Partial List of Abstracts, 1996

Dissolved organic carbon (DOC) in Lake Fryxell, 10 streams flowing into the lake, and the moat surrounding the lake was studied to determine the influence of sources and biogeochemical processes on its distribution and chemical nature. Lake Fryxell is an amictic, permanently ice-covered lake in the McMurdo Dry Valleys which contains benthic and planktonic microbial populations, but receives essentially no input of organic material from the ahumic soils of the watershed. Biological activity in the water column does not appear to influence the DOC depth profile, which is similar to the profiles for conservative inorganic constituents. DOC values for the streams varied with biomass in the stream channel, and ranged from 0.2 to 9.7 mg C/L. Fulvic acids in the streams were a lower percentage of the total DOC than in the lake. These samples contain recent carbon and appear to be simpler mixtures of compounds than the lake samples, indicating that they have undergone less humification. The fulvic acids from just above the sediments of the lake have a high sulfur content and are highly aliphatic. The main transformations occurring as these fractions diffuse upward in the water column are 1) loss of sulfur groups through the oxycline and 2) decrease in aliphatic carbon and increase in the heterogeneity of aliphatic moieties. The fraction of modern ¹⁴C content of the lake fulvic acids range from a minimum of 0.68 (approximately 3000 years old) at 15m depth to 0.997 (recent material) just under the ice. The major processes controlling the DOC in the lake appear to be: 1) The transport of organic matter by the inflow streams resulting in the addition of recent organic material to the moat and upper waters of the lake; 2) The diffusion of organic matter composed of relict organic material and organic carbon resulting from the degradation of algae and bacteria from the bottom waters or sediments of the lake into overlying glacial melt water, 3) The addition of recent organic matter to the bottom waters of the lake from the moat.

Wyoming is now at the eastern margin of westerlies originating in the Pacific, but in the Pleistocene appears to have received moisture from elsewhere, possibly the Gulf of Mexico. Oxygen isotope ratios of pedogenic carbonate in postglacial terraces correspond to ratios in equilibrium with present meteoric waters, which show a strong relation to precipitation seasonality and storm sources. In contrast, the oxygen isotope ratios of all pre-Holocene soils are significantly more positive, even though the carbon isotope composition of coexisting organic matter suggests that the carbonate formed in temperatures cooler than today. The oxygen isotope ratios of paleowaters in mid - North America appear to be more useful for identifying past storm sources than for estimating paleotemperatures.

Nonlinear regression is useful in ground-water flow parameter estimation, but problems of parameter insensitivity and correlation often exist given commonly available hydraulic-head and head-dependent flow (for example, stream and lake gain or loss) observations. To address this problem, advective-transport observations are added to the ground-water flow, parameter-estimation model MODFLOWP using particle-tracking methods. The resulting model is used to investigate the importance of advective-transport observations relative to head-dependent flow observations when either or both are used in conjunction with hydraulic-head observations in a simulation of the sewage-discharge plume at Otis Air Force Base, Cape Cod, Massachusetts, USA. The analysis procedure for evaluating the probable effect of new observations on the regression results consists of two steps: (1) parameter sensitivities and correlations calculated at initial parameter values are used to assess the model parameterization and expected relative contributions of different types of observations to the regression; and (2) optimal parameter values are estimated by nonlinear regression and evaluated. In the Cape Cod parameter-estimation model, advective-transport observations did not significantly increase the overall parameter sensitivity; however: (1) inclusion of advective-transport observations decreased parameter correlation enough for more unique parameter values to be estimated by the regression; (2) realistic uncertainties in advective-transport observations had a small effect on parameter estimates relative to the precision with which the parameters were estimated; and (3) the regression results and sensitivity analysis provided insight into the dynamics of the ground-water flow system, especially the importance of accurate boundary conditions. In this work, advective-transport observations improved the calibration of the model and the estimation of ground-water flow parameters, and use of regression and related techniques produced significant insight into the physical system.

Trends in precipitation and surface water chemistry at a network of 15 small watersheds (<10 km²) in the USA were evaluated using a statistical test for monotonic trends (the seasonal Kendall test) and a graphical smoothing technique for the visual identification of trends. Composite precipitation samples were collected weekly and surface water samples were collected at least monthly. Concentrations were adjusted before trend analysis, by volume for precipitation samples and by flow for surface water samples. A relation between precipitation and surface water trends was not evident either for individual inorganic solutes or for solute combinations, such as ionic strength, at most sites. The only exception was chloride, for which there was a similar trend at 60% of the sites. The smoothing technique indicated that short-term patterns in precipitation chemistry were not reflected in surface waters. The magnitude of the short-term variations in surface water concentration was generally larger than the overall long-term trend, possibly because flow adjustment did not adequately correct for climatic variability. Detecting the relation between precipitation and surface water chemistry trends may be improved by using a more powerful sampling strategy and by developing better methods of concentration adjustment to remove the effects of natural variation in surface waters.

The simulation-optimization approach is used to identify ground-water pumping strategies for control of the shallow water table in the western San Joaquin Valley, California, where shallow ground water threatens continued agricultural productivity. The approach combines the use of ground-water flow simulation with optimization techniques to build on and refine pumping strategies identified in previous research that used flow simulation alone. Use of the combined simulation-optimization model resulted in a 20 percent reduction in the area subject to a shallow water table over that identified by use of the simulation model alone. The simulation-optimization model identifies increasingly more effective pumping strategies for control of the water table as the complexity of the problem increases; that is, as the number of subareas in which pumping is to be managed increases, the simulation-optimization model is better able to discriminate areally among subareas to determine optimal pumping locations. The simulation-optimization approach provides an improved understanding of controls on the ground-water flow system and management alternatives that can be implemented in the valley. In particular, results of the simulation-optimization model indicate that optimal pumping strategies are constrained by the existing distribution of wells between the semiconfined and confined zones of the aquifer, by the distribution of sediment types (and associated hydraulic conductivities) in the western valley, and by the historical distribution of pumping throughout the western valley.

Water samples from several rivers were analysed for a range of herbicides by gas chromatography-mass spectrometry and the effects of 10 different sampling frequencies (e.g., annual, half-yearly, quarterly and monthly) on estimates of annual mean concentrations of atrazine, alachlor and cyanazine were tested. Time-weighted annual mean herbicide concentrations obtained from water samples at 17 locations were compared to simulated annual mean concentrations calculated for each sampling strategy, using Monte Carlo simulations. Monthly sampling was the most accurate strategy tested. Sampling quarterly, as required by the U.S. Environmental Protection Agency for surface waters used as drinking water, underestimated the annual mean herbicide concentrations in more than 40 per cent of simulations, presumably because of the seasonality of herbicide occurrence.

The chloride mass-balance method, which integrates time and aerial distribution of ground water recharge, was applied to small alluvial aquifers in the wadi systems of the Asir and Hijaz mountains in western Saudi Arabia. This application is an extension of the method shown to be suitable for estimating recharge in regional aquifers in semi-arid areas. Because the method integrates recharge in time and space it appears to be, with certain assumptions, particularly well suited for and areas with large temporal and spatial variation in recharge. In general, recharge was found to be between 3 to 4% of precipitation - a range consistent with recharge rates found in other arid and semi-arid areas of the earth.

Sierran rivers that discharge to the Lahontan basin have much lower (~4.5) ⁸⁷Sr values than the Humboldt River which drains northeastern Nevada. The ⁸⁷Sr values of tufas deposited during the last lake cycle were used to determine when Humboldt derived Sr entered the Pyramid Lake subbasin. Prior to ~15,000 yr B.P., the Humboldt River flowed to the Smoke Creek-Black Rock Desert subbasin. During the recession of Lake Lahontan, the Humboldt River diverted to the Carson Desert subbasin. This study has demonstrated that ⁸⁷Sr can be used to determine drainage histories of multi-basin lake systems if the ⁸⁷Sr values of rivers that discharge to the basins are sufficiently different.

Oxygen isotope and total inorganic carbon values of cored sediments from the Owens Lake basin, California, indicate that Owens Lake overflowed most of the time between 52,500 and 12,500 carbon-14 (¹⁴C) years before present (B.P.). Owens Lake desiccated during or after Heinrich event H1 and was hydrologically closed during Heinrich event H2. The magnetic susceptibility and organic carbon content of cored sediments indicate that about 19 Sierra Nevada glaciations occurred between 52,500 and 23,500 ¹⁴C years B.P. Most of the glacial advances were accompanied by decreases in the amount of discharge reaching Owens Lake. Comparison of the timing of glaciation with the lithic record of North Atlantic core V23-81 indicates that the number of mountain glacial cycles and the number of North Atlantic lithic events were about equal between 39,000 and 23,500 ¹⁴C years B.P.

In this paper, the fundamental importance of changes in hydrologic balance and hydrologic state on the d¹⁸O and d¹³C values of water and dissolved inorganic carbon (DIC) in lakes of the Lahontan basin is illustrated. Abrupt changes in d¹⁸O and d¹³C values of carbonate deposits (tufas) from the Pyramid Lake subbasin, Nevada, coincide with abrupt changes in lake-level and hydrologic state. Minima in lake-level at ~26,000, ~15,500 and ~12,000 yr B.P. are associated with relatively heavy d¹⁸O and d¹³C values; maxima in the lake-level record at ~14,000 and ~10,500 yr B.P. are associated with relatively light d¹⁸O and d¹³C values. We believe that the correlation between maxima and minima in the lake-level and d¹⁸O records reflect the fundamental effect of lake-level dynamics on the d¹⁸O value of lake water. Evaporation increases the d¹⁸O value of lake water, whereas, streamflow discharge and on-lake precipitation decrease the d¹⁸O value. Variation in the d¹⁸O value of lake water, therefore, indicates change in the hydrologic balance; increases in d¹⁸O accompany decreases in lake volume and decreases in d¹⁸O accompany increases in lake volume. Covariance of d¹³C and d¹⁸O indicates that change in d¹³C values of DIC also accompany change in lake volume. We offer the hypothesis (first put forward by J.A. McKenzie) that change in the productivity (photosynthesis) respiration balance is responsible for much of the observed variation in d¹³C.

Most Great Basin lakes, including Lake Lahontan, experienced changes in hydrologic state during the late Wisconsin. When a lake becomes hydrologically open, the residence time of water decreases. The greater the rate of spill, the greater the volume of evaporated (¹⁸O-enriched) water removed from the spilling lake and the more negative the d¹⁸O value of water remaining in the spilling lake. The concentration of DIC, as well as the concentrations of photosynthesis limiting nutrients (e.g., phosphorus, nitrogen, silica, molybdenum) decrease as spill increases. Increasing rates of spill, therefore, lead to overall decreases in photosynthetic rates relative to respiration rates and, as a consequence, the d¹³C values of DIC become more negative.

A tracer plot was used to determine the contribution of nitrogen fertilizer to nitrate in groundwater and runoff under continuous maize (Zea mays) production in Missouri, U.S.A., using nitrogen-15-labelled fertilizer and bromide as tracers. The tracers were applied in May 1992. Approximately 66 per cent of the labelled nitrogen was accounted for in runoff (1.3 per cent), groundwater (30 per cent), the soil and stover (5 per cent), interflow (2 per cent), and grain (27.3 per cent) at the end of the second growing season. Much of the unaccounted for nitrogen was lost through volatilization of ammonia from maize leaves. Nitrogen transport in surface runoff accounted for less than 2 per cent of the applied labelled nitrogen. The presence of labelled nitrate only in the top 2 m of the aquifer, slow horizontal transport, and winter recharge indicated that grass cover crops such as wheat or rye might be used to extract near-surface nitrogen during the winter recharge period.

Hydrological mechanisms controlling the variation of dissolved organic carbon (DOC) were investigated in the Deer Creek catchment located near Montezuma, CO. Patterns of DOC in streamflow suggested that increased flows through the upper soil horizon during snowmelt are responsible for flushing this DOC-enriched interstitial water to the streams. We examined possible hydrological mechanisms to explain the observed variability of DOC in Deer Creek by first simulating the hydrological response of the catchment using TOPMODEL and then routing the predicted flows through a simple model that accounted for temporal changes in DOC. Conceptually the DOC model can be taken to represent a terrestrial (soil) reservoir in which DOC builds up during low flow periods and is flushed out when infiltrating meltwaters cause the water table to rise into this "reservoir". Concentrations of DOC measured in the upper soil and in streamflow were compared to model simulations. The simulated DOC response provides a reasonable reproduction of the observed dynamics of DOC in the stream at Deer Creek.

The role of organic matter in binding copper was investigated by measuring the fulvic acid fraction of dissolved organic carbon (DOC) and EDTA and by comparing the predicted copper binding to binding measured in stream samples from 7 stream sites in Massachusetts. The computer model, PHREEQE, was used and there was good agreement between modelled and measured binding results. There was an inverse correlation between hardness and copper complexing properties of DOC.

Solute transport simulations quantitatively constrained hydrologic and geochemical hypotheses about field observations of a pH modification in an acid mine drainage stream. Carbonate chemistry, the formation of solid phases, and buffering interactions with the stream bed were important factors in explaining the behavior of pH, aluminum, and iron. The precipitation of microcrystalline gibbsite accounted for the behavior of aluminum; precipitation of Fe(OH)₃ explained the general pattern of iron solubility. The dynamic experiment revealed limitations on assumptions that reactions were controlled only by equilibrium chemistry. Temporal variation in relative rates of photoreduction and oxidation influenced iron behavior. Kinetic limitations on ferrous iron oxidation and hydrous oxide precipitation and the effects of these limitations on field filtration were evident. Kinetic restraints also characterized interaction between the water column and the stream bed, including sorption and desorption of protons from iron oxides at the sediment-water interface and post-injection dissolution of the precipitated aluminum solid phase.

Substantial flowpath-related variability of ⁸⁷Sr/⁸⁶Sr is observed in groundwaters collected from the Trout Lake watershed of northern Wisconsin. In the extensive shallow aquifer composed of sandy glacial outwash, groundwater is recharged either by seepage from lakes or by precipitation that infiltrates the inter-lake uplands. ⁸⁷Sr/⁸⁶Sr of groundwater derived mainly as seepage from a precipitation-dominated lake near the head of the watershed decreases with progressive water chemical evolution along its flowpath due primarily to enhanced dissolution of relatively unradiogenic plagioclase. In contrast, ⁸⁷Sr/⁸⁶Sr of groundwater derived mainly from precipitation that infiltrates upland areas is substantially greater than that of precipitation collected from the watershed, due to suppression of plagioclase dissolution together with preferential leaching of Sr from radiogenic phases such as K-feldspar and biotite. The results of a column experiment that simulated the effects of changing residence time of water in the aquifer sand indicate that mobile waters obtain relatively unradiogenic Sr, whereas stagnant waters obtain relatively radiogenic Sr. Nearly the entire range of strontium-isotope composition observed in groundwaters from the watershed was measured in the experimental product waters. The constant mobility of water along groundwater recharge flowpaths emanating from the lakes promotes the dissolution of relatively unradiogenic plagioclase, perhaps due to effective dispersal of clay mineral nuclei resulting from dissolution reactions. In contrast, episodic stagnation in the unsaturated zone along the upland recharge flowpaths suppresses plagioclase dissolution, perhaps due to accumulation of clay mineral nuclei on its reactive surfaces. Differences in redox conditions along these contrasting flowpaths probably enhance the observed differences in strontium isotope behavior. This study demonstrates that factors other than the calculated state of mineral saturation must be considered when attempting to simulate chemical evolution along flowpaths, and that reaction models must be able to incorporate changing contributions from reacting minerals in the calculations.

Following declines in submersed macrophyte populations in tidal ecosystems, revegetation of areas devoid of macrophytes may be sudden and rapid or may not occur for years. Declines of submersed macrophyte populations in the Chesapeake Bay and the tidal Potomac River have been attributed to insufficient light in the water column; however, the role of light in promoting revegetation has never been unequivocally documented. Photon irradiance was artificially increased for Vallisneria americana transplants in two unvegetated embayments in the otherwise vegetated freshwater tidal Potomac River: Pohick Bay and Belmont Bay. Pohick Bay had high nutrient concentrations and frequent algal blooms. Belmont Bay was broader and shallower than Pohick Bay with turbidity resulting from wind-driven resuspension of sediment. The total number of plants of V. americana in the lighted cages was 7.5 times higher than that in the unlighted cages at Pohick Bay and 11 times higher than that in the unlighted control cages in Belmont Bay. The biomass in the lighted cages was 11-fold higher in Belmont Bay and 38-fold higher in Pohick Bay than that in the control cages. Plants were less numerous and more robust in lighted cages in Pohick Bay than in Belmont Bay.

An important part of the water supply in the western United States is derived from runoff fed by mountain snowmelt. Snow accumulation responds to both precipitation and temperature variations, and forms an interesting climatic index, since it integrates these influences over the entire late fall-spring period. Here, effects of cool season climate variability upon snow water equivalent (SWE) over the western part of the conterminous United States are examined. The focus is on measurements on/around 1 April, when snow accumulation is typically greatest. The primary data, from a network of mountainous snow courses, provides a good description of interannual fluctuations in snow accumulations, since many snow courses have records of five decades or more. For any given year, the spring SWE anomaly at a particular snow course is likely to be 25%-60% of its long-term average. Five separate regions of anomalous SWE variability are distinguished, using a rotated principal components analysis. Although effects vary with region and with elevation, in general, the anomalous winter precipitation has the strongest influence on spring SWE fluctuations. Anomalous temperature has a weaker effect overall, but it has great influence in lower elevations such as in the coastal Northwest, and during spring in higher elevations. The regional snow anomaly patterns are associated with precipitation and temperature anomalies in winter and early spring. Patterns of the precipitation, temperature, and snow anomalies extend over broad regional areas, much larger than individual watersheds. These surface anomalies are organized by the atmospheric circulation, with primary anomaly centers over the North Pacific Ocean as well as over Western North America. For most of the regions, anomalously low SWE is associated with a winter circulation resembling the PNA pattern. With a strong low in the central North Pacific and high pressure over the Pacific Northwest, this pattern diverts North Pacific storms northward, away from the region. Both warm and cool phases of El Niño-Southern Oscillation tend to produce regional patterns with out-of-phase SWE anomalies in the Northwest and the Southwest.

The potential use of iron(III) oxide to stimulate in-situ hydrocarbon degradation in anaerobic petroleum-contaminated harbor sediments was investigated. Previous studies have indicated that Fe(III)-reducing bacteria (FeRB) can oxidize some electron donors more effectively than sulfate-reducing bacteria (SRB). In contrast to previous results in freshwater sediments, the addition of Fe(III) to marine sediments from San Diego Bay, CA did not switch the terminal electron-accepting process (TEAP) from sulfate reduction to Fe-(III) reduction. Addition of Fe(III) also did not stimulate anaerobic hydrocarbon oxidation. Exposure of the sediment to air [to reoxidize Fe(II) to Fe(III)] followed by anaerobic incubation of the sediments, resulted in Fe-(III) reduction as the TEAP, but contaminant degradation was not stimulated and in some instances was inhibited. The difference in the ability of FeRB to compete with the SRB in the different sediment treatments was related to relative population sizes. Although the addition of Fe(III) did not stimulate hydrocarbon degradation, the results presented here as well as other recent studies demonstrate that there may be significant anaerobic hydrocarbon degradation under sulfate-reducing conditions in harbor sediments.

In an attempt to better understand the microorganisms responsible for Fe(III) reduction in sedimentary environments, Fe(III)-reducing microorganisms were enriched for and isolated from freshwater aquatic sediments, a pristine deep aquifer, and a petroleum-contaminated shallow aquifer. Enrichments were initiated with acetate or toluene as the electron donor and Fe(III) as the electron acceptor. Isolations were made with acetate or benzoate. Five new strains which could obtain energy for growth by dissimilatory Fe(III) reduction were isolated. All five isolates are gram-negative strict anaerobes which grow with acetate as the electron donor and Fe(III) as the electron acceptor. Analysis of the 16S rRNA sequence of the isolated organisms demonstrated that they all belonged to the genus Geobacter in the delta subdivision of the Proteobacteria. Unlike the type strain, Geobacter metallireducens, three of the five isolates could use H sub(2) as an electron donor for Fe(III) reduction. The deep subsurface isolate is the first Fe(III) reducer shown to completely oxidize lactate to carbon dioxide, while one of the freshwater sediment isolates is only the second Fe(III) reducer known that can oxidize toluene. The isolation of these organisms demonstrates that Geobacter species are widely distributed in a diversity of sedimentary environments in which Fe(III) reduction is an important process.

Dams have major downstream effects in settings throughout the United States. These effects depend on the operating regime of the dam, the extent of flood control, the amount of sediment deprivation downstream, and the nature of riverine and aquatic ecosystems in the river. This publication provides an overview of the effects of 6 dams on the downstream environment, written in a style for lay audiences.

Environmental isotope studies were undertaken to determine the origin and the rate of movement of groundwater through the thick surficial gravel aquifer in the Szigetkoez region of northwest Hungary. Both ¹⁴C and ³H data indicate that the groundwaters in the surficial gravel aquifer are much younger (¹⁴C > 60 pmC) than underlying groundwaters of the Pannonian aquifer (¹⁴C < 7 pmC). Although some upward leakage of older waters may occur, such leakage does not appear to be a major source of water in the surficial gravel aquifer. Stable O isotope data indicate that the groundwater stored in the gravel aquifer originated mainly from the Danube river (d¹⁸O = -11.6 ± 0.5ppt), with relatively little local infiltration (d¹⁸O = -9.3 ± 0.4ppt), so the aquifer may be considered as an unusually well developed, large scale natural bank filtered groundwater system. The distribution of ³H in the aquifer confirms that the groundwater flow is mainly horizontal and directed away from the Danube. The mid-1960s ³H 'bomb peak' (>50 TU as measured in 1992-1993) has moved as much as 15 km from the Danube and indicates horizontal flow velocities as high as 500 m/a in the surficial aquifer.

Pulse-chase experiments show that two marine bivalves take optimal advantage of different types of particulate food by varying food retention time in a flexible two-phase digestive system. For example, carbon is efficiently assimilated from bacteria by subjecting nearly all the ingested bacteria to prolonged digestion. Prolonging digestion also enhances assimilation of metals, many of which are toxic in minute quantities if they are biologically available. Detritus-feeding aquatic organisms have always lived in environments naturally rich in particle-reactive metals. We suggest that avoiding excess assimilation of metals could be a factor in the evolution of digestion strategies. We tested that suggestion by studying digestion of particles containing different Cr concentrations. We show that bivalves are capable of modifying the digestive processing of food to reduce exposure to high, biologically available, Cr concentrations. The evolution of a mechanism in some species to avoid high concentrations of metals in food could influence how effects of modern metal pollution are manifested in marine ecosystems.

Water, bed sediment, and biota were sampled in streams from Butte to below Missoula as part of a program to characterize aquatic resources in the upper Clark Fork basin of western Montana. Sampling stations were located on the Clark Fork and major tributaries. Water-quality data were obtained periodically at 16 stations during October 1994 through September 1995 (water year 1995). Data for twelve bed-sediment and eleven biological stations were obtained in August 1995. The primary constituents analyzed were trace elements associated with mine tailings from historical mining and smelting activities. Water-quality data include concentrations of major ions, trace element, and suspended sediment in stream samples collected periodically during water year 1995. Daily values of streamflow, suspended- sediment concentrations in the fine-grained and bulk fractions. Biological data include trace- element concentrations in whole-body tissue of aquatic benthic insects. Quality-assurance data are reported for analytical results of water, bed sediment, and biota. Statistical summaries of water-quality, bed-sediment, and biological data are provided for the period of record at each station since 1985.

Nitrate reduction and denitrification were measured in swamp forest streams draining lowland rain forest on Costa Rica's Atlantic slope foothills using the C₂H₂-block assay and sediment-water nutrient fluxes. Denitrification assays using the C₂H₂-block technique indicated that the full suite of denitrifying enzymes were present in the sediment but that only a small fraction of the functional activity could be expressed without adding NO₃^-. Under optimal conditions, denitrification enzyme activity averaged 15 nmoles cm^-3 sediment h^-1. Areal NO₃^- reduction rates measured from NO₃^- loss in the overlying water of sediment-water flux chambers ranged from 65 to 470 mmoles m^-2 h^-1. Oxygen loss rates accompanying NO₃^- depletion averaged 750 mmoles m^-2 h^-1. Corrected for denitrification of NO₃^- oxidized from NH₄⁺ in the sediment, gross NO₃^- reduction rates increase by 130 mmoles m^-2 h^-1, indicating nitrification may be the predominant source of NO₃^- for NO₃ reduction in swamp forest stream sediments. Under field conditions approximately 80% of the increase in inorganic N mass along a 1250-m reach of the Salto River was in the form of NO₃ with the balance NH₄⁺. Scrutiny of potential inorganic N sources suggested that mineralized N released from the streambed was a major source of the inorganic N increase. Despite significant NO₃^- reduction potential, swamp forest stream sediments appear to be a source of inorganic N to downstream communities.

This report describes a computer model of the Raritan River Basin water-supply system in central New Jersey. The computer model provides a technical basis for evaluating the effects of alternative patterns of operation of the Raritan River Basin water-supply system during extended periods of below-average precipitation. The computer model is a continuity-accounting model consisting of a series of interconnected nodes. At each node, the inflow volume, outflow volume, and change in storage are determined and recorded for each month. The model runs with a given set of operating rules and water-use requirements including releases, pumpages, and diversions. The model can be used to assess the hypothetical performance of the Raritan River Basin water- supply system in past years under alternative sets of operating rules. It also can be used to forecast the likelihood of specified outcomes, such as the depletion of reservoir contents below a specified threshold or of streamflows below statutory minimum passing flows, for a period of up to 12 months. The model was constructed on the basis of current reservoir capacities and the natural, unregulated monthly runoff values recorded at U.S. Geological Survey streamflow- gaging stations in the basin.

Crude oil spilled from a subsurface pipeline in north-central Minnesota has dissolved in the groundwater, resulting in the formation of a plume of aliphatic, aromatic, and alicyclic hydrocarbons. Comparison of paired oil and groundwater samples collected along the central axis of the residual oil body shows that the trailing edge of the oil is depleted in the more soluble aromatic hydrocarbons (e.g., benzene, toluene, etc.) when compared with the leading edge. At the same time, concentrations of monoaromatic hydrocarbons in groundwater beneath the oil increase as the water moves toward the leading edge of the oil. Immediately downgradient from the leading edge of the oil body, certain aromatic hydrocarbons (e.g., benzene) are found at concentrations near those expected of a system at equilibrium, and the concentrations exhibit little variation overtime ( approximately 8-20%). Other compounds (e.g., toluene) appear to be undersaturated, and their concentrations show considerably more temporal variation (approximately 20-130%). The former are persistent within the anoxic zone downgradient from the oil, whereas concentrations of the latter decrease rapidly. Together, these observations suggest that the volatile hydrocarbon composition of the anoxic groundwater near the oil body is controlled by a balance between dissolution and removal rates with only the most persistent compounds reaching saturation. Examination of the distributions of homologous series and isomeric assemblages of alkylbenzenes reveals that microbial degradation is the dominant process controlling the fate of these compounds once groundwater moves away from the oil. For all but the most persistent compounds, the distal boundary of the plume at the water table extends no more than 10-15 m down-gradient from the oxic /anoxic transition zone. Thus, transport of the monoaromatic hydrocarbons is limited by redox conditions that are tightly coupled to biological degradation processes.

Several geochemical properties of an aquifer sediment that control metal-ion adsorption were investigated to determine their potential use as indicators of the spatial variability of metal adsorption. Over the length of a 4.5-m-long core from a sand and gravel aquifer, lead (Pb²⁺) and zinc (Zn²⁺) adsorption at constant chemical conditions (pH 5.3) varied by a factor of 2 and 4, respectively. Pb²⁺ and Zn²⁺ were adsorbed primarily by Fe- and Al-oxide coatings on quartz-grain surfaces. Per unit surface area, both Pb²⁺ and Zn²⁺ adsorption were significantly correlated with the amount of Fe and Al that dissolved from the aquifer material in a partial chemical extraction. The variability in conditional binding constants for Pb²⁺ and Zn²⁺ adsorption (log K ads) derived from a simple non-electrostatic surface complexation model were also predicted by extracted Fe and Al normalized to surface area. Because the abundance of Fe- and Al-oxide coatings that dominate adsorption does not vary inversely with grain size by a simple linear relationship, only a weak, negative correlation was found between spatial variability of Pb²⁺ adsorption and grain size in this aquifer. The correlation between Zn²⁺ adsorption and grain size was not significant. Partial chemical extractions combined with surface-area measurements have potential use for estimating metal adsorption variability in other sand and gravel aquifers of negligible carbonate and organic carbon content.

Soils are often layered at scales smaller than the block size used in numerical and conceptual models of variably saturated flow. Consequently, the small-scale variability in water content within each block must be homogenized (upscaled). Laboratory results have shown that a linear volume average (LVA) of water content at a uniform suction is a good approximation to measured water contents in heterogeneous cores. Here, we upscale water contents using van Genuchten's function for both the local and upscaled soil-water-retention characteristics. The van Genuchten (vG) function compares favorably with LVA results, laboratory experiments under hydrostatic conditions in 3-cm cores, and numerical simulations of large-scale gravity drainage. Our method yields upscaled vG parameter values by fitting the vG curve to the LVA of water contents at various suction values. In practice, it is more efficient to compute direct averages of the local vG parameter values. Nonlinear power averages quantify a feasible range of values for each upscaled vG shape parameter; upscaled values of N are consistently less than the harmonic means, reflecting broad pore-size distributions of the upscaled soils. The vG function is useful for modeling soil-water retention at large scales, and these results provide guidance for its application.

Investigating natural, potential, and human-induced impacts on hydrologic systems commonly requires complex modeling with overlapping data requirements, plus massive amounts of one- to four-dimensional data at multiple scales and formats. Given the complexity of most hydrologic studies, the requisite software infrastructure must incorporate many components including simulation modeling and spatial analysis with a flexible, intuitive display. Integrating geographic information systems (GIS) and scientific visualization systems (SVS) provides such an infrastructure. This paper describes an integrated system consisting of an orographic precipitation model, a GIS, and an SVS. The results of this study provide a basis for improving the understanding of hydro-climatic processes in mountainous regions. An additional benefit of the integrated system, the value of which is often underestimated, is the improved ability to communicate model results, leading to a broader understanding of the model assumptions, sensitivities, and conclusions at a management level.

Dissolution experiments using augite (Mg_0.87Ca_0.85Fe_0.19Na_0.09Al_0.03Si₂O₆) and diopside (Mg_0.91Ca_0.93Fe_0.07Na_0.03Al_0.03Si₂O₆) were conducted in flow-through reactors (5-ml/h flow rate). A pH of 5.8 was maintained by bubbling pure CO₂ through a solution of 0.01 M KHCO₃ at 25 degrees C. Two experiments were run for each pyroxene type. In one experiment dissolved O₂ concentration in reactors was 0.6 ± 0.1 ppm and in the second dissolved O₂ was 1.5 ± 0.1 ppm. After 60 days, augite dissolution rates (based on Si release) were approximately three times greater in the 1.5 ppm dissolved O₂ experiments than in the sealed experiments. In contrast, diopside dissolution rates were independent of dissolved O₂ concentrations. Preliminary results from the augite experiments suggest that dissolution rate is directly related to oxidation of iron. This effect was not observed in experiments performed on iron-poor diopside. Additionally, dissolution rates of diopside were much slower than those of augite, again suggesting a relationship between Fe content, Fe oxidation and dissolution rates.

Ground-water withdrawal from a confined or semiconfined aquifer causes three-dimensional deformation in the pumped aquifer and in adjacent layers (overlying and underlying aquifers and aquitards). In response to the deformation, hydraulic head in the adjacent layers could rise or fall almost immediately after the start of pumping. This deformation-induced effect is analyzed by a linear poroelasticity model using properties typical of unconsolidated sedimentary materials. Model simulations suggest that an adjacent layer undergoes horizontal compression and vertical extension when pumping begins. Hydraulic head initially drops in a region near the well and close to the pumped aquifer, but rises outside this region. Magnitude of head change varies from a few centimeters to more than 10 centimeters. Factors that influence the development of deformation-induced effects includes matrix rigidity (shear modulus), the arrangement of aquifer and aquitards, their thicknesses, and proximity to land surface. Induced rise in hydraulic head is prominent in an aquitard that extends from land surface to a shallow pumped aquifer. Induced drop in hydraulic head is likely observed close to the well in an aquifer that is separated from the pumped aquifer by a relatively thin aquitard. Induced effects might last for hours in an aquifer, but could persist for many days in an aquitard. Induced effects are eventually dissipated by fluid flow from regions of higher head to regions of lower head, and by propagation of drawdown from the pumped aquifer into adjacent layers.

Riparian vegetation and fluvial-geomorphic processes and landforms are intimately connected parts of the bottomland landscape. Relations among vegetation, processes, and landforms are described here for representative streams of four areas of the United States: high-gradient streams of the humid east, coastal-plain streams, Great Plains streams, and stream channels of the southwestern United states. Vegetation patterns suggest that species distributions in the humid east largely controlled by frequency, duration, and intensity of floods. Along channelized streams, vegetation distribution is largely controlled by variation in fluvial geomorphic processes (cycles of degradation and aggradation) in response to increases in channel gradient associated with channelization. Similarly, riparian vegetation of Great Plains streams may be controlled by fluxes in sediment deposition and erosion along braided streams. patterns of riparian vegetation in semi-arid regions may be most closely related to patterns of water availability, unlike most other streams in more humid environments. Channel-equilibrium conditions control stability of the coincident fluvial landform and attendant vegetation pattern throughout the continent. In most situations, riparian-vegetation patterns are indicative of specific landforms and, thus, of ambient hydrogeomorphic conditions.

Riparian vegetation and fluvial-geomorphic processes and landforms are intimately connected parts of the bottomland landscape. Relations among vegetation, processes, and landforms are described here for representative streams of four areas of the United States: high-gradient streams of the humid east, coastal-plain streams, Great Plains streams, and stream channels of the southwestern United states. Vegetation patterns suggest that species distributions in the humid east largely controlled by frequency, duration, and intensity of floods. Along channelized streams, vegetation distribution is largely controlled by variation in fluvial geomorphic processes (cycles of degradation and aggradation) in response to increases in channel gradient associated with channelization. Similarly, riparian vegetation of Great Plains streams may be controlled by fluxes in sediment deposition and erosion along braided streams. patterns of riparian vegetation in semi-arid regions may be most closely related to patterns of water availability, unlike most other streams in more humid environments. Channel-equilibrium conditions control stability of the coincident fluvial landform and attendant vegetation pattern throughout the continent. In most situations, riparian-vegetation patterns are indicative of specific landforms and, thus, of ambient hydrogeomorphic conditions.

The Pueblo of Isleta and the New Mexico Environment Department have established water-quality standards for the Rio Grande, which flows through Albuquerque, New Mexico. Trace-element concentrations historically have been greater than maximum permissible concentrations allowed by these standards. It is not known if these concentrations are due to sources from the Albuquerque metropolitan area or are from natural or other sources outside Albuquerque. Accurate water-quality data with lower reporting limits than have been previously available were collected, and instantaneous concentrations and loads were calculated for trace elements and other constituents in the Rio Grande during high-flow and low-flow conditions. Seven sampling sites were selected upstream from, in, and downstream from metropolitan Albuquerque. Concurrent streamflow measurements were made at the time of sampling to determine suspended-sediment loads. Samples were analyzed separately for trace elements dissolved in water (less than 0.4 micrometer in diameter) and for those contained in suspended sediment (greater than 0.1 micrometer in diameter). Sample collection and processing, analytical methods, and quality control are discussed.

Seleniferous agricultural drainage wastewater has become a new major source of pollution in the world. In the USA, large areas of farmland in 17 western states, generate contaminated salinized drainage with Se concentrations much higher than 5 mg/l, the US Environmental Protection Agency water-quality criterion for the protection of aquatic life; Se values locally reach 4200 mg/l in western San Joaquin Valley, California. Wetland habitats receiving this drainage have generally shown Se toxicosis in aquatic birds causing high rates of embryonic deformity and mortality, or have indicated potential ecological damage. Results of our laboratory flow experiments indicate that nanofiltration, the latest membrane separation technology, can selectively remove >95% of Se and other multivalent anions from >90% of highly contaminated water from the San Joaquin Valley, California. Such membranes yield greater water output and require lower pressures and less pretreatment, and therefore, are more cost effective than traditional reverse osmosis membranes. Nanofiltration membranes offer a potential breakthrough for the management of Se contaminated wastes not only from agricultural drainage, but from other sources also.

Generally accepted standards for testing and validating ecosystem models would benefit both modellers and model users. Universally applicable test procedures are difficult to prescribe, given the diversity of modelling approaches and the many uses for models. However, the generally accepted scientific principles of documentation and disclosure provide a useful framework for devising general standards for model evaluation. Adequately documenting model tests requires explicit performance criteria, and explicit benchmarks against which model performance is compared. A model's validity, reliability, and accuracy can be most meaningfully judged by explicit comparison against the available alternatives. In contrast, current practice is often characterized by vague, subjective claims that model predictions show 'acceptable' agreement with data; such claims provide little basis for choosing among alternative models. Strict model tests (those that invalid models are unlikely to pass) are the only ones capable of convincing rational skeptics that a model is probably valid. However, 'false positive' rates as low as 10% can substantially erode the power of validation tests, making them insufficiently strict to convince rational skeptics. Validation tests are often undermined by excessive parameter calibration and overuse of ad hoc model features. Tests are often also divorced from the conditions under which a model will be used, particularly when it is designed to forecast beyond the range of historical experience. In such situations, data from laboratory and field manipulation experiments can provide particularly effective tests, because one can create experimental conditions quite different from historical data, and because experimental data can provide a more precisely defined 'target' for the model to hit. We present a simple demonstration showing that the two most common methods for comparing model predictions to environmental time series (plotting model time series against data time series, and plotting predicted versus observed values) have little diagnostic power. We propose that it may be more useful to statistically extract the relationships of primary interest from the time series, and test the model directly against them.

The HYDROCOIN project was an international cooperative study on groundwater modeling. HYDROCOIN Level 1 studies aimed to "verify" the accuracy of groundwater codes for various hydrogeological problems. The Case 5 problem included significant density variations related to salinity differences. Several independent teams simulated this problem using finite difference or finite element numerical models. Comparisons showed some significant differences among model results. We applied a method of characteristics model (MOCDENSE) to this same case. These results show that MOCDENSE can reliably simulate density-dependent groundwater flow and transport for the conditions of Case 5. This analysis also revealed that the standard numerical implementation of a constant concentration boundary to represent salt release only by lateral dispersion was flawed because this boundary condition allows the release of salt into the flow field by both dispersion and advection. When the boundary is modified to truly allow salt release only by dispersion, significantly less salt is released into the flow field and the size and shape of the plume is notably different than that obtained by the HYDROCOIN participants.

This report presents a model, MOC3D, that simulates three-dimensional solute transport in flowing ground water. The model computes changes in concentration of a single dissolved chemical constituent over time that are caused by advective transport, hydrodynamic dispersion (including both mechanical dispersion and diffusion), mixing (or dilution) from fluid sources, and mathematically simple chemical reaction (including linear sorption, which is represented by a retardation factor, and decay). The transport model is integrated with MODFLOW, a three-dimensional ground-water flow model that uses implicit finite-difference methods to solve the transient flow equation. MOC3D uses the method of characteristics to solve the transport equation on the basis of the hydraulic gradients computed with MODFLOW for a given time step. This implementation of the method of characteristics uses particle tracking to represent advective transport and explicit finite-difference methods to calculate the effects of other processes. However, the explicit procedure had several stability criteria that may limit the size of time increments for solving the transport equation; these are automatically determined by the program.For improved efficiency, the user can apply MOC3D to a subgrid of the primary MODFLOW grid that is used to solve the flow equation. However, the transport subgrid must have uniform grid spacing along rows and columns. The report includes a description of the theoretical basis of the model, a detailed description of input requirements and output options, and the results of model testing and evaluation. The model was evaluated for several problems for which exact analytical solutions are available and by benchmarking against other numerical codes for selected complex problems for which no exact solutions are available. These test results indicate that the model is very accurate for a wide range of conditions and yields a minimal numerical dispersion for advection-dominated problems. Mass-balance errors are generally less than 10 percent, and tend to decrease and stabilize with time.

Twenty-seven years of data from midcontinent wetlands indicate that the response of these wetlands to extremes in precipitation - drought and deluge - persists beyond the extreme events. Chemical changes transcend such simple relations as increased salinity during dry periods because drought provides mechanisms for removal of salt by deflation and seepage to groundwater. Inundation of vegetation zones including rooted or floating mats of cattail Typha glauca can stimulate sulfate reduction and shift the anion balance from sulfate to bicarbonate dominance. Disruptions in the circulation of moisture-laden air masses over the midcontinent, as in the drought of 1988 and the deluge of 1993, have a major effect on these wetlands, which are representatives of the primary waterfowl breeding habitat of the continent.

The Modular Modelling System (MMS) is an integrated system of computer software that is being developed to provide the research and operational framework needed to support development, testing, and evaluation of physical-process algorithms, and to facilitate integration of user-selected sets of algorithms into operational physical-process models. MMS uses a module library that contains compatible modules for simulating a variety of water, energy, and biogeochemical processes. A model is created by selectively linking modules from the library using MMS model-building tools. A geographic information system (GIS) interface also is being developed for MMS to support a variety of GIS tools for use in characterizing and parameterizing topographic, hydrologic, and ecosystem features, visualizing spatially and temporally distributed model parameters and variables, and analyzing and validating model results. MMS is being coupled with the Power Reservoir System Model (PRSYM) to provide a database-centered decision support system for making complex operational decisions on multipurpose reservoir systems and watersheds. The U.S. Geological Survey and the Bureau of Reclamation are working collaboratively on a project titled the Watershed Modeling Systems Initiative to develop and apply the coupled MMS-PRSYM models to the San Juan River basin in Colorado, New Mexico, Arizona, and Utah.

The Modular Modelling System (MMS) is an integrated system of computer software that is being developed to provide the research and operational framework needed to support development, testing, and evaluation of physical-process algorithms, and to facilitate integration of user-selected sets of algorithms into operational physical-process models. MMS uses a module library that contains compatible modules for simulating a variety of water, energy, and biogeochemical processes. A model is created by selectively linking modules from the library using MMS model-building tools. A geographic information system (GIS) interface also is being developed for MMS to support a variety of GIS tools for use in characterizing and parameterizing topographic, hydrologic, and ecosystem features, visualizing spatially and temporally distributed model parameters and variables, and analyzing and validating model results. MMS is being coupled with the Power Reservoir System Model (PRSYM) to provide a database-centered decision support system for making complex operational decisions on multipurpose reservoir systems and watersheds. The U.S. Geological Survey and the Bureau of Reclamation are working collaboratively on a project titled the Watershed Modeling Systems Initiative to develop and apply the coupled MMS-PRSYM models to the San Juan River basin in Colorado, New Mexico, Arizona, and Utah.

Leith, S.D., Reddy, M. M., Ramirez, W.F., and Heymans, M.J., 1996, Limestone characterization to model damage from acidic precipitation: effect of pore structure on mass transfer: Environmental Science and Technology, v. 30, no. 7, p. 2202-2210.

The pore structure of Salem limestone is investigated, and conclusions regarding the effect of the pore geometry on modeling moisture and contaminant transport are discussed based on thin section petrography, scanning electron microscopy, mercury intrusion porosimetry, and nitrogen adsorption analyses. These investigations are compared to and shown to compliment permeability and capillary pressure measurements for this common building stone. Salem limestone exhibits a bimodal pore size distribution in which the larger pore provide routes for convective mass transfer of contaminants into the material and the smaller pores lead to high surface area adsorption and reaction sites. Relative permeability and capillary pressure measurements of the air/water system indicate that Salem limestone exhibits high capillarity and low effective permeability to water. Based on stone characterization, aqueous diffusion and convection are believed to be the primary transport mechanisms for pollutants in this stone. The extent of contaminant accumulation in the stone depends on the mechanism of partitioning between the aqueous and solid phases. The described characterization techniques and modeling approach can be applied to many systems of interest such as acidic damage to limestone, mass transfer of contaminants in concrete and other porous building materials, and modeling pollutant transport in subsurface moisture zones.

In the mountainous regions of the Upper Colorado River Basin, snow course observations give local measurements of snow water equivalent, which can be used to estimate regional averages of snow conditions. We develop a statistical technique to estimate the mesoscale average snow accumulation, using 8 years of snow course observations. For each of three major snow accumulation regions in the Upper Colorado River Basin - the Colorado Rocky Mountains, Colorado, the Uinta Mountains, Utah, and the Wind River Range, Wyoming - the snow course observations yield a correlation length scale of 38 km, 46 km, and 116 km respectively. This is the scale for which the snow course data at different sites are correlated with 70 per cent correlation. This correlation of snow accumulation over large distances allows for the estimation of the snow water equivalent on a mesoscale basis. With the snow course data binned into 1/4 degree latitude by 1/4 degree longitude pixels, an error analysis shows the following: for no snow course data in a given pixel, the uncertainty in the water equivalent estimate reaches 50 cm; that is, the climatological variability. However, as the number of snow courses in a pixel increases the uncertainty decreases, and approaches 5-10 cm when there are five snow courses in a pixel.

Winter mean 700-hectoPascal (hPa) height anomalies, representing the average atmospheric circulation during the snow season, are compared with annual streamflow measured at 140 streamgauges in the western United States. Correlation and anomaly pattern analyses are used to identify relationships between winter mean atmospheric circulation and temporal and spatial variability in annual streamflow. Results indicate that variability in winter mean 700-hPa height anomalies accounts for a statistically significant portion of the temporal variability in annual streamflow in the western United States. In general, above-average annual streamflow is associated with negative winter mean 700-hPa height anomalies over the eastern North Pacific Ocean and/or the western United States. The anomalies produce an anomalous flow of moist air from the eastern North Pacific Ocean into the western United States that increases winter precipitation and snowpack accumulations, and subsequently streamflow. Winter mean 700-hPa height anomalies also account for statistically significant differences in spatial distributions of annual streamflow. As part of this study, winter mean atmospheric circulation patterns for the 40 years analysed were classified into five winter mean 700-hPa height anomaly patterns. These patterns are related to statistically significant and physically meaningful differences in spatial distributions of annual streamflow.

Ground water in terrace deposits of the South Platte River alluvial aquifer near Greeley, Colorado, USA, had a median nitrate concentration of 1857 mmol/l. Median nitrate concentrations in ground water from adjacent floodplain deposits (468 mmol/L) and riverbed sediments (461 mmol/L), both of which are downgradient from the terrace deposits, were lower than the median concentration in the terrace deposits. The concentrations and d¹⁵N values of nitrate and N₂ in ground water indicated that denitrifying activity in the floodplain deposits and riverbed sediments accounted for 15-30% of the difference in nitrate concentrations. Concentrations of Cl^- and SiO₂ indicated that mixing between river water and ground water in the floodplain deposits and riverbed sediments accounted for the remainder of the difference in nitrate concentrations. River flux measurements indicated that ground-water discharge in a 7.5 km segment of river had a nitrate load of 1718 kg N/day and accounted for about 18% of the total nitrate load in the river at the downstream end of that segment. This nitrate load was 70% less than the load predicted on the basis of the median nitrate concentration in the terrace deposits and assuming no denitrification or mixing in the aquifer. Water exchange between the river and aquifer caused ground water that originally discharged to the river to reenter denitrifying sediments in the riverbed and floodplain, thereby further decreasing the nitrate load in this stream-aquifer system. Results from this study indicated that denitrification and mixing within alluvial aquifer sediments may substantially decrease the nitrate load added to rivers by discharging ground water.

Methane, a radiatively active "greenhouse" gas, is emitted from lakes to the atmosphere throughout the open-water season. However, annual lake CH₄ emissions calculated solely from open-water measurements that exclude the time of spring ice melt may substantially underestimate the lake CH₄ source strength. We estimated potential spring CH₄ emission at the time of ice melt for 19 lakes in northern Minnesota and Wisconsin. Lakes ranged in area from 2.7 to 57,300 ha and varied in littoral zone sediment type. Regression analyses indicated that lake area explained 38% of the variance in potential CH₄ emission for relatively undisturbed lakes; as lake area increases potential CH₄ emission per unit area decreases. Inclusion of a second term accounting for the presence or absence of soft organic-rich littoral-zone sediments explained 83% of the variance in potential spring CH₄ emission. Total estimated spring CH₄ emission for 1993 for all Minnesota lakes north of 45 degree with areas greater than or equal to 4 ha was 1.5 x 10⁸ mol CH₄ assuming a 1:1 ratio of soft littoral sediment to hard littoral sediment lakes. Emission estimates ranged from 5.3 x 10⁷ mol assuming no lakes have soft organic-rich littoral sediments to 4.5 x 10⁸ mol assuming all lakes have soft organic-rich littoral sediments. This spring CH₄ pulse may make up as much as 40% of the CH₄ annually emitted to the atmosphere by small lakes.

The response of water in the unsaturated zone to seasonal changes of temperature (T) is determined analytically using the theory of non-isothermal water transport in porous media, and the solutions are tested against field observations of moisture potential and bomb-fallout isotopic (³⁶Cl and ³H) concentrations. Seasonally varying land-surface temperatures and the resulting subsurface temperature gradients induce thermal vapor diffusion. The annual mean vertical temperature gradient is close to zero; however, the annual mean thermal vapor flux is downward, because the temperature-dependent vapor-diffusion coefficient is larger, on average, during downward diffusion (occurring at high T) than during upward diffusion (low T). The annual mean thermal vapor flux is shown to decay exponentially with depth; the depth (about 1 m) at which it decays to 1/e of its surface value is one-half of the corresponding decay depth for the amplitude of seasonal temperature changes. This depth-dependent annual mean flux is effectively a source of water, which must be balanced by a flux divergence associated with other transport processes. In a relatively humid environment, the liquid fluxes greatly exceed the thermal vapor fluxes, so such a balance is readily achieved without measurable effect on the dynamics of water in the unsaturated zone. However, if the mean vertical water flux through the unsaturated zone is very small (less than 1 mm/y ), as it may be at many locations in a desert landscape, the thermal vapor flux must be balanced mostly by a matric-potential- induced upward flux of water. This return flux may include both vapor and liquid components. Below any near-surface zone of weather-related fluctuations of matric potential, maintenance of this upward flux requires an increase with depth in the annual-mean matric potential; this theoretical prediction is supported by long-term field measurements in the Chihuahuan Desert. The analysis also makes predictions, confirmed by the field observations, regarding the seasonal variations of matric potential at a given depth. The conceptual model of unsaturated-zone water transport developed here implies the possibility of near-surface trapping of any aqueous constituent introduced at the surface.

Loss of fluids and samples during retrieval of cores of saturated, noncohesive sediments results in incorrect measures of fluid distributions and an inaccurate measure of the stratigraphic position of the sample. To reduce these errors, we developed a hollow drive shoe that freezes in place the lowest 3 inches (75 mm) of a 1.88-inch-diameter (48 mm), 5-foot-long (1.5 m) sediment sample taken using a commercial wire line piston core sampler. The end of the core is frozen by piping liquid carbon dioxide at ambient temperature through a steel tube from a bottle at the land surface to the drive shoe where it evaporates and expands, cooling the interior surface of the shoe to about -109 degree F (-78 degree C). Freezing a core end takes about 10 minutes. The device was used to collect samples for a study of oil-water-air distributions, and for studies of water chemistry and microbial activity in unconsolidated sediments at the site of an oil spill near Bemidji, Minnesota. Before freezing was employed, samples of sandy sediments from near the water table sometimes flowed out of the core barrel as the sampler was withdrawn. Freezing the bottom of the core allowed for the retention of all material that entered the core barrel and lessened the redistribution of fluids within the core. The device is useful in the unsaturated and shallow saturated zones, but does not freeze cores well at depths greater than about 20 feet (6 m) below water, possibly because the feed tube plugs with dry ice with increased exhaust back-pressure, or because sediment enters the annulus between the core barrel and the core barrel liner and blocks the exhaust.

In the past, ice-core records from mid-latitude glaciers in alpine areas of the continental United States were considered to be poor candidates for paleoclimate records because of the influence of meltwater on isotopic stratigraphy. To evaluate the existence of reliable paleoclimatic records, a 160-m ice core, containing about 250 yr of record was obtained from Upper Fremont Glacier, at an altitude of 4000 m in the Wind River Range of south-central North America. The d¹⁸O (SMOW) profile from the core shows a -0.95ppt shift to lighter values in the interval from 101.8 to 150 m below the surface, corresponding to the latter part of the Little Ice Age (LIA). Numerous high-amplitude oscillations in the section of the core from 101.8 to 150 m cannot be explained by site-specific lateral variability and probably reflect increased seasonality or better preservation of annual signals as a result of prolonged cooler temperatures that existed in this alpine setting. An abrupt decrease in these large amplitude oscillations at the 101.8-m depth suggests a sudden termination of this period of lower temperatures which generally coincides with the termination of the LIA. Three common features in the d¹⁸O profiles between Upper Fremont Glacier and the better dated Quelccaya Ice Cap cores indicate a global paleoclimate linkage, further supporting the first documented occurrence of the LIA in an ice-core record from a temperate glacier in south-central North America.

Predicting changes in alluvial channel morphology associated with anthropogenic and natural changes in flow and/or sediment supply is a critical part of the management of riverine systems. Over the past few years, advances in the understanding of the physics of sediment transport in conjunction with rapidly increasing capabilities in computational fluid dynamics have yielded new approaches to problems in river mechanics. Techniques appropriate for length scales ranging from reaches to bars and bedforms are described here. Examples of the use of these computational approaches are discussed for three cases: (1) the design of diversion scenarios that maintain channel morphology in steep cobble-bedded channels in Colorado, (2) determination of channel maintenance flows for the preservation of channel islands in the Snake River in Idaho, and (3) prediction of the temporal evolution of deposits in lateral separation zones for future assessment of the impacts of various dam release scenarios on lateral separation deposits in the Colorado River in Grand Canyon. With continued development of their scientific and technical components, the methodologies described here can provide powerful tools for the management of river environments in the future.

Improvements in the field sampling, preservation, and determination of trace metals in natural waters have made many analyses more reliable and less affected by contamination. The speciation of trace metals, however, remains controversial. Chemical model speciation calculations do not necessarily agree with voltammetric, ion exchange, potentiometric, or other analytical speciation techniques. When metal-organic complexes are important, model calculations are not usually helpful and on-site analytical separations are essential. Many analytical speciation techniques have serious interferences and only work well for a limited subset of water types and compositions. A combined approach to the evaluation of speciation could greatly reduce these uncertainties. The approach proposed would be to (1) compare and contrast different analytical techniques with each other and with computed speciation, (2) compare computed trace metal speciation with reliable measurements of solubility, potentiometry, and mean activity coefficients, and (3) compare different model calculations with each other for the same set of water analyses, especially where supplementary data on speciation already exist. A comparison and critique of analytical with chemical model speciation for a range of water samples would delineate the useful range and limitations of these different approaches to speciation. Both model calculations and analytical determinations have useful and different constraints on the range of possible speciation such that they can provide much better insight into speciation when used together. Major discrepancies in the thermodynamic databases of speciation models can be evaluated with the aid of analytical speciation, and when the thermodynamic models are highly consistent and reliable, the sources of error in the analytical speciation can be evaluated. Major thermodynamic discrepancies also can be evaluated by simulating solubility and activity coefficient data and testing various chemical models for their range of applicability. Until a comparative approach such as this is taken, trace metal speciation will remain highly uncertain and controversial.

This study investigates the episodic acidification of Reedy Creek, a wetland-influenced coastal plain stream near Richmond, Virginia. Primary objectives of the study were to quantify the episodic variability of acid-base chemistry on Reedy Creek, to examine the seasonal variability in episodic response and to explain the hydrological and geochemical factors that contribute to episodic acidification. Chemical response was similar in each of the seven storms examined, however, the ranges in concentration observed were commonly greater in summer/fall storms than in winter/spring storms. An increase in SO₄^2- concentration with discharge was observed during all storms and peak concentration occurred at or near peak flow. Small increases in Mg²⁺, Ca²⁺, K⁺ concentrations and dissolved organic carbon (DOC) were observed during most storms. At the same time, ANC, Na⁺ and Cl^- concentrations usually decreased with increasing discharge. In summer/fall storms, the absolute increase in SO₄^2- concentration was one-third to 15 times the increase observed in winter/spring storms; the decrease in ANC during summer/fall storms was usually within the range of the decrease observed in winter/spring storms. In contrast, the decrease in Na⁺ and Cl^- concentrations during winter/spring storms was much greater than that observed during summer/fall storms. Data show that while base flow anion deficit was higher in summer/fall than in winter/spring, anion deficit decreased during most summer/fall storms. In contrast, base flow anion deficit was lower in spring and winter, but increased during winter/spring storms. Increased SO₄^2- concentration was the main cause of episodic acidification during storms at Reedy Creek, but increased anion deficit indicates organic acids may contribute to episodic acidification during winter/spring storms. Changes in SO₄^2- concentration coincident with the hydrograph rise indicate quick routing of water through the watershed. Saturation overland flow appears to be the likely mechanism by which solutes are transported to the stream during storm flow.

The ability of microorganisms to degrade trace levels of the hydrochlorofluorocarbons HCFC-21 and HCFC-123 was investigated. Methanotroph-linked oxidation of HCFC-21 was observed in aerobic soils, and anaerobic degradation of HCFC-21 occurred in freshwater and salt marsh sediments. Microbial degradation of HCFC-123 was observed in anoxic freshwater and salt marsh sediments, and the recovery of 1,1,1-trifluoro-2-chloroethane indicated the involvement of reductive dechlorination. No degradation of HCFC-123 was observed in aerobic soils. In some experiments, HCFCs were degraded at low (parts per billion) concentrations, raising the possibility that bacteria in nature remove HCFCs from the atmosphere.

Quantitative analysis of geophysical logs in ground-water studies often involves at least as broad a range of applications and variation in lithology as is typically encountered in petroleum exploration, making such logs difficult to calibrate and complicating inversion problem formulation. At the same time, data inversion and analysis depend on inversion model formulation and refinement, so that log interpretation cannot be deferred to a geophysical log specialist unless active involvement with interpretation can be maintained by such an expert over the lifetime of the project. We propose a generalized log-interpretation procedure designed to guide hydrogeologists in the interpretation of geophysical logs, and in the integration of log data into ground-water models that may be systematically refined and improved in an iterative way. The procedure is designed to maximize the effective use of three primary contributions from geophysical logs: (1) The continuous depth scale of the measurements along the well bore; (2) The in situ measurement of lithologic properties and the correlation with hydraulic properties of the formations over a finite sample volume; and (3) Multiple independent measurements that can potentially be inverted for multiple physical or hydraulic properties of interest. The approach is formulated in the context of geophysical inversion theory, and is designed to be interfaced with surface geophysical soundings and conventional hydraulic testing. The step-by-step procedures given in our generalized interpretation and inversion technique are based on both qualitative analysis designed to assist formulation of the interpretation model, and quantitative analysis used to assign numerical values to model parameters. The approach bases a decision as to whether quantitative inversion is statistically warranted by formulating an over-determined inversion. If no such inversion is consistent with the inversion model, quantitative inversion is judged not possible with the given data set. Additional statistical criteria such as the statistical significance of regressions are used to guide the subsequent calibration of geophysical data in terms of hydraulic variables in those situations where quantitative data inversion is considered appropriate.

Modeling of ground water infiltration and movement in the wellhead area is a critical part of an effective wellhead protection program. Such models depend on an accurate description of the aquifer in the wellhead area so that reliable estimates of contaminant travel times can be used in defining a protection area. Geophysical and hydraulic measurements in boreholes provide one of the most important methods for obtaining the data needed to specify wellhead protection measures. Most effective characterization of aquifers in the wellhead vicinity results when a variety of geophysical and hydraulic measurements are made where geophysical measurements can be calibrated in terms of hydraulic variables, and where measurements are made at somewhat different scales of investigation. The application of multiple geophysical measurements to ground water flow in the wellhead area is illustrated by examples in alluvial, fractured sedimentary, and fractured crystalline rock aquifers. Data obtained from a single test well are useful, but cannot indicate how conductive elements in the aquifer are connected to form large-scale flow paths. Geophysical and hydraulic measurements made in arrays of observation boreholes can provide information about such large-scale flow paths, and are especially useful in specifying aquifer properties in wellhead protection studies.

Parkhurst, D.L., Christenson, Scott, and Breit, G.N., 1996, Ground- water quality assessment of the central Oklahoma aquifer, Oklahoma--Geochemical and geohydrologic investigations: U.S. Geological Survey Water-Supply Paper 2357C, 101 p.

Ground-water samples, core samples, and hydrologic measurements were obtained in the Central Oklahoma Aquifer as part of the pilot National Water-Quality Assessment Program. This report examines ground-water recharge and discharge, the potentiometric surface, the chemical and isotopic composition of ground water, and the abundances and textures of minerals in core materials to determine the rates and directions of ground-water flow and the geochemical reactions occurring within the aquifer.

A study was conducted in 1992 to assess the effects of anthropogenic activities and land use on the water quality of the San Joaquin River and its major tributaries. This study focused on pesticides and organic contaminants, looking at distributions of contaminants in water, bed and suspended sediment, and the bivalve Corbicula fluminea. Results indicated that this river system is affected by agricultural practices and urban runoff. Sediments from Dry Creek contained elevated concentrations of polycyclic aromatic hydrocarbons (PAHs), possibly derived from urban runoff from the city of Modesto; suspended sediments contained elevated amounts of chlordane. Trace levels of triazine herbicides atrazine and simazine were present in water at most sites. Sediments, water, and bivalves from Orestimba Creek, a westside tributary draining agricultural areas, contained the greatest levels of DDT (1,1,1-trichloro-2,2-bis[p-chlorophenyl]ethane), and its degradates DDD (1,1-dichloro-2,2-bis[p-chlorophenyl]ethane), and DDE (1,1-dichloro-2,2-bis[p-chlorophenyl]ethylene). Sediment adsorption coefficients (K_oc, and bioconcentration factors (BCF) in Corbicula of DDT, DDD, and DDE at Orestimba Creek were greater than predicted values. Streams of the western San Joaquin Valley can potentially transport significant amounts of chlorinated pesticides to the San Joaquin River, the delta, and San Francisco Bay. Organochlorine compounds accumulate in bivalves and sediment and may pose a problem to other biotic species in this watershed.

A preliminary assessment was made in 1992 of chlorinated organic compounds in sediments and in livers of striped bass from the San Francisco Bay-Delta Estuary. Samples of sediment and striped bass livers contained DDT (ethane, 1,1,1-trichloro-2,2-bis (p-chlorophenyl)-) and its degradation products, DDD (ethane, 1,1-dichloro-2,2-bis(p-chlorophenyl)-) and DDE (ethylene, 1,1-dichloro-2,2-bis (p-chlorophenyl)-); PCBs (polychlorinated biphenyls); alpha and gamma chlordane, and cis and trans nonachlor. In addition, the livers of striped bass contained small concentrations of DCPA (dimethyl tetrachloroterephthalate), a pre-emergent herbicide. Agricultural run-off from the Sacramento and San Joaquin Rivers, as well as atmospheric deposition, are probably responsible for a low chronic background of DDT in sediments throughout San Francisco Bay. Larger concentrations of DDT in sediment near Richmond in the Central Bay, and Coyote Creek in the South Bay may be derived from point sources. Ratios of pentachloro isomers of PCBs to hexachloro isomers in the South Bay sediments were different from those in the Central and North Bay, suggesting either differences in microbial activity in the sediments or different source inputs of PCBs. Concentrations of alpha chlordane in livers of striped bass were greater than those of gamma chlordane, which suggests a greater environmental stability and persistence of alpha chlordane. Trans nonachlor, a minor component of technical chlordane, was present in greater concentrations than alpha and gamma chlordane and cis nonachlor. Trans nonachlor is more resistant to metabolism than alpha and gamma chlordane and cis nonachlor, and serves as an environmentally stable marker compound of chlordane contamination in the estuary. Chlorinated organic compounds have bioaccumulated in the livers of striped bass. These compounds may contribute to the decline of the striped bass in San Francisco Bay-Delta Estuary.

The distribution and fate of chlorinated pesticides, biomarkers, and polycyclic aromatic hydrocarbons (PAHs) in surficial sediments along a contamination gradient in the Lauritzen Canal and Richmond Harbor in San Francisco Bay was investigated. Compounds were identified and quantified using gas chromatography-ion trap mass spectrometry. Biomarkers and PAHs were derived primarily from weathered petroleum. DDT was reductively dechlorinated under anoxic conditions to DDD and several minor degradation products, DDMU, DDMS, and DDNU. Under aerobic conditions, DDT was dehydrochlorinated to DDE and DBP. Aerobic degradation of DDT was diminished or inhibited in zones of high concentration, and increased significantly in zones of lower concentration. Other chlorinated pesticides identified in sediment included dieldrin and chlordane isomers. Multivariate analysis of the distributions of the DDTs suggested that there are probably two sources of DDD. In addition, DDE and DDMU are probably formed by similar mechanisms, i.e. dehydrochlorination. A steep concentration gradient existed from the Canal to the Outer Richmond Harbor, but higher levels of DDD than those found in the remainder of the Bay indicated that these contaminants are transported on particulates and colloidal organic matter from this source into San Francisco Bay. Chlorinated pesticides and PAHs may pose a potential problem to biota in San Francisco Bay.

Mountain glaciers, because of their small size, are usually close to equilibrium with the local climate and thus should provide a test of whether temperature oscillations in Greenland late in the last glacial period are part of global-scale climate variability or are restricted to the North Atlantic region. Correlation of cosmogenic chlorine-36 dates on Sierra Nevada moraines with a continuous radiocarbon-dated sediment record from nearby Owens Lake shows that Sierra Nevada glacial advances were associated with Heinrich events 5, 3, 2, and 1.

A mathematical model (WETSIM 2.0) was used to simulate wetland hydrology and vegetation dynamics over a 32-yr period (1961-1992) in a North Dakota prairie wetland. A hydrology component of the model calculated changes in water storage based on precipitation, evapotranspiration, snowpack, surface runoff, and subsurface inflow. A spatially explicit vegetation component in the model calculated changes in distribution of vegetative cover and open water, depending on water depth, seasonality, and existing type of vegetation. The model reproduced four known dry periods and one extremely wet period during the three decades. One simulated dry period in the early 1980s did not actually occur. Simulated water levels compared favorably with continuous observed water levels outside the calibration period (1990-1992). Changes in vegetative cover were realistic except for years when simulated water levels were significantly different than actual levels. These generally positive results support the use of the model for exploring the effects of possible climate changes on wetland resources.

Trihalomethane formation potentials were determined for the chlorination of water samples from the Mississippi, Missouri, and Ohio Rivers. Samples were collected during the summer and fall of 1991 and the spring of 1992 at 12 locations on the Mississippi from New Orleans, LA, to Minneapolis, MN, and on the Missouri and Ohio 1.6 km upstream from their confluences with the Mississippi. Formation potentials were determined as a function of pH and initial free-chlorine concentration. Chloroform concentrations decreased with distance downstream and approximately paralleled the decrease of the dissolved organic-carbon concentration. Bromide concentrations were 3.7-5.7 times higher for the Missouri and 1.4-1.6 times higher for the Ohio than for the Mississippi above their confluences, resulting in an overall increase of the bromide concentration with distance downstream. Variations of the concentrations of the brominated trihalomethanes with distance downstream approximately paralleled the variation of the bromide concentration. Concentrations of all four trihalomethanes increased as the pH increased. Concentrations of chloroform and bromodichloromethane increased slightly and the concentration of bromoform decreased as the initial free-chlorine concentration increased; the chlorodibromomethane concentration had little dependence on the free-chlorine concentration.

Water samples were collected from 12 sites along the Mississippi river from Minneapolis to New Orleans during summer and autumn 1991 and spring 1992. Trihalomethane (THM) and non-purgeable total organic-halide (NPTOX) formation potentials were determined as a function of pH (6.72-9.7) and initial free-chlorine concentration at 25C for a reaction time of 7 d. The THM and NPTOX formation potentials increased with distance upstream. The THM formation potential increased as pH increased but the NPTOX decreased with increasing pH. Both THM and NPTOX formation potentials increased with increasing initial free-chlorine concentration and increasing dissolved organic carbon concentration. The THM and NPTOX formation potentials were largest for summer samples and smallest for spring samples. The implications for water treatment are discussed.

A system of functional utilities and computer routines, collectively identified as the Time-Dependent Data System CI DDS), has been developed and documented by the U.S. Geological Survey. The TDDS is designed for processing time sequences of discrete, fixed-interval, time-varying geophysical data--in particular, hydrologic data. Such data include various, dependent variables and related parameters typically needed as input for execution of one-, two-, and three-dimensional hydrodynamic/transport and associated water-quality simulation models. Such data can also include time sequences of results generated by numerical simulation models. Specifically, TDDS provides the functional capabilities to process, store, retrieve, and compile data in a Time-Dependent Data Base (TDDB) in response to interactive user commands or pre-programmed directives. Thus, the TDDS, in conjunction with a companion TDDB, provides a ready means for processing, preparation, and assembly of time sequences of data for input to models; collection, categorization, and storage of simulation results from models; and intercomparison of field data and simulation results.

The TDDS can be used to edit and verify prototype, time-dependent data to affirm that selected sequences of data are accurate, contiguous, and appropriate for numerical simulation modeling. It can be used to prepare time-varying data in a variety of formats, such as tabular lists, sequential files, arrays, graphical displays, as well as line-printer plots of single or multiparameter data sets. The TDDB is organized and maintained as a direct-access data base by the TDDS, thus providing simple, yet efficient, data management and access. A single, easily used, program interface that provides all access to and from a particular TDDB is available for use directly within models, other user-provided programs, and other data systems. This interface, together with each major functional utility of the TDDS, is described and documented in this report.

Hydrologic and water-quality data were collected at a precipitation-collection station and from two small watersheds on Catoctin Mountain, north- central Maryland, as part of an investigation of episodic acidification and its effects on streamwater quality. Data were collected from June 1990 through December 1993. Descriptions of the water shed instrumentation, data-collection techniques, and laboratory methods used to conduct the studies are included. Data that were collected on precipitation, throughfall, soil water, ground water, and streamwater during base flow and stormflow indicate that the streams undergo episodic acidification during storms. Both streams showed decreases in pH to less than 5.0 standard units during stormflow. The acid-neutralizing capacity (ANC) of both streams decreased during stormflow, and the ANC of one of the streams, Bear Branch, became negative. The chemistries of the different types of waters that were sampled indicate that shallow subsurface water with minimal residence time in the watersheds is routed to the streams to become stormflow and is the cause of the episodic acidification observed. Three-component hydrograph separations were performed on the data collected during several storms in each watershed. The hydrograph separations of all of the storms indicate that throughfall contributed 0 to 50 percent of the stormflow, soil water contributed 0 to 80 percent, and ground water contributed 20 to 90 percent. The results of the hydrograph separations indicate that, in general, the watershed with higher hydraulic gradients tends to have shallower and shorter flow paths than the watershed with lower hydraulic gradients.

Hydrologic and water-quality data were collected from a precipitation-collection station and from two small watersheds on Catoctin Mountain, north-central Maryland, as part of investigations of acidic deposition and episodic acidification, and their effects on streamwater quality. Detailed descriptions of the site instrumentation in the watersheds, field data-collection techniques, and laboratory methods used to conduct the studies are included. Data that were collected on precipitation, throughfall, soil water, ground water, streamwater, and other surface and ground waters sampled during biannual synoptic surveys are given in tables. Data collected since October 1987 from one of the streamwater-quality monitoring sites and data collected since March 1988 from one of the ground-water quality monitoring sites are presented. Additional data collected since January 1987 from the precipitation station and data collected since June 1990 from all of the other water-quality monitoring sites are presented. Hydrologic data include tables of precipitation and throughfall quantities, streamflow, and synoptic measurements of ground-water levels. Selected hydrologic data are shown in graphs.

Although oil spills are becoming increasingly common, few geochemical studies have investigated in detail the production and fate of methane (CH₄), one of the products of the biodegradation of oil. We investigated the CH₄ geochemistry of a site near Bemidji, Minnesota, where crude oil from a pipeline break floats on the water table in a shallow sand and gravel aquifer.

Dye injection studies and direct velocity and water-level measurements were made in macrophyte stands and adjacent channels in order to observe the effects of the macrophyte stand on flow and mass exchange in the tidal Potomac River. During the summer, dense stands of submersed aquatic plants cover most shoals <2 m deep. Continuous summertime water-level records within a submersed aquatic plant stand and in the adjacent channel revealed time-varying gradients in water-surface elevation between the two areas. Water-level gradients are created by differing rates of tidal water-level change in vegetated and unvegetated areas. Results were consistent with the idea that on a rising tide the water was slower to enter a macrophyte stand, and on a falling tide it was slower to leave it. Differences in water elevation between the stand and the open channel generated components of velocity in the stand that were at right angles to the line of flow in the channel. Seasonal differences in flow speed and direction over the shoals indicate substantial differences in resistance to flow as a result of the vegetation.

The friction-slope term in the unsteady open-channel flow equations is examined using two numerical models based on different formulations of the governing equations and employing different solution methods. The purposes of the study are to analyze, evaluate, and demonstrate the behavior of the term in a set of controlled numerical experiments using varied types and combinations of boundary conditions. Results of numerical experiments illustrate that a given model can respond inconsistently for the identical resistance-coefficient value under different types and combinations of boundary conditions. Findings also demonstrate that two models employing different dependent variables and solution methods can respond similarly for the identical resistance-coefficient value under similar types and combinations of boundary conditions. Discussion of qualitative considerations and quantitative experimental results provides insight into the proper treatment, evaluation, and significance of the friction-slope term, thereby offering practical guidelines for model implementation and calibration.

The transport of many solutes in groundwater is dependent upon the relative rates of physical flow and microbial metabolism. Quantifying rates of microbial processes under subsurface conditions is difficult and is most commonly approximated using laboratory studies with aquifer materials. In this study, we measured in situ rates of denitrification in a nitrate-contaminated aquifer using small-scale, natural-gradient tracer tests and compared the results with rates obtained from laboratory incubations with aquifer core material. Activity was measured using the acetylene block technique. For the tracer tests, co-injection of acetylene and bromide into the aquifer produced a 30 mM increase in nitrous oxide after 10 m of transport (23-30 days). An advection-dispersion transport model was modified to include an acetylene-dependent nitrous oxide production term and used to simulate the tracer breakthrough curves. The model required a 4-day lag period and a relatively low sensitivity to acetylene to match the narrow nitrous oxide breakthrough curves. Estimates of in situ denitrification rates were 0.60 and 1.51 nmol of N₂O produced cm^-3 aquifer day^-1 for two successive tests. Aquifer core material collected from the tracer test site and incubated as mixed slurries in flasks and as intact cores yielded rates that were 1.2-26 times higher than the tracer test rate estimates. Results with the coring-dependent techniques were variable and subject to the small-scale heterogeneity within the aquifer, while the tracer tests integrated the heterogeneity along a flow path, giving a rate estimate that is more applicable to transport at the scale of the aquifer.

The Mammoth Corridor in and adjacent to Yellowstone National Park encompasses a N-S alignment of geothermal features that extends from the Norris Geyser Basin adjacent to the Yellowstone caldera through Mammoth Hot Springs to the Corwin Springs Known Geothermal Resources Area (KGRA). Thermal springs in this region discharge water that ranges from Na-K-Cl, silica-depositing type to Ca-Na-HCO₃-SO₄, travertine-depositing type. Although only a few relatively shallow wells have been drilled in the corridor, the region is of special interest because of the environmental issues associated with potential geothermal development adjacent to Yellowstone National Park. The U.S. Geological Survey conducted an intensive hydrogeologic study of this region during 1988-1990 and continued to collect hydrologic and geophysical data until 1994. The results of these investigations document the rates of discharge of thermal water and heat within the corridor, evidence for a magmatic heat source beneath the Mammoth Hot Springs area, and evidence for separate geothermal systems associated with Mammoth Hot Springs and with thermal waters discharging in the KGRA in the vicinity of La Duke Hot Springs. These investigations also indicate that limited development of the 70 degree C geothermal resource in the La Duke area would not affect thermal springs in Yellowstone National Park.

Sewage-contaminated groundwater currently discharges to Ashumet Pond, located on Cape Cod, Massachusetts. Phosphate concentrations as high as 60 mu mol l^-1 have been measured in groundwater entering Ashumet Pond, and there is concern that the rate of eutrophication could increase. Phosphate in the sewage plume is sorbed by aquifer sediment; the amount is a function of phosphate concentration and pH. A nonelectrostatic surface-complexation model coupled with a one-dimensional solute-transport code was used to simulate sorption and desorption of phosphate in laboratory column experiments. The model simulated sorption of phosphate reasonably well, although the slow rate of approach to complete breakthrough indicated a nonequilibrium process that was not accounted for in the solute-transport model. The rate of phosphate desorption in the column experiments was relatively slow. Phosphate could still be measured in effluent after 160 pore volumes of uncontaminated groundwater had been flushed through the columns. Desorption was partly a function of the slowly decreasing pH in the columns and could be modeled quantitatively. Disposal of sewage at this site is scheduled to stop in 1995; however, a large reservoir of sorbed phosphate exists on aquifer sediment upgradient from Ashumet Pond. Computer simulations predict that desorption of phosphate could result in contamination of Ashumet Pond for decades.

The 2 800-km reach of the Mississippi River between Minneapolis, MN, and New Orleans, LA, was examined for the occurrence and fate of linear alkylbenzene sulfonate (LAS), a common anionic surfactant found in municipal sewage effluents. River water and bottom sediment were sampled in the summer and fall of 1991 and in the spring of 1992. LAS was analyzed using solid-phase extraction/derivatization/gas chromatography/mass spectrometry. LAS was present on all bottom sediments at concentrations ranging from 0.01 to 20 mg/kg and was identified in 21% of the water samples at concentrations ranging from 0.1 to 28.2 mg/L. The results indicate that LAS is a ubiquitous contaminant on Mississippi River bottom sediments and that dissolved LAS is present mainly downstream from the sewage outfalls of major cities. The removal of the higher LAS homologs and external isomers indicates that sorption and biodegradation are the principal processes affecting dissolved LAS. Sorbed LAS appears to degrade slowly.

Five methods of developing regional regression models to estimate flood characteristics at ungaged sites in Arkansas are examined. The methods differ in the manner in which the State is divided into subregions. Each successive method (A to E) is computationally more complex than the previous method. Method A makes no subdivision. Methods B and C define two and four geographic subregions, respectively. Method D uses cluster/discriminant analysis to define subregions on the basis of similarities in watershed characteristics. Method E, the new region of influence method, defines a unique subregion for each ungaged site. Split-sample results indicate that, in terms of root-mean-square error, method E (38 percent error) is best. Methods C and D (42 and 41 percent error) were in a virtual tie for second, and methods B (44 percent error) and A (49 percent error) were fourth and fifth best.

Laboratory experiments employing radiotracer methodology were conducted to determine the assimilation efficiencies from ingested natural seston, the influx rates tram the dissolved phase and the efflux rates of 6 trace elements (Ag, Am, Cd, Co, Se and Zn) in the mussel Mytilus edulis. A kinetic model was then employed to predict trace element concentration in mussel tissues in 2 locations for which mussel and environmental data are well described: South San Francisco Bay (California, USA) and Long Island Sound (New York, USA). Assimilation efficiencies from natural seston ranged from 5 to 18% for Ag, 0.6 to 1% for Am, 8 to 20% for Cd, 12 to 16% for Co, 28 to 34% for Se, and 32 to 41% for Zn. Differences in chlorophyll a concentration in ingested natural seston did not have significant impact on the assimilation of Am, Co, Se and Zn. The influx rate of elements from the dissolved phase increased with the dissolved concentration, conforming to Freundlich adsorption isotherms. The calculated dissolved uptake rate constant was greatest for Ag, followed by Zn > Am approximately Cd > Co > Se. The estimated absorption efficiency from the dissolved phase was 1.53% for Ag, 0.34% for Am, 0.31% for Cd, 0.11% for Co, 0.03% for Se and 0.89% for Zn. Salinity had an inverse effect on the influx rate from the dissolved phase and dissolved organic carbon concentration had no significant effect on trace element uptake. The calculated efflux rate constants for all elements ranged from 1.0 to 3.0% d super(-1). The route of trace element uptake (food vs dissolved) and the duration of exposure to dissolved trace elements (12 h vs 6 d) did not significantly influence trace element efflux rates. A model which used the experimentally determined influx and efflux rates for each of the trace elements, following exposure from ingested food and from water, predicted concentrations of Ag, Cd, Se and Zn in mussels that were directly comparable to actual tissue concentrations independently measured in the 2 reference sites in national monitoring programs. Sensitivity analysis indicated that the total suspended solids lead, which can affect mussel feeding activity, assimilation, and trace element concentration in the dissolved and particulate phases, can significantly influence metal bioaccumulation for particle-reactive elements such as Ag and Am. For all metals, concentrations in mussels are proportionately related to total metal lead in the water column and their assimilation efficiency from ingested particles. Further, the model predicted that over 96% of Se in mussels is obtained from ingested food, under conditions typical of coastal waters. For Ag, Am, Cd, Co and Zn, the relative contribution from the dissolved phase decreases significantly with increasing trace element partition coefficients for suspended particles and the assimilation efficiency in mussels of ingested trace elements; values range between 33 and 67% for Ag, 5 and 17% for Am, 47 and 82% for Cd, 4 and 30% for Co, and 17 and 51% for Zn.

The kinetics of covalent binding of aniline to dissolved organic matter (DOM) at concentrations typically found in natural aquatic ecosystems (1-50 mg carbon per litre) were studied using humic and fulvic isolates from theSuwannee River, Georgia, and International Humic Substances Society soil humic and fulvic acid isolates. Data indicated that approximately 10 per cent of the covalent binding sites associated with Suwannee River fulvic acid were highly reactive and tha the reaction rate decreased with decreasing pH. The covalent binding of aniline was inhibited by pre-treatment of the fulvic acid with hydrogen sulfide, hydroxylamine or sodium borohydride.

The role of solute diffusion between ground water and granite in the Mirror Lake drainage area was evaluated by direct observation, experiment, and inference. The outcrops display ubiquitous Liesegang bands associated with fractures that clearly indicate the activity of diffusion in this system in the past. Laboratory experiments determined that the effective diffusion coefficient for ¹³⁷Cs was approximately 6 x 10-13 m²/s in granite from Mirror Lake. The ¹³⁷Cs penetrated to a depth of approximately 7 mm in 101 days, demonstrating the potential for rapid diffusion in this system. Porosities of 32 granite samples averaged of 1.46 percent with a range of 1.07 to 2.32 percent. Measurements of carbon-isotope of ground water in the fractures suggest that calcite, identified in the granite, is dissolving and the bicarbonate generated is diffusing to the fractures; that is, a significant amount of the dissolved solids in the water in the fractures are derived from diffusion of weathering products from the rock matrix. These observations taken together are consistent with the interpretation that diffusion is a major process controlling solutes in this fractured granite aquifer.