Even in Arcadia: Stories is a short story collection that follows adolescent and young adult women as they navigate growing up and growing out of the spaces they inhabit. Set in the American southwest, specifically unnamed suburban and rural cities of Arizona, the collection challenges the culturally popular narratives that surround the West-- the idealized cowboy, rugged individualism, and conquest of nature. Drawing on long-standing myths, serialized TV shows, and classic literature, the collection asks the reader to evaluate the stories they consume and are willing to take as truth. The stories range from realistic to speculative, employing horror and surrealistic elements as they descend into a sort of hellscape that draws on natural elements of desert landscapes juxtaposed against urban spaces. The collection focuses on gender, adolescence, and trauma set in the aftermath of the 2008 recession and the decline of small-town America from the perspective of youth.

Wildfire is a natural but often hazardous part of the Alaskan ecosystems. Physically based wildfire models range from simple relationships used for rapid, in-situ fire behavior analysis to complex weather models used for prediction over several days and weeks. Physical models in Alaska, however, often struggle to integrate weather forecast information to make predictions beyond just a few days. The random forest model explored here is able to leverage an array of variables to identify days of enhanced and reduced satellite fire detections. Peaks and lulls in activity are accurately identified, though exact magnitudes are often incorrect, especially when wildfire suppression efforts occurred. This study emphasizes the use of reanalysis weather variables in addition to antecedent fire activity, highlighting the usefulness of variables like vapor pressure deficit for use in quantitative prediction. By applying weather forecast data, the model generated simulated wildfire forecasts. These forecasts show some success at identifying peaks and lulls in fire activity. Effective lead time varied widely ranging between 1 and 10 days, mostly dependent on the weather model performance. By providing specific timing and using real ensemble forecasts for medium term prediction, a model likes this fills a potential open niche in fire predictive services. Machine learning may be especially useful for its relative efficiency and ease of automation.

Under the optimal oviposition theory, insects are expected to lay eggs on hosts that maximize the success of their offspring. Tree height is known to be an important factor influencing the distribution of phytophagous insects because some species perform better at a distinct range of heights. This difference in performance could lead to incorrect estimates of population parameters if surveys are only conducted on one host plant height. Aspen leaf miners (Phyllocnistis populiella) have undergone a major outbreak in interior Alaska over the last two decades. We quantified patterns of aspen leaf miner oviposition and juvenile survival over 2 years and found that aspen leaf miners were approximately 1.5 times more likely to survive on tall trees than short trees. Parasitism and both egg and larval predation were lower on tall trees. Aspen leaf miners on tall trees also had larger pupal masses than those on short trees. Although aspen leaf miners performed better on tall trees, the number of eggs laid per leaf did not significantly differ by tree height. There were no significant differences in leaf foliar nitrogen between tall and short trees. We also found little differences in wind speed between tall and short trees that could explain ovipositional patterns. Ovipositional patterns may partially reflect the difference in phenology between tall and short aspen trees. Aspen leaf miners only lay eggs on new leaves. Tall aspen trees leafed out 7 days earlier on average than short aspen trees, and tall trees, unlike short trees, ceased to produce new leaves after budburst. Consequently, there was little overlap in the availability of tall and short aspen trees for oviposition, so even if aspen leaf miners have a preference for laying more eggs on tall than short trees, they can only act on it during the short time period when tall trees are available for oviposition. The results suggest that population projections based on data collected from only short trees may underestimate future aspen leaf miner population growth due to lower juvenile survival rates and pupal masses on short trees. More broadly, the results highlight the importance of examining multiple tree heights when studying the performance and population dynamics of phytophagous insects. They also suggest that phenological differences between plants may constrain insects from using higher quality hosts.

Increasing Arctic shipping requires study of the increasing aerosol emissions impact on aerosol optical depth (AOD) in the Bering Sea and the Bering Strait. This study used Moderate Resolution Imaging Spectroradiometer (MODIS) 550 nm AOD 3 km products to study the seasonal variability over the length of the Arctic shipping season, from June to October, over 2011 and 2014. Bucket resampling was used to project and downscale the MODIS AOD to a 0:25 by 0:25 grid during overpasses. An overpass is dened as consecutive MODIS granules from both the Terra and Aqua satellites. Ship positional data obtained from automatic identication systems (AIS) was aggregated to hourly data. Collocation of ships and AOD from overpasses were determined for all quarter degree grid cells and times. Area-weighted means for both grid-cells with ship occurrence and without ships were determined for each month. AOD decreases with increasing time in the shipping season due to the increasing frequency of low pressure system and hence aerosol removal via washout and scavenging. Comparison of the AOD of 2011 and 2014 revealed that the position of the Aleutian Low not only strongly aects the sample size, but also AOD over the Bering Sea. The sea-ice area seemed to be without notable impact on the number of ships and AOD. A weak positive correlation was found between AOD and the number of ships present in the same grid cell during a overpass for 7 out of the 10 months. A strong skewing towards October occurred in 2011 due to a strong positive correlation of the number of data points.

The purpose of the research was to examine the heat transfer and fluid dynamic performance of various nanofluids in heating and cooling applications using empirical and computational methods. Two experiments were performed to characterize and compare the performance of a Al₂O₃/60% ethylene glycol (60% EG) nanofluid to that of its base fluid. In the first experiment, the nanofluid was comprised of Al₂O₃ nanoparticles with 1% volumetric concentration in a 60% ethylene glycol/40% water (60% EG by mass) solution to that of 60%EG in a liquid to air heat exchanger. The test bed used in the experiment was built to simulate a small air handling system typical of that used in heating, ventilating and air conditioning (HVAC) applications. Previously established empirical correlations for thermophysical properties of fluids were used to determine the values of various parameters (e.g. Nusselt number, Reynolds number, and Prandtl number). The testing shows that the 1% Al₂O₃ nanofluid generates a marginally higher heat rate than the 60% EG under certain conditions. At Re=3,000, the nanofluid produced a heat rate that was 2% higher than that of the 60% EG. The empirically determined Nusselt number associated with the convection inside the coil tubing follows the behavior predicted by the Dittus-Boelter correlation quite well (R²=0.97), while the empirically determined Nusselt number for the 60% EG follows the Petukhov correlation similarly well (R²=0.97). Pressure loss and hydraulic power for the nanofluid were higher than for the base fluid over the range of conditions tested. The exergy destroyed in the heat exchange and fluid flow processes were between 8 and 13% higher for the nanofluid over the tested range of Reynolds numbers. The objective of the second study was to experimentally characterize and compare the performance of a nanofluid comprised of Al₂O₃ nanoparticles with 1, 2 and 3% volumetric concentrations in a 60% EG solution to that of 60% EG in a liquid to air heat exchanger. In this experiment, the heating system was operated in a higher temperature regime than in the first experiment. As in the first experiment, the test bed used in the experiment simulated a small air handling system typical of that used in HVAC applications. Entering conditions for the air and liquid were selected to emulate typical operating conditions of commercial air handling systems in sub arctic regions (such as Alaska). In the experiment the nanofluids generally did not perform as well as expected based on previous analytical work. The performance of the 1% nanofluid was generally equal to that of the base fluid considering identical entering conditions. However, the 2% and 3% nanofluids performance was considerably worse than that of the base fluid. The higher concentration nanofluids exhibited heat rates up to 14.6% lower than that of the 60%EG, and up to 44.3% lower heat transfer coefficient. The 1% Al₂O₃/60% EG exhibited 100% higher pressure drop across the coil than the base fluid considering equal heat output. In the computational portion of the research, the performance of a microchannel heat sink (MCHS), similar to those used to cool microprocessors filled with various nanofluids and the corresponding base fluid without nanoparticles are examined. The MCHS is modeled using a three- dimensional conjugate heat transfer and fluid dynamic finite-volume model over a range of conditions. The model incorporates a fixed heat flux of 1,000,000 W/m² at the base of the solid domain. The thermophysical properties of the fluids are based on empirically obtained correlations, and vary with temperature. Nanofluids considered include 60% Ethylene Glycol/40% Water solutions with CuO, SiO₂, and Al₂O₃ nanoparticles dispersed in volumetric concentrations ranging from 1 to 3%. The flow conditions analyzed are in the laminar range (50£Re£300), and consider multiple inlet temperatures. The analyses predict that when compared on an equal Reynolds number basis, the 60%EG/3% CuO nanofluid exhibits the highest heat transfer coefficient, and the largest reduction in average base temperature. At an inlet Reynolds number of 300, and an inlet temperature of 308K the nanofluid is predicted to have an average heat transfer coefficient that is 30% higher than that of the base fluid, while the average temperature on the base of the heat exchanger is 1K lower than that of the base fluid. In contrast, the inlet pressure required for these entering conditions is 192% higher than that for the base fluid, while the required hydraulic power to drive the flow is 366% higher than that of the base fluid. The enhanced heat transfer performance potential of nanofluids comes at the expense of generally higher pumping power consumption.

Small-scale low-temperature geothermal-based electricity generation systems are under development for use as grid supporting and grid forming power sources in remote locations. Conversion of low-temperature heat to electrical energy in an organic Rankine cycle using working fluids such as refrigerants is challenging due to the low energy conversion efficiency of the process and the significantly slower thermal response rate in comparison to the time response of the electrical grid for changes in electrical generation and load. There is a need to investigate techniques for controlling the flow of the working fluid in combination with the use of a secondary heat exchanger to improve ramping response of these systems. This research project develops and models a working fluid control technique that incorporates power electronic technologies that could help to improve the ramping response of geothermal-based organic Rankine cycle generation systems. The performance of the model is examined with the aid of simulations in MATLAB® Simulink®. The results from these simulations are used to develop a functional and reliable control technique for ramping response improvement in geothermal-based electricity generation systems using organic Rankine cycles.

Ice roads and bridges are necessary routes to transport heavy equipment, supplies and food in the winter months to and from isolated cold region communities off the road system. Ice roads allow for community members to avoid the high costs of air shipments and obtain equipment and vehicles that would otherwise not be available. These ice roads traverse frozen bodies of water (e.g., rivers, estuaries, and lakes), and require extreme safety when driving over. To achieve this, calculations are frequently completed to determine the maximum acceptable loading on the ice cover. River ice tends to have increased safety concerns and uncertainty for travel that stem from warmer air temperatures and other factors such as precipitation, snow drifting, and ice cover forming differently each year. The necessity of obtaining time intensive ice thickness measurements by hand puts the responsible personnel at considerable risk of injury or fatality. Ground penetrating radar (GPR), which has gained much popularity in the last few decades, is a quicker and more effective non-invasive method for measuring ice thickness and other properties. The GPR system was tested for its accuracy in measuring ice thickness on common transportation routes on the Yukon River and the Tanana River. Identification of varying ice type layers in river ice cover using GPR was also attempted. While layers could not be identified using the 450 MHz and 750 MHz central frequency antennas, an accuracy analysis of GPR ice thickness measurements under various environmental conditions was completed. This analysis contributes to a comprehensive understanding of the safety of ice roads for community members in remote northern villages and provides the basis for further research on identifying layers in river ice cover.

Pesticides used at mid latitudes can accumulate in Arctic environments. Two commonly detected pesticides in Arctic lakes are chlorothalonil and chlorpyrifos. In surface waters, photolysis can play an important role in the attenuation of contaminants. The chemical characteristics of dissolved organic matter (DOM) can further alter the extent of photolytic degradation of pollutants. To determine the relative effect of natural Arctic lake water and its DOM on the photolysis of chlorpyrifos, experiments were conducted under natural Arctic irradiation and under artificial irradiation. Similarly, the effect of Arctic DOM was investigated for chlorothalonil under artificial irradiation. The fulvic acid (FA) fraction of DOM was isolated from Fog 1 and from Toolik Lake in May and July. Lake waters significantly enhanced the photodegradation of chlorpyrifos under natural light by up to an order of magnitude. FA's significantly increased the degradation of chlorpyrifos (>2x) and chlorothalonil (>100x) under artificial irradiation relative to 18 MΩ-cm Water. Toolik Lake FA isolated in May, significantly enhanced the photolysis of both contaminants relative to the isolate collected in July. In the presence of iron, a lower ratio of carbohydrates and peptides to aromatics in the FA's was associated with faster degradation for chlorothalonil.

Seven hydro-meteorological stations in the National Petroleum Reserve Alaska were analyzed to look at precipitation, discharge, and temperature trends. These hydro-meteorological stations included: Fish Creek, Judy Creek, Ikpikpuk, Ublutuoch, Seabee, Prince, and Otuk. A linear regression was performed on a year-by-year basis to fill in data gaps in the temperature time series, for all six stations with temperature data. The Seasonal Mann-Kendall test and a Modified Seasonal Mann-Kendall test were performed to determine if a trend appeared present in the time series and, if so, how significant the trend was. The Sen's slope analysis was then utilized to determine the magnitude of the trend, if a trend was observed in both analyses. The temperature trends showed an increasing trend in the temperature data for four stations: Judy Creek, Ublutuoch, Fish Creek, and Ikpikpuk. No trend was shown in the remaining station, Otuk station. One station, Prince, was removed from analysis due to a high percentage of missing data. The Modified Seasonal Mann-Kendall tests showed a trend in four of the five stations, and a slight positive trend in one of the five stations. The precipitation data showed 'no trend' in the Seasonal Mann-Kendall analysis. The Modified Seasonal Mann-Kendall test showed a slight trend (Fish Creek), a moderate trend (Otuk), and no trend (Ikpikpuk) for the precipitation data. Using the seasonal Mann-Kendall analysis the discharge data showed no trend in five out of seven stations and two trends (Fish Creek and Seabee). The Modified Seasonal-Mann-Kendall analysis showed and a significant trend twice (Fish Creek and Seabee), a moderate trend three times (Ikpikpuk, Prince, Otuk), a slight trend once (Ublutuoch), and no trend one time (Judy Creek) in the discharge data.

This thesis is a collection of poetry that explores the relationships between time, memory, identity, freedom, and meaning within the context of a shifting and unsteady world. The collection is organized along a trajectory not unlike that of Dante's Divine Comedy, moving through inner and outer landscapes of uncertainty and anxiety before emerging in more ambiguous, subtle spaces. Allusion is used to suggest the continuity and fragmentariness of a tradition (similar to T. S. Eliot's "The Waste Land"), reformation and response to a tradition (as in Nets by Jen Bevin), and the chancy contingencies of personal experience. Formally, the poems use a wide variety of strategies and forms including erasure and non-traditional lineation to suggest various states of being ranging from the hypnagogic and nightmarish to the nostalgic and wistful, and ultimately to something like hope. The poems also range from highly confessional modes to more abstract, imagistic modes to similarly suggest motion and change.