ISSN:0017-467X    

DOI:10.1111/j.1745-6584.1994.tb00910.x

出版商:Blackwell Publishing Ltd    

年代:1994

数据来源: WILEY

摘要:

AbstractA common effort involved in the remediation of contamination by petroleum hydrocarbons in porous media is the monitoring and volume estimation of the immiscible hydrocarbon fluid. The apparent free product thickness indicated by a standard monitoring well is typically much greater than the actual free product thickness in the surrounding soil. This results in significant errors in estimates of drainable fluid (product) volume, which in turn leads to inaccuracies (i.e., over design, improper pump sizing, etc.) with remediation system design.An equation to predict actual thickness was developed using heterogeneous fluid flow mechanics and hydrostatics. This equation is: tg= t(l – Sg) ‐ ha, where tg= actual formation free product thickness, t = apparent (well bore product) thickness, Sg= specific gravity of petroleum hydrocarbon (gasoline in this research), and ha= distance between the ground‐water table and the free product in the formation.The developed theory was compared to data collected from a physical model which simulated field conditions. The theory was used to estimate product thickness in the model, and then these estimates were statistically tested for accuracy. The theoretical slope, (1 – Sg), was not statistically different from the regression slope at test levels of α= 0.05 and α= 0.01, while the theoretical intercept (ha) was statistically different at α– 0.05 and α– 0.01. The discrepancy between the theoretical intercept and the regression intercept was probably due to either an incorrect assumption that ha= hc(hc= average wetting capillary rise), or an incorrect laboratory measurement of hc.The effects of water‐table fluctuations were also studied. A rising water table caused a decrease in apparent thickness and an increase in actual thickness, and vice versa. No theoretical equation was developed for the effects of water‐table fluctuations, but it was reasoned that hysteresis of the saturation curve and adsorption were the primary factors in the observed trends.Finally, the developed theoretical equation was compared to the results of previously published predictive methods and experiments. The comparison was made by calculating percent error and using a chi‐square statistic. The developed theory was found to be the best predictor of actual product thickness for both labo

ISSN:0017-467X    

DOI:10.1111/j.1745-6584.1994.tb00911.x

出版商:Blackwell Publishing Ltd    

年代:1994

数据来源: WILEY

摘要:

AbstractThe paper discusses the use of the “pathline counting” approach to estimate solute concentrations at an extraction well using particle‐tracking flow models. The pathline counting approach is a unique method for predicting solute concentrations due to mixing. The pathline counting method using the semianalytical module RESSQC, contained within the WHPA model, was compared to MT3D, a numerical mass transport code, for a simple hypothetical problem. Results from the comparison indicate that results from the pathline counting approach were in reasonable agreement with the MT3D results. This approach was then applied to a field study, the Park City Bioremediation Project. The migration of a conservative tracer injected into three upgradient cells was simulated to estimate the concentration of solute at two pumping wells. The calculated concentrations compared favorably with measured field concentra

ISSN:0017-467X    

DOI:10.1111/j.1745-6584.1994.tb00912.x

出版商:Blackwell Publishing Ltd    

年代:1994

数据来源: WILEY

摘要:

AbstractAn in situ surfactant washing process was previously developed and successfully pilot‐tested at a field site contaminated with polychlorinated biphenyls (PCBs) and oils (Abdul et al., 1992). A second phase of the pilot study, conducted the following year at the same site, confirmed the technical viability of this process. The results of the phase II study surpassed the predictions of the long‐term performance of this technology based on the phase I results. Soil cores taken from the plot before the phase I and phase II studies and after the phase II study showed mass reduction and redistribution of the contaminants.A continuous column washing study using the same contaminated soil showed consistency with field results. The amounts of PCBs removed in the laboratory and field studies, respectively, are 7% and 10% after 5.7 pore volume (PV) washings (phase I) and 19% and 24% after 8 PV washings (phase II). More than 85% of the PCBs were removed from the laboratory soil column after 105 PV washings. This study supports the phase I study which indicates that the in situ surfactant washing process is a viable remediation technology for hydrophobic contamina

ISSN:0017-467X    

DOI:10.1111/j.1745-6584.1994.tb00913.x

出版商:Blackwell Publishing Ltd    

年代:1994

数据来源: WILEY

摘要:

AbstractThe vertical component of specific discharge was estimated across several depth zones in the aquifer system at Cañutillo, Texas. The specific discharge was estimated using temperature and thermal conductivity data from four observation wells bottoming near the base of the aquifer system.Specific discharge was first calculated (using a steady‐state model) from the slopes of conductive heat flow versus temperature plots. The heat flow plots for all four wells suggested a zone of downward ground‐water flow from ∼ 70 to ∼ 220 feet depth and one or two zones of upward flow from ∼ 800 to ∼ 220 feet depth. Head data from observation wells did support the conclusion of an upper zone of downflow as indicated by the temperature data; however, the head data suggested that ground water should presently be moving downward from ∼ 220 to ∼ 800 feet.A computer model was developed to determine if the temperatures may be remanent and therefore reflect prepumping ground‐water flow, particularly in the deep flow zone. Computer model estimates were made of during‐pumping downflow in the upper flow zone. For the deep flow zone computer model estimates were made of prepumping upflow. The computer model estimates and steady‐state vertical specific discharge estimates were similar for the upper zone having downward flow. In the deep zone prepumping upflow estimates modeled on the computer were the same order of magnitude as steady‐state heat flow specific discharge estimates. Hence, present temperature data may demonstrate a significant remanent groundwater flow component and therefore might be used to estimate past ve

ISSN:0017-467X    

DOI:10.1111/j.1745-6584.1994.tb00914.x

出版商:Blackwell Publishing Ltd    

年代:1994

数据来源: WILEY

摘要:

AbstractUnder many circumstances during an RCRA investigation, the EPA will not allow intrusive sampling of an SWMU (Solid Waste Management Unit, or “waste dump”). This presented difficulties in characterizing the waste within a closed landfill identified as an SWMU located on land owned by Purdue University, 3 kilometers northwest of the West Lafayette (Indiana) campus because there were few records of what was discarded into this public dump during its operation from 1958–87. Chemical and isotopic data of shallow ground water and surface runoff provided a realistic means to infer movement of contaminants away from the landfill.Leachate is characterized by high concentrations of Na+and Cl−(∼ 4mN), sulfate (∼ 8 mN), total hardness (Ca+2+ Mg+2∼ 19 mN), and alkalinity (∼ 12mN). Disposal of road salt is responsible for the NaCl concentrations, while dissolution of concrete explains the high Ca+2, Mg+2, and alkalinity values for waters having a pC02only slightly above the regional level for ground waters (10−1.82atm).Deep and shallow uncontaminated ground waters in the vicinity of the SWMU have an average δ34S value of 0.5 per mil (CDT). Since the δ34S of leachate lies between + 10 and + 17 per mil (CDT), the high sulfate concentrations are most likely due to dissolution of interred plasterboard consisting primarily of gypsum.Tritium activities for the leachate in two shallow wells emplaced 15 meters from the toe of the SWMU ranged from 55 to 85 TU. Correcting for half‐lives and correlating these activities to historical records of tritium in rainfall as measured at Chicago, these3H activities suggest that the travel time between the SWMU interior and these wells is be

ISSN:0017-467X    

DOI:10.1111/j.1745-6584.1994.tb00915.x

出版商:Blackwell Publishing Ltd    

年代:1994

数据来源: WILEY

摘要:

AbstractSpatial variability of hydrologic properties was quantified for a nonwelded‐to‐welded ash flow tuff at Yucca Mountain, Nevada, the potential site of a high‐level, nuclear waste repository. Bulk density, porosity, saturated hydraulic conductivity, and sorptivity were measured on core specimens collected from outcrops on a grid that extended vertically through the entire unit thickness and horizontally 1.3 km in the direction of ash transport from the volcanic vent. A strong, geologically determined (deterministic) vertical trend in properties was apparent that correlated with visual trends in degree of welding observed in the outcrop. The trend was accurately described by simple regression models based on stratigraphic elevation (vertical distance from the base of the unit divided by unit thickness). No significant horizontal trends in properties were detected along the length of the transect. The validity of the developed model was tested by comparing model predictions with measured porosity values from additional outcrop sections and boreholes that extended 3000 m north, 1500 m northeast, and 6000 m south of the study area. The model accurately described vertical porosity variations except for locations very close to the source caldera, where the model underpredicted porosity in the upper half of the section. The presence of deterministic geologic trends, such as those demonstrated for an ash flow unit in this study, can simplify the collection of site characterization data and the development of site‐scale

ISSN:0017-467X    

DOI:10.1111/j.1745-6584.1994.tb00916.x

出版商:Blackwell Publishing Ltd    

年代:1994

数据来源: WILEY

摘要:

AbstractSlug and packer tests provide data for determining the local value of hydraulic conductivity in aquifers. However, most techniques commonly used for test interpretation disregard any anisotropy of the tested formation. Using dimensional analysis, an extension of available methods for interpretation of slug and packer tests to anisotropic conditions is developed. The technique is a transformation of the radial dimension via the well radius, and involves dividing the true well radius by the square root of the anisotropy ratio which must be specified a priori. This technique can be applied to formulae, computer programs, graphs, or tabulated data available for confined or unconfined isotropic conditions. It also reduces the number of computer computations. As an example, the technique is used to extend the Bouwer and Rice (1976) method, which was developed for isotropic conditions, to anisotropic conditions. A comparison of the Bouwer and Rice (1976) method corrected for anisotropy, with the Widdowson et al. (1990) method, yields horizontal hydraulic conductivities differing by 8% (in one typical case). Ignoring anisotropy for the same case results in a 44% underestimation of horizontal conductivity.

ISSN:0017-467X    

DOI:10.1111/j.1745-6584.1994.tb00917.x

出版商:Blackwell Publishing Ltd    

年代:1994

数据来源: WILEY

摘要:

AbstractThe contamination by salt‐water intrusion affecting the coastal aquifer at Castell de Ferro (Granada, Spain) is especially influenced by the coexistence of two coastline lithological formations with highly differing transmissivity (highly karstified limestones and detrital sediments). Despite not having a very high intrusion volume (0.1–0.2 hm3/year), the aquifer does have a high saline content. In order to explain this phenomenon, we first of all modeled the ground‐water flow (MODFLOW) in the aquifer. Two models were then used for the fresh‐water/salt‐water contact: one assuming a sharp interface (BADON2) and the other taking into account density‐dependent mass transport (MOCDENSE). Our conclusion is that the high saline content of the aquifer is due to the existence of a preferential route for ground‐water flow through the karstified materials in contact with the detrital materials and the sea. The detrital materials located beside the karstified massif therefore constitute the route for only 10%‐20% of the system's total outlet to the sea, which is not enough water to wash the salts resulting from salt‐water intrusion. We also find that in this type of aquifer two‐dimensional transport models do not provide the most realistic results, since the effect in a direction transverse to the ground‐water flow of very transmissive material cannot be simulated in this type of profile. In this case, and despite its limitations, better results are obtained using the

ISSN:0017-467X    

DOI:10.1111/j.1745-6584.1994.tb00918.x

出版商:Blackwell Publishing Ltd    

年代:1994

数据来源: WILEY

摘要:

AbstractThe exaggeration of light nonaqueous phase liquids (LNAPL), typically hydrocarbon, by monitoring wells is a well‐known problem that introduces significant errors in the estimation of recoverable hydrocarbon. Farr et al. (1990) and Lenhard and Parker (1990) show that significantly different volumes of hydrocarbon may produce the same thickness of hydrocarbon in a monitoring well, due to differences in the capillary characteristics between soil types. The purpose of our investigation was to evaluate the influence of local sediment variability on estimation of hydrocarbon volumes. Sediment samples from two sites underlain by a relatively homogeneous sandy deposit were collected within a small area. Capillary characteristic curves were determined for 10 samples from one site and 41 samples from the second site using a pressure plate. Grain‐size analysis was performed on all samples for which pressure plate data were available.The results show significant variability, even for small sites. For example, a hydrocarbon volume of three cm3/cm2could produce anywhere between 45 and 200 cm of hydrocarbon within an observation well. This suggests that use of an “average” soil sample to characterize hydrocarbon exaggeration, within even a very small site, can lead to substantial errors. It also suggests that maps of apparent hydrocarbon thickness can be extremely misleading, leading hydrologists to place remediation wells in areas of greatly exaggerated thicknesses produced by fine‐grained materials.Comparison between characteristic curves calculated from grain‐size analysis using the approach of Mishra et al. (1989) to those measured using the pressure plate shows poor correlation that introduces large errors into the estimated hydrocar

ISSN:0017-467X    

DOI:10.1111/j.1745-6584.1994.tb00919.x

出版商:Blackwell Publishing Ltd    

年代:1994

数据来源: WILEY