The transport and emplacement mechanisms of the highly energetic pyroclastic density current (PDC) generated in the blast style eruption of Soufriere Hills Volcano, Montserrat, on 26 December 1997 are examined through detailed lithological mapping and sedimentological analysis of the deposits. The PDC formed deposits which range in grain size from coarse breccias to fine ash, with distinctive bipartite layering and well-developed grading and stratification. On a large scale the PDC was highly erosive, sculpting large bedforms and depositing relatively thin deposits. However, locally, centimetre scale topographic protuberances were responsible for significant variations in deposit thickness, grain size, and the development of dune bedforms. The strong lateral and vertical lithofacies variations are attributed to well-developed density stratification, which formed during explosive expansion of the dome prior to PDC formation. Experimental modelling of stratified inertial gravity currents was carried out to investigate the effects of density stratification prior to release of the current. The degree of stratification governs the rate of mixing in the current, which in turn influences the velocity. Well·stratified currents initially move faster than homogenous currents but are slower in the latter stages of current propagation. The results have important implications for deposition from particle-laden flows, which may become stratified with coarser material concentrated at the base of the current. The role of PDCs jn the formation of unit US2-B, emplaced during the Upper Scoriae 2 eruption (79± 8 ka) on Santorini, Greece, was investigated through sedimentological analysis and mapping. Proximally, the unit exhibits features characteristic of emplacement from a flow, such as thickening into palaeochannels and erosive basal contacts. Distally, the unit is of uniform thickness and grain size parameters suggest the deposit is more characteristic of exnplacement from a fallout mechanism. Discrete lenses of fine-grained material within US2-B, and a gradational upper contact with PDC deposits suggest that there may have been contemporaneous deposition resulting the development of a hybrid deposit.

'Natural' disasters have traditionally been viewed as the result of an extreme physical environment. A radical backlash against this dominant view, in the nineteen seventies and eighties, moved the debate to the opposite extreme and in doing so replaced physical with social determinism. Vulnerability analysis is proposed as a methodology that bridges these extremes. It takes into account individual decision making, social milieu and physical hazard when describing human habitation in areas of volcanic activity. It is argued that vulnerability should be defined in terms of universal human needs in order to avoid it simply being a measure of the chance of death and injury or losing its meaning in the uncertainty of cultural relativism. Once vulnerability is identified it is important to explore why it has come to exist. A contextual theory of vulnerability change is presented. Vulnerability to volcanic activity was explored in the area around Mt. Etna in Sicily (Italy) and Furnas volcano San Miguel in the Azores (Portugal) using a case study methodology. This included: collecting data through interviews (semistructured and structured) and field surveying, utilising census and other secondary data sources, and examining historical documents and texts. The volcanic hazard on Mt. Etna is related to regular (4-7 years) effusive lava flows which threaten property and land rather than people. Living in a European state, it is likely that a victim of Mt. Etna will have their basic needs provided for in the long-term and therefore they are not vulnerable. In contrast the irregular explosive eruptions of Furnas, last eruption 1630, not only damage property and land but also endanger lives. The limited ability of individuals to protect themselves in the event of an eruption and organisations to aid them in this means that, in spite of state insurance, many around Furnas are vulnerable. The production of vulnerability around Etna and Furnas is strongly related to the socio-economic nature of the region and wider European and global contexts. Opportunities and constraints that exist across socio-physical space encourage behaviour and forms of life which, in tum, produce various levels of vulnerability. Individuals seem to cognitively diminish their perceptions of this threat within a context of social representations of low risk. They, and society as a whole, rarely seem to engage directly with the risk itself.

<p>Validerat; 20130508 (global_studentproject_submitter)</p>

Bandura (1986) posited that self-efficacy beliefs help determine what individuals do with the knowledge and skills they have and are critical determinants of how well skill and knowledge are acquired. Research has correlated self-efficacy beliefs with academic success and subject interest (Pajares, Britner, & Valiante, 2000). Similar studies report a decreasing interest by students in school science beginning in middle school claiming that they don’t enjoy science because the classes are boring and irrelevant to their lives (Basu & Barton, 2007). The hypothesis put forth by researchers is that students need to observe models of how science is done, the nature of science (NOS), so that they connect with the human enterprise of science and thereby raise their self-efficacy (Britner, 2008). This study examined NOS understandings and science self-efficacy of students enrolled in a sixth grade earth science class taught with explicit NOS instruction. The research questions that guided this study were (a) how do students’ self-efficacy beliefs change as compared with changes in their nature of science understandings?; and (b) how do changes in students’ science self-efficacy beliefs vary with gender and ethnicity segregation? A mixed method design was employed following an embedded experimental model (Creswell & Plano Clark, 2007). As the treatment, five NOS aspects were first taught by the teachers using nonintegrated activities followed by integrated instructional approach (Khishfe, 2008). Students’ views of NOS using the Views on Nature of Science (VNOS) (Lederman, Abd-El-Khalick, & Schwartz, 2002) along with their self-efficacy beliefs using three Likert-type science self-efficacy scales (Britner, 2002) were gathered. Changes in NOS understandings were determined by categorizing student responses and then comparing pre- and post-instructional understandings. To determine changes in participants’ self-efficacy beliefs as measured by the three subscales, a multivariate analysis of covariance (MANCOVA) was conducted. Findings indicated that explicit NOS instruction was effective for all students except minority (Black, Hispanic, Asian, or multiracial) male students in improving NOS understandings. Furthermore, all students that received NOS instruction demonstrated decreased anxiety towards science. Future research should focus on long-term investigations of changes in anxiety and value of research constructs with regards to NOS instruction.

This thesis describes a passive-source seismic mapping project in the Nechako Basin of central British Columbia (BC), Canada, with the ultimate goal of assessing the hydrocarbon and mineral potential of the region. The Nechako Basin has been the focus of limited hydrocarbon exploration since the 1930s. Twelve exploratory wells were drilled; oil stains on drill chip samples and the evidence of gas in drill stem tests attest to some hydrocarbon potential. Seismic data collected in the 1980s were of variable quality due mainly to effects of volcanic cover in this region. For the present study, an array of nine seismic stations was deployed in 2006 and 2007 to sample a wide area of the Nechako Basin and map the sediment thickness, crustal thickness, and overall geometry of the basin. This study utilizes recordings of about 40 distant earthquakes from 2006 to 2008 to calculate receiver functions, and construct S-wave velocity models for each station using the Neighbourhood Algorithm inversion. The surface sediments are found to range in thickness from about 0.8 to 2.7 km, and the volcanic layer below ranges in thickness from 2.3 to 4.7 km. Both sediments and volcanic cover are thickest in the central part of the basin. The average crustal thickness across the basin is about 30-32 km; it is thicker in the northern and western parts of the basin, and thinner in the southern and eastern parts. This study complements other research in this region, such as independent active-source seismic studies and magnetotelluric measurements, by providing site-specific images of the crustal structure down to the Moho and detailed constraints on the S-wave velocity structure.

The south-west portion of the main caldera was mapped and a stratigraphy for the caldera-fill was constructed. The exact timing of formation of the main caldera is unclear; However, caldera collapse either predates or was synchronous with the eruption of the Campagnola Tuff. The proximal facies of the Campagnola Tuff exists as a complex relation of ignimbrite, lithic breccia and pyroclastic surge deposits. Overlying this the Galluccio Tuff a compound ignimbrite, ~6 km3 D.R.E, forms the base of the exposed caldera fill. Caldera lakes then became well established and following activity was predominantly phreatomagmatic. Pyroclastic surge deposits possess sand wave structures of several types and their migration direction was apparently controlled by the velocity/flow regime of the surge rather than the moisutre content. The morphology of juvenile clasts from phreatomagmatic deposits indicates that the eruptions were driven by a combination of vesiculation and magma/water interaction. The uppermost pyroclastic deposits are thought to represent the early phase of dome building where water still had access to the vent. The construction of the lava domes brought activity to a close within the main caldera. The Galluccio Tuff on the flanks of the volcano may be divided into three compositionally distinct eruptive units. The Lower Galluccio Tuff, correlated with the bulk of the Galluccio Tuff filling the main caldera. The Middle Galluccio Tuff commenced with the eruption of pumice-rich pyroclastic flows followed by flows enriched in both the size and amount of lithic fragments forming lithic-rich ignimbrite and co-ignimbrite lithic breccias of which several types exist. The Upper Galluccio Tuff is composed of lithic-rich ignimbrite which possess dense pumice fragments and are thought to be the product of a combination of both vesiculation and magma water interaction. Field relations indicate that pyroclastic flows were sometimes generated in quick succession and may have overrun earlier slower moving flows. Occasionally internal shear may have caused the overriding of portions of the same flow, these often coincide with lithic breccias and represent the climax of the eruptive phases. The grading of lithic fragments indicates that the expansion and fluidization decreased and yield strength increased with time in a pyroclastic flow.

This thesis focuses on the Manyi Fault in Northern Tibet which experienced a M_w 7.6 earthquake in 1997. The remoteness of the area limits the feasibility of measurements in the field, however the climate makes it ideal to study by remote sensing, specifically Interferometric Synthetic Aperture Radar (InSAR). The mechanics of the earthquake have been well documented however there are still numerous questions about the other stages of the earthquake cycle (postseismic and interseismic) across the fault. Previous studies of the postseismic motion across the Manyi Fault using four years of ERS SAR data show the deformation can be explained by either viscoelastic relaxation of a Standard Linear Solid body with a viscosity of 4x10¹⁸ Pa s or afterslip. We use the ERS timeseries and ratemaps formed from a network of ENVISAT SAR scenes from 2003-2010 to analyse the postseismic deformation. We create a series of afterslip models based on rate-and-state frictional laws, along with series of viscoelastic models with various rheologies (Maxwell and Burgers). Our results show that an afterslip model fits the data slightly better than a Burgers rheology but not within resolvable errors. A range of afterslip models fit the data well, with frictional parameters ranging from 8x10^-4 to 2x10^-3 and a preseismic slip rate of 8 to 20 mm/yr. The best-fitting Burgers rheology has a Kelvin element viscosity of 4x10¹⁸ Pa s and Maxwell element viscosity of 6x10¹⁹ Pa s. We analyse the interseismic InSAR signal observed before the 1997 earthquake using ERS data from 1992-1997 to find that the Manyi Fault was accumulating strain at 3+/-2 mm/yr. We also find the seismic locking depth was 22+/-15 km which correlates with the maximum depth of slip during the earthquake. We show there is no significant deformation across the fault to the north of the Manyi Fault which may be an extension to the Kunlun fault. We discuss an analytical 2D thin viscous channel model from literature that has been shown to match the data in this thesis. We show that, once errors are properly accounted for, their model cannot explain both the post and preseismic datasets.

The use of machine learning techniques in tackling hydrological problems has significantly increased over the last decade. Machine learning tools can provide alternatives or surrogates to complex and comprehensive methodologies such as physics-based numerical models. Machine learning algorithms have been used in hydrology for estimating streamflow, runoff, water table fluctuations and calculating the impacts of climate change on nutrient loading among many other applications. In recent years we have also seen arguments for and advances in combining physics-based models and machine learning algorithms for mutual benefit. This thesis contributes to these advances by addressing two different groundwater problems by developing a machine learning approach and comparing this previously developed physics-based models: i) estimating groundwater and surface water depletion caused by groundwater pumping using artificial neural networks and ii) estimating a global steady-state map of water table depth using random forests. The first chapter of this thesis outlines the purpose of this thesis and how this thesis is a contribution to the overall scientific knowledge on the topic. The results of this research contribute to three of the twenty-three major unsolved problems in hydrology, as has been summarized by a collective of hundreds of hydrologists. In the second chapter, we tested the potential of artificial neural networks (ANNs), a deeplearning tool, as an alternative method for estimating source water of groundwater abstraction compared to conventional methods (analytical solutions and numerical models). Surrogate ANN models of three previously calibrated numerical groundwater models were developed using hydrologically meaningful input parameters (e.g., well-stream distance and hydraulic diffusivity) selected by predictor parameter optimization, combining hydrological expertise and statistical methodologies (ANCOVA). The output parameters were three transient sources of groundwater abstraction (shallow and deep storage release, and local surface-water depletion). We found that the optimized ANNs have a predictive skill of up to 0.84 (R2, 2σ = ± 0.03) when predicting water sources compared to physics-based numerical (MODFLOW) models. Optimal ANN skill was obtained when using between five and seven predictor parameters, with hydraulic diffusivity and mean aquifer thickness being the most important predictor parameters. Even though initial results are promising and computationally frugal, we found that the deep learning models were not yet sufficient or outperforming numerical model simulations. The third chapter used random forests in mapping steady-state water table depth on a global scale (0.1°-spatial resolution) and to integrate the results to improve our understanding on scale and perceptual modeling of global water table depth. In this study we used a spatially biased ~1.5-million-point database of water table depth observations with a variety of iv globally distributed above- and below-ground predictor variables with causal relationships to steady-state water table depth. We mapped water table depth globally as well as at regional to continental scales to interrogate performance, feature importance and hydrologic process across scales and regions with varying hydrogeological landscapes and climates. The global water table depth map has a correlation (cross validation error) of R2 = 0.72 while our highest continental correlation map (Australia) has a correlation of R2 = 0.86. The results of this study surprisingly show that above-ground variables such as surface elevation, slope, drainage density and precipitation are among the most important predictor parameters while subsurface parameters such as permeability and porosity are notably less important. This is contrary to conventional thought among hydrogeologists, who would assume that subsurface parameters are very important. Machine learning results overall underestimate water table depth similar to existing global physics-based groundwater models which also have comparable differences between existing physics-based groundwater models themselves. The feature importance derived from our random forest models was used to develop alternative perceptual models that highlight different water table depth controls between areas with low relief and high relief. Finally, we considered the representativeness of the prediction domain and the predictor database and found that 90% of the prediction domain has a dissimilarity index lower than 0.75. We conclude that we see good extrapolation potential for our random forest models to regions with unknown water table depth, except for some high elevation regions. Finally in chapter four, the most important findings of chapters two and three are considered as contributions to the unresolved questions in hydrology. Overall, this thesis has contributed to advancing hydrological sciences through: i) mapping of global steady-state water table depth using machine learning; ii) advancing hybrid modeling by using synthetic data derived from physics-based models to train an artificial neural network for estimating storage depletion; and (iii) it contributing to answering three unsolved problems in hydrology involving themes of parameter scaling across temporal and spatial scales, extracting hydrological insight from data, the use of innovative modeling techniques to estimate hydrological fluxes/states and extrapolation of models to no-data regions. / Graduate

Stable isotopes of oxygen (O18) and hydrogen (H2) are used as proxies of past climates when ice preserved in the polar icecaps. The main assumption when doing paleoclimatic reconstructions using these isotopes is that the isotopic composition of the snow that turned into ice reflect the present temperature of the planet during deposition. In this study, I investigate to what extent the composition of O18 and H2 varies as a function of precipitation. I tested the following hypotheses: i) δO18 and δH2 decreases due to fractionation during orographic fallout, and ii) the electric conductivity in the snow decreases with distance from sea due to fading marine inputs. The hypotheses were tested with field samples gathered along a precipitation gradient from the Norwegian coast to Kiruna, Northern Sweden. To support my first hypothesis, I found a decrease from -9.32% δO18 (-54.85% δH2) to -25.48% δO18 (-109.54 % δH2) due to orographic deposition over the Scandes with a drastic drop in Abisko caused by a rain shadow. Similarly, I found a considerable variation and decreasing trend of conductivity along the precipitation gradient. My study raises awareness that δO18 in snow is in fact dependent on precipitation regimes rather than temperature regimes in this area, which should be considered when interpreting environmental records of δO18 in a paleoclimatic context. Lastly, I suggest further studies targeting the effects of post-depositional processes (melting, sublimation, and anthropogenic influence) on the isotopic composition and conductivity, especially with a changing climate.

The purpose of this action research project was to determine how my practice of using graphic organizers during instruction influenced my students' ability to summarize and comprehend significant fifth grade Earth Science content regarding the Earth's changing surface. A secondary purpose was to determine the students' perceptions of how concept mapping assisted in making connections to understand the fifth grade Earth Science content regarding the Earth's changing surface. The three processes used to collect data for this research were concept maps, focus groups and the pre- and post-test results. The themes that emerged were the ability to describe, categorize and classify details, the increased accuracy of the use of vocabulary and the memory of the concepts that students' ability to recall information and understand the Earth Science concepts as evidenced through summarization and comprehension through the pre- and post-test.