42337 research outputs found
Sort by
Seismic and hydroacoustic monitoring of bedload transport in an alluvial river
No full textMonitoring the mobilisation and transport of bedload in rivers is key to understanding
landscape evolution and sediment transfer, as well as providing valuable information
for problems in the fields of ecology, river and landuse management, and
civil engineering. When rivers transport bedload their erosive capacity increases
which not only affects infrastructure such as rivers and dams, but it also influences
channel change by eroding or aggrading bars and banks altering the channel capacity
and subsequently flood risk. Bedload transport has traditionally been measured
through the use of geomorphic measurements, such as manual sediment sampling,
geomorphic mapping, and empirical equations. However, advances in measurement
methods and technologies such as drone surveys, numerical modelling, and acoustic
monitoring have been at the forefront of bedload transport studies over the last
couple of decades. These methods extend the spatial and temporal resolution of
monitoring efforts, allowing better understanding of the characteristics of sediment
mobilisation and the effects of bedload transport over longer timescales. In particular,
seismic monitoring has emerged as a valuable tool for monitoring river processes
such as propagating flood waves and the movement of bedload. This provides an
opportunity to indirectly monitor river processes over greater spatial and temporal
scales to previous methods. A significant challenge remains in independently
characterising the seismic signature of bedload transport from other sources such as
turbulence. Additionally, there remains some uncertainty in the interpretation of
seismic bedload transport signals in complex alluvial channels. This thesis examines
seismic signals recorded adjacent to an alluvial mountain river in Scotland (the River
Feshie) and presents a unique dataset combining three-component seismic data with
complementary hydroacoustic measurements, to analyse bedload transport during
high river flows.
In the first part of this thesis, I examine the characteristics of the recorded river-induced
seismic signals during high river discharge events. Using data from eight
seismic sensors located along the study reach along with local stream gauge data,
I compare the site and event variabilities and find that there is greater variability
in the signals recorded at different sites than there are over different hydrological
events. Additionally, I observe that the sensor locations have a significant effect on
the signals recorded with those on flatter ground and/or closer to the river, located
in open areas, recording clearer and stronger seismic signals with more distinct
bedload transport signals. This can inform on siting of seismic sensors in future
seismic studies.
Through the analysis of seismic signals recorded over high flow events combined
with independent interpretation of the occurrence of bedload mobilisation through
hydroacoustic measurements, I then assess the consistency of bedload transport
thresholds in response to changing flow conditions. Results indicate that bedload
entrainment thresholds vary from event to event and may show a link with event
duration and magnitude, as well as being influenced by the occurrence of successive
events due to armouring effects. This analysis also revealed that the relationship
between bedload-induced seismic power and water level varied between events with
events exceeding ∼ 1.40 m water level exhibiting distinct variations in this relationship
between the rising and falling limb of the hydrograph with increased seismic
power prolonged on the falling limb. Previous studies have used these hysteretic
relationships as an indicator of bedload transport; however, my work reveals that
bedload transport can occur without these relationships present in the data.
In the third part of this thesis, I explore the directional-component dominance of
the seismic signals recorded adjacent to the river to interpret whether river-induced
seismic signals, and specifically bedload signals, exhibit stronger signals in a single
direction. Analysis of the three orthogonal components of the recorded seismic signals
revealed that river-induced signals are strongest in the two horizontal directions,
consistent with Love-wave propagation. This suggests that existing fluvial seismic
models and studies may be incorrectly using the Rayleigh wave interpretation under
the assumption that the dominant component of bedload-induced seismic signals is
in the vertical direction. Additionally, the signals from the movement of bedload
particles generally induce a stronger stream-parallel signal as a result of downstream
particle motion.
This work adds to the growing body of research using indirect bedload transport
monitoring methods, such as seismic and hydroacoustic measurements, to understand
complex fluvial dynamics in alluvial rivers. By advancing the use of seismic
monitoring, it improves understanding of sediment movement and its effects on river
dynamics. Using a unique dataset from the River Feshie in Scotland, this research
reveals: (1) the importance of sensor placement for clear seismic signals produced
by bedload transport; (2) variability in bedload entrainment thresholds, which are
influenced by event duration and magnitude, as well as the occurrence of successive
events; and (3) a strong horizontal directional component to the seismic signals
generated by bedload transport, challenging assumptions about vertical dominance.
These findings demonstrate the value of combining independent datasets for longt-erm
monitoring of bedload transport, offering insights into the spatial and temporal
evolution of sediment mobilisation. This provides crucial information for effective
river and land-use management, particularly as climate change alters the frequency
and intensity of high-flow events
Multi- and heterometallic ProPhenol catalysts for cyclic ester ring-opening polymerisation
Get PDFPlastic is an integral part of daily life, utilised in many sectors spanning from domestic to industrial purposes. However, high societal reliance on plastic products as well as the durability of plastic have led to rising environmental concerns. This has prompted urgent research into sustainable and biodegradable alternatives. Polymers such as aliphatic polyesters (i.e. poly(lactic acid) (PLA), poly(ε-caprolactone) (PCL) and poly(δ-valerolactone) (PVL)) are attractive candidates owing to their desirable properties, biodegradability and biocompatibility. The ring-opening polymerisation (ROP) of cyclic esters is an efficient route for producing well-controlled biodegradable polyesters, typically performed using organometallic catalysts. Heterometallic cooperativity is of key interest in polymerisation catalysis, with many heterometallic complexes outperforming the homometallic counterparts in terms of activity and control. In heterometallic systems, synergistic interactions between different metals depend on the position of proximate metals within a ligand scaffold; this plays a significant role in monomer coordination and the subsequent ring-opening process. The work in this thesis aims to exploit and understand the synergistic effects stemming from heterometal interactions in the ring-opening (co)polymerisation of cyclic esters, for multimetallic complexes based on the ProPhenol ligand scaffold.
While heterometallic cooperativity has delivered enhanced catalyst performance in the ROP of cyclic esters, so far this has been almost exclusively limited to their homopolymerisation. The work in Chapter 2 outlines heterobimetallic cooperativity and trade-offs in the synthesis of PCL-PLA block copolymers, comparing the catalyst performance and control to the homobimetallic analogue. Specifically, this study shows that heterometallic Mg/Zn and Ca/Zn catalysts, based on the ProPhenol ligand, show significant activity enhancements in the synthesis of PCL-b-PLA diblock copolymers, outperforming the Zn/Zn analogue with the activity order Ca/Zn > Mg/Zn > Zn/Zn. The excellent activity accompanies good (co)polymerisation control, generating well-defined PCL-b-PLA block copolymer structures. Interestingly, the catalyst activity order is completely reversed upon the sequential addition of ε-caprolactone as the third monomer in an attempt to generate PCL-PLA-PCL triblock copolymers. This demonstrates that the features of heterometallic “ate” catalysts that boost polymerisation activity can also enhance competitive undesired transesterification processes. Overall, this chapter highlights the importance of the choice of the metal combination to generate well-defined block copolymer microstructures, especially when targeting multi-block copolymer structures.
Heterometallic complexes can be accessed through several different synthetic pathways, including transmetallation and sequential deprotonation routes. Many s-block metals exhibit high catalyst activities in polymerisation owing to the Lewis acidity of the metal (facilitating monomer coordination), yet often give poor control. In contrast, catalysts based on earth-abundant aluminium often show good polymerisation control yet lower activities. Heterometallic catalysts have the potential to combine these features to deliver improved performance. Chapter 3 describes synthetic approaches to prepare novel heterotrimetallic ProPhenol complexes based on inexpensive and low-toxicity metals, by combining an alkali metal (Na or K) with two aluminium centres (AlMe₃, AlEt₃ or Al(iBu)₃). These complexes were characterised in the solution-state by NMR spectroscopy, and in the solid-state via single crystal X-ray diffraction. The molecular structures provide insight into the metal coordination geometry and different bonding modes for the two Al centres; where one Al centre sits in a ‘pocket’ within the ligand scaffold, while the other is coordinated through a dative bond to AlR₃ from a benzylic oxygen of the ProPhenol ligand.
Chapter 4 builds upon the work in Chapter 3, and probes the reactivity of these Na/Al₂ complexes through NMR studies. The abstraction of the datively bound AlR₃ unit was investigated as a route to form bimetallic mono/trivalent Na/Al analogues of the di/divalent Mg/Zn and Ca/Zn complexes reported in Chapter 2. However, the AlR₃ group persisted in the presence of a range of Lewis donors. Therefore, heterotrimetallic Na/Al₂ complexes were directly tested for the ROP of cyclic esters using benzyl alcohol as an initiator, as aluminium analogues of a previously reported Na/Zn₂ ProPhenol complex. PCL, PLA and PVL were successfully generated with good control, albeit at relatively low activity rates (0.068 h⁻¹, 0.053 h⁻¹ and 0.195 h⁻¹). Notably, reactivity studies of the heterotrimetallic Na/Al₂ complex with Lewis base additives provided insight into the different roles of sodium and aluminium metals in cyclic ester polymerisation
Structural studies of energetic materials at extreme conditions
Get PDFA class of materials of significant technological importance are energetic materials - explosives, propellants and pyrotechnics. The performance of energetic materials is largely dependent on the crystal structure of the material under both ambient conditions and elevated temperatures and pressures. Hence, in order to understand and model the behaviour of energetic materials under operational conditions, detailed structural information for these compounds must be obtained over a range of conditions. This thesis describes the high pressure structural behaviour of a series of oxidisers namely ammonium perchlorate (AP), potassium dinitramide (KDN) and ammonium dinitramide (ADN).
Detailed structural information on AP was obtained to a pressure of ∼11.5 GPa using neutron powder diffraction. Under hydrostatic conditions, AP undergoes a first-order phase transition at ∼4 GPa to form phase II AP. Crystallographic lattice parameters, equation of states, and hydrogen-bond distances for both phases I and II were obtained. No evidence for phase III was observed despite spectroscopic evidence that had suggested the II to III transtion. Low-temperature neutron diffraction studies at ambient pressure indicate a second-order phase transition at 130 K and a previously unreported first-order phase transition at 29 K. Lattice parameters and hydrogen-bond distances are reported down to 31 K.
The dinitramide ion is of interest because of its high oxygen content. KDN - a simple dinitramide salt that is a useful model for understanding any pressure-induced structural change. The variation in the Raman spectra and crystal structure with pressure of KDN was reported here. We see evidence for a previously unobserved, subtle phase transition at ∼9 GPa. The Raman spectra were recorded up to ∼10.5 GPa where combination of mode hardening and softening is observed together with the appearance and disappearance of vibrational modes. To complement this study, a single-crystal X-ray diffraction study was also performed up to 10.5 GPa. Upon initial analysis no discontinuous structural behaviour was observed in the structural parameters, but closer analysis of the volumes of the voids and the network of intermolecular contacts indicated a subtle phase transition and can be reconciled with the observed changes in the Raman spectra. A complementary density functional theory study of KDN was performed to aid in the assignment of eigenvectors and serve as a comparison to experimentally obtained Raman spectrum.
A further dinitramide salt was studied in this thesis - ammonium dinitramide (ADN). ADN has been suggested as a environmentally benign replacement for AP as an oxidiser and has been extensively studied. This thesis reports the variation with pressure of the crystal structure and Raman spectra of ADN. No evidence of a phase transition up to a pressure of 10.2(2) GPa was observed. However, lengthening of the N–D bond distances was observed alongside a corresponding shortening of the N–D· · · O hydrogen bonds. Raman spectra have been measured up to a pressure of 7.1 GPa and corroborating evidence for the stretching of the N–H bonds was discovered. Further analysis of the crystal structure of ammonium dinitramide reveals significant reorientation of the ions, which result in the breaking and reformation of hydrogen bonds. This is supported by the N–H stretching modes for which discontinuities were observed with pressure. A complementary density functional theory study of ammonium dinitramide was performed to aid in the assignment of eigenvectors and serve as a comparison for experimentally obtained Raman spectrum
Metabolomics and machine learning to assist biotechnology culture optimisation
Get PDFOptimisation of product titre or yield in a bioprocess is crucial for the economic and technical success of its operation. This optimisation problem is usually a challenge, as it involves several factors or variables. For example, in a bioprocess, medium components are important factors in the final titre, and the concentrations of each component need to be manipulated to achieve optimal conditions. Here, we present an active learning/Bayesian optimisation framework to enhance surfactin titres by adjusting medium component concentrations. Surfactin, produced by bacteria of the genus Bacillus, is a promising biosurfactant due to its physical and chemical properties. However, reported laboratory titres are typically low because of its complex molecular assembly pathway. We used active learning to refine the culture medium composition through iterative experimentation, enhancing Surfactin C levels in Bacillus subtilis DSM 3256. Growth curves and other central metabolites were measured as part of the experimental loop. The final medium mixture resulted in approximately a 1.6-fold increase after three rounds, compared to the M9 medium standard. Reanalysis of the optimisation data reveals trade-offs when comparing the production of lipopeptides, such as Surfactin D and Iturin A, with the maximum OD in the growth curve data. Organic acids in the supernatant positively correlate with Surfactin C levels, suggesting an impact on central carbon metabolism. For some metabolites, including certain amino acids and sugars, the change in their abundance around the optimal surfactin C mix is not uniform, indicating an "anisotropy" in how metabolism reacts to shifts in carbon and nitrogen levels. Thus, our framework addresses the challenges of data handling and analysis, offering several visual tools, data analysis techniques, and analytical methods (using mass spectrometry), which promised to be a contribution to Design, Build, Test & Learn cycles.
After addressing the challenge of modifying the concentrations of two components in the culture medium, we scaled up our approach to optimise surfactin production by modifying all seven components of the M9 medium, transforming it into a multidimensional optimisation protocol. However, performing the mixing and medium preparation became technically challenging.
Thus, this experiment was made possible through a high degree of automation, both computationally and experimentally. Two pipelines were built: the first one addresses the initial sampling and first robotic experiment in a Bayesian optimisation loop, while the second one execute data analysis following data acquisition from the mass spectrometer and can couple with the concentration mixing protocol in the Opentrons OT-2 robot for subsequent iterations. The Opentrons scripts were able to calculate and transfer the correct volumes of each component based on the stock concentration and desired concentration in the wells, generating robot-ready instructions to perform the mixing. Similar protocols were employed for quenching and sample preparation, enabling a full experimental cycle in 2-3 days.
In the experimental design part, we opted for an off-line approach, whereby sufficient samples—specifically 42 combinations with four replicates each, plus quality control and biological controls—were obtained from a single space-filling design to cover the full seven-factor space. The results indicate that combinations close to the M9 reference composition are the highest producers of surfactin C, confirming the optimal carbon and nitrogen conditions from the previous 2D iterative experiment. We then generated a high-quality surrogate model of production outputs, including lipopeptide production and biomass, measured as OD. This model serves as a realistic benchmark for testing single-objective and multi-objective lipopeptide production optimisation using Bayesian optimisation. From the single objective optimisation, results showed that the optimal number of initial samples and batch size can be adjusted to achieve the maximum Surfactin C yield in fewer iterations. However, the greater number of factors and the observed variance in the measurements mean the iterations cannot be reduced further, with approximately 10 iterations required using our current experimental setup of seven initial samples and seven combinations per batch (with 4 replicates). In the case of multi-objective optimisation, A Bayesian optimisation framework was able to identify the Pareto Front between lipopeptide production and biomass in the 7-dimensional factor space, with batch sizes and number of iterations comparable to those obtained from microplate experiments. This thesis tested that Bayesian optimisation is a feasible option for optimisation of secondary metabolites such as lipopeptides and that this approach can integrate with automation for high-throughput microbial metabolism studies
Enriching sentence-level machine translation
Get PDFNeural Machine Translation (MT) has long been established as a successful paradigm
to produce high-quality MT across many languages and domains. However, it suffers
from one significant limitation – it is too often formulated as a task of translating isolated
sentences in a source language into sentences in the target language. This renders
standard MT models unable to capture any information that is not in the sentence, such
as document context, speaker information, the domain of the text, external constraints
etc. This thesis aims to study this limitation, analyse the shortcomings of sentence-level
MT, and present some approaches to enrich MT models to overcome this limitation.
The first part of this thesis introduces a method to quantify the amount of information
missing from source sentences that is needed to translate them perfectly. This method
is called “cheat codes” and it allows us to establish an upper bound on the amount
of additional information that the model needs to be provided to be able to exactly
reproduce reference translations. We find that a surprisingly small amount of leaked
information about the target in addition to the source is enough to achieve this. We
also use this method to study what parts of translation are difficult for these models to
learn correctly, even in the presence of extra information. This analysis allows us to
signpost some hard problems for neural MT for further research to focus on.
The second part of the thesis presents two examples of how MT can be augmented with
extra information to improve translation quality or overall user experience in specific
applications. The first example is using document context, which is always used by
human translators when translating text, but is rarely present in parallel corpora. We
extract and publish a large-scale dataset of parallel sentences with corresponding contexts
from existing publicly available resources, and show that this data helps improve
translation performance in terms of overall quality as well as specific document-level
phenomena. The second example is providing timing constraints to an isochronous
MT model for use in automatic dubbing. By incorporating duration information and
keeping track of it while translating, the model can produce translations that better
match the source audio, which eventually results in a better user experience when
viewing the automatically dubbed content.
On the whole, we find that even though a relatively small amount of information is
missing from sentence-level MT, enriching the models with these small pieces of
information can have a significant positive impact on the quality and usefulness of
MT systems in a wide variety of situations. We provide detailed analyses, datasets, and
methods to build better MT systems and encourage future research in this direction
Recognising, valuing and supporting clinicians who teach: a critical realist exploration
Get PDFThis project set out to investigate an apparent contradiction in the published literature on recognition and reward of doctors who teach: while medical schools report a range of apparently successful interventions designed to recognise and reward their clinical teachers, teachers consistently report feeling under-recognised and undervalued.
Using a critical realist (CR) perspective, the project aimed to establish: a) why there is a widespread perception that clinical teachers and teaching are under-recognised / undervalued; b) why this persists despite the best efforts of medical schools to resolve it; and c) what, if anything, might be done to change this within existing organisational and societal constraints. A range of transdisciplinary perspectives were used to theorise potential causal mechanisms, and a critical pluralist approach was taken to methodology, combining case study research with critical discourse analysis, followed by application of a complex, adaptive systems (CAS) perspective to model the findings at the levels of ‘society’, ‘institution’, ‘organisation’ and ‘individual’.
Key findings of this study at societal and institutional levels were that the problem is underpinned by a primary tension between healthcare (high value) and education (lower value), and that this is compounded by structural factors such as lack of resources and heavy emphasis on quality assurance and accountability. Within the case study school, evidence suggested that: teaching was being ‘backgrounded’; communication was unidirectional, fragmented and liable to break down; the rewards being offered had high potential for adverse effects; and the reliance on implied contracts was contributing to a mismatch in expectations between teachers and medical schools.
Although this problem has multiple causes, the application of a CAS perspective to map the connections between these allowed the generation of a series of recommendations for change aimed at medical school managers and institutional policy makers. This project also makes a novel contribution to the medical education research landscape by demonstrating how critical realist metatheory can be utilised within the medical education domain and by illustrating how bricolage and iterative research designs can support a flexible approach to research, while still producing work which meets demands for theoretical and methodological rigour
Contributions of maternal parenting stress and mental health to parenting practices and children’s developmental outcomes
Get PDFPrevious studies suggested that children of mothers with higher parental stress scores or mental health difficulties may be more likely than others to have developmental delays. Additionally, such parents may have a lower frequency of engagement in play and stimulating activities with their children compared to their peers. Emerging literature has also explored the bidirectional association between parenting stress and maternal mental illness. Further, studies have suggested poor child developmental outcomes in socioeconomically deprived settings and in events that could lead to the disruption of maternal and child health services, such as the COVID-19 pandemic. However, there is scarce or mixed evidence on the contributions of parenting stress to children's developmental outcomes in socioeconomically deprived settings such as sub-Saharan Africa (SSA). In addition, no study has examined the potential worsening of the association between maternal mental illness and child developmental outcomes at birth during the COVID-19 pandemic. This thesis aimed to fill these gaps by examining the contributions of parenting stress and maternal mental health on children's developmental outcomes in the context of socioeconomically deprived settings and birth during COVID-19, using linked administrative health datasets from Scotland and longitudinal studies from the SSA.
Chapter 1 reviews evidence on predictors of parental stress, its links to maternal mental health, parenting practices and children's developmental outcomes. Chapter 2 examines sociodemographic predictors of maternal parenting stress in the SSA. The results showed that mothers' income and their level of education were associated with reduced parental stress scores (PSS). However, this study found that marital status, mother's age, child's age, and the number of children under five years were not associated with PSS.
In Chapter 3, this study employed a Random Intercept Cross-Lagged Panel Model (RI-CLPM) to examine the longitudinal relationships between parental stress, caregiving stimulation, and child development. The findings showed a reciprocal association between caregivers' stimulation practices and children's developmental outcomes. However, such associations were not consistent with parental stress, stimulation practices and children's developmental outcomes.
Chapter 4 investigates whether maternal mental health (both prenatal and postnatal) mediates the relationship between socioeconomic deprivation and children's developmental outcomes in the context of Greater Glasgow and Clyde, Scotland. The results showed that maternal mental health assessed by a history of hospital admissions mediated the relationship between SED and children's developmental outcomes, but only to a small extent. Chapter 5 examines the interaction effects of maternal mental health disorders and being born during COVID-19 on children's developmental outcomes in Scotland. The results showed that being born during the COVID-19 pandemic and maternal MH influenced child development with relatively small effects, with mixed findings on their combined presence. Lastly, Chapter 6 provides an overall discussion of the findings presented in this thesis.
In conclusion, the findings align with the existing literature on the potential association between parental stress, maternal mental health, parenting practices, and children's developmental outcomes. These findings, therefore, underscore the need to invest in maternal mental health interventions and address predictors of mental health, such as parental stress and socioeconomic deprivation. In addition, there is a need to explore the potential long-term effects of both being born during the pandemic and maternal mental health on children's developmental outcomes
Somatosensation in rat models of neurodevelopmental disorders
No full textNeurodevelopmental disorders (NDDs) are a group of life-long conditions linked to intellectual
disability, epilepsy, and autism. Two of the most common monogenic causes of NDDs are Fragile X
Syndrome (FXS) and SYNGAP1 haploinsufficiency. Many individuals with these NDDs have atypical
sensory experiences, such as tactile-seeking or tactile-aversive behaviours and high or low pain
thresholds. Mouse models have been used to study the pathophysiology underlying altered
somatosensory processing but this has not been previously investigated in the rat models of these
conditions. Rats are a useful model given that they are highly social and display a rich behavioural
repertoire in response to their environment.
This thesis first aims to assess the somatosensory behavioural phenotypes in rat models of these two
monogenic NDDs. Adult male rats modelling FXS (Fmr1⁻/ʸ) exhibited an unaltered tactile and pain
phenotype and they developed injury-induced mechanical hyper-sensitivity following hindpaw surgical
incision at a time course and magnitude that did not differ from wild-type rats. In contrast, the two
studied rat models of SYNGAP1 haploinsufficiency (Syngap⁺/⁻ and Syngap⁺/Δ⁻ᴳᴬᴾ) displayed tactile
hypo-reactivity without a change in their reactivity to acute noxious stimuli as adults. This indicates a
role for the SYNGAP protein in the processing of tactile stimuli in rats. However, this finding differed in
one of the studied cohorts, where rats that did not display an altered somatosensory phenotype as
juveniles also did not have it when studied as adults.
Spinal cord processing is an important early stage in the transmission of somatosensory information
before it is perceived and acted on. The somatosensory region of the spinal cord, the dorsal horn,
undergoes activity-dependent postnatal development, which includes changes in the anatomical
organisation of primary afferent terminals and functional maturation of spinal circuits. As many
autism-associated genes, like FMR1 and SYNGAP1, are key components of activity-dependent
processes, this postnatal spinal maturation was expected to be disrupted. Overall, spinal cord
immunostaining revealed largely unaltered central termination patterns of tactile and nociceptive
primary afferents in rat models of FXS and SYNGAP1 haploinsufficiency. Similarly, hindpaw glabrous
skin structure and tactile corpuscle innervation density were not altered in any of the models.
As Syngap⁺/⁻ rats exhibited reduced tactile spinal reflex responses, which were not associated with
altered gross anatomy of tactile inputs, the possibility of underlying functional changes at the level of
the spinal cord was investigated. Importantly, SYNGAP expression was reduced by 50% in the dorsal
horn of Syngap⁺/⁻ rats. Given SYNGAP’s localisation at the postsynaptic density and its role in
regulating AMPA receptor trafficking, disrupted glutamatergic spinal synaptic transmission in the
Syngap⁺/⁻model was predicted. Compound action potential recordings from dorsal roots were
conducted to establish parameters for later characterisation of excitatory inputs in spinal neurons and
they revealed no differences in electrical properties of the Aβ-, Aδ-, and C-fibre primary afferent
components. Using whole-cell patch clamp recordings from spinal neurons, it was shown that
monosynaptic excitatory postsynaptic currents had reduced maximum amplitude in the tactile, but not
nociceptive, region of the spinal cord dorsal horn. Finally, the response threshold of spinal reflex
networks was found to be increased in dorsal root-ventral root potential recordings.
This thesis examines tactile and pain behaviours in rat models of FXS and SYNGAP1
haploinsufficiency, highlighting their differing tactile phenotypes, and provides evidence for a spinal
functional mechanism contributing to tactile hypo-reactivity in Syngap⁺/⁻ rats
Integrated geophysical imaging of the lithosphere beneath Britain: 3D magnetotelluric modelling, multi-physics inversion, and space weather applications
No full textGeophysical methods are one of the main tools to study the Earth’s interior. Through observations at the Earth’s surface, advanced numerical methods, and fundamental physical laws, geophysical inversion allows us to create models of the distribution of physical properties in the subsurface.
The magnetotelluric (MT) method uses simultaneous measurements of electric and magnetic fields at the Earth’s surface to probe the electrical properties at a broad range of frequencies and, therefore, depths. Given its ability to reveal electrical conductivity contrasts associated with geological variations and other physical parameters, the MT method is widely applied to resource exploration, geohazards monitoring, lithospheric studies, and, recently, space weather studies. During geomagnetic storms, the rapid variations in the geomagnetic field induce electrical currents in the subsurface. These so-called Geomagnetically Induced Currents (GICs) can damage ground-based infrastructure, such as electrical power networks and railways. To assess and forecast the risk of GICs, knowledge of electric fields at the Earth’s surface is necessary. The strength of the geoelectric field depends not only on the magnetic source but also on the ground electrical conductivity. Therefore, models that accurately reflect conductivity contrasts are required to improve the estimation of geoelectric fields, especially in areas of complex geology, such as the UK.
This thesis presents the first three-dimensional (3D) electrical resistivity model of the lithosphere beneath Britain (BERM-2024) and explores its geological implications and applications to space weather. BERM-2024 was derived from the v inversion of long-period magnetotelluric (LMT) data primarily acquired during a field campaign conducted by the British Geological Survey and the University of Edinburgh between 2021 and 2024, with additional legacy data incorporated to establish a nationwide dataset of 70 sites across Great Britain.
A systematic workflow for 3D MT inversion was developed through extensive testing of prior models, inversion strategies, and regularization parameters. The influence of bathymetry and marine sediments was assessed through a pilot study using legacy data from the Isle of Skye. Computational resources and data misfit criteria were also considered, resulting in an optimized methodology that can provide general guidance for 3D MT studies, and a robust and geologically meaningful model.
BERM-2024 reveals strong lateral and vertical resistivity variations that correlate with known geological and tectonic structures, and provides new insights into the lithosphere down to depths of ∼200 km. In the upper crust, high conductivity anomalies correlate to sedimentary basins in western Britain, such as the Cheshire and Welsh basins, while resistive features align with granite plutons in Scotland and Cornwall. Sharp resistivity contrasts identified at mid- to lower crustal depths in northern Scotland, the Southern Uplands, northern England, and Wales align closely with major terrane boundaries. Notably, a prominent conductor at 85-150 km depth beneath the West Midlands region is imaged for the first time, and is here termed the West Midlands Conductor (WMC). Integrated geophysical-thermochemical modelling suggests that high water contents (∼700-600 ppm) in the lithospheric mantle provide a plausible explanation for the resistivity response of the WMC.
Beyond its geological implications, BERM-2024 provides a new input for estimating geoelectric fields in the UK, a key element in modelling and forecasting geomagnetically induced currents in critical infrastructure during space weather events. Geoelectric fields modelled for major geomagnetic storms demonstrate a strong correlation with fields measured at geomagnetic observatories. Discrepancies in amplitude, however, highlight the need for denser data coverage and further research.
This thesis also pioneers the application of multi-physics joint inversion to nationwide geophysical datasets in Britain. Land gravity data and Rayleigh-wave group-velocity travel times from ambient noise seismology, not previously used to generate 3D models, were individually and jointly inverted along with the LMT dataset. Challenges related to data coverage and resolution, as well as specific parameters involved in the joint inversion methodology, are addressed and discussed. The resulting high-resolution crustal models of density, shear-wave velocity, and resistivity consistently recover robust structures such as sedimentary basins, intrusions, and terrane boundaries, and suggest potential unrecognised intrusions. The extensive parameter testing conducted provides guidance for future joint inversion studies.
Overall, this research delivers new insights into the lithospheric structure of Britain, and establishes a new baseline for regional and local geophysical research. The models and methodologies developed here can support future geological and geophysical studies, deepen our understanding of Britain’s lithosphere, and improve assessments of space weather hazards
Parameter estimation in sparse state-space models
Get PDFState-space models are a flexible framework for modelling sequential data in the
presence of noise or incomplete observations, within which we model a system via a
hidden state process and a related observation process. These processes are described
via a pair of distributions encoding the state dynamics and the observation process,
with the distribution of the current state depending only on the previous state, and the
distribution of the current observation depending only on the current state. In general,
the parameters of these distributions are unknown, and are challenging to estimate,
with conventional estimation schemes failing due to the temporal dependence of the
time series, and the resultant concentration of the likelihood function. In this work
we present several methods to estimate the parameters of state-space models, as well
as some methods for estimating the form of the model itself when this is unknown.
In particular, we focus on methods that admit interpretable estimates via promoting
sparsity in the parameters, thereby shrinking many parameter values to zero.
In the first contributing chapter of this work, Chapter 3, we propose a method
to obtain sparse Bayesian estimates of the transition matrix of a linear-Gaussian statespace
model by utilising reversible jump Markov chain Monte Carlo. We discuss the
construction of the reversible jump kernel, and how to interpret the sampled sparsity
in terms of a Bayesian causality. We demonstrate our method on several synthetic
datasets, where we have the ground truth of causality, and on real-world weather data
where we do not, comparing the performance to the existing state-of-the-art.
In Chapter 4, we propose a method to promote graphical clusters in the transition
parameters of a linear-Gaussian state-space model by utilising a sparsity promoting
estimation scheme in conjunction with a dynamically adaptive penalty. We design
a general framework to construct state clustering methods within state-space models,
and then construct a representative method as a case of this general framework,
wherein we apply ideas from network analysis to design an iteratively applied cluster
promoting penalty function. We test our method on a series of synthetic datasets,
and compare the performance to the existing state of the art.
In Chapter 5, we propose a method to construct a polynomial representation of a
general state-space model, whereby we learn a sparse approximation of the transition
function from a basis of polynomial terms. This allows us to infer the connectivity of
the hidden states, thereby providing insight into the unknown underlying dynamics.
In the final main chapter, Chapter 6, we propose a method to approximate the
intractable optimal proposal of a particle filter utilising a shallow neural network which
parametrises a Gaussian mixture distribution. We compare this proposal to several
standard proposals, and extend the work to simultaneous estimation of the transition
and proposal distributions.
Finally, we provide some concluding remarks on the techniques developed, and
present a number of potential avenues for future research
Access Repository Dashboard
Do you manage Edinburgh Research Archive? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!