Water Resources Simulation Research Articles

Assessment of the effects of agricultural practices on non-point source pollution for a coastal watershed: A case study Nif Watershed, Turkey

Best management practices (BMPs) are increasingly being used by decision makers to reduce agricultural non-point source (NPS) pollution while improving productivity for the farmers. Models are used as planning tools and are especially useful for targeting priority areas for BMP implementation. This study describes a modeling system to assess the impacts of agricultural NPS pollution on coastal environments. The approach is based on Remote Sensing (RS), Geographical Information Systems (GIS) and Hydrological Modeling (HM) and applied to a pilot watershed – catchment area of Nif Creek – in the Aegean region of Turkey. Land Use and Land Cover (LULC) map of the selected area is produced from LANDSAT MSS image based on Co-ordination of Information on the Environment (CORINE) classification scheme. Simulator for Water Resources in Rural Basins (SWRRB-WQ) was used to estimate the NPS pollution loads. The effects of agricultural management practices such as application of irrigation, fertilizer and tillage on the streamflow quality and quantity can be determined by such a modeling system.

A Bayesian method for multi-site stochastic data generation: Dealing with non-concurrent and missing data, variable transformation and parameter uncertainty

Stochastically generated stream flow and climatic data may be used as input to water resources simulation models for planning purposes. Dealing with non-concurrent and missing data, variable transformation and parameter uncertainty presents a significant challenge in the development of methods for stochastic data generation. In this paper, a Bayesian method is introduced for multi-site stochastic generation of annual stream flow and climatic data. A contemporaneous autoregressive lag-one model CAR(1) with the Box–Cox transformation is used to capture key statistical structure of multiple annual stream flow and climatic time series while keeping the number of model parameters to a minimum. The posterior joint distribution of the model parameters is formulated, allowing for inputs of historical data series that are not continuous or concurrent, thus avoiding the need to infill or truncate data records and maximising the value of available data. Parameter and uncertainty inference are solved numerically by using Markov Chain Monte Carlo simulations. Subsequent stochastic generation of data fully accounts for parameter uncertainty. In addition, a re-parameterization scheme is used to handle the problem of strong inter-parameter dependence from the Box–Cox transformation. The method was applied to the Melbourne Water supply system to demonstrate its computational feasibility.

Optimal groundwater remediation design using an Adaptive Neural Network Genetic Algorithm

Large‐scale water resources optimization often involves using time‐consuming simulation models to evaluate potential water resource designs or calibrate parameter values. Approximation models have been proposed for improving computational efficiency of the optimization. In most instances, multiple simulation runs have been done prior to the optimization, which are then used to fit an approximate model that is used during the optimization. This paper demonstrates that this approach can lead to suboptimal solutions and proposes a dynamic modeling approach, called Adaptive Neural Network Genetic Algorithm (ANGA), in which artificial neural networks are adaptively and automatically trained directly within a genetic algorithm (GA) to replace the time‐consuming water resource simulation models. A dynamic learning approach is proposed to periodically sample new solutions both to update the ANNs and to correct the GA's convergence. Different configurations of ANGA were tested on a hypothetical groundwater remediation design case, and then the best configuration was applied to a field‐scale case. In these applications, ANGA saved 85–90% percent of the simulation model calls with no loss in accuracy of the optimal solutions. These results show that the method has substantial promise for reducing computational effort associated with large‐scale water resources optimization.

A generic database and spatial interface for the application of hydrological and water resource models

The paper discusses a software package that has been designed to enhance the efficiency of applying a range of hydrological and water resource simulation models. The SPATSIM (SPatial and Time Series Information Modeling) system has been developed in Delphi using MapObjects and incorporates a spatial data interface for access to the different types of information commonly associated with water resource analyses. All of the information is stored within database tables (Paradox by default) with generic structures. The structure and components of the system are briefly described and some example applications provided. The discussion identifies some of the criticisms that have been leveled at the software, the approach of the developers to user support and some possible future developments. The conclusions suggest that the development can be considered successful in that it has substantially improved the efficiency of the Institute for Water Research to undertake one of its core functions of hydrological model development and application.

Computer software tool REALM for sustainable water allocation and management

REALM ( REsource ALlocation Model) is a generalised computer simulation package that models harvesting and bulk distribution of water resources within a water supply system. It is a modeling tool, which can be applied to develop specific water allocation models. Like other water resource simulation software tools, REALM uses mass-balance accounting at nodes, while the movement of water within carriers is subject to capacity constraints. It uses a fast network linear programming algorithm to optimise the water allocation within the network during each simulation time step, in accordance with user-defined operating rules. This paper describes the main features of REALM and provides potential users with an appreciation of its capabilities. In particular, it describes two case studies covering major urban and rural water supply systems. These case studies illustrate REALM's capabilities in the use of stochastically generated data in water supply planning and management, modelling of environmental flows, and assessing security of supply issues.

CalSim: Generalized Model for Reservoir System Analysis

The California State Department of Water Resources and the United States Bureau of Reclamation Mid-Pacific Region have developed a general-purpose reservoir–river basin simulation model for the planning and management of the State Water Project and the federal Central Valley Project. The California Water Resources Simulation Model brings a fundamental change to modeling of these systems. Model users specify system objectives as input to the model. System description and operational constraints are specified using a new water resources engineering simulation language. A mixed integer linear programming solver efficiently routes water through the system network given the user-defined priorities or weights. Simulation cycles at different temporal scales allow for successive layering of constraints. The power and flexibility of the model is demonstrated by its ability to simulate the operation of complex new environmental water accounts.

Coupled SVAT–groundwater model for water resources simulation in irrigated alluvial plains

Understanding the interaction between soil, vegetation and atmosphere processes and groundwater dynamics is of paramount importance in water resources planning and management in extensively irrigated alluvial plains. This is the case, for example, of the most important agricultural and industrial area in Italy, the Padana Plain, where intensive exploitation of groundwater for domestic and industrial supply coexists with massive diversions from surface water bodies, providing abundant irrigation to one of the most productive agricultural districts in Europe. The paper presents a simulation system which reproduces the hydrological processes relevant in alluvial irrigated plains. In particular, it allows the evaluation of the distribution of crop water consumption in space and time, as well as simulation of the interaction between recharge and groundwater dynamics. The simulation code is based on the coupling of two models: a conceptual vadose zone model and the groundwater flow model MODFLOW. Additional code was written to provide an interface which performs the explicit coupling in space and time between the two models. A geographical information system (GIS) manages the spatially distributed inputs, parameters and outputs of the system. An application of the package to a large irrigation district, of approximately 700 km 2, located in the middle of the Padana Plain, is also discussed in the paper.

Identification of physical processes inherent in artificial neural network rainfall runoff models

AbstractThe emergence of artificial neural network (ANN) technology has provided many promising results in the field of hydrology and water resources simulation. However, one of the major criticisms of ANN hydrologic models is that they do not consider/explain the underlying physical processes in a watershed, resulting in them being labelled as black‐box models. This paper discusses a research study conducted in order to examine whether or not the physical processes in a watershed are inherent in a trained ANN rainfall‐runoff model. The investigation is based on analysing definite statistical measures of strength of relationship between the disintegrated hidden neuron responses of an ANN model and its input variables, as well as various deterministic components of a conceptual rainfall‐runoff model. The approach is illustrated by presenting a case study for the Kentucky River watershed. The results suggest that the distributed structure of the ANN is able to capture certain physical behaviour of the rainfall‐runoff process. The results demonstrate that the hidden neurons in the ANN rainfall‐runoff model approximate various components of the hydrologic system, such as infiltration, base flow, and delayed and quick surface flow, etc., and represent the rising limb and different portions of the falling limb of a flow hydrograph. Copyright © 2004 John Wiley & Sons, Ltd.

Experiment with Simulation Models in Water-Resources Negotiations

A laboratory experiment was conducted to investigate the effects of various kinds of access to a simulation model on water-resources negotiation outcomes and processes. The (mock) negotiation involved determining a release schedule from two reservoirs through a negotiation between subjects representing hydroelectric, agricultural, and flood-control interests. Results revealed the following: (1) The model was of only limited value in helping negotiators understand the behavior of the environmental system; (2) the model did provide assistance in finding policies that satisfied specific task constraints; (3) the availability of the model encouraged negotiators to consider more policies; and (4) the benefits of using the model directly, rather than getting access to model results produced by an assistant, were offset by the burden of direct use. The findings are evaluated in light of a growing trend in many governmental organizations to make water-resources simulation models available to stakeholders and the public at large.

An example of using Landsat and GOES data in a water budget model

Landsat and GOES satellite data are used in analyzing monthly water budgets on a 538‐km2 watershed in central Oklahoma for 1984–1985. The budget components examined are water yield, percolation loss, evapotranspiration and change in soil water content. The hydrologic model employed is SWRRB (Simulator for Water Resources in Rural Basins) developed by the Agricultural Research Service of the U.S. Department of Agriculture. Landsat imagery is used to subdivide the basin into land covers for model application. The land cover subareas are more homogeneous than subbasins with mixed vegetation and their use allows tracking of the unique contribution of each type of cover to the water budget. Daily insolation is determined from GOES imagery. A comparison of budget components based on pyranometer insolation versus unbiased GOES estimates shows a maximum difference less than 4% and 5 mm for any component of the annual and monthly budgets, respectively. Satellite‐derived temperatures are simulated by adding systematic and random errors to observed temperatures. A temperature bias of a few degrees greatly changes the growing season and monthly budgets of winter wheat.

MODELING THE EFFECT OF BRUSH CONTROL ON RANGELAND WATER YIELD1

ABSTRACT: With the increase in water demand in Texas, attention has turned to improving water yield by brush control on rangeland watersheds. Several hydrologic models have been developed for either farmland or rangeland. However, none of the models were specifically developed to assess the impact of brush control on rangeland water yield. Yet, modeling the impact of brush control on water yield needs to be considered if alternative techniques are to be compared. Two models, Ekalaka Rangeland Hydrology and Yield Model (ERHYM‐II) and Simulator for Water Resources on Rural Basins (SWRRB) were selected. The Soil Conservation Service curve number (SCS‐CN) method is used in both models to predict surface runoff from each rainfall event. The major differences between the ERHYM‐II and SWRRB models are the evapotranspiration, soil water routing, and plant growth components. The models were evaluated on brush‐dominated and chemically and mechanically brush‐controlled range watersheds in Texas. Results indicated that both models were capable of simulating soil water and water yield from brush dominated and chemically brush‐controlled range watersheds. The models were not able to predict water yield from the mechanically brush‐controlled (root plowed) watershed with acceptable accuracy. The depressions that were caused by root plowing stored surface runoff and reduced water yield from the watershed. Information about the size of depressions was not available for further model evaluation.

Global Water Management Model of Surface and Groundwater Resources of the Souss Plain in Morocco

In the Souss plain, groundwater resources have traditionally been developed particu-lary for the agriculture sector, which had a remarkable growth during the two decades 1960 and 1970. As a consequence of the increasing groundwater withdrawal and of the occurence of an exceptionally dry period between 1972 and 1986, the groundwater level has significantly dropped during the last two decades, reflecting a progressing deficiency of the aquifer’s balance. As of 1986 development of surface water was still limited and the only sizeable impounding dam had just started operations. Although an extensive development of surface water would constitute a major action for preventing further deterioration of the aquifer’s balance, there exist no further suitable sites to impound reasonable volumes of water. The attention of planners has therefore been focused on the opportunity to build smaller impounding structures aimed at reducing flood intensity and consequently increasing infiltration and recharge of the aquifer.In order to make suitable choices for the emphasized structures and to define optimum rules for integrated surface and groundwater management in accordance with the geographical distribution of the water demand, the application of a global water resources simulation model integrating both surface and groundwater has been sought.In conclusion, the conceptual structure of the model and preliminary results of infiltration simulations are presented.

MODELING THE EFFECTS OF URBANIZATION ON BASIN WATER YIELD AND RESERVOIR SEDIMENTATION1

ABSTRACT: The dam impounding White Rock Lake was completed in 1910 to provide water for the City of Dallas. Since then, land use on the watershed has changed from entirely rural to over 77 percent urban. A model called SWRRB (Simulator for Water Resources in Rural Basins) was utilized to determine the effect of urbanization on water and sediment entering the lake. The simulation results show that, if urbanization had not occurred, then the annual surface runoff would be 135 mm rather than 151 mm and the annual sediment yield would be 4.4 t/ha rather than 4.1 t/ha. Also, the effect of urbanization on delivery ratios was shown and a positive linear correlation was found. Finally, the weather generator in SWRRB was utilized to estimate the loss of reservoir capacity until 2050 for three different land use management scenarios.

Validation of SWRRB—Simulator for Water Resources in Rural Basins

A model called SWRRB (simulator for water resources in rural basins) was developed for simulating hydrologic and related processes in rural basins. The SWRRB model was developed by modifying the CREAMS (chemicals, runoff, and erosion from agricultural management systems) daily rainfall hydrology model for application to large, complex, rural, basins. The three major components of SWRRB are weather, hydrology, and sedimentation. Processes considered include surface runoff, return flow, percolation, evapotranspiration, transmission losses, pond and reservoir storage, sedimentation, and crop growth. SWRRB has been tested on 11 large watersheds from eight Agricultural Research Service (ARS) locations throughout the United States. The results show SWRRB can realistically simulate water and sediment yields under a wide range of soils, climate, land‐use, topography, and management conditions. SWRRB should provide a versatile and convenient tool for use in planning and designing water resources projects.

Simulator for Water Resources in Rural Basins

A model called SWRRB (Simulator for Water Resources in Rural Basins) was developed for simulating hydrologic and related processes in rural basins. The objective in model development was to predict the effect of management decisions on water and sediment yields with reasonable accuracy for ungaged rural basins throughout the United States. The three major components of SWRRB are weather, hydrology, and sedimentation. Processes considered include surface runoff, percolation, return flow, evapotranspiration, pond and reservoir storage, and sedimentation. The SWRRB model was developed by modifying the CREAMS (Chemicals, Runoff, and Erosion from Agricultural Management Systems) daily rainfall hydrology model for application to large, complex, rural basins. The major changes were: (1) A return flow component was added; (2) the model was expanded to allow simultaneous computations on several sub‐basins; (3) a reservoir storage component was added for use in determining the effects of farm ponds and other reservoirs on water and sediment yield; (4) a weather simulation model (precipitation, solar radiation, and temperature) was added to provide for longer‐term simulations and more representative weather inputs, both temporally and spatially; (5) a better method was developed for predicting the peak runoff rate; and (6) a simple flood routing component was added. Besides water, SWRRB also simulates sediment yield using the Modified Universal Soil Loss Equation (MUSLE) and a sediment routing model. Tests with data from a 538km2 basin in Oklahoma and a 17.7km2 basin in Texas indicate that SWRRB is capable of simulating water and sediment yield realistically.

Design of Length of Water Resource Simulation Experiments

The research procedure was to: (1)Explore the experimental nature of simulation; (2)outline criteria for making sampling decisions about the length of simulation experiments; and (3)develop test statistics for ranking alternative system designs. Lengths of simulation runs required to achieve desired levels of reliability are shown to depend on serial and cross-covariance properties of simulated net benefit time series. Application of the procedures suggested by this study to a set of simulations of alternative plans for the Rio Colorado, Argentina, shows that the longest runs are required for alternatives with largest variance of annual net benefits. Serial correlation tends to increase required run length whereas cross-correlation tends to reduce required run length. In need of further work are: procedures to rank many alternatives having correlated net benefit series; procedures for nonstationary series; procedures to deal with non-normal benefit series; and procedures to account for the consequences of making incorrect choices.

Decision Criteria for using Stochastic Hydrology

A decision-theoretic plan is presented that provides insight into questions of interest to the system planner about to embark on a study which uses stochastic hydrology. The plan considers and integrates three broad areas: (1) the collection, interpretation, and adjustment of the data bank which is required to initiate stochastic modeling; (2) the investigation of existing, and development of new, stochastic hydrology models; and (3) the construction of water resource simulation procedures whose outputs serve as vehicles by which the importance of the other necessary tasks can be determined. The overall plan has component tasks which correspond directly to the design steps of the system planner. Emphasis is laid upon data management. A scheme for multivariate data fill-in, based on variance reduction criteria, is presented.