as a major storm is approaching is in many cases extremely difficult. In this paper the authors present a method that enables performing the steps 2, 3, and 4, in advance of a flooding crisis by determining a large set of potential “scenarios” (combination of parameters), running a model for each, and storing the results. The researchers analyze a way to create the scenarios so to maximize the uniformity of their distribution on the parameters space. They present a simple technique to retrieve the stored result that best matches the current conditions. The authors illustrate this method in a test case, for which seven model parameters were selected and more than 2,000 scenarios were generated and pre-computed.

The grain-size distribution can play an important role in the sediment movement and the bedload transport rate. However, it still remains an important and challenging issue in the study of river behavior. Accurate estimation of the grain-size distribution is desired, while simultaneously one expects to spend much less time on it. Recently image analysis and machine learning techniques facilitated grain identification and measurement on images. In this paper, a semisupervised affinity propagation model (SAPM) oriented to images method is proposed for automatic extraction of the grain-size distribution based on photographs sampled from Wenchuan and Yingxiu in China where landslides and mudslides usually take place. The model to estimate the grain-size distribution is developed and the corresponding algorithm is illustrated in detail. The experiments are finished in both lab and field, and the proposed algorithm is compared with traditional methods. The proposed algorithm produces much better results in estimating the grain-size distribution in comparison with other image processing methods and manual sieving methods. It is shown that SAPM is an efficient method for precisely estimating the grain-size distribution.

The current study presents an improved Shiono and Knight method (SKM) applying the complete secondary flow term that includes both the dispersion and the transverse convection components, which outspread the SKM solution. The model calibration coefficients for quantifying the secondary flow term are specified for the discrete lateral regions, defined based on the perturbation in distribution of lateral depth-averaged velocity and the boundary shear stress distribution. The inaccuracy in prediction of the SKM approach is attributed to poor secondary flow cells modeled in the interface zone. Improved SKM is adapted for laboratory and natural symmetrical and asymmetrical cross sections in both homogeneous and nonhomogeneous boundary conditions. The improved SKM has notably enhanced the precision of the prediction of depth-averaged velocity in the interface zone.

The kinematic wave system of shallow-water equations is posed as an initial value problem, or Cauchy problem, to investigate error analysis using the square-grid finite-element method. The consistent, lumped, and upwind models of the square finite-element method are analyzed using the Fourier (or von Neumann) stability method as separate initial value problems for two-dimensional shallow-water equations. The small difference between the Cauchy solution and integer (or half) multiples of the eigenvalue solution of amplification factors using the eigenvalue method indicates that the Cauchy solution is a possible stable numerical solution of the kinematic wave shallow-water equations within the reasonable limits of solution accuracy. The nodal amplification factors are less than or equal to unity for implicit finite-element schemes for all of the three formulations—consistent, lumped, and upwind for all of the wave numbers, implying unconditional stability. For explicit and semi-implicit finite-element schemes, at least one of the four nodal amplification factors exceeded unity for all wave numbers, thus implying that the explicit and semi-implicit schemes are unconditionally unstable (except for semi-implicit, upwind). Although element-level amplification factors may be useful in obtaining stable solutions for the interior part of the solution domain, amplification factors exceeding 1.0 might be encountered with associated errors and oscillations. Therefore, an understanding of nodal amplification factors is necessary for better solution accuracy and solution stability. The current research bridges the knowledge gap between the Cauchy and eigenvalue solutions, brings closer the accurate large-scale numerical modeling of watersheds, and delineates directions to be followed in the future for the hydrologic modeling community.

Rainfall plays a crucial role in the socioeconomic development of a country. Knowledge of both the amount of rainfall and its pattern of distribution are equally important for proper management of water resources systems. In the present study, trends of two rainfall series from different locations having different time periods and time steps have been extracted using the singular spectrum analysis (SSA) method. Data analyzed include monthly data of England and Wales precipitation (EWP) from 1766 to 2002 in which no periodic component is prevailing and daily rainfall data of Koyna watershed, Maharashtra, India from 1961 to 2009, which shows a strong periodic component of 365 days. Method of periodogram analysis has been used in order to select the components corresponding to trend in the grouping stage of SSA. The Mann–Kendall (MK) test is also used to detect trends in EWP monthly series and the performance of SSA and MK test is compared. The result showed that the MK test could detect the presence of a positive or negative trend at a significant level, whereas the proposed SSA method could extract the nonlinear trend present in the series along with its shape. Trends extracted from the England and Wales precipitation are compared with a previously published EWP trend extraction study. The comparison shows that the method of SSA in trend extraction could extract nonlinear trends along with its shape whereas the previous study extracted linear trends. The EWP monthly rainfall series showed an increasing trend during the winter season and a decreasing trend during summer. The trend extracted for the Koyna series has very small values (almost constant), implying that the rainfall series is almost stationary. The study proves the applicability of SSA for extracting nonlinear trends that provide more insight into the observed time series.

Reservoirs play an important role in modifying the natural hydrological regime. Reservoir-induced changes in flow fluctuations at monthly and hourly scales were analyzed through a case study of a tributary of the Yangtze River (China), the Qingyi River, where the Chencun reservoir is located at the upper reaches. Monthly and hourly flow data were collected from two flow gauging stations monitoring the inflow and outflow regimes, respectively. The flow fluctuations were characterized by coefficient of variation ($C_V$) and concentration degree ($C_d$). Continuous Morlet wavelet transform was applied to detect diurnal period. The influence of the reservoir was revealed comparing the postdam with the predam conditions by and comparing the affected with the natural conditions. The results indicate two opposite change patterns. Reservoir impoundment may mitigate flow fluctuations at both of the monthly and the hourly scales. At a monthly scale, reservoirs usually impound water in the flood season for the dry season. At an hourly scale, reservoirs may restrain a flood pulse and strictly control the outflow. Thus, the reservoirs reduce variation and concentration of the runoff distribution within a year and within a day. In contrast, the reservoirs may aggravate the hourly flow fluctuation by releasing hydropeaking waves for hydropower generation and superimpose a diurnal period to the downstream flow regime. But the artificial diurnal fluctuation is of some uncertainty in terms of appearance time and duration. This work may deepen the understanding of the reservoirs influence mechanism on the downstream flow regime.

Dams have great hydrological impacts on the river system. Poyang Lake, the largest freshwater lake in China, has experienced severe drought over the past decade since the operation of the Three Georges Dam. The Jiangxi Province Government proposed to build a 2.8-km-wide dam with sluice gates across the narrowest part of the channel that links Poyang Lake and Yangtze River, called the Poyang Lake Project (PLP), to solve the problem of drought. The proposal caused concerns about the negative effect of increased water level on wintering birds’ habitat and water quality. But so far, little scientific data have been provided to evaluate the hydrological impacts of the PLP. The aim of this study is to investigate the hydrological impacts of the PLP using a two-dimensional Environmental Fluid Dynamics Code (EFDC) model based on the latest regulation scheme, which will be helpful to evaluate the advantages and disadvantages of the building of PLP. The model was calibrated and verified by comparing model predictions with observed data. A typical dry season between 2006 and 2007 was used as the period to simulate the scenarios with and without the PLP. Simulated results from the EFDC model showed that the increase of water depth mainly occurred in the water channel in the north part of Poyang Lake with the regulated water level set at 10–11 m, and the increased water depth in two national nature reserves was less than 0.1 m, which has little effect on wintering birds’ habitat. However, the benefit of the PLP was limited with this regulated water level and the increased water storage was less than $6.2\times {10}^8{\mathrm{m}}^3$. The increased water storage would double when the lowest water level was set at 12 m, and the bird habitat was still little affected. The existing regulated water level reduced the flow velocity in the northern part of Poyang Lake by approximately 50% and the effect became smaller with distance far away from the PLP. The slowdown of flow velocity would increase the eutrophication risk and the water quality would deteriorate if the regulated water level was set higher. Whether the project should go ahead depends on how to balance the advantages and disadvantages.

From a water management perspective, the bulk quantity of water moved and where it has been sourced from are more important than simply estimating drawdown effects due to pumping. This note considers the problem of estimating the total quantity of water discharged between each aquifer in a multilayer aquifer model of Dudley Ward and Falle. Direct formulas are obtained in terms of the storativities and leakage coefficients. Assuming a typical disparity of scale between storativities of the phreatic aquifer and confined aquifers, type curves show that over time the vast majority of extracted water is sourced from the overlying phreatic aquifer.

The authors presents a set of automated geographic information system (GIS)-based geoprocessing tools for the implementation of one-dimensional hydrodynamic (1D-HD) models that simulate an unsteady open channel flow through a channel network. The authors’ aim is to determine the degree of automation that is granted in the process of preparing river bathymetry and model parameters for 1D-HD model simulations. The authors present two case implementations to illustrate the use of the geoprocessing tools and to demonstrate how they can simplify some of the preprocessing tasks required for preparing model parameters. The authors present a statistical analysis of the automated algorithms performance for finding bank stations and show that it performs adequately in the authors’ study basins.

Streamflow forecasting is essential for the proper management of water resources, especially when severe droughts cause water resource scarcity. Streamflow forecasting using physically based or conceptual hydrologic models is a common approach. However, these models rely on the predicted climate data, which are at times unrealistic and depart significantly from actual observed data, resulting in an unreliable forecast. Because the sea surface temperature (SST) in the Niño 3.4 region has a potential teleconnection with streamflow in the El Niño Southern Oscillation (ENSO)-affected regions, the streamflow forecasting ability of a model can be enhanced by using SST in data-driven models. In fact, conceptual models cannot incorporate SST data as input. Therefore, in this study, an adaptive neuro-fuzzy inference system (ANFIS) was used to infuse SST data (from the equatorial Pacific) with predicted precipitation and temperature for streamflow forecasting with one-to-three months’ lead time. For the forecasted climate data, two methods were used: (1) ENSO-conditioned weather sequences, and (2) climate data from the Climate Forecast System version 2 (CFSv2) model. The forecasted streamflow, after systematic error correction, was postvalidated with observed streamflow from 1982 to 1988. The streamflow forecasting at one-month lead time was found to be better than that of the three-month lead time. The root-mean square error and percentage bias for one-month lead time forecast using CFSv2 were $3.8{\mathrm{m}}^3/\mathrm{s}$ and 7%, whereas these statics using ENSO-conditioned weather-sequence data were $4.4{\mathrm{m}}^3/\mathrm{s}$ and 10.5%, respectively. This research concludes that the climate model approach is a better choice for moderately sized watersheds for streamflow forecast with a one-month lead time. Conversely, the weather-generator approach is more suitable for streamflow forecasting with a three-month lead time, especially for low-flow conditions.

In this study, an index, named as standardized precipitation anomaly index (SPAI), is proposed for the meteorological drought quantification in the context of the monsoon-dominated climatology, where the precipitation is strongly seasonal and periodic. In the computation of SPAI, the anomalies of the precipitation are normalized rather than normalizing the raw precipitation series. The SPAI is compared with the standardized precipitation index (SPI), with respect to certain shortcomings of the latter. It is shown that the SPAI, owing to its design, is able to successfully differentiate between the consequences of shortages/surplus in rainfall in the monsoon and nonmonsoon months which is not possible through SPI. The unique suitability of SPAI for monsoon dominated regions is also illustrated by comparing its premise of development with that of the standardized nonstationary precipitation index (SnsPI). Further, drought quantification through the SPAI is shown to be applicable for both periodic and nonperiodic precipitation series. This is demonstrated using a typical strongly periodic precipitation series (from India) and a typical nonperiodic precipitation series (from Arkansas, United States of America). As compared with SPI, the SPAI is found to have a better coherence with the consequences of droughts and wet spells faced by the country (India, as the study area) in the past.

This paper uses adaptive neuro-fuzzy inference system (ANFIS), artificial neural network (ANN) and nonlinear regression (NLR) for flow duration curve (FDC) estimation at ungauged sites. For this reason, all stations existing in the Namak Lake basin of Iran were considered due to their long period of data availability and minimum human activities. In total, 33 hydrometric stations were selected and the annual FDC was determined for each station. In selecting the effective factors on FDCs, 18 parameters were extracted such as physiographical, meteorological, land use, and geological characteristics using Arc/GIS. Six factors including weighted average height (H), area (A), rangeland area (RA), drainage density (DD), permeable formation (PF), and average stream slope (SS) using principal-component analysis (PCA) were selected, which illustrate 83.54% of variation of the data. The results showed that the ANFIS has generally the lower root-mean squared error (RMSE) and higher Nash criterion than the ANN and the NLR for regional analysis of FDCs.