Browsing by Author "Carlson, Kenneth, committee member"
Now showing 1 - 20 of 20
Results Per Page
Sort Options
Item Open Access Analysis of octamethylcyclotetrasiloxane and decamethylcyclopentasiloxane in wastewater, sludge and river samples by headspace gas chromatography/mass spectrometry(Colorado State University. Libraries, 2014) Zhang, Yu, author; Ömür-Özbek, Pinar, advisor; Carlson, Kenneth, committee member; Dooley, Gregory, committee memberSiloxanes are commonly used in cosmetic and personal care products, healthcare products and many industrial applications. Because siloxanes are persistent, they end up in the wastewater and go untreated through the wastewater treatment units, which lead to contamination of the surface waters through effluent discharge. Siloxanes tend to be adsorbed onto or absorbed by the activated sludge in the wastewater treatment process. In the digesters, the siloxanes volatilize and accumulate in the biogas, which leads to mechanical problems due to scaling. The two most common siloxanes detected in wastewaters and sludge are: octamethylcyclotetrasiloxane (D4) and decamethylcyclopentasiloxane (D5). For this study, the D4 and D5 in wastewater and sludge samples were monitored using headspace gas chromatography/mass spectrometry. Samples were collected from the City of Loveland Wastewater Treatment Plant (WWTP), Loveland, CO and the Drake Wastewater Reclamation Facilities (WWRF), Fort Collins, CO. The levels of D4 were in the range of 0.7-11.3 ng•mL-1 in wastewater and 0.3-1.8 µg•g-1 dry solid in the sludge from Drake WWRF; 1.0-6.7 ng•mL-1 in wastewater and 0.3-1.7 µg•g-1 dry solid in the sludge from Loveland WWTP. D5 levels were determined in the range of 0.4-10.4 ng•mL-1 in wastewater and 3.2-31.4 µg•g-1 dry solid in the sludge from Drake WWRF; 0.5-14.0 ng•mL-1 in wastewater and 2.5-18.9 µg•g-1 dry solid in the sludge from Loveland WWTP. The concentrations of D4 and D5 were higher in this study compared to other researches in other countries and the concentrations in waste activated sludge were in a comparable range. The concentrations of D4 and D5 in the receiving water body near the discharging points were below the limit of detection. The average mass loadings in the influent were 53.1 and 159.9 g•d-1 of D4 and 155.3 and 225.3 g•d-1 of D5 respectively in two plants.Item Open Access Characterization and treatment of water for unconventional oil wells with temporal variability and fracturing fluid type(Colorado State University. Libraries, 2017) Kim, Seongyun, author; Omur-Ozbek, Pinar, advisor; Carlson, Kenneth, committee member; Catton, Kimberly, committee member; Dooley, Gregory, committee memberFlowback/produced water from unconventional oil and gas wells cannot be optimized without an understanding of water quality which needs to be treated for reuse, the factors to be considered include the temporal variability and different frac fluid types used for hydraulic fracturing. Produced water treatment for reuse is becoming a critical factor for water management surrounding unconventional oil and gas industry. For this research flowback/produced water samples were collected over 200 days from two wells in the Wattenberg Field, located in northeast Colorado. One of the frac fluids had an initial pH greater than 10 and had a guar-based gel. The other frac fluid included a non-guar polysaccharide based polymer and an initial pH less than 6. Total dissolved solids (TDS) and total organic carbon (TOC) analyses were conducted as an indicator for presence of organic and inorganic solids, and the data was compared with key ions (barium, calcium, chloride, magnesium, sodium, strontium, boron and iron) with the different frac fluid types. High values of the coefficient of determinant (over 0.85) were observed between TDS and the key ions, showing that significant positive correlations between two. Despite the significant initial pH differences, the concentrations of calcium, chloride, sodium and strontium were statistically equivalent between the two frac fluids. A mass balance approach was applied to evaluate the quantity of mass of injected additives that was recovered over the 200-day period. Zirconium, potassium and aluminum were selected due either to the lack of contribution from the formation (Zr, Al) or the conservative (non-reactive) nature of the ion (K). Recoveries of these ions ranged from 3 % to 33 % after 200 days, and notable differences were observed between frac fluids. The fraction of cross-linking chemical (Zr) recovered was significantly less for the residue-free polysaccharide-based fluid than the derivatized guar-based fluid. It is hypothesized that the polysaccharide-based Zr cross-linked gel is broken down more completely than the guar-based gel, releasing the Zr metal with subsequent precipitation in the formation. Another study conducted for this dissertation involved the analysis and treatment of produced water samples from three wells that were fractured with different fracturing fluid types over 63 days. TOC analysis showed significantly higher organics composition in produced water from wells fractured by the gel and hybrid fluid (943-1,730 mg/L) compared to the well fractured by the slickwater (222-440 mg/L). TDS levels increased with time, varying from roughly 18,000 mg/L to 30,000 mg/L between 1 to 63 days at each well. Liquid chromatography–mass spectrometry (LC-MS) was applied to characterize the organic matters and similar mass spectra were observed from each well with no temporal trend. Chemical equilibrium modeling was used to predict the precipitation of metals from produced water mixed with groundwater. Chemical coagulation was successfully performed for reducing the turbidity from produced water samples at each well. LC-MS was performed to study the compositions of frac fluid prepared with fresh water (FWA) only and frac fluid prepared with recycled water (RWA) mixed with fresh water. Ethylene oxide and propylated glycol functional units were observed from both FWA and RWA samples. Qualitative analysis from FWA and RWA was performed through Agilent qualitative analysis software B.06.00 based on the exact mass of the chemical compound. Van Krevelen diagram proved FWA and RWA show highly saturated and low degree of oxidation of organic compounds. Kendrick mass defect (KMD) from ethylene oxide was below 0.1 while KMD analysis from propylated glycol were close to 1. FWA showed 32.3 average carbon number and 9.8 double bond equivalent and RWA showed 31.5 average carbon number and 9.5 double bond equivalent. For the last phase of this research, produced water samples were treated by electrocoagulation (EC), ultrafiltration (UF), granular activated carbon (GAC) and reverse osmosis (RO) in series. Total dissolved solids (TDS), total organic carbon (TOC), dissolved organic carbon (DOC), BTEX, total petroleum hydrocarbons (TPH), turbidity, propylene glycol, ethylene glycol and ethylene glycol monobutyl ether were measured after each treatment. Gas chromatography–mass spectrometry (GC-MS) with solid phase extraction (SPE) method was applied to detect propylene glycol (PG), ethylene glycol (EG) and ethylene glycol monobutyl ether (EGME) in the samples. EGME was not detected in any produced water samples. PG concentration was between 0.07ug/ml to 5.39ug/ml and EG ranged from 0.07 ug/ml to 5.52ug/ml. GAC removed both PG and EG for acceptable drinking water criteria. EC was effective at removing both turbidity (85%) and TPH (80%) and most of turbidity and 90% of TPH were removed after UF. This study confirmed that almost 95% of BTEX, TOC, and DOC in produced water samples were removed by GAC. GAC contributed approximately 15% of TDS removal while RO removed 90% of TDS (2550mg/L) which is still high for reuse for various purposes.Item Open Access Characterization of mixed linear energy transfer environments utilizing tissue-equivalent proportional counters(Colorado State University. Libraries, 2023) Fehrman, Joseph M., author; Brandl, Alexander, advisor; Johnson, Thomas, advisor; Carlson, Kenneth, committee memberThere is currently great interest in the biological impact of radiological space exposures due to manned space missions (e.g., moon) where astronauts will face the challenge of living on board spacecraft for long periods of time. Cosmic radiation of many types exists in space and creates a unique mixed linear energy transfer (LET) environment. A tissue equivalent proportional counter (TEPC) was used to produce dose-equivalent measurements and quantify neutron exposures. Three neutron sources were used to simulate high energy, mixed LET environments: californium-252, a plutonium-beryllium source, and a deuterium-tritium neutron generator. TEPC dose-equivalent measurements can be utilized for basic research, and regulatory or clinical purposes for correlation with observable health effects. The main study was to quantify and compare TEPC dose equivalent rates in microdosimetric volumes to determine if significant dosimetric differences exist between mixed LET environments generated by photons and neutrons. The findings from this experiment showed that mixed LET environments where both photons and neutrons interacted with the TEPC had lower average LET values than neutron-only exposures, and produced varying dose equivalent rates that were dependent on the source characteristics. In summary, the TEPC was capable of monitoring in a mixed-LET environment and was successful at measuring the absorbed dose of high-energy photon and neutron interactions in space-like settings.Item Open Access Effect of graywater irrigation on soil quality and fate and transport of surfactants in soil(Colorado State University. Libraries, 2012) Negahban Azar, Masoud, author; Sharvelle, Sybil, advisor; Carlson, Kenneth, committee member; Stromberger, Mary, committee member; Durnford, Deanna, committee memberWhile interest in and adoption of graywater reuse for irrigation has rapidly grown in recent years, little is known about the long-term effects of graywater irrigation. Concerns exist in relation to the presence of pathogenic organisms, fate of personal care products, and accumulation of salts. The purpose of this research was to evaluate the long-term effects of graywater irrigation to soil quality. The specific objectives were to evaluate the effects of graywater application on physical and chemical quality of soil, including surfactants, salts and boron accumulation, organic matter leaching and soil hydrodynamic properties in real environment in the field, in controlled environment in the greenhouse and column studies. In addition, fate and transport of surfactants in soil were investigated including how surfactant characteristics impacts mobility in soil of varying types. Graywater irrigation was found to significantly increase sodium in soil at households with graywater systems in place for more than five years; however SAR was not high enough in any of the sampling locations to raise concern about soil quality or plant health. There is a potential for salts, N, and B to leach through soil when graywater is applied for irrigation. A portion of the applied N is assimilated by plants, but leaching of N was observed. Graywater irrigation was also found to significantly increase surfactants in soil. Surfactants mainly accumulated in surface soil (0-15 cm) compared to depth soil. While surfactants have high sorption capacity due to their hydrophobic characteristics, they can be transported through soil if a large amount of water is applied. Among the surfactants measured in this study, AS and AES had the highest mobility. Mobility of surfactants in soil decreased when their number of ethoxylated groups increased. Adding organic matter to the soil increased sorption capacity of soil, as a result, more surfactants retained in the soil columns. Antimicrobials, including triclosan and triclocarban were detected in graywater irrigated areas only in surface soil samples, but not freshwater irrigated areas.Item Open Access Elucidating the mechanisms and developing mitigation strategies of mineral scaling in membrane desalination(Colorado State University. Libraries, 2022) Yin, Yiming, author; Tong, Tiezheng, advisor; Sharvelle, Sybil, committee member; Carlson, Kenneth, committee member; Li, Yan, committee memberMineral scaling in membrane desalination, which is referred to as the accumulation of minerals on the membrane surface, has been considered as the primary constraint that limits the water recovery and efficiency of membrane desalination significantly. The occurrence of mineral scaling results in the decrease of water flux and compromises the lifetime of membrane materials, leading to increased needs of energy consumption and facility maintenance. Furthermore, the limited water recovery of membrane desalination due to mineral scaling also results in the production of high volumes of concentrated brines, which may require thermal-based technologies to further reduce the brine volume to achieve minimal liquid discharge (MLD) or zero liquid discharge (ZLD). However, these technologies are energy- and cost-intensive. Therefore, developing feasible and effective strategies to mitigate mineral scaling in membrane desalination is urgently needed to improve the resilience and performance of desalination systems, which will ultimately facilitate the implementation of desalination to mitigate global water scarcity and reduce the environmental risks and cost associated with brine management. Gaining a fundamental understanding of mineral scaling mechanisms and the relationship between scaling behaviors and membrane surface properties are the key prerequisites to the rational design of scaling mitigation strategies in membrane desalination. First, I dedicated efforts to elucidate the scaling mechanism of silica in membrane distillation (MD), a hybrid thermal-membrane desalination technology. Three PVDF membranes with different surface wettability were used to unveil underlying scaling mechanisms of silica in MD. The experimental results revealed that homogeneous nucleation played an important role in inducing silica scaling in MD, while heterogeneous nucleation facilitated the formation of silica scaling layer on the membrane surface. Additionally, I demonstrated that tuning membrane surface wettability was insufficient to reduce silica scaling in MD. Next, I investigated the effect of membrane surface wettability on the scaling kinetics and reversibility of gypsum scaling and silica scaling in MD. Unlike the formation of silica that is regulated by polymerization reactions, the formation of gypsum is governed by crystallization reactions between Ca2+ and SO42- ions. In this work, I demonstrated that superhydrophobic membrane was able to delay the induction time of gypsum scaling and enhanced scaling reversibility, which resulted in increased total water recovery. However, this strategy was not effective to mitigate silica scaling. Such distinct experimental observations between gypsum scaling and silica scaling were attributed to their different formation mechanisms and corresponding interactions with membrane surfaces. Further, in addition to the development of novel membrane materials to resist scaling, the use of anti-scalants to mitigate gypsum scaling and silica scaling was also explored in MD. Although the use of anti-scalants has been widely adopted in the industry, the effectiveness of anti-scalants and the underlying factors that control the anti-scaling efficiency have not been systematically studied. Three anti-scalants with different functional groups were used to elucidate the efficiencies of anti-scalants in mitigating gypsum scaling and silica scaling in MD. Poly(acrylic) acid and poly(ethylenimine), which were enriched with carboxyl and amino groups, were shown to be effective to inhibit gypsum scaling and silica scaling, respectively. The mitigating effect of poly(acrylic) acid molecules on gypsum scaling was due to their effects of stabilizing scaling precursors, whereas poly(ethylenimine) facilitated silica polymerization and altered the morphology of silica scale layer on the membrane surface. This work indicates that anti-scalants with different functional groups are needed for different mineral scaling types. Finally, I compared the efficiencies of membrane surface modification and anti-scalants in mitigating mineral scaling in membrane desalination. The efficiencies of four types of membrane surface modification in mitigating gypsum scaling in reverse osmosis (RO) were compared with the use of anti-scalant poly(acrylic) acid. It was shown that membrane surface modification was only able to reduce the water flux decline caused by gypsum scaling moderately, whereas the use of anti-scalants greatly inhibited gypsum scaling. In addition, I also demonstrated that the use of anti-scalants was highly efficient in preventing gypsum scaling in a combined RO-MD treatment train, which dramatically increased the total water recovery. Therefore, a comparative insight on the efficiencies of different scaling mitigation strategies was provided, which has the potential to guide the selection of the most appropriate strategy to mitigate mineral scaling in membrane desalination.Item Open Access Elucidating the performance and mechanisms of membrane separation: the use of artificial intelligence and a case study of produced water treatment(Colorado State University. Libraries, 2023) Jeong, Nohyeong, author; Tong, Tiezheng, advisor; Carlson, Kenneth, committee member; Sharvelle, Sybil, committee member; Bandhauer, Todd, committee memberPressure-driven membrane technologies such as nanofiltration (NF) and reverse osmosis (RO) have been widely used in water and wastewater treatment because of their effective removal of contaminants and exceptional energy efficiencies. The performance of NF and RO membranes is regulated by the well-documented permeability-selectivity tradeoff, in which an increase of membrane permeability typically occurs at the expense of membrane selectivity and vice versa. To break the upper bound of this tradeoff and further enhance the efficiency of NF and RO treatment, a mechanistic understanding of the solute transport across membranes with pore sizes at the nanometer- or angstrom-scale is required. Current theoretical models relating to solute transport across membranes are limited as the models require precise acquisition of multiple parameters. Machine learning (ML) models, a data-driven approach, have been applied to predict membrane performance and elucidate the membrane separation mechanisms. However, whether the ML models possess appropriate knowledge on membrane separation mechanisms has not yet been studied. Probing knowledge of ML models on membrane separation mechanisms can enhance the reliability of the ML model, which is of great importance to the implementation of ML models for decision-making processes, such as membrane design and selection. Moreover, contrary to the well-controlled experiments for studying the mechanisms or models associated with solute transport, where a limited number of defined solutes are present, membrane treatment has been used to treat wastewater containing diverse organic and inorganic compounds. Thus, along with fundamental research on predictive ML models for membrane performance, investigating the performance of membranes for treating wastewater with complex compositions is also valuable to provide knowledge of solute transport across membranes in practical applications. In this thesis, I present both a fundamental study of probing solute transport across NF and RO membranes using ML models and an applied study that explores membrane treatment of unconventional oil and gas (UOG) produced water. First, the reliability of the ML model as a tool to predict membrane performance was investigated. Specifically, the influence of data leakage on the ML model performance, as well as the solution to prevent this issue, was explored to evaluate the prediction capability of the ML model objectively. I discovered that data leakage can lead to falsely high prediction accuracy of the ML model, and appropriate data splitting for the training, validation, and testing dataset is necessary to avoid data leakage. Second, the underlying knowledge of ML models for organic and inorganic solute transport across polyamide membranes was investigated by using a model interpretation method (i.e., Shapley additive explanation, SHAP). I not only tested whether ML models are able to possess adequate knowledge on solute transport, but also utilized the SHAP method to reveal solute transport mechanisms that are typically obtained using tedious, well-controlled experiments. For the ML model applied to predicting the rejection of organic constituents by NF and RO membranes, I found that the ML model had proper knowledge of size exclusion, but its understanding of electrostatic interaction and adsorption remains rudimentary. By using ML to predict the rejection of inorganic constituents, I elucidated that explainable artificial intelligence (XAI) can capture the major governing mechanisms of ion/salt transport across polyamide membranes (i.e., size exclusion and electrostatic interaction), which have different importance for the transport of single salt, cation, anion transport in mixture salt solution. Lastly, the performance of RO/NF membranes for the treatment of UOG produced water was explored as a case study, which comprehensively investigated the chemical composition and toxicity level of the treated water. NF permeates, which still had high salinities and high boron concentrations, were found to be inappropriate for irrigation and livestock drinking water, while RO membranes effectively removed most pollutants and met most water quality standards for beneficial reuse (i.e., irrigation and livestock drinking water). However, the chloride concentrations and sodium adsorption ratio (SAR) values of RO permeates were still higher than the recommended thresholds for irrigation. Also, surfactants with molecular weights higher than the molecular weight cut-off of RO/NF membranes were able to traverse through the membrane, indicating that NF and RO are not complete barriers against organic contaminants. The toxicity test results of NF and RO permeates demonstrated that NF permeates were still toxic to Daphnia, while RO permeates showed less toxicity than NF permeates or no toxicity. The toxicity level of NF and RO permeates showed a correlation with salinity in the permeates, which might be the main driver of the toxicity. I envision that my thesis provides a framework to evaluate the knowledge and reliability of ML model predictions, while presenting a comprehensive investigation on membrane performance and the potential risks associated with membrane treatment of UOG produced water for beneficial reuse. The knowledge gained in this thesis improves our capability for rational membrane material design and selection, which has the potential to lead to more efficient NF and RO technologies for sustainable water and wastewater treatment.Item Open Access Evaluation of a trickle flow leach bed reactor for anaerobic digestion of high solids cattle manure(Colorado State University. Libraries, 2013) Hanif Abdul Karim, Asma, author; Sharvelle, Sybil, advisor; Carlson, Kenneth, committee member; Davis, Jessica, committee memberAnaerobic digestion (AD) of cattle manure from feedlots and dairies is of increasing interest in Colorado due to its abundant availability. Colorado is the one of the highest producer of high solids cattle manure (HSCM) in the United States. Despite the available resources, Colorado currently has only one operational anaerobic digester treating manure (AgSTAR EPA 2011), which is located at a hog farm in Lamar. Arid climate and limited water resources in Colorado render the implementation of high water demanding conventional AD processes. Studies to date have proposed high solids AD systems capable of digesting organic solid waste (OSW) not more than 40% total solids (TS). Lab tests have shown that HSCM produced in Greeley (Colorado) has an average of 89.6% TS. Multi-stage leach bed reactor (MSLBR) system proposed in the current study is capable of handling HSCM of up to 90% TS. In this system, hydrolysis and methanogenesis are carried out in separate reactors for the optimization of each stage. Hydrolysis is carried out in a trickle flow leach bed reactor (TFLBR) and methanogenesis is carried out in a high rate anaerobic digester (HRAD) like an upflow anaerobic sludge blanket (UASB) reactor or a fixed film reactor. Since leach bed reactors (LBRs) are high solids reactors, studies have indicated clogging issues in LBRs handling 26% TS. Since TFLBRs are subjected to hydrolyze upto 90% TS, obtaining hydraulic flow through the reactor is a challenge. The objective of this research is to (a) ensure good hydraulic flow through the TFLBRs and (b) evaluate and optimize the performance of the TFLBR to effectively hydrolyze the HSCM. The system was operated as a batch process with a hydraulic retention time (HRT) of 42 days without leachate recirculation. A layer of sand was added as dispersion media on top of the manure bed in the TFLBRs. This promoted good hydraulic flow through the reactor eliminating clogging issues. Organic leaching potential of a single pass (without leachate recirculation) TFLBR configuration was evaluated in terms of chemical oxygen demand (COD). Manure is naturally rich in nutrients essential for microbial growth in AD. In a typical MSLBR system, the TFLBRs are subjected to leachate recirculation, conserving the essential nutrients in the system. However, in this single pass system, the leachate removal would flush out the nutrients in the TFLBRs over time. So, nutrient solution was added to the TFLBRs to provide a constant supply of essential nutrients in the reactors for the purpose of this study and would not be necessary in a leachate recirculated TFLBR. A comparison between nutrient dosed and non-nutrient dosed TFLBRs was performed. The non-nutrient dosed and nutrient dosed TFLBRs indicated a COD reduction of approximately 66.3% and 73.5% respectively, in total in terms of dry mass. A total reduction in volatile solids (VS) of approximately 46.3% and 44.7% was observed in the non-nutrient dosed and nutrient dosed TFLBRs, respectively. Biochemical methane potential (BCMP) tests indicated a CH4 potential of approximately 0.17 L CH4/g COD leached and 0.13 L CH4/g COD leached from the non-nutrient dosed and nutrient dosed TFLBRs, respectively. Concentration of inorganics leached from the TFLBR was monitored periodically.Item Open Access Evaluation of commercially available on-line analyzers for measurement of natural gas contaminants(Colorado State University. Libraries, 2021) Zineddin, Khalid Mohamad, author; Olsen, Daniel, advisor; Jathar, Shantanu, committee member; Carlson, Kenneth, committee memberThe proliferation of natural gas usage and extraction has led to never-before-seen levels of demand across the United States in several industries. Because of this increased demand for quality processed natural gas, a need has arisen to streamline its processing and distribution for faster and more efficient delivery to customers. One method to achieve this is by consolidating natural gas contaminant analyzers at processing and distribution sites; current sites typically operate with multiple separate gas analyzers, each dedicated to measuring one individual contaminant species. Recent developments in laser-based gas composition analysis (in particular Tunable Diode Laser Absorption Spectroscopy or TDLAS) as well as advances in gas chromatograph (GC) technology have given rise to commercially-available analyzing instruments which are capable of detecting and measuring multiple gas contaminants simultaneously. In order to verify the effectiveness and reliability of these new technologies, three specific gas analyzing instruments (herein referred to as Instruments A, B, and C) were selected for in-depth laboratory and field testing. The main objective of this research is to quantify the accuracy, precision and uncertainty of these new multi-species gas analyzers and compare their performance with existing gas analyzers currently in use by natural gas processing and distribution organizations. Four natural gas contaminant species were specified for evaluation of the natural gas contaminant analyzers; these are water (H2O), hydrogen sulfide (H2S), oxygen (O2) and carbon dioxide (CO2). Laboratory testing was performed first by sampling existing natural gas from three separate sources then by custom gas mixtures blended in-house from pure component gases to simulate four levels of contaminants (Low, two Mid-range blends, and High). For results verification a sample of each gas mixture (both source natural gas and custom blends) was submitted to EMPACT Analytical Inc. for composition analysis. Following laboratory testing, two instruments were selected for ongoing (through February 2022) field testing to simulate "real-world" conditions and their results are compared with those of existing on-site gas contaminant analyzers. It was concluded that Instrument A (TDLAS-based) was the most accurate and reliable of the three analyzers under test and required the least amount of user intervention to maintain satisfactory operation. Instrument B (also TDLAS-based) ranked a close second-place, having slightly less accuracy than Instrument A in measuring gas concentrations and falling short by its inability to detect O2; it also experienced multiple failures which required user intervention and/or shipping the instrument back to the manufacturer for repair and reconfiguration. Instrument C (GC-based) performed the poorest of the three analyzers with very low accuracy in measuring O2 and H2O to the point of being essentially non-detected species; it also required in-depth user intervention for initial setup and on multiple occasions during operation which was determined to be inappropriate for the intended application.Item Open Access Evaluation of oxidation and adsorption techniques for taste and odor and toxin removal(Colorado State University. Libraries, 2017) Sampath, Muthukumaran, author; Omur-Ozbek, Pinar, advisor; Carlson, Kenneth, committee member; Dooley, Gregory, committee memberThe cyanobacteria, also known as blue-green algae, owe their name to the presence of photosynthetic pigments. Cyanobacteria are a major group of bacteria that occur throughout the world. Freshwater cyanobacteria may accumulate in surface water supplies as "blooms" posing as an environmental hazard because of the release of water soluble toxic compounds, called cyanotoxins. Especially massive blooms of blue–green algae in the surface waters used as drinking water resources may lead to taste and odor problems during the summer and fall, they may also produce cyanotoxins. Since the taste and odor compounds, Geosmin (GSM) and 2-Methylisoborneol (2-MIB) can be easily detected by the human nose at low concentrations of 2-5 ng/L, the surveillance of harmful toxins such as microcystin-LR may be easily performed by sensory analyses due to the likely co-occurrences of the two types of metabolites. This research focused on removal of taste and odor compounds (GSM, 2-MIB) and microcystin-LR with five oxidants: chlorine, chlorine dioxide, potassium permanganate, ozone, mixed oxidants (MiOX) and powdered activated carbon (PAC) using Ralston Reservoir water as reagent water collected in early April, 2014. The objective of the study was to develop a bench scale treatment process efficacy information that Denver Water can utilize to decide on a treatment technique for taste and odor control. The Design Expert software was used to determine the optimum dose of the oxidants for an acceptable treatment level.Item Open Access Fate and transport of surfactants in graywater when applied to soil(Colorado State University. Libraries, 2013) Huang, Zhaohua, author; Sharvelle, Sybil, advisor; Carlson, Kenneth, committee member; Stromberger, Mary, committee memberGraywater reuse for irrigation has been considered an efficient way to reduce demand on water supply. Concerns, however exist regarding the potential impacts that graywater pose to soil quality. In particular, the fate of surfactants, the primary component in personal care and cleaning products, is not well understood. The objective for this study was to gain a better understanding of the adsorption behavior of surfactant onto soils, with particular attention on the effect of the organic matter and soil texture, then provide a suggestion about the kind of surfactants and soil be reused during graywater irrigation. Surfactants linear alkylbenzene sulfonates (LAS) (anionic), alcohol ethoxysulfates (AES) (anionic) and alcohol ethoxylates (AE) (nonionic) were applied to three different soils with varying organic matter (OM) and clay fraction column studies. Adsorption results were obtained from leachate and soil samples. The fraction ranges of leached surfactants to sorbed of LAS, AES and AE were 0.10-0.42, 0.42-2.35, 0.06-0.77 respectively. The results indicated that AES had the most potential leaching capacity, which mean they could reach deeper soil layer even groundwater systems. On the other hand, from soil properties, OM played an important role in the adsorption of surfactants, both anionics and nonionics, whereas, the clay fraction content had a negative effect on anionic surfactants sorption (p=0.006, 0.002 for LAS and AES), possibly due to an increase in negative charge, repulsion forces as clay content increasing, but not significant on nonionic surfactants with clay content increased from 33% - 46% (p=0.986 for AE). Meanwhile, AES homologues which contained different number of ethylene oxide (EO) groups were studied. Results indicated that adsorption increased as EO chain increased. Based on the results above, AE were recommended for graywater irrigation in terms of surfactants with relative high OM.Item Open Access Food waste diversion for enhanced methane gas production at the Drake Water Reclamation Facility(Colorado State University. Libraries, 2012) Robbins, Cristian Arthur, author; Sharvelle, Sybil, advisor; Carlson, Kenneth, committee member; Keske, Catherine, committee memberFood waste diversion to enhance methane gas production in municipal wastewater treatment plants is an emerging trend in the United States. The methane gas produced in anaerobic digesters of a municipal wastewater treatment plant can be used to produce renewable energy to meet electric and heating needs of the plant. The Drake Water Reclamation Facility in Fort Collins, Colorado is very interested in implementing energy generation from anaerobic digester biogas and a food waste diversion program. The objective of this study is to determine the efficacy and viability of implementing a food waste diversion program coupled with energy generation technology to provide electricity and heating generation to meet the plant's needs. A food waste characterization study of the Colorado State University's Ram's Horn Dining Facility processed food waste was conducted to determine important characteristics of a readily available food waste. An analysis of the operating capacity of the Drake Wastewater Reclamation Facility anaerobic digesters was conducted to determine the maximum amount of food waste that could be added on a daily basis. The maximum amount of food waste that could be added to the Drake Water Reclamation Facility anaerobic digesters is 37.5 tons per day. 2010 data for the Drake anaerobic digesters was analyzed and used as a baseline for analysis of the addition of various amounts of food waste ranging from 800 pounds of food waste per day to the maximum amount of 37.5 tons per day. The effects of the food waste on anaerobic digester biogas production and solids reduction in the digester were reported. Various technologies for generating energy from biogas were evaluated using reported cost data and characteristics. An economic analysis utilizing flared methane gas as fuel for the various technologies was completed which showed that microturbine and reciprocating engine technologies are economically viable options for the Drake Water Reclamation Facility to use for both electricity and heating generation. A triple bottom line analysis, with a rigorous economic analysis, of implementing a food waste diversion program at the Drake Water Reclamation Facility was conducted. Costs associated with a food waste processing facility and associated equipment was outlined and evaluated against the energy savings that enhanced methane gas production from various amounts of food waste addition provided. It was determined that it is not economically viable for the Drake Water Reclamation Facility to implement a food waste diversion program at this time. If energy prices rise and cost of equipment for a food waste diversion program decrease in the future, then the economics of this project may improve making it more viable.Item Open Access Membrane treatment of wastewater from oil and gas production: motivations and material innovation(Colorado State University. Libraries, 2022) Du, Xuewei, author; Tong, Tiezheng, advisor; Carlson, Kenneth, committee member; Sharvelle, Sybil, committee member; Kota, Arun Kumar, committee memberThe rise of shale oil and gas (O&G) via hydraulic fracturing (HF) has boosted energy production in the United States. Further, many of the U.S. shale plays coincide with water-scarce areas that suffer from prolonged drought periodically. The substantial volumes of water consumption and wastewater generation associated with O&G activities intensify local water stress and create a challenge of wastewater management, rendering treatment and reuse of O&G wastewater an essential strategy to improve water sustainability of O&G-producing regions. Herein, the main goal of this dissertation is to facilitate the reuse and treatment of O&G wastewater in order to promote water sustainability of O&G-producing regions. To achieve this goal, two sets of studies were performed, which pertain to (1) data analysis to investigate the water footprint of O&G production under hydrodynamic variation; and (2) developing novel membrane materials for more efficient O&G wastewater treatment. First, I investigated the relationship of hydroclimate variation with the activities and water footprint of O&G production in Colorado, one of the major O&G producing states in the U.S. I discovered that hydroclimate variation imposes a negligible impact on well number and water footprint of O&G production. However, the intensive water consumption by HF under arid conditions could escalate competition for water resources at the local scale. Further, I expanded the research scope to estimate the water consumption by HF activities under different hydroclimate conditions in eleven O&G-producing states in the central and western U.S. from 2011 to 2020. The results show that the water consumption under abnormally dry or drought climates accounted for 49.7% (475.3 billion gallons) of total water usage of HF, with 9% (86.1 billion gallons) of water usage occurred under extreme or exceptional drought conditions. The water usage of HF under arid conditions can translate to high densities of water footprint at the local scale, equivalent to more than 50% of the annual water usage by the irrigation and domestic sectors in 21-47 and 11-51 counties (depending on the specific year), respectively. Such water stress imposed by O&G production, however, could be effectively mitigated by the reuse of flowback and produced water. This renders wastewater reuse necessary to maintain water sustainability of O&G-producing regions in the context of both a rising O&G industry and a changing climate. Second, I focused on developing novel membrane materials for the treatment of O&G wastewater by membrane distillation (MD), which is an emerging technology showing promise for efficient desalination of high salinity industrial wastewater. I investigated the impacts of membrane surface wettability on the treatment of O&G wastewaters by MD. From this study, omniphobic membranes with high wetting resistance showed more robust performance, but they also required the use of toxic long-chain per- and polyfluoroalkyl substances (PFASs, ≥ 8 fluorinated carbons) during fabrication process and displayed lower water productivity compared to conventional hydrophobic membranes. Then, I developed highly wetting-resistant MD membranes while avoiding the use of long-chain PFASs, which is essential to improve the viability of MD for resilient and sustainable MD desalination. I demonstrated that long-chain PFASs are not required when designing membranes with high wetting resistance. Instead, the combination of hierarchical texture and (ultra)short-chain fluorocarbons are able to create MD membranes with exceptional wetting resistance. Finally, I also elucidated the fundamental relationship between membrane wetting resistance and water vapor permeability in the MD process, which needs to be taken into consideration when designing and selecting appropriate membranes for effective MD treatment of O&G wastewater. I identified that a trade-off exists between wetting resistance and water vapor permeability of MD membranes, and also unveiled the mechanism of such a trade-off by revealing the importance of water-air interfacial area in regulating water vapor transport through microporous membranes. I envision that the novel insights on omniphobic membrane fabrication and the wetting resistance-vapor permeability trade-off will pave the way for more rational design of MD membranes for sustainable O&G wastewater treatment applications.Item Open Access Modelling and analysis of systems on offshore oil and gas platforms(Colorado State University. Libraries, 2019) Grassian, David, author; Olsen, Daniel, advisor; Bradley, Thomas, committee member; Carlson, Kenneth, committee member; Marchese, Anthony, committee memberThis research examines oil and gas systems from the seemingly underutilized perspective of energy; this is counterintuitive since the energy content of hydrocarbon products is its most distinguishing characteristic and the very reason why it is valued by society. It is clear that the amount of energy required to extract crude oil is increasing over time, at the long-term global level, and at the much shorter time span of individual fields. The global trend is a well-documented phenomenon and is related to the depletion of the most energetically favorable reservoirs and a coincidental growing global demand for energy. Concerning existing fields, it is often necessary to implement increasingly higher energy intensity methods to extract the remaining crude oil resources. These trends are the impetus for the industry to gain a better understanding of the relationship between the application of energy and the production of crude oil across a wide spectrum of production methods.Item Open Access Nitrogen recovery from anaerobic digestate via ammonia stripping and absorbing with a nitrified solution(Colorado State University. Libraries, 2021) Alhelal, Ismail Ibrahim, author; Reardon, Kenneth F., advisor; Sharvelle, Sybil, advisor; Perkins, Tracy, committee member; Carlson, Kenneth, committee memberAnimal wastes cause environmental pollution, including contamination of air and water, when not managed properly. For example, stored livestock manure releases greenhouse gasses, which contribute to air pollution and global warming. Anaerobic digesters have been used for animal waste treatment in order to reduce the environmental impacts of animal wastes. However, current anaerobic digestion systems have serious economical and operational challenges such as high capital cost, low byproduct price, and ammonia toxicity. Therefore, more research is needed to increase the benefits of anaerobic digestion and reduce its challenges. The goal of this project was to improve the cost and performance of anaerobic digesters by enhancing their byproducts, biogas and fertilizer, while reducing one of their serious operational challenges, ammonia toxicity. To achieve these goals, this project investigated an integrated anaerobic digestion nitrogen recovery process that includes anaerobic digestion, nitrogen recovery and nitrification. The nitrogen produced during anaerobic digestion is volatilized in a stripper, captured in an absorber, and converted to nitrogen certified organic fertilizer in the nitrification process. Recovering the ammonia in anaerobic digesters not only produces organic fertilizer but also reduces ammonia toxicity, enhancing biogas production. Experiments and modeling were used to identify appropriate operating conditions for the stripper and absorber units of the proposed process. The objective of the nitrogen recovery system experiments was to find the best operational conditions as well as to evaluate the performance of the nitrification solution as an ammonia absorbent. Stripping and absorption columns were designed to measure the ammoniacal nitrogen recovery. The ammonia stripping and absorption extents were calculated for several operational conditions: stripping and absorption feed pH, stripping temperature and absorbent nitrogen concentration. The experimental results showed that a feed pH of 10 was optimal for ammonia stripping in the pH range 8.5–10.5, providing an ammonia stripping extent of 77%, while the optimal stripping temperature was 50 °C since it provides the highest extent of ammonia stripping in the tested range of 35–65 °C. An Aspen Plus simulation model was also developed for the ammonia stripping process to calculate the effects of the number of equilibrium stages, feed pH, and the amount of CO2 in the stripping gas. The model showed that the use of three equilibrium stages, a feed pH of 10, and having no CO2 in the stripping gas provides the most feasible operational conditions considering the stripping performance and economics. Moreover, the data suggested that the stripping units will require pH control for effective ammonia recovery since the pH of the stripper decreases with the ammonia removal. For the ammonia absorption unit, the experimental data showed that ammonia absorption was not greatly impacted by the feed pH nor by the concertation of nitrogen in the liquid feed. With a low concentration of nitrogen in the liquid feed (2 g/L NH4NO3 as N), the extents of ammonia absorption for feed pH values of 7 and 2 were 82% and 92, respectively. However, the extents of ammonia absorption using a high concentration nitrogen liquid feed (7 g/L NH4NO3) for feed pH values of 7 and 2 decreased to 70% and 85%, respectively. However, a Two-Factor ANOVA test with replication has a p-value >0.05, so there is no statistically significant difference in the ammonia absorption due to the feed pH nor in the concentration of nitrogen in the absorbent. Consequently, it can be concluded that nitrified solution can be used as an ammonia absorbent because it can affectively absorb ammonia over a wide range of its pH and its nitrogen concentration. This project demonstrated that it is possible to recover nitrogen in an integrated anaerobic digestion process and determined recommended operational conditions for the nitrogen recovery system. The novel integrated anaerobic digestion system proposed in this work decreases ammonia toxicity for anaerobic digestion, while increasing potential for revenue from increased biogas yield and recovery of ammonia fertilizer. increasing the biogas yield, producing organic fertilizer and decreasing ammonia toxicity.Item Open Access Nutrient management control regulation and preparedness of a northern Colorado wastewater treatment plant(Colorado State University. Libraries, 2013) Venkatapathi, Keerthivasan, author; Omur-Ozbek, Pinar, advisor; Carlson, Kenneth, committee member; Reardon, Kenneth, committee memberExcessive nutrients in wastewater treatment plant (WWTP) effluents instigate eutrophication of receiving water bodies. Colorado Department of Public Health and Environment (CDPHE) adopted nutrient management control regulation, also known as regulation 85, to moderate the nutrients released by point sources such as the WWTP effluents. City of Loveland WWTP was selected as the study plant to determine a new treatment process configuration to meet the new limits of total phosphorus < 1 mg/L and total inorganic nitrogen < 15 mg/L in the effluent. BioWin, a windows based modeling software, was used to model and simulate the City of Loveland WWTP. Existing activated sludge step feed process configuration was modeled along with proposed anaerobic, anoxic, oxic (A2O) process for design influent flow of 10 MGD and 5-stage Bardenpho process for future flow of 12 MGD along with A2O process. Existing configuration was modeled to establish the accuracy of BioWin. 5 stage Bardenpho process modeling indicates that higher design HRT of 2 days for anaerobic, 4 days for anoxic, 6 days for aerobic, 4 days for secondary anoxic and 1 day for reaeration has better treatment removal efficiency for nutrients with methanol dosage of 250 gal/d and 1Q internal recycle. Model simulations for A2O process reveals that aerobic reactor to anaerobic reactor volume ratios from 3 to 4 and aerobic reactor to anoxic reactor volume ratio of 2.2 along with internal recycle of 1Q has the better nutrient removal efficiency for design flow of 10 MGD. For 12 MGD influent flow, volume of reactors was increased by 20% to compensate for 20% increase in the flow. Previously mentioned reactor volume ratios are feasible for 12 MGD influent flow with volume ratios of 3 and 4 for aerobic to anaerobic reactors and volume ratios of 1.8 and 2 for aerobic to anoxic reactors. Modeling results indicates that increasing the reactor volume ratio for increased flow can result in better treatment for removal of nutrients with a conservative volume, reducing the operational and maintenance cost.Item Open Access Occurrence of cyclo-siloxanes in wastewater treatment plants - quantification and monitoring(Colorado State University. Libraries, 2012) Kulkarni, Harshad Vijay, author; Omur-Ozbek, Pinar, advisor; Carlson, Kenneth, committee member; Dooley, Gregory, committee memberSiloxanes are persistent, bio-accumulative and toxic emerging contaminants introduced to wastewater from common healthcare and biomedical products, and various industrial processes. They remain unchanged through wastewater treatment and a considerable portion ends up in surface waters through effluent discharge. 30 to 60 ng/L Decamethylcyclopentasiloxane (D5) was detected in two UK Rivers, while ~400 ng/L of D5 may be found in wastewater effluents. Hence, siloxanes are under consideration by Canadian Environmental Assessment Agency and UK Environment Agency for drinking water regulations. Siloxanes are hydrophobic and also accumulate in activated sludge and biogas, causing mechanical problems due to scaling. This research aims: to quantify the siloxanes in sludge samples obtained from Loveland, CO wastewater treatment plant (WWTP); and to study their removal. A method was developed to effectively extract siloxanes from activated sludge samples using liquid extraction followed by quantification with gas chromatography/mass spectrometry. Results for Loveland Wastewater Treatment Plant samples indicated that Octamethylcyclotetrasiloxane (D4) and Decamethylcyclopentasiloxane D5 are present up to 17.11 µg/g dried-sludge. The effectiveness of H2O2 in siloxane removal was investigated. Sludge samples were spiked with D4 and D5 at 12 mg/g and were treated with 1ml, 3ml, and 5ml of 30% H2O2 for 1hr, 2hr, and 3hr reaction time each. Results indicated a 72% reduction in D4 and D5 levels after 3 hrs.Item Open Access Reduction of methane emissions through in-cylinder methods on a lean-burn four-stroke natural gas engine(Colorado State University. Libraries, 2023) Bayer, Justin, author; Olsen, Daniel, advisor; Windom, Bret, committee member; Carlson, Kenneth, committee memberThe U.S. utilizes over 27 trillion cubic feet of natural gas per year for a wide range of uses, including heating and electricity production, according to the U.S. Energy Information Administration. Natural gas (NG) engines used to compress natural gas and generate electricity account for nearly one-quarter of the total methane emissions in the gathering and boosting sector. These methane emissions are referred to as fuel (methane) slip since they originate from the engine fuel supply and result from incomplete combustion. The primary mechanism leading to unburned methane is related to the engine crevice volume. The crevice volume is the region between the side of the piston and the cylinder wall. This region is particularly difficult for the flame to propagate into because the gap is generally smaller than the quench distance. This research evaluates multiple in-cylinder methods to attempt to reduce the methane slip. One of the mitigation strategies is hydrogen blending with the engine's natural gas fuel supply as a means of methane reduction. Converge Computational Fluid Dynamics (CFD) and a Caterpillar G3516J model are implemented to analyze the effects of hydrogen blending. The G3516J engine is a lean-burn engine commonly used for gas compression in the US NG pipeline system. Converge CFD is a solver that couples combustion chemistry and adaptive mesh refinement to model combustion accurately. Simulations of combustion with NG-hydrogen blended fuel are performed with constant fuel energy, achieved by adjusting boost pressure at a constant equivalence ratio. Combustion cycle simulations are performed at various hydrogen blending levels, and the methane emissions are evaluated at the end of the cycle and compared. In addition, the fuel mixtures' pressure is adjusted to reflect similar indicated power and a similar emission comparison is made. The second mitigation strategy that is explored is induced autoignition. This strategy involved advancing the engine's spark timing to hopefully increase the temperature and pressure in the cylinder to have the end-gas auto-ignite and thus burn more fuel that would otherwise become methane slip. This research also incorporates installing a G3516J engine at the Engines and Energies Conversion Laboratory. Advancing the spark timing in the simulations did not show signs of end gas auto-ignition. Although this is the case, the advanced spark timing showed a decrease in unburned methane compared to the baseline. The spark timing with the lowest unburned CH4 percentage decreases from 3.00% to 2.26% by advancing the spark timing by ten crank angle degrees. The hydrogen blending also showed lower unburned CH4 percentages compared to the baseline. After adjusting the simulations to match indicated power output, the lowest results decreased the unburned percentage from 3.00% to 1.20%. NOx emissions were increased by 129% compared to the baseline in the most extreme simulation case. Leaning the fuel mixture lowered the NOx emissions to within 6% of the baseline while still lowering the unburned percentage from 3.00% to 2.67%.Item Open Access Restoration of a small, shallow, eutrophic lake by submerged aeration and comparison with a similar lake(Colorado State University. Libraries, 2013) Zhang, Xiaoju, author; Roesner, Larry, advisor; Sharvelle, Sybil, committee member; Carlson, Kenneth, committee member; Catton, Kimberly, committee member; Stednick, John, committee memberA submerged aeration system was introduced to Fossil Creek Lake, a small scale, shallow, eutrophic, very hard water lake, to increase hypolimnetic dissolved oxygen levels, prevent thermal stratification and improve water quality. There has been a long-standing concentration of research effort on aeration systems in deep water lakes, but very little attention has been paid to urban shallow lakes. This research describes the investigation of physical, chemical and biological parameters of the lake before and after aeration. The aeration system gradually and permanently improved the water quality of the entire lake and eliminated many undesirable lake conditions. With the destratification caused by the aeration system, the bottom water dissolved oxygen concentrations increased significantly (p<0.05) from less than 1 mg L-1 to above 4 mg L-1. Furthermore, the vertical variations of many water quality parameters have been significantly reduced with the complete mixing of the entire lake, which are: pH, specific conductance, nutrients, chlorophyll a, alkalinity, total suspended solids, sulfate, and hardness. However, limiting nutrient, trophic status and water clarity of the lake were barely affected by the aeration system. Multiple years of continuous aeration on this lake may be necessary to achieve a lower trophic condition. In addition, this research also discusses sources of extremely high sulfate and sulfide levels of Fossil Creek Lake and provides aeration startup recommendations for lakes with high sulfide concentrations. In this study, the water quality parameters, trophic status and nutrient mass balance of Fossil Creek Lake were compared with Sheldon Lake from April to December in 2010 and 2011, a turbid, phytoplankton-dominated lake, which is restored by submerged aeration after bottom sediment removal. Monthly mean values of temperature and dissolved oxygen, as well as monthly mean total alkalinity, total inorganic carbon, total organic carbon, orthophosphate and total nitrogen concentrations showed similar distributions in the two lakes. However, the total suspended solids and chlorophyll a concentrations were significantly (p<0.05) higher in Sheldon Lake than those in Fossil Creek Lake. In addition, the Secchi disk transparency of Sheldon Lake rarely exceeded 0.5 m, which was lower than Fossil Creek Lake (>1m). Models relating water clarity to monitored water quality parameters over a period of 2 years are developed in both lakes. In Fossil Creek Lake, the water clarity is not only significantly related with inorganic carbon, nitrogen and orthophosphate, but also hardness, alkalinity, total suspended solids and chlorophyll a concentrations. But in Sheldon Lake, the lake clarity is only significantly related to two parameters, which are total suspended solids and chlorophyll a concentrations.Item Open Access Systems-based approaches for evaluating residential-based hazards to inform environmental exposure intervention design(Colorado State University. Libraries, 2022) Oke, Oluwatobi Olamiposi, author; Carter, Ellison, advisor; Magzamen, Sheryl, committee member; Carlson, Kenneth, committee member; Sharvelle, Sybil, committee memberHousing is an essential aspect of the physical built environment, where people spend most of their time, and a key determinant of health. Sub-standard and poor physical housing conditions (e.g., disrepair, deferred maintenance, and deteriorated physical environment) and exposures of inhabitants to lead, pests, air pollutants, and other indoor contaminants are associated with a wide range of health conditions, including respiratory infections, asthma, lead poisoning, injuries, chronic disease outcomes (e.g., cardiopulmonary conditions) and mental health illness. The environmental justice (EJ) research community that has focused on residential exposures has documented that adverse environmental exposures associated with residential settings and built environments are unevenly distributed and often disproportionately affect low-income and socioeconomically disadvantaged populations in the United States. Hence, work remains to be done if we intend to develop and maintain healthy residential environments for vulnerable population groups. However, a central challenge to this effort is the complex system of sources and source activities that contribute to and drive residential exposures, which makes it difficult to identify and isolate dominant sources. This dissertation sought to holistically investigate sources of three prevalent home-based exposures in the United States – i.e., flooding, lead, and indoor chemical mixtures. Through a combination of empirically-based and modeling approaches, this work brought together information on physical dwellings, their conditions and surroundings, the overlying sociodemographic characteristics of the people living in them, and the behaviors and activities that people undertake in their homes. The central hypothesis of this work has been that this approach would improve the identification of sources and their relative contributions to exposures in the residential environment. Significant findings from this work include: (i) socioeconomically disadvantaged populations in all three studies tended to have higher environmental exposures, regardless of exposure type, including natural hazards (i.e., floods), legacy pollution (i.e., lead), and pollutants driven by daily human activity (indoor air pollutants of outdoor origin); (ii) sources of environmental exposures varied within the same study and, at times, were more subtle than initially hypothesized in the literature, suggesting that the more holistic approaches taken in this work have practical value; and (iii) residential interventions to reduce adverse exposures could provide some measurable benefits to residents if customized to local occupants' needs in solving more building-related problems and providing higher residential satisfaction. In two of the three studies, we have worked closely or directly with the communities from which the data are collected. Thus, we expect outputs from this work to improve the design and delivery of home-based interventions for adverse environmental exposures through direct engagement with local decision-makers and more traditional scholarly communication channels.Item Open Access Understanding meteorological impacts on ambient PM2.5 concentrations using random forest models in Beijing(Colorado State University. Libraries, 2022) Brehmer, Collin, author; Carter, Ellison, advisor; Bond, Tami, committee member; Carlson, Kenneth, committee member; Pierce, Jeffrey, committee memberPolicymakers and non-governmental organizations have been implementing policies and interventions designed to reduce exposure to hazardous air pollution. Having knowledge of how non-policy related factors (i.e., meteorology) impact air pollution concentrations in a given study area can better inform longitudinal studies of the effects of the policy on air pollution and health. In this study, we apply a random forest machine learning approach to evaluate how meteorological factors including temperature, relative humidity, wind speed, wind direction, and boundary layer height influence daily PM2.5 concentrations in rural Beijing villages during heating months (January and February of 2019 and 2020). Ten-fold cross validation indicated good model performance with an overall r2 of 0.85 for season 1, and 0.93 for season 2. The models were able to identify variables that were the most important for predicting PM2.5 concentrations both field seasons (relative humidity) and variables that had changes in relative importance between seasons (temperature and boundary layer height). Additionally, examination of one and two-way partial dependence plots as well as interactions through Friedman's H-statistic granted insight into how meteorology variables influence PM2.5 concentrations. Findings from this work provide a basis for adjusting for meteorological variability in important indicators of air quality like PM2.5 concentrations in an ongoing real-world policy evaluation of a province-wide ban on household use of coal for space heating in Beijing, which is critical for isolating (to the extent possible) changes in measured pollutant concentrations attributable to the policy.