Journal of Water Resources Planning and Management

The processing of composite samples by the station‐year approach either coupled or not coupled to the systematic use of Hershfield's formula can be very helpful when studying design floods. Therefore, a station‐year approach is applied to rainfall and discharge data from Indonesia, Malaysia, and the Philippines to define regional distributions for observed maxima. After a cluster analysis defining groups of homogeneous stations, a double distribution concept is used to properly take into account the existence of outliers. Results are compared with information available on probable maximum precipitation (PMP) and design floods. Concerning Indonesian rainfall data, the estimate of 10,000‐year daily rainfall was consistent with the application of Hershfield's method for computing PMP. Hershfield's PMP was between 73% and 115% of the 10,000‐year daily rainfall. The processing of discharge data yields results that are confirmed by project inflow design floods. The method provides a bridge between statistical and hydrometeorological approaches. Regional analysis obliges to weigh the occurrence of outliers. Data clustering is recommended for applications involving large amounts of information.

The probabilistic‐based calibration strategy developed in this paper was briefly reviewed and applied to derive the optimum values of eight calibration parameters. Three optimum sets are obtained: estimations of peak flow, runoff volume, and both peak flow and runoff volume. Data for each of three storms were used to calibrate and verify the model for a catchment in Singapore to demonstrate the application of the calibration methodology. The calibration results indicate that: (1) The derived optimal values approximate the global optimum; (2) a close agreement between the simulated and the measured catchment responses (peak flow or runoff volume) could be achieved when only one specific objective function is considered at a time; and (3) a reasonably good match between the simulated and the measured hydrographs could be achieved, however, when both catchment responses (the peak flow and the runoff volume) are simultaneously considered. The verification simulations also indicate reasonably good comparisons between the simulated and the measured data.

Global simulation models used to predict climate change typically have resolutions too coarse for resolving small drainage basins, creating a major obstacle to evaluating the hydrologic impacts of climate variability. The writers circumvent this difficulty by using a physical model for severe hurricanes to derive changes in hurricane rainfall under a predicted global warming. The modified rainfall is then transferred to a linked hydrologic‐hydrodynamic model to compute the impact of global warming on hurricane‐induced flooding in a susceptible watershed. As an example, using simulations of tropical, $\mathrm{doubled}‐{\mathrm{CO}}_2$ climate change, the hurricane model predicts a 40% precipitation increase for severe hurricanes affecting southeastern Florida. The 40% increase is assumed for a 100‐year, three‐day event used for flood‐management planning in South Florida and applied to the Basin Runoff and Streamflow Simulation (BRASS) model of the Western C‐11 drainage system. The increased precipitation provokes substantially greater flooding that could negate most of the benefits of present water‐management practices in this basin.

Unsaturated flow models are increasingly being used in ground‐water quantity/quality studies to predict recharges due to rainfall. This paper presents a critical examination of the current practice of using numerical models in light of the approximate nature of problem identification. Three issues are addressed: (1) Sophistication of models; (2) specification of boundary conditions; and (3) selection of hydrogeological properties. To systematically study these issues, two ground‐water models are used, a simple lumped parameter model and the numerical model, UNSAT1. The models are applied to two hypothetical unsaturated domains as well as to a field site where extensive monitoring of ground‐water conditions is reported in literature. The results indicate that, in the case of spatially homogeneous soil profiles, errors in recharge predictions due to improper problem identification are more significant than those due to model simplification. The study exemplifies the importance of problem identification in modeling studies.

The objective of this work is to develop a general methodology for determination of the optimal freshwater inflows into bays and estuaries to balance freshwater demands with the harvest of various types of estuarine resources. The methodology is based upon solving a large‐scale nonlinear programming problem formulated in an optimal‐control framework. Constraints of the optimization problem include regression equations for harvest of the various species that express fishery harvest as a function of the quantity of freshwater inflow. The stochastic element of the problem (i.e. the uncertainty associated with the regression equations for harvest) is considered by expressing constraints in a chance‐constrained formulation. A nonlinear programming optimizer is interfaced with a hydrodynamic transport model to implicitly solve the hydrodynamic‐salinity constraint equations for salinity levels. An augmented Lagrangian method is introduced to incorporate the salinity constraints into the objective so that the problem size for the optimizer is significantly reduced. A computer model OPTFLOW has been developed by interfacing a simulator for the hydrodynamic‐salinity (HYD‐SAL) with a nonlinear optimizer (GRG2) to apply the methodology by the present writers in an accompanying companion paper. An efficient approximation scheme is developed for evaluation of the objective function and its reduced gradient to reduce the computational effort dramatically. The new methodology can provide a very useful tool for decision makers to quantitatively analyze various water‐management strategies.

The companion paper by the writers developed a model for the determination of optimal freshwater inflows into bays and estuaries for the purpose of balancing freshwater demands with the various types of fish harvests. This model has been applied to determine the freshwater inflow needs into the Lavaca‐Tres Palacios Estuary in Texas for various scenarios. The results of the model application are analyzed for various management alternatives, treatment of salinity constraints, and the desired levels of certainty for harvest regression equations. The comparison of the application results with previous models further illustrates the advancement of this new model. The new methodology can provide a powerful tool for analysis of “what if” type scenarios for decision makers to quantitatively analyze various water resources management strategies.

The requirements and basic components of a typical optimal control environment for water‐supply pumping systems are presented and discussed. Examined model components include hydraulic network models, demand forecast models, and optimal control models. A review of existing optimal control methodologies for water‐supply pumping systems is also provided. Examined methodologies are classified on the basis of the type of system to which the methodology can be applied (single source‐single tank or multiple source‐multiple tank), the type of hydraulic model used (mass balance, regression, or hydraulic simulation), the type of demand model used (distributed or proportional), the type of optimization method used (linear programming, dynamic programming, or nonlinear programming), and the nature of the resulting control policy (implicit or explicit). Advantages and disadvantages of each approach are presented, along with recommendations for future work. The applicability of current technology to an existing water‐supply pumping system is examined in light of existing technical limitations and operator acceptance issues.

Many of the hydrodynamic models used for estuary tidal flows and storm surges are based on a set of coupled two‐dimensional hydrodynamic equations, which are obtained from the full three‐dimensional flow equations by averaging the vertical coordinates. Various numerical techniques, such as finite difference, finite element, and the method of characteristics, have been used to solve these equations. The diffusion hydrodynamic model (DHM) has been developed to simulate two‐dimensional surface‐water and channelized flows. In a prior study, the DHM was applied to a deep‐water estuary with computational results, comparable to those obtained using the St. Venant equations and the method of characteristics. In this paper, the DHM is applied to a shallow estuary located in southern California. The main objective is to determine local flow velocities and circulation patterns in the shallow estuary caused by the incoming and the outgoing tide. Verification of the DHM is provided by comparison of tidal gage measurements and computed flow depths.