World Water & Environmental Resources Congress 2003

Most channel reconstruction projects using natural channel design principles have, to date, focused on lower to moderate gradient alluvial streams where the analog design approach (e.g., as advocated by David Rosgen) has been widely used. This method is predicated on the concept that the bankfull hydraulic geometry for a stable reference reach is largely adjusted to the one- to two-year flow and that this morphology can provide a template for channel reconstruction. While this is no doubt valid in many settings, there is abundant empirical evidence indicating that the morphology of steeper, topographically confined headwater channels developed in coarse alluvium is more closely tied to much less frequent flows. The channel architecture of such supply-limited, step-pool streams characteristically consists of transverse boulder steps with intervening pools. These typically have a quite variable spacing as well as considerable variation in such parameters as step height and length.

Two approaches are tested and evaluated to reduce the effort and time required for computation of turbulent flows that plunge and travel as density currents in open channels. These approaches include: (1) Boussinesq approximation when solving the heat transport equation instead of describing the density as a function of temperature, and (2) equivalent excess density approach which combines the two equations of heat transfer and dissolved solids transport in one equation. The computational fluid dynamics solver FLUENT is utilized for testing and evaluation of such methods by applying to a negatively-buoyant flow in a diverging channel. The numerical results show that both the Boussienesq approximation and the equivalent excess density approaches give reasonably accurate solutions and in the same time reduce the computing time.

A theory is proposed to describe contaminant particle transport over highly disordered media due to advection in the stochastic fluid velocity field characterized by long-range correlations. The problem is treated in the most general case. The analysis is based upon the concepts of scale-invariance and Feynmann diagram technique. The particle may be considered diffusing, the diffusivity field being stochastic as well, but this irregularity doesn't occur to result in affecting the power of dispersion-time dependence: it leads only to renormalization of effective diffusivity. On the contrary, velocity fluctuations may lead to anomalous ("superdiffusion") behaviour. Namely, when in fluctuation velocity correlator < v_i (r→) v_k (r→') >∼ |r→ – r→'|^–2h the index h is less than unity, the characteristic migration length R varies in time according to R ∼ t^1/(1+h), which is faster comparing with the classical law R ∼ t^1/2. The heavy superdiffusion tails are shown to behave as t²/r^(5+2h), i.e. are much weaker than in customary superdiffusion theory. The concentration equations with fractional derivatives are given, which describe the contaminant behaviour in the region adjacent to source and in the far distance region. No common differential equation exists for the whole space. For the case h > 1 in the main space region the behaviour of the system corresponds to classical diffusion, and here stochastic advection results only in renormalization of the average diffusivity without any change in Fick' law. But, except for the special values of h, at far distances power tails of concentration appear. In the borderline case of h = 1 the migration length behaves as R ∼ t^1/2 ln ^1/4 t, and the tail has the form t²/r⁷.

Students in the undergraduate Civil Engineering program at Rowan University have worked on a number of applied research projects in water resources engineering. These projects have helped students learn details about hydrologic processes that cannot easily be covered in traditional courses. In addition, students gain valuable research experience, and the projects provide positive changes to local watersheds, thus providing a service to the community. This paper presents two case study projects that have met with success.

The relationship between capacity and total net benefit of concrete gravity and roller compacted concrete (RCC) dams is investigated using benefit-cost analyses. The total capital cost of such dams is equal to the summation of costs of construction, engineering and consultancy services, right of way, and relocation of structures. Operation, maintenance and repair costs are considered as the annual costs during the service life of the structure. Implementation of dams may lead to the achievement of several benefits, such as meeting municipal water demand, irrigation, hydropower, navigation, flood mitigation, water quality control, wildlife and recreational facilities, and positive impacts on the environment. In a case study, benefit-cost analyses have been carried out for 144 different concrete gravity and RCC dam alternatives that can be implemented at Cine Dam site, Turkey. Stability analyses have also been conducted for the proposed alternatives. The information obtained through this study can be used in decision-making for optimum size and type of a gravity dam.

Spatial variability of soil properties like cation exchange capacity and saturated hydraulic conductivity affect the irrigation induced field scale alkalisation. Field data on cation exchange capacity and saturated hydraulic conductivity follow lognormal distribution. Randomly generated values of these variables are used in the UNSATCHEM model simulations for studying the solute transport over 106 days. Irrigations are applied to fallow field. The variability of saturated hydraulic conductivity is ignored while studying the variability of cation exchange capacity and vice versa. Variability of soil water retention parameters is not considered. The variability in alkalisation process is studied by developing the field average profiles of adsorbed Na concentration and ESP (exchangeable sodium percentage), and determining depth wise standard deviation values. Results reveal that spatial variations in cation exchange capacity affect the adsorbed Na concentration mostly in topsoil layer, where soil solution reaches in equilibrium with irrigation water and ESP is the maximum. For the same layer, saturated hydraulic conductivity variability does not affect adsorbed Na concentration. The ESP development is not affected by variability of both variables in the topsoil layer but in below layers, adsorbed Na and ESP development are affected. The effect of cation exchange variability is more dominant than saturated hydraulic conductivity variability. Soil variability need to be considered while planning the alkali water use for agriculture. Irrigation water quality for safe use of alkali water for particular soil should be decided in such a way that linearity of ESP profile is preferred.

Integration of engineering schematics and photos into classroom lectures previously required use of poor quality transparencies made from copied or scanned images. These supplements remained in the possession of the instructor, thus limiting the students' ability to view them outside of the classroom. More recently, faculty have begun lecturing using computerized LCD projection systems, into which images can be easily integrated. This gives students the benefit of seeing virtual on-site projects or phenomena and improves the effectiveness of classroom instruction. Unfortunately, many existing image collections do not have a strong focus on environmental and water resources engineering and, due to insufficient resolution, result in poor projection quality when enlarged in a document or through a computer-projection system. This paper describes a project supported by the National Science Foundation to develop a website and CD-ROM, Digital Images of Environmental and Water Resources Engineering, designed to enhance the pool of resources available for environmental and water resources engineering education. Through this resource, the authors aim to support the integration of technology into engineering courses and further facilitate student learning, understanding and recognition of classroom concepts. The website and CD-ROM contain a database of photographs and video clips that illustrate concepts related to hydraulic structures, open channel flow, sedimentation, and other specializations. This database was constructed by filming an array of existing engineering operations, facility renovations, and construction projects, as well as naturally occurring phenomena. The majority of images are available in three alternative resolutions, each suited towards a particular viewing medium. The image collection is ideal for use by instructors who prefer stand-alone images that supplement classroom discussion or as part of computerized lecture media. The images can also be directly accessed by students outside of the classroom or as part of distance learning courses. Additionally, the website and CD-ROM can be used as a recruitment tool to disseminate information regarding careers in water resources engineering to middle and high school students.

Civil Engineers that specialize in Water Resources have significant opportunities to undertake projects that meet the needs of underprivileged in the U.S. and developing countries. Most of the developing countries and many of the rural poor live in arid or semi-arid climates. Especially in developing countries, the problem of inadequate water supplies is endemic in spite of the various United Nations programs. This paper presents a case study that used engineering knowledge to design novel water supply source among Native Americans.

Clays and its composites have been widely used in the past as a secondary containment wall for underground storage tanks and landfills. The underground storage tanks are used commonly to store gasoline and other organic chemicals, both in the industry and the commercial market. Recent studies have shown that these secondary containment walls are failing. Gasoline and many other organic contaminants pass through clay liners without much effort. This study attempts to evaluate the changes occurring in the interactions between water molecules and clay particles at a microscopic level, using an atomic force microscope, for varying levels of water content. The pore size changes occurring in the clay particle lattice have a profound effect on the permeability of clay. The results show that the clay pore diameters reduce exponentially as the water content increases. This change of pore size can be attributed to the reduction in the size of the diffuse double layer and more extensive hydrogen bonds between clay particles and bipolar water molecules. Due to sample preparation difficulties, however, the pore diameters were obtained when the clay was dry, while interest for underground storage tank containment layer is for wet conditions.

This paper describes methods to evaluate the feasibility of monitored natural attenuation (MNA) for remediation of contaminated sediments deposited on the beds of channels and marshes. The feasibility of MNA depends on the erosion potential of contaminated sediments due to channel flows and burial of contaminated sediments due to deposition of cleaner sediments transported by the channel. Because of the empiricism and uncertainty associated with sediment transport formulas and methods to estimate non-scouring velocities, several empirical and semi-analytical methods are presented to estimate the erosion potential of channel bed sediments. Reasonably conservative judgments can be made based on the results obtained from several alternative methods presented in this paper. The potential for deposition of suspended sediments is evaluated by comparing the fall velocities of suspended sediments and flow velocity in the channel.

An EWRI Task Committee consisting of 30 members from industry, government, and academia is studying the state of the art in long-term groundwater monitoring (LTM) programs for subsurface waste sites. LTM is an on-going activity aimed at assessing remediation performance, containment integrity, and/or continued non-contamination of the subsurface and groundwater. LTM has different goals and needs than site characterization, so data collection, analysis, and modeling approaches must evolve to meet these new needs. This paper summarizes the committee's recommendations for future research and technology transfer to improve LTM design.

The Department of Civil and Environmental Engineering (CEE) at Rowan University (RU), with support from NSF and Rowan, will adopt "Sooner City", a virtual city developed by the School of Civil Engineering and Environmental Science at the University of Oklahoma (OU). A more portable version of Sooner City will be created, to ease adoption by other institutions. To be named "Garden City" at Rowan, it will be used in the CEE-RU undergraduate program. Garden City will provide design throughout the curriculum, with design projects extending over multiple classes and years. Undergraduates will go to the Garden City website to obtain assignments, data and design criteria for homework and projects. They will also be able to store their projects and designs in the city, allowing them to record their accomplishments and easily refer to them when needed for future assignment. Finally, the Garden City website will provide a central location for course webpages, tutorials, etc. Faculty will use Garden City to demonstrate the context (i.e., human communities) of many civil engineering projects. The purpose of this paper is to provide detail on the Garden City project, including how other institutions can adopt it. A multi-class and year project developed at RU is also described.

There are numerous studies related to the application of artificial neural networks to various problems frequently encountered in water resources. The training method of ANNS in the majority of these studies is the feed forward error back propagation method (or multilayer perceptrons, MLPs). In this study another neural network training method, the Generalized Regression Neural Network (GRNN) is employed for daily river fow forecasting. The simulations with GRNNs were satisfactory in terms of the selected performace criteria and compared well with MLPs.

In this study, we consider transient saturated flow in randomly heterogeneous porous media and try to obtain higher-order solutions of mean head and mean flux, as well as their associated uncertainties based on the combination of Karhunen-Loeve decomposition and perturbation methods. We first decompose the log hydraulic conductivity Y = lnKs as an infinite series on the basis of a set of orthogonal Gaussian standard random variables. The coefficients of the series are related to eigenvalues and eigenfunctions of the covariance function of the log hydraulic conductivity. We then write head as an infinite series whose terms h⁽ⁿ⁾ represent the hydraulic head of n^th order in terms of σ_Y, the standard deviation-of Y, and derive a set of recursive equations for h⁽ⁿ⁾. We assume that h⁽ⁿ⁾ can be expressed as infinite series in terms of products of n Gaussian random variables. The coefficients in these series are determined by substituting decompositions of Y and h⁽ⁿ⁾ into those recursive equations. We solve the mean head (and mean flux) up to fourth order in σ_Y and the head (and flux) variances up to third order in σ_Y ². We conduct Monte Carlo simulations (MC) and compare MC results against approximations of various orders from our moment-equation approach on the basis of Karhunen-Loeve decomposition (KLME). We also compare our results with those from first-order conventional moment-equation approach (CME). It is evident that the KLME approach with higher-order corrections is superior to first-order approximations and is computationally more efficient than Monte Carlo simulations and the conventional moment-equation approach.

Solution to groundwater inverse problems is an important if not an essential step in developing efficient groundwater remediation and management strategies. But application of groundwater inverse modeling to field problems has been limited due to the lack of efficient and/or flexible algorithms and immense computational requirements. We have developed an efficient optimization based framework for solving three dimensional groundwater inverse problems in a parallel computing environment. Our implementation is based on a hybrid genetic algorithm - local search (GA-LS) optimizer that drives a parallel finite-element groundwater simulator. The MPI (Message Passing Interface) communication library is employed to exploit data parallelism within the groundwater simulator and task parallelism within the optimizer. Data parallelism in the groundwater simulator is achieved through a domain decomposition strategy and task parallelism in the optimizer is achieved through a dynamic self-scheduling algorithm. Two types of GAs, (i) binary/integer encoded GA (BGA/IGA), and (ii) real encoded GA (RGA), and three local search approaches, (i) Nelder-Mead simplex method (NMS), (ii) Hookes and Jeeves pattern search method (HKJ), and (iii) Powell's method of conjugate directions (PWL) have been implemented.

The objective of this study is to demonstrate how spatial redundancy analysis results are influenced by the plume interpolation method used to evaluate sampling strategies. Specifically, this work investigates how the underlying sensitivities and assumptions of interpolation schemes can affect LTM decisions. The influence of plume interpolation on spatial redundancy results is demonstrated in this study by combining the Non-Dominated Sorted Genetic Algorithm-II (NSGA-II) with 3 different forms of kriging (ordinary, Multigaussian, and quantile) to quantify the optimal tradeoff between sampling costs and interpolation error. The test case used data drawn from a 50 million-node flow-and-transport simulation for a hypothetical perchloroethylene (PCE) plume in a highly heterogeneous alluvial aquifer. Results show that the cost—estimation error tradeoff drastically changes depending on which of the three kriging methods were used to evaluate designs. The differences in the tradeoffs are primarily a function of 2 factors: (1) sensitivity to extreme PCE concentrations and (2) sensitivity to the clustering of PCE samples (i.e., areas of high PCE concentrations are preferentially sampled).

This paper presents the results of the implementation of interfacing parallel particle tracking software with the FEMWATER chemical transport code. FEMWATER is a well-known three-dimensional unstructured finite element code that models subsurface flow and transport. Tracy et al. parallelized FEMWATER using METIS as the partitioner for distributed memory systems. On distributed memory systems, domain decomposition or element-by-element strategies are usually adopted for finite element parallelization, especially for unstructured mesh applications. Eulerian-Lagrangian methods (ELM) are employed to solve a variety of partial differential equations (PDEs), particularly advection/convection-dominated PDEs. The transport module in FEMWATER uses ELM in addition to the traditional finite element method. To parallelize the chemical transport code, a parallel particle tracking algorithm is required to solve for advection in the Lagrangian step. Cheng and Plassmann have developed a library of parallel particle tracking algorithms for generic scientific computing applications. Their developments were based on a framework of parallel in-element particle tracking. What this means is that only element-based operations are specified and particles are partitioned and assigned to processors by their element location within the mesh. The design goal of the particle tracking library was to provide an efficient means for legacy codes such as FEMWATER to rapidly implement parallel particle tracking capability. By using FEMWATER as a candidate code, this paper demonstrates how the use of generic parallel particle tracking libraries speeds up the development of portable parallel versions of legacy application codes.

Multidisciplinary environmental issues associated with land development and nonpoint source pollution are increasingly addressed on a watershed-integrated level. The so-called "watershed approach" recognizes the interconnected nature of all aspects of environmental quality, including land use and water resources, and the need for sustainable growth, and thus becomes an effective way of connecting the seemingly disparate aspects of environmental engineering for our students. This paper presents a case study of the introductory course in environmental engineering at Lafayette College that we have designed around a rapidly developing 200-km² watershed whose outlet is adjacent to campus. Throughout the course, students take part in a variety of field and laboratory activities designed to give a complete picture of the challenges facing the watershed and to show how civil engineers can contribute to future environmental quality of the watershed. The course attracts students minoring in environmental science as well as civil engineering majors. The success of the course has been indicated by increasing numbers of civil engineering majors specializing in environmental and water resources engineering, and increasing numbers of students conducting independent research projects focusing on the watershed.

Senior design projects for environmental engineering students are uniquely suited to incorporate deign criteria such as sustainability and social aspects. Projects that involve innovative designs can be applied to specific needs of a community. For example, at Roger Williams University, several senior design projects have addressed specific problems in nearby communities where there are septic problems and limited fresh water resources. The students designed a pressurized wastewater collection system to minimize the pollutant level in nearby estuaries. In addition, the students designed a groundwater recharge system for the collected wastewater to help mitigate water resource limitations. These projects, as well as others, allowed students not only to learn the technical aspects of design but also to examine social acceptance and sustainability of wastewater reclamation and water reuse. In addition, by designing a solution for real problems and customers, other constraints such as budget can be applied. This paper discusses these design projects and describes how non-technical design criteria were incorporated in a meaningful way.

In this paper an integrated system, named HydroGeologist was developed in Delphi 5 and runs under Windows Operating System with minimal requirements in hardware resources, in order to support the storage and analysis requirements for all hydrogeological information belonging to different public sectors. The data are stored in a distributed Oracle 8i database with embedded security and charging policies. The data search in the database is user friendly and facilitates any search queries for any field in any table of the database without any SQL knowledge. HydroGeologist can produce graphs of time series; calculate time series trend lines and statistical parameters. Geological Cross-Sections are generated on the fly from geological data and the user can change the symbols used for stones, filters or gravel packs easily. Calculation of hydrogeological parameters (T,K,S,R) from pump tests is easy, intuitive and fully parameterized. HydroGeologist can export and import its raw and processed data easily to ASCII text files and communicate with MS Office applications. HydroGeologist can also produce input files for other well-known applications such as ModFlow or Arcview.

Water resources management problems can be computationally intensive and improved methods are needed to allow solution of more complex applications. In this paper, we study a numerical algorithm designed to efficiently solve water resources management applications such as groundwater management problems. The algorithm is a combination of a simple genetic algorithm and a local search method and is called a self-adaptive hybrid genetic algorithm (SAHGA). The paper presents new ways to improve performance of this algorithm together with an analysis of different alternative local search algorithms. The paper also includes an analysis of the reduction in population size that is possible when using SAHGA relative to a simple genetic algorithm (SGA). The results show that the improved algorithm is more reliable and effective in solving the proposed problem, with average savings of 68% with respect to the SGA.

Development a senior capstone design project for any program can be a challenge. The challenge starts with meeting a goal of achieving the involvement of an entire class on one project that focuses on all aspects of the particular engineering program. Additionally, a capstone course must be meaningful, practical, and enjoyable for the students, as well as the instructor. Adding to the mix, prerequisites of prospective employers, accreditation requirements and expectations of a professor to lead such a course escalate the challenge. And finally, any capstone course should represent a gateway into the "real-world" and bring together concepts and principles taught in earlier courses; therefore, ideally, it must have as many "real-world" components possible, making the challenge somewhat daunting to any professor at any school. This paper focuses on addressing a set of defined goals and how the challenges of meeting the goals were achieved in developing a civil engineering senior capstone course at Lafayette College.

Water distribution modeling requires for not only topological and geometrical information but also junction demand data. Maintaining accurate up-to-date demand data is essential to keeping a model calibrated and thus it ensures an informed hydraulic and water quality simulation. However, deriving junction demands can require a variety of data from different data sources. This process is often conducted by retrieving the data from various databases e.g. customer billing data from Customer Information System. (CIS), land use data from Geographical Information System (GIS) and then manipulating the data to calculate demand for current and future conditions on a node-by-node basis. A general procedure entails information extraction, data analysis, service area (polygon) definition, spatial proximity/polygon analysis, consumption aggregation/projection and finally allocating calculated loading to a junction node. It is a tedious and time-consuming procedure. In addition, due to the diversity of CIS and GIS used by different water utilities, a large variety of information sources and data formats have been used. Thus, most of previous studies have been customized to their own data sources and format and generalized procedures are not available. Demand analysis and allocation methods are also separate from a water distribution modeling system. To facilitate loading a water distribution model, this paper has developed a general framework and an fully GIS integrated tool that enables hydraulic and GIS modeler to flexibly calculate, allocate, distribute and project demand by using water consumption data and a variety of GIS-based information.

Organic compounds, owing to their persistence and volatility, present unique environmental problems in the vadose zone of soils. Apart from advection and dispersion, such compounds also undergo several complex chemical reactions such as adsorption, degradation, volatilization and ionic exchange. The focus here is on the role of volatilization losses in organic compounds that have the potential to migrate to the atmosphere by vapor diffusion. A simple mathematical model is developed where the solute resides in the solid, liquid and gaseous phases, and explicitly accounts for immobile water as well as chemical losses through decay. Using moment analysis, this paper studies potential losses by volatilization and degradation for several organic compounds. The results of this study can be used as a screening tool for assessing the contamination potential of various volatile contaminants in different soils. A sensitivity analysis is conducted to assess the importance of different physical/chemical processes on volatilization, degradation, or leaching. The moment analysis also facilitates the development of effective velocity, dispersion, and degradation rates for organic contaminants.

Performance assessment practices are very much essential because of their central role in effective management. It is an utmost need to know the performance of the irrigation systems to keep pace with the population growth and food production. But the limited resources of freshwater and land are to be properly utilized for a sustainable increase in production and the conveyance structures should work efficiently with minimum conveyance losses. Upgrading of existing systems and also functioning of future systems can be improved, if the reasons are known - why a system performs well, whereas systems created under similar conditions do not. For assessing the performance, performance indicators are needed. As the irrigation is a socio-economic and technical process, many factors are to be taken into account while studying the performance of the system like infrastructure design, management, climatic condition, price, availability of inputs and socio-economic settings etc., which are difficult to obtain. However, factors like water, land, finances and crop production can easily be accounted for performance.

Acoustic techniques have been used successfully to measure suspended sediments in marine environments, but more research is needed for effective acoustic measurements in fluvial environments. The high concentrations and wide size distributions of suspended-sediment in fluvial systems present difficult challenges. Experiments have been under way with the goal of improved hardware and software for acoustic sediment measurement. Time series backscatter data 1.8 hours long were collected from a fixed position. Cross-sectional suspended-sediment concentration variation was measured acoustically by repeatedly traversing a transducer across a laboratory flume, producing cross-sectional data at discrete time intervals. Time series of acoustic data 15–30 minutes in length were collected along with pumped reference samples taken simultaneously at four vertical positions over a dune-covered bed. This data was used to continue the evaluation of simple explicit methods for converting backscatter data into concentration values.

An approach to modelling for robust model predictive control of chlorine residuals in drinking water distribution systems (DWDS) was proposed in previous paper. A model for, the next control horizon cannot be obtained by directly using the observation data in the past time due to time varying water demand. Water network simulator is used instead to generate the hydraulic and quality data needed for model identification. There are many uncertain factors that need to be considered. The uncertain coefficients and variables involved in the simulation problem are time varying and their variation rates are quite different, which hints that single estimation technique cannot handle all the estimation problems. Hence, different estimation techniques are needed. The paper proposes the DWDS simulator based design of chlorine dynamic model where the uncertainty is quantified by bounds. Algorithm for calculating the bounds based on the DWDS structure and demand prediction is derived.

The flows of the Ganges and Brahmaputra in Bangladesh are highly seasonal, and heavily influenced by the monsoon rainfall. As a result the rivers in the region swell to brims and often overflow during the monsoon months whereas during the dry months (Nov. – May) the flows in the rivers reduce dramatically. The region, therefore, faces two major hazards: floods during the monsoon and scarcity of water during the dry season. These hazards are most pronounced in the downstream regions, particularly in Bangladesh, which is the lowest riparian of the Ganges. The objects of this paper are i) to study the hydro-meteorological variability and change for the greater Ganges and Brahmaputra regions, including the headwater regions and their role in stream flow in Bangladesh, and ii) to quantify the variability of key stream flow features of the seasonal hydrographs for Ganges and Brahmaputra Rivers, and iii) to explore the climatic factors influencing these hydrologic changes.

The present paper emphasizes evaluation of the risk of subsidence caused by applications of ASR within a multiple-layer aquifer, and sensitivity analysis of the model parameters. Based on conceptual and mathematical models, the governing equation derived in terms of displacement is invoked under the initial and boundary conditions at interfaces of hydraulic separators and aquifers. In this investigation, a nonlinear viscous model is introduced and applied to the governing equation to assess land subsidence due to ASR applications. The analytical results show that the nonlinear viscous solution comprises two parts of deformation that represents both the short-term and the long-term effects of ASR applications. Two groups of the model parameters are analyzed. One group of parameters is related to the compressible clay layer, and the other is associated with the pumping-injecting activity. The analysis of the parameters is conducted using the relation of the normalized displacement versus of the normalized parameters defined in this investigation. Analysis of parameter sensitivity presented in this paper can be used as a guideline for applications of ASR that causes the potential risk of one-dimensional subsidence.

The Sulphur Branch Channelization Project Number 033 in Bedford, Texas, is one of 29 drainage projects included in a $70 million package of improvements that are still in progress. It is also one of four major channelization projects on this stream. The project goals include stormwater management, floodplain integration, soil erosion control and slope stabilization. Located in a residential area of a "bedroom community" of nearly 50,000 people in the Dallas/Fort Worth Metroplex, the project has involved many water management challenges, including ease of maintenance for the city. The existing stream had tall, moderately steep side slopes that were subject to significant erosion. Further erosion was threatening to take more of the residents' property if left unchecked. To help ensure that problem areas were addressed, Bedford's citizens were invited to provide input during the design process. The hydrology of the urban watershed was computed in a Master Drainage Study in 1992. The watershed encompassed primarily residential land use, some commercial development and a limited amount of undeveloped land. The soils were generally a mixture of clays and sands. Since the area was almost fully developed and had moderately steep slopes, the stormwater runoff within the watershed was quick and severe during a rainfall event.

A three-year site for Research Experiences for Undergraduates (REU) in the Civil and Environmental Engineering Program at Rowan University has been established through funding from the National Science Foundation and Rowan University. A common theme, unifies the Rowan REU Site activities-Pollution Prevention. Site activities focus on fundamental and applied multidisciplinary research in pollution prevention. Pollution prevention is a key element of new EPA initiatives to protect our children's health and to promote environmental justice and urban environmental quality. Selected students from all over the USA work closely with engineering faculty on funded research projects in engineering that encourage pollution prevention and sustainable development. This eight-week program exposes students not only to the values of research experiences but helps them reinforce and build other important skills such as communication and leadership. The program also exposes participants to engineering ethics and gender/race issues facing underrepresented populations in science and engineering. The ultimate objective is to provide the REU participants with an enriching research experience that will encourage them to pursue graduate studies.

The drinking water network serving are up to 100 years old in Korea. Therefore, pipelines suffer various degrees of deterioration due to aggressive environments. Water supply system is one of the major urban infrastructures based on activities of mankind. Its most expensive facility is the water distribution network which is an important component in both quantitative and qualitative sides. Therefore, its effective maintenance and operation (M&O) has essential effect on sustaining human life. Moreover, factors of pipe deterioration and models to predict failure and optimal rehabilitation time are needed to support network management strategies. This study has performed to set up the determination of optimal rehabilitation time for Cast Iron Pipes (CIP) with diameter less than 300 mm. And the determination model for optimal rehabilitation time was developed by using cost analysis and deterioration evaluation on rehabilitation alternatives. In the result from the application of the field, the renovation time was faster about 10 years than the replacement time. Especially, as the difference between rehabilitation and replacement time on E-CIP was about from 3 to 5 years, and it was thought that the replacement was effective on E-CIP. To model sensitivity analysis, the discount rate of coefficient was fixed at 0.08, and the values of initial year break rate(N(t₀)) and growth rate coefficient(A) were adjusted in values of 0.0009, 0.0018, 0.0027, and 0.05, 0.10, 0.15 respectively. When the values of N(t₀) and A was increased, the results from the time of rehabilitation and replacement was faster. It was thought that N(t₀), 0.018 was reliable values on the applied pipeline through the result of the study. In case of A, the values of A above 0.1 was thought to be proper.

Since September 11, 2001 the US EPA's water protection task force and regional offices have initiated massive actions to improve the security of drinking water infrastructure. This paper deals with the development and application of a methodology for finding the optimal layout of a detection system, taking explicitly into account the unsteady hydraulics, and the dilution and decay properties of the water quality constituents as distributed with flow. The detection system outcome is a set of monitoring stations aimed at capturing contaminant entries within a pre-specified level of service, defined as the maximum volume of polluted water exposure to public at a concentration higher than a minimum hazard level. The detection system provides an early warning system for a deliberate terrorist external hazard intrusion — a problem that currently has not been solved. The methodology is casted in a genetic algorithm framework tailored made with EPANET. An example application for Anytown U.S.A. is provided.

The effectiveness of disinfectant residuals in drinking water distribution systems against waterborne infectious disease is an open question that is attracting increasing interest. Here, a preliminary framework to analyze and assess the vulnerability of a water system against accidental and deliberate microbiological contamination is presented. Using network water quality modelling, a statistical approach is proposed to quantify the risk of delivering infected water to consumers. The methodology is applied to an example network under different disinfection practices. Results show that the risk of exposure to random pathogenic intrusions can be reduced by selecting adequate disinfection strategies. For the network considered, adding a booster station at the storage tank may improve consumer protection significantly. Moreover, a simple strategy that seeks to maintain a "detectable" disinfectant residual may not be sufficient to guarantee effective consumer protection against microbial contamination.

This paper discusses the current status of wastewater and water supply infrastructure funding in Pennsylvania. The paper focuses primarily on infrastructure funding provided through Pennsylvania's state revolving loan fund programs (SRFs). It also addresses the present and future water supply and wastewater funding needs, and the future of infrastructure funding for the Commonwealth.

In October of 2002, the City of Overland Park, Kansas adopted a new ordinance establishing criteria for stream corridor setbacks and protection. Overland Park is a rapidly growing community located in the Kansas City metropolitan area. As part of the development of setback criteria, the City wished to have a defensible technique for evaluating the migration potential of local streams undergoing urbanization. The City had become increasingly aware of stream stability concerns adjacent to new and older developments. Many of the City's watersheds are already partially developed using traditional design methods. The intent of this procedure is to establish buffer widths large enough that channels undergoing change could evolve to a post-urban dynamic equilibrium with a minimum potential for property damage and with a minimal need for intrusive bank stabilization measures or hard armor. The method developed incorporates geotechnical, geological, hydraulic and fluvial geomorphologic properties of the affected stream reach or sub-basin.

Increasing concerns for the security of the national infrastructure have led to a growing need for improved management and control of municipal water networks. To deal with this issue, optimization offers a general and extremely effective method to identify (possibly harmful) disturbances, assess the current state of the network, and determine operating decisions that meet network requirements and lead to optimal performance. This paper details an optimization strategy for the identification of source disturbances in the network. Here we consider the source inversion problem modeled as a nonlinear programming problem. Dynamic behavior of municipal water networks is simulated using EPANET. This approach allows for a widely accepted, general purpose user interface. For the source inversion problem, flows and concentrations of the network will be reconciled and unknown sources will be determined at network nodes. Moreover, intrusive optimization and sensitivity analysis techniques are identified to assess the influence of various parameters and models in the network in a computational efficient manner. A number of numerical comparisons are made to demonstrate the effectiveness of various optimization approaches.

The State of Oklahoma is assessing its drinking water and wastewater infrastructure. Since the majority of water and wastewater infrastructure projects in Oklahoma are funded through State and Federal loan programs, it is necessary to ensure the financial security of these investments. The financial and operational considerations for water or wastewater system differ for small, medium, and large systems. This research examined the infrastructure needs of systems serving various size communities, and developed a methodology for identifying systems in need of financial and/or technical assistance. The State of Oklahoma has approximately 570 Publicly Owned Treatment Works (POTW) and 860 Publicly Owned Water Systems (POWS). A survey of all the publicly owned systems was performed in conjunction with routine compliance inspections by Oklahoma Department of Environmental Quality (ODEQ). The surveys included interviews with treatment works owners and operators and visual inspection. Data obtained during the onsite inspections was combined with data obtained the ODEQ databases to identify and prioritize potential technical and financial needs.

In this paper we present a model for optimizing the placement of sensors in municipal water networks to detect maliciously-injected contaminants. Optimal sensor placement is required to ensure an adequate coverage of the network's flow for detection and remediation of contaminants. There are many possible objectives that could be formulated for optimal sensor placement, reflecting various costs and risks of an attack on a network. Previous work has focused on detection before a given volume of water has been consumed and detection within a given time limit. Additionally, Tryby et al. present related work on the placement of disinfectant booster station. We consider a sensor placement formulation for which optimal sensor configurations minimize the expected fraction of the population that is at risk for an attack. We model an attack as the release of a large volume of harmful contaminant at a single point in the network with a single injection. For any particular attack, we assume that all points "downstream" of the release point (connected by a set of directed edges) can be contaminated. In general, we do not know a priori where this attack will occur, so our objective is to place sensors to provide a compromise solution across a set of weighted attack scenarios.

A statistical approach, failure analysis, is used. It was first introduced to the water industry in the 1980s but poor data availability compromised its chances of applicability. The recent renewed interest stems from the fact that a wealth of data is becoming available through new tools (monitoring of indicators such as leak, pressure, or flow), or hydraulic simulation. Furthermore, water utilities are now acquiring information systems (such as GIS) where all the above data originally generated for operations purposes is to be organized with the potential to be analyzed for the sake of systematic rehabilitation planning. These beneficial developments are being experienced at a time when U.S. water distribution systems are in serious need of rehabilitation. The expected capital needs to both maintain and expand water distribution systems over the next 20 years are estimated to be between $77.2 billion and $325 billion.

This paper presents some preliminary results of the application of an analytical solution to a one-dimensional advective-dispersive-reactive equation accounting for input from Lake Erie and other point and non-point discharges as well as for transport and kinetic processes along the Niagara River. The objective of this paper is to compare the modeled toxic concentrations of Hexachlorobenzene (HCB) at the exit of the Niagara River, with the field measurements taken at the same location. This comparison is performed in terms of probabilistic analysis. The uncertainty of the model output is estimated from the uncertainty of the stochastic variables and parameters used in the model. The uncertainty analysis provides a better understanding of the influence the stochastic model components have on the overall final results. Estimation of this uncertainty is performed by means of the First-Order-Second-Moment Method (FOSM). Risk and reliability analysis is introduced here as a tool to aid in defining waste discharge limits at the point and non-point source level that would enable fulfillment of the established water quality criteria. The results of the uncertainty analysis of the Niagara River HCB toxic loadings can be a valuable instrument for assessing the kinetic and transport processes along the River and offers a starting point for improvement on the traditional deterministic modeling work.

Tampa Bay Water initiated water supply planning to provide for the requirements of their governing Master Water Plan and to meet projected demands for a 20-year planning horizon. Diminishing water sources combined with evolving state and federal water policies render this effort exceptionally difficult. With this in mind, even preliminary project feasibility studies need to encompass a comprehensive evaluation of issues affecting the permitting, development, and operation of the water source in conjunction with impacts on regional environmental conditions. Black & Veatch developed planning-level feasibility studies for nearly 40 potential new water sources that included vulnerability assessments paralleling Florida's Source Water Assessment Program requirements to delineate the water source boundaries, inventory contaminants of concern, and present a determination of potential vulnerability to contamination. The vulnerability assessments were primarily developed using the data available in the Environmental Protection Agency's Better Assessment Science Integrating Point and Nonpoint Sources (BASINS) program. These data provided a foundation for evaluating interrelated issues such as topography, land use, and regulated pollution sources. Consideration of these planning-level assessments provided a classification of the water source's potential vulnerability to contamination and added an important component to the feasibility studies. These studies will assist Tampa Bay Water with future water supply planning decisions.

A sediment transport model was developed for Lake Waban, MA to evaluate the likelihood that selected areas of the lake would become recontaminated after removal of contaminated sediments from this area of the lake and to assess/quantify the transport of Lake Waban sediments towards and out to Lower Waban Brook. The model developed for this study (ECOMSED) consists of three separate models that were coupled together: hydrodynamic, wind-wave and sediment transport models. The model was used to simulate a base-case scenario representative of pre-dredging and post-dredging conditions for the year 2000 to evaluate the effect of sediment removal (dredging) operations in selected areas of the Northern Shoreline and Western Cove of the lake. The results of the simulation showed that Lake Waban is a low energy environment, and seems to be mostly depositional. There is relatively little change in sediment accretion for pre-dredging conditions as compared to post-dredging conditions. Sensitivity analyses were carried out for both pre-dredging and post-dredging conditions: (1) to evaluate the impact of no sediment inflow from Upper Waban Brook, and (2) to evaluate the impact of high winds (i.e., Logan Airport winds). Model results showed that sediment resuspension from other areas of Lake Waban does not contribute to significant sediment accretion in the dredging areas. The flux of sediment from Lake Waban to Lower Waban Brook was computed to be approximately 5 – 6 % of the sediment inflow for the simulated conditions.

Contaminants can be introduced into water bodies in either dissolved or particulate form. Further reactions with natural suspended and bed sediments often produce considerable sediment contamination even when pollutants are originally emitted in dissolved form. Fine colloidal particulate matter is ubiquitous in streams, and often has a higher affinity for contaminants than larger streambed sediments. As a result, suspended sediments can carry a considerable fraction of the overall contaminant load. To understand the transport and ultimate fate of contaminants in aqueous systems, and particularly fluxes into and out of contaminated sediments, the potential for particulate-phase transport of contaminants must be considered. Contaminant behavior within sediment beds is highly dependent on the degree of association with suspended sediments, and the presence of suspended matter can favor either mobilization or immobilization of contaminants depending on the relative affinities of contaminants for suspended and bed sediments and the potential for suspended sediment deposition. Experimental results and model simulations will be used to illustrate a wide range of contaminant transport behavior and to elucidate the important processes that control multi-phase contaminant transport in sediment beds.

This study focuses on the modeling and simulation of the self-weight consolidation process of newly-sediment mud deposits occurring in a vertical sedimentation column for low stress values. It describes the derivation steps of the governing equation in an Eulerian coordinate system where material deformation is referenced to the impermeable bottom of the sedimentation column.

The U. S. Department of Interior, Bureau of Reclamation is involved in many dam decommissioning projects in the Western United States. These projects involve complete to partial dam removal. A primary factor of each project is to identify the characteristics and volume of sediment impounded by the dam and to describe the potential erosion and transport of that sediment into the downstream river channel as a result of dam decommissioning. Three Bureau of Reclamation dam decommissioning projects are: 1.) Savage Rapids Dam on the Rogue River in Oregon, 2.) Matilija Dam on the Ventura River near Ventura, California, and 3).Coleman and South Diversion Dams on Battle Creek in California. The dam decommissioning studies for each project are described and sediment characterization before and after dam removal are compared. Insights into the unique characteristics of each project are also given. Conclusions will be drawn about the dam removal projects.

To help overcome the financial burden of the engineering analyses associated with rehabilitation of privately owned dams, a unique partnership has been developed between dam owners, Rowan University, and the local engineering community. Undergraduate engineering students have been working with a faculty member to conduct hydraulic and hydrologic analyses and flood inundation mapping, as part of the Engineering Clinic program at Rowan. The results of this project are being evaluated by a team of external practicing engineers, and then submitted to the New Jersey Bureau of Dam Safety and Flood Control for regulator review. Involving students in all aspects of the dam evaluation introduces many concepts not included in traditional classroom instruction, such as the ecological benefits (and detriments) of small dams. In addition, students are exposed to the socio-economic and political realities that engineers must contend with in professional practice. Thus, having students work on a real-world analysis and design project provides an all-around win-win situation: Students work on a unique educational opportunity; the dam owner receives valuable services; State regulators are satisfied; and public safety is upheld.

With the release of HEC-RAS 3.1 (November, 2002), users can now perform dam and levee breaching analyses. This paper discusses the methodology used to perform the breaching, as well as the available options. The data required to perform such an analysis is reviewed, and available output from the model is displayed. Model stability issues are also addressed, with solutions to some of the typical problems provided. An example application of the dam break capability within HEC-RAS will be presented.

In September 2000, The Aspen Institute's Program on Energy, the Environment and the Economy invited a group of experts to engage in a two-year long dialogue about the issue of dam removal. The twenty-six experts are involved in various aspects of dam and river management, represent a full range of interests and work regularly with responsible government agencies, dam owners, nongovernmental organizations, and dam users and beneficiaries. Using an intentional, values-based method of dialogue, the group engaged in a series of eight three-day meetings over a period of two years. During this time they worked to express, inform, and integrate their values, knowledge and understanding in order to create new thinking to guide policy and practice affecting dams and rivers. The resulting report — Dam Removal: A New Option For a New Century — provides the group's recommendations and practical advice aimed at integrating the dam removal option into river management decisions, evaluating the options fairly and, if appropriate, implementing the dam removal option effectively. The imprimatur of this diverse group, with interests that are often at odds, lends a unique weight to the recommendations.

Maintaining the safe operation of thousands of aging dams in North America is a daunting task for dam owners. Whether it is the need to increase spillway capacity, solve erosion problems at outlet structures, or improve stability for seismic, sliding, and/or overturning, finding a solution that protects the public and is economical can be difficult. Of the more than 76,000 dams on the national inventory in the United States many have either exceeded their design life or will over the next decade. In addition, many dams once classified as low hazard are now being reclassified as high hazard due to development downstream of the dams necessitating a reevaluation of the dam's overall safety. Roller compacted concrete (RCC) has now been in use in rehabilitating dams for over 22 years with over 200 projects completed in North America. The economics of using RCC for rehabilitating dams is well documented and its performance proven under some of the most severe climatic and hydraulic loading conditions. Seventy four embankment dams have had overtopping protection provided with RCC to provide the safe passage of extreme flood events, aging concrete and masonry dams have used RCC as a buttress material to provide additional stability particularity during the design seismic event, and RCC has been used to extend or reinforce stilling basins subjected to high velocity and turbulent flows.

The coordinated flushing at Dashidaira and Unazuki reservoirs on the Kurobe River in 2001 and 2002 and sediment flushing of the Lower Yellow River in July 2002 are introduced in this study. The differences between the two types of flushing and conventional flushing are analyzed. The environmental impacts and relevant mitigation measures of the sediment flushing are discussed. The coordinated flushing has less impact on the downstream environment compared with the flushing at only one dam. The sediment flushing of the Yellow River performed in July 2002 is the first flushing experiment in the world to wash away the sediment deposited in the riverbed. The measured results show deposits of 36.2 million ton was scoured from the bed of the Lower Yellow River. The flushing of the Yellow River has no severe negative impact on the downstream aquatic lives. The flushing of the Yellow River expands the concept the sediment flushing and the concerted flushing at Dashidaira and Unazuki dams shows an environmentally friendly flushing example. The major difference between the flushing at a reservoir and the flushing of a riverbed is how to manage the reservoir operation.

The one-dimensional unsteady flow model UNET was used to study the propagation of flood waves in the lower Illinois River from Peoria to Grafton. The calibrated and verified model was used to simulate water surface elevations along the Illinois River for existing conditions and for managed Levee and Drainage District (LDD) storage options. To further expand the applications of the lower Illinois River UNET model, the model was extended to include Pool 26 of the Mississippi River to resolve the difficulties in handling the backwater effects from the Mississippi River. This model was used to study different managed storage options to develop systematically organized information and tools for managing floods in the lower Illinois River. It was assumed that spillways will be placed on top of the levees to simulate managed flow into selected LDD storage areas along the lower Illinois River. Two LDDs of difference sizes and locations were selected to study the relationship between the peak flood stage reduction and the sizes of the inflow section of the spillways. The study was based on the maximum water surface elevation of a 50-year frequency flood. Study of the optimal spillway widths for fixed depths and optimal spillway depths for fixed widths indicated that the optimal spillway dimensions for peak flood stage reduction vary according to locations and individual flood events.

The sediment management at Naodehai reservoir on the Liu River in China is discussed in this study. The Naodehai reservoir was operated as an uncontrolled flood detention reservoir from 1942 to 1970. It has been operating in such a scheme as storing clear water in dry season and sluicing muddy water at free flow or low water level in flood season since 1971. The equilibrium status of erosion and deposition in the reservoir was reached in early 1970s. The process of the sediment movement in the reservoir is erosion in the flood season and deposition in dry season since 1971. This mode of operation has enabled the reservoir to maintain 80% of its original storage capacity after 60 years of operation. The major negative impact of the reservoir operation mode is silt deposition in the downstream channel. The downstream deposition causes the riverbed rising and decreases the flood discharging capacity of the river channel. The measures such as water and soil conservation in the upper basin, revetments and groynes in the downstream river channel have been executed to reduce the deposition.

This paper presents seven activities that were implemented in a senior level, undergraduate course in Environmental Engineering to bring reality into the classroom. Each of the activities will be discussed along with student feedback and an assessment of the use of the seven activities will be presented.

The following is a summary of design projects that have been included in a Junior Level Water Resources Engineering course that is required for all Civil and Environmental Engineering majors. The projects are structured to encompass the fundamentals of applied fluid mechanics, while at the same time, providing the students an opportunity to be creative and bring water resources engineering "to life." The students (working in groups of 3) designed a water feature (fountain) and a golf course water system for the proposed resort, "Le Reve", on the Las Vegas strip. A field trip to the Bellagio resort provided students the opportunity to see the components of a world class water feature. The students acquired information from local sources such as water requirements from the Las Vegas Valley Water District, sprinkler and nozzle capacity from manufacturers, and evaporation rates from the National Weather Service. The phasing of the project throughout the semester provided each student an opportunity to be "Lead Engineer." The Lead Engineer responsibilities included the delegation of work on each Phase, and preparation of the progress report. The public domain software, EPANet, was the tool used for analysis of the water distribution systems and ensuring that all pressure requirements were met. Course assessments show improved student knowledge in the design of water resource systems. Lastly, the incorporation of "water themed" music into the classroom creates an active learning environment in the classroom.

This paper will outline an alternative to the cookbook approach to conducting the undergraduate laboratory experience. The concept relies on the undergraduate students, working in teams, to design and conduct experiments. The instructor acts as a mentor and resource during the process, and insures quality of the laboratory exercise without having to be the sole responsible party. By taking ownership of the experiment, the students' learning is increased and the laboratory experience is enhanced. This is due, in part, to experiments being tailored to the students' needs and interests which is a result of their direct involvement. Finally, concepts of team building and experimental design are introduced. Surveys and testing of students involved in this approach suggest a strong positive outcome. Objective outcome assessments of participating groups over several years of using this approach have defined pitfalls to be avoided by the instructor.

Educational experiences that relate social and technical subjects offer students the opportunity to reflect on the broad significance of environmental engineering. Although it is often difficult to join social and technical subjects in a classroom setting, the link may be more easily made through service learning projects in which students use classroom knowledge and hands-on service to implement solutions for a given community. As an example, a senior civil and environmental engineering student project to design a sanitary sewer for an in-need community in San Benito, Guatemala, is described. Students traveled to the site, performed a land survey, and gathered other design data while also learning about Guatemalan history and culture. The students apply knowledge from required courses and the humanities/social science class "Latin American Health, Infrastructure, and Environment" to arrive at an appropriate final design. The international design project is described as an approach to increase student appreciation of the engineering profession and support educational goals, such as increased understanding of engineering solutions in a societal context, and the ability to function on a multidisciplinary team.

A study was conducted to determine how well permittees of publicly owned treatment works in the State of Georgia are meeting the pathogen reduction criteria and vector attraction reduction criteria for biosolids being land applied. Based on reports submitted to the Environmental Protection Division, all 31 facilities evaluated met the vector attraction reduction (VAR) criteria and 25 were in full compliance with the pathogen reduction criteria.

Nitric oxide (NO) from abiotic and biotic processes is examined through the experimental measurement of NO flux from autoclaved and non-sterilized garden soil sampled from Sarah B. Duke Gardens, Durham, North Carolina. The biological nitrifier activity is examined using nitrapyrin treatment as an autotrophic nitrification inhibitor. NO emissions comprise an important fraction of unaccounted nitrogen movement from soil to lower levels of the troposphere. Particularly, agricultural soils are a major contributor of atmospheric nitrogen in the form of NO in rural areas. In several ways NO is a missing link for understanding nitrogen cycling globally, as NO is an important constituent in various global change models including ozone formation and accumulation in the lower troposphere. Results from this study indicate microbial nitrifier activity in aerobic garden soil as the primary source of NO. Mean NO emissions were reduced from 7.23 to 1.16 ng N/m²s, 2.7 to 0.63 ng N/m²s, and 4.18 to 0.47 ng N/m²s following nitrapyrin treatment of non-amended, water amended, and waste amended soil, respectively. Preliminary findings also suggest denitrifier activity as a negligible contributor to NO flux based on the lack of response following nitrogen/waste addition to nitrapyrin treated soil as compared with non-treated test soil. Further research should investigate the role of these contributors with respect to changing soil parameters and ambient conditions.

The State of Connecticut is in the process of reviewing their instream flow standards. There are serious proposals to increase instream flow standards to levels which will cause serious conflicts between nearly all surface water withdrawals and requirements to maintain instream flows. Recent proposals for instream flows could easily reduce the reliable supply from surface water reservoirs by 60% or more. The presentation will describe recent developments in the process and describe a rationale approach to balancing environmental impacts and social and economic needs for clean and reliable water supply.

As one of the sixteen regions in the Texas state water plan (Senate Bill 1), the Far West Texas Region, which is comprised of seven counties, engaged in a "bottom up" approach to develop a 50-year drought-contingency and water-supply management plan based on the general consensus of all the parties responsible for water resources management in the region. Work on the Far West Texas Regional Water Plan (FWTRWP) was originally divided between two teams: urban and rural, and water planning was conducted simultaneously because of their different interests and potential conflicts. This paper presents two major findings in terms of conflicts and opportunities for cooperation between the urban and rural teams in water resources planning in the Region. One concept is the groundwater transfer from rural areas into the larger urban area of El Paso. The hydrologic challenge is to assess whether groundwater pumping can be sustained by those aquifers, what impact will be imposed on those aquifers by the groundwater transfer mechanism, and what level of impact is accepted. The other issue to be addressed is the fact that the Far West Texas Region is located on the Chihuahuan Desert, and continuously experiences prolonged periods of drought according to traditional precipitation standards. Because the State Water Plan requires that supplies be identified to meet demands for the drought of record, the challenge is to determine what parameters should be used for drought trigger criteria to initiate drought contingences. With the cooperation of all the affected parties, a consensus in the general concept for water resources sharing was reached and reflected in the final regional water plan. However, some conflicts still remain unresolved, and should be addressed in future water resources plans.

Sanitary sewer overflows (SSO) are becoming of increasing concern to utilities and regulators as they pose a health risk. However, many overflows are not easily identifiable. A methodology is presented for detecting if a disruption in the system is occurring. The approach links neural networks as a prediction tool for expected flows and control limit theory for identifying significant deviations from the expected values that suggest an SSO occurrence. Detection depends upon the magnitude of the disruption, the relative distance between the disruption and the gage, the ability to analyze large quantities of monitoring data and the capacity of the developed methodology to compare real-time data with expected conditions. Data from Pima County Wastewater Management's monitoring system is used as a case study.

The Florida Department of Environmental Protection (FDEP) recently adopted the Ultraviolet Disinfection Guidelines for Drinking Water and Water Reuse, published in December 2000 by the National Water Research Institute / American Water Works Association Research Foundation (NWRI/AWWARF). These guidelines now serve as the design basis for new stand-alone, high-level UV disinfection systems at wastewater treatment facilities in Florida. The NWRI/AWWARF guidelines establish minimum design requirements for UV disinfection of filtered reclaimed water. An integral part of the guidelines is performance-based testing in which the dose delivery capability of the UV reactor is experimentally validated. The City of Auburndale's Raymond Allred Wastewater Treatment Plant (WWTP) recently became the first facility in Florida permitted under the new guidelines. Several issues related to UV dosage, reactor and reliability design, monitoring requirements, and manufacturers' validation studies came to light as a result of the FDEP permitting process. Unlike California, the FDEP does not "conditionally accept" specific models of UV systems. Rather, Florida approves a system on an individual basis and requires the engineer of record to provide assurance that the system design and validation comply with the NWRI/AWWARF guidelines. Key aspects of the NWRI/AWWARF guidelines and validation requirements will be presented using Auburndale's permitting experience as a "case study".

The Surface Water Treatment Rule emphasizes the need for utilities to meet minimum levels of disinfection for surface waters, whereas the Disinfection-Disinfection By-Products Rule limits the disinfectant by-products levels. Therefore, utilities will have to implement a treatment approach that balances the benefits of disinfection against disinfection by-products. After the Crptosporidium outbreak in Milwaukee during March 1993, many utilities began immediately investigating the use of more advanced disinfectants such as ozone and chlorine dioxide in order to combat the threat from the Cryptosporidium protozoan. Chlorine alone was not adequate to inactivate the new target organism. Most disinfectants have their consequences in terms of disinfection by-products causing a potentially long-term adverse health impact from cancer. Although chlorine has served the water industry well for about a century in safeguarding the public health, it is relatively ineffective against Cryptosporidium and can cause excessive trihalomethanes in distribution systems. THMs were limited in drinking water in 1979 to 100 ppb (THM Rule) and are being lowered to 80 ppb to meet impending regulations because of their potential carcinogenic properties.

Tradable Discharge Permit (TDP) programs have shown, both in practice and in theory, to have tremendous potential as a cost-effective method of pollution control. Nevertheless, there are still many uncertainties regarding TDP programs that if not adequately addressed, might impair its success. These uncertainties, of which some are quantifiable but most are not, are as random as they are diverse. Concerns range from issues of market failure that prevents optimal trading, to political agendas that differ from a typical TDP program in their priorities, to modeling difficulties that might cause erroneous predictions of cost savings and environmental performance. The authors of this study recognize the hopelessness of trying to overcome these concerns all at once. And therefore, apart from a brief introduction where the more common of these issues are identified and discussed, attention is focused only on the ones that are less abstract and more tangible. More specifically, for the time being, uncertainties of trading, politics and such are assumed negligible and only modeling uncertainties are considered. Numerous studies have been carried out to predict the potential impacts of TDP programs, whether positive or negative, on the environment they are intended to protect. These studies have been invaluable in laying essential groundwork for the further understanding and actual implementation of such programs. However, many of these studies assumed deterministic environmental models when in reality nothing is ever constant. The environment is an open system vulnerable to, amongst many other agents, weather variations and changes in microbial behavior. It is therefore, this study's goal to attempt to advance a step forward by re-assessing those same questions asked many times before, but this time without disregarding the stochastic nature of the environment.

EPA requirements minimize the risk of infectious disease from the beneficial use of more than 4 million dry metric tons of sewage sludge per year by reducing its pathogenic microorganism content to below the analytical detection limit. Disinfection is necessitated by the fact that the wastewater originates in a community with human and animal fecal discharges. Federally specified PFRP treatments/measures are known to reduce pathogen levels below the detection limit. Other processes may be able to achieve the same levels of pathogen reduction as a PFRP, but before they can be employed at a specific site or nationally must be shown equivalent. This paper discusses in detail the equivalency determination process, notes those processes which have been deemed equivalent and are currently in use, and notes some of the pending applications for equivalency illustrating new and emerging approaches to sludge disinfection. Mention is also made of how the equivalency process might be applied to stored or stockpiled material.

Significant growth has occurred in the metropolitan Atlanta region within the past several decades. Water quality downstream from development has been impacted by both point and non point discharges resulting in degradation of receiving streams. Implementation of Total Maximum Daily Load (TMDL) and watershed protection strategies associated with both point source discharge permits and MS4 stormwater discharge permits have brought additional attention to deteriorating receiving water quality. Over the past several years, Northwest Georgia has undergone a sustained drought of record, during which new studies have indicated that the metropolitan area's water supplies are not inexhaustible. Further, long standing three state water compact negotiations will dictate a limit on supply.

The Surface Mining Control and Reclamation Act of 1977 (SMCRA) established the nationwide program to protect society and the environment from the adverse impacts of coal mining. It was promulgated because there was a need to have standards to minimize the adverse impacts to the environment from coal mining while still providing for the nations need for coal as an energy source. SMCRA recognizes that economically viable coal mining is necessary to meet these energy requirements and that there will be impacts from these operations. However, the surface and underground methods used to mine coal can be designed to guarantee environmental protection, ensure mined-lands are restored to productive uses, and protect the public from danger. Prior to the promulgation of SMCRA coal mining left unreclaimed lands which caused water pollution, erosion, floods, slope failures, loss of fish and wildlife resources and dangerous highwalls. Today SMCRA requires the coal mine operations submit permits and plans prior to any coal mining disturbances. The permits have to include information about the permit and adjacent area, the mining operation plan, the reclamation plan, and predictions of the impacts to the environment. These permits are evaluated to ensure the mine plan is feasible, reclamation can be assured and the impacts to the environment have been minimized.

Methyl tertiary-butyl ether (MTBE) is used as a chemical additive (oxygenate) to gasoline. In 1996 MTBE use was increased when a federal mandate required low-polluting gasoline in states with air quality problems such as California. The rule requires the gasoline to contain a certain amount of oxygen-enhancing compounds, and various oil companies selected MTBE as an economical choice. South Lake Tahoe is one of the 50 California communities that have suffered water quality impacts to their groundwater wells due to MTBE contamination from gasoline spills, leaking tanks, and piping at distribution facilities. In 1999, Governor Gray Davis declared MTBE to be a "significant risk" to the environment and ordered its removal from California gasoline by the end of 2002. However, in March 2002, Governor Davis delayed the ban until January 1, 2004 in order to allow more time for construction of transportation and fuel-blending facilities needed to replace MTBE with ethanol, which is produced mostly from corn grown in the Midwest.

To design a sanitary sewer or treatment plant the engineer is required to estimate the daily flow and the peak flow expected in the system. To do this, an estimate of the dry weather flow and the wet weather flow is needed. To produce the most reliable estimates, measured data should be used in determining these flow rates. A procedure is developed to stochastically separate and model the dry weather flow, inflow, infiltration, and ground-water infiltration using a measured flow record of sewers and rainfall in the area. This model can then be used to estimate the inflow and infiltration for a given storm or design frequency. The modeling approach used, disaggregates the sanitary sewer flow record into its components. The sequence of desegregation is the dry weather flow, then the wet weather flow. The dry weather flow component of the model is a periodic model plus the mean daily flow. This type of model is used since the variation of wastewater flow is a function of the daily routines of the population, which is based on the diurnal cycle. A Fourier analysis is utilized to determine the predominant frequencies in the flow record to model the dry weather flow. The wet weather flow model estimates the inflow and infiltration separately. Since inflow and infiltration are attributed to different defects, a hypothesis can be made of the types of defects in the system with them separated. The inflow and infiltration use the same modeling technique, each having a different lag time estimated by the correlogram of the system. The flow response from an individual precipitation impulse is modeled using an autoregressive model for the rising and receding limbs of the response independently.

In order to improve water quality from point source discharges, the United States Environmental Protection Agency (EPA) is proposing comprehensive strategies in the new Sanitary Sewer Overflow (SSO) rule. The SSO rule will require owners of the sanitary collection system to exercise decision-making skills so that the utility, agency or municipality will be in the best position possible to comply with the rule. These strategies will have significant impacts in regards to resources required from the local, state and federal governments and citizens. This paper will address the various aspects of the SSO rule as it affects Metropolitan St Louis Sewer District (MSD) and possibly other agencies, operators and municipalities. Another paper in the US Congress will focus on a new tool for estimating rainfall dependent inflow and infiltration (RDII). The method employed is a statistical model approach to predict RDII at any point in a collection system when applied to the principal sewershed characteristics.

In 1993, the Environmental Protection Agency issued the 503 Sludge Regulations which brought increased focus and attention to residuals management. Methods of treatment and disposal are continually reviewed and revised to meet the changing needs of wastewater utilities. Aerobic digestion has been a common means of meeting the 503 regulations, especially for facilities without primary clarifiers. However, aerobic digestion typically produces a sludge that is more difficult to dewater. The City of Winter Haven, Florida, owns and operates Wastewater Plant No. 3, a 5 million gallon per day wastewater treatment plant. The City constructed modifications to the Facility in 2002 which included the redesign of the existing two-stage aerobic digester to an anoxic/oxic digester. The design considered the stabilization performance as well as conditioning and operational characteristics of anoxic/oxic versus aerobic digested sludge. Initial results indicate that the system is providing good quality sludge with conditioning that yields a dewaterable product. This paper reviews the key concepts behind the design and compares performance indicators.

Detailed flow and rainfall measurements, accompanied by long-term simulation, were used to identify the inflow and infiltration response characteristics of the sanitary sewer collection system in Vallejo CA. This response has produced sanitary sewer overflows on numerous occasions throughout various locations in the collection system. A mix of collection system rehabilitation, capacity upgrades, storage and increased treatment capacity may be used to control future wet-weather flows to a regulatory standard. A major limitation to optimizing the design of these wet-weather controls is a relatively high uncertainty regarding the effectiveness of collection system rehabilitation to control wet-weather flows. The ability of collection system rehabilitation activities to control wet-weather flows will in turn affect the performance of all downstream controls, including conveyance, storage and treatment. If rehabilitation effectiveness is over-estimated, overall control performance will not meet design standards, and if underestimated, significant over-expenditure of resources is possible. A risk-based approach was used to identify the importance of rehabilitation effectiveness on the overall design. Probability density functions of wet-weather pollutant control for storage, treatment and rehabilitation were derived from direct observations and used to estimate the overall reliability of various mixes of design alternatives. Based on this analysis, a concentrated small-scale rehabilitation project was used to reduce the uncertainty associated with estimating the performance of rehabilitation for sanitary sewer overflow (SSO) control.

An optimization model for river water quality management is presented in this study to provide the monthly treatment policies for point sources. The dimensionality of the problem is tackled by developing a Sequential Dynamic Genetic Algorithm (SDGA). This optimization model is linked with an unsteady water quality simulation model and minimizes the total losses in the optimization time horizon including the treatment cost and the cost associated with the violation from water quality standards. The proposed model has been applied to the water quality management of Karoon River in Iran. The results show how effective the model is for seasonal waste load allocation and the significant values of seasonal policies for quantitative management of the river.

Significant Water Issues in Georgia (graphics and charts only).

The Los Angeles River watershed covers 824 square miles, of which, approximately 364 square miles are covered by forest or open space land including the area near the headwaters, which originate in the Santa Monica, Santa Susana, and San Gabriel Mountains. The remaining area is highly urbanized including the San Fernando Valley and portions of the City of Los Angeles. Major flood events at the turn of the century initiated flood control projects in the watershed that led to dramatic alterations of the natural hydrology of the river. Presently, most of the Los Angeles River, and its major tributaries, are concrete lined with numerous dams, spreading grounds, and other hydraulic control structures. The development of the Los Angeles River nutrient and pathogen Total Maximum Daily Loads (TMDL) required that the full range of pollutants, sources, and flow conditions, typical of heavily urbanized areas, be addressed for a single water body. Although the climatic issues are specific to the southern west coast region, the processes and types of issues encountered can be viewed as typical for urban TMDLs across the country. The development of modeling tools for use in simulation of the water quality conditions, and ultimately for use in determination of allowable loadings, needed to address many issues typical of urban areas.

Recent developments in water resource management include efforts to use the resource in a sustainable manner. Sustainable water use supports social objectives into the indefinite future without undermining environmental and hydrologic integrity. Key aspects are incorporation of mechanisms to insure ecological viability and recognition of the economic value of water. California is a wealthy state, with a complex mix of rules and infrastructure to move water to areas of demand deficit. Many mechanisms exist to allocate water and protect river ecology. There is continued degradation of river habitat due to an imbalance in water allocation. Markets mechanisms have been slowly introduced and are advocated to reallocate water to higher valued uses and the environment. There is widespread support for environmental concerns in the state, reflected in a willingness to pay for environmental water, water use behavior change, and legislative mandates. The CALFED Bay-Delta program and Central Valley Project Improvement Act include provisions to better reach a balance between the environment and other uses, including water transfers for instream flows using public and private money. Most water transactions are subject to environmental impact assessment, with most involving public agencies due to water passage through the Delta and/or wheeling through federal or state project infrastructure. Water allocation and distribution management in Chile is characterized by broad private-sector involvement based on historical participation in agricultural communities and reinforced in the Water Code of 1981, which encourages market transactions to redistribute water.

Dublin principles and Rio + 10 at the international levels besides Bolivarian Venezuela's Constitution and future water law, at the national level, indicate that water has an economic value. To understand such precept will help deciding about water resource use. We need to decide about water structural works and water political issues. There are many methods to estimate water economic values. Which to apply will depend on audience, water value significance and time and money to do the study. Delphi method can be use to determine water value significance. In this sense, Delphi method was used to determine water economic value at the Chama River upper watershed. After validating the questionnaire, two rounds of the Delphi method were applied at 19 "water resources experts". The experts gave the hierarchy of the use and non use water values. Urban an irrigation value were high in 4 out 12 units studied. Option value was high in 3 units and tourism and recreational values were high in 1. When the upper watershed was considered as a unity, indirect use of water (habitat) was the highest. Urban, recreational and tourism were the following values.

The Huntington District of the Corps of Engineers is working to develop better operating strategies for lakes in the Muskingum River Basin in order to respond to the conflicting desires of multiple stakeholders in the basin, including flood control, water supply, and water quality. The Basin includes sixteen lakes and thirteen downstream gage controls. A modeling tool was developed using STELLA software to assist the Corps in quickly and easily evaluating operating scenarios and investigating impacts to the various stakeholders. A post-processor has also been developed that calculates performance statistics for lake storage and instream flow. This post-processor will calculate the required lake storage allocation for a user-defined operating alternative. The Muskingum Basin has several conflicting uses for water and this model is being used to better address the needs of all stakeholders.

Water quality regulations continue to evolve as new research and treatment technologies are developed. Recent regulatory changes focused on control of disinfection by-products (DBPs) have affected the operating strategy for many water treatment plants. Historically, treatment plants focused disinfection and turbidity removal practices to conservatively provide for pathogen inactivation. However, the interaction of disinfectants with naturally occurring organic matter in the source water resulted in formation of DBPs such as trihalomethanes and haloacetic acids in the finished water. These DBPs are suspected to pose a health hazard in drinking water and the Stage I Disinfectant/Disinfection By-Product Rule was put into effect as of January 1, 2002 to limit the levels of DBPs in the distribution system, as well as to require increased removal of DBP precursors (measured as TOC) through the treatment process. This paper describes how the City of Wilmington, Delaware proactively planned for and successfully implemented low cost equipment, training and operational changes at its two treatment plants to better comply with the new regulations while still balancing the need for adequate turbidity and pathogen removal.

Historically, humans have controlled rivers for various applications such as transportation, hydro-power generation, flood control, and water supply, including agricultural irrigation. Though these activities proved to be economically beneficial, the altering of natural riverine systems resulted in negative affects on riparian ecosystems' natural processes. The continued trend of river modification has resulted in damage to the biological integrity of many river ecosystems. Researchers have shown that natural streamflow variability over the full range of flows—from drought to flood flows—is critical to ecosystem dynamics. The variability of the flow regime, characterized by magnitude, frequency, duration, timing, and rate of change of hydrologic conditions, has been identified as integral in the sustaining of a natural ecosystem. Thus, the prevention of significant alteration to ecosystems needs to address a full regime of flow conditions. While progress is being made in assessing ecological integrity of riverine systems, a significant gap exists in the application of these advances into a watershed management setting.

Recently, a renewed interest in risk analysis in the potable water supply industry has been brought about by an increasingly stringent regulatory environment and a more acute sense of the consequences of failure events. The analysis of reliability, both of system mechanics and of system operations, is a key part of risk analysis methodologies. A wide variety of reliability analysis methodologies are currently available to help utilities quantify the probability of system and sub-system failure. In order to truly minimize the risk of system failure, however, utilities need to be able to not only quantify the probability of failure, but also to determine the operational conditions that are the most likely to impact the probability of failure. The conditional reliability approach introduced in this paper addresses both of these needs and therefore offers a useful alternative to traditional probabilistic reliability analyses. The utility of the approach is demonstrated through several case studies using data from North American systems.

With the recent focus on lowering the operating costs of environmental remediation systems, including groundwater pump and treat systems, there has been increased interest in algorithmic approaches to optimization. The Department of Defense and United States Environmental Protection Agency (USEPA) sponsored a study to evaluate the benefits and utility of transport optimization algorithms, operable on desktop computers, against trial and error approaches for pump and treat system design. This follows upon an earlier USEPA effort to determine the efficacy of a hydraulic optimization algorithm designed for MODFLOW. Mathematical optimization problems are formulated to minimize an objective and satisfy a set of constraints. Hydraulic optimization typically applies to plume containment problems, while transport optimization typically applies to problems associated with reduction of contaminant concentrations or contaminant mass. Transport optimization can consider life-cycle costs and cleanup time as components of the objective or as part of the constraints. Application of transport optimization could potentially reduce the costs for pump and treat systems by millions of dollars over the life of the projects, as this study has demonstrated. Note that these types of optimization tools have previously been (and currently are being) applied to Air Force projects.

The chlorination of drinking water distribution networks is usually carried out at the supply source. However, chlorine may disappear in some portions or at distant points within a network. In this case, one or more booster chlorination stations must be built in the network in order to observe detectable chlorine residual at all levels of branching. Network hydraulic values, tank water levels and chlorine concentrations may vary over the course of one day because of changes in consumer demand. For this reason, the optimal location of a booster chlorination station, injection rates and scheduling must be considered together. In this research, the locations, injection rates and scheduling of chlorine booster stations were studied using genetic algorithms. The results indicate that booster disinfection can significantly increase the desired residual concentrations above the minimum limit while helping to reduce variability in nodal concentrations. The objectives were satisfied with a small increase in chlorine consumption compared to conventional disinfection only at the source. In order to find a hydraulic solution and chlorine concentration distribution in a network, EPANET software was used. In the solution phase, genetic algorithms and EPANET software were run interactively. The algorithm developed was used on an existing network given in the literature and solutions were compared with the current status of the network.

There exists much scientific evidence showing that our planet is undergoing a global climate change and that the climate change will threaten the status quo of regional water resources that support ecosystem and human society. An assessment was conducted to evaluate the vulnerability of water resources to climate change in the Mackinaw watershed in central Illinois, USA. In this study, the SWAT (Soil and Water Assessment Tool) model was used to predict stream flow and crop yields in a watershed under different future climate scenarios that represent possible climate outcomes predicted by well-known global circulation models. The potential effects of climate change and irrigation incurred by the change, if any, on low-streamflow and profitability of crop growing operations were investigated, for currently customary crops for the region.

This paper develops a Bayesian approach to Generalized Least Squares regression allowing computation of the posterior distributions of the parameters and the model-error variance using a quasi-analytic analysis and Markov Chain Monte Carlo simulation. Bayesian GLS analysis is illustrated with two regional skew estimation examples. The examples demonstrate that regional skew models can be relatively precise with effective record lengths in excess of 50 years. The marginal posterior distribution of the model error variance and the mean and variance of the parameters can be computed directly thereby providing a simple but important extension of the regional GLS procedures popularized by Tasker and Stedinger.

Total Maximum Daily Load (TMDL) studies often require several years of continuous flow data for model calibration. Unfortunately, historical flow information is highly limited for some watersheds. Synthetic flow generation methods could be used to extend the available flow records. This research tests an artificial neural network (ANN) model for synthetic flow generation by using a small observation set of 122 daily flows randomly selected over a 4-year period. To generate multiple training and testing data sets, a bootstrap method is used to sample with replacement the available data. Thus, all available data is used to train and validate the model. Determination of an effective split in the data set between training and testing data is not required, which is especially important when the amount of available observations is limited. Daily flows from three nearby flow gauges and precipitation are the potential inputs for the ANN model. The ANN model structure that results in the lowest bootstrap prediction error is chosen as the best model for the available data. The bootstrap method also provides a prediction interval for each flow prediction. The Levenberg-Marquardt algorithm is used to train these ANN models. For flows within the range of observations, the ANN model provides good predictions. However, the ANN model does not extrapolate well; flow predictions above the highest observed flow are not accurate. These high flows are replaced by the results from a maintenance of variance (MOVE) model. When input data are selected over a 4-year period, the combined ANN and MOVE model performs better than MOVE alone. This study shows that, even with a small data sample, the combined ANN and MOVE model can be used to create an accurate synthetic flow series, and the bootstrap method can be used to guide the model selection and generate reasonable prediction intervals.

The most common mathematical problem related to a water distribution network is not the design process but the calibration process of an existent network due to the leaks presence. The calibration process gives a model that offers a functional tool to do, among others things: to detect the problem areas in the network, to simulate possible operational and maintenance changes to reach possible solutions to the quality problems of the water received by the final users, to detect high leak concentrations and to test optimal operational pressure methods. None of these tasks, used in a program of integral management of potable water distribution network can be implemented without the calibrated model. This article summarizes most of the techniques as experiences of CIACUA (Water Supply and Sewerage Systems Research Center at Universidad de los Andes, Bogotá, Colombia) in solving the calibration problem of water networks under losses environment, in particular with genetic algorithms. The genetic algorithms have proved its effectiveness at the time of developing the calibration of a potable water distribution network that operates under a leaks environment. However, this kind of algorithms requires a lot of iterations, which makes a slow process. The article process the development of new methods to make use of the best characteristics of each one of the artificial intelligence available tools instead of generating a new algorithm more effective than the ones used currently. In the calibration process the values of some measured field variables are compared between them for which certain pipe characteristics are adjusted, such as, roughness, diameters, coefficients of minor losses, leaks, etc. Each change that is carried out upon a characteristic is done upon specific group, not upon an individual pipe; this procedure is made in order to keep the hydraulic criteria and to do not arrive with meaningless physical solutions; also each variable is related to a probabilistic distribution determined by some criteria. In this article is shown the methodology applied in a Bogotá area where satisfactory results were obtained.

Large and complex water supply and irrigation canal networks exist in arid regions of the United States, as well as in many arid locations in other countries. It is necessary to establish management strategies for the operation of a water supply and irrigation canal network to meet the water demands at various locations in the network over time without causing problems such as water shortage or flooding in the areas served by the network. A previous study by El Gamel and Harrell demonstrated that by incorporating an unsteady flow model into a genetic algorithm-(GA-) based framework, efficient strategies for scheduling gate operations and allocating crops can be determined to address the various objectives (such as profits from crops and costs of damages due to flooding and water shortages) and constraints (such as maintaining stability of gates and adequate water levels at water supply intakes) over a period of time using deterministic models. However, the actual values of many of the input parameters for this type of problem are not known with certainty due to variability over space and time, as well as lack of information. The GA framework is flexible to incorporate estimates of reliability to search for efficient solutions that satisfy the constraints with a specified likelihood under conditions of uncertainty. Because a GA evaluates the fitness of each potential solution, it is a straightforward extension to evaluate each solution repeatedly using a set of Monte Carlo (MC) realizations of uncertain parameters generated based upon their probability distributions.

Chlorine disinfectants can dissipate dramatically within distribution systems. The purpose of this study is to examine the loss of free chlorine to reactions with concrete pipe surfaces. The distribution system examined in this study is owned by the Central Lake County Joint Action Water Agency (CLCJAWA), a water distributor with 16 discrete delivery points supplying water to several suburban communities north of Chicago. Field measurements taken from the distribution system included free and total chlorine residual, pressure, and flow rates measured at the treatment plant and the 16 water delivery points. The bulk water reactivity of the chlorine was determined using batch tests. A hydraulic and water quality model of the distribution system was established and calibrated for the specific days when the measurements were taken from the distribution system. The bulk reactivity coefficient derived from the batch tests was input to the model and the wall reactivity was calibrated to fit the measured data. The results of this study showed that an average of 23 percent of the free chlorine loss could be attributed to wall reactions with concrete pipes. Since the finished water is very low in organics, the wall reactions can be attributed directly to the concrete materials. Concrete is a composite material that contains a high fraction of aluminates and silicates. Aluminates and silicates, over time, can form acidified surfaces by splitting water, supplying a free proton that can catalyze acid-sensitive reactions. Free chlorine exists as HOCl or OCl–, depending on the pH, with a pKa of 7.5. Because the pH is buffered much higher than 7.5, OCl– is the dominant form of free chlorine in the CLCJAWA distribution system. If the concrete surfaces donate protons, OCl– would shift to the more reactive HOCl, explaining the accelerated loss of free chlorine at the concrete wall surfaces.

This paper describes the approach taken in a study exploring water resources impacts associated with climate change scenarios in the Snake River basin in the Pacific Northwest. The Snake River is an extremely important river within the Columbia River basin that supports significant agricultural activity. Recently, a series of dams on this river have been targeted for removal because of their impact on salmon. Transient climate ensembles are used as inputs to the Variable Infiltration Capacity (VIC) macroscale hydrologic model to predict daily transient streamflow scenarios throughout the basin to 2050. Water resource simulation models are used to predict, on a monthly time-step, the implications of the altered climates on changes in streamflow timing and volume in approximately 91.5% of the basin's 17.9 BCM of total storage. Preliminary results from these models indicate that both short and long-term climate change will have important implications for future policy and operational decisions in the region. Of particular interest are the effects of increasingly stringent river outflow regulation on irrigation deliveries, fish endangerment, hydropower generation, and flood control. Early analysis indicates that the predicted changes in magnitude and timing of future streamflows under climate change conditions will impact reliability in these areas. To address these concerns, a series of operational alternatives and optimizations are developed that mitigate future agricultural, economic, hydropower, and environmental impacts on the system.

Genetic Algorithms have been successfully applied to numerous water resources problems, including problems with multiple objectives or uncertainty (noise). GAs tackle multi-objective optimization by following three basic principles — advancing the non-dominated frontier; maintaining diversity in the population (through various techniques like sharing, niching, and crowding); and using an elitist. However finding Pareto-optimal solutions becomes complicated when we add uncertainty to the problem. It was found that the solutions obtained using existing multi-objective solvers, although Pareto optimal were not the most robust or reliable solutions. In single-objective problems noise has typically been dealt with using Monte-Carlo-type sampling and some form of aggregate statistics (e.g., the average of the sample fitness). With multiple objectives the noise can interfere in determining non-domination of individuals, diversity preservation, and elitism (the three basic steps in multi-objective optimization). This paper proposes and tests several approaches to tackling some of these problems. These approaches strike a balance between finding the most optimal and the most reliable solution to the problem, thus giving decision makers and designers a practical and robust optimization tool.

The Water Infrastructure Network Coalition concluded that over the next twenty years the water supply and wastewater treatment infrastructure would need approximately $1.9 trillion to maintain current service performance levels. The implications to the water supply and wastewater treatment services are to lower service performance levels and/or increase service rates during this period of locality resource scarcity, limited infrastructure funding, and increasing capacity expansion needs to meet the projected demand requirements over the next twenty years. The purpose of this paper is to compare and contrast the old supply-side at least cost municipal sanitation services (MSS) planning and management paradigm with the new demand-management at full cost integrated municipal sanitation systems (IMSS) planning and management paradigm. The objectives of the paper are to: 1) expose the failures of the MSS planning and management paradigm when the water supply, wastewater/sewage, and solid waste management functions are managed as disjoint services with localized service goals and objectives; 2) promote the advantages of the new IMSS planning and management paradigm when the water supply, wastewater/sewage, and solid waste management functions are managed as a single system with integrated systems goals and objectives; and 3) propose the matrix organizational structure with flexible interrelating Architecture component linkages as a vehicle for the new IMSS planning and management paradigm.

A computational fluid dynamics (CFD) model has been applied to the study of flow within several hydraulic structures and water treatment facilities. In particular, the model has been successfully applied to the hydraulic design of clearwells, especially for optimizing chlorine contact times. The model has also been applied to optimize water quality by improving disinfectant residual in water storage reservoirs. In some of the applications, the model was favorably compared to measured experimental and field data. Currently, the model is being used in lieu of experimental testing for the hydraulic optimization of various hydraulic structures including clearwells, storage reservoirs, and pump wetwells. Due to limited space requirements, this paper addresses the application of the model to one clearwell in Palmdale, California. At the symposium, additional examples will be provided.

Denver Water (DW) serves over 1.1 million customers and supplies about 500 million gallons per day on a maximum day basis. The DW system includes three treatment plants, 17 major pump stations and over 160 pressure zones with over 400 pressure reducing valves. DW's primary responsibility is to supply high quality water to its customers at the lowest cost possible. This paper will present the process and results of a three-phased approach DW used to complete a suite of distribution system models, ultimately geared to meet the Stage II DBPR as well as provide guidance to operations for effective energy management. In 1997 DW began a process of creating steady state and extended period simulation (EPS) transmission models for a comprehensive Capital Improvement Program (CIP). This CIP planning was completed in late 2000. The transmission model was expanded in a second phase by creating a steady state all-pipes model in 2002. This more detailed model allows hydraulic analysis of smaller distribution mains. The third phase of the modeling transformed the steady state model into an all-pipes EPS model. This last model was calibrated using two system-wide tracer tests at maximum day and fall average day conditions. The all-pipes EPS model allowed accurate determination of detention times within the distribution system as well as determining efficient water quality sampling locations to meet regulatory requirements. The EPS model was also used to refine operational energy requirements.

This paper presents the results of a forecasting procedure developed for the reservoirs that supply water to the cities and suburbs of Everett, Tacoma, and Seattle, Washington. A six-month streamflow forecast is developed that incorporates weather forecasts provided by the National Center for Environmental Prediction (NCEP). The NCEP data are bias corrected, spatially distributed, and temporally downscaled to produce monthly forecasts for temperature and precipitation for four river basins. These forecasts are then used as inputs into a distributed hydrology model that translates the future climate conditions and the antecedent conditions into an ensemble of future streamflows. These streamflows are then incorporated into systems operation models to determine operation of the four regional water supply reservoirs. This paper also evaluates of the quality of the forecasts and describes how these ensemble forecasts can be incorporated into decision making.

This work enhances the multi-objective robust genetic algorithm (MRGA) for groundwater remediation design under uncertainty presented by Chan Hilton and Beckford. A new genetic algorithm (GA) is developed, which couples the MRGA for optimization under parameter uncertainty with a fast elitist non-dominated sorting genetic algorithm for multi-objective optimization (NSGA-II). This work is significant due to the innate uncertainty involved in ascertaining aquifer parameter values and the multi-objective nature of groundwater management problems. This parameter uncertainty makes it difficult to identify an optimal remediation design based on a deterministic description of the aquifer. The overall MRGA simulation-optimization approach unites a GA with a contaminant fate and transport simulation model and a spatially correlated random field generator for hydraulic conductivity values to determine tradeoffs between expected remediation design cost and reliability. The GA is a stochastic search method that mimics survival of the fittest through the selection, crossover, and mutation operators. One advantage of the enhanced MRGA (EMRGA) over other GA optimization approaches is its ability to efficiently solve optimization problems that are both multi-objective and include parameter uncertainty.

Approximation ("meta") models have been used in coupled optimization and simulation models to improve computational efficiency. In most instances, multiple simulation runs have been done before the optimization, which are used to fit an approximate model that is then used for the optimization. In this study, we propose a dynamic meta-modeling approach, in which artificial neural networks (ANN) and support vector machines (SVM) are embedded into a genetic algorithm (GA) optimization framework to replace time-consuming flow and contaminant transport models. Data produced from early generations of the GA are sampled to train the ANN and SVM and the numerical models are periodically called to dynamically update the ANN and SVM. This allows the meta model to adapt to the area in which the GA is searching and provide more accuracy. Preliminary results show that a well trained ANN or SVM can achieve satisfactory accuracy. Different approaches to dynamic training will be presented at the conference.

Since the 1960s, an increase in the public's environmental ethics, federal species preservation, water quality protection, and interest in free flowing rivers have evolved to the current concern for stewardship and conservation of natural resources. This heightened environmental awareness creates an appetite for data, models, information management, and systematic analysis of multiple scientific disciplines. A good example of this information and analysis need resides in the Green and Yampa Rivers, tributary to the Upper Colorado River. These rivers are home to endangered native fish species including the pikeminnow and razorback sucker. Two dams, Fontenelle and Flaming Gorge, impound the Green River headwaters. The respective reservoirs store water supplies as well as generate hydropower. Conversely, the Yampa River is considered unregulated and encompasses most of Dinosaur National Monument. Recreation is highly regarded on both rivers including fishing, whitewater rafting, and aesthetic values. Vast areas of irrigated agriculture, forestry, and mineral extraction also surround these rivers. To address this information need, we developed a prototype Environmental Resources Analysis System (ERAS) spreadsheet-based decision support system (DSS). ERAS provides access to historic data sets, scientific information, statistical analysis, model outputs, and comparative methods all in a familiar and user-friendly format. This research project demonstrates a simplified decision support system for use by a diverse mix of resource managers, special interest groups, and individuals concerned about the sustainability of the Green and Yampa River ecosystem.

A discrete-time optimal computational methodology is developed to minimize water level fluctuations within multi-reservoir river networks while meeting desired stages at user defined cross-sections, desired reservoir storage levels, bounds on dam releases and gate openings, and minimum pool levels behind locks and dams. The methodology integrates a statistical preprocessor tool; the National Weather Service's FLDWAV model, which is an unsteady, one-dimensional hydraulic simulator; and evolutionary search algorithms, namely genetic algorithms (GA) and a simulated annealing (SA) algorithm, as optional optimizers. The preprocessor determines the minimum and maximum water level bounds at desired cross-sections based on historical water level data, while the simulator solves the governing hydraulic equations for open channel flow. The optimizer is charged with solving the overall control problem and identifying optimal dam operation policies. Operational constraints such as water surface and storage level bounds and minimum pool level requirements for navigation are handled by an additive penalty function method. The methodology is capable of modeling reservoirs under three modes of operations: (i) head independent controlled releases, (ii) gate/channel controlled flow, suitable for simulating U.S. Army Corps of Engineers' "Wicket-type" dams, and (iii) head dependent release, suitable for modeling Tainter gate flows. The capabilities and robustness of the methodology are demonstrated through applications to a hypothetical three-reservoir river system and a major portion of the Illinois River Waterway. The developed methodology has the potential to assist engineers and water resource managers in establishing improved, coordinated reservoir operation rules.

Variations in water use at short time scales, seconds to minutes, produce variation in transport of solutes through a water supply network. However, the degree to which short term variations in demand influence the solute concentrations at different locations in the network is poorly understood. Here we examine the effect of variability in demand on advective transport of a conservative solute (e.g. chloride) through a water supply network by defining the demand at each node in the model as a stochastic process. The stochastic demands are generated using a Poisson rectangular pulse (PRP) model for the case of a dead-end water line serving 20 homes represented as a single node. The simple dead-end network model is used to examine the variation in Reynolds number, the proportion of time that there is no flow (i.e., stagnant conditions, in the pipe) and the travel time defined as the time for cumulative demand to equal the volume of water in 1000 feet of pipe. Changes in these performance measures are examined as the fine scale demand functions are aggregated over larger and larger time scales. Results are compared to previously developed analytical expressions for the first and second moments of these three performance measures. A new approach to predict the reduction in variance of the performance measures based on perturbation theory is presented and compared to the results of the numerical simulations. The distribution of travel time is relatively consistent across time scales until the time step approaches that of the travel time. However, the proportion of stagnant flow periods decreases rapidly as the simulation time step increases. Both sets of analytical expressions are capable of providing adequate, first-order predictions of the simulation results.

The recently developed expected moments algorithm EMA does as well as MLEs at estimating LP3 flood quantiles using systematic and historical information. Needed extensions include use of a regional skewness estimator and its precision to be consistent with Bulletin 17B and to make use of such hydrologic information. Another issue addressed by Bulletin 17B is the treatment of low outliers. A Monte Carlo study illustrates the performance of an extended EMA estimator compared to estimators that employ the complete data set with and without use of regional skew, conditional probability adjustment from Bulletin 17B, and an estimator that uses probability plot regression to compute substitute values for low outliers. Estimators that use a regional skew all do better than estimators that fail to use an informative regional skewness estimator. For LP3 data, the low outlier rejection procedure results in no loss of overall accuracy, and the differences between the MSEs of the estimators that used an informative regional skew were generally negligible in the skew range of real interest.

Up-to-date flood-frequency prediction methods for unregulated, ungaged rivers and streams of Tennessee have been developed. Prediction methods include regional-regression equations and the newer region-of-influence method. The prediction methods were developed using streamgage records from unregulated streams draining basins having from 1 percent to about 30 percent total impervious area. However, these methods should not be used in heavily developed or storm-sewered basins with hydraulically-effective impervious areas greater than 10 percent. The methods can be used to estimate 2-, 5-, 10-, 25-, 50-, 100-, and 500-year recurrence-interval floods of most unregulated rural streams in Tennessee. A computer application was developed that automates the calculation of flood frequency for unregulated, ungaged streams of Tennessee. Regional-regression methods were derived by using both single-variable and multivariable regression analysis. Contributing drainage area is the explanatory variable used in the single- variable equations. Contributing drainage area, main-channel slope, and a climate factor are the explanatory variables used in the multivariable equations. Average deleted-residual prediction errors for the single-variable equations ranged from 32 to 65 percent. Average deleted-residual prediction errors for the multivariable equations ranged from 31 to 63 percent. These equations are included in the computer application to allow easy comparison of results produced by the different methods.

The paper considers an integrated approach to control of quantity and quality in drinking water supply and distribution systems. The integrated control consists in optimising the operational cost, meeting a demand on water of desired quality and maintaining the constraints. A repetitive control strategy based on a short-term demand prediction and a mathematical model of the system is used to operate a water distribution system. The basic control structure consists two layers. The optimising controller at the upper layer operates according to a receding horizon strategy. The paper presents a hybrid approach to the optimising controller design at the upper layer. The hybrid solver combines the mixed integer linear and genetic solver.

Hydrologic Simulation Program FORTRAN (HSPF) is a lumped-parameter watershed model implemented in EPA's Better Assessment Science Integrating Point and Nonpoint Sources (BASINS) system. In order to perform water quality simulations in specific study areas, HSPF's hydrology must first be calibrated to observed streamflow data. PEST is a model-independent parameter optimization program which minimizes an "objective function" quantifying the misfit between model outputs and corresponding field measurements. PEST also performs nonlinear predictive confidence interval estimation in which PEST determines maximum and minimum limits on key model predictions. A PEST/HSPF Windows interface is being developed for incorporation in the BASINS GIS-based watershed and water quality modeling system. This study seeks to identify successful PEST/HSPF calibration protocols that successfully minimize key model predictions for incorporation in the interface design. We present specific details of PEST implementation for HSPF, including model setup, identification of sensitive parameters, and parameter transformations required to facilitate the parameter estimation process. Calibration parameter set nonuniqueness is demonstrated. PEST utilizes this non-uniqueness to define nonlinear predictive confidence interval estimates. PEST varies parameters, while maintaining the objective function at or below a user-defined threshold, to define the range of predictions that satisfy calibration constraints of the objective function.

A Poisson Rectangular Pulse (PRP) process is used to simulate instantaneous indoor and outdoor residential water use during a 55-hour period at 135 demand nodes in an all-pipes water distribution system. The corresponding network equations are then solved with an ad hoc program to generate unsteady flow rates in each pipe. This computational approach yields an unprecedented level of detail, affording new insights into the hydraulic behavior of municipal pipe networks. Results of simulation experiments include the frequency of stagnant, laminar, transition and turbulent flow, the number of flow direction reversals, the coefficient of variation of the flow rate, the minimum and maximum rates of flow and other hydraulic statistics for any link in the pipe network. These hydraulic characteristics are compared against similar data generated by a standard network progam using a one-hour extended period time resolution.

The mountain water supply of the City of Nanaimo (population 78,000) in British Columbia, Canada, provides the potential for generating hydroelectric energy. The hydro plants would be connected directly to the two supply pipelines — head available for power therefore will vary inversely with water consumption. The hydraulic analysis made effective use of the capability of the Shamir and Howard network analysis program SDP to determine the hydro energy by explicitly solving for unknown conductance, and combinations of head and consumption at the same node. Optimization of the installed capacity also considered opportunities for power sales, variability in present and future water consumption, hydrology of the ungauged river sites, instream flow requirements to protect aquatic habitat, and hydroelectric licensing requirements.

The Texas Commission on Environmental Quality (TCEQ), its partner agencies, and contractors developed during 1997–2003 a water availability modeling system for the 23 river basins of the state, pursuant to comprehensive water management legislation enacted by the Texas Legislature in 1997. The modeling system is motivated by water allocation concerns and is designed to support a broad range of planning and water rights regulatory activities. Implementation of the statewide water allocation modeling system illustrates key technical and institutional aspects of this type of effort. The generalized modeling system and the lessons learned in its development and application in Texas are applicable to river basin management throughout the world.

The purpose of this study was to improve the Ensemble Streamflow Prediction (ESP) probabilistic forecasts by combining them with streamflow persistence information. ESP was applied to one-month ahead inflow forecasting for the Chungju multipurpose dam in Korea. This ESP system ran a rainfall-runoff model called TANK in the beginning of a forecast month on a daily basis. The historical rainfall data for the 36-year period from 1966 to 2001 were employed as the meteorological input in order to generate the corresponding 36 inflow scenarios each month. To issue probabilistic forecasts, the range of the generated inflow scenarios was divided into three flow categories and a probability was assigned to each flow category from the fitted probability density function. Using the optimal linear combination (OLC) approach, the ESP forecast was then combined with the persistence and naive forecasts. The persistence forecast represents a conditional probability of a flow category, provided that the inflow category of the previous month is given. The naive forecast assigns a 33.3% probability to each flow category. Split sampling simulation proved that OLC improves the accuracy of ESP forecasting for the non-flood season.

Under the water withdrawal regulations and water rights based on a concept of sharing, allowable withdrawals by individual abstractors vary directly with streamflow. The day-to-day administration of such systems, therefore, requires that withdrawal rates be tied to observed streamflows. For many water users, whose reference stream gauge is downstream from the point of abstraction, this poses the possibility of time delay between water withdrawal and streamflow measurement. This in turn induces the possibility of unstable operation of stream water withdrawal systems such that, for example, with a one-day time lag in the stream, the allowable withdrawal may be high one day, resulting in streamflow subceeding the standard the next day, resulting in withdrawal curtailment orders the following day, resulting in rebounding of the streamflow the following day, etc. Such instability exacerbates violations of the streamflow standard and inconvenience in implementation of water withdrawal regulation. Thus resolution of this problem is very significant for successful water withdrawal regulations. This paper explores the application of process control theory to resolve the instability problem through a hypothetical example, in pursuit of meeting minimum in- stream flow requirement and withdrawal demands of abstractors. Since feedback control systems adopted here do not require great accuracy in the model connecting stimulus and response, difficulties in modeling time delay may be overcome to a great extent by a specification of feedback controllers. This paper reveals that with this control technique, we are able to maintain a stable streamflow that meets the minimum in-stream flow standard. This result promises an improvement of the overall reliability of the water withdrawal system and convenient implementation of the withdrawal scheme.

The Bang Pakong river basin is part of Thailand's Eastern Seaboard. It has been the target of considerable economic development given its proximity to the Bangkok Metropolitan Area. Several new reservoir projects have been planned or already constructed to support the Basin's increased industrial, residential and agricultural water needs. Among those projects, the Bang Pakong diversion dam, located about 70 kilometers from the river mouth, was constructed to regulate the upstream inflow for multiple uses and to minimize salinity intrusion. Unfortunately, the diversion dam did not function as planned, leading to a number of problems related to bank erosion, flooding and poor water quality.

Nowadays, the operation of a drinking water supply network can be assessed by using hydraulic computer models, such as "Porteau", developed by Cemagref. These models are accurate for high volume flows (such as emptying or filling of tanks) but less so for minimum flows. Partially closed valves, small diameter pipes and overestimation of consumption lead to difficult numerical problems. In these cases, optimization is required to prevent non- convergence of solutions. Besides, for such a model, consumption is an input and has to be estimated without considering the available pressure. Thus, demand overestimation is emphasized at the time of a network dysfunction, such as pipe breaks, pump failures and missing water resources. For about ten years, authors have proposed demand depending on available head. Such a model is known as a "Pressure Driven Model" (PDM), but theory and solving tools have not been fully developed yet. Firstly, two problems of optimization are associated to (PDM): one is written in terms of pressure heads, and the other in terms of flow rates. As with the classical models, they correspond to principles of least action. Next, equivalence of formulations and solution existence and uniqueness are demonstrated. Then, three descent algorithms of Newton type have been proposed. Finally, the efficiency of these algorithms is proved in solving former problematic study cases.

The Upper Rio Grande Basin Water Operations Review and Environmental Impact Statement (EIS) is a multi-agency effort that is using a Geographic Information System (GIS) base to evaluate operational changes to improve river system management, The Upper Rio Grande Water Operations Model (URGWOM-Planning version) represents the backdrop of an institutionally and physically complex river system. There are three joint-lead agencies, five cooperating agencies, and other participants involved in this formal EIS planning process. With a host of different scientific disciplines on ten technical teams, GIS was the choice of the Interdisciplinary Team to "somehow put it all together and make sense of it", a Rosetta stone for the Tower of Babel. This paper documents hidden pitfalls, such as converting ArcInfo data to ArcView, trials, tribulations and most importantly, techniques for successfully using GIS for individual resource analysis and integrating these into a basis for making management decisions.

An efficient operating plan for multiple reservoir systems should be developed based upon the effect of long-term storage and short-term release policies. The short-term planning model is supposed to attain the long-term storage target meeting the changes in hydrologic conditions and water demand. In this study, we developed a mathematical model for daily coordinated multiple reservoir operation to find the efficient daily operating policies. The model is formulated as a multiple objective mixed integer programming (MOMIP) model based on multiple period dynamic network flow optimization. For the daily operation model, short-term operational characteristics such as river routing, and the nonlinear effect on hydropower generation should be considered. For the daily river routing, the concept of lagged time water conveyance is developed as a function of the volume of water released. The notion is incorporated into the mixed integer programming model to simulate the process of determining progressively the timing and the amount of water delivered at successive points along a river.

Pollution transport models are based on field data and the calibration of the parameter values that cannot be directly measured. This combination of field data and calibrated model can provide an important investigative and planning tool in environmental analysis. However, the calibration process for such models can be computationally very demanding if it is done thoroughly. An even more serious computational burden is parameter sensitivity associated with models that have a large number of parameters. This paper discusses computationally efficient methods for sensitivity analysis. It also discusses applications to the large and highly significant Cannonsville watershed, which provides drinking water for New York City. The watershed model used is SWAT. The sensitivity analysis method is applied to 160 parameters simultaneously. The methodology described can also be applied to other types of models arising in water resources and in hydraulics.

A methodology that establish the optimum water head pressure at the entrance of each hydraulic sector of the potable water distribution system, allow an efficient management and a free asset cost planning and operation of the network. Also reduces the commercial water losses (Non revenew water), reduces the amount of burst pipes, raises the pipes life time due a less operational pressure, and allows fast and easy identification of zones with commercial problems, spatial localization of users with consumption (0), spatial localization of zones with high rates of technical losses to detect leaks and illegal connections, renewal and rehab programs according to the real needs of the water distribution networks and long term assets investment and operational costs of the water distribution system.

Aging water distributions systems face a variety of problems in terms of disruptions resulting from lack of good quality water and pressure, and increased leakage, and breakage. These problems are attributed to corrosion, unanticipated overburden load, frost load, thermal load, expansive soil load, and water hammer pressure. Corrosion reduces the strength of a pipe in terms of reduced thickness and increased loads worsen the load bearing ability. In this paper, we review pipeline conditions that result in failure. The methods reviewed are physically based as opposed to the lumped statistical procedures. The methods are implemented in the form of a user-friendly decision support system software PipeADDIN. The software capabilities are also presented.

Management of groundwater contamination is a very cost-intensive proposition filled with conflicting objectives and substantial uncertainty. A critical source of this uncertainty in groundwater problems comes from the data for the conductivity values for the aquifer on which the flow and transport of the contaminant is dependent. For a remediation solution to be reliable in practice it is important that it is robust over the error in modeling data. This paper presents our efforts to model the uncertainty for the Umatilla Chemical Depot site at Oregon, a difficult task given that the scarcity of available data precludes use of stochastic hydraulic conductivity generation techniques. The installation's modeling team has divided the site into conductivity zones. We use the results from various pumping tests to establish plausible ranges for the hydraulic conductivity value in each zone. Realizations for each zone are then generated randomly from these ranges. The hydraulic head conditions resulting from each realization are then compared with measured head conditions. To incorporate spatial as well as quantitative differences in the comparison, the first moments of the hydraulic head scenarios are also compared. Unrealistic realizations are eliminated and the remaining realizations are ranked based on the moment values. The ranked realizations will be used for efficient sampling using Latin Hypercube Sampling within the framework of an advanced stochastic multi-objective genetic algorithm to obtain robust, reliable and optimal solutions.

The conversion of natural and agriculturally dominated watersheds to industrial, commercial and residential developments leads to a cascade of adjustments in runoff quantity and stream quality at locations further downstream. The use of sophisticated hydrologic simulation models and Geographic Information Systems (GIS) has become the standard for evaluating these impacts of urban sprawl on water resources systems. Simulation and GIS models alone, however, are incapable of directly revealing optimal land development patterns that meet specified objectives. This paper describes the development of a multi-objective Spatial Decision Support System (SDSS) designed to overcome this limitation. The SDSS is created by integrating the U.S. Department of Agriculture's Soil and Water Assessment Tool (SWAT) for comprehensive hydrologic simulation, a GIS for generating input and visualizing output, and a genetic algorithm (GA) for identifying weighted, optimal land use patterns. In addition to the GA, future research will involve the integration of a second search mechanism, the artificial life algorithm, to verify optimal results. The optimal landscape is that which minimizes sediment yield in subsequent streams, while simultaneously maximizing approximate anticipated profit from urban development. The SDSS could be a useful visualization tool for land use managers and watershed management institutions in planning new developments. The SDSS has been tested on the Lower Kaskaskia watershed, located in the Metro East area of southwestern Illinois. Evidenced by a historical survey of population growth and hydrologic and water quality variability, this basin is an example of a watershed that is undergoing extensive water resources changes as a result of urbanization. An investigation of watershed planning activities and stakeholder groups in the watershed has also been undertaken. Meetings with these individuals have allowed direct dissemination of the research to affected groups and have been useful for generating feedback on future work and model modifications.

Brazil is a country plenty of water resources. Due to that 90% of its electricity is produced by hydropower plants. The Brazilian hydropower system has four major sub-systems, North, Northeast, Southeast and South, with 61,000 MW of installed capacity. The São Francisco river system is located in the Northeast area and has seven power plants with three storage and six run of river reservoirs. It has 10,229 MW of installed capacity and it is the biggest system in the Northeast, Brazil. The São Francisco drainage area is 640,000 square km and its water resources are used for energy, irrigation, water supply and navigation. During the rainy season the three storage reservoirs are used for flood control. Part of the São Francisco basin is located in a very dry and poor area. There are many irrigation projects designed for the dry areas, including a huge water transfer system from São Francisco River to the neighbor basins. Due to that the conflict between irrigation, flood control and energy is increasing. In order to analyze these conflicts a multi purpose non linear model for multi reservoir operation was developed. The multi objective model uses the constraint methodology. Considering different scenarios for irrigation demands and flood control volumes a set of trade-off curves were calculated. These graphs show the relationship between three purposes: energy, irrigation and flood control and can support the decision processes on the São Francisco system operation. This paper shows the multi purpose model, its methodology and its preliminary results. Some analyses on the São Francisco water resources conflicts are also presented.

Genetic Algorithms (GAs) face major computational bottlenecks when numerical models are used for estimating the fitness of the objective function. Especially in large-scale water resources design problems, where the scale of the spatial grids is important in determining the numerical accuracy of the design, a tradeoff exists in the precision of the fitness function and the computational expenses endured. This paper discusses multiscale strategies that can be utilized for improving the performance of GAs when working with spatial grid dependant fitness functions. The strategy uses fine grid and coarse grid fitness functions strategically to maintain the accuracy and computational speed of the problem and drive the GA towards better and more accurate solutions faster. The algorithm's efficacy is tested using a groundwater remediation design case study.

Drinking water systems should invariably be operated on continuous pattern for twenty four hours a day. But due to financial constraints it is not practically possible to operate drinking water systems for twenty-four hours a day in developing countries. Generally a period of eight hours or less is considered adequate to supply the drinking water. Due to present faulty design practice, which is derived from continuous water supply the consumers are deprived of adequate water quantity and pressure. Intermittent water supplies are unique in itself and need different approach for design and modeling other than the continuous water supplies. Demand of water at particular node is important characteristic in the continuous water supply, but it is SUPPLY of water at a node what drives the intermittent systems. Many factors like charging of water mains at the start of supply each day, ingress of contaminant due to negative pressure, application of peak factors, changing resistance coefficients, equitable water distribution using zoning valves, leakages etc are peculiar features of intermittent water supply. A comparative study of existing town in northern India is presented for the design of new water supply system using present design practices and a proposed design method. The attempt has been made using "EPANET" software and it is suggested that new method is more practical and realistic for the design of intermittent drinking water systems.

An alternative transport layer for water quality models is described which relies on actual flow data obtained from a real-time SCADA system. Two of the central concepts of the new transport layer are operating modes and the connectivity matrix. An operating mode is defined as a pattern of flows in which all links in the network are assigned a flow direction. Criteria for determining the feasibility of operating modes are incorporated in a search algorithm that finds all the feasible operating modes for a network without recourse to full enumeration. The algorithm is able to make inferences in flow directions in links that have not been established directly from the SCADA data. The connectivity matrix summarizes the combined effects of these feasible operating modes. The resulting connectivity matrix can be used to determine potential sources and the dispersal of water quality constituents as demonstrated in an example. Finally, the impacts of the proposed SCADA-driven transport model on the selection (design) of network geometry, the placement of valves, and placement of monitoring equipment are discussed briefly.

The Geum River basin, one of most important river basins in Korea, is located in the South Central portion of the country. Water conflicts have erupted in the Geum River basin twice recently, during the droughts of 1995 and 2001. These recent events have accentuated the need for careful planning in the basin and focused attention on the conjunctive operation of Daechong Dam and the Yongdam Dam (constructed in 2001). This paper describes the development of an optimization model that enhances understanding of water planning in the basin and provides a disciplined method for developing and comparing robust and reliable reservoir operation alternatives for the Geum River basin. The mathematical objective function of the model is to minimize the maximum difference between supply and demand subject to constraints representing: (1) conjunctive dam operation, (2) instream flows downstream of the dams, and (3) increasing water demands in Jonju area. Stochastic constraints, as well as deterministic constraints, are included to develop robust operating alternatives. Most significant among the uncertainties that must be included in the models are those associated with future water demands and streamflow forecasts.

The Riparian Decision Support System combines hydrologic, ecologic and economic models to provide the best distribution of riparian buffers along the edges of a stream. The Soil and Water Assessment Tool and the Riparian Ecosystem Management Model were combined within a single interface, providing input for an economic model that searches for the least cost width, and buffer distribution constrained by monthly pollutant threshold levels. This optimization problem was solved using a heuristic method. A tabu search algorithm was developed to obtain the least cost buffer widths. The Riparian Decision Support System was applied in a watershed in Delaware County, New York to estimate the effectiveness of alternative buffer widths for the reduction of phosphorus.

This paper demonstrates how the genetic algorithm technique can be used to optimize the operations of a water distribution system considering both hydraulic and water quality considerations. The methodology developed is demonstrated through application to a simple water distribution system consisting of a reservoir from where water is pumped into a high level tank. It is then gravity fed into a distribution system consisting of 19 pipes and 6 loops. The pump is activated by water levels in the tank called "trigger" levels. When the water level falls below the lower trigger level, the pump switches on and stays on until the water level reaches the upper trigger level. As there is a different electricity tariff for peak and off-peak pumping, different trigger levels may be set for peak and off-peak periods. Hydraulic optimization involves choosing these four trigger levels so as to minimize the pumping cost for the system for a defined daily demand pattern. Water quality is considered in terms of the concentration of free chlorine in the water. Cost penalties are imposed if this concentration falls outside the range of 0.1 mg/L to 1.0 mg/L in the distribution system. The concentration of chlorine after dosing (just downstream of the pump) is an additional decision variable in the problem. The results obtained indicate that different trigger levels are required when water quality is taken into account. The paper demonstrates a methodology that can be used to minimize pumping costs in a distribution system while still ensuring that desirable water quality is achieved.

Water systems are responsible for conducting monitoring of drinking water to ensure that it meets all drinking water standards. EPA has established pollutant-specific minimum testing schedules for public water systems. Typical sampling and monitoring sites in distribution systems may include points close to water treatment systems, core business locations, secondary water storage reservoirs, repumping or retreatment facilities and change in water quality. Sampling location is also dependent on investment in labor, transportation requirements, and supplies to carry out sampling commitments. This adds to a water utilities financial commitment. Monitoring and/or predicting the fate and transport of introduced contaminants in water distribution systems is therefore a challenging proposition, involving the identification and operation of numerous hydrological and water quality-related factors. A common and important question in the design of a monitoring network is how many samples should be collected and where? The answer is often based upon best professional judgment and financial considerations. This paper describes a methodology to select nodes for monitoring in case of an intrusion. The monitoring location selection methodology uses a combination of extended period simulation (EPS) model and GIS data.

California has long learned from and led water management for semi-arid and arid regions. Sustained growth in environmental and urban demands and decreasing agricultural commodity prices have increased stress in California water policy, planning, and operations. These stresses, combined with a nation-wide retreat of Federal and State governments from major infrastructure projects, have led local and regional agencies to reconsider and change long-held management and operating policies. Examples can be seen in the areas of demand management/water conservation, conjunctive use of ground and surface waters, wastewater reuse, environmental restoration, water markets and exchanges, and the integration of these newer management techniques with traditional water supply infrastructure options. This paper discusses these ongoing changes in California water management with examples at local, regional, and statewide scales. The results suggest myths and new realities for California water management, which also have relevance for water management in semi-arid and arid regions. Many of these experience-based practical lessons also are supported by computer modeling studies. These modeling studies suggest potential for continued re-assembly of California water management and provide a basis for identifying promising local, regional, and statewide initiatives for improved water management.

Knowledge of water-distribution system historical operations is required to obtain information on the frequency and duration of exposure to drinking water for epidemiologic investigations. For example, groundwater well and booster pump on-off cycling patterns over a typical 24-hour demand cycle, for periods that can exceed 30 years, are required information that must be known or estimated. Prior to the generalized use of supervisory control and data acquisition systems, however, this type of detailed information was rarely available from operators' records or notes. Therefore, an approach was developed whereby missing and unknown historical operating data could be estimated and quantified using a set of "Master Operating Criteria" (explicit conditions and standards based on hydraulic engineering principles necessary to successfully operate water-distribution systems similar to the one investigated by the authors) along with generalized operating practices of the water utility. Using the aforementioned approach, however, does not guarantee a unique or optimal set of system operations owing to parameter uncertainty and variability. To address this issue, sensitivity analyses are conducted using genetic algorithm optimization to search for and select a set of alternate operating conditions that result in the satisfactory operation of the historical water-distribution systems. In this paper, a case study of the water-distribution system serving the Dover Township area, New Jersey, from January 1962 through December 1996 is presented. In this system, contaminated drinking water from two of ten municipal well fields was distributed to the community. Using the approaches described above, historical system operations were reconstructed. The reconstructions allowed epidemiologists to compute exposure indexes required for conducting a case-control study.

This study uses genetic algorithms (GAs), a relatively new approach to optimization, to refine roughness values of pipes and demand values associated with nodes in a network. The approach is used to obtain more precise hydraulic calibration of an existing network. The methodology introduced here uses field measurements of fluoride of an existing network in the U.S. as the basis for calculation of network model parameters such as roughness of pipes and nodal demands.

The control and management of discolouration of potable water is one of the biggest problems to face the international water supply industry. Knowledge and understanding of the processes, materials and other factors leading to discolouration events is improving and this has lead to enhanced operational and management techniques to reduce the risk of discolouration events. However there is little knowledge with regards to the regeneration of the materials responsible, hence return periods for maintenance are uncertain. Results of fieldwork undertaken to investigate the regeneration of discolouration risk within a live distribution system are presented. The results clearly demonstrate the regeneration of material causing discolouration. The change in roughness of CI (Cast Iron) pipes with time, and the impact of this on hydraulic capacity is well documented. Several studies have compiled datasets and derived empirical equations to predict the rate of roughness increase with time. This paper compares discolouration regeneration rates evaluated from monitored field data with predicted increase in pipe roughness using these empirical methods. Such comparison shows good agreement. Hence from the results presented it is possible to evaluate maintenance return intervals necessary to control discolouration risk in CI pipes based on knowledge of water quality parameters. The methodology has been applied to derive a cleaning frequency for the control of discolouration risk in a typical distribution system.

Data of 12 stream flow gauging sites of Sone Subzone-1 (d) of India are screened using the discordancy measure (D_i) and homogeneity of the region is then tested employing the L-moments based heterogeneity measure (H). It is observed that the data of 10 out of 12 stream flow gauging sites constitute a homogeneous region. Comparative regional flood frequency analysis studies are conducted using the various L-moments based frequency distributions. Based on the L-moments ratio diagram and |Z_i^dist| -statistic criteria, GEV distribution is identified as the robust distribution for the study area. Regional flood frequency relationships are developed for gauged catchments using the L-moments based GEV distribution. For estimation of floods of various return periods for ungauged catchments, a regional relationship between mean annual peak flood and catchment area is developed and the same is coupled with the regional flood frequency relationship derived for the gauged catchments. Flood frequency estimates of gauged and ungauged catchments obtained using the data of 10 gauging sites are compared with those obtained using the data of 12 gauging sites, without meeting the criteria of regional homogeneity.

This article presents analysis results of degradation modeling for water distribution systems in urban environments. The system is based on data from New York City (1982–2002) and application of statistical modeling techniques stemming from parametric and non-parametric survival analysis. Artificial Neural Network techniques are also employed for the identification of the underlying risk factors and their relevance to water main breaks (about 12 factors were reviewed) including age of pipe, diameter, material, location, previous breaks, etc.). The statistical models used were calibrated for different groups of pipes, and a combination of parametric (Weibull distributions) and non-parametric models (Kaplan-Meier survival curves, Epanechnikov kernels) were utilized to investigate the suitability of them to the estimation of pipe degradation and the forecasted "time to next failure (break)". The research aims the development of a framework of risk analysis models for the degradation of water distribution systems, and it is a prelude to a wider effort to combine such models with geographical information systems (GIS) in the operations and maintenance of large water distribution networks in urban environments.

Many water resources problems require careful balancing of fiscal, technical, and social objectives. Informed negotiation and balancing of objectives can be greatly aided through the use of evolutionary multiobjective optimization (EMO) algorithms, which can evolve entire tradeoff (or Pareto) surfaces within a single run. The primary difficulty in using these methods lies in the large number of parameters that must be specified to ensure that these algorithms effectively quantify design tradeoffs. This paper addresses this difficulty by introducing a multi-population design methodology that automates parameter specification for the Nondominated Sorted Genetic Algorithm-II (NSGA-II). The NSGA-II design methodology is successfully demonstrated on a multiobjective long-term groundwater monitoring application. The design methodology fully exploits the efficiency of the NSGA-II to enable the solution of a new class of high order multiobjective applications in which users can balance more than two performance objectives. Using this methodology, multiobjective optimization problems can now be solved automatically with only a few simple user inputs.

Groundwater simulation/optimization techniques are often formulated using many candidate well locations. Each candidate well location represents the potential location of a well in the final optimal solution. Execution of the optimization solution algorithm yields active pumping at a subset of these candidate locations. It is often convenient to use binary variables to represent the presence or absence of each well. Use of binary variables makes it possible to include well construction costs in the optimization formulation and to include constraints such as the number of active wells and non-zero lower bounds on pumping rates. Inclusion of binary variables produces a mixed-binary optimization problem that can require significant computational effort to solve. In the present work, the algorithmic advantage of a modification to the branch and bound algorithm for mixed-binary linear programs is considered. The approach involves resetting the binary variables in subproblem based on the optimal solution to the subproblem. Application of the modification to a small test problem yields substantial reduction in the number of iterations required for the branch and bound algorithm.

The pump affinity laws are widely used in adjusting pump performance curves from one pump or speed to another. In most cases, they work fairly well and can be reliably used in modeling studies. This paper describes an experiment undertaken to determine how well these laws apply to a variable speed pump, in which the speed is controlled by a variable-frequency drive, the most common technology used today. The pump speed was varied from full speed to roughly 30 percent of full speed. Over that range, the affinity laws worked quite well for adjusting the pump head characteristic curves. However, the pump efficiency curves fell significantly lower than those predicted by the affinity laws. The explanation for this is most likely the loss of efficiency in the variable-frequency drive itself. Since energy costs are the largest component of pump life-cycle costs, the implications for pump selection are significant. With variable-speed drives costing almost as much as the pumps themselves, it becomes essential for life-cycle cost estimates, whether done manually or with models, to accurately account for the loss of efficiency.

Control of integrated quality and quantity in Drinking Water Distribution Systems within recently proposed hierarchical framework is considered in the paper. A supervised nonlinear Indirect Model Reference Adaptive Controller is derived for the lower control level of the control structure to operate as the fast feedback controller of chlorine residuals in the monitored nodes. The major supervisor role is to manage switching between linear adaptive controllers. The controller is tested on an example DWDS and it achieves a superior performance over known linear adaptive controllers.

Random intermittent laminar flow is common in the periphery of a municipal distribution system. The Poisson Rectangular Pulse (PRP) process is employed to generate realistic residential water demands. These demands are converted into sequences of random intermittent laminar flows, statistically identical to flows through a typical pipe in a peripheral zone of the network. The intermittent laminar flow sequence is used to drive unsteady transport of a conservative solute through the pipe network. Adopting a high-resolution two-dimensional advection-diffusion model as the benchmark, a one-dimensional advection-dispersion model is compared to examine the effect of the temporal averaging scale and the initial dispersive condition on the spread of the solute cloud.

WEST Consultants, Inc. was tasked to provide analysis regarding floodplain impacts and bank protection for two developments in the floodplain near Phoenix, Arizona. The developments were in areas that one-dimensional models show as only partially in the floodplain. The one-dimensional models also show either very limited or no impacts resulting from the new developments. One development, however, was in an area of very complex flows while the other development was in an area that appeared to be one dimensional. Two-dimensional models of both areas, however, showed unexpected flow patterns and impacts due to the developments. In one of the areas flood elevations along the project boundary as predicted by the two-dimensional model were as much as 2 ft above the non-encroached flood elevations obtained from the one-dimensional model. This difference in water surface elevations would have serious implications for flood protection levels if not noted and accounted for during development. One of the projects was designed such that adverse impacts from development were eliminated while the other project used a combination of drainage easements and channels to mitigate adverse impacts caused by development. This paper presents summaries of both projects and show the necessity for two-dimensional modeling in cases where complex flows or geometry exist as well as in some areas where overbank flows appear to be one-dimensional. Both projects were designed to minimize adverse impacts on neighboring property owners.

Stochastic optimization techniques allow the decision making process to incorporate variability and uncertainty leading to decisions which perform satisfactorily under a range of conditions. When considering probabilistic constraints, there are both robust and reliable formulations. Robust formulations may not be reliable when simulating a different representation of the variable and uncertain parameters. Utilizing the Bayesian Binomial-Beta conjugate pair to represent the distribution of sample violations, a reliable constraint formulation is developed. Applying the reliable constraint formulation to particulate removal through conventional drinking water treatment plants increased the design cost by 5.1% relative to the robust formulation. Unlike the robust solution, the reliable solution satisfied the probability constraint under multiple simulation sets. The reliable formulation was then modified to terminate the simulation loop within the stochastic optimization routine when the proportion of violations was determined to be statistically greater or less than the desired value (under-or over-designed, respectively); thus removing wasted computational effort. The over-designed option did not reduce the computational efficiency. The under-designed option required one more iteration in the optimization routine, however, the total number of process model calls was reduced by 19.4%.

Encrustation and biofouling causes loss in carrying capacity of water mains and impacts water quality. In this work, the decrease in pipe diameter and increased pipe roughness are computed as a function of service age of water mains. The method is applied to the water distribution network in Walkerton, Ontario and the effects of encrustation and biofilm build-up with age and their effects on the hydraulics are estimated. Hydraulic analysis was performed by simulating the flow through the water supply network, assumed to be 100 years old, using EPANET2. The results indicate that changes in the hydraulic parameters of a network can cause 1. a change in a flow pattern from reservoirs, 2. water to take a longer time to reach the far end of network and 3. increased energy cost. It is also suggested that the available empirical model to compute pipe roughness and decrease in pipe diameter after a certain service time cannot be standardized, because of its dependence on the local and transient quality of water from various supply sources.

Small systems are often plagued by a lack of resources, outdated equipment, unreliable raw water supplies, and minimally trained staff. A study has been undertaken to document the transition of a small system from conventional treatment to microfiltration. The Town of Carthage, North Carolina (population 2,300) completed construction in May 2002 of a 1.0-MGD microfiltration treatment plant. The study involves the analysis of monthly samples of raw water, finished water, and water collected at six different locations in the distribution system, both before and after the startup of the microfiltration plant. Water quality in the distribution system has been highly variable. To better analyze and interpret water quality data for this system, a water quality model of the distribution system was used to predict water age and disinfectant residual concentration. The model demonstrated that tank sizing and operation along with long retention times significantly affect water quality in the Carthage system.

In Korea, water quality variations (WQVs) in water distribution system (WDS) are characterized by seasonal tendency of water temperature variations ranging from 4 to 29°. Therefore, establishment of special administration plan by season will be necessary for the water quality management in WDS. This research monitored WQVs to evaluate the effect of water temperature in WDS of Y water treatment plant (WTP). 11 points were selected for sampling through the distribution system. Samples were analyzed for 52 parameters such as residual chlorine, TOC, TTHMs, and others affecting on the corrosion and microbial growth. Water temperature variation is more than 20° through seasons generally. As a result, decay of residual chlorine and microbial detection, metal release becomes an issue. Therefore, management of residual chlorine is essential to control microbial re-growth in summer when exceed water temperature 20°. Also, because corrosives of water increases in winter when water temperature is average 7°, control of aggressive water through adjustment of pH and alkalinity is needed.

The U.S. Global Change Research Program (USGCRP) includes, among other issues, the impact of global change on water quality. This field study evaluates the impact of global changes (land-use change and climate change) on source water quality. Changes in source water quality can change the cost of drinking water treatment. Obvious changes to the upstream environment such as land development or natural resource production effect source water quality. Other changes that happen at a global scale, such as atmospheric warming or drought frequency alter the hydrologic cycle through atmospheric or groundwater processes which, in turn, may impact source water quality. Using data from two carefully selected watershed sites, this study weighs the relative merit of land-use practices as a means to mitigate impacts to source water from the large-scale atmospheric changes. This paper will present a unique framework for the site selection process. The goal of this process was to find sites for which we could define a "boundary" condition. "Boundary" in our context means more than the areal extent of the watershed; we need to define the boundary of the model state for some set of hydrologic parameters in time and space.

One of the first steps in setting up an optimal groundwater remediation design problem is developing an appropriate objective function, which represents the primary goals of the design. Selecting appropriate objective functions can be challenging. A more realistic objective function, which is a cost function applied to a realistic site without simplification, may yield more accurate results but at the same time it will require more time and effort to develop the appropriate function for a particular application. On the other hand, a simple function will save setup time but may sacrifice the accuracy of the results. This research seeks to identify what situations encountered in remediation design would make the development of a realistic objective function necessary. It also examines tradeoffs among three objectives: total cost, risk, and total cleanup time. A pump-and-treat system is designed for a case study to explore these questions. The model used here is NSGA II (Non-dominated Sorting Genetic Algorithm-II) combined with two numerical models (Modflow and RT3D) and an exposure and risk assessment model. Four different cost functions are applied, ranging from simple to complex. The results show that the realistic cost function generally found better solutions than the simplified ones, especially for shorter-term cleanups. These findings are now being tested for a field-scale application at Umatilla Army Depot in Oregon.

So far Artificial Neural Network (ANN) has gathered popularities in many engineering fields because of its ability to nonlinearly relate independent and dependent variables. This paper deals with the application of the ANN to the decision support system (DSS) for the pipe condition assessment in water distribution systems. Also the window-based interface is developed using visual tool for enhancing user-friendliness.

The Chestnut Branch of Mantua Creek runs through Rowan University's campus, located in the town of Glassboro, NJ. The Chestnut Branch is part of the Mantua Creek Watershed in southern New Jersey. Currently, the Chestnut Branch serves as a drainage system for the university's storm water runoff. More recently, the land surrounding the stream has undergone substantial residential and commercial development, resulting in a large area of impervious ground cover and high discharges of storm-water runoff into the stream. In addition, even more development is planned for Rowan University over the next ten years. Excessive runoff into Chestnut Branch has had a significant impact on water quality as well as the physical characteristics of the stream. Severe erosion has become evident along extensive portions of the stream's banks. Trees that were once firmly rooted are now holding on by small roots, which are growing at a severe angle to the banks. Despite the severity of erosion present in Chestnut Branch, many of the above-mentioned problems could be repaired or stalled with restoration and stabilization techniques. The goal of this project is to develop long-term restoration measures that will counter the effect of continued land development around the Chestnut Branch in addition to providing college students with invaluable experience. This goal is to be met through the completion of five components as follows: Water Quality Monitoring, Hydrologic Studies, Hydraulic Analyses, and Design and Implementation of Restoration measures. Each of these will be discussed further.

The majority of artificial neural network (ANN) algorithms applied to different water resources problems are mainly the multi layer perceptrons with feedforward back propagation algorithm (FFBP). This method is sensitive to the initial random weight selection and is likely to be trapped by local minima. In this study another ANN algorithm, evolutionary artificial neural network (EANN) is employed in daily river flow forecasting. EANNs are the neural network structures into which the algorithms inspired of evolution are incorporated. In this implementation EANN has a multi-layer perceptron network structure but connection weights are adjusted by breeder genetic algorithm. The simulations showed that problems faced by back propagation algorithm were overcame and the EANN forecasts compared well with FFBP algorithm and the conventional stochastic model in terms of the selected performance criteria.

In this study, an optimization model is developed for monthly operation of a multi-purpose hydropower reservoirs using genetic algorithm. The real value encoding approach is used considering alternative representation, selection, crossover, and mutation schemes. The constraints are handled using the Multiplicative Penalty Method (MPM) function, in order to evaluate the objective function in deferent conditions. The reliability of water allocation to different demands and hydropower generation are evaluated using an economic objective function which has been calculated based on the actual value of water and energy of Karoon-I Reservoir in southwestern part of Iran. The results of this study have shown the importance of selecting a suitable mutation operator for reducing the computational run time of the optimization model. The robustness and efficiency of genetic algorithm in developing the operation policies for a multi-purpose hydropower reservoir is discussed in the paper.

The impact of leakage can be incorporated into the transfer functions of the complex head and discharge. The impedance transfer functions for the various leaking pipeline systems can be derive. The leak calibration can be performed by the Genetic Algorithm (GA) and Harmony Search (HS). The objective functions for the leakage detection can be made using the pressure-head response at the valve, or the pressure-head or the flow response at a certain point of the pipeline located upstream from the valve. The proposed method is not constrained by the Courant number to control the numerical dissipation of the method of characteristics. The limitations associated with the discretization of the pipeline system in the inverse transient analysis can be neglected in the proposed method.

Artificial neural networks (ANNs) are applied to estimating nitrate concentrations in a typical Midwestern river, i.e., the Upper Sangamon River in Illinois. Throughout the Midwestern U.S., nitrate in raw water has recently become an increasingly important problem. This is due to recent changes in the USEPA nitrate standard and to the increasingly widespread use of chemical fertilizers in agriculture. Back-propagation neural networks (BPNN) and radial basis function neural networks (RBFNN) are compared as to their effectiveness in water quality modeling. Training of the RBFNN is much faster than that of the BPNN and yields more robust results. These two types of ANNs are compared to traditional regression and mechanistic water quality modeling, based on overall accuracy and on the frequency of false-negative prediction. The RBFNN achieves the best results of all models in terms of overall accuracy, and both BPNN and RBFNN yield the same false-negative frequency, which is better than that of the traditional models.

Development of a standard or standard guideline takes considerable time of the contributors and also requires considerable patience on the part of the editor. The changing structure of ASCE, heavy workloads of those involved, and the balloting process require patience on the part of the editor. Some of the problems encountered by the editor are described as well as ways to prevent or deal with the problems as they arise.

In an effort to develop consensus standards for the evaluation of aeration devices in both clean and process waters, the ASCE established a subcommittee on Oxygen Transfer Standards under the Technical Council on Codes and Standards in 1977. An initial meeting of a core group of interested individuals in 1997 determined that such an undertaking would be significant. Current standards for oxygen transfer testing were inadequate and often misused by consultants, manufacturers, and users. In order to conduct this work, funding was obtained from USEPA. A cooperative research agreement was obtained in 1978 to conduct preliminary review and assessment of oxygen transfer standards and to identify critical areas of uncertainty in the existing procedures. A workshop was held in 1978 whereby a number of experienced engineers and scientists began the process. This paper describes the chronology of the process and illustrates the importance of USEPA funding throughout the development of three important standards for oxygen transfer testing.

The economic effects of subsidence are primarily a result of increased maintenance or replacement of permanent facilities and buildings. There is also a secondary economic effect resulting from loss of buildable land and decay of neighborhoods. Once the cost of maintenance and replacement are known there is a problem of assigning blame and responsibility for costs. Assignment is difficult, requiring the development of pumping schedules through special district creation or through court action. The development of guidance documents, such as standard guidelines, on the control of land subsidence due to groundwater withdrawal would be helpful to the engineering community.

The Artificial Recharge of Groundwater Standards Committee is starting work on a Standard Guideline on Land Subsidence Related to Groundwater Overdraft. Part of the task will be to examine areas that have suffered land subsidence and to draw conclusions from the event. One of the case histories to be examined is in Santa Clara County, California where land subsidence caused by overdraft of a ground water basin was arrested. The studies conducted by the Santa Clara Valley Water District (SCVWD) initially in cooperation with federal and State agencies and later by it's own staff are described

The American Society of Civil Engineers (ASCE) encourages broader participation by its Institutes and the other technical councils in the development of manuals of practice, codes and standards, conferences, and other appropriate continuing education materials. This Chair of the EWRI Standards Development Council (SDC) recommends the continuing development of a private sector-government-academia partnership, including financial support by all sectors, in order for professionals to participate fully in the development of all EWRI/ASCE documents that will benefit partnerships for the entire world. The need to convert many of the current ASCE or other manuals of practice to Environmental & Water Resources Institute (EWRI) standards is emphasized. Updated data analyses are needed for many river basins; along with knowledge about water conservation, reclaimed water, replacement of pipes, land use planning, flood control, coastal protection, desalinization, evaporation suppression on reservoirs, dam safety, aquifer storage and recovery, and atmospheric water management that can be further developed in the 21^st Century.

The approval process for the development of the new Water Infrastructure Security Enhancement (WISE) Standards Committee (SC) from 2002 through 2003 is described. The complete approval process by the Environmental & Water Resources Institute (EWRI) Standards Development Council (SDC) and its EWRI governing groups is provided along with the required approval of the American Society of Civil Engineers (ASCE) Codes & Standards Activities Council (CSAC). The development of the proposal for federal agency support and funding is provided. The information on the creation of three subcommittees of the WISE SC is recognized. The subcommittees are now called the Methodology & Characteristics (M&C) Subcommittee, the Water Supply (WS) Subcommittee, and the Wastewater & Stormwater Management (W&SM) Subcommittee. The estimated tasks for the pre-standards and standards development assignments for the next three years under this WISE SC for United States federal agencies is described. This brief proposal information going to the agencies should be reviewed and accepted sometime in 2003.

This paper is aimed at characterizing turbulence in bubble plumes undergoing a transient motion, a topic of primary importance in many environmental applications. Mathematical models for dilute mixtures, derived from the two-fluid model equations (multiphase flow theory), are initially presented. These models include diverse degrees of complexity, and they handle turbulence via a k - ε model and a Large-Eddy Simulation (LES) approach, in a consistent way. The models are implemented in a parallel code, which is then used to numerically simulate the dynamic behavior of bubble columns in three dimensions, using the LES approach. The results of high-resolution simulations show the presence of coherent structures whose distributions in space, and number, change with time. These structures are characterized in the paper via several indicators, connecting them with potential events of breakup of bubbles.

Pretreatment of slow sand filter influent is necessary to remove colloidal clay contaminants resulting from source watershed erosion. In pilot-scale studies using slow sand filters fed from a natural river source, horizontal-flow roughing filters containing basaltic river rock and calcite limestone media were tested for water chemistry changes over sixty-day periods and turbidity removal during clay challenge studies. Roughing filters with calcite or basalt media alone failed to adequately pretreat source water, but a calcite-amended rock roughing filter in combination with slow sand filtration enhanced effluent pH, alkalinity, and saturation index values, and this pretreatment filter system produced effluent compliant with a 1 ntu turbidity regulatory requirement for potable water.

The Natural Resources Conservation Service (NRCS), formerly known as the Soil Conservation Service (SCS), has developed hydrologic procedures for ungaged watersheds. NRCS initially used manual procedures to design conservation measures at the farm level. The agency was authorized to make small watershed studies in 1954 and this encouraged NRCS to develop uniform manual procedures that could be used for ungaged watersheds across the United States. As computers became available, NRCS used them to develop and refine hydrologic procedures. This paper will explain the evolution of the current hydrologic procedures.

In the early 1960's the Stanford Watershed Model (SWM) was instrumental in introducing the civil engineering profession to the concept of continuous hydrologic modeling. With wide distribution and application of the SWM in the late 1960s, civil engineers recognized the value of digital continuous simulation for hydrologic applications. By the early 1970s the developers of SWM expanded and refined SWM to create the Hydrocomp Simulation Program (HSP), which also included general nonpoint source loadings and water quality simulation capabilities. Hydrocomp had demonstrated the utility of quantity/quality simulation by modeling a range of water quality constituents in a large basin. During the early 1970s EPA sponsored development of the ARM (Agricultural Runoff Management) and the NPS (Nonpoint Source) pollutant loading models to address pollution from agriculture, urban, and other land uses; the SWM approach was selected as the hydrologic foundation for an expanding suite of models of nonpoint pollution impacts. In the late 1970s EPA recognized that the continuous process simulation approach contained in all these models would be needed to analyze and solve many complex water resource problems. Grant money from the agency to Hydrocomp resulted in the development of the Hydrological Simulation Program - FORTRAN (HSPF), a non-proprietary system of simulation modules in standard Fortran that handled essentially all the functions performed by HSP, ARM and NPS, and was considerably easier to maintain and modify. HSPF simulates the hydrologic and associated water quality processes on pervious and impervious land surfaces and in streams and well-mixed impoundments. Since the first public release (Release No. 5) of HSPF in 1980, the model has undergone a continual series of code and algorithm enhancements producing a succession of new releases, leading up to the most recent Release No. 12 in 2001. Since 1981, the U.S. Geological Survey has been developing software tools to facilitate watershed modeling by providing interactive capabilities for model input development, data storage and data analysis, and model output analysis including hydrologic calibration assistance.

The Hydrologic Engineering Center, HEC, was established in 1964 to provide a center of expertise for US Army Corps of Engineers field offices in the newly emerging area of hydrologic engineering. The mission was to combine research, technical assistance, and training efforts to develop and implement the latest hydrologic engineering technology for Corps offices and their partners in the profession. Research emphasis would be on bridging the gap between basic research at universities and the practical applications needs of the field offices. This applied research would focus on particularly difficult hydrologic engineering problems in field offices, thus both developing new technology and providing technical assistance to the field. As the new technology was developed, formal training courses, field workshops, and professional society publications would make it available to all Corps offices and the profession. This paper will discuss the formation of HEC, its organization and philosophy, as pertains to development and use of hydrologic engineering technology over the last 39 years. Emphasis has always been on development of generalized, non-location-specific computer programs that are distributed in the public domain. This results in worldwide testing and information sources for new and improved software capabilities and applications methodologies. Emphasis is also placed on standardizing software design and operational methods to minimize learning time and errors in application. The impact of the growth of computer resources on the evolution HEC's software is described as well as recent advances in a new generation of software.

This paper focuses on the rainfall-runoff (surface hydrology) routines of the EPA Storm Water Management Model (SWMM). These include methods for generation of overland flow as well as precipitation input options and loss (abstraction) options.

Lighthill and Whitham developed the kinematic wave theory but Iwagaki also independently conceived the kinematic wave concept. Over a decade later, the theory started to receive wide acceptance. By the 1980s, the theory became an accepted tool for modeling not only surface runoff but also for subsurface flow, soil moisture movement, macropore flow, snowmelt runoff, and soil erosion. Many watershed models began incorporating the kinematic wave method for modeling overland flow. The decade of the 1990s also witnessed applications of the theory to a variety of other hydrologic processes. This study provides a historical perspective of the kinematic wave theory-based models in hydrology.

The South Florida Water Management District operates a system of over 1000 miles of primary canals, several million acre-feet of active water storage and over 200 primary water control structures to manage multi-objective demands of flood control, water supply, and environmental enhancement. This system is growing very rapidly in scope and complexity in response to both State and Federal environmental restoration projects. A prototype operations decision support system has been under development for several years and is currently used as an important component of managing the water control system. This paper will focus on the design and operational philosophy of the decision support software.

Before the decade of the 1960s, watershed modeling was primarily confined to modeling of individual components of the hydrologic cycle. This was principally due to limited computing capability and the availability of limited data. Since the advent of computers in the 1960s and the exponential growth of computing capability in the decades to follow, the watershed modeling started becoming more comprehensive. Since the groundbreaking development of Stanford watershed model by Crawford and Linsley in 1966, numerous watershed models have been developed. The development of more and more integrated and comprehensive models continues today. This study traces the evolution of watershed models before and during the computer era. Watershed models have not only become more comprehensive but have also been integrated into larger ecosystems and watershed planning and management models.

Hungary is situated in Europe, in the deepest part of the Carpathian Basin. Regardless of the basin character, hills rising up to 1 000 m cover slightly more than one-half of the country the valleys of which are farmed and accommodate a number of communities. These facts lend special importance to water management in these mountain-hilly areas. As a consequence of adverse natural conditions, flash floods and the inundations in their wake, poor land use and neglect of soil conservation, close to one-quarter of the country's area is affected more or less severely by erosion. Watershed control, training of mountain streams and rivers, soil conservation are therefore the problems that require urgent solution.

NPDES and TMDLs have brought a new awareness of stormwater quality issues to communities across America the last several years. As these communities have moved forward with stormwater utilities and begun implementation of retrofit projects, there have been numerous questions regarding the selection of the sometimes bewildering array of newly developed BMPs. Many of these smaller end-of-pipe BMPs are proprietary systems marketed by vendors who purport product effectiveness based on a wide range of testing programs. To date there has been no standardized testing protocols or programs to provide a realistic comparison of these BMPs to verify manufacturer's claims of pollutant removal efficiencies. Historical BMP testing has consisted of field monitoring of installations under a wide range of pollutant loadings, storm events, monitoring procedures, and product sizes. This has led to obvious limitations on pollutant loadings and repeatability of measured results. Any testing program which is rainfall driven will be handicapped by these limitations. Due to these highly variable conditions, it has not been possible to make accurate predictions of BMP effectiveness over the wide range of real world conditions that confront a community's engineer tasked with designing BMPs to meet regulatory goals.

This paper summarized the results of computational modeling study for the confluence of the Kankakee and the Iroquois Rivers. This project aims to study the effectiveness of engineering alternatives on reducing sedimentation at the confluence. The hydraulics and sediment transport patterns of three management scenarios, which are keeping natural situation without engineering structures, constructing three short dikes at the left banks of the Kankakee River, and constructing three longer dikes at the left banks are studied by applying the EnSed2D model. The sediment transported in the Kankakee and the Iroquois Rivers are primarily suspended sediment. Channel bed has a thin layer of bed material, and occasionally bed rocks are exposed. Therefore, this study focused on the simulation of suspended sediment transport in the system. Two methods were applied to simulating suspended sediment deposition and erosion processes. One method assumes that bed material layer is too thin to allow suspended sediment concentration reach equilibrium at the bottom that only deposition occurs, the other method assumes there is a sufficient amount of sediment that can be entrained from channel bed so that the change of bed elevation is the difference between the rate of deposition and entrainment.

Transmission losses along ephemeral channels are an important, yet poorly understood, aspect of predicting runoff rates. Channel transmission loss research at the U.S. DOE/NNSA, Nevada, began after a February 1998 storm caused a large runoff event to occur within a small, isolated watershed. Two transmission loss field experiments are on-going at the NNSA. These experiments have utilized water discharged from unrelated drilling activities. The first experiment, conducted in 2001, used an incised channel, created by water discharged during well drilling. The second experiment, conducted in 2002 and 2003, uses a channel constructed several decades ago. Transmission losses are quantified during well development and aquifer pump test discharges. Transmission losses are quantified using three different methods. Method 1 uses discharge measurements at several flumes located along the channels. Flume locations were chosen in relation to geomorphic surface types and ages, vegetative cover and types, subsurface indurated layers (calcrete), channel slopes, etc. Method 2 uses heat as a subsurface tracer for infiltration. Numerical modeling is used to estimate infiltration from soil temperature data. Method 3 uses hydraulic gradient and water content in a Darcy's Law approach to calculate one-dimensional flow rate. Heat dissipation and water content data were collected for this analysis. This paper describes initial transmission loss calculations and soil hydraulic property responses for the first field experiments. Preliminary results indicate that up to 70 % of flow is lost in reaches of 1000 to 2000 m. Transmission loss is less in reaches where geomorphic surface ages are older and calcrete is present. Thermal gradients correspond well with channel and surficial geomorphology and measured losses. Larger thermal gradients are measured on older geomorphic surfaces with buried calcrete, suggesting heat transfer by conduction, and lower infiltration rates. Smaller thermal gradients, suggesting advection-dominated heat transfer and higher infiltration rates, are observed in reaches on younger geomorphic surfaces.

The Hydrologic Modeling System (HEC-HMS) is designed to perform deterministic hydrologic simulation in support of engineering studies. It is intended to assist in planning, designing, and operating projects by providing information about current and future runoff from watersheds, with and without water control structures. It is the integration of models for simulating the hydrologic cycle with models for simulating the operation of reservoirs, levees, diversions and other structures that make the program useful. Until recently, the program lacked the capability to address some types of studies. Changes in the program framework have been made and new features added to make it more flexible, robust, and complete. The reservoir element included in the program is used to represent any lake, detention basin, or reservoir with a storage-outflow relationship. The program now allows for explicit, physical description of each outlet from the reservoir. Transient events such as a dam failure can also be simulated. The new capabilities are implemented using an adaptive time-step, iterative solution of the continuity equation following Brent's method.

Estimates of sediment yield are required in a wide spectrum of water resources engineering problems. The non-linear nature of suspended sediment time series necessitates the utilization of non-linear methods for the forecasting study. In this study artificial neural networks, a well known non-linear method, are employed to forecast the daily total suspended sediment amount on rivers. The neural networks are trained using the rainfall data, recorded on the river catchment, the river flow and suspended sediment data belonging to Juniata Catchment in USA. The simulations provided satisfactory forecasts in terms of the selected performance criteria comparing well with conventional methods.

Water management decisions frequently have significant impacts on the public. These decisions are usually founded on a wide variety of complex technical data and information, coupled with values often embodied in shared beliefs. The challenge is to ensure that the public understands and is involved in the decision-making process. The International Joint Commission, a binational organization established by the Boundary Waters Treaty of 1909 to help prevent and resolve disputes primarily concerning water quantity and water quality in U.S. - Canada transboundary waters, controls flows at 19 structures along the boundary. In making its decisions, the Commission emphasizes joint fact-finding among technical experts in both countries and consensus decision-making by the Commission boards that actually control these flows. Over the past decade, the Commission has worked to merge this joint technical work with a major effort to increase public education and involvement. This paper describes how the Commission reconciles differing technical methodologies, units of measurement, and approaches used by different organizations in a process that integrates the public into the process. It highlights new approaches currently being explored to help the public better understand and become more involved in the complexities of water management decisions.

Sri Lanka's Coastal Zone has its landward jurisdiction over rivers, streams, lagoons, estuaries, wetlands or any other body of water connected to the sea either permanently or periodically. The coastal Zone, is very rich in environmental resources such as water supply & flow, lagoon fishery, mangroves, shrub forests with there wildlife, agricultural land and dynamic beach environment etc. These highly productive ecosystems which comprised unique coastal habitats serve as a vital base for the country's growth. The coastal zone, while extremely valuable to the country's economy, is also very fragile and highly vulnerable to many dynamic processes which occur on both land and the sea, shaped by natural and man induced forces. As a result of this process and unsustainable extractive practices a number of critical issues have been appeared within the coastal zone. The inability of the resource base to renew itself under heavy exploitation and growing population has led to create social and economic problems within the coastal zone during the last few decades. As a policy response to the above situation, Sri Lanka government placed great emphasis on adopting more technical oriented strategies. With the realization of low level of positive results of the interventions and increasing capital expenditure on engineering based management interventions Sri Lanka government shifted its policy approach from coast protection to Integrated Coastal Resource Management (ICM) by preparing and implementing National Coastal Zone Management Plan (CZMP) from 1990.

The year 2003 marks the tenth anniversary of the signing of the North American Agreement on Environmental Cooperation by Canada, Mexico and the United States (NAAEC). The purpose of the treaty is to promote environmental protection and sustainable development in North America and increase cooperation between the signatories. The Commission for Environmental Cooperation of North America (CEC) was created to implement this transcontinental environmental cooperation. Similarly, 2003 also marks the tenth anniversary of the Agreement between the Government of the United States and the Government of the United Mexican States Concerning the Establishment of a Border Environmental Cooperation Commission and a North American Development Bank (BECC-NADB Agreement). This agreement created the Border Environmental Cooperation Commission (BECC) and the North American Development Bank (NADB). The purpose of BECC is "to promote the conservation protection and improvement of the environment in the border region, by developing and certifying environmental infrastructure projects through an open and transparent process and with broad public participation." The NADB is described as "a financial institution established and capitalized in equal parts by the United States and Mexico for the purpose of financing environmental infrastructure projects along their common border." NADB only finances projects certified by BECC. This paper reviews the activities and programs that resulted from the North American Agreement on Environmental Cooperation and the BECC-NADB Agreement and how they have evolved over the past nine years. This work in progress supports a future EWRI publication entitled "Crossing Boundaries, Transboundary Environmental Management in Mexico, Canada and the USA after NAFTA" under the sponsorship Transboundary Environmental Management in Canada, Mexico, and USA after NAFTA Task Committee.

A variety of statistical tests were executed on 64 annual maximum discharge series observed in Israel during the last six decades. The period of observation at each of the stations selected is at least 25 years, with an average of 42 years per station. The tests aimed at detecting trends in the central values of the discharges, as well as in their variation. Except for one region, the detected changes are of local extent, if any. A decline in the annual maximum discharges and in their variability is found for the Upper Jordan River and its tributaries. A general rise in the variability is found over most other regions. These appear resulting from changes in the meteorological causes of high discharges. Few high discharges occurred in the 1970s and in the early 1980s, whilst an abundant number occurred in the 1990s.

The EU member States have started to implement the long discussed European Water Framework Directive. This Directive seeks to protect and improve the terrestrial environment through a sustainable use of water resources and introduction of protective measures. The water resources authority of each defined river basin district is required to analyze the water environment and introduce measures to improve it if necessary. Such an analysis is carried out for the Piemonte Region, Italy, with emphasis on agricultural and industrial load to the groundwater and surface water system. The region covers an area of more than 25.000 km² and several water resources numerical tools have been applied as support for the planning. Models for plant growth and nitrogen consumption, groundwater and surface water flow and quality and overall water and solute balances are introduced. The models are applied to investigate various schemes for reorganization and/or rationalization of the water use in the region with a focus on the overall water management and water budget as well as water quality. The paper describes the development of the various models for the Piemonte Region.

Although the classical irrigation efficiency concept is appropriate for irrigation system design and management, it could lead to erroneous conclusions and serious mismanagement of scarce water resources at a larger scale. This is because the classical approach ignores the potential reuses of irrigation return flows. To overcome these limitations, researchers have proposed the concept of water use efficiency at the basin scale, or basin efficiency, which takes into account the potential recycling and reuse of return flows. This paper discusses the implications of this new concept for hydrologic science and water management policies. An example for the assessment of basin efficiency is presented for the Yellow River Basin. Basin efficiency (BE) depends on the level of recycling and reuse of return flows as well as on local water use efficiency. Engineering measures such as conjunctive surface and ground water uses, storage regulation, and drainage treatment can increase BE through enhancing water recycling and reuse. Economic incentives can also improve the BE level by facilitating the reduction of non-beneficial water consumption or water losses. For water management and allocation decisions the BE indicator should be used together with economic efficiency indicators. For basins with competition among off-stream water demand and in-stream or in-site ecological water demand, the evaluation of water management in the basin should take into account BE in addition to other indicators, such as the criticality ratios and consumption ratios, as well as BE.

A detailed study of the fertilizer runoff from 26 ha of rain-fed rice cultivation in a small (100 ha) catchment was carried out to determine the contribution to nutrient fluxes observed in rivers. The discharge of water and the concentrations of NO3-N, NH3-N and PO4-P in the stream draining the fields were measured. The fluxes of Nitrogen and Phosphorous obtained by multiplying these quantities together were compared with data on fertilizer application obtained by a detailed weekly survey of the 77 farmers in the catchment. The results showed that about 874 kg (52%) of the total of 1665 kg of nitrogen applied was lost as NO3 and NH3 in the surface runoff. However, the runoff was very variable and strongly related to the rainfall with nearly 90% of applied N being lost during two weeks of heavy rain. The reason for the high losses was that the fields were flooded to overflowing by the heavy rains. The proportion of applied phosphorous lost as PO4 was about 6%.

Spatial-Stochastic Neural Networks Model(SSNNM) is used to estimate long-term streamflow in the parallel reservoir groups. SSNNM employs two kinds of backpropagation algorithms, based on LMBP and BFGS-QNBP separately. SSNNM has three layers, input, hidden, and output layer, in the structure and network configuration consists of 8-8-2 nodes one by one. Nodes in the input layer are composed of streamflow, precipitation, evaporation, and temperature with the monthly average values collected from Andong and Imha reservoir. But some temporal differences apparently exist in their time series. For the SSNNM training procedure, the training sets in input layer are generated by the PARMA(1,1) stochastic model and they covers insufficient time series. Generated data series are used to train SSNNM and the model parameters, optimal connection weights and biases, are estimated during training procedure. They are applied to evaluate model validation using observed data sets. In this study, the new approaches give outstanding results by the comparison of statistical analysis and hydrographs in the model validation. SSNNM will help to manage and control water distribution and give basic data to develop long-term coupled operation system in parallel reservoir groups of the Upper Nakdong River, South Korea.

Qatar is a rich country due to oil and gas reserves. It lies in Arabian Gulf that is an arid region and as a result it is freshwater deficient. Due to lack of fresh water resources and the growing water requirements, the country relies heavily on water desalination. Water is supplied free of cost to the nationals and partially subsidized to the expatriates. The per capita supply of water is over 600 liters/person/day that includes domestic, commercial, public and industrial uses. Precious drinking water is not only used for municipal and industrial purposes but also for irrigating lawns and gardens. It is a real challenge to continue to meet such a demand in future, that too free of charge. Huge sum of money is required for desalination and more investment is needed for quality control and maintenance, that makes the tap water very precious. In fact a liter of oil in the country is cheaper than a liter of water. The objective of this paper is to describe the existing situation and the problems associated with subsidized water supply. Necessary line of action has been proposed herein to minimize the impact particularly on domestic and public water consumption.

International water resources management programs and projects (hereafter, projects) historically have sometimes failed to meet increasingly complex societal demands and multiple objectives. The international agencies, banks and national governments have detailed project review criteria, but such reviews often have been narrowly focused. There are few systems in place that assist with formulating, assessing, implementing and monitoring these international projects in an integrated fashion, directed toward the goal of water resource sustainability. Projects are also being assessed for `environmental due diligence' by the project proponents, but without a comprehensive system or procedure. The International Organization for Standardization (ISO) has formulated the ISO 14000 series of standards to improve environmental performance of an organization. ISO 14001 is a standard that provides the "core" requirements for developing and implementing Environmental Management Systems (EMSs) that can be "certified" or registered by an external third party. This paper proposes the use of ISO14001 gaps analyses (a gap defines the difference between the existing condition and ideal EMS compliance), using water-related EMSs, to improve the economic/environmental performance and sustainability of projects addressing or affecting regional or national water resources at various stages of project development. Development of comprehensive EMSs could be undertaken by organizations such as the World Bank or United Nations agencies. The process of defining sustainability will involve many facets and levels of society and could consume decades to become integrated in decision-making. Local, regional and national viewpoints from many disciplines are required to fully develop sustainable economic planning for a particular country, as well as resolve international water conflicts. This paper is intended to begin a dialogue on the development of a systematic review process that leads to more comprehensive tools for improving international water project performance relative to sustainability goals.

A Bayesian methodology is used to assess the information content of categorical, probabilistic forecasts of seasonal-mean precipitation derived from General Circulation Model (GCM) forecast ensembles. A "prior" forecast given by climatological probabilities of each category is combined with categorical probabilistic forecasts derived from several GCM ensembles to develop posterior categorical probabilities. The combination algorithm assigns a weight to each particular model forecast and to the climatological forecast. Several improvements are described, that produce more robust results when many models are combined together.

From the General Circulation Models(GCMs), it is known that the increases of concentrations of greenhouse gases will have significant implications for climate change in global and regional scales. The GCM has an uncertainty in analyzing the meteorologic processes at individual sites and so the "downscaling" techniques are used to bridge the spatial and temporal resolution gaps between what, at present, climate modellers can provide and what impact assessors require. This paper describes a method for assessing local climate change impacts using a statistical downscaling technique. The method facilitates the rapid development of multiple, low-cost, single-site scenarios of daily surface weather variables under current and future regional climate forcing. The construction of climate change scenarios based on spatial regression(transfer function) downscaling and on the use of a local stochastic weather generator is described. Regression downscaling translates the GCM grid-box predictions with coarse resolution of climate change to site-specific values and the values were then used to perturb the parameters of the stochastic weather generator in order to simulate site-specific daily weather values. In this study, the global climate change scenarios are constructed using the YONU GCM control run and transient experiments.

Each month, the International Research Institute for Climate Prediction (IRI) makes forecasts for the global temperature and precipitation for several overlapping 3-month periods extending to 6 months in advance. The forecasts are based mainly on the indications of five dynamical atmospheric general circulation models, whose behaviors are governed mostly by predicted global sea surface temperature (SST) anomalies. The IRI's forecast procedure begins with the prediction of SST anomalies, which are then used as a basis for prediction of the climate over land. The historical skill of the atmospheric models in hindcast mode, and of the real-time forecasts over 4 recent years, are discussed. Due to the large uncertainty in most climate forecasts, the IRI's forecasts are expressed as probabilities of pre-defined tercile categories (below, near, and above normal). The real-time precipitation forecast for an upcoming 3-month period is used as an example for illustrative purposes. The meaning and utility of such forecasts will be discussed in the context of the users' cost/loss contingencies, combined with the fact that most forecasts do not deviate drastically from the climatological distribution. Forecast formats that would be most suitable for the largest proportion of the user communities will be explored, together with prospects for their production.

We present an ensemble local hydrologic forecast derived from the seasonal forecasts of the International Research Institute (IRI) for Climate Prediction. Three-month seasonal forecasts were used to resample historical meteorological conditions and generate ensemble forcing datasets for a TOPMODEL-based hydrology model. Eleven retrospective forecasts were run at a Florida and New York site. Forecast skill was assessed for mean area modeled water table depth (WTD), i.e. near surface soil wetness conditions, and compared with WTD simulated with observed data. Hydrology model forecast skill was evident at the Florida site but not at the New York site. At the Florida site, persistence of hydrologic conditions and local skill of the IRI seasonal forecast both contributed to the local hydrologic forecast skill. At the Florida site, we have also quantified the link between modeled WTD (i.e. drought) and the amplification and transmission of St. Louis Encephalitis (SLE) virus. We derive an empirical relationship between modeled land surface wetness and levels of SLE virus transmission associated with human clinical cases. Using this relationship, we then produce year-round retrospective forecasts of epidemic SLE virus transmission in Florida. Epidemic SLE virus transmission forecast skill is demonstrated and a real-time epidemic SLE virus transmission forecast is presented. A second, probabilistic, epidemic SLEV transmission forecast for June to August, the critical early transmission phase of the disease, is in development.

The hydraulic performance graph (HPG) is an efficient means for summarizing the backwater profiles of all possible gradually varied flow conditions in an open-channel reach, expressed in the form of water surface elevations (stages) or depths at the ends of the channel reach for different constant discharges. HPGs have been used for determining the flow carrying capacity of channel system. Further, Gonzalez and Yen have also carefully assessed the applicability of HPG for unsteady flow routing. Theoretically, the HPG for a channel reach is a graphical representation of the relation between discharge and the downstream and upstream stages in the reach according to gradually varied flow. In this paper we explore the applicability of HPGs for estimating the flow in an open-channel reach from observed water stages at its two end stations. We describe the methodology for developing the HPG for a site in South Florida based on actual channel geometric and roughness properties and compare the discharges estimated with the reach's HPG for actual observations of upstream and downstream stages with field-flow measurements concurrently collected with an acoustic Doppler current profiler. This comparison shows that the HPG of channel reach is a plausible and reliable alternative for estimating flows in open channels. It is also shown that the accuracy of the HPG for flow rating is a function of S_f L / H, where S_f = friction slope, L = length of the channel reach and H is the error in estimating the water stages. HPG ratings are found relatively accurate for S_f L / H >> 1, but their accuracy decreases as S_f L / H → 1.

Grid-based daily soil moisture balance model coupled with dry spell analysis of the soil moisture condition were used to assess the sustainability of the practice of rainfed agriculture in Pangani Basin, Tanzania. On the basis of 20% exceedence probability as dependable soil moisture condition for rainfed agriculture, the analysis results show that rice cultivation is impossible without irrigation in the basin. The probability of failure during the rain season exceeds 80% almost in the entire basin area. However, maize cultivation may be possible in areas around the foot of Mt. Kilimanaro and Mt. Meru, Kirua swamp and Further to the mouth of the river. Beans may generally be considered for rainfed agriculture without any supplementary irrigation in the basin. The probability of failure was well below 20% in the vast majority of the basin. It has been learnt that beans can grow in low water availability areas and is considered as drought resistance crop. It can generally be concluded that mixed agriculture may be sustainable to produce main food items like rice and maize in the basin.

As drought sweeps through the Southwest of the United States, water conservation strategies have become more urgent for all entities responsible for water resources management in the Paso del Norte Region. Preliminary results have shown that there is a great potential for improvement of water delivery efficiency (water savings or reduction in water diversions for a given level of surface water diversion) through lining of canals within this region. However, water losses need to be determined accurately prior to the execution of the project since canal lining is a project that requires a significant investment of monetary resources. This paper describes results of water losses studies in the region using ponding tests. Five ponding tests were conducted on the Franklin Canal in El Paso, Texas to determine water losses at selected canal sections. The results of ponding tests indicate that the water losses for the Franklin Canal range from 0.0042 to 0.0132 cubic meters per second per kilometer. The results also indicate that seepage rates vary spatially and possible temporally because of different soil settings of canal beds and hydrological conditions along the canal. Using ponding test results, authors assess amount of water that could be salvaged by lining the Franklin Canal, and conclude that there is a great potential for water conservation and improvement of delivery efficiency by lining canals.

A tool is presented for modeling complex stream-aquifer interactions by integrating an artificial neural network (ANN) with a geographic information system (GIS) and a numerical groundwater model. The proposed methodology allows development of dynamic, spatially dependent relationships between measurable aquifer stresses and river return flow. A well calibrated finite difference groundwater model is used to represent the system stream-aquifer interaction for training the ANN. ArcGIS (ESRI) is used to process and prepare the extensive spatial data sets for training, validation and testing of the ANN. MATLAB (MathWorks, Inc) performs training and validation of the ANN, with VBA interfaces embedded in the GIS for manipulating the finite difference model and preparing the required ANN spatial data set. A significant feature of the proposed tool is the ability to generate the ANN input/output data set based on physical-spatial system characteristics. The tool is applied to the Lower Arkansas River Basin in Colorado as a case study. Preliminary results show the feasibility of using the proposed methodology for developing more robust stream-aquifer modeling tools for incorporation into river basin management models.

Reuse of treated wastewater in irrigation is gaining recognition as a vital element in the water resources management plan of developing countries, especially those situated in arid and semi-arid regions. An understanding of the transport of residual pollutants from treated wastewater, such as bacteria, in soil as a result of irrigation is critical to assessing health risks and the possible contamination of limited groundwater resources. In this work, retention of E. Coli by adsorption is evaluated for a soil that is irrigated by treated wastewater for growth of non-food crops near Egypt's Red Sea coast. In particular, the effect of soil organic content (SOC) was investigated using the batch isotherm method. Five different organic fractions ranging from 0.03–2 percent were tested, with soil amendment achieved using uncomposted horse manure. A linear isotherm model provided an adequate description of the data, with the value of the partitioning coefficient, K_P ranging from 1.8 ml/g (0.03 percent SOC) to 10.3 ml/g (2 percent SOC). A semi-log linearization of K_P versus SOC, similar to that developed for adsorption of organic compounds onto soils, produced an empirical expression to predict K_P, namely: K_P = 1.89 ln (SOC) + 8.38, with a correlation coefficient, R² = 0.95.

The project area comprises the 70 km² headwater drainage basin of Slippery Rock Creek in northwestern Pennsylvania. Slippery Rock Creek and its tributaries have been adversely impacted by acid mine drainage for over a 100 years and at present, over 75 groundwater discharges from abandoned coal mines have been identified within the watershed and of these, 59 are contributing acid load to receiving streams. The pH of these streams varies from 2.5 to 6.0 with acidity exceeding alkalinity and heavy metals varying from 2 to over 100 mg/L. To date, 15 passive treatment systems are removing 30% of the acid loading and contribute 193 kg/day of alkalinity to receiving streams with the treated effluents maintaining pHs between 6.8 and 7.2. In addition, these systems are removing over 27% of the metal loading to receiving streams. Biological surveys of the Slippery Rock Creek and its major tributaries revealed that fish and macroinvertebrates are beginning to recolonize the streams below the discharges from the passive treatment systems.

Oxygen levels are depleted seasonally in the Gulf of Mexico where the Mississippi River waters enter the gulf. This hypoxic zone, or "dead zone", extends thousands of square kilometers and causes a decrease in aquatic habitation. In addition, the seafood industry suffers a loss of income from the decreased availability of healthy aquatic life. The primary cause of this oxygen depletion is theorized to be the addition of fertilizer, especially nitrogen, in the Mississippi River watershed. This watershed encompasses 41% of the contiguous United States and contains a large portion of the nation's agricultural land. The additional nitrogen increases algal blooms that consume valuable dissolved oxygen, thereby making less available for other aquatic life, including seafood crops such as shrimp and fish. Nitrate can also replace oxygen in the bloodstream and thereby internally suffocate aquatic life. The situation has worsened in the last 50 years and there are few regulations governing the situation. The issue is politically charged between midwestern state's agricultural production and the seafood industry and environmental agencies. Possible solutions include regulation of nitrate addition, buffer zones next to waterways, or the development of organic fertilizers.

Soil aquifer treatment is the managed infiltration of water, often treated wastewater in recharge basins. Optimal operation of those basins is related to the rate of infiltration and achieving desired water quality improvement. Most often, a mass balance is used to compute the average basin infiltration rate. This averaging method does not provide insight into potential temporal changes in the infiltration rate during the recharge period and spatially varying rates of infiltration across the basin surface. Knowledge of both these factors is critical to engineering optimal basin operation. In order to address this shortcoming, automated infiltrometers have been developed and tested which can measure local infiltration rates (areal scale of ∼0.19 m² or ∼2 ft²) in near real time (time steps of ∼1.0 min). The automated infiltrometers were tested in a recharge basin operated by Tucson Water in Tucson, Arizona and performance data is presented.

Phosphorus management is required to reduce agricultural runoff and improve water quality. The phosphorus management options discussed in this paper include developing alternative uses for poultry litter, reducing the phosphorus content of the litter and reducing the runoff of soluble phosphorus. This paper evaluates the following alternative uses in terms of their technical and economic feasibility: compost, use as a cattle feed supplement, pelletizing, biogas generation, and electricity generation via direct combustion or gasification. The use of the enzyme phytase and high availability phosphorus corn as feed ingredients to reduce the phosphorus content of broiler litter are evaluated in terms of their effectiveness and market constraints. The use of alum as a soil amendment to bind soluble phosphorus to soils is evaluated relative to its efficacy in the soils of the Delmarva region. The research concludes that at this time, the use of phytase, and the development of poultry litter compost and pelletizing products offer the most viable means of managing phosphorus on the Delmarva Peninsula.

A simple model for event-based transient flow to a subsurface tile drain was developed. Green-Ampt theory was used to describe redistribution of infiltrated water in the vadose zone. Semi-analytical solutions in terms of Fourier series were sought for the Boussinesq equation describing subsurface saturated flow subjected to time-dependent recharge. Model results were compared with observed hydrograph data from the Purdue Water Quality Field Station (WQFS) in West Lafayette, Indiana. In general, the models were able to capture the sharp peak of the tile drain hydrograph, as well as the time to peak and the time of initial response to rainfall. The models performed particularly well for rainfall events with single-burst hyetographs, and in the prediction of the first hydrograph peak from a multiple burst hyetograph, though subsequent peaks could not be captured as well. We conclude that the solution developed here holds promise for use in larger watersheds where the hydrology is governed by tile drains.

Submerged weir flow is one of the most difficult flow conditions to accurately predict at gated spillways. Previous research work has mostly focused on submerged weir flow measured in small-scale laboratory models. In this paper, we develop a submerged weir flow equation based on field measurements collected with an Acoustic Doppler Current Profiler at several prototype gated spillways in southern Florida. The proposed method is developed using dimensional analysis. This method is compared with field measurements and several previously published submerged weir equations.

Climate change and its impacts on water resources are becoming increasingly significant in long term forecasting, planning, and management of water resources. The impacts range from erratic patterns of the hydrologic cycle to sea level rise in coastal areas. The rise in sea level and the associated advance in the saltwater-freshwater interface farther inland are investigated using a three-dimensional numerical code called SEAWAT. The impacts of pumping are coupled with the rise in sea level and prescribed recharge based on selected climate change scenarios. A case study considers a hypothetical aquifer of specified parameters and pumping conditions. Based on the results obtained, groundwater pumping coupled with a long term sea level rise exhibits a quantifiably significant accelerated shift of the saltwater-freshwater interface as compared to the scenario where no sea level rise is assumed.

It is well documented that children determine if they are "interested" in math and science by the time they are in middle school. Therefore, it is important to reach children at that age with fun activities that help them learn about math, science, engineering, and their environment. This paper will discuss several hands-on experiments used to teach children about erosion, run-off, pollution, watershed management, and water resources. The two primary apparatus are a river meander demonstration table and a soil erosion box. These devices were constructed using relatively inexpensive materials and could be easily replicated by other interested parties. Both are portable and have been used for outreach activities as well as on-campus instruction. The river table is a 20" wide by 48" long by 4" deep Plexiglas box filled with sand. Students are given the opportunity to interact with the model (with instruction) such that they investigate river dynamics, sediment transport, erosion, and erosion control (using model gabions, riprap, and sponges). The soil erosion box is an 18" wide by 36" long by 3" deep wood box that can be filled with various materials and placed at variable inclines. The simulated rainfall run-off and eroded sediment are caught in glass jars at the end of the box. The amount of water and sediment can then be measured and conclusions drawn by students on the effect of slope, soil type, and ground cover on erosion and erosion control techniques. These experiments have been performed by middle school, high school, and college students with degree of complexity and expectations increasing with age. Finally, this paper will also discuss the success and failures of these models in teaching children and make recommendations for implementing similar outreach programs.

The Penn-Alexander School's new building opened in September 2002, with classrooms for grades Pre K-8, and state-of-the-art specialized learning spaces. The school is located in a park-like setting featuring a playground, a playing field with a grass amphitheater seating area, a rain garden and a science garden. Located on a five-acre campus site between 42nd and 43rd streets from Locust to Spruce Streets, this unique public school has already been hailed as a "gold standard" by the National Association of Independent Colleges and Universities for bringing the most effective, research-proven educational practices into the classroom. The Penn-Alexander School opened its new building to children in grades K, 1, 2, 5 and 6 in September 2002, as well as a Head Start classroom, and should eventually house 700 students in grades PreK-8 by 2004. Already serving a diverse student body from the local community, the school will be a centrally located hub of neighborhood life and an active educational partner for other schools in West Philadelphia.

The U. S. Bureau of Reclamation is a federal agency in the Department of Interior that provides water management in the 17 western states. The U. S. Bureau of Reclamation (Reclamation) has been involved in environmental education to students in the Denver Metropolitan area through school visitations, water festivals, science fairs, and laboratory tours of the water, structural and environmental laboratories. Environmental demonstrations have been completed on meteorology, snow storms, ground water movement, wetlands, water quality, sedimentation, and dam failure. A number of diverse presentations have been made to students and other volunteer organizations to try to impart basic scientific information and to provide an early interest in science and engineering career paths. This paper will describe Reclamation's environmental education program and future possibilities with the program. Extensive effort by many scientists and engineers has resulted in a successful program through time.

Post-construction monitoring data at groundwater remediation sites need to be assessed to determine whether the remedies are performing satisfactorily and, if not, what might be done to improve performance. Such assessments may include determining whether there is adequate evidence that groundwater containment/restoration remedies contain the extent of target plumes of contamination when one analyzes all of the available information, including groundwater head and chemistry measurements, groundwater flow directions and travel-time frames derived from flow and transport simulation, and the constantly changing rates of pumping groundwater that are typical real-world conditions of pump and treat remediation. We discuss statistical and physics-based assessment of monitoring data from groundwater remedies. Our approaches are colored by the need to accommodate inconsistencies, errors, non-detects, and missing values present in real-world data sets. Statistical comparisons of pre- and post-construction data provide a starting point. Spatial correlation between field data may be used to evaluate redundancies. Data assimilation (DA) characterizes the true state of a groundwater system by appropriately (in some cases optimally) combining field data with predictions made by simulation models and error estimates. The field data set and flow and transport simulation programs are linked and coordinated through DA. The resulting analysis leads to improved estimates of state, which leads to improved performance evaluation on an ongoing basis and provides estimates that fill gaps in the information necessary for continuous optimization of remediation systems. Examples are presented from currently active sites.

The Coastal Aquifer in Israel, and particularly in the area of Tel Aviv, Israel, supplies a great deal of the drinking water needed in the Coastal Zone of the country. Various sampling operations that have taken place over the last few years in the Tel Aviv area have shown it to be contaminated with organic micropollutants (chlorinated hydrocarbons) in concentrations of the order of magnitude of, and even higher than, their threshold values. However, none of the sampling operations covered all of the relevant wells in the area, due to limitations in sampling resources and particularly in laboratory capacity. Therefore, each one of the sampling operations presented a totally different description of the contaminant distribution in the aquifer, together providing different pictures of the possible contaminant plumes. This study concerns the development of a new methodology for devising a monitoring plan for groundwater contamination. The objective of the research is to locate accurately and quickly the contaminant sources and plumes by taking a limited number of samples during each sampling operation. Furthermore, the results' reliability should be maximized for a given number of sampling operations per year, and a specified number of sampled wells per operation.

This paper presents a set of methods, data management and visualization tools, incorporated in the MAROS 2.0 software, for the analysis and optimization of existing long-term groundwater monitoring plans. The set of methods were designed to assess concentration trends, optimize the monitoring network and sampling frequency, analyze plume stability with spatial moment analysis, and evaluate well or risk-based site cleanup status using statistical power analysis. Data management tools provided within the software have the capability of site database management and automated raw data filtering, reduction, and smoothing. Visualization tools facilitate the interpretation of monitoring data and analysis results using automated, updateable database software that can be integrated with a geographical information system (GIS). These methods and tools can help formulate an updated monitoring plan that is more effective and efficient. At the Fort Lewis Logistics Center, Washington, this system was applied to assess possible improvements in the current groundwater monitoring network for a chlorinated solvent plume. Preliminary monitoring optimization results reduced the monitoring well network as well as the sampling frequency which provided a basis to make more cost effective, scientifically based future long-term monitoring decisions.

Best management practices (BMPs) for erosion and sediment control are measures designed to reduce the amount of sediment leaving a construction site and prevent that sediment from entering nearby surface waters. Failure modes of these BMPs were used to develop a Failure Modes and Effects Analysis (FMEA). By conducting a FMEA, all possible failure modes and resulting consequences were determined for the selected control measures. Risk priority numbers (RPN) were used to determine the relative risk of failure for each measure. These RPNs for the different control measures could be used to aid the user in the selection of an appropriate set of BMPs for the given situation.

Spillways at dams on the Columbia River and main tributaries are of the tainter-gate type, which discharge water beneath the gate structure when open. An experimental overflow weir was installed in one of the Lower Granite Dam (LGD) spillway bays in the Snake River in the Fall of 2001. Called the Removable Spillway Weir, or RSW, its purpose is to take advantage of the surface oriented behavior of migrating juvenile salmonids to provide a more natural entrance for downstream passage. This paper presents the first year of a multiyear study of turbulent flows adjacent to the RSW. The goal of this project is to relate turbulence characteristics in the LGD forebay to fish movement, fish behavior, and hydrodynamic modeling results considering the operation of the RSW. Tasks included in the first year of study include site preparation, stand construction, protocol testing, and velocity collection. Field data characterizing flow paths were collected in the near field of the LGD forebay using an Acoustic Doppler Velocimeter (ADV). Measurements were taken from the face of the RSW using a sampling platform equipped with several motors to allow for a full range of movement. Three dimensional velocity vectors were measured at a sampling rate of 25 Hz. Due to low suspended solids in the water, turbulence parameters could not be defined; only mean velocities were obtained. These and future data will be used to assist in CFD model validation, corroborate fish tracking data, assess juvenile salmonid response to turbulence levels, and to compare field data to natural turbulence levels in natural rivers upstream. Results from this study can be used to improve fish passage technology.

This paper presents a conjunctive 2-D surface and 3-D subsurface flow model, which deals with the interaction between surface and subsurface flows. The model uses the non-inertia approximation of the Saint-Venant equations for two-dimensional unsteady surface flow and a modified version of the Richards equation for three-dimensional unsteady unsaturated and saturated subsurface flows. The equations are written in the form of 2-D and 3-D heat diffusion equations respectively. In the model the surface and subsurface flow components are coupled interactively using the common boundary condition of infiltration through the ground surface. Overland flow is generally classified into Hortonian and saturation overland flows in watershed hydrology. The conjunctive model was, after verification with experimental data, applied to a hypothetical watershed to simulate examples of 2-D saturation overland flow and 3-D saturated and unsaturated subsurface flow. The model demonstrates that overland flow is initiated when rainwater supply exceeds infiltrability, not only in Hortonian but even in saturation overland flow. The model also simulates overland flow, infiltration, seepage, subsurface flow, and the interaction between subsurface and channel flows. The conjunctive model, as a general model to deal with the interaction between surface and subsurface flows, contributes as a tool for improved detailed simulation of 2-D surface and 3-D subsurface flows and their interaction.

It was studied four soil topossequences in the flooded area of the Amazon region. The results showed similar chemical and physical fertile soils, as well as reactive mineralogical thin components. Meanwhile, the use of the land as the "roças" cultivation system, along with physical limitations due to the fluctuation of the water table, points to a unsustainable development of the region.

There is a need to accurately and robustly simulate the hydrologic response of large catchments where insufficient data exist for detailed model calibration. Our research is aimed at development of a robust two-dimensional hydrologic modeling capability for forecasting stages (depths), discharges, and velocities along rivers and streams using a minimum of input data. The two-dimensional, physically-based, distributed-parameter Gridded Surface-Subsurface Hydrologic Analysis model accurately predicts runoff when sufficient data exist for model calibration and verification. However, when required calibration data do not exist or are unavailable, the assignment of meaningful parameter values is practically impossible for all model parameters. Given this fact, it is relevant to consider conceptualizations of the most uncertain portions of the hydrologic cycle while retaining physics-based approaches where most applicable.

The majority of the 16 federal reservoir projects operated by the Pittsburgh District of the U.S. Army Corps of Engineers were authorized by the Flood Control Act of 1936. These dams were built primarily for the purposes of reducing flood stages (flood control) and providing low water regulation for water quality control along the Allegheny, Monongahela, and upper Ohio Rivers. After these dams were placed in operation, Congress granted general authority to all Corps reservoirs for the following limited purposes: recreation; municipal and industrial water supply; fish and wildlife conservation; water quality control; and threatened and endangered species preservation. However, none of these laws specifically allocated water in the Lakes or changed operational criteria for any of these added purposes. The legislation generally states that the dam will be operated in such a manner as will protect and enhance these purposes and to the extent that they are compatible with other project purposes.

This paper discusses different issues that appear when characterizing turbulence in two-phase, air-water flows, in relatively deep containers. The flows correspond to bubble plumes with relatively large bubble diameters. Results of velocity time series obtained with the help of Acoustic Doppler Velocimeters (ADVs) are presented and analyzed. The post-processing of those signals is discussed in detail, presenting a methodology to separate air velocity from water velocity using ADV. Turbulence statistics, obtained from the time series, are also presented. Specifically, spatial distributions for the turbulent kinetic energy are analyzed. An interpretation of the behavior of the plume in the tank is also provided.

This research proposes a generalized parameterization method for parameter structure identification in groundwater modeling. Identification of distributed parameters invariably involves a large number of unknowns. Because of data limitation, the formulated inverse problem is inherently non-unique and unstable. Parameterization reduces the degrees of freedom of the unknown distributed parameter and defines the spatial attributes of a distributed parameter in terms of parameter dimension, parameter pattern and parameter values; thus, making parameter structure identification possible. Traditional parameterization methods include zonation, kriging, spline and other interpolation schemes. In this research, we develop a universal parameterization (UP) method that unifies various continuous and piecewise constant (zonation) structures through a set of weighting coefficients. By adjusting the weighting coefficients, UP will produce either a zonation structure, a continuous distribution structure, or a hybrid structure. The UP method shows greater flexibility not only in manipulating the highly complex spatial distribution but also in optimizing the parameter structure. When investigating a field with a high degree of discontinuity, UP outperforms other traditional parameterization methods. With UP, parameter structure identification seeks to identify the parameter dimension, parameter pattern, parameter values as well as the weighting coefficients. The developed methodology is demonstrated by a numerical example in which the distributed transmissivity in a two-dimensional confined aquifer is identified. We use the natural neighbor (NN) method for interpolation. The zonation structure is determined by Voronoi tessellation (VT).

The magnitude of the lateral variations in the concentration of suspended sediment over dunes in an alluvial sand-bed channel are poorly known. Characterizing the lateral distributions of suspended sediment is important for understanding its causes and for accurate measurement of the rate of sediment transport. A series of laboratory experiments was conducted in a laboratory flume to characterize the lateral variations of suspended sediment over dunes. Forty experimental runs were made using flow depths of 0.33 and 0.13 m in a 1.2 m wide flume channel. The Froude number was 0.5 and the median diameter of the bed material was 0.52 mm. Suspended sediment concentrations were simultaneously sampled at two points located at the same depth but spaced laterally at intervals of 0.40, 0.20, 0.10 and 0.05 m (one third to one twenty fourth of the channel width). Also, acoustic instruments were used to obtain high- resolution bed profiles and depth-integrated suspended sediment concentration for the lower region of the flow. Mean absolute differences between paired suspended sediment concentrations were shown to increase and then level off to values between 25 and 40 percent of the mean concentration as the lateral spacing between samples was increased to between 1 to 2 times the flow depth. An analysis of the sediment concentration distributions indicated that from 8–16 sample positions across the channel would be required to define the mean suspended sediment concentration with a confidence interval of ± 4 to 7.5% of the mean concentration with a probability of 0.95. This study provides new information for understanding lateral variations in the concentration of suspended sand over dunes and new information for devising effective strategies for sediment sampling.

Rainfall distributions from rain gages are typically estimated by assuming a spatial geometry tied to point rain gage observations using, for example, Thiessen polygons, inverse distance squared weighting, or statistical Kriging techniques. Unfortunately, the spatial distributions inferred by these approaches have little connection with how rain actually falls. Since the release of the WSR-88D (NEXRAD) radar in the early 1990s, many hydrologists and engineers have begun using gage-adjusted radar rainfall estimates for hydrologic and water resource modeling. Over large areas under multiple NEXRAD radar coverages, the quality of radar rainfall estimates can vary significantly from one location to another. Visible discontinuities can develop at the limits of coverage of a single NEXRAD site because of slightly different performance or calibration techniques used at the different radar sites. Using a variety of GIS procedures for this study, these discontinuities were eliminated and locations of ground clutter were suppressed, yielding a seamless map of unadjusted radar rainfall estimates. These data were adjusted with over 400 rain gages located throughout the state using a modified spatial adjustment technique originally developed by Brandes at the National Severe Storms Lab in the mid-1970s. This approach was able to retain the volumetric rainfall estimates from the gages while maintaining the spatial signature of the rainfall. Use of this technique greatly improves gage-adjusted radar rainfall estimates.

The South Florida Water Management District is responsible for managing water resources in a 46,439 square-kilometer (17,930 square-mile) region. The area extends from Orlando to Key West and from the Gulf Coast to the Atlantic Ocean and contains the country's second largest lake. The area is also the site of a major environmental restoration project. The District operates 2,898 kilometers (1,800 miles) of canals, 22 major pump stations and 2,200 water control structures. Rainfall-based management plans are becoming more prevalent in operation of these pumps and structures. The District uses a network of over 300 active raingage stations that cover the more populated and environmentally sensitive areas under its management and provide data for this purpose. Five NEXRAD (Next Generation Weather Radar) sites operated by the National Weather Service cover the region. NEXRAD data have been used for several years at the District for weather reporting that targets operational issues. However, their use has been largely limited to visual interpretation of data by the staff as opposed to quantitative analysis. In July 2002 the District, in conjunction with three of the other four water management districts in Florida, began to acquire NEXRAD data coverage and develop a corporate database and methods for data access. Corporate access of 15-minute, raingage-adjusted NEXRAD data was a major objective of the acquisition. Four major uses of the NEXRAD data at the District were identified: operations; modeling; planning and analysis; and reporting. Each use had unique data-storage and access needs. This paper details the database design; integration with the corporate database, DBHYDRO; access methods established, including those for GIS and modeling applications; data QA/QC; data verification; and applications of the NEXRAD database.

This study evaluates potential biases in radar rainfall data being considered for use in hydrologic models for Great Lakes forecasting. Precipitation estimates from Weather Surveillance Radar — 1988 Doppler (WSR-88D) are input to the Great Lakes Environmental Research Laboratory Large Basin Runoff Model (GLERL-LBRM), and results from this model with both radar (MAPX) precipitation estimates and Thiessen-weighted rain gage (MAP) data are compared to observed flows. Case study results are presented for several watersheds in Michigan, Ohio, and Wisconsin. Long-term goals are to recalibrate the existing lumped-parameter models based on MAPX data, develop more distributed hydrologic models, and utilize radar precipitation data to estimate precipitation over the lakes and in watersheds with sparse rain gage networks.

The overall objective of this study is to use the GPS Integrated Precipitable Water (IPW) data at Blacksburg VA (BLKW) to predict significant convective rain events capable of producing flash flooding. It is found that for frontal events there is a long lead-time; summer time convective events have short lead times. The wintertime GPS-IPW behavior is very similar between stations under suitable transformation of data. The methodology shows potential in gaining lead-time for significant rain events.

In engineering design and analysis, proper assessment of stochastic features of model outputs is essential to account for the existence of various uncertainties. Monte Carlo simulation provides a viable tool to reproduce the stochastic features of the model outputs by generating statistically plausible realizations for random model inputs and parameters. Although there are a variety of generators developed for problems involving univariate random variable or independent random variables, generating multivariate random variates is more restricted to a few distributions whose joint probability density functions (PDFs) are known. Since most engineering problems involve multiple correlated random variables whose joint PDF is normally very difficult to establish, practical algorithms are needed to generate correlated random variables with a mixture of marginal distributions and correlation structure or even less information. In this paper, the third-order polynomial normal transform (TPNT) technique is applied to the Monte Carlo simulation, which enables one to generate correlated random variables with much less information than joint PDF. Using the TPNT, arbitrary univariate or multivariate random variates can be generated from univariate or multivariate normal distributions utilizing orthogonal transform and normal transform. The information needed is the marginal distributions and correlation structure, or sample data of the involved random variables. As compared with the existing model, the proposed algorithm shows comparable results with less information required.

Drinking water supply utilities are responsible for providing a safe and reliable supply of potable water to their customers. Their ability to do so can be hampered by natural disasters, anthropogenic disasters, and performance failure. These events constitute the major risk vectors in potable water supply. The frequency and severity of disaster-type events and their impacts on water supply have been well studied. As such, the focus herein is on the analysis of the mechanical and operational failure events that result in performance risk. While tools for the analysis of the likelihood and severity of performance risk vectors currently exist, a general framework for the performance risk analysis across entire water supply systems remains to be developed. This paper presents a framework for the analysis of performance risk in water supply that can be applied to the entire system or to individual sub systems.

Rosenblueth's probabilistic point estimate method is a computationally straightforward technique for uncertainty analysis of engineering problems. It is capable of estimating the statistical moment of any order of a model output involving several stochastic variables that are correlated or uncorrelated and symmetric or asymmetric. However, in multivariate problems with more than two stochastic variables involved, the Rosenblueth's method is not able to provide a unique solution, rather an approximate solution to indeterminate problems. This is attributed to the fact that the number of unknowns to be solved is larger than the number of governing equations that are provided. An improved point estimate method is proposed to circumvent this drawback and increase the computational efficiency as well. It is demonstrated that the proposed point estimate method preserves the capabilities of the Rosenbleth's PE method but resolves the associated non-uniqueness problem.

This study addresses the role of the uncertainty and variability associated with groundwater aquifer used for water supply and health risk parameters, their propagation, and their influence on the probability of developing cancer due to subsurface migration of chlorinated ethenes in the water supply system of a community. The risk analysis is performed for a hypothetical aquifer that has a tetrachloroethene (PCE) plume with a downstream drinking water well and a 100 year time horizon. PCE can be biodegraded into trichloroethene (TCE), which biodegrades into dichloroethene (DCEs), which biodegrades into vinyl chloride (VC). PCE, TCE, and DCE are suspected carcinogens and VC is a known carcinogen. VC can further be biodegraded into ethane which is non toxic. Bioremediation is enhanced by addition of a hydrogen donor through injection wells and occurs for a few years at the beginning of the time horizon and brings the contaminant concentrations down to accepted maximum contaminant levels. A human exposure model is developed and a methodology of linked groundwater contaminant transport and human exposure models is applied in order to propagate population variability of cancer susceptibility from exposure to the contaminants and aquifer uncertainty (in hydraulic conductivity and in microorganism distribution) separately using a two-stage Monte-Carlo simulation technique. The analysis computes upper bounds on risk for percentiles of both the aquifer uncertainty and population variability. A stochastic sensitivity analysis is used to assess the relative importance in predicting human health risk of the variable and uncertain parameters for three age categories of the population; adults, adolescents, and children.

Improvement in computer hardware and software over recent years have enabled water networks to be modeled in detail in order to simulate their hydraulic operation. It is common, for large-size networks to be modeled comprising of several thousand nodes. Therefore, it is interesting to study the level of modeling to be used during hydraulic model construction. We propose a method allowing for consideration at the same time the problems studied by the model and the hydraulic differences between the sub-zones of the network. The method consists firstly of splitting the network into sub-networks according to the districts or of the towns supplied by the network. Secondly, for each sub-network, we define three or four levels of possible modeling. The third part of the method consists of making a hydraulic simulation on the basis of level 1 then by introducing the other levels gradually. With each new level, we compare the flows obtained in entries and exit of the district considered. We can then decide, by using a fixed criterion, to stop the iterations. This method was used on the water network of the urban community of Strasbourg which supplies approximately 400 000 inhabitants. It was possible to represent the network with a model comprising of only 950 nodes.

The National Oceanic and Atmospheric Administration installed a Physical Oceanographic Real Time System (PORTS) to provide water surface elevation, currents at National Ocean Service prediction depth (∼4.6m below MLLW) as well as near-surface and near-bottom temperature and salinity, and meteorological information at six-minute intervals. To complement the PORTS, a nowcast/forecast system has been developed for Galveston Bay using a modified version of the Blumberg-Mellor three-dimensional circulation model and a one-way coupled fine resolution Houston Ship Channel model. Daily 24-hour nowcasts and 36-hour forecasts are performed using both bay and channel models in a pseudo- operational setting. This paper investigates the incorporation of a real time wave model component within the present system. Initial test results for two parametric USACE wave models as well as for the finite difference-based Donelan wave model are presented. These tests were performed using constant wind speed and direction to study the behavior of wave growth in each of the models. Next a period in January 1997 was used to evaluate/modify and calibrate the models. A swell open boundary condition was added to the parametric models. For the deep-water Donelan wave model, swell effects were not included, but rather a depth dependent wind to wave momentum transfer coefficient was used to incorporate shallow water effects. Based on these studies a combined parametric model for significant wave height prediction and CW-167 for significant wave period prescription) is advanced for incorporation within the present nowcast/forecast system. Finally, data requirements and plans for incorporation of the combined parametric wave model as a subroutine within the circulation model program are discussed.

The hydrodynamic transport characteristics of the Kill van Kull and the Arthur Kill, which connects the New York Bay to the Raritan Bay, have been investigated by using a three-dimensional, time dependent hydrodynamic model, ECOM. The objective of this study is to determine volume and salt transport through the Kill van Kull and the Arthur Kill and to obtain a basic understanding of the physical factors driving the salt transport through these important water bodies. The current model is an enhanced version of the original System Wide Eutrophication Model, which has been re-calibrated and re-validated in the New Jersey tributaries including the Hackensack, the Passaic and the Raritan Rivers. Volume and salt fluxes were determined by the decomposed correlation terms using the model computed salinity, temperature and currents. Results indicate that the net long-tem volume and salt flux is directed west through the Kill van Kull and south through the Arthur Kill. The peak water flux through the Arthur Kill is in excess of 400 m³s^–1 Stokes transport term contributed most towards upstream salt transport in Newark Bay and the Arthur Kill. In the Kill van Kull, the upstream salt transport is minimal. Salt flux through the Arthur Kill appears to be dominated by the elevation gradient between entrance to the Kill van Kull (from New York Harbor) and Perth Amboy. The salt flux through the Kill van Kull is influenced to a considerable extent (but not dominated) by the elevation gradient between the entrance to the Kill van Kull (from New York Harbor) and Shooters Island. The density gradient does not appear to be a predominant driving factor for the salt transport through the Kill van Kull and the Arthur Kill.

Recent advances in incorporating artificial intelligence concepts into real time control have reduced the barriers to successful implementation. A neural-optimal control algorithm is presented that fully incorporates the complexities of dynamic, unsteady hydraulic modeling of combined sewer system flows and optimal coordinated, system-wide regulation of in-system storage. The neural control module is based on the recurrent Jordan neural network architecture that is trained using optimal policies produced by a dynamic optimal control module. The neural control algorithm is demonstrated in a simulated real-time control experiment for the West Point combined sewer system, Metro Seattle. The algorithm exhibits an effective adaptive learning capability that results in improved performance of the control system.

A random-walk particle-tracking model was developed for the East River, the New York Harbor, the Long Island Sound and the New York Bight to predict the distribution of ichthyoplankton and their probability of entrainment at the cooling water intake of the Charles Poletti power plant. The model was configured into a three-dimensional hydrodynamic and particle tracking model, ITM, which uses neutrally buoyant passive particles as surrogates for ichthyoplankton. In ITM, particles are transported by multidimensional advection and dispersion processes. The model was driven by time-dependent water levels, temperature and salinity along open boundaries, meteorological forcing and freshwater inflows. It was calibrated using observed surface and bottom salinity and temperature, water surface elevations and surface, mid-depth and bottom currents at locations across the New York harbor and the East River during 1994 and 1995. Results indicate that estuarine circulation greatly determined the distribution of particles in the East River. The probability that ichthyoplankton would be entrained into the Charles Poletti plant intake structure was highly dependent on the location and time of their release and the assumption that ichthyoplankton behave like neutrally buoyant particles.

Linking a hydraulic model to telemetry systems allows for an integrated simulation of hydraulic, water quality and economic parameters and is of growing interest to water system operators and owners. Coupling a mathematical hydraulic and water quality model to a Supervisory Control and Data Acquisition (SCADA) system allows the model to be calibrated and validated in real-time with data obtained by the SCADA system. Using this integrated approach to modeling, pressure and flow conditions in the water distribution network can be monitored in an automated fashion, and economic parameters can be evaluated in parallel to the hydraulic and water quality parameters, enabling their optimization. This approach also allows the operator to see how the entire system is operating, instead of relying on the feedback from a few SCADA sensors placed in the network, thus allowing for any breakdown in the network to be readily identified, evaluated, and rectified. This modeling approach also assists the operator in contingency planning to meet the EPA requirements for water security emergency action plans and determining the most appropriate response to unusual operating conditions. The state of art of the available couplings between a hydraulic model and a SCADA system, as well as the experience from an existing implementation is discussed in this paper.

The National Ocean Service (NOS) installed a Physical Oceanographic Real Time System (PORTS) in June 1996 in Galveston Bay, Texas. The PORTS water surface elevation, currents at National Ocean Service prediction depth (∼4.6m below MLLW) as well as near-surface and near-bottom temperature and salinity, and meteorological information are scheduled to be made available at six-minute intervals. Due to biofouling, cable snagging, and an assortment of maintenance issues, these data are either not available or degraded part of the time. A Continuous Operational Real-time Monitoring System (CORMS) has been implemented to quality control the PORTS data. Under the present operational scheme, if data are suspect, data dissemination is suppressed. This paper investigates the possibility of statistically nowcasting these missing data to at least provide an estimate to the mariner for missing data values. A computer program is developed to acquire the PORTS screen data every six-minutes and to statistically nowcast missing water level, current, near-surface salinity and temperature, as well as wind and atmospheric pressure. A proof of concept is demonstrated for two PORTS screen datasets, each with missing data. The statistical nowcast procedures involving linear regression of subtidal water level and along channel axis currents are evaluated for a typical dry season month (July 2001) and a typical wet season month (January 2002). The evaluation results are discussed and demonstrate the utility of the statistical nowcast for both water levels and currents when contrasted to harmonic analysis. Water level and current statistical nowcasts are compared with nowcasts produced by the NOS Experimental Nowcast/Forecast System, which is based on a three-dimensional hydrodynamic model for Galveston Bay and a one-way coupled fine resolution Houston Ship Channel model. Next, the role of statistical nowcasting within an operational nowcast/forecast system for Galveston Bay is considered. Finally, conclusions are drawn and additional statistical nowcast enhancements are discussed.

The risk of fungal and bacterial diseases in a crop canopy can be predicted using disease risk models with specific environmental conditions (microclimate data such as temperature, relative humidity, solar radiation, wind speed, and surface wetness duration (SWD)). Unfortunately, the inconvenience and uncertainty associated with monitoring key variables such as SWD at the local crop scale prevent existing disease risk models from being used with reliability. A suite of recurrent Artificial Neural Networks (ANNs) has been developed to estimate key environmental variables (specifically SWD) at local crop scales from local and regional weather station data and site specific sensing data. The selected ANN combines two statistical methods to accomplish this spatial mapping (a K-nearest means classifier and a Bayesian classifier), while the recurrent nature of the ANNs provide a means of forecasting in time. The overall goal is the development of web-based Geographical Information System (GIS) software incorporating ANN models that ultimately can be combined with existing disease models to improve maps of disease risk at fine spatial and temporal scales.

Chlorination has been the major strategy for disinfection of drinking water in Taiwan. The objective of this study is to classify and evaluate the factors of temporal-spatical explicitly for the influence of total trihalomethanes (TTHMs) concentration. The temperature for finished water increasing from northern to southern ranged from 20.4–27.1 °C. pH ranged from 6.7–8.3. TCM is the main content of THMs, the range was from 1.3–18.5 μ g/L, and the DCBM range was from 0.2–6.2 μ g/L. That conformed to the tendency of temperature of spatical factor in Taiwan. DBCM and TCM were predominant in Peng-Hu (11.2 and 24.8 μ g/L, respectively) because the water source was groundwater which near the coast and had higher bromide concentration.

Increasing demand and continuing land development in the eastern U.S. are raising concern for the long-term ability of small watersheds to sustain baseflow during ever more frequent periods of drought. To develop accurate predictive models of baseflow recession and prioritize watershed management/protection efforts, it will be critical to develop a stronger understanding of the processes that link water storage and stream baseflow at the watershed scale. Prior research has shown that streamflow recession often follows several distinct stages (seen as slope breaks in the log-transformed recession curve), with the slope generally declining as recession proceeds. One hypothesis is that these multiple slopes correspond to different storage reservoirs and drainage processes, such as unconfined aquifer drainage and deep aquifer discharge. This hypothesis is tested by determining the correlations between successive hydrograph recession slopes (recession rate constants) and watershed hydrogeomorphic properties for 25 small (<150 km²) watersheds across the major physiographic provinces of Pennsylvania. It is shown that the second rate constant is correlated to geology, soil type, drainage density, and relief; however, additional work is necessary to address variability in the first rate constant. Based on the magnitude of the second rate constant, it is determined that in Pennsylvania, the Appalachian Plateau region streams are most susceptible to drought. Those in the Great Valley and Piedmont region streams are least susceptible; however, many watersheds in these areas are already highly developed.

Streamflow-rating curves showing the relation between water-surface stage and the flow discharge have been a widely used tool in hydrology for over a century. Despite their wide-spread application, several problems with stage-discharge ratings have been recognized, resulting in development of many empirical methods to adjust or shift the rating to improve the agreement between ratings and measurements. Among these are methods to account for looped ratings during unsteady flow, stage-slope-discharge ratings to account for variable downstream backwater conditions, and shifts to account for changes to the "control" for the rating. Although these methods have been commonly applied to compute discharge records for streamflow-gauging stations for over half a century, they have achieved only a limited degree of success because of simplifying assumptions on the flow hydrodynamics. Most of these methods are not widely known or understood, other than by those actively applying them to streamflow gauging stations. Furthermore, the underlying physics and scientific justification for these methods has not been examined. Examination of the simplifying assumptions that these methods apply to the momentum and continuity equations clearly illustrates the reasons for the limited success of these methods. In this presentation, methods to shift or adjust stage-discharge ratings are described from the viewpoint of the fundamental hydrodynamics of unsteady nonuniform flow. The simplifying assumptions inherent in these methods are presented and conditions for which the error introduced by these assumptions becomes significant are discussed.

A fully integrated regional simulation model (RSM) has been developed to simulate the complex hydrologic system of South Florida. The South Florida system consists of a large integrated overland flow system, a groundwater flow system and a canal flow system. Implicit solution methods, efficient sparse solvers, and object oriented methods have made it possible to solve large complex systems simultaneously with practically any time step and cell size, even when the numerical errors can be extremely large. The current study is aimed at understanding the numerical error due to a boundary disturbance under a variety of spatial and temporal discretizations and solution characteristics. The errors are compared with the analytical estimates obtained by Lal for finite difference problems. Results of the study are useful in determining optimal spatial and temporal discretizations for model applications.

A regional estimate of potential evapotranspiration (ETp) for central and south Florida is provided based on cumulative literature and lysimeter studies of evaporation and evapotranspiration measurements and estimations. General estimates of annual ETp for the south Florida Water Management District range from 122 cm in the north to 137 cm in the south. Open water and wetland systems evaporate at the potential rate. Otherwise, actual evapotranspiration will be lower due to limited availability of water or moisture. A network of weather stations and an ETp computation software developed in-house is currently being used at the South Florida Water Management District to develop a daily ETp database. A simple ETp estimation model that was previously calibrated with lysimeter data is being applied.

The tools available for teaching hydraulic engineering classes have changed considerably in the 46 years that this professor has been at it. In the 1950s, who would have thought that the slide rule would become obsolete! Or that civil engineering students would no longer have to master the Moody diagram. Civil Engineering students at Montana State University all take a first course in Fluid Mechanics, then a junior level "water resources" course (typically using a text such as Linsley & Franzini "Water Resources Engineering"). From there they take one or more senior elective "applied hydraulics" classes, which include a "closed conduit flow" (affectionately referred to by students as "pipes and pumps") and an "open channel hydraulics" class. My teaching assignments at MSU over the years have usually included one or more of these "fluids" or "water" courses, so I have participated in and been awed by the evolution of new teaching methods in my chosen areas of expertise. In graduate school in the mid-1950s my textbooks included "Elementary Mechanics of Fluids" by Hunter Rouse and "Steady Flow in Open Channels" by Woodward and Posey. The word "computer" wasn't in our vocabulary. I'll share a few observations in each of the areas "closed conduits" and "open channels," and how teaching of undergraduate students has changed through the years.

In the past 30 years, lumped-parameter watershed models have been employed for integrated surface and groundwater modeling to calculate surface runoff and total maximum daily loads (TMDL) on various temporal and spatial scales of hydrologic regimes. Physics-based, process-level models that have the design capability to cover various scales have been practically nonexistent until recently. It has long been recognized that only such models have the potential to further the understanding of the fundamental biological, chemical and physical factors that take place in nature hydrologic regimes, and give mechanistic predictions. However, there are severe limitations with these models that inhibit their use. These are, among other things, the ad hoc approaches of coupling between various media and the excessive demand of computational time. This paper presents the development of a first principle, physics-based watershed model to address these issues. A rigorous coupling strategy is described for interactions among overland regime, rivers/streams/canals networks, and subsurface media. Introductions of non-physics parameters such as linkage terms are avoided. The cultivation of innovative numerical algorithms to increase the computational speed are discussed.

A new integrated surface-groundwater model, the integrated hydrologic model or IHM, is currently being developed which significantly advances the technology from its predecessor models FHM and ISGW. IHM combines the EPA surface water model HSPF and the USGS groundwater model MODFLOW with near simultaneous integration and physical interpretations of HSPF vadose zone process dynamics. The two model components are linked through an interface that provides a surface-water-to-groundwater transition that maintains strict water balance and smoothness in physical processes. Water table transitions from seasonal deep to at or above land are simulated. The interface can accommodate a specific yield that varies as the capillary fringe approaches land surface, which is an important phenomenon in shallow water-table environments.

Restoration of the South Florida ecosystem is a major undertaking for the U.S. Army Corps of Engineers and the South Florida Water Management District in the federally approved Comprehensive Environmental Restoration Plan (CERP). Many competing entities have an interest in the restoration process which will probably include physical changes to the land surface and adjustments to water deliveries. Restoration plans will be developed using numerical hydrologic models that optimize the benefits among the various stakeholders. Since the health of the South Florida ecosystem is impacted by subtle exchanges of surface and subsurface water, there is a need for great accuracy in the models that will identify the best restoration plans. The accuracy needed for design level simulations of the various projects approaches +/– 0.1 ft in the water elevations. Models with insufficient physical processes and/or highly schematized models with large grid or mesh induced error will not be sufficient to make design level simulation of South Florida hydrologic processes. Additionally, since a large number of models will be set up and analyzed, effective and efficient graphical user interfaces are needed for the models. This paper presents the results of testing WASH123D code to simulate surface water and subsurface interactions in the groundwater system.

Modern engineering design takes advantage of computerized solutions to a great extent. Preparing students for employment should involve exposing them to the use of software in design. Using commercial software in engineering classes has the advantage of giving the student a better appreciation of the kinds of work that must be done in engineering design. The more realistic the student experience, the better prepared the student will be for real employment. A good design problem contains the following features: complexity/realism, cost considerations, permitting issues, deadlines, imperfect data and multiple "correct" decisions. Problems are useful to the extent they contain these features. Additional discussion examines the use of a lab rather than homework environment to solve these problems.

Several municipalities located west of the City of Miami experience frequent flooding due to high water tables, inadequacies in the primary drainage system, and relatively low land surface elevations. These jurisdictions are located within the C-4 drainage basin of the South Florida Water Management District. Most of the urban development within this basin is situated in closed subcatchments that drain primarily through infiltration of rainfall to the water table aquifer. The surficial aquifer is drained by an extensive canal network. Hydraulic interactions between canals and the surficial aquifer are very significant due the highly transmissive nature of the aquifer. Stages in the canal network are controlled by structure S-25B, a gated spillway located near the eastern (downstream) end of the C-4 canal. During wet hydrologic conditions, S-25B must maintain canal stages low enough to allow drainage of the urban areas. S-25B, however, is several miles downstream of certain affected municipalities and has a limited ability to affect drainage in these areas. The purpose of this study was to evaluate the flood control benefits of lowering the target wet season headwater elevation at S-25B.

When is hydraulic modeling not really modeling? It's when a refugee from the water industry takes on the challenge of teaching a water system design course to college seniors in Urban Engineering studies. The extensive use of modeling software (WaterCAD, EPANet, etc.) in the hands of imaginative students can make a design laboratory an understandable learning experience without getting wet. When the drudgery of calculation is removed, the creativity of experimentation surfaces. The results can be stunning!

A challenge in teaching undergraduate engineering courses is incorporating current software packages that are commonly used in engineering practice. This is especially difficult for small undergraduate colleges and universities due to budget constraints and a lack of a graduate program to support potentially expensive software packages. For a senior-level engineering hydraulics course, commercial software is used by the students to complete course projects. The students are provided a city's water and sewer system layout. These programs are used to design a network to deliver water from the existing water distribution system to a proposed new industrial building; determine the effect of the added sewage from the proposed industrial building to the sewer collection system; and redesign a pumping station to deliver water to a water tower. This presentation will show how these projects are delivered to the students and how they are assessed.

Solutions using spreadsheets and simulation software are presented and discussed as illustrations in undergraduate water resources engineering curricula. Students use the spreadsheet solutions to reinforce fundamental engineering principles through individual development of applications. Simulation software is utilized to facilitate the study of realistic engineering systems while exposing students to a more professional working environment. Assessment of the technology that is embedded within the simulation software and the introduction of computer modeling skills enhance the students' understanding of complex water resource systems while developing aptitude in an emerging technology field.

While equipped with knowledge of the most recent technological advances and problem solving tools in their field of study, today's engineering students often possess an inability to effectively solve new problems. This inability to solve new problems likely arises from either students' difficulty in grasping basic concepts, or applying these concepts to new situations. Resolving this predicament must therefore involve activities that focus on improving knowledge of fundamental concepts, while emphasizing application of these concepts to problem solving exercises involving both relevant and interesting issues. One method that addresses this issue is the use of interactive software. Interactive software can assist students in visualizing the effects of different parameter inputs on water resources processes, without the cumbersome tasks of writing extensive computer programs. A second method involves the use of case studies. Case studies provide students with open-ended problems that reflect those encountered in the real world, and the use of case studies in class lectures provides students an opportunity to study and learn the instructor's problem solving approach. This paper addresses the employment of interactive software and case studies in an introductory course in water resources engineering. Through use of a progressive approach to water resources engineering problem solving, students are first exposed to fundamental concepts of pressurized pipe and open channel flow hydraulics through review of basic fluid mechanics principles, followed by extension of these principles to other water resources topics including pressurized pipe networks, pumping systems, and open channel flow systems including culverts, storm sewers, and sanitary sewers. Following fundamental subject matter development, each area is addressed in greater detail through illustration of in-class case studies and use of a suite of interactive software via in-class tutorials preceding students' use of the software during execution of out-of-class design projects oriented towards real-world case studies.

The lower Colorado River below Parker Dam is highly influenced by the needs of mankind, with several dams and diversion structures controlling flow for a variety of purposes. Along this reach of river, many natural and manmade backwater areas provide habitat for a diverse community of aquatic and aviary species. The environmental, ecological, engineering, and biological communities have expressed concern over the effects of river operations on the habitat in these backwater areas. Engineers from the Bureau of Reclamation, in conjunction with Colorado State University, have utilized a geographic information system (GIS) in association with a numerical hydraulic model to assess these effects. The system estimates the water surface elevations within the backwaters during the passing of a hydrograph created by reservoir releases, then quantifies and visualizes the change in habitat for many species of life living in the areas. Geostatistical analysis techniques are applied in the GIS to refine the geographic data for analysis purposes, and an object-oriented computer program is developed for automating the analysis and visualization processes.

The application of information collected by remote sensing satellites to urban related problems has developed at a fast pace during the last few years as a result of the availability of commercial satellites proving high spatial resolution data. These systems obtain image data with 4- m and 1m spatial resolutions that are one order of magnitude better than what was available until recently. These data can be processed automatically to yield information related to locations of interest (site-specific information), information about the surrounding areas (neighborhood information), and can be combined with other types of information such as census data (contextual information) to provide insight into current population distribution characteristics and the potential environmental impact of alternative urban development scenarios. This study was carried out within the City of Columbia, Missouri to assess the ability of using satellite remote sensing data and a Geographic Information Systems (GIS) to characterize and analyze community development and population distribution to inform decisions regarding the implementation of the Environmental Protection Agency's (EPA) Phase II storm water requirements. The study utilized the multispectral data collected by the Ikonos satellite with 4 m spatial resolution to produce a land cover map for the city. The land cover information was the starting point to identify land use in the urban setting and to evaluate the potential impact of storm water runoff that might carry different types and amounts of pollutants from the land surface to streams.

Presently, one of the most challenging problems faced by Metropolitan Regions in developing countries is environmental degradation due to inadequate land use. In order to alleviate environmental damages and detain illegal or irregular occupations, river basins have been often used as a unity in urban/environmental planning and, especially, as a land use managerial tool. With the intention to assess river basin water quality, this research has adopted the Açungui River, Curitiba's Metropolitan Region (CMR), Brazil, as a case study because it has been considered that its river basin will be used as a source of supplying water to CMR by the year 2020. For the water quality study physico-chemical data on pH, dissolved oxygen, biochemical oxygen demand, fecal coliforms, total solids, turbidity, temperature, conductivity, total nitrogen, total phosphate and flow rate were used. Through the development of a database the values of the physico-chemical parameters were input to two GIS software (SPRING and ARCVIEW). The cartographic spacialization of the information on water quality was associated to other data on land use, topography, presence of a natural forest and roads among others. This was important to produce a general view of all factors which contribute to the comprehension of the environmental system being studied. Results were compared to values established by the Brazilian Surface Water Quality Standards (CONAMA Resolution n° 20/86), and final results of this research are expected to contribute to a better understanding of this environmental system and also to help decision makers to improve the environmental management of the region.

In this paper, the use of the Arc Hydro data model for integrating hydrologic computer applications is presented. Arc Hydro is a template for the storage of spatial and temporal hydrologic information, in which the different objects of the system are assembled together by virtue of their connection to the river network or to each other. Additionally, its use helps expedite multi-user/agency/company collaboration on a project. Customarily, simulation of the behavior of the surface waters of a hydrologic system is conducted with the aid of mathematical models and/or computer applications, each of which addresses a different natural or man-driven process (i.e., soil water balance, flow routing, non-point source pollution, reservoir operation, etc.). Whenever these processes take place — whether in sequence or simultaneously — interchange of data among applications is required. This data interchange has been called here integration of hydrologic application.

This report documents the selected treatment method and laboratory experimentation of arsenic removal from drinking water in small water delivery systems and domestic water systems. The objective is to expand upon research of new and existing arsenic removal technologies or promote a new, alternative process. Several treatment technologies have been considered to perform this function, but cost and reliability concerns prompted the decision to analyze small-scale, home filtration units, specifically. Various filter media including cuttlebone, diatomaceous earth (DE), iron oxide, and activated carbon were evaluated for efficiency in arsenic removal for the proposed home filters. Based upon initial test data, the use of iron oxide in the adsorption of arsenic has proven to be much greater in capacity and more reliable than the other media examined for use in the filtration units. Additionally, further experimentation with iron oxide exploring the affect of particle size and mass on the potential adsorption of arsenic has been performed to further determine the best possible filter design.

Thousands of children in developing countries needlessly die as a result of infections and diseases contracted by touching and drinking contaminated water each day. I say needlessly because there are simple solutions to prevent a large majority of these diseases in basic civil engineering and education. Children stand in puddles where water catches as they fill up their water jugs to bring back home as part of their daily chores, completely unaware that microscopic worms are boring into their bodies through their feet to infect them for the next twelve months. People in developing countries also drink water from nearby sources without disinfecting it because boiling it requires too much fuel and they are unaware of alternative options that are affordable. This paper proposes the importance both routing the water away from water sources so people do not stand in it as well as explains how water can be disinfected with virtually no resources or cost using solar disinfectant. This paper does not claim to discuss the discovery of any new technology or engineering, but rather the importance of the issue and the ease at which the problem can be solved and hundreds of thousands of lives can be saved each year.

Effective collection and treatment of urban and industrial wastewater is a critical problem in countries such as Sri Lanka and India. Current solutions to these problems have been dominated by "western" approaches such as centralized sewerage based on traditional design criteria and treatment using aerobic processes such as the activated sludge process. However, the solutions are expensive to construct and operate and pose technical and financial challenges for the operating authorities. This paper examines alternative concepts — such as distributed systems, re-use of treated wastewater and community participation - and technologies — such as simplified sewerage, settled sewerage, septic tanks for primary treatment, anaerobic filters for secondary treatment and constructed wetlands for secondary and tertiary treatment - for the collection and treatment of urban wastewater and describes their performance in various situations in Sri Lanka. It is concluded that these alternative approaches have been successful in a wide variety of applications. In particular, treatment to levels that permit re-use in gardening and vehicle washing has been achieved in several hotels. However, some issues related to long term-operation and maintenance remain to be resolved in the case of application to low-income settlements.

A three-dimensional hydrodynamic and pathogen fate model has been developed for Quincy Bay, MA. The model has been used to determine the water contamination due to sewage outfalls in the bay and to evaluate pollution control strategies to meet water quality criteria for recreational purposes. The Quincy Bay model has two parts: a hydrodynamic model and a pathogen fate model. The hydrodynamic model provides a quantitative description of water circulation and mixing of Quincy Bay. The pathogen fate model uses the transport fields generated by the hydrodynamic model to predict the distributions of specific pathogens associated with fecal contamination. The modeling framework, used to determine the abundance of enteric organisms in Quincy Bay, includes pathogen loads from the outfalls and shoreline sources, dispersion and loss processes that include phototoxicity, temperature, salinity, predation or die-off and settling. The ability of the model to simulate advective and dispersive processes in Quincy Bay was assessed by comparing model results against observed temperature, salinity and fecal coliform. Overall the model performs very well. The model response to the atmospheric cooling event observed in mid-September, 2001 and to the rainfall events was very good. After the model was calibrated and validated, six alternatives of outfall locations were evaluated under four different storm scenarios.

This paper summarizes the results of the pilot constructed wetland in Warangal city in the State of Andhra Pradesh in India. This wetland was monitored for 12 months to understand the removal efficiencies of various pollutants in the municipal wastewater during different seasons in a year. This paper examines the potential of constructed wetlands for secondary municipal wastewater treatment especially in removing pollutants like BOD, COD, TSS, and nutrients like nitrates and phosphates etc. Since sufficient data pertaining to various seasons of the year are available, mathematical models to predict the behaviour of the system also are developed. A k-C* model, suggested by Kadlec also has been fitted for various parameters taking account the monthly data. The study shows that the system performs well and could be adopted as a viable alternative, which would be low in cost and at the same time environmentally and ecologically friendly, in upcoming cities in developing countries where sufficient land is cheaply available. It is to be noted that the successful performance of the pilot plant has motivated the city municipal authorities in Warangal to go in for a decentralized treatment wetland system for treating municipal wastewater in different parts of the city. Wetlands are widely regarded as the best low cost alternative eco-technology for treatment of municipal, industrial and agricultural effluents. Although developed countries have been using this technology, the use of these systems in treating wastewater has not been very popular in most tropical/semi arid countries like India. The study reported in the present paper is the result of an attempt to use this technology to demonstrate to the authorities in charge of municipalities in less developed countries that the use of treatment wetland could be one of the most viable alternatives for municipal wastewater and stormwater management.

A decentralisation structure is proposed in the previous paper for robust model predictive control (MPC) of chlorine residuals in drinking water distribution systems (DWDS). The water network is divided into a number of interacting zones. Each zone is directly controlled by a local controller. It is assumed that the local controllers can exchange information about their control actions. The communication is asynchronous. This information is used to predict the zone interactions. Interaction prediction errors are unknown but can be bounded. The interaction prediction errors are rejected by robust design of the model predictive local controller. The design extends an approach developed previously. Simulation results are presented and discussed.

The effect of variable demands on pressure and water quality in a water distribution system is presented. An analytical solution for the branched system and a probabilistic model for a looped system are described. The study was done for both steady and unsteady flow situations. The analytical solution for branched systems is derived using the concepts of hydraulics and probability. The analytical solution can give the mean and variance of pressure and water quality with a known changing demand. The Monte Carlo analysis was developed using Monte Carlo analysis and the simulation program EPANET. It was demonstrated on the municipal water distribution system of Milford, Ohio. The variations of demands are assumed to be uniformly distribution. In the steady flow case, when the demands are uniform, it was found that pressure is uniformly distributed; with Gaussian demands, pressure is a "spiked" Gaussian. In unsteady flow, with demands changing uniformly, pressure is pseudo-normally distributed and travel time of contaminant is an unidentified distribution. With Gaussian demands, pressure is Gaussian. In the Monte Carlo generation of demands, however, some negative demands will be generated. When these negative Gaussian demands are set to zero, the pressure is a "spiked" Gaussian distribution. The means of pressure and travel time can be estimated by calculating pressure and travel time with mean demands and the variance by using probabilistic theory with multiple simulations to get the needed derivatives. These results can be useful to distribution system managers when trying to make decisions regarding the amount of time and money to spend on quantifying demand values to meet pressure and water quality objectives.

Dissolved oxygen content in water is one of the most important water quality parameters. In most wastewater treatment plans a lot of oxygen is utilized. Thus it becomes necessary to add oxygen to water in order to comply with the minimum dissolved oxygen concentration in the State's Water Pollution Control Regulations before the treated water is discharged into rivers or streams. One of the most cost effective post-aeration designs is a stepped-cascade aerator. There are many existing equations for modeling the oxygen transfer at stepped cascade aerators. These equations are mostly empirical and limited to certain situations. Existing models are evaluated against field data taken at four wastewater treatment plants. All models are found to need improvement. The need for a new equation specifically for stepped cascade aerators is evident since existing models were not developed for stepped cascade aerators, but for weirs or other hydraulic structures.

The kinetics of Escherichia coli (ATCC 25922) inactivation by irradiation with ultraviolet (UV) light was evaluated in batch reactors using pH of 6.0, 7.0, and 8.0, at a constant temperature of 25 C, and an exposure time of 30 minutes. The microbial inactivation rate was first order with respect to ultraviolet inactivation. Regression analysis of the survival data showed a clear influence of pH on the disinfection kinetics with pHs of 6.0 and 8.0 being most conducive for inactivation of the bacterium.

The Metropolitan Syracuse Wastewater Treatment Plant (METRO) discharges ∼80 MGD of treated effluent to the waters of Onondaga Lake, the outflow of which is received by the Seneca River. Trophic conditions in the lake are severely degraded by this input. A dual discharge strategy has been proposed under which the effluent would be diverted to the river. This, together with companion management actions, would serve to restore water quality conditions in the lake. Periodically (i.e. under certain low flow conditions), it would be necessary to return some of the diverted effluent to the lake in order to avoid violation of the river dissolved oxygen standard. Return of the effluent to the lake has the potential to increase phosphorus concentrations above the guidance value established to protect water quality there. A modeling analysis of the dual discharge strategy will be conducted, over a range of river flow conditions, to establish the frequency and magnitude of violations of the lake phosphorus guidance value. This requires specification of a suite of companion management options which would contribute assimilative capacity. These include actions focusing on the lake (e.g. P removal at METRO, reducing sediment P release through hypolimnetic oxygenation, reductions in nonpoint P contributions, and discharge of the METRO effluent to the lake's hypolimnion) as well as on the river (e.g. elimination of density stratification in the river, oxygenation of the METRO effluent, and attention to levels of CBOD/NBOD removal at METRO). Here we describe those companion options, examine their potential for contributing assimilative capacity, and establish a base management scenario to support further analysis.

In drinking water distribution systems, maintaining chlorine residual provides protection against contamination from leaks, regrowth of microbial contamination, cross connections, and other breakdowns. There are a number of factors affecting chlorine decay in the water distribution system both within the bulk fluid and with the material on the pipe wall. While the reactions within the bulk fluid are predominantly affected by water temperature and quality, the reactions at pipe wall are related to the corrosiveness of the ferrous pipe materials and biofilm. The reaction rate of chlorine is highly variable, even for waters of similar quality. Some of this variability is likely due to changes in temperature and chlorine dose. Furthermore, the availability of residual chlorine in a water distribution system varies significantly with time or space. Since, the reactions do not exactly follow first order decay, the value of the decay constant changes with respect to time as well.

In an Urban water supply scheme the normal mode of supply in a developing country such as India, is of intermittent nature with restricted supply hours to each area. The supply to different areas takes place in rotation. The operating conditions in such a system can be very varied depending on the zonation pattern for rotational supply and duration of supply for each zone. Normally available models of distribution system such as EPANET are mainly applicable for continuous supply. While this as well as other models may also contain a provision for considering pressure dependent consumption, in an intermittent supply system, the network operation may shift from pressure dependent consumption to demand controlled consumption, both in space and time. There are several other modelling issues in intermittent water supply such as the effect of initial filling of pipes in several segments when the supply is started. Field experiments have shown, that it may take even one to two hours for air to be fully vented out from a distribution system, after which the normal modelling conditions will apply. Observations indicate that the Hazen-Williams C value in such a system may be low in the initial period of supply (associated with presence of air packets) and reaches the true value after a period.

In the use of artificial neural networks (ANNs), probably, the most difficult part is the determination of the number of data patterns to use in training the network. This paper presents a method of choosing the right data patterns to use in training in conjunction with power mixing of turbine mixers. Turbine mixers are used in a variety of ways from mixing of chemicals to the aeration of an activated sludge process. The network used is the back propagation neural network. First, a data set consisting of 20 patterns of turbine power arranged in increasing order of the power were generated. A plot of the power target value as the abscissa and the ordinate was then constructed. This produced a straight-line plot, inclined 45 degrees with the horizontal. Plot of predicted values obtained from various derived weights will coincide with this plot if the weights were derived correctly. The first and the twentieth patterns were then used to train the network. When the resulting predicted output was superimposed on the plot, the figure produced a large bump on the curve. Henceforth, the production of this bump guided the choice of the next pattern. The process then consisted of eliminating the bumps by the choice of the next training patterns guided using the respective preceding bumps. The method succeeded in producing network weights that produced a perfect overlay inclined at 45 degrees with the horizontal, as expected. The pattern sets of turbine power target values were then increased from 20 to 40 and, then, to 100. Using the derived weights, the network predicted the 40 and the 100 target values at an extreme accuracy corresponding to the same coefficients of determination for both of 0.9996.

The uses of biomass sorbents, which are less expensive and more biodegradable than synthetic sorbents, for oil removal from gas station runoff were investigated. Bench-scale flume experiment was conducted to evaluate the oil removal and retention capabilities of biomass sorbents which included kapok fiber, cattail fiber, Salvinia sp., wood chip, rice husk, coconut husk, and bagasse. Polyester fiber, a commercial synthetic sorbent, was also experimented for comparison purpose. Oil sorption and desorption tests were performed at a water flow rate of 20 L/min. In the oil sorption tests, a 50 mg/L of used motor oil-water mixture was synthesized to simulate the gas station runoff. The mass of oil sorbed for all sorbents, except coconut husk and bagasse, was greater than 70%. Cattail fiber and polyester fiber were the sorbents that provided the least average effluent oil concentrations. Oil selectivity (hydrophobic properties) and physical characteristics of the sorbents were the two main factors that influenced the oil sorption capability. The used sorbents from the sorption tests were employed in the desorption tests. Results indicated that oil leached out of all the sorbents tested. Polyester fiber released the highest amount of oil, approximately 4% (mass basis) of the oil sorbed.