Browsing by Author "Fischer, Emily V., committee member"
Now showing 1 - 9 of 9
Results Per Page
Sort Options
Item Open Access Air quality impacts from unconventional oil and gas development(Colorado State University. Libraries, 2024) Ku, I-Ting, author; Collett, Jeffrey L., Jr., advisor; Fischer, Emily V., committee member; Carlson, Kenneth H., committee member; Kreidenweis, Sonia, committee memberUnconventional oil and natural gas development (UOGD) has expanded rapidly across the United States raising concerns about associated air quality impacts. While significant effort has been made to quantify and limit methane emissions, relatively few observations have been made of emitted Volatile Organic Compounds (VOCs). Extensive air monitoring during development of several large, multi-well pads in Broomfield, Colorado, in the Denver-Julesburg Basin, provides a novel opportunity to examine changes in local concentrations of air toxics and other VOCs during drilling and completions of new wells. With simultaneous measurements of methane and 50 VOCs from October 2018 to December 2022 at as many as 19 sites near well pads, in adjacent neighborhoods, and at a more distant reference location, we identify impacts from each phase of well development and production. In Part 1, we report how emissions from Broomfield pre-production and production operations influence air toxics and other VOC concentrations at nearby locations. Use of weekly, time-integrated canisters, a Proton Transfer Reaction Mass Spectrometer (PTR-MS), continuous photoionization detectors (PID) to trigger canister collection upon detection of VOC-rich plumes, and an instrumented vehicle, provided a powerful suite of measurements to characterize both transient plumes and longer-term changes in air quality. Prior to the start of well development, VOC gradients were small across Broomfield. Once drilling commenced, concentrations of oil and gas (O&G) related VOCs, including alkanes and aromatics, increased around active well pads. Concentration increases were clearly apparent during certain operations, including drilling, coil tubing/millout operations, and production tubing installation. Emissions of C8-C10 n-alkanes during drilling operations highlighted the importance of VOC emissions from a synthetic drilling mud chosen to reduce odor impacts. More than 90 transient plumes were sampled and connected with specific UOGD operations. The chemical signatures of these plumes differed by operation type. Concentrations of individual, O&G-related VOCs in these plumes were often several orders of magnitude higher than in background air, with maximum ethane and benzene concentrations of 79,600 and 819 ppbv, respectively. Study measurements highlight future emission mitigation opportunities during UOGD operations, including better control of emissions from shakers that separate drill cuttings from drilling mud, production separator maintenance operations, and periodic emptying of sand cans during flowback operations. In Part 2 OH reactivities (OHR) were calculated to examine the potential of emitted VOCs to contribute to regional ozone formation. NO2 was the largest contributor to OHR during winter when OHR values peaked, while VOCs dominated OH sinks during summer. Oxygenated VOCs and C3-C7 n-alkanes, closely associated with O&G activities, were primary contributors to OHR levels during the summer ozone season. In Part 3 we leverage observations from Broomfield and other Colorado O&G air quality studies to examine relationships between O&G emissions of methane and VOCs. A key goal is to determine whether more commonly measured methane emissions can serve as a surrogate to estimate emissions of less frequently measured compounds such as benzene, a key air toxic. While strong correlations are observed between benzene and methane emissions in some situations, considerable variability is observed in this relationship across locations and operations suggesting caution in assuming that reductions in methane emissions will yield proportionate reductions in releases of air toxics.Item Open Access Characteristics sources, and formation of organic aerosols in the central Rocky Mountains(Colorado State University. Libraries, 2014) Schurman, Misha Iris, author; Collett, Jeffrey L., advisor; Kreidenweis, Sonia M., committee member; Henry, Charles S., committee member; Fischer, Emily V., committee memberTo view the abstract, please see the full text of the document.Item Open Access Composition of fine particles in Carlsbad Caverns National Park and implications for sources and visibility impacts(Colorado State University. Libraries, 2022) Naimie, Lillian E., author; Collett, Jeffrey L., advisor; Benedict, Katherine B., committee member; Fischer, Emily V., committee member; Jathar, Shantanu, committee memberThe Carlsbad Caverns Air Quality Study (CarCavAQS) was designed to examine the influence of regional sources, including urban emissions, increased oil and gas development, wildfires and other biogenic sources, and soil dust on the park, including impacts on fine particle haze, ozone, and nitrogen deposition. Field measurements of aerosols, trace gases, and deposition were conducted from 25 July through 5 September 2019. Here the focus is on observations of the composition and concentration of fine particles and key trace gas precursors to understand important contributing species, their sources, and associated impacts on haze. Measurements focused on fine particulate matter (PM2.5) including mass, major ions, water soluble organic carbon (WSOC), and black carbon (BC) from various high time-resolution instruments as well as an Interagency Monitoring of Protected Visual Environments (IMPROVE) sampler. Supplemental measurements included denuder-filter pack sampling for inorganic gases (HNO3 and NH3) and a Picarro cavity ring down spectrometer for methane (CH4). High-time resolution (6-minute) PM2.5 mass ranged up to 31.8 μg m−3, with an average of 7.67 μg m−3. The main inorganic ion contributions were sulfate (avg 1.3 μg m−3), ammonium (avg 0.30 μg m−3), calcium (Ca2+) (avg 0.22 μg m−3), nitrate (avg 0.16 μg m−3), and sodium (avg 0.057 μg m−3). The WSOC average concentration was 1.2 μg C m−3. Inorganic ion concentrations had significant, sharp spikes in Ca2+, consistent with local dust generation and transport. Ion balance analysis suggests one period of acidic aerosol, the importance of ammonium and calcium in neutralizing sulfate, and significant reactions of nitric acid with sea salt and soil dust. The sums of PILS ion and WSOC concentrations, the latter multiplied by a factor of 1.8 to account for elements other than carbon, were not enough to reach mass closure with the TEOM PM2.5 mass concentrations, suggesting that insoluble species are also an important component of the aerosol at CAVE. IMPROVE sampler data, including insoluble species had good agreement between total PM2.5 mass and speciated PM2.5 aerosol mass. Sulfate is the major contributor to modeled light extinction in the 24-hour IMPROVE data set. Higher time resolution data had periods of significant light extinction from black carbon as well as sulfate, with a maximum 1-hour extinction value of 90 Mm−1. Analysis of transport patterns indicated clear enrichment of sulfate, BC, and CH4 during periods when transport came from the southeast, the direction of greatest abundance of oil and natural gas development. Air masses transported from the northeast, a region of high agricultural activity, were enriched in ammonia.Item Open Access Earth, humans, and metals: investigating the role of iron and other metals in the atmospheric, oceanic, and energy systems(Colorado State University. Libraries, 2022) Rathod, Sagar D., author; Pierce, Jeffrey R., advisor; Bond, Tami C., advisor; Denning, A. Scott, committee member; Fischer, Emily V., committee member; Scott, Ryan P., committee memberMetals such as iron and copper have been an integral component of the Earth system since its beginnings and have formed the basis for modern human civilization growth since the Bronze and Iron Ages. Human activities include metals at various levels, from burning coal in power plants and mining ores lead to emissions of particulate and gaseous metallic products into the atmosphere. While suspended in the air, metal oxides such as hematite and magnetite absorb solar radiation, thus warming the atmosphere. After falling into the oceans, metals such as iron and magnesium act as important nutrients for oceanic biota, and thus affect the marine nutrient and carbon cycles. Human activities have increased many-fold since the beginning of the Industrial Era, and as the world moves from fossil fuel to renewable energy to reduce carbon emissions, the demand for metals is also projected to increase many folds. Yet, the past, present, and future impacts of anthropogenic activities on the atmospheric and marine metal cycles, particularly iron, remain poorly understood.In Chapter 2, I estimate the atmospheric radiative and oceanic biological impacts of anthropogenic iron emissions over the Industrial Era. I perform simulations using a mineralogy-based inventory and an Earth System Model and estimate the 1850-to-2010 global mean direct radiative forcing by anthropogenic iron to be +0.02 to +0.10 W/m2. I estimate that the enhanced phytoplankton primary production due to anthropogenic soluble iron deposition over the last 150 years caused carbon dioxide (CO2) sequestration of 0.2-13 ppmv. This sequestered CO2 also led to an 'avoided' CO2 forcing of -0.002 to -0.16 W/m2. While globally small, these impacts can be higher in specific regions; the anthropogenic iron oxide direct radiative forcing is +0.5 W/m2 over areas such as East Asia and India with more coal combustion and metal smelting. Anthropogenic soluble iron sustains >10% of marine net primary productivity in the high-latitude North Pacific Ocean, a region vulnerable to thermal stratification due to climate change. In Chapter 3, I focus on evaluating anthropogenic total iron emissions using observations and models. Performing the model-observation comparison only at sites where the modeled anthropogenic contribution is the highest, I find that the current emission inventory underestimates anthropogenic total iron emissions from North America and Europe by a factor of 3-5. Further isolating anthropogenic sectoral emissions over North America using Positive Matrix Factorization, I find that smelting and coal combustion emissions are overestimated by a factor of 3-10 in the current emission inventory, whereas heavy fuel oil emissions from ships and industrial boilers are underestimated by a factor of 2-5. By comparing modeled concentrations of iron oxides with observations from Japan, I find that the current smelting and coal combustion emissions from East Asia are only slightly overestimated in the inventory, by a factor of 1.2-1.5. Finally, in Chapter 4, I explore the regionality and magnitude of PM2.5 emissions from metal mining and smelting to meet projected global renewable energy demand. I estimate future PM2.5 (particulate matter smaller than 2.5 μm diameter) emissions from mining and smelting to meet the metal demand of renewable energy technologies in two climate pathways to be 0.3-0.6 Tg/yr in the 2020-2050 period, which is projected to contribute 10-30% of total anthropogenic primary PM2.5 combustion emissions in many countries. The concentration of mineral reserves in a few regions means the impacts are also regionally concentrated. Rapid decarbonization could lead to a faster reduction of overall anthropogenic PM2.5 emissions but also could create more unevenness in the distributions of emissions relative to where demand occurs.Item Open Access Preliminary development and testing of an open-path hydrocarbon sensor for oil and gas facility monitoring(Colorado State University. Libraries, 2019) Farris, Betsy M., author; Yalin, Azer P., advisor; Fischer, Emily V., committee member; Jathar, Shantanu H., committee memberWe developed an open-path laser absorption sensor for detection of unspeciated hydrocarbons for oil and gas production facility fence line monitoring. Such sensors can aid in maintaining air quality standards by quantifying greenhouse gas emissions and detecting emissions that cause adverse health effects. Our initial design employs a single-path detection system, though future implementations may use multiple paths for large-scale facility monitoring. The sensor uses a compact mid-infrared laser source in the spectral region of ~3.3 µm to measure absorption of several hydrocarbon species and is intended for open-paths of ~100 m to 1 km. Spectral simulations show that for typical conditions the hydrocarbons cause a transmission reduction of ~10% allowing for a robust measurement. The initial prototype system uses a helium-neon (He:Ne) laser at 3.391 µm for which signal contributions from methane and non-methane hydrocarbons are comparable. Closed-cell tests were performed with diluted methane (~150-250 ppm) to validate the transmission signals and showed good agreement with expected (calculated) values to within ~10%. The system employs a reference leg, with a 2nd detector (near the source), to normalize for laser power fluctuations. For improved signal-to-noise, particularly for detection of small concentrations and transmission changes, we employ phase-sensitive detection with a mechanical chopper and software based lock-in amplifier. This detection scheme, when employed in the field, allows measurement of transmission signals with stability <0.5% (based on coefficient of variation over 60 s). The portable field sensor system uses two refractive telescopes (2" diameter optics), a transmitter and receiver co-located on a mobile optical breadboard, and a reflector dictating the pathlength. We performed initial tests with pathlengths up to ~25 m (one way), though the design should allow paths in excess of 100 m. Methane was released for initial field tests at known flow rates near the center of the beam path. Transmission signals in agreement with expectations (given uncertainties in the wind and plume dispersion) were observed. The system should allow detection of leaks (emissions) for mass flows as low as ~0.1 g/s of methane (or equivalent optical signal from other species resulting in a 1% change in signal) for the case where the source is ~150 m from the beam path and under typical atmospheric conditions. Recommendations for future modifications are provided based on potential shortcomings identified by initial field testing. Initial field testing also proved that this technology could be a viable low-cost solution for hydrocarbon detection.Item Open Access Summertime ozone production at Carlsbad Caverns National Park, New Mexico: influence of oil and natural gas development(Colorado State University. Libraries, 2023) Marsavin, Andrey, author; Collett, Jeffrey L., Jr., advisor; Fischer, Emily V., committee member; Willis, Megan D., committee memberSoutheastern New Mexico's Carlsbad Caverns National Park (CAVE) has increasingly experienced summertime ground-level ozone (O3) levels surpassing the US Environmental Protection Agency's National Ambient Air Quality Standard (NAAQS) of 70 parts per billion by volume (ppbv). The park is located in the western part of the Permian oil and natural gas (O&G) basin, where production rates have more than tripled in the last decade. We investigate O3–precursor relationships by constraining a zero-dimensional (0-D) model to an hourly nitrogen oxides (NOx = NO + NO2) and speciated volatile organic compound (VOC) data set collected at CAVE during the summer of 2019. O&G-related VOCs dominated the calculated VOC reactivity with hydroxyl radicals (OH) on days when O3 concentrations were primarily controlled by local photochemistry. Radical budget analysis showed that NOx levels were high enough to impose VOC sensitivity on O3 formation in the morning hours, while subsequent NOx loss through photochemical consumption led to NOx-sensitive conditions in the afternoon. Daily maximum O3 was sensitive to both NOx and O&G-related VOC emission reductions, with NOx reductions generally being more effective. The model could not reproduce a 5-day high O3 episode when constrained to observed NOx and primary VOCs, likely due to influence from O3 produced during air mass transport from regional O&G basins as indicated by back-trajectory analysis, low i/n-pentane ratios consistent with O&G emissions, increased concentrations of secondary VOCs, and extensive oxidation of emitted NOx. Constraining the model with observed total oxidized reactive nitrogen (NOy), which approximates NOx at the time of emission, greatly improves model-observation agreement during this episode, reaffirming NOx-sensitive conditions in photochemically aged air masses.Item Open Access Using modelling tools to advance the understanding of ammonia dry-deposition and bidirectional flux processes next to large animal feeding operations(Colorado State University. Libraries, 2020) Lassman, William, author; Pierce, Jeffrey R., advisor; Collett, Jeffrey L., Jr., advisor; Fischer, Emily V., committee member; Ham, Jay M., committee memberAmmonia in the atmosphere is a trace gas that can play a big role in the Earth's climate, as well as human and ecological health. Due to its stickiness and solubility, ammonia can enter the biosphere via wet and dry deposition, where excess ammonia input often results in soil acidification, disruption of natural ecological equilibria, and loss of biodiversity. Additionally, ammonia is the most abundant alkaline species in the atmosphere and can react with atmospheric acids to form aerosols, which can affect the earth's radiative balance as well as human health. Ammonia emissions tend to be associated with agricultural sources, such as fertilized fields or animal waste at concentrated Animal Feeding Operations (CAFOs). Consequently, ammonia emissions tend to be dynamic and highly heterogeneous, and ammonia surface-fluxes are difficult to measure. However, in regions with many large CAFOs, ammonia can be an important regional pollutant, especially if there are sensitive ecosystems or other regional sources of atmospheric acids present. In this dissertation, I study ammonia dry-deposition fluxes immediately downwind of CAFOs using a variety of modelling tools. First, I discuss original research where I use a coupled a K-epsilon model with a Lagrangian-Stochastic ammonia bidirectional exchange surface model to simulate the dispersion and deposition of ammonia downwind of an idealized CAFO. Based on these simulations, the amount of ammonia that undergoes dry deposition depends greatly on the land surface downwind of the CAFO; replacing bare soil or unmanaged grassland with leafier surfaces such as cropland or forests can increase the fraction of total ammonia emissions that deposits from 2 - 10% to 30 - 50%, though this is sensitive to the ammonia emission potential in the model plant canopy. Next, I describe a separate study where I use a 3-D Large-Eddy Simulation model to simulate the dispersion of ammonia and methane from a CAFO with a time-resolved modelling tool. I use this modelling system to produce synthetic observations, which are used to develop an inversion approach to quantify the ammonia dry deposition near a CAFO with colocated mobile measurements of ammonia and methane. While I demonstrate that such an inversion technique is feasible with surface-based measurements, considerable value is added, in terms of minimizing method bias and increasing method precision, by mounting measurements on a small Unmanned Aerial System (sUAS). Finally, I present measurements of PM2.5 concentration and composition that were made in Palapye, Botswana. Botswana is a developing country with a hot and arid climate. Beef and livestock production are important economic activities in Botswana; however, the agricultural practices differ considerably from the CAFOs discussed in the rest of the dissertation. Furthermore, these livestock activities occur against a backdrop of emissions and air pollutants that differ considerably from the United States and Europe. The measurements show that PM2.5 concentrations were on average 9 μg m-3 during the 5-week measurement period. While below levels that are typically considered hazardous, there was considerable variability in the measured concentrations, and the measurement period is too short to conclusively determine that air pollution is not a public health concern in this region. The aerosol composition is dominated by carbonaceous species, probably from biomass burning, though inorganic sulfate also is abundant in the aerosol phase. As Botswana continues to undergo economic development, the types of emissions and pollution present will continue to change.Item Open Access Volatile organic compound and methane emissions from well development operations in the Piceance Basin(Colorado State University. Libraries, 2016) Hilliard, Noel G., author; Collett, Jeffrey L., advisor; Fischer, Emily V., committee member; Ham, Jay M., committee member; Hecobian, Arsineh, committee memberThe natural gas industry in Colorado has experienced significant growth in the last decade due to widespread use of unconventional natural gas extraction technologies. Garfield County is located in the Rocky Mountain Region on the western slope of Colorado above the Piceance Basin. Natural gas wells in this region penetrate the William’s Fork formation, located approximately 4,000 ft. below the surface, which is a tight sand formation known to be rich in natural gas. Horizontal drilling increases the extraction potential of natural gas stored in several sandstone lenses. Hydraulic fracturing is a stimulation technique used to maximize the flow and efficiency of natural gas transport to the surface from unconventional reservoirs. Once the formation is adequately cracked, 10-50% of the hydraulic fluid flows back to the surface . Our field team collected samples in Garfield County between 2013-2015 to measure methane, ozone precursors, and air toxics associated with natural gas extraction activities. Very few studies have provided direct observations of VOC emissions from individual well development activities. Emission rates of 48 VOCs and methane were determined using the tracer ratio method for three well development operations: drilling, hydraulic fracturing (fracking), and flowback for a subset of samples collected. Methane had mean emission rates of 1.57, 6.78, and 25.6 g s-1 for drilling, hydraulic fracturing, and flowback operations respectively, while toluene had mean emission rates of 1.24, 0.469, and 0.437 g s-1 for these operations. Measured emission rates were used to determine if specific VOCs were well correlated with each other and/or methane emission rates. Strong correlations between individual VOC emission rates and methane were investigated to assess whether methane emission rates might serve as useful surrogates for emission rates of individual VOCs, which are less easily measured. We found that methane and ethane appear to be emitted from the same sources for all operation types indicating that methane emission rates may be useful surrogates for ethane emission rates. Methane emission rates appear not to be very useful surrogates for heavier VOCs, including C5-C10 alkanes, alkenes, and aromatics. Concentration ratios of source-specific tracer compounds were investigated to determine the source signatures of individual operation types. We found that drilling emissions appear to be primarily influenced by combustion, while flowback emissions are primarily influenced by the release of natural gas and other substances from the well.Item Open Access Volatile organic compound concentrations and the impacts of future oil and natural gas development in the Colorado Northern Front Range(Colorado State University. Libraries, 2018) Weber, Derek T., author; Collett, Jeffrey L., advisor; Fischer, Emily V., committee member; Jathar, Shantanu H., committee member; Hecobian, Arsineh, committee memberRecent advances in unconventional extraction of oil and natural gas (O&NG) have caused an increase in the number of wells in the Colorado Northern Front Range (CNFR) which has doubled Colorado's natural gas production over the last 15 years. Increased O&NG activity can lead to increased emissions of Volatile Organic Compounds (VOCs) which may negatively impact air quality and human health. This study looks at five sites (an elementary school, residential area, Fossil Creek Natural Area, Soapstone Natural Area, and a gas station) in Fort Collins and Timnath with the objectives of determining the gradient of VOC concentrations across a subsection of the CNFR, providing a baseline to compare potentially elevated VOC concentrations from future O&NG development, and a better understanding of the influence of O&NG emissions on air quality in the CNFR. Whole air samples were collected at all locations using an evacuated 6L stainless steel canister equipped with a calibrated flow controller that sampled at a constant flow rate for approximately 1 week. Sampling began at the elementary school and gas station in the summer of 2015 and concluded in November of 2016. Sampling at the two natural areas and the residential area took place in the fall of 2015. VOC concentrations were analyzed using an online gas chromatography flame ionization detector (GC-FID) system. An in-situ real-time GC was also deployed along with an All-In-One (AIO) weather station at the residential area providing hourly VOC and meteorological measurements for approximately 3 weeks in the fall of 2015. A suite of 48 VOCs were measured in this study. Ambient concentrations of BTEX compounds (Benzene, Toluene, Ethylbenzene, and Xylenes) are often of particular interest due to their carcinogenic effects and toxicity; therefore, they were studied in-depth as part of this thesis. Benzene was found to have median ambient concentration at the elementary school, residential area, Fossil Creek Natural Area, Soapstone Natural Area, and the gas station of 0.18, 0.14, 0.32, 0.09, and 0.55ppbv, respectively. Through the use of VOC correlations with propane and acetylene and VOC ratios, it was determined that O&NG emissions have a large influence on ambient VOC concentrations in the CNFR. The mean ratio of i-pentane to n-pentane found at the elementary school, residential area, Fossil Creek Natural Area, Soapstone Natural Area, and the gas station was 1.07, 1.17, 1.16, 1.05, and 2.35, respectively. This indicates that the elementary school and Soapstone Natural Area are strongly influence by O&NG emissions while the residential area and Fossil Creek Natural Area have a mixed influence from O&NG activity as well as vehicular emissions. In contrast the gas station, displayed a clear combustion signature, as expected. Additional VOC ratios were utilized; however, i-pentane to n-pentane ratio was determined to be the most robust tool to assess source apportionment in the CNFR. In addition, through the use of meteorological data coupled with the real-time GC VOC measurements, there is strong evidence that local O&NG sources can have a large impact on air quality at the residential area. The OH reactivity at each location was evaluated in order to compare the ozone production potential by the VOCs measured at each site. Fossil Creek NA showed the largest total OH reactivity in the fall while Soapstone NA displayed the lowest. At Soapstone NA, 66.7% of the total OH reactivity resulted from aromatics, which is the highest, and 11.4% resulted from alkenes, which is the lowest compared to each group's contribution at other sites. At the elementary school, 3.2% of the OH reactivity in the summer was attributed to isoprene, whereas in the fall, winter, and spring only 2.0%, 0.41%, and 0.76% of the OH reactivity resulted from isoprene, respectively. Development of new unconventional O&NG wells is ongoing in the CNFR and there are plans to develop wells in close proximity to the elementary school. The American Meteorological Society (AMS)/Environmental Protection Agency (EPA) steady-state dispersion model AERMOD was utilized to project the potential increased concentration of benzene as a result of this development. The model was run utilizing the 5th, 25th, median, 75th, and 95th percentile emission rates of benzene found by a past study at production sites in the CNFR. The annual average concentration increases above background at the school (0.18 ± 0.08ppbv) for the 5th, 25th, median, 75th, and 95th percentile emission rates were found to be 0.0067, 0.11, 0.33, 0.89, and 6.7ppbv, respectively. The strongest benzene enhancement at the school occurred 0:00 (midnight) - 08:00 and 17:00 - 23:00 (0.46ppbv); however, during school attendance hours (08:35 - 15:13) the concentration increase was 0.024ppbv.