16580 research outputs found
Sort by
Track Fitters for Track Triggers A Linearized Track Fit for the Event Filter of ATLAS at the High-Luminosity LHC
No full textThe upcoming High-Luminosity LHC (HL-LHC) upgrade of the Large Hadron Collider (LHC), will result in an average pileup of hμi = 200 proton-proton collisions per bunch-crossing. This is an unprecedented challenge for the Trigger and Data Acquisition Systems (T-DAQs) of the main experiments. Since track triggers are less susceptible to high-pileup than calorimeter triggers, a track trigger was developed for a low level of the ATLAS T-DAQ. The Linearized Track Fit (LTF) is a non-iterative algorithm that reconstructs the goodness and helix parameters of track candidates based on a linear approximation around a reference track solution. It requires minimal computational resources but a large amount of fast memory to store the linearizations of the reference tracks. Furthermore, it is highly parallel because it processes every helix parameter of every track candidate independently and with constant runtime. Therefore, it is uniquely well suited for implementation on parallel hardware. The Hardware Track Trigger (HTT) is a flexible custom hardware system that contains a full track reconstruction workflow; It was developed to be able to serve as both a low level track trigger and a track reconstruction coprocessor for the event filter. The basic building block of the HTT is the Pattern Recognition Mezzanine (PRM), which consists of a single large Field Programmable Gate Array (FPGA) that runs the LTF and holds the linearization constants in a High-Bandwidth Memory (HBM) as well as several Associative-Memory ASICs (AM-ASICs) that perform pattern recognition.
This study presents the finalization of the PRM firmware development, particularly the implementation of the Track Fitter Block (TF), of the AM-ASIC interface, and of the HBM Interface (HBM-IF). The PRM was confirmed to run on prototype hardware using an Intel Altera Stratix 10 MX 2100 FPGA (Intel Stratix 10) FPGA and custom made emulators for the AM-ASICs. The track fitting quality (fit goodness, parameter resolution, and reconstruction accuracy), trigger quality (efficiency and purity), and the technical performance (latency, throughput, and resource usage) were evaluated using register transfer level (RTL) simulations. Additionally, the LTF was implemented on an Nvidia Tesla T4 GPU (Nvidia T4) Graphics Processing Unit (GPU) to investigate which hardware architecture is better suited to the algorithm. The results of those implementations are compared to pure software results and to the results of a completely different track fitting algorithm, the General Triplet Track Fit (GTTF).
This study found that it is possible to implement an LTF track fitter on the Intel Stratix 10 so that it fulfills the HTT baseline requirements for the event filter scenario. As of now, the latency target of 1 μ s for the low level scenario was missed by 30 %. Because of its exceptionally low computational demands and its reliance on fast memory access and
wide connections, the LTF has been found to be best suited to be implemented on a large number of low performance GPUs. This enables a cost effective track trigger architecture using consumer-grade or even outdated devices. This study successfully demonstrates the feasibility of hardware-accelerated track reconstruction for next-generation
particle physics experiments and the multi-platform evaluation offers valuable insights into the trade-offs between different computing architectures for high-throughput scientific computing applications. The study contributes both to the immediate needs of ATLAS and to the broader field of real-time data processing in scientific applications, establishing a foundation for future developments in hardware-accelerated track reconstruction
Wrapped Floer Homology in the Circular Restricted Three-Body Problem
No full textApplications of symplectic geometry to the Three-Body Problem have slowly begun appearing in the past few years, allowing one to approach the problem with tools from Floer theory.
In this thesis, we introduce a new variant of Floer theory: Local Wrapped Floer Homology, which generalises the previously-existing 'Wrapped Floer Homology' to degenerate settings. We use this new machinery to prove a generalisation of the famous Poincaré-Birkhoff theorem to open-ended paths with exact Lagrangian ends in a Liouville domain, assuming a twist condition first stated in [arXiv:2011.10386].
We then proceed to improve the applicability of our theorem to real-world problems, by replacing the constraining 'twist condition' mentioned above by a 'Weakened Twist Condition', and by adapting the setup to degenerate Liouville domains.
Finally, we deduce applications to the Three-Body Problem: first to prove existence of infinitely many trajectories of collision, and then trajectories bi-normal to the xz-plane; under the assumption of the Weakened Twist Condition
The structure of the innermost regions of circumstellar discs with VLTI/MATISSE
No full textCircumstellar discs serve as the progenitors of planetary systems. Their inner regions (from <1 au to a few astronomical units) are where terrestrial (i.e. Earth-like) planets form. Structures closely connected to planet formation, such as spirals, rings, or gaps, have been observed in the outer disc regions (∼20– 400 au). These outer regions are being intensively studied with instruments like SPHERE at near-infrared wavelengths or ALMA in the (sub-)millimetre. The innermost regions, on the other hand, have only been made accessible recently by interferometers with high spatial resolutions in the near- and mid-infrared (e.g. AMBER, MIDI, PIONIER, and GRAVITY on the VLTI). In this thesis, I use spectro-interferometric observations with the VLTI/MATISSE instrument, covering a broad wavelength range in the mid-infrared from the L/M (2.8–4.2 µm/4.5–5 µm) up to the N band (8– 13 µm), to investigate the structures present in the innermost regions of planet-forming discs. MATISSE provides spatial resolutions sufficient to resolve the innermost disc regions for sources in nearby star formation regions (∼100 pc). I apply data reduction and parametric modelling to give astrophysical interpretations of the interferometric observations. First, I explore the circumstellar discs of the HD 142527 binary system, where a close passing (periapsis∼5 au) stellar companion greatly perturbs the disc around the primary. Using a a disc model with an off-centre Gaussian asymmetric component, I find a time-variable structure around the primary at a separation of∼1– 1.2 au over the course of our observations (2021, 2022, and 2023) close to periapsis. The complex structure in the disc seems to be strongly influenced by the companion passage and warrants further observations to enable more detailed modelling. Subsequently, I perform a statistical analysis of∼60 sources from the MATISSE GTO YSO survey. I derive L- vs. N -band sizes from modelling and combine them with simple disc models. In doing so, I find that many discs in the survey sample show signs of gaps or truncations. Here, the sizes increase when the gaps in the disc become larger, up to the point of no significant N -band emission (i.e. truncated), which show smaller sizes than for continuous discs. The sample contains many strongly asymmetric sources from which disc substructures are deduced
Straggling in Underdepleted Ultra-Thin Silicon HV-CMOS Sensors
No full textThe pursuit of increasingly rare processes with higher precision in modern high-energy particle physics experiments drives the development of silicon pixel detectors offering excellent spatial and timing resolution, radiation hardness, and minimal material budget. However, a lower material budget necessitates thinner detectors, which provide less active volume for energy deposition, thereby reducing the signal-to-noise ratio.
A promising approach is represented by HV-MAPS (High-Voltage Monolithic Active Pixel Sensor). These monolithic pixel sensors integrate the active pixel matrix and readout electronics on a single die, allowing the silicon sensors to be thinned down to 50 μm. The sensor studied in this work is the MuPix10, a full-reticle prototype of about 2×2 cm^2, developed for the Mu3e experiment. Each pixel features a large fill factor electrode design of a deep n-well embedded in a low to medium resistivity p-substrate.
HV-MAPS are generally biased from the top through guard-ring structures, and cannot be operated beyond full depletion. However, variations in substrate resistivity and manufacturing tolerances lead to varying depletion depths across sensors, mandating a better understanding of charge collection in under-depleted operation.
To investigate the corresponding charge collection spectrum, a dedicated testbeam campaign was performed using 350 MeV c^−1 pions. This is done for different sensor thicknesses, ranging from 50 μm to 100 μm, as well as substrate resistivities and reverse bias voltages. By measuring the hit efficiency as function of the applied detection threshold, the integrated charge spectrum is reconstructed. To unfold the impact of detector effects on these measurements, a dedicated calibration procedure was developed based on charge injection to the pixel electronics and Fe-55 source measurements for absolute charge calibration.
The charge spectrum is modeled by a convolution of a Landau distribution with a normal distribution to account for energy deposition fluctuations and detector effects, respectively. This approach has been evaluated using simulated charge deposition spectra and has proven to be valid even for thin silicon sensors. By applying theoretical predictions for the scale and the most probable energy loss for incident particles of known kinematics, the effective charge deposition thickness can be extracted.
The measured charge collection exceeds expectations based solely on the depleted region, indicating that the non-depleted volume significantly contributes to the charge collection in under-depleted operation. Even for low bias voltages, where charge losses due to charge sharing are expected, the observed charge collection exceeds the depleted thickness expectation. Comparing an almost fully depleted 50 μm sensor with a 100 μm sensor at −20 V shows a 40 % higher most probable charge collection, corresponding to an excess effective deposition thickness of about 12 μm. The relative excess observed under equivalent bias conditions increases with the sensor thickness, implying that the size of the non-depleted region governs the overall charge collection behavior in this thickness range. This suggests that even in thicker sensors, a notable fraction of the non-depleted region participates in charge collection
Towards Self-Configuring Radiomics for Robust and Reproducible Predictive Performance
No full textIntroduction:
Radiomics aims to extract quantitative features from medical images that capture underlying biological and clinical characteristics. Despite its promise for precision oncology, radiomics research continues to suffer from poor reproducibility and limited generalization across studies, software, and imaging modalities. This thesis addresses
these fundamental limitations by systematically analyzing how methodological design choices—such as feature extraction, preprocessing, and model selection—affect
the robustness and transferability of radiomic biomarkers. To enable this large-scale methodological investigation, I developed the Radiomics Processing Toolkit (RPTK),
a fully automated and open-source framework that standardizes radiomics experimentation and benchmarking across heterogeneous datasets. Using RPTK, I conducted comprehensive evaluations on seven open-source cancer imaging cohorts and demonstrated the framework’s applicability in two clinical studies on lung cancer immunotherapy response prediction and colorectal neoplasia detection.
Materials and Methods:
My work integrates radiomics analyses performed on retrospective data, including seven public datasets and two proprietary cohorts, comprising 3,189 Computer Tomography
(CT) and Magnetic Resonance (MR) scans from 3,116 patients with a total of 3,273 segmented regions of interest (ROI). The seven open-source datasets include retrospective MR and CT cancer imaging data concerning different tasks for
cancer classification from 931 patients. The proprietary data collection include a multi-timepoint (prior treatment and during treatment) CT dataset for lung cancer
immunotherapy treatment response prediction (the Predict study) consisting of 73 patients and a large-scale CT liver imaging datasets with 1,997 patients investigating
in colorectal neoplasia detection (LiverCRC study). The mean patient age was 62 ± 17 years, and 53.6 % of participants were male. Radiomics features were extracted using two independent feature-extraction tools, PyRadiomics and Medical Image Radiomics Processor (MIRP), enabling standardized cross-extractor comparisons in compliance with the Image Biomarker Standardisation Initiative (IBSI). The
RPTK framework integrates adaptive preprocessing, standardized feature extraction, and robust feature stability filtering to enhance reproducibility and robustness for subsequent model training. Six machine learning models were trained to predict tumor malignancy, treatment response, colorectal neoplasia, or cancer subtypes based on the selected feature sets from each extractor. The performance of RPTK was tested against a state-of-the-art radiomics tool (AutoRadiomics) and six different deep learning models.
Results:
Across seven open-source datasets, RPTK outperformed both AutoRadiomics and deep learning models (Residual Networks (ResNet) and Densely connected convolutional Networks (DenseNet)), achieving a mean test Area Under the Receiver Operating Characteristic curve (AUROC) of 0.81 ± 0.12 compared to 0.68 ± 0.15 and 0.60 ± 0.16, respectively. In the Predict study, longitudinal delta-radiomics analysis
with RPTK improved early prediction of immunotherapy response compared to single-timepoint analyses with RPTK, and the inclusion of clinical variables further
enhanced model performance in RPTK. RPTK achieved a test AUROC of 0.75 ± 0.10 using delta radiomics, outperforming AutoRadiomics (0.51 ± 0.14) and the best deep learning model (0.56 ± 0.14). In the LiverCRC study, RPTK reached a test AUROC of 0.86 ± 0.04, significantly exceeding AutoRadiomics (0.65 ± 0.03) and deep learning (0.60 ± 0.03), demonstrating scalability and generalization in large multi-thousandsample datasets. Beyond these comparisons, RPTK also matched or outperformed 12 additional published test AUROC values reported on the integrated open-source datasets.
Conclusion:
Collectively, the results demonstrate that RPTK provides robust, state-of-the-art predictive performance across diverse imaging datasets and clinically relevant tasks. Its modular design enables fair cross-framework benchmarking while maintaining flexibility for clinical data integration and ensuring methodological transparency. The open-source release of RPTK fosters community-driven validation and supports future clinical implementation. This work thus represents both a methodological advancement and a step toward reliable, reproducible, and clinically translatable radiomics
Blick auf die Veränderungen: Eine Eye-Tracking-Analyse der Lebensmittelwahrnehmung nach bariatrischem Eingriff
No full textDie global steigende Prävalenz der Adipositas stellen Gesellschaft und Medizin vor große Herausforderungen. Bei der multifaktoriellen Genese der Erkrankung spielen für einen unklaren Anteil von Menschen mit Adipositas dysfunktionale Verhaltensweisen mit Parallelen zu klassischen Abhängigkeitserkrankungen eine Rolle. Die vorliegende Arbeit soll dazu beitragen, den Einfluss dieser Verhaltensweisen herauszuarbeiten, sowie deren Veränderung im Experiment zu beobachten, um ein besseres Verständnis für diese Patientengruppe zu schaffen. Hierzu wurde eine pseudorandomisierte Kohortenstudie durchgeführt, bei der 23 Probandinnen und Probanden vor und nach bariatrischem Eingriff (Roux-en-Y Magenbypass und Schlauchmagen) eine subjektive Bewertung auf den Ebenen Valenz, Craving und Arousal mithilfe eines standardisierten Bild-Sets abgaben. Überdies erfolgte die Erfassung von Eye-Tracking Parametern, sowie eine Psychometrie mit Erfassung des SKID nach DSM-V, Beck-Depression Inventory II, State-Trait Angst Inventar und der Yale Food Addiction Scale. Die untersuchte Kohorte zeigte eine Gewichts- und BMI-Reduktion, sowie einen Abfall der subjektiven Bewertung, insbesondere bei Lebensmitteln aus der Subkategorie des „schmackhaften Essens“, eine Lebensmittelgruppe, die dafür bekannt ist, abhängige Verhaltensweisen zu begünstigen. Zudem zeigte sich eine Korrelation dieses Abfalls mit dem BMI-Verlust der Probandinnen und Probanden. Psychometrisch zeigte sich eine Reduktion der Scores für Depressivität, Angst- oder Essstörungen, die zu den typischen Komorbiditäten der Adipositas zählen. Des Weiteren zeigte sich ein Abfall der Yale Food Addiction Scale, einem Score, der nicht nur eng mit Substanzabhängigkeit vergesellschaftet ist, sondern auch mit dem Gewichtsverlust, der Lebensqualität und dem Risiko einer erneuten Gewichtszunahme bei bariatrischen Patienten korreliert. Die vorliegenden Ergebnisse lassen daher die Schlussfolgerung zu, dass individuelle Lebensmittelwahrnehmung, Verhaltensweisen mit Parallelen zu abhängigem Verhalten und psychiatrische Komorbiditäten nicht nur die Entwicklung einer Adipositas begünstigen, sondern auch bei deren Therapie eine Schlüsselrolle spielen. Es ist die Aufgabe zukünftiger Studien, geeignete Scoring-Verfahren und Prädiktoren für Prävention, Indikationsstellung und Therapieplanung von Patientinnen und Patienten mit Adipositas zu entwickeln, um einen Beitrag zur Lösung des gesamtgesellschaftlichen Problems der Adipositas zu leisten
The time alignment of the Scintillating Fibre Tracker and a LFU study with at LHCb
No full textThis thesis describes two different scientific topics: the time synchronisation of the Scintillating Fibre (SciFi) Tracker and a Lepton Flavour Universality (LFU) study with the rare baryonic decay () at the LHCb experiment.
The SciFi Tracker is a part of the LHCb Upgrade I. In the context of its commissioning and early operation, the time synchronisation procedure, hereinafter referred to as the time alignment, was developed to align the detector signal integration window to the incoming physics signal and thus to optimise the hit detection efficiency. The strategy of the time alignment is introduced and the system timing characteristics study based on the early data (2023-2024) is discussed.
LFU is a feature of the Standard Model (SM) of particle physics, which refers to the equal couplings of the electroweak interaction to the 3 generations of leptons. Assuming LFU, the relative branching fraction between the baryon decays and is expected to be close to 1.0. A deviation from this expectation can be caused by possible new physics. Using the data collected by the LHCb experiment from 2011 to 2012 (Run1) and from 2015 to 2018 (Run2), which corresponds to an integrated luminosity of 9 , this relative ratio can be probed. The procedure of this LFU study is described and, depending on the different kinematic regions, the preliminary expected sensitivity, either for the ratio observable or for the observation of , is reported
Design and Synthesis of Molecular Structural Stabilizers for the ALS-relevant SOD1 Protein
No full textAmyotrophic lateral sclerosis (ALS) is a rare, devastating, currently incurable neurodegenerative disease. Despite numerous efforts in recent years, there is still no widely effective treatment available. In a subset of ALS patients, aggregation of the mutant human superoxide dismutase (hSOD1) protein drives the development and progression of ALS. Stabilization of hSOD1’s catalytically active, native homodimer state prevents dissociation into aggregation-prone and disease relevant monomers. To this end, the development of dimer-stabilizing non-covalent ligands was pursued. Small molecule ligands were designed to target the Val148 cavity of the hSOD1 dimer interface. The ligands lead structure, comprised of a central functionalized biphenyl moiety with a third functionalized benzene moiety connected via an amide linker, was retrosynthetically disconnected into three structural fragments. These fragments were systematically varied and combinatorially reconnected in silico, affording a library of 320 virtual ligands. Aiming to identify both, the most essential fragments, as well as the most promising ligand structures, molecular docking was utilized to screen the virtual ligand library. Based on the in silico prediction, twelve small molecule ligands were synthesized. Featuring amide coupling and Suzuki–Miyaura coupling as key derivatization steps, a concise modular organic synthesis route was developed. In order to provide an in vitro model of hSOD1-ALS pathology, recombinant expression of the hSOD1 and hSOD1 A4V proteins was established. Aiming to closely mimic physiological conditions, emphasis was placed on ensuring the presence of SOD1’s native maturation modifications. Mass spectrometry confirmed native Cu2+ and Zn2+ metalation and the formation of the intrasubunit disulfide bond. Demonstrating the enzymatic activity of hSOD1 and hSOD1 A4V proved correct folding and dimer-association. Taken together, this verified that hSOD1 and hSOD1 A4V were obtained in their native states. The ligands inhibitory effect on hSOD1 A4V aggregation was investigated in multiple complementary in vitro assays. A thioflavin T based aggregation assay served as a screening tool. Eleven ligands and two literature known compounds were assessed and three ligands were identified as putative hits. Dynamic light scattering confirmed these three hits and in addition, provided estimates of the particle sizes of the formed hSOD1 aggregates in solution. Electron microscopy revealed hSOD1 A4V aggregate morphology under the incubation conditions utilized in this work. The initial molecular docking study was evaluated in light of the experimental observations. This highlighted it as a valuable asset, as it had correctly predicted two of the in vitro identified hits. Furthermore, ligand binding modes were proposed by molecular docking and two hit ligands shared a proposed binding mode. In conclusion, three of the developed hSOD1 stabilizing small molecule ligands were demonstrated to reduce the formation of ALS-relevant hSOD1 A4V aggregates. Their effectiveness was confirmed by two complementary in vitro assays and rigorous evaluation of experimental results. These ligands provide valuable tools for the elucidation of hSOD1-ALS disease pathology and related research. One ligand in particular was highlighted as a potent lead compound for the development of a therapeutic treatment of hSOD1-ALS
Potenzial der quantitativen Computertomographie in der Diagnostik und Klassifizierung der chronisch obstruktiven Lungenerkrankung (COPD)und Asthma
No full textDie auf der Spirometrie basierende Diagnostik und Klassifizierung der COPD und differentialdiagnostische Abgrenzung zu Asthma weist zahlreiche Einschränkungen auf. Zum einen lassen sich Frühformen der COPD kaum erfassen, zum anderen können die obstruktiven Atemwegserkrankungen, COPD und Asthma, nicht immer klar voneinander abgegrenzt werden. Dies liegt vor allem an der ausgeprägten Heterogenität der beiden Atemwegserkrankungen und ihrer Vielzahl an phänotypischen Ausprägungen und Überlappungen. Die konventionellen Lungenfunktionstests (Spirometrie, Bodyplethysmographie und Transferfaktorbestimmung) und das aktuelle Klassifikationsmodell sind nicht in der Lage die phänotypischen Ausprägungen der Erkrankungen, deren Unterschiede und Überschneidungen sowie die Variabilitätzwischen einzelnen Individuen zu erfassen. Dadurch können unterschiedliche klinische Verläufe und entsprechende therapeutische Ansätze nicht erfasst oder untersucht werden.Um die Genauigkeit der Diagnostik zu steigern haben sich bisher neue Lungenfunktionstest wie die Impulsoszillometrie (IOS) und das Multiple-Breath-Washout (MBW) als vielversprechend bewährt. Neben diesen beiden Untersuchungen ist auch die bildgebende Diagnostik in Form der quantitativen Computertomographie (qCT) in der Vordergrund gerückt. In zahlreichen Studien konnte bisher gezeigt werden, dass mit Hilfe dieser neuen diagnostischen Verfahren eine frühe Erfassung der obstruktiven Lungenerkrankungen und eine detailliertere Darstellung der beiden Krankheitsbilder erzielt werden kann.
In der quantitativen computertomographischen Diagnostik stehen vor allem die Erfassung der Atemwegsparameter und der Parenchymparameter der Lunge im Vordergrund. Diese können individuelle Aufschlüsse über Lokalisation von Pathologien, Ausprägung, Verlauf und therapeutischen Ansprechen geben und ermöglichen dadurch ein breiteres Informationssprektrum der Erkrankungen. In der aktuellen klinischen Studie wurde untersucht, ob die qCT in der Lage ist Unterschiede zwischen den beiden obstruktiven Atemwegerkrankungen zu detektieren und in wie weit sich die qCT mit den konventionellen Lungenfunktionstests und den neuen Verfahren, MBW und IOS, vergleichen und entsprechend in den aktuellen klinischen Alltag einordnen lässt. Hierbei wurden die durch die Spirometrie, Bodyplethysmographie, Transferfaktorbestimmung, MBW, IOS und qCT erhobenen Daten von insgesamt 132 Probanden (102 COPD und 30 Asthmapatienten) ausgewertet und in unterschiedlichen Modellen untersucht. Es konnte gezeigt werden, dass die qCT mit ihren Parametern in der Lage ist Unterschiede zwischen den beiden obstruktiven Lungenerkrankungen darzustellen und sich auch in den aktuellen klinischen Kontext eingliedern lässt, da sich gute Korrelationen zu den anderen Lungenfunktionstests dargestellt werden konnten. Anhand einer Clusteranalyse wurde außerdem untersucht, ob ein alternatives Klassifikationsmodell möglich wäre und wie dieses aussehen könnte. Die Ergebnisse zeigten, dass vor allem in Kombination mit den neuen diagnostischen Verfahren, MBW und IOS, die qCTeine detaillierte Einteilung vornehmen kann und einen Mehrgewinn an Informationen liefert, welche für weitere Phänotypisierungen genutzt werden könnte. Hierdurch ergibt sich die Möglichkeit gezielte und individualisierte Therapien entsprechend der phänotypischen Ausprägungen zu entwickeln und ihr therapeutisches Ansprechen im weiteren Verlauf zu untersuchen
Zircon Petrochronology and Thermochemical Modeling of Quaternary Magmatic Systems in Western North America (The Geysers, Eastern Snake River Plain, Blackfoot)
No full textZircon (ZrSiO4) is an accessory mineral commonly present in a wide range of igneous, metamorphic, and sedimentary rocks. When crystallized in-situ, it is a powerful tool for reconstructing chemical and thermal processes in the Earth’s crust at high fidelity and with fine temporal resolution. In this dissertation, zircon from a range of mafic to silicic volcanic fields in the western United States was investigated by combining U-Th-Pb geochronology, trace element geochemistry, and δ18O and ɛHf isotopic data. This integrated petrochronological approach of studying zircon age and composition in combination with textural constraints yielded new insights into magma origins, evolution, and geothermal potential associated with Quaternary volcanism.
The first part of this study investigates zircon crystallization in the Eastern Snake River Plain (ESRP) and the adjacent Craters of the Moon volcanic field (COM) in southern Idaho. While the ESRP is dominated by voluminous olivine tholeiite lava flows with subordinate trachyandesite and rhyolite lavas, the COM field comprises mainly Fe-rich trachybasalts and rare evolved basaltic trachyandesites. In both settings, zircon is present in intermediate–silicic host rocks. High-resolution U-Pb and U-Th dating and trace element analysis confirm that zircon in these magmas formed autocrystically during protracted fractional crystallization, rather than being inherited from interstitial melts in a crystal mush. Equivalent evolutionary trends in zircon and whole-rock chemistry support this conclusion. Ultimately, zircon saturation in these systems was facilitated by strong enrichment of Zr via fractional crystallization. In the resulting high-Zr, compositionally intermediate ESRP and COM melts, zircon became saturated at unusually high temperatures and in comparatively primitive magma compositions. Since ESRP and COM zircon formed demonstrably in situ, zircon δ¹⁸O and εHf isotopic compositions are also reliable source indicators. These data suggest that ESRP and COM magmas originated from a similar mantle source, but subsequently evolved along different pathways. In the case of the ESRP, magma assimilated hydrothermally altered low-18O felsic rocks in the upper crust, whereas for the COM magmas, two sequential stages of assimilation occurred involving Archean lower crust and normal-δ¹⁸O upper crustal rocks, respectively. These findings underscore the importance of detailed petrochronological analyses of zircon and demonstrate that its presence in magmatic systems dominated by mafic melts has petrogenetic significance.
Building on this framework, the second paper shifts focus to the Geysers Plutonic Complex (GPC) in California, where a rare combination of deep plutonic and cogenetic volcanic rocks is accessible due to extensive geothermal drilling. The GPC and its volcanic equivalents represent a compositionally evolved intrusive complex emplaced between ~700 and nearly 4000 m depth, based on geothermal drill well penetration. Zircon in these rocks records a history of magma evolution marked by early formation of highly fractionated microgranite porphyry, followed by emplacement of larger volumes of less evolved granite and granodiorite. Trace element and isotopic data (δ¹⁸O and εHf) from zircon reveal multiple magma types with different degrees of crustal assimilation. Using previously reported zircon age distributions, a thermal model was developed to reconstruct the magmatic history of the GPC. The model relates zircon saturation to magma temperature and successfully reproduces the observed zircon age distributions for the GPC. In the best-fit model scenario, the main volume of the GPC intruded during a brief, intense flare-up phase lasting for c. 50,000 years. This magma pulse was embedded into a longer interval of low-flux magmatism which lasted for at least c. 900,000 years. The agreement between the modeled volume of the intrusion and the extrapolated GPC volume validates the potential of zircon-based thermal modeling for reconstructing upper crustal magma reservoirs and their thermal histories.
The final case study returns to southern Idaho and evaluates the geothermal potential of the Blackfoot Volcanic Field (BVF), a Pleistocene bimodal volcanic field located ~60 km southeast of the Snake River Plain-Yellowstone hotspot track. Despite hosting some of the youngest topaz rhyolites globally, the geothermal significance of this field remained uncertain. Zircon geochronology and geochemistry were used to constrain its eruptive and magmatic history. U-Pb and U-Th dating revealed two distinct eruptive phases: an early (1,006–785 ka) episode forming mainly the northern domes and a late (63–55 ka) pulse during which the southern domes erupted. Corresponding isotopic data for the BVF indicate a strong Archean crustal component in the parental magmas, consistent with their off-axis position relative to the hotspot. A thermal model for a 120 km³ magma reservoir emplaced 95,000 years ago and at 6 km depth successfully reproduces the timing of zircon crystallization and magma volumes for the southern domes and their subvolcanic equivalents. It also predicts that temperatures at ~4 km depth still exceed 300 °C today. The lack of surface geothermal manifestations within the BVF is attributed to a structurally complex hydrologic system that dilutes and diverts geothermal fluids.
Collectively, these three case studies illustrate the broad applicability of zircon petrochronology in characterizing the temporal, thermal, and compositional evolution of magmatic systems underneath Quaternary volcanic fields. Zircon saturation can occur in intermediate magma compositions associated with voluminous basaltic magmas, extending the potential of zircon petrochronology for refining models of magma generation, evolution, and emplacement to systems beyond those conventionally investigated. When paired with thermal modeling, zircon data can provide insights into the thermal state of crustal magmatic systems, including those with geothermal energy potential. This thesis demonstrates the integrative power of zircon as a tool for decoding crustal magmatism and guiding geothermal exploration