Browsing by Author "Sale, Tom, advisor"

Now showing 1 - 18 of 18

Open Access
Analysis of contaminant mass in place in transmissive and low-k zones
(Colorado State University. Libraries, 2020) Roads, Eric, author; Sale, Tom, advisor; Sharvelle, Sybil, committee member; Sutton, Sally, committee member
Contaminant hydrology has been challenged by the common perception of homogeneous subsurface media. Previous sampling methods neglect the importance of differentiating between transmissive and low-k zones. Cryogenic core collection is a high-resolution sampling technique that can highlight the occurrence of transmissive and low-k zones as well as the distribution of contaminants in transmissive and low-k zones. Cryogenic core collection uses a CSU patented process that preserves core samples downhole using liquid nitrogen. Frozen cores are shipped to CSU on dry ice and always kept at -80ᵒC. Cores are cut into subsamples and analyzed to determine geology, physical properties, contaminant concentrations, and microbial ecology. The data is processed into Excel™ and then stored in gINT™, a relational database. Herein, consideration is given to 390 feet of collected core from 31 boreholes from 5 hydrocarbon and 2 chlorinated solvent sites. Data analyses include comparisons within a site, intra-site comparisons, and between sites, inter-site comparisons. Tools are developed in gINT™ to automate transformation of collected data into vibrant visual graphical outputs. First, for every borehole, a graphic is generated that includes a comprehensive panel of geology, contaminants of concern and fluid saturations properly presented by depth. Building on this, distributions of contaminants as a function of transmissive or low-k zones are resolved. Lastly, key attributes of mass distribution are compared across individual sites (intra-site comparisons) and between sites (inter-site comparisons). Our analysis presents a first-ever quantification of distribution of contaminant mass in transmissive and low-k zones. The analysis begins with processing concentration data-by-depth to produce the total mass of contaminants in each borehole, the mass of contaminants in transmissive zones, and the mass of contaminants in low-k zones. The contaminant mass in a borehole is presented for each contaminant individually and as sum of all contaminants. The visualization of this data is not intuitive due to the ranges of contaminant mass in place. Hydrocarbons contaminated sites have contaminant masses that range from less than half a kilogram to about 30 kilograms of contaminants per m2. Chlorinated solvent contaminated sites have contaminant masses that vary from less than 240 micrograms to right under 2.5 kilograms of contaminants per m2. The data is processed such that boreholes and sites with broad ranges of conditions can be compared. Data is presented as percent of contaminant mass in transmissive zones by borehole; the percent of contaminant mass in low-k zones by borehole, the percent of borehole that is transmissive, and percent of borehole that is low-k. Unlike previous data that required a y-axis formatted to a log scale, this data is visualized on a plot with the y-axis set at 0-100%. The fraction of a borehole that is low-k ranges between 0% and 94% with a median value of 52%. Secondly, the fraction of total contaminant mass stored in low-k zone ranges from 1% to 96% with a median value of 46%. Illustrations of the tendency for mass storage in low-k zones are presented through difference in percent of borehole that is low-k and percent of contaminants in a borehole in low-k zones. The calculations defined a positive difference as preference for transmissive zones and a negative difference as preference for low-k zones. Data presented characterized the 18 hydrocarbon contaminated boreholes, 12 chlorinated solvent contaminated boreholes, and all 30 contaminated boreholes respectively. Key insights include • Hydrocarbon contaminated boreholes showed statistically significant preference for low-k zones if the unit difference of percent of borehole that is low-k and percent of contaminants in a borehole in low-k zones was less than -24%. • Chlorinated solvent contaminated boreholes showed statistically significant preference for low-k zones if the unit difference of percent of borehole that is low-k and percent of contaminants in a borehole in low-k zones was less than -11%. • Remediated chlorinated solvent boreholes presented a preference for low-k zones where their non-remediated counterparts showed preference for transmissive zones. • All contaminated boreholes showed statistically significant preference for low-k zones if the unit difference of percent of borehole that is low-k and percent of contaminants in a borehole in low-k zones was less than -19%. • As an example, this thesis provides a unique documentation of benzene persisting in low-k zones. The presence or absence of benzene in low-k zones will have a large implication with respect to the longevity of benzene in monitoring wells and the efficacy of remedial measures that address the longevity of benzene in monitoring wells. Overall, cryogenic core collection and advanced analytics provides a practical means of quantifying contaminant occurrence in transmissive and low-k zones and an improved basis for anticipating the benefits of site remedies.
Open Access
Combined source infrastructure assessment model
(Colorado State University. Libraries, 2012) Maurer, Anne, author; Sale, Tom, advisor; Bau, Domenico, committee member; Sanford, William, committee member
Integrated utilization of surface and groundwater is a promising strategy that has the potential to reduce the costs associated with water system infrastructure projects and improve the sustainability of yields from finite water resources. Planning and design of conjunctive use systems can be complicated. Key challenges include resolving the timing of withdrawals, timing of storage, sizing of infrastructure components, and efficiently estimating costs. A combined source infrastructure assessment model (CSIAM) has been developed in this study using a decision programming approach. The CSIAM is designed for single-and multi-source water systems including surface water-only, groundwater-only, and combined surface water and groundwater sources. Aquifer Storage and Recovery (ASR) via groundwater injection wells is a primary component of the CSIAM when there is surplus surface water available to store. Primary model inputs include project life-span, per capita demands, initial population, population growth rate, surface water treatment capacity, number of existing wells, inflows, reservoir stage-storage, evaporative and seepage losses, and unit costs for capital expenditures and operations and maintenance. Model outputs include project water demands, surface reservoir storage, volume of monthly surface water treatment, groundwater extraction and/or injection volumes, cumulative groundwater extraction and/or injection volumes, number of wells, capital costs, operation and maintenance costs, life-cycle costs, and present value. The model is demonstrated via analysis of three scenarios involving groundwater-only, combined groundwater and surface water, and surface water-only. The scenarios are predicated on data provided by the town of Castle Rock, Colorado. While the Town of Castle Rock provides a basis for applying the model, the results should not be viewed as having direct bearing on future actions in the Town of Castle Rock. Many of the key issues that will ultimately drive the Town's water supply plans are not included in this analysis. Use of a combined groundwater and surface water system is $91 million and $231 million less than a groundwater-only system and streamflow-only system, respectively. Furthermore, the use of a combined groundwater and surface water system reduces groundwater depletion by 55%, relative to a groundwater-only system. In addition, a total of 107 pumping wells will need to be installed in a groundwater-only system versus 67 pumping wells in a combined groundwater/surface water system. Both deterministic and stochastic inputs are used in the model, wherein the principle stochastic input is urban irrigation demands. The differences between results using deterministic and stochastic inputs vary depending on the output. In general, analyses using stochastic inputs lead to a need for infrastructure with greater capacities and higher costs. The CSIAM also can be used to resolve costs as a function of groundwater depletion by testing different surface water treatment plant sizes.
Open Access
Continuous NAPL loss rates using subsurface temperatures
(Colorado State University. Libraries, 2015) Stockwell, Emily Beth, author; Sale, Tom, advisor; Blotevogel, Jens, committee member; Ham, Jay, committee member
To view the abstract, please see the full text of the document.
Open Access
Effects of capping material on longevity of degradable contaminants in sediments
(Colorado State University. Libraries, 2017) Campbell, Calista Emily, author; Sale, Tom, advisor; Blotevogel, Jens, committee member; Butters, Greg, committee member
To view the abstract, please see the full text of the document.
Open Access
Enhancing oleophilic biobarriers for non-tidal sediments impacted with petroleum hydrocarbons
(Colorado State University. Libraries, 2021) DeBiasi, Marina Ann, author; Scalia, Joseph, IV, advisor; Sale, Tom, advisor; Sutton, Sally, committee member
The objective of this study is to develop tools to prevent petroleum hydrocarbons trapped in non-tidal sediments from causing detrimental effects such as sheens. Oleophilic biobarriers (OBBs) provide a robust, low-cost solution for managing petroleum hydrocarbon contamination at groundwater-surface water interfaces in tidal zones but are untested in non-tidal zones. This study evaluates enhanced OBB remedies for petroleum hydrocarbon contamination in non-tidal zones by incorporating amendments within the OBB. The amended OBB is intended to serve as an engineered bioremediation tool to enhance microbial growth and degradation of petroleum hydrocarbons by supplying the system with a resource of electron donors and nutrients while simultaneously mitigating petroleum hydrocarbon releases to surface water. Complementary laboratory and field studies were conducted to test non-tidal OBBs (NOBBs) with six amendment types: (1) hematite (H), (2) greensand (GS), (3) greensand + hematite (GS+H), (4) gypsum (GYP), (5) hematite + greensand + gypsum (ALL), and (6) blank (B). The laboratory study was constructed as a series of chemostats using sediment and water samples from the field site. This study observed the productivity of petroleum hydrocarbon degradation through biweekly headwater extractions analyzing alkalinity, dissolved inorganic carbon (DIC), and pH as well as continuously monitored oxidation reduction potential (ORP). Results from these tests indicated that the GYP amendment was most effective in degrading petroleum hydrocarbons while the B and ALL amendments were least effective. However, all systems exhibited increased effluent DIC characteristic of enhanced petroleum hydrocarbon degradation. The field study was constructed as a series of OBB disks deployed atop petroleum hydrocarbon impacted sediments in a non-tidal setting. Results from the laboratory and field study illustrated abundant microbial growth after six months. The NOBBs with the top three highest numbers of microbial abundance were found in the field (F): F-GS+H, F-B, and F-GS. The overall results of both lab and field studies suggest that NOBBs, whether amended or not, provide effective media for petroleum hydrocarbon-degrading microorganisms. This study illustrates the promise of the non-tidal OBB as a bioreactive barrier for petroleum hydrocarbon impacted sediments. Further study is needed to evaluate the rate of petroleum hydrocarbon degradation in a non-tidal OBB relative to the rate of loading.
Open Access
Exposing new compositional coverage of weathered petroleum hydrocarbons through a tiered analytical approach
(Colorado State University. Libraries, 2019) Bojan, Olivia, author; Blotevogel, Jens, advisor; Sale, Tom, advisor; Denef, Karolien, committee member
Petroleum hydrocarbon spills are a widespread source of contamination that may threaten ecosystem services and human health, especially due to modern society's dependence on petroleum-based fuels. Remediation mainly relies on natural source zone depletion (NSZD) processes, which may generate partially oxidized transformation products of the spilled hydrocarbons through weathering or biodegradation processes. These byproducts containing one or more heteroatoms (N, S or O) – referred to as "polar hydrocarbons" – have increased water solubility and mobility in the environment. The unknown fate and toxicity of these complex mixtures of polar metabolites are causing growing concern. The objectives of this thesis were (1) to use a tiered analytical approach to investigate polar transformation products from various sources and (2) to identify common marker compounds that can be used for a more focused characterization of weathering processes at petroleum-contaminated sites. Previous studies have shown that the majority of weathered petroleum hydrocarbon compounds could not be detected by the GC-based analyses currently required by the United States Environmental Protection Agency due to their low volatility and high molecular weight. Therefore, standard methods may yield misleading characterizations of plumes and impede effective risk management. Fourier-transform ion cyclotron resonance mass spectrometry (FT-ICR MS), an emerging analytical technique in the field of "petroleomics" (the characterization of petroleum at the molecular level) offers unrivaled resolving power and mass accuracy; here it was used to determine the elemental composition of highly complex petroleum mixtures present in hydrocarbon-impacted sediment samples collected from field sites with varying redox and hydrogeological conditions. The tiered analysis revealed that GC-based techniques could only detect select nonpolar, low-molecular weight species (
Open Access
Identification and validation of screening methods for assessment of the sheening potential of embedded oil in sediments
(Colorado State University. Libraries, 2020) Sitler, Katherine, author; Scalia, Joe, advisor; Sale, Tom, advisor; Sutton, Sally, committee member
Sediments impacted with petroleum hydrocarbons (oil) may sheen due to ebullition-driven transport or sediment disturbance. The goal of this project was to develop a screening method that can be deployed on a small autonomous watercraft that will provide a reliable indication of sheening potential of embedded oil in shallow sediments. Different potential probes and methods were explored to penetrate sediments and determine sheening potential. Preliminary probe identification focused on development of a standardized laboratory column to test different probes and penetration methods to determine which probe has the highest probability to generate a sheen. Column tests were performed that consisted of different combinations of five crude oil types and a control (no oil embedded), seven probe candidates, two types of oil deposits, two targeted sheen levels, and with or without embedded air. Based on the data collected, a direct push rod with water injection had the greatest potential to generate a sheen.
Open Access
LNAPL longevity as a function of remedial actions: tools for evaluating LNAPL remedies
(Colorado State University. Libraries, 2013) Skinner, Anna Meryle, author; Sale, Tom, advisor; Ronayne, Michael, committee member; Shackelford, Charles, committee member
To view the abstract, please see the full text of the document.
Open Access
Management of contaminants in low permeability media
(Colorado State University. Libraries, 2014) Saller, Kevin R., author; Sale, Tom, advisor
To view the abstract, please see the full text of the document.
Open Access
Method comparison for analysis of LNAPL natural source zone depletion using CO₂ fluxes
(Colorado State University. Libraries, 2015) Tracy, Melissa Kay, author; Sale, Tom, advisor; Blotevogel, Jens, committee member; Butters, Greg, committee member
Accidental releases of subsurface petroleum hydrocarbons, widely referred to as Light Non-Aqueous Phase Liquids (LNAPLs), are a common occurrence in the industrial world. Given potential risks to human health and the environment, effective remediation approaches are needed to address impacts. Natural source zone depletion (NSZD) is a remedial approach gaining wide acceptance, wherein natural mechanisms in the subsurface act to deplete LNAPL in the source zone. Recent research indicates biodegradation of contaminant-related carbon results in a predominantly upward flux of carbon through the vadose zone. Building on this concept, three methods have recently emerged to quantify rates of NSZD using soil gas fluxes; these include the gradient, chamber, and trap methods. Unfortunately, side-by-side field applications of the methods have shown differing estimates of NSZD, leaving concerns about method comparability. The primary objective of this thesis was to conduct a laboratory comparison of the gradient, chamber, and trap methods using uniform porous media, constant environmental conditions, and a known CO₂ flux (i.e., ideal conditions). Given these experimental conditions, challenges associated with field comparisons could be minimized and the fundamental accuracy of the methods could be resolved. Preliminary efforts were also made to understand the effect of surface wind on the accuracy of the methods. A large-scale column (1.52 m high x 0.67 m ID) was filled with dry, homogenous, well-sorted fine sand. Known CO₂ fluxes were imposed through the bottom of the column spanning a range typical of contaminant-related CO₂ fluxes observed at field sites (3.3-15.2 μmol/m²/s). Results under ideal experimental conditions indicated that on average, the chamber and trap methods accurately captured the imposed flux to within ± 7% of the true value, and the gradient method underestimated the imposed flux to within 38% of the true value. Accuracy of the gradient method was largely dependent on estimates of effective diffusion coefficients. Consistent underestimation of the true flux using the gradient method was attributed to the method only quantifying diffusive gas transport. Considering the accuracy of measurements for other subsurface processes (e.g., hydraulic conductivity), the range of accuracy observed among all methods is not surprising. Surface winds were simulated by placing a fan on top of the column; achieved wind speeds ranged from 2.2-5.4 m/s. Laboratory studies identified that all methods were adversely affected by wind; however, the magnitude of laboratory results may have been exaggerated relative to what would be expected at field sites due to the laboratory sand being dry. Wind speeds within the tested range caused the gradient method to further underestimate the true flux to within 44% of the true value. The chamber method underestimated the true flux by 45-47% and 78% for wind speeds ranging from 2.2-3.6 m/s and 4.5-5.4 m/s, respectively. Wind had the opposite effect on the trap method, causing overestimations of the true flux by 60% and 122% for wind speeds ranging from 2.2-3.6 m/s and 4.5-5.4 m/s, respectively. Given similar results under ideal experimental conditions, wind and other environmental factors common to field conditions are suspected to be the primary cause of disagreement observed in side-by-side comparisons of the methods at field sites. Each method has advantages and limitations for field application. Method selection should be predominately driven by site-specific attributes, including environmental factors that may make one method more applicable over another for a given field site. Further consideration of all methods under environmental conditions may provide greater insight into potential biases and support additional recommendations for method selection. Secondary objectives included efforts to test design features specific to the trap method to support continued method development and to advance a model to describe steady-state advective and diffusive transport of a compressible gas through porous media. Results from trap modification studies suggested certain design features of the trap method may have affected the accuracy of measurements. Additional research and method development for the trap method could be undertaken to resolve issues raised in this thesis. Results from modeling efforts suggested gas transport was primarily diffusion driven, accounting for approximately 58-79% of transport, depending on estimates of the effective diffusion coefficient. Analytical modeling did not indicate an appreciable difference in advective and diffusive contributions to gas transport as the imposed flux was varied; however, measured concentration gradients counterintuitively indicated the advective contribution to transport increased as the imposed flux decreased.
Open Access
Method development for long-term laboratory studies evaluating contaminant assimilation processes
(Colorado State University. Libraries, 2016) McSpadden, Rachael Lynne, author; Sale, Tom, advisor; Blotevogel, Jens, committee member; Butters, Greg, committee member
Remediation technologies for soil and groundwater that are impacted by chlorinated solvents are limited when reducing contaminant concentrations below maximum contaminant levels (MCLs) established by the US Environmental Protection Agency (EPA). The limited effectiveness of current technologies is partly due to well-documented contaminant back diffusion from low-permeability (k) zones causing long-term impacts on water quality. Back diffusion out of low-k zones for extended periods of time, give strong evidence that assimilation processes are driving the fate and transport of chlorinated solvents within low-k zones. The direct impacts assimilation processes, such as sorption and degradation, have on contaminant concentrations may be slow and negligible on shorter time scales. But for longer time scales assimilation processes could have consequential effects on sites where groundwater concentrations are predicted to exceed MCLs for decades to centuries. Research studies located in the field have been carried out to study assimilation processes in low-k zones. The challenge of such field studies is capturing complete data sets from complex field environments. The challenges include inability to close the mass balance, confidently identifying assimilation mechanisms at work, and are limited to short term studies. Thus, the overall objective of this research is to advance the current knowledge of assimilative processes within low-k zones through the application of long-term (~5-10years) laboratory studies. The goal of the research presented herein, is to create a starting point for long-term laboratory studies in the hopes to quantify assimilation processes within low-k zones. Prior to conducting long-term laboratory experiments, a necessary step of establishing and testing methods need to be conducted. The research described within this thesis applies the use of short-term laboratory studies conducted over a 2 to 3 month time span to test preliminary methods, establish baseline data, and test applicability of mathematical models. The model contaminant used for the short-term laboratory experiments was tetrachloroethene (PCE). For the beginning stages of method development, the assimilation process that was isolated and focused on was sorption. Sorption was evaluated in porous media of differing properties, which included four field soils (Soil A, B, C, and D) and one lab grade soil (LGS). Two short-term column studies were tested to evaluate for viability in collecting data to be used in capturing transport and assimilation processes for use in long-term laboratory studies. The two short-term column study methods are identified throughout this document as headspace vials and ampules. The design setup for both column studies were constructed to utilize diffusive transport of contaminant with a saturated lower boundary layer of PCE, an initially clean water saturated soil column, and headspace at the upper boundary layer. For each column study design, the contaminant is transported via passive diffusion, starting from a volume of high concentration (at the lower boundary layer) to a place of low concentration (throughout the clean soil and the top of the headspace to the clean upper boundary layer). The difference between the two short-term column studies is the method of data collection. The headspace vial method allows for non-destructive sampling of the headspace over time to quantify the diffusive transport of PCE through the soil column. The ampule method utilizes a completely closed system with a destructive sampling technique where the entire ampule is extracted within methanol to help eliminate the potential for mass lost from the system due to volatilization. In addition to the two short-term column studies, batch sorption studies were conducted to gain independent measurements of sorption parameters for the four field soils used throughout the column experiments. Lastly, a numerical solution to the diffusive transport partial differential equation was developed using Mathcad™. Three sorption models are employed: linear, Freundlich and Langmuir models. The parameter values from the batch sorption study were used as inputs for the mathematical model and results were compared to the short-term column study headspace vial experiment. Results from the short-term column studies show that losses from headspace vials may limit the values of the method over time periods greater than one week, but ampules are more stable than headspace vials and show the most potential for application in long-term laboratory studies. Batch sorption studies can complement the diffusive-transport studies by allowing for resolution of sorption parameter values that are independent of transport rates. The validity of the model appears to be challenged by unaccounted losses from the headspace vials, and was therefore unable to estimate experimental data results. The results of the ampules and batch sorption studies are suggested to be used to aid in the design of the long-term studies. The laboratory experiments and modeling described herein will, in hopes, be a step closer to advance the knowledge of assimilative processes and assist in determining the assimilative capacity of low-k zones. Ultimately, this work will hopefully contribute to improved decision-making at contaminated sites, possibly allowing money spent on ineffective remedies to be directed toward more productive solutions.
Open Access
Oleophilic bio barriers (OBBs) for control of hydrocarbon sheens at groundwater-surface water interfaces
(Colorado State University. Libraries, 2015) Chalfant, Marc William, author; Sale, Tom, advisor; Butters, Greg, committee member; Gooseff, Michael, committee member
Sheens are a common problem at petroleum facilities located adjacent to surface water bodies. Thin, iridescent films of Non-Aqueous Phase Liquid (NAPL) can form on surface water sporadically and unpredictably via three processes: seeps, ebullition, and/or shoreline erosion. Because the appearance of sheens can elicit a notice of violation of the Clean Water Act, a suite of remedies has been used to address them. Common remedies are often predicated on physical barriers and sorbent barriers, both of which can be expensive and/or prone to failure due to bypass and/or finite storage capacities. Groundwater-Surface water Interfaces (GSIs) are active biological zones where NAPL fluxes are attenuated via aerobic biological degradation. Physical and sorptive barriers can inhibit aerobic degradation processes by causing NAPL to accumulate, preventing oxygen delivery or introducing organic matter that exerts an oxygen demand. Shortcomings of current sheen remedies motivate the research presented herein, exploring the concept of aerobic reactive barriers at GSIs. Specifically, the concept of an Oleophilic Bio Barrier (OBB) is advanced. An OBB prevents sheens due to seeps, ebullition, and erosion by employing 1) an oleophilic geocomposite to sorb NAPL, 2) aerobic degradation of NAPL via naturally occurring microbes, and 3) structural cover to mitigate erosion. A full US patent detailing these concepts was submitted to the US patent office in September 2014 (Zimbron et al., 2014). The work presented herein includes laboratory studies, a preliminary field study, a full-scale field demonstration and a general estimate of construction costs. Results of the lab studies provided proof-of-concept that a geocomposite material in an OBB could prevent sheens. The geocomposite was shown to have a capacity of 3L of NAPL/m². The geocomposite was also shown to reduce dissolved hydrocarbon concentrations by up to 77%. The preliminary field study showed that an OBB could be used to prevent sheens in a field setting. Four 1m x 1m OBBs were installed in March 2013 and monitored through August 2013. In August, NAPL saturations of up to 1.6 L/m² were measured in the OBBs, demonstrating their ability to prevent sheens. The geocomposite maintained structural integrity, suggesting chemical compatibility with the NAPL. A low redox potential (62 mV) and the presence of dissolved iron (9.0 mg/L) at 90 cm depth showed that subsurface sediments were anaerobic. Redox potentials ranging from 302 to 423 mV were measured in the OBB water, demonstrating that aerobic degradation could occur and deplete NAPL on the OBBs. Results from the full-scale (36 ft x 18 ft) OBB module study demonstrated sheen prevention and microbial activity. Of 26 visual inspections for sheens, no sheens were observed sourcing from the OBB, while 3 inspections yielded sheen observations on adjacent shoreline. Seasonal changes in sorbed NAPL composition were consistent with patterns of microbial degradation and correlated to decreased redox potentials and warm temperatures. Microbial populations in the OBB were comparable to adjacent and underlying sediments but showed increased diversity of hydrocarbon-degrading microbes. In addition, structural cover was shown to mitigate erosion associated with ice-scour, while sustaining minimal damage and sedimentation. Costs for OBB construction were estimated to be on the order of $100,000 per acre, making more affordable than organoclay barriers and sheet pile barriers. The primary conclusion of this thesis is that OBBs are a viable technology from both cost and performance perspectives. Recommendations for future work include OBB design modifications for improved sediment control, greater compatibility with natural environments, and enhanced NAPL retention capacity. Simplified performance monitoring, research on governing processes, methods for characterizing sheen sources, and the development of a model to support OBB design optimization are also recommended. Ongoing consideration of expanding the full-scale OBB module and active consideration of OBB remedies at other sites provide promising opportunities for further development.
Open Access
Optimizing operation and design of aquifer storage and recovery (ASR) wellfields
(Colorado State University. Libraries, 2019) Alqahtani, Abdulaziz, author; Sale, Tom, advisor; Grigg, Neil, committee member; Bailey, Ryan, committee member; Ronayne, Michael, committee member
Sustained production of groundwater from wells in wellfields can lead to declining water levels at production wells and concerns regarding the sustainability of groundwater resources. Furthermore, minimizing energy consumption associated with pumping groundwater is a growing concern. Aquifer Storage and Recovery (ASR) is a promising approach for maintaining water levels in wells, increasing the sustainability of groundwater resources, and minimize energy consumption during groundwater pumping. Therefore, studying the importance of ASR in sustaining water levels and minimizing energy consumption is critical. In the first part of this dissertation, an analytical model relying on superposition of the Theis equation is used to resolve water levels in 40 wells in three vertically stacked ASR wellfields. Fifteen years of dynamic recovery/recharge data are used to obtain aquifer and well properties. Estimated aquifer and well properties are used to predict water levels at production well. Close agreement between modeled and observed water levels support the validity of the analytical model for estimating water levels at ASR wells. During the study period, 45 million m³ of groundwater is produced and 11 million m3 is recharged leading to a net withdrawal of 34 million m³ of groundwater. Rates of changes in recoverable water levels in wells in the Denver, Arapahoe and Laramie-Fox Hill Aquifers are 0.20, -0.91, and -3.48 m per year, respectively. To quantify the benefits of recharge, the analytical model is applied to predicting water levels at wells absent the historical recharge. Results indicate that during recovery and no-flow periods, recharge has increased water levels at wells up to 60 m compared to the no-recharge scenario. On average, the recharge increased water levels at wells during the study period by 3, 4, and 11 m in the Denver, Arapahoe, and Laramie Fox-Hills Aquifers, respectively. Overall, the analytical model is a promising tool for advancing ASR wellfields and ASR can be a viable approach to sustaining water levels in wells in wellfields. In the second part of this dissertation, a simulation-optimization model (ASRSOM) is developed to optimize ASR wellfield operations. ASRSOM combines an analytical hydraulic model and a numerical optimization model to optimize wellfield operations. The objective function used to minimize energy consumption φ (L⁴) is the temporal integral of the products of temporally varying total dynamic head values and pumping rates. Comparison of ASRSOM results to work by others for idealized aquifer operations supports the validity of ASRSOM. Four scenarios were simulated to evaluate the role that optimization of operations and aquifer recharge play in reducing the energy required to lift groundwater out of aquifer. A 10-year study period is considered using data from a municipal ASR wellfield. Optimization decreased φ by 19.6%, which yields an estimated reduction of 2,179 MW hours of power and 1,541 metric tons of atmospheric carbon. For the condition considered, recharge reduced power by 1%. The limited benefit of recharge is attributed to the small recharge volume in the case study and the short duration of the analysis. Additional opportunities to address economic and environmental impacts associated with lifting groundwater out aquifer include optimizing the position of wells and factors controlling total pumping head. In the third part of this dissertation, the sensitivity of well-spacing in ASR wellfields to critical parameters is studied. The parameters studied are aquifer transmissivity and storativity, wells flowrate and the frequency of recharge and recovery. It has been found that larger well-spacing are appropriate for lower transmissivity and storativity, and larger wells flowrate and frequency. More work is needed to fully understand the optimal well-spacing of wells in ASR wellfields associated with more realistic storage and recovery schedules, and more complex wellfields. Overall, work supported the possibility that wells in ASR wellfields can be spread more closely than wells in conventional production wellfields.
Open Access
Processes controlling the behavior of LNAPLs at groundwater surface water interfaces
(Colorado State University. Libraries, 2013) Hawkins, Alison M., author; Sale, Tom, advisor; Zimbron, Julio, committee member; Ronayne, Michael, committee member
Releases of Light Non-Aqueous Phase Liquids (LNAPLs) are a significant problem at many sites. This thesis explored governing processes pertaining to LNAPL releases at groundwater surface water interfaces (GSIs). Governing processes were investigated via laboratory studies and a preliminary analysis of forces controlling LNAPL occurrence in unsaturated media. A total of six laboratory sand tank experiments were conducted using novel applications of fluorescing dyes. The results of these experiments provide unique insights regarding LNAPL behavior in porous media. Key insights include: * LNAPLs occur in three distinct zones, herein referred to as Zone 1, 2, and 3. Zone 1 refers to the area below the water capillary fringe where LNAPL is a discontinuous nonwetting phase. Zone 2 refers to the area below the LNAPL capillary fringe where LNAPL is a continuous nonwetting phase. Zone 3 refers to the area above the LNAPL capillary fringe where LNAPL is a continuous intermediate wetting phase. Each zone has unique attributes controlling LNAPL mobility * Solutions for LNAPL releases at GSIs need to address transport of LNAPL in all three zones * Modeling fluid saturations versus height in a porous media using a force balance is more complex than two forces and requires further research A common theme with current solutions for LNAPLs at GSIs is their failure with time. Failure is defined as the observation of LNAPL down-gradient of the solution. A better understanding of these failures is advanced through a volume balance on a representative elementary volume (REV) of porous media at a GSI. Key factors controlling releases to surface water include inflows, natural losses, enhanced losses, and recovery of LNAPL in the REV. Furthermore, the timing of failure is dependent on the capacity of the REV to store LNAPL prior to releases to surface water. A novel solution demonstrated in this thesis was the use of capillary barriers to limit LNAPL lateral migration. Herein, capillary barriers are defined as vertical walls of fine-grained media that preclude lateral movement of LNAPL via a capillary pressure less than the displacement pressure in Zone 2 and an elevated water capillary fringe in Zone 3. A capillary barrier alone can delay releases; however, the barrier will fail when LNAPL storage capacities are exceeded. In contrast, the use of a recovery well to deplete accumulating LNAPL, in combination with a capillary barrier, provides a sustainable solution. During a laboratory experiment, 92% of the delivered LNAPL held behind the capillary barrier was recovered by aggressively pumping at low water stages. A second strategy explored to control LNAPL releases at GSIs was organoclay barriers. Herein, organoclay barriers are defined as vertical walls of organoclay-sand mixtures. Organoclay is hydrophobic and retains LNAPL via sorption. Using a "simple" organoclay barrier, breakthrough to surface water was observed when only 11% of the organoclay was saturated with LNAPL. Early failure was attributed to preferential pathways and slow water drainage. Adding vertical baffles and vertical coarse-grained drains improved the efficacy of organoclay barriers. Fractions of the clay contacted at breakthrough were 43% and 34%, respectively, for baffles and drains. A concern that arose from the sand tank studies was the necessary water capillary rise in the capillary barrier to preclude LNAPL migration in Zone 3. This led to an attempt to develop a force-based model describing LNAPL (intermediate wetting phase) saturations in Zone 3. The model would be beneficial to determine the vertical rise of LNAPL at sites with non-tidal conditions. Key factors included in the model include spreading coefficients and gravity. The model developed (Model 1) was compared to three-phase data. It was found that Model 1 had poor correlation to the data and lacked some key factor affecting saturations. The model was altered by raising Model 1 to the power of lambda and adding the residual saturation, resulting in Model 2. Model 2 was compared to two-phase data and the Brooks-Corey equation and showed promising similarities. The work described in this thesis provides a basis for future work on remediation solutions and mathematical models for LNAPLs at GSIs. Work could include development of strategies to enhance natural losses of LNAPLs at GSIs and further refinements to Model 1 and Model 2 to better capture factors controlling fluid saturations in Zone 3.
Open Access
Processes governing the performance of oleophilic bio-barriers (OBBs) – laboratory and field studies
(Colorado State University. Libraries, 2018) Tochko, Laura, author; Sale, Tom, advisor; Scalia, Joe, committee member; Sutton, Sally, committee member
Petroleum sheens, a potential Clean Water Act violation, can occur at petroleum refining, distribution, and storage facilities located near surface water. In general, sheen remedies can be prone to failure due to the complex processes controlling the flow of light non-aqueous phase liquid (LNAPL) at groundwater/surface water interfaces (GSIs). Even a small gap in a barrier designed to resist the movement of LNAPL can result in a sheen of large areal extent. The cost of sheen remedies, exacerbated by failure, has led to research into processes governing sheens and development of the oleophilic bio-barrier (OBB). OBBs involve 1) an oleophilic (oil-loving) plastic geocomposite which intercepts and retains LNAPL and 2) cyclic delivery of oxygen and nutrients via tidally driven water level fluctuations. The OBB retains LNAPL that escapes the natural attenuation system through oleophilic retention and enhances the natural biodegradation capacity such that LNAPL is retained or degraded instead of discharging to form a sheen. Sand tank experiments were conducted to visualize the movement of LNAPL as a wetting and non-wetting fluid in a water-saturated tank. The goal was to demonstrate 1) the flow of LNAPL as a non-wetting fluid in sand, 2) the imbibition of LNAPL as a wetting fluid on the geocomposite, and 3) the breakthrough of LNAPL after saturating the geocomposite to the point of failure (sheens in the surface water). Dyed diesel was pumped through a tank with sand and geocomposite and photographed to document movement. Diesel was the non-wetting fluid in the sand and moved in a dendritic pattern. Diesel was the wetting fluid on the geocomposite and uniformly imbibed horizontally across the geocomposite before breakthrough to the overlying sand layer. A second set of laboratory experiments was designed to estimate the aerobic and anaerobic OBB degradation rates of LNAPL in field-inoculated sediment. Unfortunately, due to a flaw in the experimental design, the mass balance could not be completed, and degradation rates were not calculated. The setup was designed to emulate field conditions as best practically possible and to observe the effects of water table fluctuations, different loading rates, and iron. The effluent pumping system designed to remove water in the water fluctuation columns also inadvertently removed LNAPL, creating a mass balance discrepancy for the aerobic columns. Though degradation rates could not be calculated from this experiment, the experiment did visually document the changing redox conditions in the columns, such as formation of a black precipitant (likely iron sulfides) around LNAPL. Ideally, the limitations of this experimental design can be addressed for future research to eventually resolve degradation rates for OBBs. The success of a 3.8 m by 9.3 m demonstration OBB at a field site on a tidal freshwater river resulted in replacing the demonstration OBB with a 3.8 m by 58 m full-scale OBB. The construction event provided a unique opportunity to sample the demonstration OBB after a four-year deployment. The sampling results advanced the mechanistic understanding of how OBBs work to reduce LNAPL releases at GSIs. Sampling revealed the material was suitable for field LNAPL loading rates and was not compromised by field conditions such as ice scour or sediment intrusion. LNAPL analysis showed no LNAPL on the geocomposite or in the underlying upper sediment (0-10 cm). Diesel range organic (DRO) concentrations in the low 1,000s of mg/kg were observed in the sediment 10-20 cm below the geocomposite. LNAPL composition analysis suggests that the majority of the compounds are polar in the lower sediments (10-20 cm), providing a line of evidence that petroleum liquids have been oxygenated. Microbial data show the average number of bacterial 16s transcripts in the geocomposite is larger than in the sediment layers, confirming that the geocomposite is suitable substrate for microbe growth. The observation of ferric iron suggests that ferric/ferrous iron cycling may play a role in degradation processes, where the ferric iron acts as a "bank" of solid-phase electron acceptors. This sampling event suggests that LNAPL biodegradation rates in and below the OBB are comparable to the LNAPL loading rates.
Open Access
Remediation of soil impacted with chlorinated organic compounds: soil mixing with zero valent iron and clay
(Colorado State University. Libraries, 2014) Olson, Mitchell R., author; Sale, Tom, advisor; Dandy, David, committee member; De Long, Susan, committee member; Shackelford, Charles, committee member
Chlorinated solvents in the environment continue to present an enormous remediation challenge. A primary reason for the difficulty in cleaning up chlorinated solvent source zones involves heterogeneous distributions of permeability and contaminants in natural porous media. A method that can be used to overcome heterogeneity involves use of soil mixing techniques to deliver reagents and homogenize soils. A typical soil mixing application involves admixing contaminated soil with zero valent iron (ZVI) and bentonite (clay). This technology, herein referred to as ZVI-Clay, combines ZVI-mediated degradation of chlorinated solvents with bentonite-induced stabilization. As of December 2013, ZVI-Clay has been applied in 13 field applications, all of which have been viewed as being successful in achieving site remediation objectives. However, our understanding of the processes governing treatment in the ZVI-Clay mixed soil system is rather limited. The overarching goal of the research presented herein is to broaden our understanding of the processes controlling degradation and transport in soils treated via soil mixing with ZVI (or similar reactive media) and bentonite. In support of this objective, research included (a) analysis of field data, (b) initial rate studies, (c) hydraulic conductivity testing, (d) reactive-transport modeling coupled with column experiments, and (e) treatment of hydrophobic compounds. Field data analysis was based on performance data from a ZVI-Clay field application at Camp Lejeune, NC, in which 23,000 m3 of soil initially contaminated with trichloroethene (TCE) and 1,1,2,2-tetrachloroethane (TeCA) were treated with 2% ZVI and 3% bentonite. Within one year of treatment, total chlorinated organic compound (COC) concentrations in soils were decreased by average and median values of 97% and >99%, respectively. Total COC concentrations in groundwater were reduced by average and median values of 81% and >99%, respectively. Total COC reductions by 99.9% or greater were observed in most soil and water sampling locations. Hydraulic conductivity in the treated soil zone was reduced by an average of about 2.5 orders of magnitude. To explain the variations in kinetic data observed following ZVI-Clay field applications, initial-rate batch-reactor studies were conducted under a range of initial TCE concentrations and ZVI concentrations. When TCE concentrations were less than solubility, the Michaelis-Menton kinetic model provided an excellent fit of experimental data. When TCE concentrations were above solubility (i.e., NAPL was present), the degradation rate was independent of the amount of NAPL in the system. The presence of NAPL appears to have had a minor impact (~20% reduction) on the TCE degradation rate. A linear relationship between TCE degradation rate and ZVI amount was observed. Hydraulic characteristics of soils mixed with bentonite were evaluated by conducting column studies and MODFLOW modeling of field-scale systems. Experiments were conducted to evaluate the hydraulic conductivity, K, in two soils types mixed with 0.5 to 4% bentonite (i.e., the range of values typical for ZVI-Clay field applications). In a well-sorted fine sand, with a moderately high initial K (10-4 m/s), the value of K was reduced by about a factor of 10 for each 1% bentonite added to the soils. In a moderately-sorted fine sand with silt, with a low initial K (10-8 m/s), addition of up to 4% bentonite had only minor impacts on K. MODFLOW modeling indicated that surrounding groundwater flow patterns tend to bypass the treated soil body, under steady state conditions, given a reduction in K by at least an order of magnitude. Within the treated soil body, contaminant residence time is extended in approximate proportion to the reduction in K. The concepts of NAPL dissolution, ZVI-mediated degradation, and flow reduction were combined in a mathematical model. The model was then tested using column reactor studies containing NAPL-phase TCE and soils treated with 2% ZVI. The model adequately described TCE elution and formation of degradation products. The model was then used to predict treatment performance following field-scale implementation of ZVI-Clay. Model output predicts that the benefits of reaction are most effectively utilized with a reduction in flow rate by at least 2 orders of magnitude. Finally, enhancements to the ZVI-Clay treatment process were evaluated for treatment of polychlorinated biphenyls (PCBs). Due to their strong hydrophobicity and stable molecular structure, PCBs in the environment have been shown to be much more difficult to degrade than many of the common chlorinated solvents. Thus, alternative types of reactive media were evaluated. Batch experiments were conducted to evaluate zero valent metals (ZVM), ZVM + Pd-catalysis, and emulsified zero valent iron (EZVI) for dechlorination of PCBs in systems with and without soil. In water-based systems, ZVM with a Pd-catalyst facilitated rapid destruction of 2-chlorobiphenyl (half-life < 2 hr), while ZVM alone did not achieve any measurable degradation. In the presence of soils, EZVI was the only approach that resulted in a clear enhancement in PCB dechlorination rates. The results suggest treatment of PCBs in the presence of soil presents a much greater challenge than treatment of aqueous phase PCBs; however, treatment of PCBs in soil can benefit from enhanced desorption and a persistent reactive media.
Open Access
Subsurface water storage assessment model
(Colorado State University. Libraries, 2015) Alqahtani, Abdulaziz A., author; Sale, Tom, advisor; Bailey, Ryan, committee member; Ronayne, Michael, committee member
Water storage is an essential part of water resources management schemes. Due to the high cost and escalating risks of building new surface reservoirs, water managers are becoming interested in employing more effective alternatives. Subsurface water storage is getting attention as one of these alternatives. However, due to lack of experience and tools to estimate the cost and effectiveness of subsurface water storage, water managers are reluctant to adopt this alternative. Available tools/models are only case specific; hence in this study, we develop a general model for subsurface storage and recovery. The model estimates the cost of the subsurface water storage and recovery using wells in bedrock. The model takes monthly river flow, population, and per capita demand as inputs to determine capital cost and operation and maintenance costs for the lifespan of the proposed project. To account for uncertainty in the input parameters, the model has the capability to perform stochastic analyses as well. Furthermore, the model includes the option of modular expansion of infrastructure through time, potentially reducing total and operation and maintenance costs. An application of the model is advanced based on the conditions in the vicinity of Fort Collins, Colorado. Critically, work presented herein should not be taken as a rigorous analysis of the issues faced by the city of Fort Collins. The application is simply a demonstration of what can be done with the tools developed in this thesis. The general premise of the application is creating new water storage in the Fountain Formation, north of Fort Collins. This model uses either deterministic or stochastic inputs. Since the deterministic model's inputs and outputs are both fixed numbers, the model is relatively simple. However, this type of input will yield specific results and does not consider the possibility of inputs varying through time. It misses a key challenge of water projects, the temporal variability in available water and demand. In Stochastic Analysis, inputs are varies from year to year and from month to month, allowing the system to accommodate wet or drought years, making the model more reliable for calculating the cost of system. One hundred simulations were performed using the stochastic model to estimate the range of variability of outputs. Except total pumping and additional storage, other outputs have small coefficients of variation, which show that they are less sensitive to uncertainity in input variables. The coefficient of variation for cost variables are around 0.1 (i.e., costs are expected to vary within ±10% of the estimated mean cost). As different cost components estimated by deterministic model are within ±10% of estimated mean cost from stochastic model. Therefore, we conclude that the deterministic model estimates different cost components fairly well. Both models, deterministic and stochastic, have been applied to a scenario predicted on conditions faced by the city of Fort Collins. At thirtieth year, the system can deliver 7.8×10⁶ m³/year of water (6.4×10³acre-ft/year) in an average year and up to 15.7×10⁶m³/year of water (12.7×10³acre-ft /year) in a drought year. The estimated present value cost from deterministic and stochastic models of the entire project was $ 23.1 million U.S and $ 22.5 million U.S., respectively. Not considered in the cost analysis is the value of the water saved due to reduced losses of evaporation and seepage losses, inherent with surface water storage. The model shows high reliability in meeting demand through the lifespan of the project, with no failure anticipated. The deterministic model added 9.12 million m³ to the aquifer, while the stochastic model shows an average addition of 16.8 million m³ to the aquifer. Greater stored water with the stochastic model is attributed to less pumping of groundwater. Further study is needed to resolve the basis for the stochastic model pumping less groundwater. The capital cost of the project is predicted to be approximately $ 6.0 million U.S. by both models. Both models estimated the need for 10 ASR wells and two alluvium inflow drain units through the lifespan of the project. The case study of Fort Collins shows the potential of subsurface water storage as a viable and cost effective alternative to surface water storage.
Open Access
Thermal aspects of STELA (sustainable thermally enhanced LNAPL attenuation)
(Colorado State University. Libraries, 2013) Akhbari, Daria, author; Sale, Tom, advisor; Bau, Domenico, committee member; Ronayne, Michael, committee member
Extensive bodies of light non-aqueous phase liquids (LNAPLs) are commonly found beneath petroleum facilities. Related concerns include lateral spreading of LNAPL, impacts to groundwater, and impacts to indoor air. Recent studies have shown that natural losses of LNAPL can be on the order of thousands of gallons per acre per year and temperature is a primary factor controlling rates of natural losses. Results of the laboratory and field experiments suggest that LNAPL impacted media in the range of 18-300C can have loss rates that are an order of magnitude greater than media at temperatures less than 18ºC. The vision that has emerged from recent work is that passive thermal management strategies could enhance natural losses of LNAPL and significantly reduce the longevity of LNAPL. Owing to this new understanding, plans were developed for a small-scale field demonstration of sustainable thermally enhanced LNAPL attenuation (STELA) at a former refinery in Wyoming, located adjacent to the North Platte River. The overarching objective of the STELA initiative is to develop a new technology for LNAPLs that is more effective, faster, more sustainable, and/or lower cost than current options. The primary objective of the field demonstration is to collect data needed to evaluate cost and performance at field sites. In November 2011, seventeen multilevel sampling systems were installed in a 10m by 10m area. Preheating temperature and water quality data were collected through the multilevel samplers over a period of 10 months. In August 2012, ten heating elements, including submersible heat trace wires wrapped around 7.6 cm ID PVC pipe with thermostat controls, were installed upgradient of the sampling network to deliver heat to sustain subsurface temperature in an LNAPL body. The heating elements were energized in September 2012. Subsequently, effects of the heating elements on the subsurface temperature were monitored using 17 multilevel sampling systems equipped with 6 thermocouples for 10 months. Preheating data indicates that in the absence of heating, subsurface temperatures are in the range of 18-30°C for 40 days per year. Data collected from September 2012 to July 2013 indicates that with heating, conditions can be maintained in the target range for 60 to 200 days per year depending upon proximity to the heat source. A principle challenge is heat loss to the surface in the winter. Minimum and maximum power inputs have been 15 kw-hr/day and 30 kw-hr/day occurring, respectively in October and May. Assuming an energy cost of 0.10 kw-hr, this equates to costs of 1.5 $/day to 3 $/day. An independent experiment using Geo-net layer showed that using Gas Permeable Insulation/Heat Sink (GPIHS) system has the potential to enhance the ability of the heating system to sustain temperature beneath the ground surface, and, potentially decrease the power costs. A primary challenge with evaluation and design of STELA systems is anticipating the appropriate spacing of heating elements and necessary energy inputs. Herein this challenge is met by developing a model, calibrated to field data, which can be used to design a full-scale STELA remedies. The overarching objective of the modeling is to demonstrate methods that can be employ to evaluate and/or design full-scale STELA systems. At 5m downgradient of the heating elements, the developed model, accurately, predicted 60 days of the effective season in 2012. Also, the simulation results anticipate that by keeping the heating system activated for three years, the effective season will increase each year. At 5m downgradient of the heating elements, model results suggested 120 days and 150 days of effective season for 2013 and 2014, respectively as compared to 60 days in the first year. The ability of the model to anticipate the effective season for the next years makes the model a useful tool to design and evaluate the future STELA systems. Calibration of the model to the field data shows that exothermic reactions associated with LNAPL losses can change the heat distribution at the system. In addition, the simulation results indicate that the losses at the subsurface are in the range of 5,000 to 10,000gal/acre/yr. These anticipated loss rates are consistent with the previous values reported by McCoy (2012) in 2012 (~900-11,000gal/acre/yr.) A conceptual STELA design is developed in the last chapter to explore the cost of a STELA system at a 1-hectare site. The design is based on condition at the former refinery in Wyoming where the STELA field demonstration was conducted. The cost analysis study indicated that the primary cost is the heating elements installation. The second significant cost is the operation costs, and the third significant cost that can be reduced is the energy source. The cost estimates normalized to common units indicated that the total cost ranges between $590,000 to $720,000 per hectare, $11.9 to $14.4 per cubic meter of treated soil, and $1.3 to $1.5 liter of LNAPL removed depends on the energy source, heating system and the degradation rate. Cost of this magnitude support the hypothesis that STELA has the potential to have cost that is lower than other options employed for LNAPL remediation.