Critical Transitions in Water and Environmental Resources Management

The Mesopotamian Marshlands previously constituted the largest wetland ecosystem in the Middle East and Western Eurasia. They play a key role in the intercontinental flyway of migratory birds, support endangered species, and sustain fisheries of the Persian Gulf. In addition to these important ecological benefits, these marshlands represent a unique element of our global heritage and resources. They have been home to indigenous human communities for millennia and are regarded as the site of the legendary "Garden of Eden." The marshlands once covered over 20,000 km² of interconnected lakes, mudflats, and wetlands within modern-day Iraq and Iran. However, in the past thirty years, over 90% of the marshlands have been desiccated through the combined actions of upstream damming in Syria, Turkey and Iraq as well as the development of extensive downstream drainage projects within Iraq. Before the recently concluded war in Iraq, no close-up monitoring of conditions in the marshlands has been possible for the international scientific community. Nevertheless, independent organizations have raised international awareness since 1991. After the end of major combat operation in the region in early May 2003, several reconnaissance surveys have been carried out and preliminary work has begun the long-term restoration of the marshlands. The work currently being carried out includes modeling of flow regimes, ecological surveys and monitoring demonstration projects in various locations. Past and present marshlands conditions are described here along with an overview of the work currently being undertaken.

On July 2003, the Italian Ministry of Environment financed a project called "New Eden" to the Iraq Foundation, an NGO active in the reconstruction effort in Iraq. The objective of the project focuses on the identification and prioritization of key water management problems, and on development of a comprehensive policy using a participatory process for water resources management in Iraq. Through resource planning, the project has identified actions to bring major improvements to the welfare of the Iraqi population and condition of their environment, both of which have significantly deteriorated over the last decade. The project would guide donor assistance for development and contribute beneficially to water management, including investments from other economic sectors. The first phase of the project was dedicated to data gathering to create an organized and comprehensive overview of the biophysical and socio-economical data. On the basis of collected data, the development of an integrated water management model is important to understand Iraq's water needs and to prepare an action plan for the management of water resources in Iraq. A major objective of the project includes concrete proposals for immediate actions that do not require substantial financing, but which can bring significant and rapid results to the Iraqi people and their environment. These analyses define the Priority Action Plan in agreement with local authorities and stakeholders to outline the capital investment program of emergency water works in selected critical areas. Successful completion of this project entailed the collaboration and cooperation of the existing engineering and scientific community in Iraq through missions of Italian and U.S.A. technicians coupled with the efforts of local Iraqi Foundation experts. Our approach included collaboration with local experts to identify and assist in solving key water- and environment-related problems.

This study focuses first on the macro-scale modeling of marsh hydrodynamics. Simulation results have provided a preliminary look at overall circulation patterns arising from inflows/outflows to the system. Our analysis benefited from access to reasonably detailed topographic maps and exceptional satellite imaging products. This enabled us to include in our model most of the hydraulic structures that are responsible for causing the desiccation of the marshes. The model also provided the means to assess the effects of hypothetical water releases into the marshlands. The results of the macro-scale circulation model are now used to develop micro-scale models for pilot projects at a more detailed level. The objective is to examine the interaction of marsh conditions (e.g., bathymetric features, vegetation, etc.) and hydrodynamic forcing (freshwater inflows, meteorology, salinity and temperature gradients, the presence of hydraulic structures, etc.). With the aid of newly acquired, highly accurate topographic data, the micro-scale analysis are providing insights on circulation patterns and velocity distributions and are applied to evaluate alternative restoration scenarios and to plan/design constructed wetlands to meet water treatment needs for villages in the region.

Tibetan Plateau plays an important role in past and present global climate; its modern environment change is both locally and worldwide significantly. In this article, we analyse the reason and consequences of the rapidly developing of desertification of Northeast Plateau grassland. It is found that the reason of desertification lies in the rapid growing population and under-developed economy; the affect of local human factors may be magnified through the desertification process and air circulation; through the surface albedo change and increased dust emission, this local environmental change may affect East Asia, the north hemisphere or even the global climate.

Regional decadal precipitation reconstructions for the arid and semi-arid zones of northwestern China were established by the use of different palaeoclimate archives such as ice-cores, tree-rings, lake sediments and written historical documents. Local rainfall reconstructions from single sites were averaged to obtain regional precipitation records for a western and eastern region of an arid and semiarid zone of northern China, respectively. All established regional precipitation curves display 5 dry periods, each lasting about 50 years. Meanwhile, precipitation reconstructions show regional dissimilarities. During the last 500 years, the trends of precipitation change in the eastern arid region are basically consistent with those in the western and eastern regions of the semiarid zone. Precipitation variations in the western arid region are unique, showing significant local patterns of rainfall variability. Maximum entropy method (MEM) spectral estimates show that each regional precipitation series contains stationary century-scale periodicities of about 120a. Singular spectrum analysis was applied to isolate the century-scale oscillation signals from the regional proxy precipitation series. Significant periods with wavelengths of 121.4a, 154.6a, 124.3a, 118.6a, 108.5a and 121.4a were found that explain 26.56%, 26.44%, 28.87%, 18.67%, 33.48% and 34.04% of the variances of the original series for the western arid zone, the eastern arid zone, the whole arid zone, the western semiarid zone, the eastern semiarid zone and for whole northern China, respectively.

Large areas of moraine sediments and cryogenic weathering products, formed by glacial action and a cold environment, are the main source of aeolian sand in the high and cold region of the Qinghai-Tibet Plateau in China. The evolution of aeolian sand geomorphology was closely related to the environmental conditions. Owing to freezen dune surface, the evolution of sand dunes was dominated by expanding dune bases and vertical accretion, and thereby forming huge barchan dunes. The migration rate of huge barchan dunes was very slow at an average rate of 1.8–0.5cm.a^–1, and their vertical accretion rate was 0.3–0.08cm.a^–1. The temperature mainly controls the environmental changes in the adjoining region of Puruogangri ice sheet. The dating of humus layers in studied sand dune was 10780±130, 9549±130, 8320±110, 7450±100, 5970±95, 5330±90, 4420±80, 3460±80, 2440±80, 2280±70, 1820±70, 980±70 aBP respectively. Viewed from the region's thermal mainly coming from the latent thermal of the summer monsoon, the changes of the Southwest monsoon intensity might be an important leading factor. As long as both water and thermal conditions are suitable, the plants will grow well, and sand dunes (land) will be in a stabilization state and form humus layers. Otherwise, sand dunes are bare and cause the re-activation of stabilized sand dunes.

Dust samples collected monthly for 15 years from 50 sites in 10 urban cities in Gansu Province, China, provide information on modern rates of dust deposition in the desert/Gobi and loess areas. The dust deposition is highest during spring months and lowest during autumn months, in both the desert/Gobi and loess areas. There is a significant positive correlation between dust deposition and dust event, and an inverse correlation between dust deposition and precipitation. The 15-year mean maxima in the desert/Gobi area and loess area are 498.64 and 327.02 t km^–2yr^–1, respectively, and the mean minima 290.22 and 180.86 t km^–2 yr^–1, respectively. Drought may have a widespread, major influence on the modern rates of the dust deposition in the desert/Gobi and loess areas in Gansu Province, China.

Results from remote sensing and field survey show that desertification of Sunite Steppe increased rapidly since the 1980s' and 26 percent of the 6092533 hectare grassland has become desertification land, of which there are 1073600 hectare slight desertification land, 311900 hectare moderate desertification land, 122700 hectare severe desertification land and 54700 hectare very severe desertification land. Three types of desertification land exist here: coarse gravel surface resulting from wind erosion, shrub-coppice dunes and sand sheets from sand deposit, and fixed dunes reactivation. It is found that the cause of desertification in Sunite Steppe lies in the non-equilibrium between changeful natural environments and human activities. In the arid and semi arid human-nature ecological system, there exists some kind of hysteresis that people adjust actions themselves with the change of environment factors. That should be why desertification develops rapidly today in a flourish steppe at one time.

Decreases in streamflow, particularly in the lower flow regime in perennial streams, are a significant concern to water-resource planners, especially in the Southwestern United States, where perennial streams are relatively few in number but provide valuable water supply, wildlife habitat, and recreational opportunities. Rapid population growth and the ensuing ground-water development have created the potential for a decrease in low flows in the upper Verde River Basin, Arizona. As a result, the performance of three U.S. Geological Survey streamflow-gaging stations at meeting low-flow monitoring objectives was evaluated on the basis of uncertainty in the streamflow data. In downstream order, the stations are Verde River near Paulden, Verde River near Clarkdale, and Verde River near Camp Verde.

A flood damage assessment study was recently completed for the proposed North Levee at the confluence of Salt and Gila Rivers in Phoenix, Arizona under the U.S. Army Corps of Engineers' (Corps) Tres Rios environmental restoration project. This paper presents the methodology applied to estimate stage discharge uncertainty and levee freeboard under regulated flows. U.S. Army Corps of Engineers' Flood Damage Reduction Analysis software HEC-FDA was used to obtain necessary freeboard for the proposed levee at a minimum 95% level, i.e., a non-exceedance probability of 0.95, for levee overtopping. The knowledge of high discharges (> 100 year) was important for setting the levee heights. A rating curve was generated in HEC-RAS using peak discharges of up to 500-year return period, and stage uncertainties were computed per Corps of Engineers' Engineering Manual EM 1110-2-1619 at several locations along the study reach based on the 100-year water surface elevations. Model uncertainty was computed based on Manning's n sensitivity analysis using HEC-RAS. As a second step in the levee freeboard determination, graphical exceedance probability functions were introduced in HEC-FDA. The levee freeboard was then calculated to be sufficient to maintain the minimum 95% reliability for the 100-year event along the Salt-Gila river being protected.

A method is proposed to delineate hazard maps for flooding and mud and debris flow events, based on the application of a two-dimensional flood routing model FLO-2D. The method defines hazard levels based on flood event frequency and intensity. The FLO-2D model has been enhanced with pre- and post-processor programs to automate data input and to generate hazard maps in ArcView GIS format. The proposed methodology was tested in twenty three sites in the Caracas and Vargas State region in Venezuela. This paper describes one application of the proposed method to the community of Tanaguarena on the Cerro Grande alluvial fan. The model results compare very well to the maximum flow depths and area of inundation observed during the December 1999 Vargas debris flow disaster. The hazard maps for the region are being used by local agencies to develop emergency plans and new land use policies. The methodology is being expanded to other flood hazard regions in Latin America.

Planning, design, construction and operation of lakeshore structures require information about the future likely extremes of the lake levels at a given risk percentage. The estimation of average return periods of hydrological events and the corresponding risks of failure of hydraulic structures that are associated with such events are important aspects in many water resources studies. Small changes in the elevation of the lake surfaces can lead to enormous changes in the amount of land surface lost when the lake level rises or exposed when lake level falls. Persistence is the most important property in any hydrologic design concerning the storage capacity of reservoirs, average return periods, failure risks, drought properties. Its consideration in analytical derivations of design criteria presents difficulties and for this reason most often the analytical expressions are obtained on the basis of non-persistent (independent) hydrological processes. Both return period and risk of failure can be easily calculated assuming the lake levels are independent. But it is well known that persistence is more important for lake level fluctuations. By using autorun function which is probability distribution free and robust parameter, return period and risk is determined. The application of the methodology is presented for the Great Salt Lake monthly water level data in western US.

The Great Salt Lake wetlands are one of the most important inland wetlands in the United States. The playa lake topography of the Great Salt Lake basin and the climatic affects on the inflow and evaporation rates produce large fluctuations in the water surface elevations. These fluctuations produce cycles of destruction and regeneration of the adjoining wetlands that may have long term benefits to the overall diversity of the wetland habitats, especially for water birds. The limited land available in Davis County and subsequent development may impact the wetlands around Farmington Bay and have created some controversy over the proposed construction of a highway (Legacy Parkway) through the wetlands. The impact of the Legacy Parkway was assessed by analyzing the affects the highway may have on the wetlands with regards to the fluctuating lake and diversion of the water sources to the wetlands. The results indicate that the highway will not act as a dike since the drainage design of the Legacy Parkway requires that the inflow from the existing streams and channels from a 100-year storm be conveyed under the roadway. This allows large flows in both directions. The water source for most of the wetlands in the area of the highway is groundwater. The legacy Parkway project will destroy 46 hectares (114 acres) of wetlands within the highway footprint and may have an adverse affect on the adjoining wetlands if the existing hydrologic conditions are not maintained with their hydraulic design. The major problems affecting the wetlands are water quality (pollution from surface water discharged into the bay), effluent from water treatment plants, unnatural vegetation and invasive species and hydrological modification.

The presence of water on Rosamond Dry Lake presents potential operational hazards and environmental issues for Edwards Air Force Base (EAFB). This research tests the utility of threshold precipitation in predicting inundation of Rosamond Lake, which is located within the boundaries of EAFB. Climatic data from the Western Regional Climate Center and remotely sensed data from the U.S. Army Corps of Engineers Cold Regions Research and Engineering Laboratories were used to determine the correlation between winter precipitation events and surface water inundation of Rosamond Lake.

A performance evaluation of flow predictions from four semi-arid watersheds in Arizona is presented and analyzed. The watersheds vary in size from 30 to 800 sq mi and belong to different climatic divisions with mean annual precipitation ranging from 9 to 19 inches. Daily streamflows were computed by the semi-distributed USGS Modular Modeling System (MMS) that uses a conceptual bucket type model to represent the local water balance. GIS tools have been used to characterize all the watersheds in a similar and consistent manner using the same predefined criteria. Simulations have been carried out in continuous mode using a daily time step (daily mode).

The presence of water on Rosamond Dry Lake presents potential operational hazards and environmental issues for Edwards Air Force Base (EAFB). This research tests the utility of threshold precipitation in predicting inundation of Rosamond Lake, which is located within the boundaries of EAFB. Climatic data from the Western Regional Climate Center and remotely sensed data from the U.S. Army Corps of Engineers Cold Regions Research and Engineering Laboratories were used to determine the correlation between winter precipitation events and surface water inundation of Rosamond Lake. The Poisson probability distribution was then used to estimate the probability of a specified number of precipitation events exceeding the threshold precipitation for the Rosamond Lake watershed in any given year. The Chi-square test for goodness of fit was used to assess whether or not precipitation can be represented by the Poisson distribution. The statistical methods were used in concert with remotely sensed data to provide insight into the validity of the approach.

Real time rainfall runoff modeling is rapidly advancing and soon will be the framework of a predictive early flood warning system. The Flood Control District of Maricopa County has supported the development of spatially and temporally variable rainfall simulation in the FLO-2D model. FLO-2D is a two-dimensional flood routing model that can simulate rainfall-runoff. Initially, FLO-2D was designed to simulate uniform rainfall on a finite difference grid system of a watershed or floodplain. The model system has been expanded to interpolate ASCII grid file rain data (such as NEXRAD rain data or the Maricopa County rain gage data) to incorporate spatially and temporally variable rainfall data. FLO-2D can also simulate a moving storm system. The variable rainfall-runoff can be simulated with multiple inflow flood hydrographs routed over urbanized alluvial fans and floodplains. Spatially variable rainfall losses are computed with the Green-Ampt model. The assignment of the Green-Ampt parameters are automatically generated by a processor program. Simulating spatially and temporally variable rainfall enables monitored rain storms to be replicated, design storms to be predicted, or real time network rain gages data to be simulated as a projected flood event. The new rainfall components in the FLO-2D flood routing model set the stage for integration for a predictive early flood warning system.

The effect of teleconnection climate signals on climate variability of different regions in the world has been the focus of interests of many investigations since early 90s. The feasibility of long-term climate prediction through understating of how climate signals may affect the regional rainfall, have made these studies important for strategic watershed management. This study is consisted of a regional assessment of rainfall and the long-term variation of climate in the southwestern part of Iran. In this region, there are considerable water resources in major rivers of Karoon, Dez and Karkheh. The climate regionalizing of the study area is done based on the long-term variations of observed rainfall during dry spells at each station over the region. Drought characteristics such as duration and severity are considered as the main criteria in the regionalizing process. Fuzzy clustering by equivalence relation method is used to cluster 23 stations over the region. The clusters show the regions with similar pattern of drought events. The relation between extreme events of North Atlantic Oscillation (NAO) and El-Nino Southern Oscillation (ENSO) and the seasonal rainfall at each station is assessed. This analysis is also done for the Sea Surface Temperature of Persian Gulf (PG-SST) as an indicator of rainfall in the southwestern part of Iran. The method of Fuzzy clustering is also used for this purpose. This method provides three regional maps, each showing the regions with similar behavior during the extreme events of ENSO, NAO, and PG-SST. Then clusters of similar signal-affected stations are compared with the clusters of similar-drought pattern. The similarity between these two kinds of clusters shows that drought pattern in which stations is affected by climate signals. The results show that the global circulation patterns such as ENSO, NAO, and PG-SST could be indicators of meteorological drought in the study area.

California is one of our nation's largest states, stretching from the arid and hot south at the Mexican border to the cooler and wetter state of Oregon (Figure 1). With a population of 34 million people, it is the most populous of any state in the Union. Its trillion-dollar economy (the nation's largest) would rank it as the fifth largest economy of any nation on earth. Within the continental United States, Mt. Whitney (4418 meters (m) above mean sea level) is the highest mountain, and Death Valley (86 m below mean sea level) is the lowest point. Most of the water supply in the state comes from snow and rain that fall on the Northern coastal area and high mountains, particularly the Sierra Nevada. The statewide average precipitation is approximately 0.08 m per year (20 inches per year). Much of the runoff is conveyed by canals and pipelines to the more populated southern part of the state and to irrigate agricultural areas in the Central Valley. One of the earlier conveyance systems was built over 85 years ago to bring Owens Valley water to Los Angeles. Because of the diversity of ecosystems in California, the generally semi-arid climate over much of the state, and the large and growing population (17 million people are expected to be added by 2030 to the 34 million people currently in California), California's water resources are particularly vulnerable to climate change. In this paper, an overview of the current and projected status of California's water resources is provided. Completed, ongoing, and anticipated work on climate change activities within the state are summarized. Then, the impact of projected climate change on water resources is discussed, along with potential adaptation strategies.

This study analyzes the influence of large-scale climate patterns on precipitation variability in the Colorado River Basin. Large-scale climate patterns, like ENSO and PDO, are associated with regional hydrologic variations. In order to link large-scale climatic variations to the variations of precipitation in the basin, a conceptual influence index is developed, and its spatial variability in the basin is examined. Various possible combinations of teleconnections are examined using the conceptual climate phase representing extraordinary climatic fluctuation to investigate their influences on seasonal precipitation variations in the basin.

The lower Colorado River basin is located in an area of known El Niño-Southern Oscillation (ENSO) influence. A streamflow forecast is developed using Pacific Ocean Sea Surface Temperatures (SSTs) as predictors in addition to a traditional ENSO predictor, such as the Southern Oscillation Index (SOI). Significant regions of SST influence on streamflow were determined using linear correlations (LC). These significant SST regions are then used as predictors in a statistically based exceedance probability model previously applied to streamflow stations in Australia and the U.S. Long lead-time (3 and 6 month) streamflow forecasts were developed for El Niño, La Niña and non-ENSO years for the winter-spring (January-February-March — JFM) season. The use of the SSTs resulted in improved forecasts, based on cross-validated skill scores, when compared to forecasts using the SOI. Additionally, forecast lead-times were increased when using the SSTs as predictors due to the inability of the SOI to provide an acceptable forecast. Also, the use of SSTs provided an improved forecast for all lead times for non-ENSO seasons when compared to the SOI forecasts. Following the methodology presented, water resource planners in ENSO influenced areas are provided a useful tool for forecasting streamflow.

The Virgin River in southwestern Utah and adjacent Arizona and Nevada is habitat for a number of endangered and threatened species of fish. The river also has significant loads of sediment that change the characteristic of the stream channel with time. The Virgin River transports large quantities of sand. Some sections of the river store the sand in the stream bed following a high stream flow event; the sediment is then removed by lower streamflows that can cause a wave of sand to pass through river channels in other sections of the river. The Hurricane Bridge on the Virgin River had a sand wave that passed through the channel during a low flow event that followed a high flow event. This paper demonstrates that antecedent conditions are important in the analysis of physical habitat in sand-bed rivers because the relation between the streamflows and habitat will change depending on these antecedent conditions.

Using historical data, the purpose of this study was to determine the occurrence and distribution both temporally and spatially of salt in the Rio Grande within the Rio Grande Project (RGP) in New Mexico and west Texas. The concentration in the river of total dissolved solids (TDS), major ions, and flow were analyzed generally on a monthly time step for river metering stations at San Marcial, below Elephant Butte Dam, below Caballo Dam, at Leasburg Dam, and below Mesilla Dam (all in New Mexico), and at El Paso and Fort Quitman, Texas. In addition, time series models have been constructed that may assist in better understanding the behavior of the system and also be used for forecasting purposes.

Phosphorus (P) transport to streams from furrow irrigation tailwater is a significant water quality issue in the Pacific Northwest. Under this NRI funded research project, we have collected field data to document phosphorus loads that occur under furrow irrigation conditions in Kimberly, Idaho. Since the majority of phosphorus transported off the field is associated with eroded sediments, we have developed an erosion detachment, transport and deposition component for the surface irrigation model SRFR. This model has the ability to track particle-size distributions in the furrow flow. A modification to the Laursen transport formula was required for small particle (silt and clay) sizes. Laboratory batch studies were conducted to determine the amount of phosphorus released when soil is entrained in water, as occurs under furrow erosion. There appear to be influences from both sediment:water mass ratios and P concentration in the water, as well as soil P indices. Flume studies were conducted to determine the amount of phosphorus released from the soil surface under non-erosive conditions. Preliminary results suggest that soil P release is influenced by the flow regime (turbulent or laminar) at the interface between the soil and the water in the furrow. Turbulent flow results in more P released, and potentially desorbed and diffused from deeper soil layers. A phosphorus transport component is being added to SRFR to model the transport of P with eroded soil and the desorption of P into the irrigation water from both the eroded soil and the soil surface. Several new modeling approaches are being explored. This paper presents a status report on this modeling - field verification effort on P transport from furrow-irrigated fields.

Irrigation impoundments, canals, drainage ditches, irrigated fields, and the incidental wetlands created by irrigation and drainage, provide valuable habitat for birds in and around Delta, Utah. Documentation of the creation of these areas and observations of bird use from 1995–2004 during migration, breeding, and wintering are reported. 147 species of birds were observed using incidentally created habitat sites.

The Salt Lake Valley (Valley) is the largest metropolitan area in Utah and is located along the front of the Wasatch mountain range. Wholesale domestic water supply is provided to the Valley by two main agencies, namely, Metropolitan Water District of Salt Lake City and Sandy (Metro) and Jordan Valley Water Conservancy District (Jordan Valley). Rapid growth and drought conditions have placed pressure on water supply and delivery. The cost of meeting the annual water supply demand of all agencies was analyzed to minimize the total cost.

Population and urban water need increases in the Utah portion of Cache Valley is forcing restrictions on groundwater use. We compute maximum sustained yield pumping strategies for two management scenario groups, differing in constraints. We apply constraints to prevent unacceptable heads, spring flows, and river-aquifer seepage (RAS). Scenarios differ in the limits imposed on head and/or pumping. Goals are to provide water for 18 municipalities using one new well for each. Computed maximum sustainable pumping can be as much as 38% greater than 1990 pumping, depending on applied bounds. However, imposing projected 2020 water needs as upper bound on pumping reduced the maximum total pumping rates. Optimal pumping is most sensitive to RAS restrictions. Significantly constraining RAS to the river significantly reduces the amount of sustainable pumping increase.

While much of the western United States may be considered arid, some regions receive more water annually than most agricultural zones in the rest of the US. The problem inherent to agriculture in a region such as California's north-central coast is that the timing of water availability is opposite that of need: roughly 45 inches of rain falls in the Russian River watershed, but all between November and March. Vintners (representing the most common form of agriculture in the region) therefore must meet water needs by obtaining it from supplies available during summer growing months, principally groundwater, surface streams, and artificial impoundments. Diversion of flow from small streams may conflict with flow needed by federally protected salmonids for spawning and rearing habitat. Historical stream flow data and water rights records highlight the disparity between stream flow and water needs: streams varying in size may dry at different points after the rainy season, but the demand for water from these streams continues throughout the summer. The difference between water demanded from small streams and the amount of water they can provide may be the primary hurdle to restoring coastal river fisheries.

"Keep it pure" is the slogan for the Salt Lake City watershed protection program. The City has put effort into protecting the water quality of the watersheds east of the valley that provide much the potable water used in the developed areas along the Wasatch Front. This paper describes the history of watershed management over the past 150 years.

Texas began statewide water planning in response to the devastating effects of the drought that ended in the 1950s. The 1950s drought prompted the Texas Legislature to create the Texas Water Development Board (TWDB) for the purpose of implementing statewide water planning. Since its inception, the TWDB has produced seven statewide water plans — in 1961, 1968, 1984, 1990, 1992, 1997, and 2002. In the mid 1990s, the State of Texas endured a significant drought, with some areas entering a new drought of record. In 1997, the 75^th Texas Legislature responded to concerns about the adequacy of water supplies by passing Senate Bill 1 (SB1) to provide for a new approach to water supply planning in Texas. In the past, the State planners had been responsible for determining what strategies and projects each region or city should implement to meet future water supply needs. SB1 is a radically different approach to statewide water planning, setting forth a grass-roots regional water planning process that calls for local representatives to develop strategies to meet water supply needs over a 50 year time period. SB1 emphasizes public participation in the planning process.

In this paper, the authors use an analytical model, GFLOW 2000, to simulate groundwater flow under different pumping conditions, with special emphasis on the analyses of well spacing and their proximity to canals and drains, and the effects of pumping on surface water flow. In the first pumping scenario, new wells and some of the existing wells were included in the model under pumping stresses of the year 2003. Additional simulations with different pumping stresses and different well spacing were then assumed to evaluate impacts of alternate pumping scenarios. Preliminary results indicate that wells located closer to canals and drains have strong impacts on surface water flow, whereas wells further away from the canals and drains have lesser impacts on surface water flow. When wells in close proximity to canals and laterals are pumped, seepage from unlined canals and laterals is significantly increased and some of drains can also dry up.

Work in progress is presented as part of an on-going investigation of mercury transport in the Carson River in west-central Nevada. The Carson River is one of the most mercury contaminated fluvial systems in North America. Most of its mercury is affiliated with channel bank material and floodplain deposits, with the movement of mercury through this system being highly dependent on sediment transport processes. In January 1997 a rare, high-magnitude flood generated significant geomorphic change to the system and transported large quantities of mercury. Calibrated bank erosion functions developed for the US EPA water quality model (WASP5), suggest that bank erosion is significantly greater at flows above bankfull discharge when compared to flows confined to the main channel with a single major flood event transporting nearly 87% of the total mass eroded during the period from 1991–1997 (Carroll et al., submitted). Preliminary mercury transport results for the 1997 flood are encouraging as it over predicts the observed value by less than 20%. However, mercury is under predicted following the flood as flows recede back into the channel and highlights the need to revisit Hg transport at moderately high flows.

The purpose of this paper is to answer the question of why it is so difficult to obtain reliable performance data for urban stormwater best management practices (BMPs) such as retention/detention ponds, wetlands, swales, infiltration devices, proprietary devices, and others. Our assessment of this question is based on extensive experience developing, implementing, and conducting stormwater monitoring programs, as well as development and oversight of the U.S. Environmental Protection Agency (USEPA)/American Society of Civil Engineers (ASCE) National Stormwater BMP Performance Database, on behalf of ASCE's Urban Water Resources Research Council. The paper begins with a list of factors that contribute to the difficulty of BMP monitoring, and concludes with a review of representative problems that we have experienced when reviewing BMP performance data sets that have been submitted to the National BMP Performance Database Clearinghouse for integration into the database.

The US EPA/ASCE National Stormwater BMP Database has grown significantly since the first evaluation of BMP performance data in the database was completed in 2000. The project team has conducted a limited re-evaluation of the data contained in the database to assess the overall performance of BMPs as well as compare BMP design attributes to performance. The evaluations have included the assessment of various BMP types as categorized in the database with regards to their ability to reduce runoff volumes as well as the effluent quality they can achieve. It is apparent that certain BMP types can reduce the volumes of runoff through soil soaking and resulting evapotranspiration and/or infiltration as opposed to BMPs that are more "sealed," such as wet ponds, wetlands, vaults and hydrodynamic devices. Runoff reductions contribute to pollutant loadings reductions as does improved effluent quality. On average, dry detention basins were found to reduce runoff volumes by an average of 30 percent (comparison of inflow to outflow), while biofilters reduced volumes by almost 40 percent. As expected, wet ponds, wetlands, hydrodynamic devices, and retention ponds show little or no runoff volume reductions. BMP types vary with regards to effluent quality that is achieved. BMPs such as wet ponds and wetlands appear to achieve lower concentrations in effluent quality than other BMPs such as detention ponds (dry) and hydrodynamic devices for some parameters. These differences vary with pollutant type. Finally, analyses of BMP design vs. performance are beginning to result in statistically valid results. For example, a relationship (ratio) between the treatment volume of retention ponds (wet ponds) vs. the average size measured storm event volume monitored has been established, showing that those with a ratio of 1 or more have been observed to achieve better effluent quality. Based upon these findings, this paper discusses potential BMP selection, design, and BMP design standards implications. A set of potential recommendations for how communities develop and specify BMP requirements such that more local goals for improved water quality is discussed.

Enhancement of simulation options for stormwater best management practices (BMPs) and hydrologic source control is discussed in the context of the EPA Storm Water Management Model (SWMM). Options for improvement of various BMP representations are presented with emphasis on incorporation of fundamental unit processes from environmental engineering into the algorithms.

So-called "best management practice" (BMP) performance for urban runoff control may be characterized on the basis of BMP effectiveness for control of both hydrologic and water quality impacts of stormwater and urban dry-weather flows. This paper deals with hydrologic impacts through the use of continuous simulation with the Storm Water Management Model (SWMM), in order to evaluate volume-based and flow rate-based controls. Case studies at sites in Greenville, NC and Austin, TX were used to demonstrate the methodology.

Bioretention is the process of using biological processes and rapid infiltration along with the storage of water to reduce contaminants. A bioretention cell creates an aesthetically pleasing water treatment area to help meet the National Pollutant Discharge Elimination System (NPDES) pollutant standards in, for example, the petroleum producing, refining, and distribution industry, parking lots of shopping centers, and residential homes. Pollutants that may be impacted consist of petroleum products (BTEX, cyclohexane, paraffins, and olefins), metals (cadmium, copper, lead, zinc), and nutrients (nitrogen and phosphorus). Currently there are no design models for bioretention cells that combine the removal of nutrients with infiltration, sorption of metals, and degradation of organic compounds. The model to be discussed will account for these constituents, as well as take into account specific site characteristics like climatic location, area drained, bioretention cell properties, and various loads of pollutants entering. The resulting output from the model would include the dimensions of an infiltration layer, an organic layer, and an anoxic layer, pollutant levels after moving through the cell with associated trapping efficiency, and water quantity leaving the cell. Results of preliminary column tests performed in the laboratory will determine the infiltration and sorption properties of various organic materials.

The Villanova Bio-Infiltration Traffic Island is part of the Villanova Urban Stormwater Partnership Best Management Demonstration Park. The site has been in operation for three years. It was designed to capture the first inch of runoff from the 50,000 ft² watershed. The Traffic Island is the subject of a continuous monitoring project. The site is fully instrumented to monitor water quantity and has recently been modified to collect water quality data. Ongoing research has shown that the site infiltrates 69% of the annual runoff produced in its 52% impervious watershed. Data from the site will be used to support the implementation of similar innovative urban stormwater management practices.

This talk will discuss methodology for data analysis, modeling, and phosphorous management analysis in complex watersheds and describes the application of this methodology to the assessment of best management practices for phosphorous in the 1200 km² Cannonsville Watershed, which is a catchment for one of the City of New York's drinking water reservoirs. Our group has used the model SWAT2000 to model hydrology, sediment, and Phosphorous movement over the watershed. Calibration and independent validation results obtained were excellent given the difficult of measuring and modeling these constituents. Current results indicate Phosphorous loadings will increase substantially in the future, even if economic activities and phosphorous management program do not change. We evaluate practices for reducing the loading of phosphorous including source control methods like precision feeding. We discuss the importance of incorporating soil storage of phosphorous in analysis of phosphorous BMPs.

The U.S. Environmental Protection Agency (USEPA) has funded the development of a decision support system for selection and placement of best management practices (BMPs) at strategic locations in urban watersheds. The primary objective of the system is to provide stormwater management professionals with a BMP assessment tool based on sound science and engineering that helps develop, evaluate, select and place BMP options based on cost and effectiveness. The system is called the Integrated Stormwater Management Decision Support Framework (ISMDSF) and is being designed through a systematic review of modeling needs, technical requirements, current and emerging data management technology, and available watershed and BMP models. The ISMDSF will be applied to a real urban watershed to evaluate its ability. There are four major design aspects for the ISMDSF development. First, the system provides a robust computer platform for BMP selection, sizing, and placement in the context of several integrated watershed factors and influences. Second, it is applicable to mixed land use urban watersheds, and can perform watershed simulation based on watershed size, scale, anthropogenic, and natural characteristics. Third, it incorporates hydrologic/hydraulic and water quality modeling, integrating surface runoff and direct discharges to surface water bodies, based on relevant data collection. Finally, it will have the capability to objectively evaluate multiple solution alternatives based on cost and the desired water-quality objectives.

Silt fences are among the most common BMPs implemented for sediment control at construction sites. Field evaluation of the efficiency of silt fences in removing sediment in highway construction runoff showed that sediment removal was not attributable to filtration by the fabric, but was the result of particle settling in the pool of water behind the fence. The median concentration of solids discharged from the silt fence controls was approximately 500 mg/L. This concentration is significantly lower than the 3000 mg/L of TSS often reported in construction runoff and suggests that approximately 75% of the particles in the runoff were removed by sedimentation in the water impounded bythe fences. Geotextile silt fences proved to be ineffective in reducing turbidity. The particles responsible for creating turbidity are mainly silt and clay and the particle diameter normally is much smaller than the size of the openings in the fabric. The observed data indicated that silt and clay size particles comprised 92% of the total suspended solids that reach the face of the silt fence. High sediment removal efficiencies were achieved with silt fences in flume studies. The geometry of the flume created a large ponded area behind the controls resulting in long detention times and significant particle settling even with the fine grained sediment used in the tests. Mean efficiency for sediment removal in the flume ranged from 68 to 90% and was highly correlated with the detention time. These data indicate that silt fences should be sited in the field to maximize the volume of water impounded behind the fence. Flow rates through the silt fence fabric of sediment-laden runoff were two orders of magnitude less than those typically specified by many regulatory agencies. The flow rates of a sediment slurry through geotextile fences are a function of apparent opening size as well as permittivity (or other measures of clean water flow rates). The fabric that resulted in the longest detention times in this series of flume tests had the highest reported permittivity, but the smallest apparent opening size, suggesting that clogging of the fabric with sediment affected the hydraulic performance.

This paper explores the factors affecting pollutant removal performance of retention (wet) ponds using data contained in the International Stormwater BMP Database. While the Database contains numerous studies with varying amounts of detail, this comparison is based on the performance of only those sites with reported basic design characteristics and water quality data (event mean concentrations) so that constituent concentrations can be accurately determined and related to the design of the individual BMP. The use of selected BMPs has a number of advantages. Some of the variability in performance observed for facilities of a specific type can be explained by differences in design and/or watershed characteristics; consequently, the expected performance for a given set of conditions can be predicted more accurately.

Retrofitting existing stormwater facilities is of critical importance to restoring impaired watersheds. Many existing storm water detention basins were designed solely for stormwater runoff peak discharge magnitude abatement or were under-designed for current water quality control standards. These basins can be retrofitted to trap additional amounts of non-point source pollutants. This study analyzed a detention basin in Morris Township, NJ as an example to demonstrate the tradeoff between water quality benefits and increases in flooding problems resulting from retrofit. As part of the first retrofit, an outlet structure with a small-diameter circular orifice at the bottom and a rectangular weir on the top was placed in front of the existing large diameter outlet pipe. Diameter of the orifice and height of the weir were varied during hydraulic routing. Quantitative relationships were obtained between the increase in stormwater detention time (pollutant removal efficiency) for a defined, small, infrequent water quality storm and the maximum increase in peak outflow rate (flooding impact) for a full range of storms. It was found that beyond a specific weir height, the detention time no longer increased whereas the maximum peak outflow rate increased greatly. The longest detention time that could be achieved was also calculated for different orifice diameters. In addition, effect of different definitions of the water quality storm on the achievable detention time was quantified.

Due to the considerable profit made by planting tea trees on the high mountains in Taiwan, farmers fertilized their tea farms with about 600 kg of N/ha/y and 210 kg of P/ha/y to maintain a high production. However, the tea trees utilized only 5% of the fertilizer, and the remaining 95% of nitrogen and phosphorus were lost. The lost nitrogen and phosphorus caused a problem of eutrophication in the downstreams of a reservoir. The best management practices of the tea farm is to reduce 20% dosage of the fertilizer. Also, the reduction of the 20% can still ensure not to damage the good quality of the tea production.

The nonpoint source pollution control is in the beginning stage in Taiwan. Currently, the soil and water conservation law provides regulations that the land developer should retain the exceeding stormwater runoff and sediment due to any significant development in the watershed. Most of stormwater controls have been designed at the local, or on-site level, but not at the regional or watershed level in a comprehensive and systematic manner. The purpose of this study is to establish an optimization model for the optimal placement of structural Best Management Practices (BMPs) at the watershed scale. The complete model consists of two interacting components, i.e. an optimization model base on dynamic programming (DP) and a zero-dimensional reservoir water quality model. To search for the optimal placement strategy of BMPs system, a preliminary procedure is needed to find suitable sites for various BMPs construction in Fei-Tsui Reservoir watershed in northern Taiwan. The landuse type, terrain and land ownership are taken into consideration by using a geographic information system (GIS). In the optimization model, the objective function is to minimize the total construction cost of BMPs and constraints consider water quality standards for total phosphorus and suspended solid concentrations in the reservoir. Also, the zero-dimensional reservoir water quality model of Vollenweider type is embedded in the whole optimization framework to estimate the average pollutant concentrations after pollutants from nonpoint source pollution flow into Fei-Tsui Reservoir. Due to the advantages of DP, the most cost-effective BMPs placement strategies can be found in this study. The trade-off curves between the resulting optimal cost and benefit of water quality improvement can be found which provide the decision makers with clear and useful information for nonpoint pollution control and management at the watershed scale.

The present paper describes the planning and implementation of a best management practice (BMP) system for a clay mining facility in Northern Taiwan. The mining factory is a 32-hectare operation located next to the Masu River, whose watershed is still relatively undeveloped except for the clay mine and some chicken and hog farms. Processed water and stormwater runoff from the mining areas have been found to cause sediment and acidity problems for the Masu River. The Water and Environment Research Center (WERC), working with Taiwan Normal University, initiated a study in 2002 on using BMPs for mitigating the impact of the clay mining operation on the Masu River water quality. The long-term BMP plan includes the use of coral rock ditches, swales, dry and wet ponds and constructed wetlands. The first phase BMP plan called for the use of three existing ponds in series, modified to include coral rocks, wetland vegetation and swales, to form a BMP treatment train. Field sampling results show that the ponds are very efficient in removing suspended sediment (more than 90%), but not as efficient in raising the pH level in the effluent water. Further tests are continuing at the present time.

Environment protection or restoration of the endangered or polluted plateau lakes in Yunnan, China, has brought intentions of mass activities that raise a great need for best management practices (BMPs). By relating algal bloom and nutrient status in the lakes plus cost-effective and sustainability analysis, it is recommended that phosphorous should be intentionally cut first to control algal bloom, pollutant source reduction should start from the source on water surface and ecological modules should be promoted to guarantee sustainability of environmental protection measures. Part of above strategy has been implemented and led positive results.

Xili reservoir is one of the major water sources for Shenzhen City, China. The reservoir has long been impacted by pollutions from a number of sources including fruit garden runoff which surrounds the reservoir and the reservoir bottom deposits. To identify the pollution source specifically, the qualities of bottom deposit and overlying waters were investigated through a year-long period. A number of pollution sources were investigated including fruit garden, flower garden, vegetable land, fish pond, villages, express roads, river water and a long-distance diversion channel. More than 27 thousand data were collected. It is concluded that the water qualities were mainly influenced by the bottom deposits through pollutant releasing under anoxic conditions, while the bottom deposits accumulated through deposition of nonpoint pollution sources. About 1.8 millions kg of SS were poured into the Xili reservoir through runoff. It is found that the accumulation of phosphorous in bottom deposit has been brought to control, however the accumulation of organic nitrogen is becoming even more serious. A newly finished diversion project can improve the water qualities of Xili reservoir by renewing the water every 15 days. However, it is also found that the diversion project may bring 3.8 millions kg of suspended materials a year into the Xili reservoir which will influence the Xili reservoir significantly in many aspects. Finally five control strategies are suggested to further improve the Xili reservoir.

With the rapid economic development in the last two decades many Chinese cities are facing serious urban water pollution problems, such as uncontrolled point source and non-point source (NPS) pollution, polluted urban lakes, rivers as well as the damaged urban aquatic ecosystems. In order to develop and demonstrate the urban water environment quality improvement technology which is suitable to most Chinese cities, the Ministry of Science and Technology of China launched a key research project, "to improve the quality of urban water environment: technology and demonstrations" in 2002. Eleven cities across the country, including Shenzhen, Guilin, Suzhou, Dalian, Wuhan, Beijing, Shanghai, Tianjin, Daqing, Zhenjiang and Zibo were selected as demonstration cities. More than twenty research institutes and universities are involved in this effort to develop technologies and processes to control NPS pollution, to treat and reuse the wastewater, to restore impacted aquatic ecosystems and others. This paper will describe the main urban water environmental problems in China and report on the research progress of the ongoing key project.

Excess stormwater runoff is a serious problem in a large number of urban areas, causing flooding, water pollution, groundwater recharge deficits and ecological damage to urban streams. It has been posited that to mitigate the effects of excess stormwater runoff, policy makers could use economic incentives to reward those who install detention technologies. This paper builds on research investigating the use of a tradable allowance system to promote placement of dispersed, smaller-scale best management practices (BMP) for water detention to control excess runoff. Using continuous hydrologic modeling and cost information on BMP that include opportunity costs of land dedicated to the BMP, we investigate the outcome of scenarios that include diverse price and spatial constraints on BMP placement.

Stormwater samples collected from storm sewers draining small municipal separate storm sewer systems shown to be free of cross connections within an urban watershed dominated by a single land use were analyzed for pathogens (Pseudomonas aeruginosa and Staphylococcus aureus) and indicator organisms (total coliforms, fecal coliforms, fecal streptococci, enterococci, and E. coli). The samples were collected from three land-use designations (high-density residential, low-density residential, and landscaped commercial) in all four seasons. Flow-weighted samples were collected using automatic samplers connected to flow meters. Rain gauges in the drainage basin recorded the event while the flow meter recorded flow velocity, water level, pH, dissolved oxygen, conductivity, and temperature of the runoff. The concentrations of microorganisms in the stormwater runoff were found to be similar to, but less variable than, those reported in the National Pollutant Discharge Elimination System (NPDES) database. Organism concentrations from high-density residential areas were higher than those associated with low-density residential and landscaped commercial areas. Concentrations of organisms were significantly affected by the season during which the samples were collected. The lowest concentrations were observed during winter. The results will help watershed managers target BMPs or treatments to land uses having higher runoff loads having greater impact on lowering the total load to a receiving water.

This paper describes different approaches for controlling pathogen contamination in urban watersheds for contamination resulting from point or diffuse sources. Point sources of pathogens can be treated by a disinfection technology of known effectiveness, and a desired reduction of microbial concentration is usually achieved. Diffuse sources of pathogens are commonly managed with the use of structural and nonstructural best management practices (BMPs). The primary function of BMPs is reduction of flow volume, sediment, and/or nutrients, while reduction of pathogens is a secondary function and not always quantified.

Best management practices (BMPs) are used to mitigate the impacts of development on the environment. A life cycle analysis (LCA) is a systematic method for identifying and quantifying the effects of various factors upon BMP field efficiencies. Two life-cycle factors for 2 different LCAs are studied. The first factor is the installation probability for designed and approved silt fences during the construction phase of individual lots in a residential subdivision. Ownership change of the lots during construction is also investigated to determine its effects on this factor. The second factor is the probability a permanent residential detention facility will receive uncontrolled sediment loads from individual lots during the construction phase. The study focused on 14 residential subdivisions from which 184 construction phase lots were field visited and the files of 330 lots were researched on land records. Designed and approved silt fencing was only installed at 27% of the field visited lots. Of the 330 records researched, 76% of the lots underwent a construction phase ownership change. In the 14 subdivisions, 47.4% of the permanent detention facilities were receiving uncontrolled sediment loads from 1 or more individual lots. Typical damages to the facilities included partial or full clogging of basin inlet and outlet structures, loss of reservoir volume, side-wall erosion, and vegetation damage.

Although the objectives of urban stormwater management practices are generally the same internationally, domestically within the United States (U.S.), and regionally for a given area of the U.S., the best management practices (BMPs) used to accomplish those objectives can vary substantially. The purpose of this paper is to compare and contrast those best management practices. Although the authors have reviewed literature on practices throughout the world, the primary basis for commentary on international practices has been obtained through the primary author's service as principal investigator on a project to evaluate performance and costs of BMPs and sustainable urban drainage systems in the U.S. and the United Kingdom (U.K.). This project is jointly funded by research foundations in both countries. Commentary on national practices has been derived from Black & Veatch BMP projects carried out in states throughout the country, including Oregon, California, Nevada, Arizona, Colorado, Texas, Oklahoma, Kansas, Nebraska, Iowa, Missouri, Louisiana, Illinois, Michigan, Ohio, Tennessee, Florida, North Carolina, Virginia, Maryland, New Jersey, and Massachusetts. The area of the U.S. used to demonstrate a regional perspective on BMPs in the Kansas City metropolitan area, where Black & Veatch has performed numerous watershed studies and stormwater facilities master plans, and recently prepared a regional BMP guidance manual.

The Kansas City metropolitan region includes two states, eight counties and over 100 municipalities with populations ranging from less than 1,000 to over 400,000. Each entity has its own vision, priorities, and needs. Not surprisingly, the region possesses significant variability in terms of local policies, codes and standards. The textured geographic, economic and political terrain limits the ability of communities to work together to address regional issues such as stormwater management. Despite the many hurdles, the community has developed shared resources to manage public works projects through the local chapter of American Public Works Association (APWA). The chapter was formed in 1963. In 1981, it produced a set of design criteria and specifications for storm drainage projects that were adopted by ordinance in many communities. They were updated in 1990 to incorporate lessons learned during their use in the 1980s. While these standards functioned well, evolving local priorities led the community to revisit how stormwater was being managed. As a result of severe flooding in October 1998, a strong regional consensus reframed the practice of stormwater management to address water and environmental quality, in addition to flood control and public safety. The rationale underlying the new approach includes substantial long-term savings in the development and maintenance of public infrastructure, better protection of natural resources, and compliance with new federal water quality regulations. A regional "Water Resources Action Plan" developed in 2001 by area public works and planning directors from communities throughout the metro area called for more regional cooperation on watershed management through an institutionalized Water Resources Policy, Planning and Practices (WRP3) committee structure. A key short-term objective of the committee was to revise local stormwater engineering standards and planning guidelines.

Kansas City, Missouri traces its genesis back to the year 1821, when the Missouri Territory was admitted to the Union as a state. That same year, a Frenchman named Francois Chouteau traveled up the Missouri River from St. Louis and established a trading post on the south bank of the river. Other French families gradually joined him. From the beginning, the fledging settlement suffered from flooding. In 1826, M. Chouteau rebuilt his trading post on higher ground after a large, destructive flood. Now more than 1,800,000 people live in the Greater Kansas City Metropolitan Area. The area straddles the border between the states of Kansas and Missouri. Moreover, the area consumes part or all of eight counties and all or part of one hundred seventeen incorporated municipalities. By far the largest municipality is Kansas City, Missouri with a population of approximately 450,000. Spread out over approximately 320 square miles, the City now includes parts of four counties and continues to grow in both population and area as new territory is annexed. Roughly half of the City lies north of the Missouri River and half south of it. The area south of the river generally is older and much of it is served by combined sewers. The City extends from the old downtown and industrial areas along the river southward some 18 miles characterized by newer and newer developments. The Northland generally is much younger in character and continues to undergo rapid urbanization.

While many types of inlet devices may capture some stormwater debris, care must be taken in their design. Catchbasins with sumps may remove up to about 30% of suspended loads that enter the intet, but much of this material is relatively coarse and in many cases would not have moved to the outfall. The sumps do minimize sediment accumulation in the sewerage and reduce maintenance. These should probably be considered as grit traps, more than pollutant trapping devices. Some devices can also trap floatables. However, if not frequently maintained, clogging and ponding may occur. In addition, if water is forced through the trapped debris (especially leaves), degradation of the organic material may occur, actually causing the production of some pollutants. Some new inlet devices have been recently designed and are undergoing testing that promise more effective control of stormwater pollutants, along with better retention of bed load material and floatables.

The authors have been involved in numerous street cleaning research projects for over 30 years. This paper will give us a unique opportunity to highlight the historical use of street cleaning as a method for reducing stormwater pollutants, and will examine newly emerging street cleaning technologies. There have been many misconceptions concerning this potential stormwater management control. This paper will examine the limitations of street cleaning, and describe how it can be more effective. Street cleaning plays an important role in most public works departments as an aesthetic and safety control measure. Street cleaning is also important to reduce massive dirt and debris buildups present in the spring in the northern regions. Leaf cleanup by street cleaning is also necessary in most areas in the fall.

A major storm sewer outfall was retrofitted with a Continuous Deflective Separation (CDS) unit and a linear marsh to help treat stormwater discharged from an urban drainage basin in Tampa, Florida. The CDS unit was very effective for removing the gross solids including litter, trash, debris, leaves and sediments larger than 64 microns. These pollutants have not usually been evaluated in stormwater studies that use automated water quality sampler. Most of the gross solids were deposited between February and August 2003. During the first year of this ongoing study, about 300,000 cubic meter of flow passed through the system including both storm and base flow. The CDS unit removed 11.69 m³ (413 ft³) of gross solids from the flow stream including toxic levels of Polycyclic Aromatic Hydrocarbons (PAHs).

The California Department of Transportation (Caltrans) as part of the Best Management Practice (BMP) Retrofit Pilot Program evaluated the performance of a BMP installed at 2 sites on I-210 in Los Angeles County to capture and retain floatables, trash, debris, fine sand and larger particles of sediment. Runoff and rainfall data were collected during the 2000–01 and 2001–02 seasons at each of the sites and gross pollutant and trace element data were collected during a total of 10 events at one site and 17 events at the other site. Floatables were characterized during each season while all solids captured by the BMPs where quantified and characterized in 2001–02. The efficiency of the BMPs was determined using a Caltrans Scoping Study Methodology adapted from a Federal Highway Administration 1989 study and a Mass Balance Approach that considers the mass of material captured by the BMP. These two methodologies produced significantly different results in the BMP efficiency. These material differences were attributed to limitations of the Scoping Study Method when small data sets are used, incorrect use of sediment data, use of calculated rather than actual measured data and limitations of automatic samplers to collect representative samples of solids in storm water runoff. The Mass Balance Approach produces a better method of characterizing BMPs when there are solids in the storm water runoff.

One type of Best Management Practice (BMP) available is the use of street storage systems to prevent combined sewer surcharging and to mitigate basement flooding. A case study approach, based primarily on two largely implemented street storage systems, will be used to explain the concept through construction and operation aspects of street storage systems. Street storage refers to the technology of temporarily storing stormwater in densely developed urban areas on the surface (off-street and on-street) and, as needed, below the surface close to the source. Close to the source means where the water falls as precipitation and prior to its entry into the combined, sanitary, or storm sewer system. The idea is to accept the full volume of stormwater runoff into the sewer system but greatly reduce the peak rate of entry of stormwater into the system. System components include street berms, flow regulators, and surface and subsurface stormwater storage facilities. By eliminating surcharging in combined sewer systems, street storage has the potential to cost effectively and simultaneously mitigate basement flooding and combined sewer overflows (CSO). Other possible benefits of street storage are mitigating sanitary sewer overflows (SSO), eliminating surface flooding, reducing peak flows at wastewater treatment plants (WWTP) and controlling non-point source pollution. This technology, in the two case studies discussed here, has proven to eliminate surcharing, complied with regulations, proved to be cost effective and earned public support.

This paper presents a systematic approach for the economical design of rain-water quality control systems. For the design of runoff quality control system (RQCS), the rainfall-runoff process requires the local rainfall data recorded continuously. In this study the rainfall probability distribution is assumed to follow an exponential decay function. Applying the exponential decay function, the normalized curves are derived to explain the non-exceedance probability distributions. The optimal curves for the determination of the RQCS size are derived based on the overflow risk. Comparison of the optimal capture volume and peak runoff rate to those computed by an urban rainfall-runoff model (ILLUDAS) demonstrates that the optimal curves derived in this study can be utilized for the design of rain-water quality control systems in Korea avoiding an excessive computational effort by the rainfall-runoff model.

A pilot scale investigation was undertaken at the Allegheny County Sanitary Authority (ALCOSAN) for approximately 14 months to evaluate the feasibility of cross-flow ceramic membrane microfiltration for the treatment of primary sewage effluent simulating combined and sanitary sewer overflows. Ceramic membranes of various pores sizes (0.05 – 1.4 microns) were tested to understand the impact of cross flow velocity, transmembrane pressures, and inlet suspended solids on permeate water quality and flux rate. A 0.2 micron membrane operated with a 1.8 m/s cross flow velocity, a transmembrane pressure less than 2.1 bar and a backpulse frequency of 60 seconds was selected as the preferred operating conditions. Permeate from this membrane had virtually undetectable levels of fecal coliforms, e-Coli, enterococci, and levels of BOD₅ and SS commensurate with secondary treatment expectations. Elevated membrane-feed suspended solids concentration reduces steady state flux rates; a quantification of the influence of feed suspended solids is for the design of multi-stage membrane systems.

Continuous Deflective Separation (CDS) screens are being used around the world in a number of different ways to treat CSO and SSO wet weather flows. In addition to fine screening (effectively to less than 100 microns) with direct discharge to the environment, CDS units are being used as pretreatment for constructed wetlands. This fine screening has also been demonstrated to provide sufficient removal of solids to allow effective disinfection with UV light. CDS screening has been demonstrated for clarification of chemically flocculated raw sewage, leading to the proposed use in providing a higher level of treatment for wet weather CSO/SSO flows. The regulatory environment in which these applications units are utilized is discussed; performance data for these various applications will be presented, along with design criteria and descriptions of several actual facilities.

Recently, the City of Milwaukee, Wisconsin embarked on a long term planning effort based on the Environmental Protection Agency's watershed approach. The use of this integrated approach was prompted by the understanding that if every sanitary sewer overflow was stopped tomorrow, non-point pollution would still cause beaches to close and make fish unfit to eat. The Milwaukee Metropolitan Sewerage District (MMSD) launched this effort to develop an approach to reduce non-point water pollutants. To insure success, MMSD believes it must increase public awareness and public educationof the sources of these pollutants in order to reduce non-point pollution. The use of good stormwater management principles by the general public is a major component of this effort. For these principles to be effective, the public must be informed about how water becomes polluted and what manmade influences induce flooding. This public education must start with the basic understanding of what happens to a drop of water when it lands on an individual's property. Knowing how that drop of water becomes polluted, how it can cause flooding, and how it can cause sewer overflows is critical to any stormwater management program. Public outreach, public education and public collaboration are the tools to making the leap to less flooding and cleaner waterways; stormwater management cannot be effective without it.

The Urban Watersheds Management Branch of US Environmental Protection Agency Edison, NJ, Laboratory is in the process of developing design guidelines for a number of BMP techniques. This paper provides a summary of the design guidelines that are being developed for a group of onsite BMP control practices that are referred to as vegetative biofilters and includes the following BMP control practices: grass swale, grass swale with filter media (dry swale), wet (wetland swale), filter and buffer strip, bioretention cell.

Today, many municipalities are implementing best management practices (BMPs). The most commonly used structural treatment BMPs that will be discussed in the presentation are ponds (detention/retention) and vegetated biofilters (swales and filter/buffer strips). Historically, structural BMPs were employed to capture peak flows, assist in local drainage, and manage the quantity of runoff produced during wet-weather flow (WWF), i.e., flood control. Current design typically still focuses on rainfall events that range from 2 to 10 inches of daily rainfall and occur at much longer return periods ranging from 2- to 100-year storm. These storms may contain significant pollutant loads, but their contribution to the annual average pollutant load is really quite small due to the infrequency of their occurrence. In addition, longer periods of recovery are available to receiving waters between larger storm events allowing receiving water systems to flush themselves and the aquatic environment to recover. Water quality control of urban runoff is still a relatively new and developing area of engineering and science. The addition of water quality considerations in the design of BMPs has introduced a new dimension to the traditional hydrologic considerations for BMP design. Water quality considerations have created a shift from flood events to a continuous long-term rainfall-runoff BMP design volume approach and the pollutant loads associated with these volumes. To treat the bulk of the pollutant loads from stormwater runoff, many states and municipalities specify a treatment volume that is designed to capture the initial component of the stormwater runoff. In practice this is achieved by specifying a rainfall amount (e.g., the first ½-inch or 1-inch) or the capture of a stormwater runoff volume that correlates to a design storm (e.g., 6-month, 1-year, or 2-year frequency storm). BMPs that encompass both peak discharge hydrology and small storm hydrology would optimally use a system that incorporates on-site treatment and storage of stormwater for the smaller storms while protecting downstream from floods. By including supplemental measures using either distributed and/or centralized controls, the peak discharge control strategies can be upgraded to perform water quality control.

Increased urbanization has increased the amount of impervious area resulting in larger quantities of stormwater runoff. This runoff can contribute significant amounts of debris and pollutants to receiving waters. Urban watershed managers often incorporate best management practices (BMPs) to treat these pollutants. Although many studies have evaluated the capabilities of two popular treatment methods — retention ponds and constructed wetlands — to reduce pollutant concentration and load in stormwater, few have focused on the internal mechanisms controlling the efficacy of these stormwater BMPs. Experimental mesocosms were constructed in Edison, NJ to simulate retention pond and wetland BMPs and to evaluate, describe, and model key processes that occur within these treatment systems. Understanding the hydrologic character of the mesocosms, and BMPs in general, is necessary to model the performance of these systems. Preliminary data suggest constructed wetland mesocosms outperformed retention pond mesocosms for removal of total suspended solids and total Kjeldahl nitrogen. However, retention pond mesocosms were the most effective at treating dissolved nitrates. Complete data that includes hydraulic modeling, multiple nutrient parameters, microbial indicators and metals will provide more information on the mechanisms controlling BMP effectiveness. A primary goal is developing models for constructed wetland and retention pond BMPs that will allow for better stormwater BMP design leading to overall improvements in receiving water quality to meet state and federal water quality standards.

Greater public awareness and the enactment of increasingly stringent regulations have made reliable and cost-effective erosion control methods or best management practices (BMPs), particularly on construction sites, a national priority. At present, silt fence is the most widely used BMP, but unfortunately, the instances of failure are widespread. We believe that with proper assessment, following recommended installation methods, and conducting routine maintenance, better erosion control can be achieved with current silt fence technology. The current phase of our research is to investigate the current technology and develop design tools to assist users to get better results through avoiding structural failures and also provide a means of assessing performance in terms of trapping efficiency or effluent concentration. The next phase of our research will be to develop and evaluate several structural, material, and chemical enhancements to the current technology. This paper describes the progress made to date in investigation of current technology and development and application of the design tool.

EPA is developing national effluent guidelines for discharges of stormwater from construction sites to augment the existing Phase I and II NPDES stormwater regulations. As part of this effort, EPA is required to evaluate the environmental benefits of the rulemaking. In evaluating the benefits of regulations, EPA attempts to both quantify the degree of reduction of a range of pollutants or indicators, as well as monetize the expected environmental improvements that result. Due to the national scope of this evaluation and limited data available on both the quality of discharges from construction sites and the locations of those discharges, as well as the limited range of methodologies available to quantify and monetize in-stream improvements, EPA relied on a modeled approach in lieu of using actual field-sampled data. For this evaluation, EPA estimated sediment export from construction sites using the Revised Universal Soil Loss Equation (RUSLE) and modeled sediment reductions using SEDCAD. EPA then developed a methodology for determining the geographic distribution of construction sites and modeled in-stream water quality improvements and monetized benefits of several regulatory scenarios using EPA's National Water Pollution Control Assessment Model (NWPCAM).

This study evaluates the potential impact of watershed development on sediment transport in a prototype headwater stream subjected to typical residential development. Event based and continuous simulations, using 50 years of hourly rainfall records were performed for two climatically different locales. The first is the semiarid climate of Fort Collins, Colorado and the other a typical southeastern climate, Atlanta, Georgia. Five scenarios were evaluated for the study watershed, including: current (undeveloped) conditions, fully developed conditions, without stormwater controls, and fully developed conditions with stormwater controlled using (a) the City of Fort Collins flood control standard, (b) the City of Fort Collins flood control standard and water quality capture volume (WQCV) criteria, and, (c) using common standards of practice in the United States: control of the 100- and 2-year storms to historic peak discharge rates and control of the WQCV. For each scenario examined, sediment transport potential is evaluated for two noncohesive soil types: medium gravel and medium sand.

Asphalt surfaces have greatly increased the amount of pollutant-carrying runoff entering surface waters. To counteract this, permeable pavement can be installed to allow water to infiltrate, thus reducing runoff and acting as a filter. This study tested the surface infiltration rate of 27 permeable pavement sites in North Carolina, Maryland and Delaware. One of these sites in North Carolina was monitored to compare pollutant loads of asphalt runoff to those in infiltrate. Concrete grid pavers (CGP) and permeable interlocking concrete pavers (PICP) were tested with pavement ages ranging from six months to 20 years. Two infiltration tests were run on 14 CGP lots filled with sand. The initial test was on the existing condition of the surface and second test was run after the removal of the top layer of residue (1.3 – 1.9 cm) to simulate maintenance. Maintenance improved the infiltration rate on 13 of 14 sites. Analysis of the data showed that maintenance improves surface permeability at a confidence level of 99.8%. The median average infiltration rate increased from 5.0 cm/hr., for existing conditions, to 8.0 cm/hr after maintenance. Eleven PICP sites were also tested. Sites built in close proximity to loose fine particles had infiltration rates significantly less than sites free of loose fines. Averages of each condition are 60 cm/hr and 2000 cm/hr respectively. Even the minimum existing infiltration rates were comparable to those of a grassed sandy loam soil. Water quality data included in this study shows the results of six storms from June to October, 2003. With only a few storms to compare, only Zinc has been identified as having a statistically significant difference between infiltrate and runoff.

Clean Water Services (District), formerly Unified Sewerage Agency, is a County Special Service District operating in the urban area of Washington County, Oregon. The District is in charge of the public storm water and sanitary sewer systems, and we administer the erosion control and water quality programs in the urban areas. There are 12 member cities and a large unincorporated area within our boundary. In 2002, the Field Operations Division started the construction of a permanent field yard. The site is 2 hectares (5 acres) in size with a 1,672 m² (18,000 sq. ft.) administration office, 1,394 m² (15,000 sq. ft.) of indoor vehicle parking, and 1,208 m² (13,000 sq. ft.) of outdoor carport parking. The $9 million facility houses 58 employees and more than 70 pieces of equipment including street sweepers, dump trucks, crew vehicles, sewer cleaning equipment, excavators, and more. The Field Operations crews maintain 1,156 km (718 miles) of sanitary sewer lines, 591 km (367 miles) of storm lines, sweep public streets monthly, clean storm drains and catch basins, maintain more than 300 stormwater detention and water quality facilities, and provide 24-hour emergency response to sewer overflows and flooding within our 316 km² (122 square-mile) service area. It was essential that the construction be done to the highest standards relative to erosion control and water quality. One of the project goals was to incorporate innovative erosion control and water quality treatment aspects, both during construction and that would be in place in the long term. The project goals called for the site to be a demonstration project so contractors, engineers, plan review staff, and the public could see these techniques in action.

Site specific characterization data are very important in estimating the effectiveness of wet-weather controls (WWCs) in removing pollutants from highway and urban runoff. An analysis of urban runoff sedimentation data is used to develop first order reaction rate constants using linear regression techniques. The reaction rate constants and associated initial concentrations are used to evaluate the dependency of detention basin designs on removal rates. A continuous simulation spreadsheet model is used to evaluate the pollutant removal performance of a detention basin as a function of storage volume and release rate for a one year period. The design is optimized by minimizing the total cost of the detention basin while maintaining a specified level of pollutant load removal. Simulations were run with different reaction rate constants and results from the optimized solutions indicate that storage-release strategies are heavily dependent on the assumed reaction rate constant. Different performance levels were also evaluated and the results indicate that attempts to achieve pollutant load reductions above 80% are much less cost effective.

The objectives of this paper are to document the pollutant removal of existing vegetated areas adjacent to highways and to identify the minimum design requirements needed to obtain substantial pollutant reduction. The pollutant removal performance of highway shoulders (buffer strips) is based primarily on eight test sites located on highway rights-of-way in California that were not designed for stormwater treatment. At each location, multiple stormwater collection systems (over 30 systems total) were installed and equipped with automated samplers to collect highway runoff after passing through vegetated buffer strips of varying widths. The vegetated buffers at each of the eight sites differed in characteristics such as slope (up to 50 percent), width, vegetation type and coverage, soil, and hydraulic loading. The data collected between 2001 and 2003 indicate that concentration reductions consistently occur for TSS and total metals and frequently for dissolved metals. Conversely, concentration increases were observed for dissolved solids and occasionally for organic carbon. Nutrient concentrations were generally unchanged by the buffer strips. The pollutant removal and design of roadside swales is based primarily on the performance of two sites located in Austin, Texas, and which were constructed solely for stormwater conveyance. The data collected between 1994 and 1997 indicate that the majority of the pollutant removal occurs on the side slopes of the swales rather than along the length. Minimum discharge concentrations observed at these sites were similar to those observed from buffer strips in California. The greatest pollutant removal occurs when the geometry of the swale maximizes the length of the side slope; consequently, the optimum cross-section geometry for highway medians is V-shaped rather than the trapezoidal geometry normally illustrated in guidance manuals. In addition, the relatively constant side slopes facilitates mowing and is safer for the traveling public than channels with abrupt changes in side slopes. Swales along highways are normally sized to convey large, infrequent events, which when combined with the low side slopes implemented for safety reasons results in a broad bottom that reduces the concentration of runoff constituents. Consequently, a flat bottom is not required for effective operation.

Stormwater runoff from highways can be detrimental in terms of the water quality and quantity of runoff produced. The roadside area can be improved to filter more suspended pollutants from the runoff and also to infiltrate more of the annual stormwater volume through the use of controls such as biofilters. A simple continuous simulation example was developed to illustrate the process of setting up a spreadsheet-based design optimization problem. Accurate site-specific infiltration estimates are critical for the continuous simulation. Optimization procedures generated the least-cost combination of storage to infiltrate 80% of the annual runon volume of stormwater. The model shows the conceptual framework for integrating continuous simulation, optimization, and cost estimation into the design process.

Due to recent fish kills along the Mid-Atlantic coast, new regulations require all future land disturbing activities — including the urbanization or development of land — to be environmentally friendly. Consequently, stormwater management practices are being designed to remove nutrients, particularly nitrogen and phosphorus, before runoff reaches streams. One stormwater practice, bioretention, is gaining popularity in commercial development because it can easily be sited in the required natural areas of parking lot medians; however, the current design criteria used for bioretention have not been tested extensively. Pollutant removal, especially for species of nitrogen, has not been tested under various conditions. Currently used bioretention monitoring protocol assumes bioretention devices retain little water. Current designs do not reduce nitrate-nitrogen levels sufficiently, as bioretention is constructed without any zone designed to be saturated. For nitrate-nitrogen to be converted to nitrogen gas, thus enhancing total nitrogen (TN) removal, an anaerobic zone may be necessary. This research determined the effect of an anaerobic layer within bioretention devices on the concentrations and loadings of TN, nitrate-nitrogen (NO₃-N), and other nutrient and pollutant species in stormwater runoff including ammonia-nitrogen (NH₃-N), total kjeldahl nitrogen (TKN), total phosphorus (TP), ortho-phosphate (Ortho-P), zinc (Zn), iron (Fe), copper (Cu), lead (Pb), and total suspended solids (TSS). Four field bioretention areas were installed and monitored in two North Carolina cities. The field sites were used to test the pollutant removal abilities and hydraulic retention of bioretention devices that ranged in depth from 0.6 m (2 ft) to 1.2 m (4 ft) and were designed with or without an anaerobic layer. The field studies in North Carolina confirmed high annual TN mass removal rates at two conventionally drained bioretention cells (40% reduction). NO₃-N mass removal rates varied between 75% and 13%. The presence of an anaerobic zone was tested at the Greensboro field site and significantly (p<0.10) reduced TP and Fe outflow concentrations. Annual metal mass removal rates at the Greensboro site were over 98% for both Zn and Cu and 81% for Pb. A very large mass of Fe (nearly 17 Kg from a 0.20-ha (0.5-acre) watershed was added to the storm sewer network by the bioretention cell. TP and TSS removal rates were highly variable from cell to cell.

Low-impact development (LID) techniques can be used for controlling stormwater impacts from new development and redevelopment / urban retrofit. Prince George's County has developed a Best Management Practices (BMPs) evaluation module to assist in assessing the effectiveness of LID technologies. This module uses process-based algorithms to simulate BMP function and removal efficiency and accepts flow and water quality time series either from observation or generated by runoff models such as the Hydrologic Simulation Program, FORTRAN (HSPF) as input data. Process-bases algorithms include weir and orifice control structures, storm swale characteristics, flow and pollutant transport, flow routing and networking, infiltration and saturation, evapotranspiration, and a general loss/decay representation for a pollutant. The module offers the user the flexibility to design retention style or open-channel BMPs, define flow routing through a BMP network, and compares BMP controls against pre-development conditions. BMP effectiveness can be evaluated and estimated over a wide range of storm conditions, site designs, and flow routing configuration approaches. The simulation processes of the model have been validated using available monitoring data and literature values. The BMP Module has been applied to several real land development projects using actual long-term meteorological data recorded at Washington National Airport. These development projects were designed based on the State stormwater management requirements and regulations. The simulation results indicate that LID can actually minimize water quantity and quality impacts of the developments. For all cases, the results indicate that post-development condition with LID BMPs exceeds existing condition (assumed to be completely forested) in terms of hydrologic attenuation benefit and pollutant control benefit. This paper provides an overview of the model and demonstrate the simulation results. The results also raise an interesting question: Is using the Natural Resources Conservation Services (NRCS) Rainfall Type II Distribution appropriate for the design of stormwater BMPs? With the U. S. Environmental Protection Agency's support, the model is currently being enhanced to include the following: linkage to the CSO model, development of an optimization module, and integrating with GIS platforms.

Impervious surface is a basic parameter used in hydrologic studies, particularly in urban watersheds. The availability of a new generation satellite image collection systems providing data with a high spatial resolution, in digital format, comparable to what is obtained from aerial photographs, creates a new source of data available to civil engineers involved in hydrologic or stormwater projects. Aerial photography provides a spatial resolution (less than 0.5 m for a large scale photo) that is higher than what can be obtained from civilian satellites (i.e., 0.6 to 4.0 m range), however, there are a number satellite image characteristics that make them a preferred source of impervious data for some applications. The major advantages of satellite images over aerial photographs are the capability (1) to provide up-to-date information for an area of interest, (2) to allow for digital data processing and integration with other digital datasets and (3) to produce color displays. This paper presents results of the use of medium and high spatial resolution satellite image data to estimate the impervious surface areas in Springfield, Columbia and Ashland, Missouri and Overland Park, Kansas urban watersheds. The results provide an important overview of quality of the results that can be expected for urban watersheds characterized by different development patterns, building densities and varying amounts of vegetation (trees and grass). The results provide guidance on the use of satellite image data to produce impervious surface estimates required by most hydrologic studies. The results of this study are valuable to the profession because existing regulatory programs such as the Federal Emergency Management Agency's (FEMA) National Flood Insurance Program and the Environmental Protection Agency's (EPA's) Phase I and Phase II storm water regulations require proactive actions to measure the impact of land use/land cover and storm water runoff on water quantity and quality. The results presented in this paper are a contribution to the state-of-the-art methodology applicable to urban hydrology projects by formulating guidelines to determine what satellite data sources are appropriate for a given application or modeling effort.

The paper describes the development of a planning tool for comparing different LID-type BMP implementation scenarios in a watershed and selecting the optimal spatial placement strategy for the BMPs. The framework of the planning tool includes the following components: A watershed model, which simulates nonpoint pollutant loads from all parts of the watershed under given wet-weather conditions; A sub-program that performs computations of the water quality impact of the implementation of LID-type BMPs such as bioreention cells in a watershed; An "optimization" tool, which will search for the optimal scenario for BMP placement in the watershed. In the present study, a watershed model developed at the University of Virginia, called WINVAST, was used. WINVAST is a comprehensive, event-based stormwater model, which simulates pollutant loads generated from a given storm event and has interfacing capability with GIS databases. A sub-program to the WINVAST model was developed to simulate certain BMP processes (currently wet detention ponds and grassed swales) in a watershed. The optimization tool used in the present study was the "Scattered Search" methodology. A tributary watershed to the Swift Creek Reservoir near Richmond, Virginia was used as a case study. Several BMP implementation scenarios were tested, namely, on-site or micro-scale such as LID-types, sub-regional and regional placement of BMPs. The results show that the on-site placement scheme would cause a larger reduction of watershed runoff peaks, but the water quality results were mixed. On the other hand, when watershed water quality requirement is higher, the placement of BMPs tends to favor locations closer to the watershed outlet over upstream locations on a cost-effectiveness basis. The results also show that when the maintenance cost was tied with the amount of sediment accumulation in a pond BMP, the overall system cost would be lower.

Stormwater reduction practices (SWRPs) such as rain barrels, rain gardens, porous pavement or roof gardens have become popular approaches to reducing the volume and peak rate of stormwater runoff. The effectiveness of such measures is an important factor in decisions concerning investment of public funds in such practices. There are two issues in the assessment of SWRP effectiveness: (1) how well do specific SWRP implementations work in the field and (2) what are the long term "end of pipe" benefits of a broad implementation program. The analyses presented in this paper seek to answer the second question. Continuous hydrologic simulation is employed to determine the benefits of various SWRPs in the combined sewer service area (CSSA). The effectiveness of individual SWRPs are assumed based on various papers and websites.

Filterra represents the latest advancement in Bioretention technology for treatment of contaminated urban stormwater runoff. Americast, a Division of VALLEY BLOX, Inc., working with the University of Virginia's Civil Engineering Department has developed a best management practice (BMP) that optimizes the physical, chemical and biological pollutant removal mechanisms found in the plant / soil / microbe complex of the Bioretention BMP. It was the objective of Americast to standardize the application of Bioretention technology taking it out of the realm of experimental devices by developing a much more reliable, dependable and adaptable urban BMP. In working towards this goal, Americast has developed a patented specially engineered high flow rate treatment media in a structurally sound container or simply "Bioretention in a box" system. Laboratory data demonstrates that Filterra can treat 90 to 95 % (higher if desired) of the total annual rainfall volume with maximum pollutant removal rates of 95% for total suspended solids, 82% total phosphorous, 76% total nitrogen and 91% heavy metals (measured as Cu). Filterra's excellent pollutant removal efficiency is primarily due the multiple treatment systems inherent in Bioretention's plant / soil / microbe treatment complex. The actual total annual pollutant removal for an application will vary depending on site conditions, rainfall intensity and the annual volume of runoff treated. Americast has also devised a simple and unique design /sizing strategy that is adaptable to the unique rainfall distribution / intensity of any hydrologic regime. This unique approach is based on developing a rainfall intensity probability distribution from actual rainfall information taken over many years from local weather stations (e.g. 50 years or more of airport rainfall data). From an area's intensity distribution a simple surface area to drainage area ratio and curve can be developed to size the device to treat the desired annual volume. By multiplying the annual volume treated by the pollutant removal efficiency of Filterra, the total annual pollutant removal rate can be calculated. This paper will outline the wide array of pollutant removal mechanisms and the unique design sizing strategy for Filterra.

The purpose of Low Impact Development (LID) is to minimize the effect of changes in land use on the post construction site hydrology. Specifically this encompasses minimizing increased stormwater volumes, peak flows and non-point source pollutants generally associated with land development. Infiltration is a key component of LID in accomplishing these goals. The author discusses the relationship between goals and capture volume, placement, and general usage of infiltration Best Management Practices (BMP) incorporated within the LID Framework.

As we move into the new century approximately 50% of the world population lives in the built environment and these demographics will increase to 60% of the population by the year 2025. With this growth is a commensurate increase in impervious surfaces, in particular pavement, associated with modern society's dependence on personal and commercial vehicular transportation. Well-established consequences of this increase in imperviousness in our modern built environments are deleterious impacts to water quantity (local flooding as well as local scarcity), water quality (specifically metals and particles) and local climate. In addition, impervious surface impacts to quantity and quality are related. Impervious surfaces alter the local hydrologic cycle (increased flow rates, increased volume and decreased lag time) and promote increased conveyance of anthropogenic constituents. Even with such understanding, conventional transportation and urban infrastructure engineering design philosophy and construction practices utilize impervious pavement almost exclusively. Conventional practice and current regulatory developments are at odds, resulting in a significant environmental and economic dilemma in our built environments. Permitting requirements for NPDES Phase II stormwater regulations have been in place as of March 2003. In large part, this significant cost associated with these regulations, is due to the spatial extent of impervious surfaces, alteration of the rainfall-runoff relationship and generation of anthropogenic constituents from activities on these impervious surfaces. While no one solution is the single answer to this complex problem, the development and application of multifunctional engineered porous pavement materials can fulfill modern built environment requirements for relatively smooth load-transmitting surfaces while serving as a more environmentally-conscious infrastructure material for in-situ rainfall-runoff quantity and quality control.

Urban rainfall-runoff is capable of entraining and transporting a wide gradation of particulate matter and constituents such as metal elements and organics. Under conditions where rainfall-runoff pH is relatively high (> 7) combined with higher alkalinity (> 50 mg/L as CaCO₃), long residence times and high particulate loadings, cationic metal elements transported in urban land use rainfall-runoff are often adsorbed on or incorporated with entrained particles that are ubiquitous in such runoff. Infiltration-exfiltration can be an effective structural in-situ particle separation and quantity control unit operation and process (UOP) that also will allow runoff to return to soil after passive physical-chemical treatment. The in situ partial exfiltration reactor (PER), which combines the surficial straining of a porous pavement (CPP) layer with filtration of oxide coated sand (OCS) media beneath, provides control of water quantity and quality. In this study, particle analyses were carried out for both influent and effluent to examine filter efficiency as a function of particle size and hydrology. Influent d_m/d_p ratios suggest that the dominant PER particle separation mechanisms were physical-chemical filtration with the CPP layer functioning as a straining surface. Particle size distributions were modeled based on a 2-parameter cumulative power-law function. PER filter performance was shown to be a function of unsteady hydrology and for all events examined the PER functioned as a variably saturated system. Temporal variation in the filter coefficient and the volumetric particle fraction remaining were directly related to the unsteady influent loading rate. Particle removal efficiency by the PER based on concentration ranged from 71 to 96% on a mass-based concentration and 92 to 99% on a number based concentration. Results suggest that a properly designed PER can provide effective in-situ control for particles and could be combined with or function separately from source control (i.e. pavement cleaning or a mass trading framework).

The processes involved in the relationship between rainfall and runoff are directly impacted by urban and transportation infrastructure and anthropogenic activities. Consequently, the results of such interactions impact the quantity and quality of receiving waters and the surrounding environment. Systems that combine both infiltration and exfiltration such as an in-situ partial exfiltration reactor (PER) loaded by transient sheet flow, have the potential to attenuate the impact of both the quantity and quality of urban runoff. These in-situ systems are subject to highly variable water quality and quantity and generally function under variably saturated flow conditions. To improve the understanding of the hydrologic response of a field-scale PER that has been monitored for water quality and quantity control, a two-dimensional numerical model was used to simulate the effluent hydrograph and water content profiles under transient hydraulic loadings for the PER. Richard's equation was applied in the model using parameters estimated from laboratory and field experiments; as well as event-based hydrographs measured in situ for the PER. The dynamics of the water content with time illustrated the function of PER to lower peak flow, capture and redistribute runoff volume, and attenuate temporal aspects of the inflow hydrograph despite surrounding clayey soils of low hydraulic conductivity. Simulation of the same historical loadings for different surrounding soils demonstrated the effect of the surrounding soil hydraulic conductivity on the hydraulic performance of the PER. Simulations for a series of design storm events showed that the exfiltration capacity of the surrounding clayeysoil was limiting for peak flows during the 5 year return storm. Depending on the climate, evaporation could be the dominant mechanism for the drying process in the top layer of the PER. During the summer climate in Cincinnati, Ohio approximately 2 days was required to evaporate all but residual moisture from the upper layer of the PER.

This 3-year study was funded by the Water Environment Research Foundation (WERF) to investigate innovative heavy metal removal from stormwater. This research focused on two major areas of heavy metal control: media filters and swales. This presentation will focus on the filter tests. The test stormwater used for most of the laboratory tests was collected from a campus parking lot at the University of Alabama during many storm events. The characteristics of the stormwater indicated that the most prevalent metals detected in the runoff were iron, zinc, copper and small amounts of particulate bound lead. Ranges of the metals were within national ranges reported by other studies. Twelve media were chosen for initial evaluation. Equilibrium and kinetic studies were formed on these media to assess their performance in capturing metals from urban runoff. The three best performing media: peat-sand mix, compost, and zeolite were then selected for in-depth study. The results of this investigation emphasize the importance of characterizing the stormwater before selecting a treatment media since the type and quantity of metals, pH, and other runoff characteristics can vary a great deal between sites. Upflow columns proved more effective than downflow columns in the control of detention time and reduction in clogging of the media by solids and associated head loss in the column.

The University of Alabama and the Center for Watershed Protection were awarded an EPA Office of Water 104(b)3 grant in 2001 to collect and evaluate stormwater data from a representative number of NPDES (National Pollutant Discharge Elimination System) MS4 (municipal separate storm sewer system) stormwater permit holders. The monitoring data collected over nearly a ten-year period from more than 200 municipalities throughout the country have a great potential in characterizing the quality of stormwater runoff and comparing it against historical benchmarks. This project is creating a national database of stormwater monitoring data collected as part of the existing stormwater permit program, providing a scientific analysis of the data, and providing recommendations for improving the quality and management value of future NPDES monitoring efforts. Each data set is receiving a quality assurance/quality control review based on reasonableness of data, extreme values, relationships among parameters, sampling methods, and a review of the analytical methods. The statistical analyses are being conducted at several levels. Probability plots are used to identify range, randomness and normality. Clustering and principal component analyses are utilized to characterize significant factors affecting the data patterns.

This presentation reviews the results of several recent and related stormwater research projects that focused on the treatability of stormwater heavy metals.

In this paper, a methodology for conjunctive use of surface and groundwater resources is developed using the combination of the Genetic Algorithms (GAs) and the Artificial Neural Networks (ANNs). Water supply, reduction of pumping costs and controlling the groundwater table fluctuations are considered in the objective function of the model. In the proposed methodology, the results of the PM groundwater simulation model are used to train the ANNs based simulation model. This model is then linked to the GA based optimization model to develop the monthly conjunctive use operating policies. The proposed model is applied to the surface and groundwater allocation in the irrigation networks in the southern part of Tehran, the capital city of Iran. The results of the proposed model show the significance of an integrated and systems approach to surface and groundwater resources allocation in the study area.

NOAA's Great Lakes Environmental Research Laboratory (GLERL) developed their Large Basin Runoff Model (LBRM) as a serial and parallel cascade of linear reservoirs representing moisture storages within a watershed. Each reservoir represents a moisture storage: surface, upper soil zone, lower soil zone, and groundwater zone. GLERL adapted the LBRM from its lumped-parameter definition for an entire watershed to a two dimensional representation of the flow cells comprising the watershed. This involved changes to the model structure to apply it to the micro scale as well as organization of watershed cells and an implementation of spatial flow routing. GLERL modified the LBRM continuity equations to allow upstream inflow when the model is applied to a single cell within a watershed and found the modifications in terms of corrector equations to be applied to the original solution. They began by considering flows between adjacent cells' surface storages while keeping the upper soil zone, lower soil zone, and groundwater zones in each cell independent. Thus each cell's upper soil zone, lower soil zone, and groundwater zone connected only to that cell's surface zone and not to any other cell, but the surface zones connected between adjacent cells. GLERL further modified the model to allow subsurface routing between cells of surface runoff (from the upper soil zone), interflow (from the lower soil zone), and groundwater flows (from the groundwater zone). Now surface and subsurface flows interact both with each other and with adjacent-cell surface and subsurface storages. This involved adding additional flows out of the various subsurface storages in a watershed cell and additional flows (from upstream watershed cells' subsurface storages) into the storages. The continuity equations are again modified in terms of corrector equations applied to the original solution and are derived handily. GLERL then organized LBRM applications to constituent watershed cells into a flow network by identifying the network flow cascade and automatically arranging the cell computations accordingly. They identified required characteristics of any flow network map and designed system checks to guarantee them. They devised a scheme for ordering computations and routing surface flows throughout the watershed.

Software for formally optimizing groundwater and conjunctive water management has improved dramatically in recent years. Historically utilized have been a range of classical and heuristic optimization methods, and of simulator and surrogate simulator techniques. Different combinations of optimizers and simulators are best for different types of optimization problems—groundwater supply, groundwater plume management, and conjunctive use. Methods for speeding the optimization process include linking a heuristic optimizer to Tabu Search or artificial neural networks.

In many cases the ability to determine and quantify groundwater and surface water interaction based upon hydraulic observations is an arduous task. The task is further complicated by possible transient interactions that can obscure chemical leakage signals observed in the groundwater system. The impetus of this paper is to provide an introduction to two studies where groundwater is being recharged from surface water sources on a relatively fast time of travel. The two test cases demonstrate the research plan for identifying groundwater-surface water interaction in an unconsolidated semi-confined aquifer and a karst setting. The study identifies emerging techniques and hypothesis testing for determination of groundwater under the direct influence of surface water and the potential role of these techniques in possible quantification of the flux. The tracer techniques used in the research include geochemical testing, tritium/3-helium, and microbiological molecular techniques. These inter-disciplinary studies are currently underway and this paper will provide a summary of the methodology, preliminary data and interpretation of the variant signals observable in a groundwater system from surface water interaction.

Principles of sustainable groundwater exploitation are presented in this article. The renewable and inappropriable nature of groundwater is examined in light of the process of recharge. An example illustrates the interplay among groundwater extraction, recharge, natural recharge, and storage. It demonstrates the aquifer-specific characteristics of overdraft and replenishment, which are driven by climatic variability and the rate of groundwater mining. A second example uses game theoretic methodology to quantify the roles of cooperation and non-cooperation on groundwater extraction. The economic and environmental advantages of cooperative groundwater extraction are demonstrated with data from a coastal aquifer.

To cost-effectively restore contaminated sites to productive use requires integrated planning of cleanup and reuse. Ultimate reuse objectives should guide remediation decisions, instead of being dictated by pre-described remediation plans. This study links a robust groundwater remediation approach with reuse planning. A genetic algorithm optimizer that is integrated with a subsurface contaminant transport model (The Modular Groundwater Optimizer) is modified to allow for a robust search. A robust approach is used to increase the likelihood of successful remediation in the context of the uncertainties in the system. The genetic algorithm objective is to find a low cost and reliable remediation plan. The remediation costs are then reduced by the expected reuse benefits. Reuse benefits are dependent on the time required for the remediation stage and on the endpoint of remediation. As more complete remediation is reached additional reuse options may become feasible. An analysis approach is developed and will be demonstrated to explore the trade-offs between increasing costs for more complete, faster, and more reliable remediation versus the potential reuse benefits gained from implementing more profitable reuse options and from beginning reuse sooner. The case study shown is based on the Emmell's septic landfill on the National Priorities List.

Management of groundwater contamination often involves conflicting objectives and substantial uncertainty. A critical source of this uncertainty in groundwater problems often stems from uncertainty in the hydraulic conductivity values for the aquifer. For a remediation solution to be reliable in practice it is important that it is robust over such potential errors. This paper presents the application of a robust multi-objective optimization method on a field-scale pump-and-treat design problem at the Umatilla Chemical Depot site at Hermiston, Oregon. A simple methodology is used to establish plausible realizations of hydraulic conductivity that are then efficiently sampled within the optimization framework using Latin Hypercube sampling. A noisy multi-objective genetic algorithm, developed and tested earlier on a hypothetical aquifer, is then applied to this field-scale case to come up with a set of robust and Pareto-dominant design solutions for the clean up of contaminants (RDX and TNT) in the groundwater. Interactions between the various trade-offs and the inherent uncertainty at the site are analyzed. Finally it is demonstrated that by using such robust multi-objective optimization schemes, it is possible to increase robustness of the optimal solutions without significant increases in costs.

The objective of this study to analyze the effects of parameter uncertainty on the optimal design of cost-effective and reliable remediation plans that use natural attenuation with active remediation. An optimization framework is used identify appropriate remediation strategies that are inexpensive and effective under several scenarios of parameter uncertainty, thus assessing the potential of natural attenuation with active remediation. The overall simulation-optimization framework combines a contaminant fate and transport simulation model and spatially-correlated random field generator with the enhanced multi-objective robust genetic algorithm (EMRGA) optimization approach. The optimization problem minimizes the expected cost of the natural attenuation-active remediation system while minimizing the expected dissolved contaminant concentration, given uncertainty and heterogeneity in system parameters. Decision variables are the extraction well locations and rates of the active remediation system. The EMRGA evolves a set of Pareto-optimal solutions that represent the best designs that identify the trade-off between the expected cost and clean-up level. The optimization model is applied to a problem based on a field site located in Eglin Air Force Base, Florida, contaminated with benzene. The optimization model is subjected to a series of cases, which vary the model parameters and degree of uncertainty associated with these parameters. The uncertain parameters examined in this study are the heterogeneous hydraulic conductivity, hydraulic gradient, and first-order benzene degradation rate. The results indicate that hydraulic conductivity is the most sensitive parameter, increasing the difficulty in achieving lower cleanup levels and higher remediation reliabilities.

In the federally approved Comprehensive Everglades Restoration Plan (CERP, http://www.evergladesplan.org/), the restoration of the South Florida ecosystem is a major task for the U.S. Army Corps of Engineers and the South Florida Water Management District. 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. The Biscayne Bay Coastal Wetlands (BBCW) Project is one component of more than 60 restoration plans and has a goal to restore or enhance freshwater wetlands, tidal wetlands, and near shore bay habitat. In an effort to restore wetlands, several structures, and management plans and scenarios are considered. One of the plans is to deliver fresh water from the existing canals through a shallow spreader canal system that is to distribute fresh water through wetlands into the Biscayne Bay. To achieve this, a tool is needed to design this complicated shallow spreader canal system. This paper presents how a spreader canal system, which includes 1D canal network routing, 2D overland flow, 3D subsurface flow, and flow through the interface of any two sub-domains of the spreader canal system, is simulated with the WASH123D computer code. A brief description of the model will be given. A hypothetical example that includes two cases and uses topographic data and a high-resolution computational mesh for one project area will be considered to demonstrate how WASH123D can help design a spreader canal system. A couple of issues concerning run time and numerical convergence of the coupled flow model will also be discussed in this paper.

WASH123D model is a first-principle, physics-based model, where water flow and/or contaminant and sediment transport within a watershed system are computed. In the WASH123D model, a watershed is conceptualized as a coupled system of 1-D canal/stream network, 2-D overland regime, and 3-D subsurface media. It is designed to answer the environmental issues concerning both water quantity and quality. To reach numerical solutions with reasonable and tolerable computer time for a regional scale watershed simulation, numerical algorithm improvement and code parallelization are two essential tasks when a distributed numerical model, WASH123D, is used. This paper presents the code parallelization approach, followed by demonstrating its scalability performance. The test problem of a large mesh uses the topographic data in the C-111 Spreader Canal Project.

This case study of groundwater policy regulation and documentation analyzes the roles of state and local agencies in the protection of public health and water resources at the former Fort Ord, California. State requirements for documents restricting land use represent a possible redundancy to a local ordnance that establishes a "Special Groundwater Protection Zone" at the former Fort Ord. Is this redundancy necessary to ensure the protection of public health and the environment?

The Everglades Agricultural Area (EAA) Storage Reservoir is a critical element of the Comprehensive Everglades Restoration Plan (CERP) being jointly implemented by the United States Corps of Engineers (USACE) and the South Florida Water Management District (SFWMD). The project purposes are to improve environmental conditions in Lake Okeechobee by reducing harmful high and low water levels; reduce the frequency of damaging flood releases from Lake Okeechobee to the estuaries; provide agricultural flood protection and water supply in the EAA; improve the performance of the large constructed wetland treatment systems (stormwater treatment areas) by attenuating inflows; and improve the timing of environmental water deliveries to the Everglades. Because the EAA is centrally located within the massive south Florida water management system and the EAA Project is addressing multiple goals, comprehensive computer modeling is needed. The South Florida Water Management Model is being utilized to simulate major hydrologic/hydraulic processes, including overland, canal, and groundwater flows, water control structure operations, and evapotranspiration. The model will help evaluate alternative plans using evaluation criteria that address benefits/impacts to Lake Okeechobee, the estuaries, agricultural and environmental water deliveries, and performance of stormwater treatment areas. The results of the hydrologic modeling will be applied to evaluation criteria and interpreted using Criterium Decision Plus, a multiple criteria decision support model. The paper will present a summary of alternative descriptions, evaluation criteria, and the approach used to compare alternatives and select the preferred plan.

In the past fifteen years, the numerical model, TETRAD, has become a standard numerical tool at Idaho National Engineering and Environmental Laboratory (INEEL). TETRAD is a multi-phase, multi-component numerical model used extensively in reservoir engineering and environmental modeling studies at the INEEL. Typical environmental studies take about 10-CPU weeks to run and there are approximately 120 studies at INEEL each year. So, the time involved to complete environmental studies is significant. The applications of TETRAD involve iterative solution to the conservation equations, with convergence realized when all equation residuals are less than some user-selected convergence tolerance. Because of the vast differences in mole fractions in many applications (order 10^–8 for tracers or pollutants), this can lead to numerical difficulties including round off in the solution of the larger mole fractions (e.g., water, air). We implemented and tested revisions to TETRAD that include species-dependent convergence criteria, such that all conservation equations are solved to similar accuracy. Various test cases are discussed, whose run times are much faster than the original version of TETRAD, with excellent comparison in species distribution.

The Environmental Modeling Research Laboratory (EMRL) at Brigham Young University in partnership with the U.S. Army Engineer Research and Development Center (ERDC) is currently developing a generic data format for multi-dimensional models. The goal of this exercise is to develop, promote, and deploy a common modeling format that facilitates data storage, exchange, access, analysis, and discovery of scientific and engineering data. The project encompasses one-, two-, and three-dimensional models including river cross-sections, scatter points, unstructured (finite element) grids, and structured grids. The objective of the project is to define a standard file format for all computational models developed at ERDC. The new model format is called XMDF (for Generic Model Data Format) and consists of a file format and an object code library (Application Programming Interface (API)). The API consists of a series of subroutines in both C/C++ and FORTRAN that can be used to read and write model geometry and data sets to the XMDF format. Model developers within ERDC will be encouraged to adopt the format for all existing and future models. Numerous benefits will be derived from the standardized model format including highly compact and efficient file i/o. Using a common format makes it possible to easily share data between models, link models, and gain access to powerful visualization tools.

This paper describes the concept, methodology, and application of a groundwater contamination monitoring plan using measurements at pumping wells. The monitoring strategy relies on the best accumulated engineering knowledge data and judgment of the contaminant concentrations on site, annual well-pumping locations and flows, and pumping radii of influence. The monitoring objective is to accurately and quickly locate the contaminant sources and plumes, subject to a maximum annual number of samples. The decision variables are the wells to be sampled at each sampling round. The methodology suggested is a heuristic adaptive algorithm, which is limited at this stage to a single organic compound. The methodology was tested on a synthetic computer-generated aquifer, and on real data of the Coastal Plain Aquifer in Israel. Comparisons made to previously developed monitoring algorithms indicated a much better convergence of the description of the contaminant distribution in the aquifer to its real image.

Growing interest in long-term groundwater monitoring stems from the significant challenges that stakeholders and regulators face while negotiating sampling strategies. This study presents a new optimization tool developed to aid this negotiation process by enabling decision makers to rapidly assess tradeoffs between their conflicting objectives (e.g., minimizing sampling costs and minimizing uncertainty). This paper demonstrates how the Nondominated Sorted Genetic Algorithm-II (NSGA-II) and ε-dominance archiving can be combined so that stakeholders can bias multiobjective search towards key tradeoffs. The archiving strategy for the NSGA-II serves three purposes: (1) it ensures that high quality solutions are not lost, (2) it allows stakeholders to bias search towards important design tradeoffs, and (3) it reduces the NSGA-II's computational demands. In this study, results are presented for a previously published groundwater monitoring case study in which ε-dominance archiving helped to reduce the NSGA-II's computational demands by more than 80-percent relative to prior results. The techniques presented in this study will aid practitioners in incorporating their preferences into multiobjective search while also reducing the computational demands associated with quantifying long-term monitoring tradeoffs.

Traditional multivariate statistics and geostatistics were used to analyze groundwater hydrochemistry (alkalinity, Mn, Fe, Si, Al, Mg, NH₄, Ca, K, Na, Cl, SO₄, H₂S, NO₂, NO₃, CH₄, pH, DO, EC, and DOC) and microbial data (16S rDNA DGGE community profiles of Bacteria and Archaea) previously collected from a landfill leachate contaminated aquifer. Variograms and cross-variograms developed for principal components formed from hydrochemistry and rDNA correlation matrices showed spatial correlation between 30m–55m. Kriged principal components indicate the combination of hydrochemistry and Bacteria data may provide better estimates for unknown locations, particularly along the boundary of the plume. The principal components between these two types of data appear to complement each other, as hydrochemistry PC1 appears to separate sample locations vertically and Bacteria PC1 separates sample locations horizontally. These combined data with a multivariate-geostatistical approach may be useful for delineating leachate contaminated zones at waste disposal sites and provide managers with better estimates of risk at unknown locations.

Groundwater remediation projects require long-term monitoring (LTM) to assess compliance of active remedial systems and post-closure sites where groundwater contamination is still present. LTM can be costly given the large number of sampling locations, frequency of monitoring, and number of constituents monitored at a given site. This work presents the development of a methodology to optimize a groundwater-monitoring network in order to maximize cost-effectiveness without compromising program and data quality. We propose method that combines ant colony optimization (ACO) with a genetic algorithm (GA). The ACO method is inspired by the fact that ants are able to find the shortest route between their nest and a food source. This is accomplished by using pheromone trails as a form of indirect communication. Ant colony simulation techniques are adapted to minimize the number of monitoring locations in the sampling network without significant loss of information.

This research develops a generalized parameterization method that couples zonation and geostatistics for characterization and identification of a stochastic field. The coupled zonation-geostatistical method honors distinct pointwise measurements in a field to create a zonation structure, a kriging field, or a mixed distribution. In this study, a Voronoi tessellation (VT) is adopted to create a zonation structure of Voronoi cells over a set of sample points. VT determines the boundary, shape, and distribution of zones completely and uniquely for a given set of sample points. Other than zonation approach, this study develops a natural neighbor kriging (NNK.) method that honors natural neighboring sample points for estimation. The natural neighbors of an estimation site are determined by VT. Specifically, we combine VT and NNK together as a coupled VT-NNK method by introducing a set of binary weighting coefficients to the sample points. The coupled VT-NNK method possesses greater flexibility to characterize the randomness of parameter heterogeneity and makes the best linear unbiased estimation (BLUE) over a set of binary weighting coefficients. In the numerical example, we identify transmissivity heterogeneity with coupled VT-NNK by seeking the optimal binary value of weighting coefficients such that the fitting residual of observed groundwater heads is minimized. The nonlinear binary integer minimization problem is accomplished by a genetic algorithm (GA) to obtain a near-global optimal solution. Results show that coupled VT-NNK. is able to capture the non-smoothness of true heterogeneity and gives small groundwater head fitting residual.

The Air Force spends millions of dollars annually on groundwater Long-Term Monitoring (LTM) networks at installations around the country. Because LTM is costly and its well networks so extensive, the Air Force is actively pursuing testing and implementation of optimization strategies for its LTM networks at a number of sites. One such strategy developed in coordination with AFCEE is a decision-logic statistical optimization scheme named the Geostatistical Temporal-Spatial algorithm or GTS. GTS uses known statistical and geostatistical techniques in a novel manner to answer two questions: given an existing LTM network, 1) what is the optimum number and placement of wells in that network (i.e., is there spatial redundancy and/or is there ‘undercoverage' within the spatial network)?, and 2) what is the optimal sampling frequency for wells in the network (i.e., is there temporal redundancy)? Optimized networks in typical applications of GTS have resulted in an estimated cost savings off the total LTM budget of over 30%. To measure spatial redundancy, GTS has heretofore applied kriging and robust spatial modeling techniques in an iterative fashion in order to identify optimal subsets of the existing monitoring network. Optimality is measured by balancing the costs of sampling, analyzing, and maintaining the ‘optimal' network against 1) deterioration in estimated site maps compared to the baseline, 2) increases in global uncertainty associated with map estimates, and 3) increases in localized areas of uncertainty. To maximize the algorithm's flexibility, numerous changes and improvements to GTS have been tested. Some of these changes have been applied at LTM networks on three different Air Force bases in California, New Hampshire, and Maine. More specifically, comparisons have been made between the iterative mapping approach — where increasingly optimal well location subsets are identified via changes in the global estimation weights — and a newer approach based on the use of genetic algorithms. The specific genetic algorithms developed for GTS are designed to perform the overall spatial optimization more efficiently and in fewer analysis steps.

Ground water is an important natural resource that provides drinking water to more than half the population in the United States. Unfortunately, this important natural resource is vulnerable to variety of contaminants. Some of these contaminants are nitrates, pesticides and heavy metals. Heavy metals like arsenic, mercury, lead, cadmium and chromium contaminate ground water from natural and anthropogenic sources. Few studies have been done to characterize heavy metal contamination on a regional scale. A new ordinal logistic regression (LR) model was used to predict the probability of heavy metal contamination with respect to two thresholds- natural background concentration and maximum contaminant level (MCL). Sumas-Blaine Aquifer, located in Washington State, was used as the study area for the model. The model considered variables representing soil, land uses, permeability, depth, clay content and surface elevation. Ordinal LR model was applied to relate the probability of contamination by individual heavy metals to the independent variables. The model showed a good fit. The model results are in agreement with observed data. Spatial analysis was performed using GIS technology to better understand the spatial distribution of probabilities of heavy metal contamination with respect to thresholds. Ordinal LR model was applied to predict the probability of heavy metal contamination, in general. The model results are close to the observed probabilities (r² = 0.90). Estimating the probabilities of occurrence of concentration with respect to background concentration and MCL for heavy metals individually and in general, help in managerial decisions as well as understanding the geochemistry of heavy metals.

The River Indus system comprises of seven rivers. Due to upstream storage and diversion activities in the upper basin, there has been a steady decline of the water outflow to the lower basin, the Indus delta, and to the Arabian Sea. This has created a huge water shortage problem in the lower basin, resulting in devastating economic, social, and ecological losses. This paper evaluates some of these impacts and offers some recommendations.

Integrated environmental and economic planning that considers both regional economic development and the environmental impacts of such development has rarely been used by governments to plan regional growth largely because of the absence of a methodology that would allow future environmental impacts of proposed development plans to be quantified in monetary terms and used by Economists (E1) in their overall cost-benefit analysis of projects/sectors targeted for investment. This paper discusses a new approach for achieving integrated economic-cum-environmental planning (IEEDP). This paper involves use of "unit load curves" (ULCs) for major pollutants, such as BOD, SS, and Toxics, expressed in terms of pollutant load (tons)/unit Gross Regional Product (GRP - $) to determine current and projected loads of pollutants and their cost to the economy. ULCs are also used to determine the costs of all other known environmental impairments including reduction in dry season river flows due to deforestation, peak flow changes and flooding, additional health costs, river water quality deterioration costs, and lost fisheries, These are considered to be the main threats to sustainable development in Northeast Thailand which was used for a case study. The development of the IEEDP method is based on work undertaken as part of an Asian Development Bank (ADB) project in the Mekong River Basin. The results show that conventional or "business-asusual" planning for Northeast Thailand would cost some 28.1 billion baht between 2000–2030 in water related damages due to economic growth. An integrated economic-cum-environmental plan is developed to assess and cost environmental damages (EDs) of conventional development planning to prepare and cost a correction program for these damages, and to modify the investment plan to include only projects will favourable cost/benefit resulting in a sustainable development plan both economically and environmentally. The results show that IEEDP would lower water related damages in Northeast Thailand to some 12.6 billion baht between 2000–2030. The investment in the correction program would have an approximate not present value of 6.8 billion baht with an IRR of 104%.

During the 1960s the Tuy river supplied 4 cubic meters per second to the water supply system of Caracas, Venezuela. Forty years later, the river is so highly polluted that water is only extracted during emergencies (droughts mostly) and other rivers have been tapped to replace the Tuy and to balance demand from a growing population. The paper resumes years of data collection for the basin, its main polluters and degrees of pollution along the river. It analyzes costs of recovery and the benefits to be obtained, specially from the reuse of the river as a water supply source. Costs are compared with the building of yet another dam in a river about 200 miles from the city and at a lower elevation, increasing pumping costs. The paper demonstrates that the renewal of 4 cubic meters per second of potable water to the city justifies , from the economic viewpoint the restoration of the upstream basin from the intake, and delays in time the building of the new dam.

Development and impediments to integrated water resource management in Chile are presented. Sustainable water use supports social objectives into the indefinite future without undermining environmental and hydrological integrity. International consensus exists that integrated water management and watershed management form a sound basis for sustainable water use. Increased water user conflicts, water quality degradation, recognition of economic inefficiencies, acknowledgment of ecological flows requirements, and public good issues are leading toward a more integrated and watershed based approach to water management in Chile. Despite these issues and concerted effort since the early 1990s, however, integrated water resource management and watershed management has not yet materialized and progress remains slow. Impediments in Chile to integrated water resource management include opposition due to political dogma and special interest groups, as well as lack of stakeholder participation and enabling legislation.

Some conflicts on water resources management in Indonesia are escalating nowadays - conflicts between privatization and heavy state intervention in managing water resources among NGOs and also among government institutions such as conflicts between State Ministry of Environment and Ministry of Settlements and Regional Infrastructure. Whether the best way in managing water resources is by inviting private sector or through heavy state intervention, the fact is that the victim of the two conflicting ways in managing water resources is community-based water resources management. Actually in urban areas such as in Semarang, the capital of Central Java Province in Indonesia, there are many community-based management in providing safe water for a community in certain area. Sinar Waluyo Estate, Tanah Mas Estate, Bukit Semarang Baru Estate, and Jangli Permai housing area are some of them. Sinar Waluyo Estate and Tanah Mas Real Estate for example have supplied safe water for their residents for more than twenty years. Because of the bad services of Semarang City Regional Public Water Utility Company, the residents of Jangli Permai and the developer of Bukit Semarang Baru have also managed to provide safe water by themselves. As one of the alternatives in managing water resources, specifically in providing safe water for household activities, it is interesting to do a thorough research on how they manage safe water by themselves: how they practice participatory management for the urban safe water, the efficiency that is the attempt to maximize the economic and social welfare, the equity in the allocation of scarce water resources and services to different economic and social groups, and the environmental sustainability of the three actors in water management, the government as regulator, the management of urban safe water as policy maker, and the behavior of the community. Many people forget that community-based management in providing safe water has longer history than both private and heavy state intervention. The result of the research could give picture of community-based urban safe water management model in developing countries.

An overview of indigenous rights and modern water management in Chile are presented. Fundamental cultural and spiritual differences exist between the dominant and indigenous cultures. The prevailing European or Western societal view of natural resource management embodied in current legislation is of ownership and use. Indigenous communities in Chile traditionally viewed themselves as part of nature and often treated water as a communal good. Historic interaction of indigenous communities and European immigrants in Chile followed a similar world pattern: spread of disease, armed conflict, subjugation to the dominant culture, and usurpation of indigenous land and water rights. Many indigenous communities were further marginalized during the military regime (1973–90), perhaps due to their communal society being associated with communism. Increased water conflict has occurred during the last 30 years due to socio-economic government policies and rapid economic growth that has stressed the resource. Insufficient protection of indigenous rights and lack of targeted policies to enable indigenous communities to sustain themselves have occurred. Important national steps have been taken recently to redress and reconcile a multi-ethnic society. It remains to be seen whether this process continues and produces tangible results.

In order to mitigate water crisis in northern and north-western China, a huge project of South-to-North Water Transfer is going to be built, which will divert water from upper, middle, and lower reaches of Changjiang River to the north part and northwest parts of the country. The project consists of 3 water diversion route projects, i.e. Western Route Project (WRP), Middle Route Project (MRP) and Eastern Route Project (ERP). The 3 routes will come across with 4 major rivers, i.e., Changjiang River, Huaihe River, Huanghe River and Haihe River. They will jointly form a general frame called as "Four-plus-Three". By the target year of 2050, the total water volume to be diverted will be 45 billion m³. The whole project will be executed in 3 stages depending on real situation of the project development.

Annual average precipitation in Taiwan is 2,500 mm which is about three times the world average. However, because of its high population density (23 million), the average annual precipitation per capita is 4,500 m³ which is only sixth of the world average. Due to other natural factors, such as uneven temporal and spatial distribution of precipitation and steep slope of rivers, which make water resources management in Taiwan more difficult. Taiwan also is facing other problems, such as people's opposition of building new large dams. This paper reports the stages of water resources development in Taiwan, discusses the important issues of water resources management and suggests some solutions.

China is lack of water resources, especially in its arid regions. So the management of water resources in these areas is very important. Remote sensing technology has its special advantage in this aspect. The paper introduces the application of remote sensing, including GIS, in this field, such as surface water resources investigation, groundwater exploration, dynamic monitoring of ecology, salinisation, water environment and deserfication, drought monitoring, planing of water diversion projects and so on. It shows that remote sensing technology can play important role for economy development in West China, especially in Northwest China.

The Yellow River Basin is characterized with ‘different sources of water and sediment, less water and much sediment, mismatching of flow and sediment and uneven distribution of precipitation both in a year and in years'. With the development of economy and society, the contradiction between water supply and water requirement in Yellow River basin has become sharp increasingly and water resources conflicts among different water users occur frequently. Especially in 1990s, the problem of zero-flow in its lower reaches became seriously year after year. This paper introduced the basin water features, analysed the present situation of water utilization and water problems in the Yellow River basin. Then, as for as existing problems presently of water shortage and water pollution, the authors analysed the importance of developing water right management, studies the legal base and problems to implement water right management. At last, the basic thinking of water right management was proposed.

The authors analyzed the situation of water resources quality in China. Results showed that the major rivers are generally confronted with organic pollution and non-point source pollution, which is getting serious. Dianchi Lake, Taihu Lake and Chaohu Lake have seen a rising trend of eutrophication. The ground water in most cities has suffered with point-sources and non-point sources pollution and the contents of some elements in some places exceeded the limitation. A summary of the progress of studies was made mainly in 4 aspects: water purification and treatment technology; water resources quality and safety; water resources quality planning and assessment; and water resources quality simulation and calculation.

The north part of China is the most serious area of water shortage in China. The total amount of water use was increased from 243.7 billion m³ in 1994 to 254.2 billion m³ in 2002. The annual increase rate of water use was much lower than the growth rate of the population. The water use per capita was also decreased. The over-development of water resources has caused severe eco-environmental problems so it is very difficult to further augment the capability of water supply. Therefore optimizing the distribution of water resources, improving the efficiency of water use, enhancing the saving of water use as well as protection and rehabilitation of eco-environment have become the key factors for the sustainable development and utilization of water resources in the northern China.

The compensation for water resources restoration aims at promoting the natural water system. The authors analyzed water resources decline problems and discussed the compensation mechanism for water resources restoration, for the purpose of solving problems, such as water shortage, water environment degeneration, and ecology destroy. The connotation of water restoration and compensation was discussed. Four types of compensation and two scopes of compensation as well as between generations were also discussed.

Stepped spillway is one of the idea forms to increase oxygen content in water body for hydraulic and environmental engineering. The effect factors on aeration rate in the water flow is closely related with the height, slope of step and water depth of the flow over stepped spillway. The relation between aeration rate and the effect factors mentioned above is developed. The aeration rates of 35 working conditions in the water flows over stepped spillways are simulated using numerical model. The simulated results are compared with experimental data by Chanson. It has shown that the simulated results are fairly coincident with experimental data. The study provides solid theoretical fundament on optimum design of spillways to increase oxygen content in water body.

Tseng-Wen Reservoir is the largest reservoir in Taiwan. It is also the most important water resource of Taiwan's largest plain - Chianan Plain. Due to the development of recreation and agriculture applications in the drainage area, the water quality was achieving a eutrophic state. The concentration average of the total nitrogen (TN) was 0.92 mg/L. The concentration average of the total phosphorus (TP) was 40 (μ/L, and the concentration average of chlorophyll-a was 5.84 μg/L. The nutrient released to the reservoir primarily came from non-point pollution sources. Regardless of the forecast which occupied 77% of the drainage area, agriculture lands contributed 16.4% of the TN and 11.3% of the TP to this reservoir. Fishing also contributed 7.6% of the TP. Therefore, regulating inappropriate forest applications, promoting best management practices (BMP) in agricultures, and forbidding the fishing activities are recommended to improve the water quality in this reservoir.

Considering the actuality and problem faced the Flood Control Decision Support System, an integrated design about a flood control system based on middleware technology is introduced in this paper to realize the reuse of the existing results. A simple instance based on the design is also implemented so as to advance a new thought on improving software reusability and on accelerating the industrialization of software produce.

Based on the theory of regionalized variables, the spatial variability of water level in river network of the Pearl River Delta (PRD) was analyzed. The semi-variagram revealing the spatial structure of water level in river network of the PRD was given. The directionality and the range of correlation for water level in the river net were worked out. With the analysis of spatial distribution of water levels, an example for optimized interpolation for water levels in river net of the PRD with ordinary kriging was given. Comparison between kriged and measured stages at 15 stations in the PRD river net showed that the maximum relative error is less than 1%.

Tarim River is the biggest inland river in China. Its problem of eco-environment is worsening in the lower reach. For keeping this trend within limits, emergent water diversion to the lower reach was carried out. In this paper, the remote sensing technology was applied to the analysis of eco-environment effect after water diversion. The result show that the vegetation index and cover ratio increased but not markedly, the eco-environment situation can not been improved obviously up to now. Its effectiveness was temporary. The continuity, quality and quality of water source for the Tarim River must be ensured.

The political boundaries between the Palestinian and Israelis make the water issues critical and sensitive, Groundwater is the primary source for the Palestinian in the West Bank. Although, there are many studies on the shared surface resources (Jordan River Basin), there are few studies on the groundwater shared resources between the Palestinians and Israelis. There are three primary groundwater basins underlying the West Bank (Eastern, Northeastern and Western Basins) as shown in Figure 1, Both the Northeastern and Western basins are shared between Israelis and Palestinians. The Palestinians have a limited access to the Northeastern Basin and strictly limited access to the Western Basin. In addition to the quantity of available water resources, the quality of water is emerging as a critical issue. Threats to ground water quality include disposal of untreated wastewater, increasing salinity due to agricultural activities and intrusion of native groundwater of poor quality. Widespread use of herbicides and pesticides also represent a threat to drinking water supplies. The Declaration of Principles signed in Washington D.C in 1993 was a major step toward resolution of the political conflict between the Israelis and Palestinians. However, the water issue is part of the final status negotiations, which still unresolved. As resolution to political conflict is pursued, it is clear that water resources management issues remain at the forefront because of the transboundary nature of the hydrologic regime. Cooperative management on the technical level appears to be the only alternative to further conflict and degradation of the region's scare water resources. This paper will study the impact of the transboundary resources on both sides and explore some of the most significant groundwater management issues facing both the Palestinians and Israelis.

This paper examines the history that lead to the present-day allocation of the waters that are jointly claimed by Palestinians and Israelis. It describes the status of the water sector in Israel at the time the Oslo II Peace Accords were signed, and contrasts these Israeli conditions with the development of water resources in the Occupied Palestinian Territories under the auspices of the Oslo II accords. The paper reports on the impacts that have been sustained by the Palestinian water infrastructure during the current Intifada, and speculates on the conditions that will have to prevail before Palestinian-Israeli water issues can be resolved. The paper is based in part on personal experiences of its authors in working in the Palestinian water sector since 1996. It documents a sampling of problematical situations and conditions related to water management in the region that are not frequently reported in the western media, that do not command the attention of academic researchers, but that provide a glimpse into reasons why the joint management of the waters shared by Palestinians and Israelis was not possible under the framework of Oslo II. In particular, conditions "on the ground" that are peculiar to the water sector and that continue to prevent an integrated approach to the management of transboundary water in the Occupied Palestinian Territories are discussed.

The Jordan River is a relatively small watercourse with many historic, cultural, hydrologic, ecologic and political significance. It is shared by five independent countries, Lebanon, Syria, Jordan, Israel and Palestine. During the last thirty years, most of the Jordan River was diverted for different uses which leave less than 10% of the long-term average flow into the Dead Sea. As a result, the Dead Sea surface area reduced from 1,700 km² to less than 500 km² and the level dropped more than 25 meters. To save the Dead Sea from disappearing and to produce more water available, a canal among many other solutions has been proposed to transfer water to Dead Sea from adjacent water bodies using the difference in elevation between the Dead Sea and Red Sea or the Mediterranean. This paper discusses the concept of intersea transfer and the feasibility of this approach.

Palestine is among the countries with the scarcest renewable water resources due to both natural and artificial constraints, amounting to only 100 cubic meters per capita per annum. This amount is far below the available water in other countries in the Middle East and the World. At present, the water demand exceeds the available water supply. The available supply is constrained mainly due to artificial non-economic factors. The gap between the water supply and water demand is growing due to the population growth, improving standard of living standard, and the need to expand irrigated agriculture and industrialization. This paper will give an estimate of the projected Palestinian water demand in light of the proposed national targets and will assess all potential water sources both conventional and non-conventional.

The Zaravshan River Basin is one of the most densely populated areas in Central Asia. The length of Zaravshan River is 877 km, the watershed area is 17.7 thousand km², and the population is more than 5 million. The current social and economic situation in this region requires a significant increase of water supply to communal and household needs, developing industry, irrigated agriculture and other fields of the national economy. Although the Zaravshan River is the sole source of potable water supply, it receives communal, household, industrial and agricultural sewage in Leninabad province of Tajikistan, and the Samarkand and Navoi provinces of Uzbekistan. All along, the river is characterized by intensive pollution by agricultural, stock-breeding, industrial and sewage waste flow. Consequently, it is one of the most polluted rivers in Central Asia. Impoverishment of water resources in this area has been caused by significant consumption of water for irrigation and drainage of agricultural lands. Drainage water brings herbicides, pesticides and defoliants. Salinity of water increases 6–8 times from the river head to the river outlet - from 0.27–0.30 g/l to 2–2.4 g/l. Particularly in its lower part it becomes unusable for drinking purposes due to chemical contamination. Since there are no water and sewage systems in rural areas, population of the region receives water that does not meet any health related requirements. At the same time, a tense situation with water utilization and ecology is developing in the region which no doubt manifests itself in the health situation of the region. The situation is also worsening because a number of residents in rural areas use water from open reservoirs and canals for household needs and drinking. According to the report of the Central hydro-meteorology service of Uzbekistan, the quality of water in the Zaravshan River is of class IV by the Surface Water Pollution Index, which means polluted and heavily polluted water. The most contaminated water is in the area below the Navoi city. The maximum concentration of pollutants, which are the major indicators of contamination of water reservoirs with organic and biogenic substances in this area is: phenol - 0.005 mg/L (5 times higher than the maximum allowable concentration (MAC)), petroleum products - 0.64 mg/L (12.8 MAC), nitrates - 0.202 mg/L (10.1 MAC). Such concentrations of these substances are mostly due to the existing anthropogenic pressures on the water resource.

Current Egyptian regulations for discharging oily water from production operations to the sea allow a maximum of 15 ppm oil-in-water. International standards call for a maximum of 40 ppm, which will eventually decrease to 25 ppm by 2006. The rationale behind setting up the limit to a maximum of 15 ppm has not been evaluated previously. The intention of this research was to verify the need for such a highly restrictive limit. The study has been confined to the Gulf of Suez, being a major oil exploratory zone as well as a worldwide tourist attraction. Its rich marine life ranks it as one of the most sensitive environmental habitats in the world. Findings were based on their direct impact upon Gulf of Suez oil and gas production operations and categorized as ‘political', ‘technical' and ‘commercial'. Although extensive research in this field has taken place, there is no globally recognized set of standards to govern the process of produced water disposal. Environmentally sensitive areas within the Red Sea have been least affected by oil and gas produced water discharges. The main threat to the ecology of the Red Sea is oil lost by ships, from construction and dredging, and from wastes produced by activities in the coastal zone. Moreover, emphasis on the overall monthly or yearly loadings rather than daily concentration of oil-in-water discharges is considered the best approach in managing discharges to a marine body. Fate and transport modeling confirmed that adopting a discharge requirement of 25 ppm (maximum) will have negligible impact on environmentally sensitive areas in the Gulf of Suez. Although a sizable fraction of the oil-to-sea will be deposited in the sediments, a considerable portion will exit the ecosystem altogether, with a resulting negligible effect on water quality. From a commercial perspective, treatment of produced water to 15 ppm or lower would cost operators between 5 to 6 times over the normal industry practice for 40 ppm, rendering any associated environmental gains unjustifiable. In summary, relaxing the discharge criteria to 25 ppm (which is the current global industry aspiration) for new fields will have a negligible impact on the environmental conditions and almost no negative impact on the biological or marine life in the Gulf of Suez, while at the same time promoting conditions for sustainable development in Egypt.

Large gaps exist in the understanding of how P forms, in what quantities, to what depths, with what mobility and availability in the Brahmaputra river basin. Almost 90% of the suspended load in the Brahmaputra is carried during monsoon as medium and fine sands. Concentrations of suspended sediments were found to vary from 500mg/l to 1800mg/l over a midstream stretch of 700 km. Suspended sediments ranged in size up to 600 μm. 70–80% of the monsoon load was coarser than 40 μm. The total P in suspended load varied between 950 μg/g to 1330 μg/g with net increase in down-river concentration. Sequential extraction revealed about 80% P to be extractable. Grain size differentiation in sediments, particularly in case of detrital-P was a major cause of decrease of organic-P and total-P after deposition. Apatite-P formed significant part of sediment associated P during high flow except at locations where agricultural, industrial and municipal loads were large. Demographic and socio-economic factors indicated that the factors driving environmental changes in the Brahmaputra watershed are likely to continue, and problems for sustainable management of key nutrients like P would also prevail. Considering that the Brahmaputra watershed is only next to the Yellow river in terms of average sediment yield, this would subsequently contribute significantly to the modification of the global cycles of P. The total flux of P from the Brahmaputra (8.4 x 10⁴ tons/year) constituted nearly 5% of the global flux and was about three times greater than that carried in dissolved form. However, low P bioavailability in the suspended load suggested that curtailing sediment input to the Brahmaputra might not be critical compared to other rivers, as minimizing sediment input to the river would produce little reduction in bioavailability of P.

After reduction of the external phosphorus load by phosphorus elimination plants, Lake Tegel and Schlachtensee in Berlin underwent a significant trophic improvement. The phosphorus elimination plants work by precipitation/coagulation/flocculation — sedimentation — post precipitation — filtration. The external load was reduced by one to two orders of magnitude down to 10 – 20 μg P L^–1. The inlake phosphorus concentration followed. Below a threshold value of about 50 μg P L^–1 the development of algae and cyanobacteria was clearly phosphorus-limited. In Lake Tegel, the external load reduction was counteracted by internal load, related to the nitrate concentrations above the sediment. A current R&D project aims at determining patterns of the oligotrophication process, including thresholds for reactions, using mass balance calculations and numerical models.

On January 1, 1994, Mexico, Canada, and the United States implemented the North American Free Trade Agreement (NAFTA). As part of that process, the three nations also began to operate under NAFTA's Environmental Side Agreements. Ten years have elapsed and it is worthwhile to examine issues associated with the decade's efforts to improve environmental conditions in North America and enhance environmental health along the Mexico-U.S. border. This paper is an initial product of the Environmental and Water Resources Institute's (EWRI) Task Committee on Trans-boundary Environmental Management in Canada, Mexico and USA after NAFTA (NAFTA Task Committee), which was set up in 2002 to monitor developments under NAFTA. It is beyond the scope of this paper to address the environmental consequences under NAFTA. This paper does describe the activities in three trans-boundary environmental institutions established under the NAFTA environmental side agreements: the North American Commission on Environmental Cooperation (CEC), the Border Environmental Cooperation Commission (BECC), and the North American Development Bank (NADBank). These comments are based in part upon visits by the co-authors to Washington, D.C. to discuss NAFTA with officials of the U.S. State Department and the U.S. Environmental Protection Agency in February 2003 and their participation as EWRI representatives at the Tenth CEC Joint Public Advisory Committee (JPAC) Meeting in Washington DC, held June 23 through 25, 2003.

Irrigation has been considered the hope for economic recovery in many agriculturally-based developing countries. However, irrigation has not satisfied the economic and social needs of these countries. Why hasn't irrigation, despite many development projects and its status as a priority in these societies, been able to significantly improve living conditions and decrease poverty? There are external reasons, internal reasons, and the element of human capital involved in each. In order to make irrigation a true key to development and reduction of poverty in the Third World, an understanding of underlying factors, with corresponding substantial changes, is necessary. An integrated approach to the situation should address both internal and external problems that impede the benefits that irrigation should bring. External factors are related to the market. Are developed nations really sincere in their desire to help the developing world? Two suggestions—reduction of tariffs and reduction of farm subsidies in developed nations seem to be the main avenues to open markets to farmers in the developing world. In fact, these possible solutions have been discussed in WTO meetings in Doha and were on the decision-making agenda for the WTO meeting in Cancün, where agricultural trade was discussed. Indications were that globalization would be implemented as of 2005. Globalization will come as an economic shock to the developing world. However, the next question is: Is the developing world ready and flexible to meet the demands of and rapid response required by globalization, when and if it happens? The answer is that it generally is not. In this article, strategies to prepare the agricultural sector in the developing world for this coming shock wave are discussed. Structural and nonstructural alternatives are prioritized.

There is an urgent need to improve the quality of life for the 2.5 – 3 billion people now living in poverty. More than 1.1 billion are undernourished. Many are water poor and/or energy poor. It is urgent that efforts be increased to improve the knowledge of renewable, sustainable, and environmentally friendly resources that can be used to provide a better quality of life for the present population and for the probable doubling of the population within this century. Water is available as precipitation, surface water, and groundwater. Knowledge of the adequacy of these sources of water is based on estimates. These estimates are based on available data that is usually incomplete, fragmentary, or of poor quality. The international agencies should make a strong effort to improve the quantity and quality of the available information on how much, when, and where water is available. Some estimates of the availability of water are included in this paper. A moisture adequacy index (MAI) and some of its uses are described. The World Water and Climate Atlas and its use in determining water availability are described. The World Commission on Dams and the United Nations (UNESCO) have provided guidelines on how to meet the needs for increased electrical energy and for water used for other purposes. A small increase in the use of electrical energy can produce a large reduction in poverty, reduce malnutrition, improve health, and decrease population growth. This paper strongly supports the recommendations of the World Commission on Dams and The United Nations (UNESCO). The world wide potential for green energy is enormous. Laws that decrease the use of non-renewable energy and encourage the use of renewable resource energy should be strengthened. A brief overview of some of the legal issues is included.

The Njoro Watershed, typical of the semi-arid basins in the Rift Valley of Kenya, is undergoing a new phase of rapid land use change in the uplands portion of the watershed, and on-going significant growth in both rural and urban populations. Considerable negative environmental impacts are occurring, in particular to the quantity and quality of river water. Domestic, livestock, commercial, industrial, and institutional water users in the basin are affected, as well as important downstream habitat in Lake Nakuru, a large shallow saline lake designated a Ramsar wetlands site of international importance. In response to these concerns, ajoint US-Kenyan multidisciplinary applied research project called Sustainable Management of Watersheds Collaborative Research Support Program (SUMAWA-CRSP) was established. SUMAWA aims, over the next 3–5 years, to demonstrate improved and integrated management of water and environmental resources in the Njoro Watershed through local stakeholder participation and action supported by scientific information and analyses. Preliminary results from elements of the SUMAWA planned process for stakeholder participation in management of the Njoro Watershed are presented. A mixture of participatory methods, discussion forums, awareness-raising activities, and local capacity building are being undertaken to address some of the challenges involved in engaging local stakeholders and communities in watershed action planning. The first phase of activities adapts Participatory Rural Appraisal methods to assess local residents' interests in and perceptions of the current condition of river water quality and quantity, of problems and their causes, and opportunities for local action in six different communities along the length of the watershed.

We describe the application of an integrated economic-agronomic-hydrologic river basin policy simulation model, intended to assist water managers in designing and evaluating strategies for improving the physical and economic productivity of water. The model is specified for the Brantas Basin in East Java, Indonesia, a densely populated catchment of 12,000 km² containing extensive irrigated agriculture. The Brantas is seasonally water constrained, and municipal and industrial users compete with irrigated agriculture, and with environmental quality demands, for increasingly scarce water. As options for supply augmentation in the Brantas are limited, water managers will depend increasingly on demand management strategies, particularly those effective in limiting irrigation abstraction. We describe the results of three simulation scenarios, including the introduction of volumetric water charges, the reduction of paddy and sugarcane price supports, and the introduction of formal water use rights in combination with market instruments. The effectiveness of these policies is evaluated on the basis of two criteria: the extent to which net irrigation withdrawals are reduced, particularly during the dry season, and the financial impact on the irrigation sector. We discuss key policy findings based on the results of these scenarios.

In order to advance water resources research and practice that addresses sustainability concerns, this paper identifies and explores the use of a few metrics for water use sustainability. Examples include the ratio of water withdrawal to total supply, the percentage of income spent on water and sanitation, the incidence of waterborne disease, and indices related to a managed system's ability to cope with extreme events. The spatial and temporal, scales over which these metrics can be calculated using available data are assessed, and additional data requirements are identified. Some of the challenges faced in predicting future values of these sustainability metrics are also discussed, along with research needs to improve predictability and adaptability.

In recent years, water resources planning and management activities have expanded from narrowly focusing on flood defense, water quantity improvement and quality control to the better integration of ecosystem needs in management protocols. Because ecosystem needs are still subject to discussion, a reasonable surrogate is the use of the natural flow regime. The natural flow regime integrates ecological, geomorphic, and hydrologic functions in a watershed and provides a flow-based approach to protecting ecosystems (ecohydrology). This paper identifies a suite of ecohydrologic indicators, called the Taiwan Ecohydrology Indicator System (TEIS). This system of indicators considers seasonal change, typhoons, and ecological requirements of resident species in hydrologic indicator selection. The final set of 60 TEIS hydrologic indicators captures the magnitude, frequency, and duration characteristics of natural flow regimes. To facilitate use in policy analysis, a reduced set of indicators was selected. This set includes six parameters that can be used to minimize the complexity of optimization procedures and still reflect ecological considerations in decision making and planning.

Coupled atmosphere-ocean general circulation models (GCMs) are widely used to study the current climate system as well as its possible changes in the future. The precipitation output from these models has been compared to observations in order to verify the reliability and accuracy of GCMs in simulating the precipitation process. Depending on the specific interest and emphasis, different yearly, seasonal, and extreme condition statistics derived from modeled and observed precipitation time series have been compared. So far, however, there is no report of comparison between modeled and observed rainfall events for any specific locations. In this paper, rainfall events as observed at a individual site are defined first. Event-related statistics are then calculated using modeled and observed precipitation time series. Comparison of these event-based statistics shows that while GCMs cannot model the rainfall process at the two selected locations on a year-by-year, or event-by-event basis, they do model fairly well the long-term average number of rainfall events and event volume distributions. Results presented in this paper may render more confidence in the use of certain information derived from GCM precipitation output, and at the same time assist in identifying directions for further improvement in the modeling of precipitation processes.

A study of the effect of Houston on rainfall patterns at multiple spatial and temporal scales is being performed. Rainfall data from a variety of sources, from the precipitation radar on the TRMM satellite to rain gauges in the Houston Flood Alert System, have been collected and are being analyzed. The combined rainfall data set covers a range of spatial scales, from tens of kilometers to thousands of kilometers, and temporal scales, from minutes to months. The insight gained from studying data sources with this range of scales is unique for analysis of rainfall modification in Houston. This paper summarizes previously completed Houston rainfall data analyses, introduces new results from the long-term trend analysis of Houston-area rain gauge records, and presents a sensitivity study of drainage design that indicates the potential impact of urban-induced rainfall modification on drainage system components for Houston. The results from the trend analysis suggest a weak increasing trend of warm season rainfall totals in the urban area and urban-affected area, while no trend was noted for an upwind control region. The results from the drainage design sensitivity analysis indicated that changes in design rainfall intensity similar in magnitude to changes attributable to urban-induced rainfall modification would be sufficient to modify some design sizes of pipes in simple drainage systems.

In this paper, a method for seasonal precipitation prediction is presented, which consists of three steps. In the first step, the effects of large scale climate signals on the low and high precipitation spells in the Southwest of Iran are investigated. The long-term records of Southern Oscillation Index (SOI) and North Atlantic Oscillation (NAO) as well as variations in Sea Surface Temperature (SST), Precipitation, and Sea Level Pressure (SLP) in the Middle-East and Europe are investigated. In order to quantify the effects of climate signals on the wet and dry spells, specific thresholds are defined for very high, high, normal, low, and very low precipitation periods jointly based on long-term records of average areal precipitation and cumulative streamflows in Winter and Spring. The variations of average SLP and SST during the period of June through November are considered in a number of characteristic locations in Persian Golf, Red Sea, Indian Ocean at north of Australia, Arabian Sea, Mediterranean Sea, Black Sea, Caspian Sea, and some parts of Atlantic Ocean. In the second step, the fuzzy regions and the membership functions for the SLP gradient in the selected locations are developed. Finally, the fuzzy rules are developed to incorporate the SLP variations in precipitation prediction. The fuzzy rules are used to predict the range of total precipitation in the period of 1990 to 2002. The results have shown that the actual precipitation have been within the forecasted range of precipitation in 68 percent of the years and therefore, the developed rules can effectively be used in predicting high, normal, and low ranges of precipitation in winter and spring in the Karoon River basin.

Tampa Bay Water, Florida's largest wholesale water supplier, is required to cut back groundwater pumping from 158 million gallons per day (mgd) to 121 mgd in 2003, with further reductions to 90 mgd by 2008. To meet the new groundwater pumping limits it was necessary for Tampa Bay Water to develop alternative supply sources including surface waters. Tampa Bay Water utilizes an integrated hydrologic simulation model and an optimization model to manage pumpage from groundwater and surface water facilities. Weekly streamflow forecasting is required for estimating treated surface water availability for input into the optimization model. This paper describes the development of a suite of computer modeling tools that forecast daily flows and levels for a one-week period at key locations along the Hillsborough River and Tampa Bypass Canal system (HRTBC).

This paper examines flood management for an urbanizing floodplain with increasing flood risk due to climate warming. The lower American River floodplain in the Sacramento, California metropolitan area, is used as a preliminary example. A stochastic dynamic programming model suggests economically promising adaptations to flooding problems given simultaneous changes in climate and urban land values. Economic-engineering optimizations were done for several climate warming and urbanization scenarios. Methodological and policy conclusions are drawn, based on the results, for floodplain planning considering interaction of climate, costs, and regional economic growth.

Procedures are described for assembling ensemble streamflow forecasts into a tree structure for use in a multi-stage stochastic linear programming model. Such an optimization model is the basis for a decision support system that has been developed for the Lower Colorado River Authority in Central Texas. In prior work, streamflow ensembles suitable for direct use in a two-stage model have been generated, and procedures were derived for conditioning the ensemble members on observable climate indicators. Additional work is needed to develop scenario trees that properly represent forecast uncertainty in a multi-stage model. Verification methods for scenario trees are also discussed.

Operational climate forecasts for 3-month mean temperature are issued monthly by the NOAA Climate Prediction Center, for lead times from 0.5 to 12.5 months. Among these forecasts, the Probability of Exceedance forecasts present information on expected shifts in the probability distribution of mean temperature relative to climatological distributions. The forecasts appear to support a wide range of applications in water resource management and agriculture, especially in regions where growing season length, snow pack, and lake or river ice processes are important. These temperature forecasts have been reported to have modest skill when skill is assessed across the entire United States. However, the adoption and use of the forecasts has been limited, partly due to a lack of user-oriented measures of forecast utility on a regional basis. In this presentation, we will examine forecast performance over the years 1997 – 2002, for all forecast divisions in the contiguous U.S. at the shortest lead time (0.5 months). Performance is assessed using two measures of forecast utility: usefulness (degree and frequency of forecast departures from climatological distributions), and dependability (correspondence between the direction of forecast departures and actual occurrences). A similar analysis on seasonal precipitation forecasts showed large variations in utility with region, ENSO state, and season. This analysis shows that the mean temperature forecast performance does not vary as much with region and ESNO state. Forecasts for warmer than average conditions have been very dependable for much of the contiguous United States, especially in the northern Great Plains and Rocky Mountains. This success appears to be the result of persistently warmer than average conditions during this period. There were relatively few forecasts offered predicting cooler than average conditions, and most of those were not dependable.

In this study, the magnitude of streamflow response to a precipitation forecast is examined to establish the conditions under which a probabilistic precipitation forecast, if skillful, is likely to translate into useful streamflow information. The hydrologic model SWAT was used to simulate streamflow response to several hypothetical precipitation forecasts and antecedent hydrologic conditions. The test watershed was a 32.9 km² subwatershed in the Little Washita River watershed in central Oklahoma. The streamflow characteristics were expressed in terms of runoff volume, change in runoff volume, odds of exceeding a specified threshold volume, and probability of exceedance curves of volume. While the numerical findings of this study are specific to the climatic and physiographic conditions of the watershed, the findings suggest that combinations of antecedent hydrologic and precipitation forecast conditions produce a broad range of streamflow responses that are suitable for decision making. It was also found that strong wet/dry antecedent hydrologic conditions can dominate over the impact of typically modest precipitation forecasts. In addition, dry antecedent hydrologic conditions tend to limit the utility of wet/dry forecasts, because any forecasted precipitation is mostly absorbed in the soil and produces little immediate impact on streamflow, though it may influence the antecedent conditions for subsequent precipitation-runoff events (delayed impact). Finally, streamflow response to a forecast is best expressed in probabilistic terms, and decisions framed in terms of change in odds of success or failure for key indicators. Recommendations are also given for further research.

The Midwest is the largest agricultural area of the United States. Historically, the climate, characterized by moderate temperatures and ample rainfall, has been suitable for un-irrigated agriculture. However, the specter of climate change has created concerns about the future of Midwestern agriculture, regional fresh water resources and the relationship between the two. Implications of climate change for the Midwest are revealed at the river basin scale in a recent study on the Mackinaw River Basin, a typical agricultural Midwestern watershed of central Illinois, through modeling exercises. Generally in this study a future climate with more frequent droughts is envisioned based on the outcome of one of the major General Circulation Models (GCMs), the Canadian Climate Centre model. Climate change in and of itself will affect the vulnerability of regional fresh water resources, by altering the low flow frequencies of stream at reference gauging stations. Moreover, the threats of droughts may motivate farmers to introduce irrigation in this traditionally rain-fed area to maintain high and stable yields. Such irrigations, if any, could exacerbate the effects of the changes in climatic factors. This study shows that the changes in climatic factors of temperature and precipitation will reduce the agricultural productivity, trigger irrigation and increase the low flow frequencies at reference gauging stations. However, the adverse effects of changes in temperature and precipitation may well be counteracted by the effects of elevated CO₂ concentration in atmosphere. Thus, the opposing effects of climate change could very well leave agriculture in central Illinois more or less unchanged.

Water resources and environmental assessment using hydrologic models are difficult tasks in ungaged basins. A methodology for reconstructing distributed rainfall data suitable for use in hydrologic models based on the data stored in atmospheric databases such as the U. S. National Center for Atmospheric Research/National Center for Environmental Prediction (NCAR/NCEP) global reanalysis data has been developed. The NCAR/NCEP global reanalysis data provides gridded atmospheric data at six hour intervals at a scale of approximately 210 km by 210 km that is consistent with global observations of weather related variables such as temperature and pressure. For regional and watershed scale hydrologic problems, detailed information such as orographic enhancement of rainfall at a scale down to several kilometers is required instead of the original 210 km data. In order to test the suitability of atmospheric model at regional scale (IRSHAM) and atmospheric model at watershed scale (MM5) to reconstruct rain fields at a scale suitable for use in hydrologic models, test cases have been run over two regions — the Lake Tahoe Basin in the western United States and the Ara/Tone Basins in Japan. Results of the studies indicate that the methodology shows promise for reconstructing rain fields over ungaged basins for use in hydrologic modeling for water resources and environmental assessment.

The Okanagan Basin in south-central British Columbia is one of the most arid regions in Canada. Presently, water resources in the area are under stress due to recent population growth, intensification of irrigated agriculture and recreational activities, and extensive logging at higher elevations, leading to increased concerns about water quality and fish habitat. In addition the stressors listed, climate change impacts could alter water supply patterns significantly. A major focus of this research is to encourage the local community to consider climate change in their long-term water plans. Between November 2003 and February 2004, the research team conducted stakeholder workshops to evaluate water management and adaptation options. Participants investigated the local and regional adaptive capacity, determined preferred adaptive strategies, and identified economic, institutional, social, and political barriers to implementing these strategies. The results of these workshops will be presented.

Rainfall input for urban wastewater collection system modeling is critical. The rainfall accuracy and reliability directly affect the confidence of the model results. In order to characterize the rainfall across a large metropolitan area, a network of ground rain gages is typically installed. Networks of ground rain gages have weaknesses in accurately defining the spatial variability of rainfall events across the entire metropolitan area. The density of the rain gage network often leaves large areas where rainfall is not measured and must be estimated. A number of mathematical methods have typically been used to estimate rainfall at points between ground rain gages. Advances in radar rainfall technology have led to the improved characterization of rainfall across larger study areas. Calibration of radar, even with sparse rain gages, results in aerial rainfall estimates that are better than either system alone. This paper inspected the simulated I/I (inflow/infiltration) responses in the sewer system from using four methods to define the rainfall across a portion of the service area of the Metropolitan Sewer District of Greater Cincinnati (MSDGC). The four methods are, 1) closest ground rain gage, 2) the Thiessen polygon method, 3) the inverse of the square of the distance method (IDS), and 4) radar-rainfall technology. The calibrated sanitary sewershed models were used as areas to estimate precipitation and to compare the simulated results with the rainfall input defined by the four methods.

The abstract of this paper states the paper will be about rainwater harvesting and indicates rainwater harvesting is a field that should be accorded serious attention by many disciplines. Within the body of this paper there will be numerous illustrations demonstrating that rainwater harvesting on any major scale is considered "outside the box" by those who could affect the status of this overlooked subject. The first time the term "outside the box" was applied to my efforts was just after I received an invitation to try out for the New York Giants baseball team. It was the bottom of the ninth in an important game for the team and myself. There was one out and we had a runner on third. I was at bat and the coach just called for a suicide bunt. The other pitcher and I had been having a real duel and the score was tied at zero-zero. They must have picked up the sign because his pitch was very high, something he had not done all game. My teammate was on his way and was definitely dead meat if I did not do something. All I could think of was to go after the ball, so I jumped up and laid down a perfect bunt. The action was so unusual everyone was safe. That other coach went running out to the umpire (a major league umpire) screaming "he was out of the box!" It turned out there was no rule against jumping up in the air and hitting the ball while both feet are off the ground. That umpire ruled that the "box" is three dimensional, not two. By going "outside the box" we were able to win the game.

Multiple linear regression (MLR) analysis was performed to develop mathematical relationships to predict highway runoff pollutant concentrations from various site and storm event predictor variables. Partial correlations were used to select the independent variables for the MLR models. The final "optimized" MLR model was used to generate a new fitted variable calculated as the cumulative effects of the significant predictor variables for each constituent. This fitted variable was then included as the single covariate and final MLR models were developed for each constituent. Predictor variables found to have significant impacts on highway runoff pollutant concentration include: total event rainfall (TER), cumulative seasonal rainfall (CSR), antecedent dry period (ADP), contributing drainage area (DA), and annual average daily traffic (AADT). The MLR models then were validated using actual highway runoff measurements that were not included in model development.

The Aswan High Dam Reservoir, formally known as Lake Nasser, is one of the largest man-made lakes in the world. It has played a major role in protecting Egypt against natural disasters due to shortage of rainfalls at the source of the Nile, during extended draughts or due to the hazards of the very high floods. Since 1964, 4.95 billion cubic meters (bcm) of sediments have deposited in Lake Nasser with almost 85% in Sudan and 15% in Egypt. Studies indicate that up to 134 million cubic meters a year end up in the lake, with 130 million sedimenting and 4 million passing through the Aswan High Dam to the valley north of Aswan. This paper proposes to mine these rich sediments using modern lake mining techniques. Current technology for hydraulic dredging using jet and slurry pumps, floating pipelines in High Density Polyethylene (HDPE) would allow such a solution. Sediments would be pumped in a slurry form to engineered mud ponds, similar to tailings dams to throw the foundations for new communities. The de-silting of the Nasser Lake may be executed at local, national and international levels (between Egypt and Sudan). Small-scale dredgers should be used for local farming. Preliminary calculations indicate that a 20 Megawatts pumping system would be recommended to convey to shore 10 million cubic meters of sediments. From the shore, a system of dedicated slurry pipelines, booster stations with local ponds for slurry discharge would fertilize lands closer to population centers.

Site selection procedures have improved rapidly with the acceptance of GIS technologies that can quickly evaluate the properties of thousands of potential sites. However, when more than a single parcel of land is required or when the certainty of the decision requires a more deliberate selection procedure, a systematic methodology is needed to analyze multiple scenarios and select an optimal or near-optimal solution. This need is especially true for military installations where unique land uses require calculated decision making that reflect more than "best judgment" decisions. Examples of these unique land uses include the location of hazardous storage facilities and the location of live-fire exercises. These distinctive activities are further complicated by the desire to increase the level of readiness by conducting additional training exercises on limited resources. An additional factor influencing the location of activities on military installations is that many of these sites have become prime habitat for threatened and endangered species. This paper describes the use of a site selection model to identify and evaluate and generate optimal regions with specific shape properties. In addition a heuristic algorithm that uses the "greedy" principle is applied to find solution regions with specific shape properties and non-inferior cost values. A case study using the model will be used to identify least cost areas for training activities on the Fort Hood military reservation.

Clays and their composites have been widely used in the past for secondary containment walls for underground storage tanks and landfills. These tanks are used commonly to store gasoline and other organic chemicals, both for industry and the commercial market. These materials also called liners are an inexpensive method of preventing seepage of chemicals into the groundwater. The behavior of clay liners for underground storage tanks or landfills is examined here on a microscopic level. Images from an Environmental Scanning Electron Microscope are presented with ramifications for clay modification to reduce seepage of chemicals through the clay liner.

Utility of a theoretically- and computationally-robust mathematical model for decoding the movement patterns of individual fish in 3-D space-time is described. Hydrodynamic stimuli queried from a 3-D CFD model (U²RANS) are used to elicit spatially-explicit virtual fish behavior. The modeling scheme, coupled Eulerian-Lagrangian agent individual-based modeling (CEL Agent IBM), is intuitive and based on well-established principles in computer science, fluid and water quality dynamics, computational fluid dynamics (CFD) modeling, and game and foraging theories. We demonstrate the utility of a prototype CEL Agent IBM (the Numerical Fish Surrogate) developed for outmigrating juvenile salmon in the Pacific Northwest. The Numerical Fish Surrogate is used by the US Army Corps of Engineers to decode behavior and then, using results from back-casting analysis, forecast the response of salmon to virtual designs of proposed alternative bypass structures to aid project selection and design. CEL Agent IBMs and the Numerical Fish Surrogate provide the means to integrate high fidelity CFD and water quality modeling and individual- based modeling to improve biological simulation of aquatic wildlife for water resource decision-support and reduce reliance on the build-and-test paradigm.

The 1989 Exxon Valdez oil spill showed that spill prevention and response actions in Prince William Sound and most other seaports were inadequate. Spills have continued to occur, but major improvements have been made in oil spill prevention and response planning. Federal programs have been enacted to deal with emergency spill and response measures including the Oil Pollution Act of 1990. The Act has incentives to prevent spills, with the U.S. EPA as the lead enforcement agency. EPA issued a final rule in 1992 amending the Oil Pollution Prevention regulation under the Clean Water Act setting requirements for Spill Prevention Control and Counter-measure Plans (SPCC Plans). During the same 15-year period, stormwater runoff management became a major concern across the country. The National Pollutant Discharge Elimination System (NPDES), put in place under the U.S. Clean Water Act (CWA), addresses discharge of pollutants to the nation's waterways. The 1987 Amendments to the CWA led EPA to develop Phase I of the NPDES Storm Water Program for industrial sources of stormwater runoff. Phase II, added a few years later, requires municipalities to control runoff. This paper deals with the combined issues of spill prevention and response under SPCC Plans and of industrial spills managed under the NPDES program. It identifies pollution prevention and spill control measures that prevent contaminated stormwater and runoff from entering water bodies. Successes in dealing with spills, spill prevention and NPDES compliance efforts are presented. The paper attempts to broaden the scope of spill prevention by looking at the combined efforts of spill response and recovery measures with stormwater runoff management related to potential industrial spills.

For the long-term strategic water resources planning, forecasting the future streamflow change is important to meet the demand of a growing society. The streamflow variation to the decade-long precipitation was investigated for the two major stage gauging stations in Korea. Precipitation and runoff characteristics have been analyzed at Yongwol stream stage in the Han River as well as Sutong stream stage in the Kum River for the future water resources management strategies. Monte Carlo method has been applied to estimate the future precipitation and runoff. Based on the trend line of 10-year moving average of runoff depth for the historical runoff records, the relation between runoff and the time variation was examined in more detail using regression analysis. This study showed that the surface flows have been significantly decreased while precipitation has been stable in these basins. Decreasing in runoff reflects the regional watershed characteristics such as forest cover changes. The findings of this study could contribute to the planning and development for the efficient water resources utilization.

Salt Lake County (SLCo), Salt Lake City (SLC) and Utah department of Transportation (UDOT) have been permitted under the Municipal Phase I National Pollutant Discharge Elimination System (NPDES) program since 1995. Stormwater runoff monitoring and wet and dry weather screenings have been conducted since 1992 on nine outfall locations located throughout Salt Lake County. A total of 13 storm events have been sampled up to 1999. Annual pollutant loads and event mean concentrations (EMC) were calculated for total suspended solids (TSS), trace metals, total phosphorus, and biological oxygen demand for three primary landuse types residential, transportation, and commercial/industrial. Annual pollutant loads for transportation and commercial/industrial landuses were determined to be the greatest contributor to all major pollutants. EMCs for all six major pollutants were found to fall within the range of values reported by several municipalities across the country with a similar climate type. The calculated EMCs also compared reasonably with the National Urban Runoff Program (NURP) data for Salt Lake County.

Sanitary sewer overflows (SSO) are becoming of increasing concern in the sewage system management to utilities and regulators. Many overflows are not easily identifiable. If undetected, they can adversely impact public health and are an environmental hazard. The number and severity of SSOs can be reduced through the proper sewerage maintenance and the detection-correction of ongoing events. A methodology is presented to locate meters to maximize the likelihood of detecting a system blockage. When a blockage occurs the discharge rates in the system are affected. If the changes are greater than one would expect due to random variability, they can be attributed to a blockage. The proposed method attempts to optimize the monitoring locations to maximize the number of blockages that can be identified by a given number of meters. The problem is solved using a set-covering approach. The approach allows the tradeoff between the number of detections and cost of gages to be easily assessed. Data from Pima County Wastewater Management's monitoring system is used as a case study. To introduce blockages and analyze system hydraulics a full hydraulic model for the sewer system is applied.

In 1998, the H.M.G. team was awarded a consultant contract with Salt Lake County Engineering for the NPS Public Education program. We inherited an up and running program looking for a new direction. We mapped out a 5-step process to fashion a new era in the UPDES Education Program — a process that led us to where we are today. The five steps are: 1 ) evaluation, 2) problem identification, 3) create a brand identity, 4) target a message to an identified audience and 5) select media and methods of delivery of that message to reach the audience.

Section 303(d) of the Clean Water Act and EPA's Water Quality Planning and Management Regulations (40 CFR Part 130) require states to develop total maximum daily loads (TMDLs) for their water bodies which are not meeting designated uses under technology-based controls for pollution. The TMDL process establishes the allowable loadings of pollutants or other quantifiable parameters for a water body based on the relationship between pollution sources and in-stream water quality conditions. Currently, most TMDLs are developed using a continuous simulation approach that employs a traditional deterministic rainfall-runoff model (e.g. HSPF). Use of such an approach in developing pathogen TMDLs can be problematic due to the water mass-balance errors which normally remain following even the "best" hydrologic calibration effort, and due to the challenge of calibrating the predicted pathogen loads to the erratic pattern of most observed pathogen data. In the current study both flow and pathogen loadings are modeled using probability distributions that are evaluated using a system dynamics modeling environment (e.g. STELLA) and Monte-Carlo simulation. The proposed approach has the advantage of eliminating mass-balance errors by using observed stream flow as opposed to rainfall as well as a better way of characterizing the probability of success of associated management strategies.

The determination of how pollutant loads should be distributed among different pollutant sources in the watershed, which is known as pollutant load allocation, is a critical step in Total Maximum Daily Load (TMDL) development. Under current TMDL practices, TMDL allocations are typically determined through a trial-and-error approach of reducing pollutant loadings until the watershed simulation model predicts that water quality standards will be met given a Margin of Safety. Unfortunately, because there may be many feasible combinations of load reductions and/or significant uncertainties, it is difficult and time-consuming to compare different allocation scenarios using a trial-and-error approach. A robust optimization model is developed in this study to incorporate the uncertainty of water quality predictions and to minimize pollutant load reductions given various levels of reliability with respect to the water quality standards. The Generalized Likelihood Uncertainty Estimation is used to explicitly address the uncertainty of a watershed simulation model, Hydrological Simulation Program — Fortran. The uncertainty is integrated into TMDL allocations using a robust genetic algorithm model linked with a response matrix approach. The developed robust optimization model is demonstrated on a case study based on. the Moore's Creek fecal coliform TMDL study. The trade-offs between reliability levels and total load reductions of allocation scenarios are evaluated, and the optimized load reduction scenarios are compared with the scenario generated by a trial-and-error approach and approved by the USEPA. The results show that the optimized load reduction scenario requires 30% less load reductions than the scenario approved by the USEPA at the same reliability level.

A TMDL under Section 303(d) of the Clean Water Act (CWA) is the maximum amount of a pollutant that a water body can assimilate without violating water quality standards (WQSs) and the allocation of that amount to contributing sources of the pollutant (anthropogenic and natural) plus a margin of safety.

The Belle Fourche River has been identified by the 2002 South Dakota 303(d) Waterbody List as impaired due to total suspended solids (TSS). The river has been designated as a warm-water permanent fishery. Based on this designated beneficial use, the criteria states that the TSS should not exceed the 30-day maximum of 90 mg/l or a daily maximum of 158 mg/l. A 3-year study was completed to determine and document sources of impairments to the Belle Fourche River Watershed (2.1 million acres) and to develop feasible recommendations for rehabilitation. The flow in the Belle Fourche River is heavily influenced by the Belle Fourche Irrigation District (BFID) project servicing approximately 51,000 acres. The primary cause of the elevated TSS concentrations is stream incising from alteration of the hydrologic regime due to nonused irrigation water discharging to the surrounding waterways Biologic indicators can be used to indicate long-term quality of a water body, reflecting overall ecological integrity. Macroinvertebrate biological samples were collected at 16 sites to support the water quality sampling and analysis. A good reference site was not available for this watershed. Thus the Environmental Protection Agency's (EPAs) methodology documented in the Rapid Bioassessment Protocols for Use in Wadeable Streams and Rivers was applied with modifications to develop a measure of impairment. Nonparametric statistical analyses (Kruskal-Wallis) were used to compare the sites and to determine the relative level of impairment. A robust regression equation was also developed relating TSS and seven biologic indicators.

The stormwater runoff is a primary carrier of non-point source pollutants to surface waters such as streams and reservoirs. Therefore, water quality varies highly with flow rate in the non-point source dominated surface waters. The existing water quality criteria have commonly been developed for the flow condition under which point source pollutants are dominant but not for the flow condition under which non-point sources are dominant. This study took Whippany River watershed, NJ as an example to demonstrate how the fecal coliform count varied greatly with the stream flow rate, and how the required reduction level for fecal coliform load would vary greatly depending on the flow condition chosen to meet the numerical water quality criteria. A calibrated watershed model was used to calculate the required load reduction levels for given flow conditions. A statistical method used to develop fecal coliform TMDLs was reviewed and its application results were compared with the watershed modeling results. It is recommended that the flow condition under which a TMDL is developed be quantified, using a calibrated watershed model, in order to achieve a better consistency in the required pollutant load reduction and to better educate the stakeholders.

Stormwater Treatment Areas (STAs) in South Florida are large-scale constructed wetlands operated as flow-through treatment systems to reduce phosphorus levels entering the Everglades in order to promote ecological restoration. The removal mechanism is through vegetation and periphyton uptake and sediment accretion. As part of the Everglades Construction Project, 16,800 hectares (ha) of constructed wetlands are being built to reduce phosphorus load from stormwater runoff and Lake Okeechobee discharges into the Everglades Protection Area in South Florida. At this time, four constructed wetlands with a total area 7,930 ha are operating, while construction is nearing completion in the remaining two totaling 8,864 ha. This study presents the performance of STA-1 West from May 1, 1999 through April 30, 2003, corresponding to the fifth through eighth year of operation. The 2,700 ha constructed wetland has five cells with three parallel treatment systems. Surface vegetation cover consists mainly of cattails, submerged aquatic vegetation, open water and other mixed vegetation.

Maroun River in Khouzestan province in south of Iran is one of the most important Iran water resources and it enters to international Shadegan Wetland (Biggest Iranian Wetland). It is very important that we have Fishways with high efficiency for each obstruction across the river such as Diversion Dams and Regulator Dams. The dominant family of fishes, which live in Shadegan wetland, is Cyprinidae family, which is resident at this wetland and Maroun River. On this river, there is one Reservoir Dam and four Diversion Dams that have been constructed by Khouzestan Water & Power Authority (K.W.P.A.). The Reservoir Maroun Dam has no facilities for fish migration and here is some efforts for finding a possibility for installing a suitable fish facility for this dam. Fishway of Jaeezan diversion dam on Maroun River constructed about 6 years ago and dose not have good efficiency and it is not functioning properly. During the migration season, fishes are trapped downstream of the dam. The first part of this paper belongs to results of study on finding rehabilitation for this Orifice-Weir type Fishway. Behbahan diversion dam on Maroun River, which has been constructed in 20 years ago, has no Fishway and the second part of this paper includes the results of study on finding a possibility for installing a Fishway at this dam.

In 2002, netting surveys in the Cedar River upstream of Spalding, Nebraska revealed very low numbers of native fish such as channel catfish. At Spalding, a low-head (∼3.6 m) dam blocks the Cedar River, and the species that exhibited low numbers usually migrate upstream to breed. The low fish counts are not atypical of streams in Nebraska that are straddled by low-head dams. In 2003, because of necessary repair work, the Spalding reservoir was drained, and the Cedar River ran freely through the reservoir. A brief netting survey in the river revealed a dramatic increase in channel catfish and other migratory species. While the reservoir was empty, small fish were observed migrating up the tailraces. Velocities in the tailraces were measured and will be used, along with other information, to design a low-velocity bypass for the dam. Critical velocities determined for the fish in the study and the resulting low-velocity bypass design will provide a practical solution for fish migration problems throughout Nebraska.

The United States waterway system finds itself facing increasing commercial traffic while the system's infrastructure has essentially reached its design life. The result is an increased demand on the nation's navigation locks. Financial constraints require innovative design concepts on infrastructure improvements. The overall savings estimated from the use of these innovative construction concepts for the US Army Corps of Engineer's top 11 high-priority navigation projects (lock replacements) are between 1.4 and 1.8 billion dollars. The Greenup Lock in the Ohio River is one of these projects. One of the innovative designs being considered for this site is extending the lock walls of the existing 204.2 m auxiliary lock and converting it into a 408.4 m long lock. Two alternatives have been proposed within this lock extension project. A cost effective design alternative being considered for the filling and emptying system is the use of the existing 204.2 m lock system to fill the 408.4 m lock. This type of system is hydraulically unbalanced and could be prone to extremely slow filling and emptying times, and higher hawser forces. If the filling and emptying times are slow enough to adversely affect the total transit time, significant economic losses could occur. Also, if the hydraulic performance of the system is poor, the risk of accidents increases due to excessive hawser forces and surface disturbances. Another alternative is to build a separate filling and emptying system for the extended portion of the lock. This will provide a more balanced environment within the chamber during the filling and emptying cycles. While being a more costly alternative, it can significantly reduce operation times. Research on these systems has determined minimum filling and emptying times, with acceptable hawser forces, minimal surface drift, and disturbance. Design guidance is developed from a combination of laboratory flume data supplemented with computational modeling results. This paper presents the modeling methods used to establish design criteria. This criterion is necessary for design engineers investigating the 204.2 m extension of the Greenup Auxiliary Lock.

Hydro-induction sediment discharge technique (HSDT) aims to prevent reservoir sedimentation by withdrawing the discharge water from the reservoir bottom to suck up and discharge sediments, and by inducing vertical vortexes and undercurrents in the reservoir bay to keep sediments agitated and moving. The technique is facilitated with an innovative discharge conduit system having bottom intakes that are engineered and configured to suit individual dam situation for maximum process efficacy. Analyses of the hydraulics and operating principles of the bottom intake discharge system are presented. Guidance for designing and configuring the discharge conduit system is provided. Several key issues concerning the practical application of the technique are discussed briefly. The process feasibility and functionality of the bottom intake discharge system have been demonstrated with a small test setup. Testing with large scale models and field tests are needed to develop design data for big dam application.

The Red River watershed modeling study was conducted by WEST Consultants, Inc. for the U.S. Army Corps of Engineers, St. Paul District. The primary purpose of the study was to develop hydrologic, water control, and hydraulic models for the Red River of the North above Fargo, North Dakota / Breckenridge, Minnesota. The modeling work included development, calibration and verification of the HEC-ResSim (reservoir simulation) model, unsteady HEC-RAS (flood routing) model, and a hydrologic model that provides local inflows from all the contributing areas in the watershed. The models will be incorporated into the Corps Water Management System (CWMS), a nationwide integrated framework of hardware, software, and other resources that acquires, analyzes, and stores data; develops decision support information; and allows user access to any data and information on the system. This paper presents individual model components of that integrated system for the Red River watershed.

One of the strongest typhoons ever struck the Korean Peninsula, Typhoon Rusa, from the 31st of August to the 1st September 2002, caused the historical precipitation and floods on some parts of Korea. In a region, a historical rainfall amount of 880 mm for 24 hours exceeding the PMP (840 mm within 25 km²) was recorded. The resulting floods left 213 dead and another 33 missing in the whole country. Damages caused by the typhoon were as much as $4.5 billion. During the floods, morphologies of many streams flowing especially to the East Coast from the mountainous regions were dramatically changed by the high-speed and highly concentrated flows. Channel response to such extreme flood clearly shows the inertia-dominant behavior of channel straitening, widening, and steepening. Re-channelization is to restore the stream channel with new channel forms and new floodplain management practices after a stream channel and surrounding floodplain are completely or near-completely altered by the extreme flood. It would usually follow, in case that the channel and socio-economic conditions allow, the channel forms newly shaped by the extreme flood.

The Hanford Reach is the only remaining unimpounded reach of the Columbia River in the United States above Bonneville Dam. Discharge in the Hanford Reach is regulated by several dams and is often subject to rapid changes. Sharp flow reductions have led to the stranding or entrapment, and subsequent mortality, of juvenile Chinook salmon (Oncorynchus tshawytscha) and other important fish species within the Hanford Reach. A multi-block two-dimensional depth-averaged hydrodynamic model was used to simulate time-varying river velocity and stage in a 37 km portion of the Hanford Reach. Simulation results were used to estimate time-varying juvenile Chinook salmon habitat area, and the part of that habitat affected by discharge fluctuations. Affected habitat area estimates were made for the Chinook salmon rearing period of four years. These estimates were used, along with other important factors, to establish a statistical relationship between discharge fluctuation and impacts on juvenile Chinook salmon.

The importance of including floodplain effects in one-dimensional river channel hydraulic computations is well known. However, in morphodynamic modeling, these are often partially ignored by using a single effective discharge to drive bed elevation changes. This neglects both changes in the flow and sediment transport fields at overbank discharges and long term change in channel capacity (and thus effective discharge) caused by deposition on or erosion from either the channel bed or the floodplain. This paper presents a model for reach-averaged channel bed and bank top elevation evolution that specifically accounts for changes in channel depth over time. The model considers two grain sizes: one for sand, which interacts primarily with the bed, and one for mud, which interacts only with the floodplain. The model also describes the evolution of the proportion sand and mud in the floodplain deposits. Sediment transport and floodplain deposition are driven by a simple gradually varied flow solution. Erosion from the floodplain is represented as a net loss associated with channel migration. Because overbank deposition is strongly affected by flow, effective floodplain deposition and in-channel sediment transport are obtained by integrating results from an entire flow duration curve. In the absence of bed elevation changes, the channel and floodplain co-evolve toward a stable bankfull geometry where overbank deposition just equals floodplain erosion.

Numerous flood events have historically devastated the Tillamook Valley in Tillamook County, Oregon. The City of Tillamook and both the commercial and agricultural activities in the Tillamook Valley are highly susceptible to flooding events, as the local topography is flat and broad before rising abruptly to the steep mountains of the Pacific Coast Range. As a result of the February 1996 flood, Congress authorized a request from Tillamook County to investigate water resource problems in the region. Under a contract with the U.S. Army Corps of Engineers, Portland District (District), WEST Consultants Inc. (WEST) developed one-dimensional unsteady flow hydraulic models of the system, originally applying the Danish Hydraulic Institute (DHI) MIKE 11 model and then later converting the model to the U.S. Army Corps of Engineers Hydrologic Engineering Center River Analysis System (HEC-RAS). These two hydraulic computer simulation models are commonly utilized for unsteady flow application in the both the public and private sectors in the United States; however, it is in error to assume that the construction of these models should always be developed in the same manner and with the same modeling assumptions. WEST recognized the differences between these two models, as part of the study, and compared specific aspects between HEC-RAS and MIKE11 including the model equations and variables, approaches for model construction and development, and the effects of these variations on computation results as they relate to the Tillamook Valley. This paper discusses specifically the differences in computation of the channel roughness.

As part of a larger study of juvenile fish passage, a micro-Acoustic Doppler Velocimeter (ADV) was used to characterize the three-dimensional velocity and turbulence characteristics in a full-scale culvert with spiral corrugations. The culvert was set up in a test bed constructed to examine upstream passage success of juvenile salmon under various culvert conditions. The test culvert was 12.2 m long and 1.83 m in diameter, and set at a 1.14% slope. The corrugations were 2.54 cm deep by 7.62 cm peak to peak with a 5° right-handed pitch. Tailwater elevation was adjustable with a stop-log system and was set slightly above the water surface level at the culvert exit. Cross-sectional grids of ADV measurements were taken at discharges of 28, 57, 113, 227, and 453 lps at nine locations within the culvert barrel and just inside the headwater and tailwater tanks. Results revealed asymmetries in the velocity and turbulence distributions. These asymmetries caused a Reduced Velocity Zone (RVZ) on the right side of the culvert as seen looking upstream. Velocity and turbulence levels in the RVZ were found to be less than in mid channel or on the left side of the culvert, and the difference became greater as flow rates increased. Lateral and vertical velocity components were very small relative to the axial component, while lateral and vertical turbulence intensities were comparable to the axial component. Inlet loss coefficients were calculated as well and ranged from 0.32 to 0.42. Relationships between the average velocity and the velocity and turbulence intensity in the RVZ were developed, which may be useful for evaluating whether the barrel of a culvert is passable for juvenile fish. The RVZ could be beneficial to juvenile salmon during upstream passage.

Two geometrically undistorted models of St. Louis Water Company's Meramec Plant Intakes 1 and 2 were constructed at a scale of 1:5 in order to solve chronic pump-vibration problems. Because of the unconventional layouts of small, deep, rectangular pump pits with three pumps each, vertical-pump inverted drafttubes (VPIDs) were developed. Each intake chamber was divided into three individual bays using two vertical partition walls. Each VPID included a two-segment straight-line suction scoop, a floor splitter, a backwall splitter, two sidewall floor-corner fillets, two vertical backwall corner fillets, flow-turning vanes, and a vertical guidewall at the VPID entrance. In this paper only modeling efforts on Intake 1 are described. This is the second VPID model designed by IIHR — Hydroscience and Engineering (IIHR), The University of Iowa.

The Louisville Water Company (LWC) has investigated the alluvial aquifer upstream of the Falls of the Ohio River since the late 1940s with an interest in securing an adequate volume of high quality water from this resource. A high-yield horizontal bank filtration collection well was put into operation at the B.E.Payne Water Treatment Plant in 1999 (Ohio River Mile 592), with a design capacity of 15 million gallons per day (MGD) and an initial maximum capacity estimated at 27 MGD. In the first 3 years of operation, the yield has decreased to an apparently stable safe yield of 18 MGD. This paper seeks to better define the stochastic relationship between riverbed plugging and riverbed scouring for the nearly 4-year operating history of the riverbank filtration system at Louisville, and to project the performance of this system through a 20 year forecast. If successful, the technique used to determine this model will be applied to several other bank filtration systems in the US and Europe to gain a better understanding of the influence of riverbed scouring in bank filtration systems.

Summer temperatures in the Lower Snake River can be altered by releasing cold waters that originate from deep depths within Dworshak Reservoir. These cold releases are used to lower temperatures in the Clearwater River, a major tributary to the Lower Snake River, and to improve hydrodynamic and water quality conditions for migrating salmon. This project monitored the complex three-dimensional density currents at the Clearwater and Snake River confluence and the processes that led to stratification of Lower Granite Reservoir (LGR) during the late spring, summer, and fall of 2002. In addition to monitoring the LGR environment, a three-dimensional hydrodynamic and water quality model was also applied. By utilizing both field data and a numerical model, a more holistic view of the 3-D density currents was discovered than by using either method alone. During this process, it was discovered that several predictable stratification patterns would develop depending upon the discharge ratio and the thermal gradient between the two rivers. These results illustrate the complex hydrodynamic structure at the confluence of the Clearwater and Snake Rivers, which has previously been shown by fish biologists to be a difficult passage zone for migrating salmonids of various life stages.

In free-surface flows the turbulence in a liquid may be sufficient enough to disturb its surface to the point of entraining air into the flow. The process of entraining air can be important in many applications including hydraulics and other areas. For example, in water treatment air is needed to sustain microorganisms for water purification and in rivers and streams to sustain a healthy fish population. Air entrainment is typically engineered into spillways downstream of hydropower plants to reduce the possibility of cavitation damage at the base of the spillway. There are also situations where air entrainment is undesirable such as in the sprue and runner systems used by metal casters, and in the filling of liquid containers used for consumer products. CFD can be used as an alternative to modeling such processes or as a compliment to experimental processes because it can quickly provide a large amount of information about flow. What makes it even more attractive is when conditions are difficult to reproduce experimentally or data is difficult to obtain (for example, regions that cannot be seen or are inaccessible to measurement). Unlike physical experiments, it is easy to measure quantities throughout the computational domain because all regions within that domain are accessible.

Determination of Manning's coefficient, n, for natural streams remains a challenge in practices. One source for determining the n-values that has received practitioners' attention is presenting the n-values determined from field data (measured discharge and water-surface slope) in combination of photographs and site descriptions (ancillary information). Further improvements in the visual approach can be made in presenting site characteristics and describing site ancillary information. In this manner, users can use the presented information for sites of interest with similar features. This approach in a current project on the subject for Illinois streams is discussed.

Screens may be utilized efficiently for dissipating energy of water. In this study, water flowing beneath a gate is used to simulate the flow downstream of a hydraulic structure and screens are used as an alternative tool for energy dissipation. Investigations are done conducting a series of experiments. The porosity, thickness, and the location of the screens are the major parameters together with the Froude number of the upstream flow. The experiments covered a range of Froude numbers between 5 and 18, porosities between 20% and 60%, and location of the screen up to 100 times of the undisturbed upstream flow depth. The thicknesses of the screens used are in the order of the undisturbed upstream flow depth. The results show the importance of each parameter on the energy dissipating performance of the screens and the system. It is observed that screens dissipate more energy than a jump within the Froude number range covered in the present study.

Using dimensional analysis and experimental data from three previous studies, two relationships for predicting depth-averaged velocity distributions in straight trapezoidal channels were developed. The data had a range of discharges (8.0 to 4,248 l/s), velocities (0.10 to 1.03 m/s), bottom widths (0.30 to 3.6 m), flow depths (5.2 to 80.5 cm), and sideslopes (1.0 to 3.0, horiz./vert.). The first relationship is general and requires measured velocity data for calibration. When measured data are not available, the second model, for which specific coefficients are prescribed, must be used. The first model is more accurate and should be used when possible. The second model also yields acceptable results. Among other things, the models can be employed in stable channel designs, streambank stability analyses, and to evaluate streambank stabilization designs including riprap and bioengineering designs. Brief reviews of the models are presented along with a demonstration application.

By means of the software "Excel", the critical and normal flow depths can be determined through the use of analytical solutions and procedure presented in this paper. Four channel shapes are included: they are triangular, trapezoidal, parabolic and circular. For the triangular and trapezoidal channels, the critical and normal depths can be determined for channels with unequal side slopes.

Natural or introduced vegetation often comprises the largest component of composite channel roughness in flood plains, flood control channels, and naturally occurring rivers and streams. Recent research has yielded numerous techniques for predicting Manning's "n" for flexible vegetation. This paper presents a comparison of several of these techniques. New data obtained from recent laboratory studies involving actual vegetation is used to independently validate and compare the various techniques.

River restoration and bioengineering programs commonly use vegetation to stabilize banks. For this, it is important to know the forces that the flowing river water exerts on the vegetation. To this end, a laboratory study was performed at the USDA-ARS National Sedimentation Laboratory. Dowels were used to simulate emergent rigid vegetation. Dowels were systematically placed in the laboratory flume, beginning with a single dowel connected to a load cell to measure the drag force and then placing dowels around the single dowel to determine any shielding effect. It was found that the drag coefficient of a single dowel ranged from 4.5 to 6.5, which is higher than that published for a 2-D cylinder of infinite length. A momentum balance analysis shows that (1) for the single dowel case, the momentum balance approach determines C_D>>1 contrary to the low values for a cylinder and (2) the C_D for the measured dowel is about two times higher than the momentum formulation. This difference is attributed to the presence of large standing waves upstream of the dowel and the breaking of these waves. When additional vegetation elements were added to form a matrix, it was found that as vegetation elements were added, the drag forced increased initially due to the vegetation concentrating the flow towards the instrumented dowel, and then decreased due to dowels being directly upstream of the instrumented dowel and thereby blocking the flow. These results have implications for managed planting arrangements in stream corridor rehabilitation programs.

This paper presents a Reynolds stress modeling of compound open-channel flows with vegetation on the floodplain. In the Reynolds stress model, we use the SSG model by Speziale et al. for pressure-strain correlation term, Mellor and Herring's model for the turbulent diffusion term, and Hanjalic and Launder's model for the dissipation term. In order to take into account the anisotropy of turbulence due to the free surface, the combination of Shir's model and Gibson and Launder's model is included in the pressure-strain correlation model. Model validations are carried out for the compound open-channel flows without vegetation. Then, the model is applied to the compound open-channel flows with vegetated floodplains. The mean flow and turbulence structures are simulated and the impact of vegetation on the floodplains is investigated.

Field work on the Athabasca part of the Peace Athabasca Delta, in Northern Canada, has shown the usefulness of employing new equipment to understand channel hydraulics in remote areas. A differential GPS system was used for accurate surveying work, and an Acoustic Doppler Profiler (ADP) was used to determine flow rates. Channel bank heights for an 81km distance upstream from Lake Athabasca were determined, as were river surface levels at medium discharge in the Athabasca River and its distributary channels. As many as 15 channel sections were surveyed per day with a boat-mounted ADP. For the same daily discharge value, cross-sectional average velocities in the main channel were found to be preserved at about the same value for a remarkably long distance in the final part of the delta, all the way to the lake. Comparison with old data shows a remarkably rapid rate of channel building. Flood magnitudes that cause bankfull flow conditions were found to change by a large factor from the head of the delta to the lake.

The Environmental Modeling Research Laboratory (EMRL) at Brigham Young University in partnership with the U.S. Army Engineer Research and Development Center (ERDC) and the Federal Highway Administration (FHWA) has developed a tool for automating the process of incrementally loading boundary conditions into numerical hydrodynamic engines. The process of incrementally loading includes making revisions of model parameters from unrealistic but numerically stable values to realistic values that require similar initial conditions in order to be numerically stable. Variables that are traditionally used for incremental loading include water surface elevation, flowrates and eddy viscosity. This study involves extensive model runs with two different numerical models, TABS-RMA2 and FESWMS-FLO2DH. The effectiveness of incrementally loading each variable individually and various combinations of loadings are compared. Results include guidelines for efficiently applying incremental loading strategies to numerical modeling applications.

A depth-averaged 2-D numerical model for unsteady tidal flow in estuaries is established using the finite volume method on non-staggered, curvilinear grid. The 2-D shallow water equations are solved by the SIMPLEC algorithm with the Rhie and Chow's momentum interpolation technique. The convection terms are discretized by one of the hybrid upwind/central difference scheme, exponential difference scheme, QUICK scheme and HLPA scheme. The algebraic equations are solved using the strongly implicit procedure (SIP). The model is capable of handling the drying and wetting problem due to the variation of water surface elevation. The model has been tested in Tokyo Bay and San Francisco Bay. The tests show that the present model is very stable and efficient. The simulated water elevation and flow velocity are in good agreement with the measured data.

The Netherlands are situated at the downstream end of the Rhine River. A large part of the country can be supplied with water from the river in the case of precipitation deficits. For the assessment of the economical damage due to drought it is necessary to consider the rainfall and river inflow simultaneously. Transformed normal distributions as well as Gumbel distributions have been fitted to the observed precipitation and discharge deficits. The sensitivity of joint probabilities to the choice of the marginal distributions, the dependence structure and the ‘failure region' is investigated. It is found that the bivariate normal distribution underestimates the probability that both the rainfall and runoff deficit are extreme due to its asymptotic independence.

Probabilistic modeling of activated sludge wastewater treatment systems was performed to compare the uncertainty, with respect to attaining pollutant removal targets, associated with specific reactor configurations (conventional, completely mixed, and step feed activated sludge systems) and operating variables (solids residence time, hydraulic residence time, recycle ratio, and influent distribution). Stochastic simulation results showed that different reactor configurations inherently had different levels of uncertainty associated with them. Whether a specific operating parameter had an effect on uncertainty, and the extent of that effect, was dependent on the process configuration. In allof the configurations considered, process uncertainty could be reduced by increasing the solids residence time. Treatment uncertainty was also reduced when the hydraulic regime of the reactor was closer in character to plug flow rather than to completely mixed. Hence, the treatment uncertainty for the conventional activated sludge system was lower than that for the completely mixed activated sludge system. Furthermore, operational changes that enhanced mixing, such as increasing the recycle ratio in a conventional activated sludge system, or distributing the influent along the length of the reactor basin in a step feed activated sludge system, also tended to increase process uncertainty. In every case, system response could be traced to the effect of the imposed configuration and parameter changes on the hydraulic regime and reaction kinetics within the treatment system.

This paper describes development of a hydrologic functional assessment model for depressional wetlands. Model output is a normalized metric that ranges from 0 to 1 and indicates the level of alteration to the wetland hydrologic regime relative to the unaltered condition. The metric selected to characterize the hydrologic regime is the number of days (0–365) with standing surface water. The model includes four variables, each corresponding to an alteration type common in the study area. One variable reflects alterations to wetland morphology that increase or decrease storage capacity. Two variables reflect alterations to the wetland watershed that result in increased surface runoff. The remaining variable reflects changes in evapotranspiration losses resulting from a timber harvest. Model variable sensitivity and interaction are examined by simulating 47 years of hydrology for a depressional wetland site. Simulation results are used to develop simple relationships between model variables and the normalized metric.

An integrated surface water/groundwater model of the western Big Cypress Basin in southwest Florida was developed using the MIKE SHE/MIKE 11 modeling system to evaluate the effectiveness of structural measures for restoration of the wetlands in Southern Golden Gate Estates (SGGE). The terrestrial and the estuarine ecology of the SGGE its receiving estuaries have been severely damaged by ill-planned development of roads and drainage canals. The hydro-geologic system of the basin is characterized by a high water table, low topographic relief, and a vast network of man-made canals that were excavated to lower the water table Restoration of over 135,850 hectare (55,000 acres) within the area is planned as one of the priority projects of the Comprehensive Everglades Restoration Plan. The restoration plan includes plugging of canals and introduction of water into previously drained wetlands. An initial study was conducted that evaluated the benefits of one scenario. In this study, the model was updated with more accurate canal information and ground water demands, and the re-calibrated model was superior to the original MIKE SHE model. Twenty-five alternative restoration strategies were evaluated to identify a strategy that maximized wetland restoration and minimized construction cost and flooding to rural residential areas. The model results indicate that a number of the alternative restoration strategies will significantly improve wetland hydrology. Model results will be presented to demonstrate the expected changes in hydrology in the proposed wetland system.

Increased federal regulation of stormwater runoff in recent years has concomitantly increased the interest of engineers, planners, and municipal officials towards the effective planning and design of stormwater management programs. Given the increased attention to stormwater management, the demand for technical tools to solve problems efficiently has arisen. To meet the demand for planning tools that can take advantage of spatial data, a Visual Basic for Applications tool for use within the ArcGIS 8.x geographic information system (GIS) software package has been created. The GIS tool offers a construction site erosion and sediment control module and a post-construction stormwater management module to aid the development of best management practice (BMP) plans. Within the GIS environment, BMPs can be placed on a site map and the automatic processing can be activated to determine the cost and pollutant removal performance of the BMP plan given spatial data describing site topography, land use/cover, and soil type. Site specific BMP performance and cost information can be input or a database of national cost and performance data can be used by default. BMP plans can be revised and the relative cost and performance of alternative plans determined and compared. The construction site erosion and sediment control module implements an erosion prediction algorithm based on the Revised Universal Soil Loss Equation (RUSLE). The post-construction stormwater management module implements Curve Number hydrology for long-term runoff volume prediction, as well as Event Mean Concentration associated with land use for pollutant load prediction. Monte Carlo water quality simulation and a BMP optimization routine are also being developed for later implementation.

ArcGIS-SWAT, a GIS interface for the Soil and Water Assessment Tool SWAT has been developed. It is an ArcView 8.x extension that uses objects defined according to the ArcObjects standard, which conform to the Component Object Model (COM) protocol; and its code has been written entirely in Visual Basic (VB). It consists of a data model in geodatabase format specifically designed for SWAT, and the necessary tools to populate it, prepare the SWAT input files and write the SWAT simulation results back to it. The data model approach has the advantage of being the repository of all the data (including the geographic data) of a SWAT simulation.

An important data requirement for a typical one-dimensional hydraulic flood model is accurate cross section data describing both channel bathymetry and the floodplain at multiple locations along a river reach. Sources of obtaining these data include traditional surveying, light detection and ranging (LIDAR) datasets, and digital elevation models (DEM) produced from remotely sensed elevation datasets. Required cross section data can then be extracted from a digital terrain model (DTM), which is created using data from one or more of these sources. High density datasets can be merged with datasets of a lower resolution to create a DTM. Sparse data in a DTM may exclude important terrain features such as accurate river bathymetry and levees while data that is too dense may present challenges because of computer processing times. Many rivers have existing levees, either built for flood control or formed by structures like roads, which affect water levels during a flood event, but are often not well represented in a DTM. It is possible to manually edit each cross section to represent a levee or another topographic feature, but it would be a tedious task. Faulty or unnecessary data points should be edited or removed from high resolution DTMs in order to obtain optimum input cross section data. Some rivers do not include levees presently, but it is desirable to analyze and compare the results of flood events with and without proposed levees or roads. Editing terrain data to accurately include river bathymetry, or structures such as levees is a process that will improve the quality of flood event analysis from a digital terrain model as well as open the door to analyzing multiple scenarios with ease.

Floods cause loss of life, property, and economic activity. In January 1997, California experienced one of the most geographically extensive and costly floods in the State's history. Of the State's 58 counties, 48 were declared disaster areas. Nine people were killed, 120,000 people were evacuated from their homes, and 300 square miles were flooded. Damages approached $2 billion, and floods impacted over 23,000 homes as well as numerous businesses, agricultural lands, bridges, roads, and floodwater management infrastructures. Estimated indirect costs associated with the disruption of the State's economy exceeded $5 billion. California is expected to witness an increase in its population by an estimated 40 to 50 percent over the next 25 years, leading to development pressures on floodplains that have not been mapped (United States Census Bureau). In 2001, the California Department of Water Resources (DWR) focused its attention on these flood-prone areas by initiating the statewide Awareness Floodplain Mapping Program (AFMP) where floodplains were analyzed using approximate methods supplemented with field verification. DWR noted that mapping all floodplains to Federal Emergency Management Agency (FEMA) standards would be a formidable task as well as cost prohibitive. DWR's priority is to identify potential flood hazard areas by producing quality maps that maximize the extent of coverage and makes the most efficient use of funding. A goal of the State's AFMP is to compliment FEMA's mapping program and not supersede it. The AFMP will only study streams located outside existing FEMA floodplains. The State's intent is to map potential flood hazard areas for these floodplains within the next 10 years. This will enable California communities to better protect its citizens and their properties.

All measurements of physical quantities are subject to uncertainties. These may be due to bias errors in the equipment used for the measurement or to random scatter caused by a lack of sensitivity of the equipment. Every day throughout the world numerous hydrometric measurements are made of components of the water cycle without any explicit report on the uncertainty of the measurement. The uses made of these measurements in civil engineering design and operation of river works and in water resources management generally requires an assessment of the reliability of the measurements. The result of a measurement is only an estimate of the value of the measurand and only complete when accompanied by a statement of the uncertainty. The arithmetic mean value of the observations is the basic statistic and starting point, but a measure of the scatter, or dispersion, about the mean is required. To determine this dispersion in measurable terms is the object of uncertainty analysis of hydrometric data. The "true value" of a measurand is unknown, but an estimate can be made by the use of simple uncertainty techniques, uncertainty being defined as the range in which the true value is expected to lie within specified limits. In this respect a confidence limit, or level, of 95% is normally adopted for hydrometric data which infers that 95% of the observations should, on average, lie within the specified limit of two standard deviations from the mean. It should be noted, however, that no uncertainty value can be assigned to the remaining 5%. Examples will be presented of uncertainty analysis of river flow measurements that follow the procedure adopted in the International Standards Organization's "Guide to the Expression of Uncertainty in Measurement"

Regional analysis of annual maximum precipitation data was conducted by using index-flood method based on L-moments. Annual maximum rainfall series for the durations of 1, 2, 6, and 24 hours were collected at 337 gauging stations in Korea. The identification of homogeneous regions was performed based on cluster analysis such as hierarchical and nonhierarchical techniques. As the results, various cluster analyses can be used to group appropriate sites into a region for regional analysis and to estimate more reliable quantiles.

The Illinois State Water Survey is developing detailed hydrologic and water quality simulation models of the entire Illinois River basin. Under this plan, a hydrological simulation model was developed for the Vermilion River watershed (one of the major tributaries of the Illinois River) using 1970–1995 observed streamflow data from two U.S. Geological Survey gaging stations in the watershed and weather data from six representative stations. The model was developed using the Hydrological Simulation Program - FORTRAN. The watershed was subdivided into 25 smaller hydrologically connected sub-watersheds and their stream reaches. Hydrologic response units were created within each sub-watershed based on land use and hydrologic soil groups. The model was calibrated using 1987–1995 streamflow data from the gage at Pontiac, IL, and then verified using 1972–1986 data from the same gage and using 1972–1995 data from the streamflow gage at Leonore, IL. The model simulated the mean monthly streamflows with correlation coefficients and Nash-Sutcliffe Efficiency near or greater than 0.8 during calibration and verification periods. Flow-duration curves of the daily observed and simulated streamflow data indicated good simulation for all flow conditions except for some very low flow periods. The model underestimated the flood year of 1993, but generally overestimated some very low-flow years.

The Cache River is located in the extreme southern part of Illinois, just north of the confluence of the Ohio and Mississippi Rivers. In 1915, a cutoff to the Ohio River was constructed east of Karnak that resulted in sub-dividing the Cache River watershed into the Upper and Lower Cache River watersheds. The Upper Cache River watershed consists of the eastern part of the watershed draining directly to the Ohio River through the Post Creek Cutoff. The Lower Cache River watershed consists of the western part of the watershed draining to the Mississippi River through a diversion channel at the outlet. The Cache River levee was built in 1952 along the west bank of the Post Creek Cutoff to prevent runoff from the Upper Cache River watershed from entering into the Lower Cache River. Because of these and other alterations in the watershed, the hydraulics of the Lower Cache River are very complex. Smaller, local tributaries provide the source of water for the upper portion of the Lower Cache River. Once water from tributaries enters the upper part of the Lower Cache River, it can flow in an easterly direction towards the Cache River levee or flow in a westerly direction towards the Mississippi River. A combination of several factors determines the direction of flow in the upper part of the Lower Cache River. Hydrologic and hydraulic models were developed to simulate the hydraulics of the Lower Cache River and to evaluate different alternative measures that can be implemented to reduce or eliminate the unnatural reverse flow in the upper portion of the Lower Cache River.

Long-term climate variability has been shown to affect flood frequency in New South Wales, Australia. This study extends this work by developing a Bayesian hierarchical model for predicting floods at ungauged locations. Flood records were stratified using a threshold value of the Interdecadal Pacific Oscillation (IPO), which characterises the low frequency component of sea surface temperature anomalies in the Southern Pacific Ocean. It is shown that the stratified flood distributions differ significantly with ratios of flood magnitudes being approximately 1.6 times greater during the IPO negative years. Moreover, the evidence suggests that the log-normal distribution fits the stratified flood records satisfactorily. The stratified flood records were then subjected to regional flood analysis using a Bayesian hierarchical approach. The regional model considers at-site floods to be spatially correlated via an intersite distance correlation function. The hierarchical model proposes that the parameters of the flood frequency distribution for any site are random samples from a regional probability model. This allows the inclusion of catchment characteristics, while also explicitly allowing for intersite variability. Model calibration was performed using Markov chain Monte Carlo methods. An important outcome is the quantification of predictive uncertainty at an ungauged catchment.

Accurate predictions of rainfall-runoff relationship have been the goal of numerous hydrologic studies. Despite the availability of data, the usefulness of physically based, distributed parameter hydrologic models over lumped models remains uncertain. The purpose of this study is to evaluate the performance of a physically based, distributed model and a lumped model, and shed some light on their capabilities and limitations for a mid-size catchment. The watershed selected for this study is the Goodwin Creek Experimental Watershed, located in northern Mississippi. The drainage area of Goodwin Creek is 21.2 km². The hydrologic modeling system HEC-HMS developed by US Army Corps of Engineers, and MIKE SHE developed by the Danish Hydraulic Institute were used as the lumped and disturbed hydrologic models, respectively. The evaluation criterion of model performance is the ability of each model to predict time and magnitude of peak discharges, and runoff volume.

Hidden Markov models (HMMs) offer a plausible representation of long-term hydroclimatic persistence in rainfall and streamflow observations. Persistent climate processes influence hydrological observations at various time scales. This paper develops the stochastic framework of two-state HMMs to better represent climate-rainfall interactions at both monthly and annual levels. Two new models, a hierarchical HMM and a non-homogeneous HMM are introduced, and fitted to monthly rainfall and streamflow observations from Australia. The value of these models to identify two-state persistence is compared to that of existing two-state HMMs.

This study presents a numerical examination of surface and subsurface solute transport processes over spatially heterogeneous hillslopes with special focus on investigation of role of spatial heterogeneity on the run-on process. The high degree of heterogeneity exhibited by the soil parameters that govern the mechanisms of infiltration on hillslopes necessitates a stochastic approach. In this study, numerical techniques are used to generate error maps that serve as guidelines as to when spatial variability has to be explicitly considered because run-on is substantial, versus those cases where the nature of spatial variability is such that run-on can be essentially ignored. When run-on is not significant, mathematical analysis can be utilized to determine field-scale infiltration, runoff, sediment transport, and surface and subsurface solute transport.

The Washington D.C. metropolitan area's (WMA's) PotomacRiver water supply intakes are located downstream of the Potomac's Point of Rocks gage and upstream of the Little Falls gage. In early July of 2002, it was noted that there was an apparent water loss between these two gages. Little Falls flow was estimated based on Point of Rocks gaged flow, with adjustments for intervening withdrawals and inflows. The estimated Little Falls flow was as much as 150 million gallons per day (MGD) higher than the recorded flow at the Little Falls gage. This discrepancy implied that water was being lost between the Point of Rocks and Little Falls gages. Management of streamflow in the Potomac River is critical during drought conditions to ensure an adequate water supply in the WMA, which is the main source of water to 3.6 million people and to ensure recommended environmental in-stream flow. Water is released from several key reservoirs to maintain minimum flows and to supply water for the WMA during times of drought. These releases are managed carefully in order to maintain sufficient reservoir storage; therefore, water loss is a major concern. Gage error, permitted withdrawals, evaporation, and transpiration by trees along the river were investigated to account for the loss. The Interstate Commission on the Potomac River Basin (ICPRB) calculated potential gage error and examined permits to determine permitted levels of consumptive withdrawals from the Potomac. Gage error may account for some loss but was thought unlikely to explain a significant portion of the loss In total, about 24 -MGD of the loss could be attributed to permitted withdrawals during the summer.

This study proposes a drought warning system based on long-lead climate forecasting data to provide information for risk management of water resources. The proposed method was applied to the Taipei areas in Taiwan. The system consists of a water resources system dynamic model, a streamflow model, and an irrigational water demand model. The GSM climate forecasts were downscaled by the RSM model in another study with a lead time of three months. Then the downscaled climate forecasts were entered into the streamflow and water demand models. The water resources system dynamics model predicts deficits in water supply after entry of forecasted streamflow and water demands. Drought warnings will be issued when cumulative water deficits reach an optimized threshold. The water supply and demand from 1978 to 1999 were simulated and used as input to test the system. Drought warning based on long-lead climate forecasting can provide decision makers more information in advance to determine water resources management strategies.

The Simulation/Optimization MOdeling System (SOMOS) has modules adapted for a range of international water resource management needs. The SOMOA module is best for field groundwater and conjunctive water management situations. It employs several analytical simulation models and several optimization methods to give appropriate optimal guidance for situations having sparse data. SOMOA requires understanding of the limitations of common analytical equations. It does not require experience in finite difference or element numerical modeling. SOMOS also has modules ideal for situations in which there is sufficient data for numerical modeling or aquifer and stream-aquifer systems. Emphasized here are SOMOA and its applications for common field-level water management problems. Sample situations include optimizing: minimizing required surface reservoir size for a period of low surface flows in a stream-aquifer system; conjunctive use for blending irrigation waters of different qualities; and artificial injection.

Three new herbicides, a fungicide, and an insecticide were tested for their ability to leach in various tropical soils in Hawaii. In addition, atrazine and bromide ions were used as two reference chemicals in the study. Label rates of these chemicals for agriculture or insecticidal use were applied to tilled soils. Five test sites at differing topographic, climatic and soil conditions were selected in the Hawaiian islands: three on Oahu, one on Maui and one on Kauai. Each site had four test plots; two for herbicides and two for insecticide/fungicide. No crops were grown and the soil surface was covered with a thin layer of mulch to prevent weed growth and decrease water evaporation from the soil surface. An irrigation system was set up at each site to apply water at close to potential evapotranspiration rate to induce leaching. Disturbed soil samples were collected on the day after pesticide application and in one to four week intervals for four months. Soil hydraulic and climatic data from each site were collected during the same period. Soil samples collected in the field to a depth of 2.1 m were analyzed for content of the chemicals, yielding concentration profiles for elapsed times of 1 day to 16 weeks. Prior to pesticide application, soil profiles have been also characterized for pH, ΔpH, organic carbon content and basic nutrients: P, K, Mg, Ca. Undisturbed soil core samples have been taken and sets of water-retention curves have been determined in the laboratory. Field tests for unsaturated and saturated hydraulic conductivity using tension disc infiltrometers have been carried out at all sites. In addition, laboratory tests on the sorption and degradation of the chemical compounds are being carried out.

Ground water hydrologists use specific yield to estimate the total volume of recoverable ground water in storage in an aquifer and measure pumped or drained water from water bearing material dewatered. Specific yield is defined as the volume of water that a saturated rock or soil will yield by gravity per unit area of soil per unit change in water table elevation. Accurate estimates of specific yield and other hydrogeological parameters are critical for management of water resources and can be used to control well yields and resulting water level depths and rates of water level declines. The study area is in Lithia, near the planned Tampa Bay Regional Reservoir in the southeastern portion of Hillsborough County, Florida. Data from four soil probes were studied. The objective of this paper is to obtain precise estimates of the specific yield by methodological fitting procedure and determine functional relationships between measured soil and site characteristics. A methodology for estimating specific yield in shallow water environment using continuous soil moisture data is introduced. The ultimate specific yield values estimated varied from 0.23 to 0.37 within the sensors and varied with depth in the profile of shallow water vadose zone. The results will be incorporated into the large-scale integrated surface and ground water models that have been used successfully in west central Florida.

In modeling furrow irrigation, it is often assumed in furrow irrigation modeling that roughness and cross-section are constant along the furrow length; and as a result, wetted perimeter decreases as the flow decreases. Several researchers have attempted to determine the relationship between infiltration in furrow irrigation and wetted perimeter. Fangmeier and Ramsey, Strelkoff and Souza, Blair and Smerdon, Izadi and Wallender, Schwankl and Wallender, and Oyonarte et al., have concluded that infiltration varies, either linearly or non-linearly, with wetted perimeter. However, Trout has argued that such a relation exists only in the case of stagnant tests, not in tests conducted under real conditions. The basis of Trout's argument is that field measurements clearly show that cross-section and roughness change along the furrow. Infiltration is a function of wetted perimeter and is typically modeled through compensations for wetted perimeter. A problem arises between field analyses attempting to measure infiltration and the subsequent use of the results in simulation and design models. Field studies typically give either very limited point measurements or an average of the entire furrow length. Point measurements using various infiltrometers usually lack sufficient coverage to identify spatial variability. Tests which deduce infiltration from advance, runoff, and/or tailwater provide a single value in which all field variations are consolidated, and if they include the adjustments for wetted perimeter, they do so with a formulated relationship and assumptions regarding the cross-section and roughness.

Oregon State University in collaboration with the ASCE/EWRI task committee on Soil and Crop Hydraulic Properties for surface irrigation has begun conducting a comprehensive evaluation of various algorithms proposed for the determination of infiltration parameters from surface irrigation measurements. The procedures under development will include the following elements: (1) Establish specifications and guidelines for assembling and formatting data sets used by the various infiltration parameter estimation models to be tested. These input specifications will include different types of field data commonly collected, the units to be used, formatting of field data to be consistent with the units used, and any constants and other prior assumptions to be employed; (2) Formulate procedures for assessing input data quality and for determining whether the functional forms of infiltration models that are assumed either explicitly or implicitly in the various algorithms of interest are in fact consistent with input data patterns; (3) Establish criteria for characterizing the performance of algorithms as analytical tools when used specifically for decision making in surface irrigation; and (4) Conduct prototype evaluations of two algorithms for determining infiltration parameters; specifically, the one-point and two point methods.

The two-point method is a simple, compact, and relatively accurate inverse solution technique widely used to estimate the parameters of the Kostiakov-Lewis infiltration function. However, the applicability of the method is limited to sloping free-draining furrows. In this paper, the two-point method is modified to extend its application to close-end level-bed furrows. The modifications deal with the calculation of the basic intake rate and the inlet flow cross-sectional area during the advance phase. With the modified method, basic intake rate is calculated as the quotient of the change in volumetric surface storage between two selected time lines during the steady state period and the product of the furrow length and related time interval. In addition, the inlet flow cross-sectional area can be calculated as a function of flow depth at the inlet or using the Manning equation and an assumed linear flow depth gradient. The proposed approach was tested by comparing advance trajectories computed by SRFR, on the basis of infiltration parameter estimates by the modified two-point method, with field observed advance. A satisfactory agreement was obtained between SRFR predicted and field observed advance.

The volume balance model has been an important tool in the evaluation and design of surface irrigation systems. Perhaps it's most useful application has been in evaluating the intake parameters of a sloping border or furrow during the advance phase of irrigation. Unfortunately, these volume balance estimates of intake parameters may not provide a unique enough solution to allow accurate simulation of runoff and recession. There are three important problems with a volume balance approach. First, the inflow during the advance phase must be steady. This is often difficult to achieve in the field. Second, if an intake equation with steady-state or cracking terms is used, the related parameters must be determined outside of the volume balance analysis and then input. These values are difficult to measure accurately in the field. And third, unless the analysis specifically ignores the surface volume during advance, these models use a uniform flow equation, like Manning, to describe the cross-sectional area of flow at the field inlet and then an assumption regarding the shape of the flow profile downstream — generally assuming the cross-sectional area is constant. The assumption of a constant cross-sectional area is known to introduce substantial errors.

A recurrent dynamic neural network (RDNN) based on Hopfield model was applied to linear quadratic regulator controller of an irrigation canal. The Saint-Venant equations of open-channel flow were linearized using the Taylor series and a finite difference approximation of the original nonlinear, partial differential equations. Using the linear optimal control theory, a Linear Quadratic Regulator (LQR) controller was developed for an irrigation canal with a single pool and results were observed. And then LQR controller was redesigned with applying Hopfield recurrent dynamic neural network (RDNN) to solution of Riccati equation for calculating the LQR control gains. The main advantage of Hopfield neural network model is its robustness and flexibility when dealing with unknown perturbations and parameters. The results of this study shows that an RDNN based LQR controller (RDNN/LQR) for irrigation canals offers an alternative to traditional LQR controller when dealing with the disturbances in the irrigation canals.

Farmers in Bangladesh have significantly extended agricultural activities into the dry winter-summer season over the past 2 decades to produce more through double and triple cropping and also to remain safe from damages due to floods. Agriculture in winter-summer season is fully irrigated. Ground water is now meeting the lion's share of this irrigation need. However, limit of abstraction of ground water within the capacity of technical and financial management by individual farmers, as it is now, will soon be reached at. It is, therefore, necessary that available surface water be harnessed, as far as possible, to support irrigation need in the dry season. Further to that, detection of arsenic contamination of ground water in some parts of the country in the recent past has made conservation of surface water all the more important. Also, surface water is cheaper and of better quality and so farmers have definite priority of choice of surface water over ground water for irrigation. There is no scope to develop reservoirs for storage of water in this flat country. With this background, Bangladesh has adopted Rubber Dams for conservation of water in the channels of its small and medium rivers to support winter-summer irrigation. Since introduction of the technology in 1995, Rubber Dams have been recognized as a successful method of water conservation under the conditions in Bangladesh. The Paper presents the technology of Rubber Dams in some details together with last 7 years' experience of Bangladesh with Rubber Dams. References have been made of the impact of Rubber Dam Projects on rural poverty reduction through the impact on agriculture of small and marginal farmers and through the scope of certain other economic activities for the local poor. Reference has also been made on the potential of application of Rubber Dams in regional countries — India and Sri Lanka together with interests indicated by some concerned agencies.

Residential irrigation water use is a large component of total water use in Florida. With one of the largest state populations in the U.S. that is rapidly growing, competition between urban, agricultural, and other water users is increasing. This project measured residential irrigation water use and irrigation system uniformities in central Florida. Three treatments were established as follows: typical irrigation practices (T1), irrigation based on historical evapotranspiration (T2), and water wise landscape plus irrigation designed to minimize water use (T3). T1 homes averaged 143 mm of irrigation per month in 2003 while T2 and T3 averaged 103 and 74 mm, respectively. This corresponds to a 28 and 48% reduction in irrigation water use compared to the typical practices, respectively. Distribution uniformity (DU_1q) of homes tested averaged 0.40–0.50 which was statistically lower than irrigation heads tested under controlled conditions and also lower than industry standards. The preliminary results indicate that design for uniform application of water is difficult in small non-uniform areas of residential yards with current commercially available equipment and that irrigation water use can be reduced by proper scheduling and with landscapes that are designed to minimize irrigation.

The California Department of Water Resources (DWR) conducted feasibility-level engineering and environmental studies under the Integrated Storage Investigations Program. The U.S. Bureau of Reclamation (USBR) was the lead federal agency for the EIS/R. As part of the investigation, DWR and USBR evaluated alternatives using the Tehama-Colusa canal to provide up to 5,000 cubic-feet-per-second (cfs) of conveyance capacity to deliver water to the proposed Sites Reservoir. To evaluate the existing canal's capacity, and to assist in the design of alternatives to increase the capacity, a hydraulic model was developed. HEC-RAS was applied between Mile Posts (MP) 0.51 and 66.07, and included 16 check structures, 20 inverted siphons, 33 bridges, and numerous other crossings. The canal was designed and built in reaches using standard cross sections for each reach. A spreadsheet was developed to automatically create the HEC-RAS geometry file from these standard sections and individual station invert elevations, obtained from construction plan/profile drawings. After the various structures were added, the model was calibrated to recent moderately-high flows, and then applied to evaluate a range of alternatives. The alternatives included raising the canal's lining and structures, and widening the canal to achieve new design flow capacities.

An instance-based heuristic evolutionary algorithm for the parameterization of the 2D surface water quantity and quality model CE-QUAL-W2 is presented. The methodology developed is a hybrid genetic algorithm (GA) - k-nearest neighbor algorithm (GA-kNN). To reduce computational efforts, a "hurdle race" approach was developed, with a "hurdle" being a predefined value of the parameterization objective function computed at different points during a simulation run. The "hurdle race" is formulated for accepting/rejecting a proposed set of parameters during a CE-QUAL-W2 simulation; the k-nearest neighbor algorithm (kNN) for approximating the objective function response surface; and the genetic algorithm (GA) for a mutual linking. The proposed methodology overcomes the high non-applicable, computational effort required if a conventional parameterization search technique is used, while retaining the quality of the final solution.

Use of optimization to calibrate a water distribution model has been the subject of considerable research. Such models have not been systematically studied on realistic sized systems to any great extent. This paper presents an evaluation of Darwin Calibrator, a genetic algorithm approach to achieving optimal calibration. It shows that the Calibrator works well when given a sufficient amount of good quality data and examines the relationship between the quality of the data and the quality of results.

Long-lead streamflow forecasts are very important for the operation of water resources systems. Calibration and validation of long-term forecasting models are needed to increase the accuracy of forecasted values, especially in the regions with limited observed data. This study discuses the procedure of model calibration and validation in two conceptual and statistical models. The forecasting techniques applied in this study are artificial neural networks (ANN), and non-parametric K-Nearest Neighbor (K-NN) method. The procedure of calibration and validation of these models is also discussed. The processes of model selection, calibration, and cross validation are combined through an optimization procedure. The best architecture and the number of training epochs in ANN, and the best weights of independent variables beside the number of neighbors in the K-NN method are determined by this optimization procedure. The models are applied to the Zayandeh-rud reservoir basin in Iran to forecast the monthly streamflows of Zayandeh-rud River, one of the greatest rivers in the central part of Iran. Hydroclimatic data, such as observed soil moisture, snow budget, rainfall, air temperature, and streamflow data in the previous month are considered as the predictors of streamflow in the models. The results of forecasting using the calibrated models indicates that the presented procedure for the time series analyses techniques provides significant improvements in the accuracy of streamflow forecasting.

This paper will discuss the application of a range of algorithms to optimization of calibration. The application is to the Cannonsville Reservoir, which has an area of 1200 square kilometers. The Cannonsville supplies water to NY City. There is a danger of excessive phosphorous over the watershed causing euthrophication of the reservoir and the water supply. The model predicts flow and the transport of sediment and phosphorous. For this problem, a modified simulated annealing algorithm and a global greedy search were the best algorithms.

In practice, water distribution model calibration has been based on modeler judgment. Recent research has examined model parameter and predictive uncertainties and how those uncertainties can guide future data collection experiments. This paper presents a methodology that extends previous work by considering the impact of uncertainty on the final goal of the model of system design.

This paper presents a computational fluid dynamics (CFD) model developed using the commercial software FLUENT, and presents predicted hydrodynamic force coefficients for a disc-shaped butterfly valve at different valve operating angles. The predicted force coefficients were used to provide estimates of hydrodynamic loads on the butterfly valve during closure against flow. The results of this study expand the knowledge base of butterfly valve hydrodynamic forces. The use of FLUENT for estimating force coefficients was validated by comparing CFD predictions to experimental results for the drag coefficient of a coin-shaped disc placed perpendicular to flow in an infinite flow field. The sensitivity of the results to turbulence model selection, accuracy of discretization schemes, grid quality, and grid dependence was studied as part of the validation. CFD models of the butterfly valve were used to compute drag coefficients, lift coefficients, moment coefficients, and discharge coefficients for the valve at seven different opening positions. The computed discharge coefficients compared very well with experimental discharge coefficients available in the literature, which provided confidence in the accuracy of the predicted force and moment coefficients as well, and suggests that CFD analysis can be a useful tool for predicting drag forces and lift forces on similar hydraulic structures.

A set of three-dimensional, computational fluid dynamics (CFD) models were developed and used for the Bonneville Project tailrace to study the impact of a proposed outfall structures on the tailrace hydraulics; these structures were designed to improve the survival of (juvenile) salmon migrating downstream. Flows were simulated by solving the Reynolds-Averaged Navier-Stokes equations together with the k-∊ turbulence model in a commercial CFD code. The numerical model was validated using field-measured velocity data. The model results identified undesirable combinations of outfall location and operational scenario, which aided in selecting where the outfall structure was built. The numerical model provided an effective tool to rapidly simulate and visualize the flow field for multiple proposed outfall locations over a large number of operational scenarios. The visualizations of the results from the CFD model provided insights to hydraulic engineers and fisheries biologists working on the design and placement of the outfall structure.

This paper presents a simulation/optimization model based on artificial neural networks and genetic algorithms for solving multi-objective conjunctive water use problems. The model simulates and optimizes water flows in a hydraulically connected reservoir-stream-aquifer system. It is a powerful tool to help water managers and authorities in developing conjunctive water management strategies and evaluating tradeoffs between conflicting goals.

Ecosystem services such as sediment and nutrient load reductions, flood control, nutrient cycling, regulation of atmospheric gases, and expansion of wildlife habitat and biodiversity are increasingly recognized as society essentials and as having significant economic value. With the considerable successes that have been achieved in controlling industrial point source pollution, the target for improving ecosystem services in the United States and elsewhere is increasingly centered on landscapes. Control of non-point source water pollution, for example, is an issue that must be addressed by better managing land use patterns. This study aims to identify multi-seasonal land uses and tillage practices that most cost-effectively produce improved ecosystem services within a watershed and ultimately enable examination of the tradeoffs between commodity production and generation of those services. It focuses on three particular ecosystem service related objectives, namely reduction of sediment, phosphorous, and nitrogen loads to improve water quality in rivers and streams. To solve this multi-objective problem, an optimal control computational model is developed by interfacing a widely-used distributed watershed model, the U.S. Department of Agriculture's Soil and Water Assessment Tool, with a Pareto-based multi-objective evolutionary algorithm known as the Strength Pareto Evolutionary Algorithm (SPEA). The model is applied to the Big Creek watershed in southern Illinois. The results demonstrate that the SPEA is capable of handling the complexities posed by the multi-objective watershed management problem. Furthermore, the overall integrative methodology is effective in identifying a set of optimal solutions and gives the user access to important information regarding the tradeoffs among competing objectives. In addition, the model could be used to determine the most cost-effective means for meeting Total Maximum Daily Load (TMDL) criteria. Future work will emphasize the inclusion of additional ecosystem services and further investigation and comparison of alternative multi-objective search methods.

In standard optimization approaches for water resources management problems, the designer is responsible for correctly formulating mathematical equations to describe the system objectives and constraints. The search for optimal or near-optimal solutions is made under the assumption that these formulated objectives and constraints completely describe the system. However, in real systems that is often not true. Many qualitative criteria can be integral parts of the design analysis that numerically based algorithms cannot capture. For such problems, designer interaction with the search algorithm can help the search be more creative and inclusive. Genetic algorithms are ideally suited for incorporating such interaction in their usual search process, and can successfully evolve solutions that are optimal with respect to both qualitative and quantitative objectives. Under an interactive approach, the genetic algorithm performs the usual operations of selection, crossover, and mutation, but the user evaluates the suitability (‘fitness') of candidate solutions, enabling objectives that cannot be quantified to be included in the search process. In multi-objective problems, where quantitative objectives can be as important as qualitative fitness of designs, analysis of designs is done based on tradeoff fronts made from both quantitative and qualitative information. In this paper, we demonstrate the use of interactive genetic algorithms for long term groundwater monitoring problems, which have multiple numerical and subjective objectives. We also analyze the effects on the optimal monitoring designs of using an interactive optimization approach instead of more traditional numerical optimization approaches.

This work compares alternative genetic algorithm (GA) methods and applies the enhanced multi-objective robust genetic algorithm (EMRGA) for groundwater remediation design under uncertainty. The research presented in this study investigates modifications to the EMRGA and optimization problem. In the EMRGA, the overall fitness of a string is based on its performance over multiple generations. This work is significant due to the multi-objective nature of groundwater management problems and the inherent uncertainty involved in obtaining aquifer parameter values. The overall EMRGA 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. One advantage of the EMRGA over other GA optimization approaches is its ability to efficiently solve optimization problems that are both multi-objective and include parameter uncertainty. The EMRGA is evaluated on a pump-and-treat remediation design problem with the goal of identifying the optimal locations and pumping rates of extraction wells given uncertainty in hydraulic conductivity. The EMRGA evolves Pareto-optimal sets, which identify trade-offs between the conflicting objectives: minimizing both cost and contaminant concentration. Future plans for the EMRGA include evaluating its performance on a real field site, comparing the EMRGA with other hybrid methods, and using GIS to present and analyze the results.

An optimization algorithm derived from the artificial life paradigm is introduced. The algorithm was first tested on standard optimization test functions and subsequently applied to determine the optimal dam operations in a multi-reservoir river system. The optimal dam operation scheme is that which indirectly minimizes environmental impacts caused by short-term water level fluctuations, while maintaining desired stages within the river network, desired storage levels in the reservoirs, bounds on releases from dams/gate openings, and minimum pool levels behind locks and dams. The dam operation problem is solved using a discrete-time optimal control methodology in which an artificial life algorithm is interfaced with NWS-FLDWAV, an unsteady, one-dimensional, hydraulic simulation model. The hydraulic simulation model is used to solve governing hydraulic constraints, while the artificial life algorithm is employed to solve the overall control problem. Constraints on water surface and storage levels are handled using an additive penalty method. The developed model is applied to a portion of the Illinois River, for which the hydraulic simulation model was calibrated and verified. The model may prove to be a useful tool that assists engineers and managers in establishing coordinated reservoir operation rules.

Strategies for water supply and irrigation canal systems management must be established to meet water demands throughout the network over time while preventing problems such as water shortage or flooding of sensitive areas. A previous study demonstrated that by incorporating an unsteady flow model into an evolutionary algorithm- (EA-) based framework, efficient strategies for scheduling gate operations and allocating crops can be determined to address the various objectives and constraints over a period of time. However, as implemented previously and applied to a case study in Egypt, this method did not find solutions that satisfied all system constraints. Therefore, other constraint handling techniques are investigated to improve the EA performance for this problem. The most common technique for handling constraints in an EA is the use of penalty functions that are incorporated into the fitness criterion to degrade the fitness of solutions that violate one or more constraints. Alternative techniques have also been reported in the literature, including a multiobjective optimization technique that treats the constraints in single objective problems as additional objectives and a stochastic ranking procedure, both of which have been shown to perform efficiently for other engineering applications. In the previous study, a static penalty function was used to handle constraints in the EA. This paper investigates the performance of six constraint-handling techniques (five techniques from the literature, and a new proposed technique) for a water supply and irrigation canal network operation problem.

Nonpoint source management using wet detention ponds to reduce pollutant loading into receiving water bodies is a common practice. Design of these ponds is typically carried out individually to meet a target total suspended solids (TSS) removal level. An improvement to this approach is to generate cost-effective pond configurations that meet system-wide targets for removal of pollutant loadings, corresponding to a specific build-out land use plan. The amount of pollutant loading can be managed better through appropriate land use allocation planning conducted simultaneously with the design of detention ponds. Harrell and Ranjithan presented a modeling approach to address this problem and its solution using a evolutionary algorithm- (EA-) based search procedure. The model includes constraints on the system-wide land use distribution, which were handled using penalty functions. Most penalty function implementations, which are incorporated into the fitness criterion to degrade the fitness of solutions that violate constraints, require the user to specify and fine-tune penalty factors that determine the amount of penalty assigned as a function of the level of constraint violation. Alternative techniques have also been reported in the literature, including a multiobjective optimization technique that treats the constraints in single objective problems as additional objectives and the stochastic ranking procedure, which have been shown to perform efficiently for other engineering applications. This paper investigates the performance of three penalty function implementations and two alternative techniques for constraint handling in a watershed management problem.

The design of urban drainage networks is complicated by the need to consider a number of issues that conflict and compete with the goal of managing flood impacts. These issues primarily include environmental considerations, but may also include issues such as developable land impacts, system reliability, wetland impacts, aesthetics, etc., some of which may not be modeled explicitly. Modeling to generate alternatives (MGA) is a formal optimization-based technique to find near optimal alternatives that are maximally different from one another with respect to their decision attributes. This paper presents a new evolutionary algorithm (EA)-based technique, the Solution Set Algorithm (SSA), for performing MGA and its application to design problem involving the design a least-cost drainage network using the EPA's Storm Water Management Model (SWMM).

Development of stage-discharge relationships for coastal low-gradient streams is a challenging task. Such relationships are highly nonlinear, nonunique, and often exhibit multiple loops. Traditional methods, such as parametric regression, usually fail to model these relationships. Instead, this study takes advantage of advanced computation-intensive techniques, such as neural networks and local nonparametric regression, to model such complex stage-discharge relationships. Results indicate that both neural network and local regression models are able to reproduce the multiple loops that are observed at the outlet of a low-gradient stream in southwestern Louisiana. The models have higher prediction accuracy with high flows than with low flows. The study discusses the effect of issues such as sample size and data division on the performance of the developed models.

Design of water quality monitoring network could be a fairly complicated problem, where some questions such as number of gages required, frequencies of sampling, and benefits over cost ratio of monitoring should be determined. These issues are in case of sampling river water quality variables, as they are more error-prone, costly, and time consuming. The basic difficulty underlying the design and evaluation of monitoring systems is the lack of an objective criterion to assess the efficiency and the cost effectiveness of the monitoring network. In this paper, a combination of the Genetic Algorithm (GA) optimization model and a one-dimensional water quality simulation model is proposed for design of river water quality monitoring network considering the economic constraints. First, the locations of sampling stations are determined so that the water quality data can provide the best fit to the historical spatial distribution of different water variables considering their relative weights. Then the efficiency of the proposed model is evaluated using the water quality data of Karoon River in Iran. The results show that the model can be used as an effective tool for optimal design of water quality monitoring networks. This study also shows the capability of the GAs to overcome dimensionality problems, which can not be easily solved by other optimization methods such as dynamic or non-linear programming.

Multireservoir system consists of several reservoirs which are connected serially or parallel in the same basin. To optimize such a complex multireservoir system, the dynamic programming (DP), linear programming (LP) and non-linear programming (NLP) have been widely applied to real problems. However, when DP is applied to multireservoir system it has a major problem, so called ‘the curse of dimensionality' and LP and NLP have essential approximation problems dealing with discontinuous, nondifferentiable, non-convex, or multi-modal objective functions. Recently, there has been an increasing interest in a biologically motivated adaptive system for solving optimization problems. The genetic algorithms (GAs) are one of the most promising techniques in natural adaptive system field and receiving many attentions because of their flexibility and effectiveness for optimizing complex systems. Optimization of multireservoir system is to solve multi-dimensional and multi-objective problems and GAs are appropriate optimization methods to multireservoir system. GAs are not restricted by a number of dimensions because computer memory increases by dimensions linearly, not exponentially. Thus, there is no ‘curse of dimensionality'. Especially classical optimization methods such as DP, LP, and NLP are not proper to multi-objective optimization because these methods use a point-by-point approach, in which the outcome of classical optimization methods is a single optimal solution. However, GAs use a population of solutions in each iteration instead of a single solution, so they are called as population-based approaches. This is one of the most striking differences between classical optimization methods and GAs.

Given a set of data patterns; the objective is to determine which of these patterns should be used to train a back propagation neural network (BANN or ANN). To determine these training patterns, in this study, two types of data sets were investigated: a data set on turbine power and a data set on water quality data. As applied to each of the type of pattern set, the following procedure was applied. The set was arranged from the lowest to the highest, in terms of the target values. The first (lowest value) and the last (highest value) patterns were then used to train the network; this produced the first set of weights. These weights were then tested to see if they could predict the rest of pattern set, which served as the target values. The predicted values were then plotted against the actual values. These predicted values produced bumps on the plot. The pattern that produced the biggest bump was then chosen as the third pattern in the training set; this, in turn, produced the second set of weights. These weights were again tested if they could give a good prediction of the target data, which, of course, again produced bumps on the plot. The pattern that produced the biggest bump was again chosen and added as the fourth pattern to the training set, producing the third set of weights. The process was repeated for the fourth set of weights, the fifth and so on, until the plot of the predicted and the actual data produced a straight line. The patterns that finally produced the straight-line plot are selected as the training patterns.

This paper demonstrates how existing parameterization techniques for binary coded genetic algorithms can be extended to real-coded evolutionary algorithms. An easy-to-implement framework for automating parameter setting for real-coded evolutionary algorithms is demonstrated in this study using Differential Evolution (DE), a real-coded evolutionary algorithm. DE was selected because the algorithm has been successfully demonstrated on a wide range of applications as a result of its inclusion in both the Mathematica and MATLAB optimization toolboxes. Preliminary results show that the framework successfully eliminates trial-and-error parameter analysis, minimizes the algorithms' computational demands, and ensures that DE reliably attains high quality solutions. The parameterization framework is not application specific and can be easily implemented on other real-coded evolutionary algorithms.

Genetic algorithms have been shown to be powerful tools for solving a wide variety of water resources optimization problems. Applying these approaches to complex, large-scale applications, which is usually where these methods are most needed, can be difficult due to computational limitations. Large grid sizes are often needed for solving field-scale groundwater remediation design problems. Fine grids usually improve the accuracy of the solutions, but they are also computationally expensive. Multiscale parallel genetic algorithms have been shown to improve the performance of engineering design problems that use spatial grids. In this paper we present multiscale island injection genetic algorithms (IIGAs), in which the optimization algorithm has different multiscale populations working on different islands (group of processors). Each island has a fraction of its population on the coarse grid and a fraction on the fine grid. Different islands exchange the best individuals, at the same scale, after a fixed number of generations and thus drive the GA towards better and more accurate solutions faster. The performance of this approach is compared to a single population multiscale approach developed previously, using a field-scale pump-and-treat design problem at the Umatilla Army Depot.

Water distribution network calibration is mathematically unsolvable. Then, trial and error, optimization and artificial intelligence methods have been used to solve it. Artificial intelligence has given some effectiveness. However, some artificial intelligence methods are too slow, static or hard to adapt to changes in problem, etc. Therefore, some experiments are proposed to determine advantages and disadvantages of each AI method used to calibrate water distribution networks.

The calibration of water distribution networks is a problem with no mathematical solution. Fuzzy logic is used in this research to capture knowledge from people who manually calibrate those networks. This way, process could become automatic. Fuzzy logic is used to calibrate simple pipes, pipes series and paths inside networks. Several initial configurations are needed in the calibration process. Fuzzy system defines and calculates the changes that should be making to topological variables, in order to improve calibration. This procedure is repeated until fuzzy system establishes that no more changes must be made. A satisfactory calibration is obtained. Finally, fuzzy sets configuration is refined to improve the system accuracy.

Approximation ("meta") models have been used in coupled water resources optimization and simulation models to improve computational efficiency. In most instances, multiple simulation runs have been done before the optimization, which are then used to fit an approximate model that is used for the optimization. In this study, we propose a dynamic meta-modeling approach, in which artificial neural networks (ANN) is 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. We propose a dynamic learning approach that periodically re-samples new solutions both to update the ANN and correct the GA's converging route. This allows the meta model to adapt to the area in which the GA is searching and provide more accuracy. The results show that a proper sampling strategy can benefit both GA's searching and ANN'S retraining. In our test case, more than 90 percent of the numerical model calls were saved with no loss in accuracy of the optimal solution.

A potable water supply network design methodology was implemented using a Genetic Algorithm (GA). Information about location and network connectivity, pipe materials and minor losses coefficients is required. The GA searches for designs (diameter combinations) optimal from the point of view of cost of the networks' pipes while satisfying supply for demanded volumes and minimum pressures in the network junctions. Adjustment of GAs to this problem is shown: functions of adaptation (construction cost), recombination (crossover) and mutation are modified to suit it. A scheme (Cost-Restriction function) is introduced to induce the algorithm to generate designs that fulfill the required hydraulic restrictions. The variation of the cost of resulting design based on probability of mutation, generation size, and coefficients specific to the adaptation function are studied. Special attention is given to the influence of the factors that vary the diversity within each generation of the algorithm (represented by the probability of mutation and one of the variables of the reproduction function). Finally the designs obtained through the algorithm are compared with the designs obtained according to Featherstone criterion.

Traditional trial-and-error methods make the calibration and verification of a model considerably time consuming. In addition, it is doubtful whether the best results will be achieved. However, by use of optimization techniques in calibration and verification, the best kinetic parameter estimates can be obtained in a shorter time period. In this study, a genetic algorithm (GA) is used to determine the reaeration coefficients for QUAL2E. An objective function, defined by sum-of-least-squares, is used in order to describe the difference between the observed and simulated dissolved oxygen concentrations. Simultaneous calibration and verification is carried out by treating the verification of the calibrated parameters as a constraint. The efficacy of GAs in predicting the model parameters is evaluated for perfect and biased measurements. The results show that GAs can successfully carry out simultaneous calibration and verification, and estimate the best set of reaeration coefficients to be used in stream water quality modeling with QUAL2E, especially for the accurate observation data.

Numerical models are a useful tool to evaluate and design NAPL remediation systems. Traditional models are complex and expensive on its application. For this reason this paper presents the application of a simplified Lagrangian transport model. For the application of this model, the hydraulic and NAPL structure is defined by means of probabilistic models. The parameters of these models are evaluated by means of a hybrid genetic algorithm, the SAHGA algorithm developed previously for the design of pump-and-treat remediation systems. The paper also includes one example of the application of this model to analyze tracer movement at a PCE contaminated aquifer.

Recent advances in the availability and accuracy of multi-dimensional flow models, the advent of precise elevation data for floodplains (LIDAR), and geographical GIS allow the creation of hazard maps that more correctly reflect the varying levels of flood-damage risk across a floodplain when inundated by floodwaters. Using intensity scales for wind damages, an equivalent water-damage flow intensity scale has been developed that ranges from 1 (minimal effects) to 10 (major damages to most structures). This flow intensity scale, FIS, is portrayed on a map as color-coded areas of increasing flow intensity. This should prove to be a valuable tool to assess relative risk to people and property in known flood-hazard areas.

A systematic review of the literatures, stream flow and sediment data pertinent to the Yazoo Basin tributaries in theDemonstration Erosion Control (DEC) project was conducted to identify the hydraulic roughness during flow process in the unstable channels. The stream flow process was identified by the temporal variation of stage and discharge parameters. The 15-minute stage-discharge data and the field survey information were used to analyze the stage-discharge and the hydraulic roughness parameters. Several functional relationships in the forms of normal rating curve (stage vs. discharge) and roughness impacted rating curve (stage vs. roughness*discharge) were conceptualized to identify these parameters during a channel flow process. Regression analysis was performed to identify the best parametric relationships. By this method two types of parametric relationships, in the forms of polynomial and power were identified for each site. The polynomial relationships exhibit the highest correlation coefficient, followed by the power relationships among the other parametric relationships. The simultaneous application of normal rating curve (H vs. Q) and the roughness impacted rating curve (H vs. nQ) can provide an effective monitoring tool for the temporal changes of n during a channel flow process.

A physically intuitive derivation of the Muskingum-Cunge method is presented, based on work by Cunge, Perumal and others. Common misunderstandings related to "prism and wedge" storage and negative values of X are clarified. A connection is made between the Modified Puls method of channel routing and Muskingum-Cunge, and guidance is given for evaluating the hypothetical number of reservoirs to use in Modified Puls for both prismatic sections and complex natural stream reaches.

Two-dimensional model solving advection-dispersion equation was constructed with modified governing equation using finite element method. This model was testified by the comparison with the experimental results of the mixing in the laboratory S-curved channel and also RMA4 was run at the same condition and compared. These results show the improved simulation result for the meandering channel.

River planform, restoration, and naturalization processes have attracted many research efforts in the last decades. Meandering rivers are defined by a sinuous characteristic, and they are considered the most common type of natural streams and their studies are governed by the complexity of the physical processes that are involved in these natural environments. This complex system covers phenomena such as secondary flows, turbulence flows and sediment transport. Additional complexity is introduced by the irregularity of the boundaries presented in rivers, and by the continuous planform migration due to erosion and sedimentation processes, which makes even more difficult to obtain accurate results. Several rivers have been channelized in the past causing environmental and ecological problems, where restoration techniques evolve as natural solutions to channelization. These restoration techniques require the knowledge of the complex phenomena of rivers described before, in which the analysis of sedimentation and erosion, bank erosion, vegetation and of course flow field behavior in the river is required. Then, a methodology to analyze the evolution of meandering rivers is needed for the prediction of planform migration. This prediction is indispensable for economic and social reasons such as development of urban areas close to rivers, for prevention of damages in civil structures (bridges, harbors, etc), for prevention of agricultural land losses, and for the maintenance of biological diversity in the river. The most significant finding of this study is to present a conceptual and mathematical model for evolution of meandering rivers that can be used in restoration and naturalization processes. Based on the conceptual and mathematical model for meandering evolution, a methodology for river restoration process is also presented. In this methodology, main criteria for designing stream restoration are presented.

Bank retreat rate, the receding rate of bank top-lines through basal erosion and bank failure, was calculated based on the analysis of bank stability and sediment mass conservation near banks. This study indicated that the gradient of sediment transport rate in the longitudinal direction and the strength of secondary flow determined the rate of bank retreat. An analytical model was developed based on this result and applied to simulating the evolution of sine-generated meandering channels. Not like the conventional methods which assumed that the rate of bank retreat was proportional to the magnitude of excessive near-bank velocity or shear stress, this approach in simulating bank retreat and meandering channel evolution has advantages by taking into account the gradient of longitudinal velocity and the strength of secondary flow, and will avoid the estimation of the erodibility coefficient in traditional bank erosion models.

Serious scouring may take place around bridge abutments during floods, which can lead to considerable damage or structural collapse. Design of abutment footings must, therefore, rely on the estimation of maximum possible depth of scour through the bridge opening. Deterministic scour-prediction equations are based on simplified assumptions and they do not account for randomness of the scouring variables, which leads to an unavoidable risk in footing design. This study is concerned with the development of a new reliability model for scouring around bridge abutments using the resistance-loading methodology. The validity of the proposed model is tested using the Advanced First Order Second Moment (AFOSM) and the Monte Carlo simulations. In an example, it is shown that the results of the proposed model are in good agreement with these simulations. It is also observed that the execution of this model is less time-consuming than the latter approaches. A sensitivity analysis is carried out to examine the effects of coefficients of variation and probability distributions.

As an alternative to the more conventional ARMA models, we explore the potential of hidden state Markov models for simulating hydrologic droughts. In the simplest case, the hidden (unobserved) Markov process has two states, wet and dry, representing the general climate regime prevailing at the time. Each state has a particular distribution of streamflow associated with it. Conditional upon the value of the state variable at time t, the distribution of flow at time t is independent of previous flows. Autocovariance is provided for through the Markov process. A brief discussion of model properties and parameter estimation is given. The results of an application of the hidden state Markov model to drought frequency analysis are presented along with a comparison with more conventional models.

The traditional approach for determining the safety of a levee has been to find the flood depth for a storm with a design flood and to add three feet of freeboard as a factor of safety. In recent years, however, risk analysis has been used to size levees and other flood control structures in a floodplain. These risk analysis procedures, defined by the US Army Corps of Engineers (USACE), use a series of curves to determine the expected annual damage (EAD) for a computation unit called a "damage reach". The process of using risk analysis for determining the return period for a flood is not a new concept, but the procedures for evaluating the engineering and economic performance of flood control structures are still new and are under evaluation. The National Research Council (NRC) has published a report on how the current procedures for risk analysis can be improved. One recommendation for improvement to the current methods of evaluating the performance of flood control structures involves the need to consider spatial variability. A technique involving a Monte Carlo-style stochastic simulation can be run using an integrated hydrologic (HEC-1), hydraulic (HEC-RAS) and digital terrain model to evaluate the annual exceedance probability (AEP) in a floodplain using spatially-defined data. This paper will introduce methods used in floodplain delineation studies and will describe new procedures that can be used for risk analysis in floodplain delineation studies. A case study will demonstrate how these procedures can be used to evaluate the uncertainty of the floodplain boundary.

Introduction of the Point Estimate Methods (PEMs) has recently attracted some attention in the field of environmental engineering. Differently from other uncertainty analysis methods, the PEM requires a substantially smaller computational effort for a comparable degree of accuracy in the estimation of the first few statistical moments of a model output. The application of PEMs such as the Rosenblueth method or the Tsai and Franceschini method to the modeling of toxic concentrations in the Niagara River can provide significant insights into stochastic mechanisms of the transport process of selected contaminants. This paper presents the result of the uncertainty analysis on the concentrations of Chrysene and PCBs in the Niagara River. It also proposes a potential application of the uncertainty analysis to quantifying the effective risk associated with exceeding established water quality standards for contaminants of concern.