Showing papers in "Water-Resources Investigations Report in 1987"

Documentation of computer program VS2D to solve the equations of fluid flow in variably saturated porous media

Abstract: This repprt documents a computer code for solving problems of variably saturated, single-phase flow in porous media. The mathematical model of this physical process is developed by combining the law of conservation of fluid mass with a nonlinear form of Darcy's law. The resultant mathematical model, or flow equation, is written with total hydraulic potential as the dependent variable. This allows straightforward treatment of both saturated and unsaturated conditions. The spatial derivatives in the flow equation are approximated by central differences written about grid-block boundaries. Time derivatives are approximated by a fully implicit backward scheme. Nonlinear storage terms are linearized by an implicit Newton-Raphson method. Nonlinear conductance terms, boundary conditions, and sink terms are linearized implicitly. Relative hydraulic conductivity is evaluated at cell boundaries by using full upstream weighting, the arithmetic mean, or the geometric mean of values from adjacent cells. Saturated hydraulic conductivities are evaluated at cell boundaries by using distance-weighted harmonic means. The linearized matrix equations are solved using the strongly implicit procedure. Nonlinear conductance and storage coefficients are assumed to be represented by one of three closed-form algebraic equations. Alternatively, these values may be interpolated from tabulated data. Nonlinear boundary conditions treated by the code include infiltration, evaporation, and seepage faces. Extraction by plant roots is included as a nonlinear sink term. The code is written in standard ANSI Fortran. Extensive use of subroutines and function subprograms provides a modular code that is easily modified. A complete listing of data-input requirements and input and output for a one-dimensional infiltration problem and for a two-dimensional problem involving infiltration, evaporation, and evapotranspiration (plant-root extraction) are included.

Hydrogeology, chemical quality, and availability of ground water in the Upper Floridan aquifer, Albany area, Georgia

Abstract: Large withdrawals of ground water in the 1,500square-mile Albany area of southwestern Georgia have lowered water levels in deep aquifers as much as 140 feet and raised concern about the aquifers' ability to meet increasing demands. This study was conducted to evaluate the development potential of the shallow Upper Floridan aquifer as an alternate source of ground water, especially for public supply. The study area lies mainly within the Dougherty Plain district of the Coastal Plain physiographic province. The Upper Floridan aquifer is the shallowest major ground-water reservoir, generally covered by only 20 to 80 feet (ft) of overburden. The aquifer includes units of sand, clay, limestone, and dolomite of middle Eocene age and younger, that form, in ascending order, the Lisbon Formation, the Clinchfield Sand, the Ocala Limestone, and the Suwannee Limestone. The aquifer is overlain by undifferentiated sediments of Miocene age and undifferentiated overburden of Quaternary age. The Upper Floridan ranges in thickness from about 50 ft in the northwestern part of the area to more than 370 feet in the southeastern part. The Upper Floridan stores and transmits large quantities of water, mainly in a zone of high permeability in the lower part of the aquifer. The transmissivity of the aquifer ranges from less than 10,000 feet squared per day northwest of Albany, to as much as 150,000 feet squared per day south and southeast of Albany. Twenty-eight years of agricultural and industrial pumping has not produced a long-term decline of the water level in the Upper Floridan; the aquifer system remains at equilibrium. The Upper Floridan yields hard, calcium bicarbonate-type water that contains no constituents in concentrations that exceed State drinking water standards. The Upper Floridan aquifer is the primary source of irrigation, industrial, and rural domestic water supplies in the study area. The aquifer has not been developed as a public-supply source, however, largely because of concern over possible ground-water contamination by agricultural and industrial chemicals and landfill leachate. The development potential of the aquifer as a public-supply source depends on the quantity and the chemical quality of water available to wells. Near Albany, active and abandoned landfills, industrial and commercial sites, railroad yards, and gasoline and chemical storage tanks are potential sources of contaminants and, thus, make the area unsuitable for well sites. The areas of high transmissivity southeast of Albany, east of the Flint River, and southwest of Albany, west of the river, have the greatest development potential for public water supply. East of the river, yields of 12to 16-inch-diameter wells reportedly exceed 2,000 gallons per minute (gal/min). West of the river, yields of 800 to 1,200 gal/min can be sustained by wells that tap the lower part of the aquifer, and some wells are reported to produce more than 2,500 gallons per minute. In these areas, it may be possible to develop several fields of properly spaced wells capable of supplying tens of millions of gallons of potable water per day without overstressing the aquifer. In most of the study area, contaminants applied to or spilled on the land surface eventually can be expected to percolate through the overburden and reach the aquifer. Thus, it is important that wells be sited away from areas that have been used for the storage and disposal of potential contaminants and, probably to a lesser extent, the application of agricultural chemicals. In the northern part of the study area, the upper part of the Upper Floridan aquifer acts as a leaky confining unit, and wells that derive water exclusively from the lower part of the aquifer probably would have added protection against contaminants that penetrate the overburden. To the south, the confining unit is missing and the entire aquifer is permeable; contaminants on the land surface that percolate through the overburden and reach the aquifer in this area are more likely to be drawn into a pumped well. In the area of greatest development potential east of the Flint River, wells may penetrate major groundwater conduits. Where this occurs, contamination from distant sources that recharge the aquifer, such as losing streams, is possible, because conduit flow in the aquifer is comparatively rapid and contaminants can be transported long distances without natural filtration or purification. Water in some conduits could become turbid, especially during wet periods, and cause quality problems. Also, wells located near the river could draw river water into the aquifer. In karst terrane, such as the Dougherty Plain, drawing the water level in the aquifer down below the top of limestone by pumping could initiate sinkhole development. By limiting drawdown during well development and during production, the likelihood of causing sinkholes to form can be minimized.

Ground-water resources of the Laura area, Majuro Atoll, Marshall Islands

Abstract: ----------------------------------------------------------

Hydrologic and geochemical monitoring in Long Valley Caldera, Mono County, California, 1985

Abstract: Hydrologic and geochemical monitoring, by the U.S. Geological Survey, to detect changes caused by magmatic and tectonic processes in the Long Valley caldera has continued through 1985. The monitoring included the collection of the following types of data: chemical and isotopic composition of waters and gases from springs, wells, and steam vents; temperatures in wells, springs, and steam vents; flow rates of springs and streams; water levels in wells; and barometric pressure and precipitation at several sites. In addition, reservoir temperatures for the geothermal system were estimated from computations based on chemical geothermometers applied to fluid samples from wells and springs. Estimates of thermal water discharged from springs were made on the basis of boron and chloride fluxes in surface waters for selected sites in the Casa Diablo area and along the Mammoth-Hot Creek drainage. These data are presented in tables and graphs. The Long Valley area was relatively quiescent throughout 1985 in terms of geodetic changes and seismic activity. As a consequence, the hydrologic system varied mainly in response to seasonal influences of temperature, atmospheric pressure, and precipitation. However, spring flows near Casa Diablo were influenced by pumping at the geothermal production well field nearby.

Hydrologic and geologic factors affecting land subsidence near Eloy, Arizona

Abstract: At an extensometer site near Eloy, Arizona, 1.09 meters of land subsidence caused by ground-water withdrawal were measured by leveling in 1965-83. The extensometer, which partially penetrates the compressible sediments, recorded 0.82 meter of compaction during the same period. By use of a one-dimensional model, cumulative daily compaction values were simulated to within an average of 0.0038 meter of the actual values. Land subsidence was simulated to within an average of 0.011 meter using the same model in conjunction with geohydrologic data of the sediments below the extensometer. A highly compressible clay layer that is 24.38 meters thick was partially penetrated by the extensometer. The simulation indicated that the layer was driving compaction and land subsidence linearly with respect to time, despite the presence of other compacting layers. Because of its thickness and compressibility, this layer can be expected to continue to compact after applied vertical stresses have stopped increasing and other layers have stopped compacting. The sensitivity analysis indicated that the compressibility of fine-grained sediments (expressed as specific storage) is one of the factors to which compaction is most sensitive. The preconsolidation stress and hydraulic conductivity also affect land subsidence near Eloy, Arizona.

Ground-penetrating radar study of the thickness and extent of sediments beneath Silver Lake, Berlin and Meriden, Connecticut

Abstract: A short-pulse ground-penetrating radar system was used to determine the extent and thickness of organic-rich lake-bottom sediments in Silver Lake in south-central Connecticut. Four miles of ground-penetrating radar profiles were obtained along traverses of the frozen lake during March 1984. The radar waves penetrated 6 inches of snow, 1 foot of ice, an average of 4 to 5 feet of water, and 5 feet of soft organic and inorganic deposits. A large area of the lake bottom is underlain by soft sediment that exceeds 5 feet in thickness. No radar reflections were obtained from the hard sub-bottom in these areas because the overlying sediment likely contains large proportions of silt and clay. Coring along two radar profile lines confirmed the depths of soft sediment calculated from the radar data. Boring logs around the perimeter of the lake indicate that the eastern side may be underlain by till or poorly sorted sand and gravel, and that the rest of the lake is probably underlain by fine sand and silt, with some discontinuous layers of sand and gravel.

Hydrogeology and geochemistry of the unsaturated zone, radioactive waste management complex, Idaho National Engineering Laboratory, Idaho

Abstract: To assess the potential migration of low-level radioactive waste in the shallow subsurface, a study on the geochemistry of the unsaturated zone at the Radioactive Waste Management complex (RWMC), Idaho National Engineering Laboratory, on the eastern Snake River Plain in southeastern Idaho was done. Stable isotope and chemical data suggest that the perched water obseved beneath the RWMC is not due to vertical infiltration through the ground surface, but is due to lateral flow of water that infiltrated through the diversion ponds. The water accumulates as a perched mount on the thick, laterally continuous sedimentary interbed at a depth of 73 meters (m) and then moves about 1.5 kilometers to the northweast beneath the RWMC. Infiltrating water can move clay, silt, and sand downward through sedimentary material and open fracturs, at least to the interbed at a depth of 73 m. Oxygen isotope exchange and clay mineral alteration caused by extruded lava have been observed in the upper 0.86 m of the sedimentary interbed at a depth of 34 m an in the upper 2.65 m of the sedimentary interbed at a depth of 73 m. An examination of the sediment-basalt interrelation shows that the flows overlying the interbed atmore » a depth of 73 m are substantially thicker than the flows overlying the interbed at a depth of 34 m (16 to 23 m comapred to 6 to 10 m). Sedimentary material at the RWMC shows isotopic and soils evidence of at least two major climatic changes within the last 200,000 years. 65 refs., 18 figs., 19 tabs.« less

Cenozoic stratigraphy and geologic history of the Tucson Basin, Pima County, Arizona

Abstract: .......................................................... 1

AQMAN; linear and quadratic programming matrix generator using two-dimensional ground-water flow simulation for aquifer management modeling

Abstract: 7 Introduction 8 Ground-Water Management Modeling 10 Background 10 The response matrix method 10 General Problem Formulation for Linear and Quadratic Programs 13 Linear and quadratic objectives 13 Constraints and system linearity 14 Time parameters -15 Stress and response 16 Drawdowns 16 Gradients and velocities 16 Head, gradient, and velocity definitions 18 The MPS file and the solution 19 Problem Formulation with AQMAN 20 Objective function 20 Linear objective 20 Quadratic objective 20 Constraint set 23 Pumping and recharge constraints 23 Head constraints 24 Gradient and velocity constraints 26 Head and velocity definitions 30 Nonlinearities 34 Time parameters 34 The unit stress and scaling 36 Quadratic objective 37 The MPS file 37 User changes 37 MPS format conventions 37 Naming conventions 41 Size of the MPS file 42 Program Description 43 A^aiu \" \" \" \"\" \" \" \"\"\" \"\" \"\"\" \" \"\" \"\" \" \"\"\" \"\"\" \"\"\" \" \"\"\"\" \"\"\" \"\" \" \"\"\"\" \"\" \" \"\" \"\"\"\"»\" \"U Subroutine PRE 43 Subroutine CHKDAT---------------------------------45 Subroutine GRADS 45 Subroutine MPSFMT . . _.45 Subroutine QUAD 45 Subroutine READ1 46 Subroutine WRITE 1 46 Data Files 47 Sample Problem 48 References 55 Appendix I, Definition of Variables57 Appendix II, Date File Instructions 61 Appendix III, Quadratic Objective Subroutine FUNOBJ67 Appendix IV, Data Files for Sample Problems76 Appendix V, AQMAN Program List 94

A procedure for estimating reaeration coefficients for Massachusetts streams

Abstract: An equation for estimating stream reaeration coefficients from easily measured physical characteristics was developed for moderately sloped streams in Massachusetts. To define the equation, multiple-regression techniques were applied to 30 data sets containing 9 physical, hydrologic, and water-quality characteristics. Data on mean depth, water-surface slope, and mean streamflow velocity ranged from 0.4 to 6.3 feet, from 0.00017 to 0.015 feet per foot, and from 0.13 to 2.15 feet per second, respectively. Reaeration coefficients were measured using a steady-state, propane-gas tracer technique during mediumand low-flow periods in 1983 and 1984. Measured reaeration coefficients ranged from 0.4 to 67.7 base e units per day at 20 degrees Celsius. The regression analysis defines the relation between the reaeration coefficient, K2 , and the independent variables mean depth, D, water-surface slope, SL, and mean streamflow velocity, V, given by: K2 = 252.2 D~°' 176 70-355 5Lo-43 8j in which 252.2 is the regression constant for the equation. The equation was limited to three variables that were the most significant at the 95-percent confidence level because of the small size of the data base. The standard error of estimate for the reaeration equation is 37 percent. An error-analysis technique was used to compare the proposed equation with 19 published reaeration coefficient equations. The error analysis indicates that the proposed equation has the second lowest error (37 percent) for 20 stream reaches with slopes greater than 0.002 feet per foot. INTRODUCTION Stream reaeration coefficients are used in stream-water quality models to forecast the effects of organic loadings on DO (dissolved-oxygen) concentrations in streams. State water-pollution-control agencies, such as the MDWPC (Massachusetts Department of Environmental Quality Engineering, Division of Water Pollution Control) rely on stream DO models for decisions relative to the maintenance of stream water-quality standards set by each state in conjunction with the U.S. Environmental Protection Agency. The self-cleaning capacity of a river depends on DO concentrations and the capacity to replace oxygen removed by the reduction of organic wastes. When attempting to model the concentration levels of a nonconservative substance such as DO in open-channel flow, allowances must be made for dispersion, decay, and reaeration at the surface, and for deoxygenation resulting from biochemical demands, algal respiration, and interaction with benthic deposits. The use of a steady-state river model is based on the assumption that the dispersion coefficient is constant for a given stream reach over the time period being simulated. All of the other processes can be measured directly, except for reaeration, which can be determined indirectly or estimated from equations. Measured values for the reaeration coefficient can be determined from the dissolved-oxygen balance, distributed equilibrium, and by tracer methods. The dissolved-oxygen balance method consists of measuring the various sources and sinks of DO and determining the amount of reaeration needed to balance the equation. The disturbed-equilibrium method consists of artificially producing DO deficits by adding sodium sulfite to the stream and subsequently measuring upstream and downstream concentrations of DO at two different concentration levels. The tracer method consists of correlating the rate of desorption of a tracer gas with the rate of absorption of oxygen. However, in addition to being costly and time consuming, the balance and equilibrium methods are indirect determinations of oxygen transfer and are subject to measurement errors. These indirect methods of calculating the DO contribution from reaeration may be no more accurate than reaeration coefficients calculated from theoretical or empirical equations (Bennett and Rathbun, 1972). Recent advances in the tracer method, such as the U.S. Geological Survey's steady-state propane-gas tracer method (Yotsukura and others, 1983; 1984), offer a less costly, more accurate, reliable, and reproducible method of measuring reaeration coefficients in place. The gas-tracer method is useful for determining reaeration coefficients, because it also eliminates interferences of photosynthetic oxygen production and respiration of the suspended and attached aquatic plants (Bennett and Rathbun, 1972; Rathbun and others 1978; Yotsukura and others, 1983; 1984). The reaeration coefficients used in predictive models generally are estimated from theoretical, semiempirical, or empirical equations. Bennett and Rathbun (1972) state that the theoretical models of the dissolved-oxygen absorption process generally are not suitable to predict the reaeration coefficient in streams because the model parameters have been inadequately related to bulk-flow hydraulic variables. Semiempirical and empirical equations developed from experimental data predict reaeration coefficients adequate for streams of the type on which the equations are based, but large errors may occur when the equations are applied to other types of streams or to conditions outside the range of independent variables considered in the equations derived from empirical data. Because of stringent water-quality standards imposed on State streams and the high costs of determining stream-reaer4tion coefficients, there is a need to develop an equation that will estimate reaeration coefficients reliably from easily measured physical, hydraulic, and water-quality characteristics. In response to this need, the U.S. Geological Survey, in cooperation with the MDWPC, has developed regionalized predictive equations using the Survey's steady-state propane-gas tracer method. This gas-tracer method provides the data needed to derive a reaeration.-estimating equation. Purpose and Scope igression This report describes the multiple-re an equation for estimating reaeration The equation is based on easily measured tics of stream channels. The equation is com estimating equations by using measured data to the equations to predict reaeration coefficients coefficients physical pared techniques used to derive of Massachusetts streams, and hydraulic characteristo other commonly used determine the accuracy of all in Massachusetts streams. Tracer studies were performed on 16 stream reaches representative of most streams in Massachusetts. The Survey's newly developed steady-state, gas-tracer technique was was used to measure in-situ reaeration coefficients. Reaeration coefficients, mean streamflow velocities, and nine easily measured physical, hydraulic, and water-quality characteristics were measured in 30 studies during mediumand low-flow periods between August 1983 and December 1984. Approach As a result of searching reaeration literature, nine physical, hydraulic, and water quality characteristics were selected to be correlated with the reaeration coefficient: Water-surface slope, mean velocity, depth, width, roughness coefficient, color, methylene blue active substances concentration, specific conductance, and suspended-solids concentration. Sixteen reaches on 11 rivers (fig. 1) were selected for combined time-oftravel and reaeration-tracer studies on the basis of their regional location, consistency of reach characteristics, and accessibility. Reach and study-site descriptions can be found in Appendix A. Thirty combined tracer studies were conducted on the 16 reaches. When possible, two or more tracer studies were performed on the same reach at different discharge rates. All studies were conducted at a steady discharge with little or no wind. Reach physical, hydraulic, and water-quality characteristics were measured during each tracer study and samples were collected for water-quality analysis. Water-surface slope was determined using differential leveling between reference points established at the ends of each study reach. All studies were initially evaluated for completeness and accuracy of time-of-travel data. Initial propane-gas desorption coefficients were determined using steady-state, gas-tracer injection techniques outlined in Yotsukura and others (1983). For each tracer study, the transfer of measurement error was estimated. The reaeration-coefficient estimating equation was developed from reaeration coefficients and the corresponding reach characteristics determined for the 30 combined tracer studies. Step-forward, multiple-regression analyses were conducted to relate reaeration coefficients to channel characteristics. Only those characteristics significant at a 95-percent confidence level were retained. Reaeration coefficients were estimated from widely used equations using the channel characteristics determined for all 30 tracer studies. All estimating equations, including the equation developed for this report, were ranked according to the accuracy of their predicted reaeration values compared to measured values from the 30 tracer studies. Acknowledgments The authors wish to acknowledge and thank the many persons and organizations who have kindly given their time, information, and guidance to this study in particular, personnel from the Massachusetts Division of Water Pollution Control's Westborough, Massachusetts, office who assisted with the W e st B ra n ch N o rt h R iv e r ne ar G ri sw o ld vi lle A ss a b e t R iv er n ea r W es t C o n co rd S ud bu ry R iv er at C o n co rd W e st B ra n ch N o rt h R iv er ne ar A da m sv ill e N o rt h R iv er a t G ri sw o ld vi lle M id dl e B ra nc h W e st fie ld R iv er / ne ar M id d le fie ld ^ ^ 4 2 3 0 / \ -, -> M id dl e B ra nc h W e st fie ld R iv er at N o rt h C h e st e r W es t B ra nc h W e st fie ld R iv e r at C h e st e r A b e rj o n a R iv er

Hydrogeology of the Socorro and La Jencia Basins, Socorro County, New Mexico

Abstract: Every fall since 1950, the New Mexico Geological Society (NMGS) has held an annual Fall Field Conference that explores some region of New Mexico (or surrounding states). Always well attended, these conferences provide a guidebook to participants. Besides detailed road logs, the guidebooks contain many well written, edited, and peer-reviewed geoscience papers. These books have set the national standard for geologic guidebooks and are an essential geologic reference for anyone working in or around New Mexico.

Nonpoint-source agricultural chemicals in ground water in Nebraska; preliminary results for six areas of the High Plains Aquifer

Abstract: This report describes the reconnaissance phase of a study to determine the occurrence of agricultural chemicals from nonpoint sources in ground water in six areas, which are representative of the major provinces of the High Plains aquifer in Nebraska. Nitrate and triazine-herbicide concentrations in ground water were assessed to establish preliminary relations between these constituents and selected hydrogeologic, climatic, and land-use variables. In 1984, water from 82 wells in the 6 study areas was analyzed for nitrate, and water from 57 of the 82 wells was analysed for triazine herbicides. Data for 9 of the 21 independent variables suspected of affecting concentrations of nitrate and triazine herbicides in ground water were compiled from the 82 well sites. The variables and their ranges are: hydraulic gradient (XI), 0.0006-0.0053; hydraulic conductivity (X2), 5-149 feet per day; specific discharge (X3), 0.0128-0.2998 foot per day; depth to water (X4), 3-239 feet; well depth (X5), 40-550 feet; annual precipitation (X6), 12.0-39.3 inches; soil permeability (X7), 0.76-9.0 inches; irrigation-well density (X8), 0-8 irrigation wells per square mile; and annual nitrogen fertilizer use (X9), 0-260 pounds of nitrogen per acre. Nitrate concentrations ranged from less than 0.1 to 45 milligrams per liter as nitrogen. Triazine-herbicide concentrations were detected in samples from five of the six study areas in concentrations ranging from less than 0.1 to 2.3 micrograms per liter. Statistical tests indicated that there were significant differences in nitrate concentrations among the six study areas, while no significant differences in triazine-herbicide concentrations were found. Concentrations of nitrate and triazine herbicide were determined, by use of nonparametric statistics, to be significantly larger in more intensively irrigated areas than in less intensively irrigated areas. Preliminary correlations with the independent variables and nitrate concentrations indicated significant relations at the 95-percent confidence level with variables X2, X5, and X8. Correlations with triazine-herbicide concentrations indicated significant relations with variables X2, X3, X5, X6 and X8, and with nitrate concentrations (X10). By use of a simple multiple-regression technique, variables X5, X8, and X9 explained about 51 percent of the variation in nitrate concentrations. Variables X3 and X5 explained about 60 percent of the variation in triazine-herbicide concentrations. With the addition of nitrate concentration as an independent variable, two variables, X10 and X3, explained 84 percent of the total variation in triazine-herbicide concentrations.

Method for estimating the magnitude and frequency of floods at ungaged sites on unregulated rural streams in iowa

Abstract: This report provides techniques and procedures for estimating the probable magnitude and frequency of floods at ungaged sites on Iowa streams. Physiographic characteristics were used to define the boundaries of five hydrologic regions. Regional regression equations that relate the size of the drainage area to flood magnitude are defined for estimating peak discharges having specified recurrence intervals of 2, 5, 10, 25, 50, and 100 years. Regional regression equations are applicable to sites on streams that have drainage areas ranging from 0.04to 5,150 square miles provided that the streams are not affected significantly by regulation upstream from the sites and that the drainage areas upstream from the sites are not mostly urban areas. Flood-frequency characteristics for the mains terns of selected rivers are presented in graphs as a function of drainage area. INTRODUCTION This report represents the fourth update in the last 33 years of the magnitude and frequency of floods in Iowa. Flood reports were updated to provide more dependable flood-data and more accurate and reliable methods for estimating the magnitude and frequency of floods. These estimates are needed to implement efficient flood-plain management strategies and for economical design of highway structures, levees, and buildings in the flood plain. Economic design criteria require the availability of adequate data with long periods of record, improved analytical methods, and better understanding of the flood hydrology of Iowa. The first of three previous reports (Schwob, 1953) was limited to an analysis of the magnitude and frequency of floods that was based on data collected in Iowa before 1950. A second report (Schwob, 1963) included updated data and a method for estimating magnitude and frequency of floods at sites on ungaged rural streams that had drainage areas of 10 square miles or more. The third report (Lara, 1973) was prepared using data updated through 1972. This report also included a method for estimating the magnitude and frequency of floods at sites on ungaged rural streams that had drainage areas of two square miles or more. The purpose of this report is to present a simple and effective method to estimate the magnitude and frequency of floods at ungaged sites on unregulated rural streams in Iowa. The regional flood-frequency equations and techniques presented in this report, which were defined from an updated data set, should provide flood estimates with increased reliability compared to previous reports. REGIONAL ANALYSIS Methods of estimating flood magnitudes ind frequencies applicable to an entire region rather than to a single gaging station are developed through regional analysis. Many structures ire built across or adjacent to streams at sites where there is no record of streamflow. For this reason, methods are needed to extend information pertaining to flood magnitude and frequency based on gaging-station data from gaged to ungaged sites. Flood data for a single station are relatively short-term random samples and may not be representative of the long-term distribution of floods at that station. Combining records for stations in a hydrologically similar area decreases errors associated with relatively short-term nonrepresentative samples. The magnitudes of floods in Iowa vary considerably from one region to another as a function of drainage basin efficiency. River basins with minimal drainage efficiency, such as those in north-central Iowa, are characterized by flat terrain. Streams draining these basins have considerably smaller peak discharges than do streams draining basins having steep terrain and well developed drainage systems, such as the basins in the Paleozoic Plateau Escarpment area of eastern Iowa (Prior, 1976). Typically, the discharge per square mile of a stream in the escarpment area is about 1230 cubic feet per second during a 100-year-flood, whereas the discharge per square mile of a stream in ftorth-central Iowa is about 230 cubic feet per second during a 100-year flood. These two areas are about 100 miles apart and the climatic differences such as temperature, average precipitation, and relative humidity are not significant enough to account for the differences in flood magnitude. However, the physiographic differences between these two areas are significant (Prior, 1976, p.23, fig. 22). The largest floods per unit area occur within 100 miles west of the Des Moines Lobe (north-central Iowa) along the rugged bluffs and steep ridges that border the Missouri River valley., Flood data collected in this area from both Iowa and Nebraska indicate that the discharge of a stream draining a 1-square-mile basin is as much as 1880 cubic feet per second during a 100-year flood. Therefore, it seems reasonable to delineate hydrologic regions based on the landform and physiographic characteristics of the State. Five hydrologic regions have been identified and delineated in Iowa using physiographic regions of Iowa as a guide. Prior (1976) gives a detailed description of the physiographic regions of Iowa, discussing the shape of the land surface, materials that[underlie the land surface, and the geologic history. 2 «*««. l/ Hvdrologic Region 1 Hydrologic region 1 (fig. 1) extends north and south along the bluffs that border the Missouri River valley, with limits approximating those of the physiographic area known as the Western Loess Hills (Prior, 1976). The landscape has a corrugated appearance of alternating waves and troughs. Hills are sharp-featured, with narrow broken ridge-crests, intersecting spurs, and steep-sided slopes; the landscape is conducive to rapid runoff. The western border of the region is well defined and easily distinguished on topographic maps and in the field. The eastern border is more difficult to define and merges gradually with the landscape of hydrologic region 2. Hvdrologic Region 2 The bluff area that borders the Mississippi River valley is typical of the landscape in hydrologic region 2 (fig. 1). The landscape can vary from rugged to rolling topography, where runoff may be rapid, commonly causing flash flooding. Bluff-like areas are not only located in the vicinity of the Missisippi River, they also are present along the divide between the Mississippi River and Missouri River basins; in parts of the Iowa and Cedar River basins, in areas that border the Western Loess Hills, and in the headwater parts of basins of streams in south-central Iowa. Hvdrologic Region 3 Hydrologic region 3 is the largest hydrologic region (fig. 1). Most of the area in this region is typical of landscapes in Iowa. The topography of this region can be described as steeply to gently rolling hills interspersed with areas of more subdued topography. The area has a well-established drainage system. Physiographically, it covers most of the lowan Surface, a large part of the Southern Iowa Drift Plain, and the Northwest Iowa Plains (Prior, 1976). Hvdrologic Re2ion 4 This hydrologic region, which is located in west-central Iowa (fig. 1), is characterized by level terrain and a poorly developed drainage system. The region coincides approximately with the southern two-thirds of the Des Moines Lobe (Prior, 1976). Many clusters of ponds and marshes with no drainage outlets are present in this region. Small streams in level areas are shallow and sluggish. Hvdrologlc Region 5 This hydrologic region in north-central Iowa (fig. 1) coincides approximately with the northern part of the Des Moines Lobe (Prior, 1976). The magnitude of floods in this region are the smallest per unit area in the State. This is due to the flat topography and flood-attenuating effect of abundant bogs, swales, and circular depressions.

Chemical characteristics of water in the surficial aquifer system, Dade County, Florida

Abstract: ...................................................................

Technique for estimating flood-peak discharges and frequencies on rural streams in Illinois

Abstract: Flood-peak discharges and frequencies are presented for 394 gaged sites in Illinois/ Indiana, and Wisconsin for recurrence intervals ranging from 2 to 100 years. A technique is presented for estimating flood-peak discharges at recurrence intervals ranging from 2 to 500 years for nonregulated streams in Illinois with drainage areas ranging from 0.02 to 10/000 squares miles. Multiple-regression analyses/ using basin characteristics and peak streamflow data from 268 of the 394 gaged sites/ were used to define the flood-frequency relation. The most significant independent variables for estimating floodpeak discharges are drainage area/ slope/ rainfall intensity/ and a regional factor. Examples are given to show a step-by-step procedure in calculating a 50-year flood for a site on an ungaged stream/ a site at a gaged location/ and a site near a gaged location. INTRODUCTION The purpose of this report is to provide updated station flood-peak discharges and frequencies and to provide improvement to the previous techniques for estimating flood-peak discharges and frequencies of floods for sites on most streams where flood discharges are not significantly affected by regulation or urbanization. Flood-peak discharges and frequencies are presented for 394 gaging stations in Illinois/ Indiana/ and Wisconsin for recurrence intervals of 2/ 5/ 10/ 25/ 50/ and 100 years. A technique using drainage area (A)/ slope (S)/ rainfall intensity (I)/ and regional factor (Rf) was developed for estimating flood-peak discharges at ungaged sites in Illinois. Equations using these variables are applicable for estimating flood-peak discharges for recurrence intervals of 2 to 500 years for drainage areas ranging from 0.02 to 10/000 square miles (mi 2 ) on nonregulated rural streams. Estimates of future floods are necessary for the proper design of engineering projects such as bridges/ culverts/ highways/ and flood-control structures; for establishment of actuarial flood-insurance rates; and for proper flood-plain management by State and local agencies. Previous techniques for estimating flood-peak discharges and frequencies in Illinois have been provided by Mitchell (1954)/ Speer and Gamble (1965)/ Wiitala (1965)/ Patterson and Gamble (1968)/ Ellis (1968)/ Cams (1973)/ Curtis (1977a)/ and Alien and Bejcek (1979). Techniques were developed by Cams (1973)/ Curtis (1977a)/ and Alien and Bejcek (1979) using ordinary least squares multiple-regression analyses as recommended by Thomas and Benson (1970). Additional data and improved analytical methods used in this report increase the confidence in estimating techniques over those published in earlier reports. This report was prepared under a cooperative agreement between the State of Illinois, Department of Transportation, Division of Water Resources, and the U.S. Geological Survey (Survey). Streamflow data were collected in cooperation with the U.S. Army Corps of Engineers and State and local agencies. TECHNIQUE FOR ESTIMATING FLOOD-PEAK DISCHARGES Annual peak discharges from gaging stations having a minimum of 10 years of record through the 1985 water year were used to define station floodfrequency relations. Locations of these stations are shown in figures 1 and 2. The map number, identification number, name, geographic location, and station flood-peak discharges for the stations are listed in table 1. All figures and tables are grouped in the back of the report for easy reference. Station flood-frequency relations were defined using the Hydrology Subcommittee of Interagency Advisory Committee on Water Data (1982), formerly U.S. Water Resources Council, guidelines. These guidelines outline procedures to fit the logarithms of observed annual peak discharges to the Pearson Type III frequency distribution. Peak discharges of various recurrence intervals and basin characteristics for gaging stations were used in multiple-regression analyses to develop estimating equations for flood-peak discharges and frequencies on nonregulated rural streams in Illinois. Data from stations affected by either urbanization or by regulation were not included in the regression analyses. Relations were developed for estimating flood-peak discharges corresponding to the 2-, 5-, 10-, 25-, 50-, 100-, and 500-year recurrence-interval flood (T-year flood or Qrp). The regression analyses indicated that the independent variables drainage area (A), slope (S), rainfall intensity (I), and regional factor (Rf) are the most significant variables to use in estimating flood-peak discharges for Illinois streams. One estimating equation for each recurrence interval provides a straightforward technique to compute flood-peak discharges for both small and large Illinois streams. Flood-peak discharges and frequencies, basin characteristics and other pertinent data, and regional factors are tabulated in tables 1, 2, and 3. In table 1, two sets of station flood-peak discharges are presented for two stations (Nos. 385 and 389) on the Big Muddy River. The first set of discharges are for periods of nonregulated flow and were used in the regression analyses. The second set of discharges are for periods of regulated flow and were not used in the regress: .on analyses. The reliability of flood-frequency estimates is uncertain for very large recurrence intervals. Because of this uncertainty, the 500-year flood discharges are omitted from table 1. An estimating equation for the 500-year flood is provided primarily for planners who are required to compute this event for special purposes such as flood-insurance studies. Only those stations used in the regression analyses are listed in table 2. The flood-frequency and the regression analyses, used to develop the estimating technique, are defined in detail in the data-analyses section. Flood-peak discharge equations, applicable statewide, for estimating Qrp on nonregulated rural streams are as follows: Q2 40.3 A0 ' 790 S°' 481 (1-2.5)°' 677 Rf (1) Q5 66.4 A0 ' 786 S°' 513 (1-2.5)°' 719 Rf (2) Q 10 83.0 A0 ' 785 S°' 532 (1-2.5)°' 742 Rf (3) Q25 = 103 A0 * 786 s°* 552 (1-2.5)°' 768 Rf (4) Q50 = 118 A0 * 786 s°* 566 (1-2.5)°' 786 Rf (5) Q 100 132 A0 ' 787 S°' 578 d-2.5)°' 803 ^ (6) Q500 = 162 A0 ' 789 S°' 601 (1-2.5)°' 838 Rf (7) The four variables required to solve the equations are drainage area (A), slope (S), rainfall intensity (I), and regional factor (Rf). Drainage area and slope are determined from topographic maps. Drainage area is the area contributing to surface runoff. Slope is determined between points 10 percent and 85 percent of the total distance measured along the low-water channel from the site to the basin divide. The rainfall intensity is determined from figure 3. The regional factor is determined by first selecting the region number from figure 4 and then the appropriate regional factor from table 3. Flood-peak discharge equations for recurrence intervals between 2 and 100 years, other than those in equations 1 to 7, may be developed by interpolating the regression constant and coefficients from the graphs in figure 5. APPLICATION OF ESTIMATING TECHNIQUE The technique for estimating flood-peak discharges and frequencies is applicable to either ungaged or gaged nonregulated rural streams. Figure 6 shows the sequence to follow for estimating a flood-peak discharge at a site. Step-by-step procedures for applying the estimating technique are given in the examples that follow. Site on Ungaged Stream Flood frequency estimates at sites on ungaged streams are calculated using equations 1 to 7. Example 1: Computation of the 50-year recurrence interval flood at a site on an ungaged stream: 1. Determine the size of contributing drainage area (A), in square miles. The area can be planimetered on topographic, county, or other maps suitable for delineating the basin boundary. For this example, assume A = 625 mi^. 2. Determine the slope (S), in feet per mile (ft/mi). Slope is based on the difference of elevations divided by distance between points 10 percent and 85 percent of the total distance measured along the low-water channel of the stream from the site to the basin divide. For this example, assume S = 2.5 ft/mi. 3. Determine the rainfall intensity (I), in inches, from figure 3. The value of I should be an average for the basin. For this example, assume I = 3.1 inches. 4. Determine the region (R) and the regional factor (Rf) from figure 4 and table 3, respectively. For this example, R is III and Rf is 0.862. 5. Select equation 5 from page 3 and compute the flood magnitude. Q50 = 118 A0 786 s°' 566 (1-2.5)°' 786 Rf = (118)(625)°786 (2.5)°566 (3.1-2.5)°786 (0.862) = (118)(157.6)(1.68)(0.669)(0.862) = 18,000 ft 3 /s. Site at Gaged Location Flood frequency estimates at gaged sites are combinations of the gaging station frequency curve and the equation estimates. The equivalent years of record concept (Hardison, 1971) was used to obtain weighted estimates of peak flow at gaged sites using estimates obtained from station records and from equations 1 to 7. This procedure was described by the Hydrologic Subcommittee (1982) and is expressed in the equation Om Yrs of record (log sta. QT ) + Eg yrs record (log regional QT ) t Yrs of record + Eq yrs record In equation 8, station QT is obtained from the first line of discharge values in table 1 and converted to a logarithm (log). mined from table 2. The regional Qip is computed using the desired regional estimating equation on page 3 or obtained from The years of record are deterthe second line of discharge values in table 1 and then transformed into logs. The station equivalent years of record (Eq yrs record) for the equation are also given in table 2. The antilog of the result from equation 8 is the weighted estimate of the station flood discharge. Example 2: Computation of the weighted 50-year recurrence interval flood at the gaging station 05572000 Sangaraon River at Monticello, Illinois (map No. 307) : Yrs of record (log sta. Qcn) + Eq yrs record (log eq

Geohydrology and water quality of the Inyan Kara, Minnelusa, and Madison aquifers of the northern Black Hills, South Dakota and Wyoming, and Bear Lodge Mountains, Wyoming

Abstract: ...............................

Geohydrology and susceptibility of major aquifers to surface contamination in Alabama; area 6

Abstract: ...........................................................

Area of influence and zone of contribution to superfund-site wells G and H, Woburn, Massachusetts

Abstract: Groundwater contamination by chlorinated volatile organic compounds detected by the Massachusetts Department of Environmental Quality Engineering in 1979 forced the closing of public supply wells G and H in the City of Woburn, MA. The EPA has ranked the wells G and H site on the National Priorities List as a CERCLA (Superfund) site and currently is conducting a feasibility study to determine a remedial action cleanup plan for the site. A 30-day aquifer test was conducted to determine the hydraulic properties of the stratified drift aquifer in the vicinity of the wells, and to determine the area of influenced and zone of contribution to wells G and H under pumping conditions. The area of influence of wells G and H described in this report is considered to be a snapshot representative of the hydrologic and pumping conditions of the 30-day aquifer test. Most of the water pumped by the wells is obtained directly from the part of the aquifer immediately surrounding both wells and from induced infiltration of surface water from the overlying river and wetland. The remaining part of the zone of contribution is that area of the Aberjona River drainage basin upgradient and outside the area ofmore » influence of wells G and H. A small amount of the surface water in the river entering the northern end of the study area, which is derived from groundwater discharge and surface water runoff in the up-gradient drainage area, is induced from the river to the wells under pumping conditions. The size of the area south of wells G and H that contributes water to the wells is variable, dependent on pumping rates and hydrologic conditions. 14 refs., 6 figs., 1 tab.« less

Hydrogeology of stratified-drift aquifers and water quality in the Nashua Regional Planning Commission Area, south-central New Hampshire

Abstract: The Nashua Regional Planning Commission area in south-central New Hampshire is a 12community area that is experiencing increased demands for water supply because of increases in population. The study area is underlain by 129 square miles (40 percent of the area) of stratified drift which, where sufficiently saturated and permeable, form the most productive aquifers in the area. At present, eight towns use the stratified-drift aquifers for municipal water supply. The saturated thickness of stratified drift in the study area ranges from 0 or less than 20 feet near aquifer boundaries to more than 100 feet in the Souhegan and Merrimack River valleys. The transmissivity of stratified drift ranges from less than 2,000 fta/d (feet squared per day) throughout much of the area to more than 8,000 ft2/d in the communities of Amherst, Brookline, Hollis, Hudson, Litchfield, Merrimack, Milford, Nashua, and Pelham. Directions of ground-water flow are generally from valley walls to surface waters, which act as drains for the stratified-drift aquifers. At present, the estimated total yield of community water-supply systems in the study area (surface and ground water combined) is 22 Mgalld (million gallons per day). Analytical modeling indicates that an additional 12 Mgalld could be obtained from six aquifers located in the communities of Amherst, Litchfield, Merrimack, Milford, and Pelham. Other aquifers in the area, not modeled in this study, also could provide increased amounts of water especially where yields could be augmented by induced recharge of surface water. Ground-water quality in the study area is characterized by naturally elevated levels of iron and manganese. Of 32 wells sampled, 7 exceeded USEPA (US. Environmental Protection Agency) recommended drinking-water limits for both iron and manganese, and 3 wells exceeded the manganese limit only. The average total dissolved-solids concentration for 32 samples was 121 mgIL (milligrams per liter). Ground water in the area is slightly corrosive; pH's ranged from 5.0-7.3. Ground-water contamination has been detected at two USEPA "superfund" sites in the study area located in Milford and Nashua. At both sites, contamination of ground water has caused shutdown of municipal and private water-supply wells. The widespread effect of application of highway deicing chemicals on ground-water quality is reflected by sodium concentrations that average 24 mg/L throughout the study area. At 11 of 32 sites sampled, the USEPA recommended limit for sodium (20 mg/L) was exceeded.

Techniques for estimating regional flood characteristics of small rural watersheds in the plains region of eastern Colorado

Abstract: Recorded and synthetic flood data for 52 watersheds (35 in Colorado and 17 in adjoining States) were analyzed to develop regional techniques for estimating the magnitude, frequency, volume, and hydrograph shape of floods that typically occur on small rural watersheds in the plains region of eastern Colorado. The analysis of flood magnitude and frequency included 21 floodfrequency relations that were based on recorded annual peak discharges, 2 flood-frequency relations that were based on synthetic annual peak discharges, and 28 flood-frequency relations that were based on recorded and synthetic annual peak discharges (a relation could not be determined for 1 watershed). Similarly, the analysis of flood volumes included volumes for 103 recorded floods and 4,391 synthetic floods. Synthetic flood data were generated from long-term rainfall data from National Weather Service stations and a rainfallrunoff model calibrated for each watershed. The 5-, 10-, 25-, 50-, and 100-year peak discharges were regionalized using ordinary least-squares and generalized least-squares regressions. The smallest errors of prediction were obtained using the generalized least-squares regressions, and the relations developed included the independent variables of effective drainage area, relief factor, and 24-hour, 100-year rainfall intensity; standard errors of prediction ranged from 35 to 50 percent. A relation was developed to estimate flood volume from peak discharge; the standard error of prediction was 78 percent. To develop a flood hydrograph from estimates of peak discharge and flood volume, a dimensionless-hydrograph technique is presented that produces synthetic flood hydrographs very similar in shape to recorded flood hydrographs. INTRODUCTION Flood characteristics, such as magnitude of peak discharges, frequency of occurrence, and volumes are major considerations in the design of highway bridges and culverts. Extensive discharge data available for large perennial streams generally have provided flood information necessary for the design of major drainage structures. Previous reports on the estimation of flood characteristics of Colorado streams include Patterson (1964, 1965), Patterson and Somers (1966), Matthai (1968), Livingston (1970), Hedman and others (1972), McCain and Jarrett (1976), U.S. Soil Conservation Service (1975, 1977) and Kircher and others (1985). However, except for the reports by McCain and Jarrett (1976) and the U.S. Soil Conservation Service (1975, 1977), the methods described by these reports generally do not apply to very small watersheds, particularly to watersheds that have ephemeral streams. McCain and Jarrett (1976) presented regression equations applicable to watersheds that have drainage areas greater than 1 mi 2 in the plains region of eastern Colorado (fig. 1). Their equations, however, were based only on limited data for small watersheds; only 2 of 36 watersheds studied had drainage areas less than 30 mi 2 . Procedures for estimating flood characteristics described by the U.S. Soil Conservation Service (1975, 1977) apply to small watersheds, but primarily are based on empirical rainfall-runoff relations developed for regions encompassing many States, rather than local hydrologic areas within any particular State such as Colorado. In 1968, the U.S. Geological Survey, in cooperation with the Colorado Department of Highways and the Federal Highway Administration, began a study to: (1) Collect data during thunderstorm-caused floods in small (generally

pH measurements of low-conductivity waters

Abstract: pH is an important and commonly measured parameter of precipitation and other natural waters. The various sources of errors in pH measurement were analyzed and procedures for improving the accuracy and precision of pH measurements in natural waters with conductivities of < 100 uS/cm at 25 C are suggested. Detailed procedures are given for the preparation of dilute sulfuric acid standards to evaluate the performance of pH electrodes in low conductivity waters. A daily check of the pH of dilute sulfuric acid standards and deionized water saturated with a gas mixture of low carbon dioxide at partial pressure (air) prior to the measurement of the pH of low conductivity waters is suggested. 14 refs., 2 figs., 2 tabs.

An assessment of low flows in streams in northeastern Wyoming

Abstract: This report is a brief summary and assessment of low flows in the following basins in northeastern Wyoming: Little Bighorn, Tongue, Powder, Little Missouri, Belle Fourche, Cheyenne, and Niobrara Rivers, and about 200 river miles of the North Platte River and its tributaries. Only existing data from streamflow stations and miscellaneous-observation sites during the period, 1930-80, were used. Data for a few stations in Montana and South Dakota were used in the analysis. Data were available for 56 perennial streams, 38 intermittent streams, and 34 ephemeral streams. The distribution of minimum observed flows of record at all stations and sites and the 7-day 10-year low flows at mountain stations and main-stem plains stations are shown on a map. Seven-day low flows were determined by fitting the log Pearson Type III distribution to the data; results are tabulated only for stations with at least 10 years of record that included at least one major drought. Stations installed since about 1960 are considered not to have included a major drought. Most streams that originate in the foothills and plains have no flow during part of every year, and are typical of much of the study area. For stations on these streams, the frequency of the annual maximum number of consecutive days of no flow was determined, as an indicator of the likelihood of extended periods of no flow or drought. For estimates at ungaged sites on streams in the Bighorn Mountains only, a simple regression of 7-day 10-year low flow on drainage area has a standard error of 64 percent, based on 19 stations with drainage areas of 2 to 200 square miles. The 7-day 10year low flow in main-stem streams can be interpolated from graphs of 7-day 10-year low flow versus distance along the main channel. Additional studies of low flow are needed. The data base, particularly synoptic baseflow information, needs considerable expansion. Also, the use of storage-analysis procedures should be considered as a means of assessing the availability of water in streams that otherwise are fully appropriated or that are ephemeral.