Roger Nielsen - Academia.edu (original) (raw)
Papers by Roger Nielsen
Application of Computer Simulation of Magmatic Processes to the Teaching of Petrology
Journal of Geological Education, 1993
A series of computer programs called COMAGMAT-3.0 has been developed to simulate basaltic magma d... more A series of computer programs called COMAGMAT-3.0 has been developed to simulate basaltic magma differentiation processes. The programs are designed for use by igneous penologists to show the chemical consequences of magma crystallization. This package is a powerful tool that can be used to help students better understand how pressure, oxygen fugacity, and initial melt composition effect the resultant liquid lines of descent for 10 major and 20 trace elements. The model calculates the mineral proportions and compositions of olivine, plagioclase, pyroxenes and iron-ore phases crystallizing at different pressure and redox conditions. The effect of each parameter (including mineral-melt distribution coefficients) can be investigated numerically and output to the screen or a printer. Moreover, new methods of igneous rock interpretation can be applied to magmatic systems based on the package. The COMAGMAT programs operate in a user-friendly environment with the built-in help, and graphics support makes it usef...
News and Short Contributions
Journal of Field Archaeology, 1978
Petrogenesis of Luna 16 aluminous Mare basalts
Geophysical Research Letters, 1979
The Flagstaff Lake Igneous Complex is a member of an array of Acadian plutons which extend from t... more The Flagstaff Lake Igneous Complex is a member of an array of Acadian plutons which extend from the Katahdin batholith in the northeast to New Hampshire in the southwest. It is characterized by four major rock types: (1) gabbro, (2) granite, (3) garnet tonalite, and (4) trondhjemite. Field, petrographic, and geochemical evidence support the hypothesis that these diverse rocks existed as contemporaneous liquids, and that extensive mixing occurred at the time of emplacement. The dominant mafic mineral pairs in the gabbros are olivine-clinopyroxene, two pyroxenes, and pyroxenehornblende. The high Fe, low Ni and Cr contents of most of the gab bro sampled imply that it underwent substantial fractional crystallization prior to and during emplacement. The granitic rocks of the Flagstaff Lake Igneous Complex are heterogeneous, ranging from two-mica granites with equal amounts (-30%) of orthoclase, plagioclase, and quartz, to mafic diorites. These K-feldspar-bearing silicic rocks are exposed...
Lithos, 2015
Abstract Snow Peak, Oregon, is a moderate size basaltic shield volcano (50–52 wt.% SiO2, > 7.4... more Abstract Snow Peak, Oregon, is a moderate size basaltic shield volcano (50–52 wt.% SiO2, > 7.4 km3) located within the forearc of the Cascadia subduction zone, ~ 50 km west of the current arc front. Herein we present new whole rock geochemistry, mineral chemistry and 11 new 40Ar/39Ar ages, together with petrologic modeling that allow us to constrain the timing and origin of volcanism. In contrast to previous K–Ar ages that suggested volcanism occurred at ~ 3 Ma, our new 40Ar/39Ar ages show that Snow Peak formed between 5.3 and 6 million years ago. The volcano lies unconformably on ~ 30 Ma volcanic rocks of the Western Cascades. Volcanism occurred over a total duration of 14 wt.% MgO. Snow Peak lavas have trace element characteristics transitional between the calc-alkaline basalt (CAB) and low-K tholeiite (LKT) primary magma types recognized throughout the Cascade Range, but are closer to LKT and are classified as such. Estimates based on phase equilibria models and plagioclase hygrometers suggest that the primary magmas contained moderate amounts of water (1.5–2 wt.%), consistent with LILE/HFSE ratios that are greater than MORB values. Snow Peak is part of a widespread suite of LKT magmas that erupted between 5–8 Ma throughout the central Oregon Cascade Range in response to intra-arc rifting, and Snow Peak shows that LKT magmatism at this time extended well into the forearc of the central Oregon Cascade Range. Overall LKT magmas of this age occur over a distance of ~ 120 km normal to the arc front. We speculate that the presence of Snow Peak may relate to the western extension of the Brothers Fault Zone into the Cascade forearc.
Geochemistry and petrogenesis of the Belmont diabase sheet, northern Culpeper Basin, Virginia
Geol. Soc. Am., Abstr. Programs; (United States), 1985
The Belmont diabase sheet (BDS) is one of numerous hypabyssal tholeiitic intrusions common to the... more The Belmont diabase sheet (BDS) is one of numerous hypabyssal tholeiitic intrusions common to the early Mesozoic rift valleys of the mid-Atlantic states. Petrographic, microprobe and whole rock major and trace element analyses in conjunction with limited field and geophysical evidence indicate that the BDS is a westward dipping, irregular basin-shaped intrusive sheet of high-TiO/sub 2/ quartz-normative type diabase. The bulk of the exposed portion of the BDS is fractionated relative to the chill zone composition. A semi-circular outcrop pattern of relatively high MgO rocks surrounds the main body of the BDS. These high-MgO rocks share a distinctive petrography characterized by large, anhedral OPX grains often mantled by inverted pigeonite and are interpreted as a basal cumulate zone. Drill core samples collected from the western portion of the sheet penetrate 500' of fractionated diabase without intersecting the high-MgO layer or a lower chill margin, however,subhedral grains of inverted pigeonite appear in the lowermost core samples. This suggests that the cumulate layer underlies the sheet of >500' depth on the W side and crops out up-dip to the E, N and S where the sheet margins are turned upward. Mass balance calculations indicate that the exposed high-MgO rocks may be derivedmore » from the chill margin composition by accumulation of 13% OPX + 2% CPX + 5% PLAG (AN52) and the remaining rocks of the BDS are derived through fractional crystallization of varying amounts of CPX, OPX and PLAG. Magmatic features characteristic of similar tholeiitic intrusions such as rhythmic layering and a Mg-olivine layer were not observed. However, examination of the BDS and other hypabyssal intrusions offers insight into the limitations and circumstances under which these features may be developed.« less
Simulation of igneous differentiation processes
Reviews in Mineralogy & Geochemistry, 1990
One of the primary goals of igneous petrology is the definition and evaluation of the roles of th... more One of the primary goals of igneous petrology is the definition and evaluation of the roles of the processes responsible for chemical differentiation. Quantitative simulation of these processes not only requires an understanding of the systematics of mineral-liquid equilibria in natural systems but also some understanding of how different processes interact. The programs presented here consist of empirically calibrated expressions of mineral-melt equilibria that are built around a thermodynamic framework. The framework of the programs (which are designed to model phase equilibria based igneous differentiation) have been published in earlier works (Nielsen and Dungan, 1983; Nielsen, 1985; Nielsen, 1988 a, b; Nielsen et al., 1988). This presentation is to show how modeling can be used to gain otherwise unobtainable insights into the mechanics of igneous differentiation, in particular, for processes that are difficult to simulate in the laboratory. In this chapter we will simulate the major differentiation processes that are characteristic of shallow igneous magma chambers. These processes include crystallization, recharge, assimilation .and eruption. In each example, we will examine how observations of natural systems can be converted into the mathematical constraints necessary for the construction of computer based models. Since our primary goal is to model natural igneous systems, the examples show how phase equilibria modeling can be used to help solve problems related to interpretation of the chemical and mineralogic diversity of suites of natural lavas.
[
Pyroxene-melt equilibria. [for lunar maria basalts
Plagioclase Ultra-Phyric Basalts (PUBs): Implications for the nature of the plumbing system at ultraslow-spreading ridges
AGU Fall Meeting Abstracts, Dec 1, 2010
ABSTRACT Plagioclase ultra-phyric pillow basalts (PUBs) of the Southwest Indian Ridge contain up ... more ABSTRACT Plagioclase ultra-phyric pillow basalts (PUBs) of the Southwest Indian Ridge contain up to 35% anorthitic plagioclase crystals. Plagioclase crystals exhibit compositional and textural heterogeneities reflecting a history of complex crystallization conditions. Phenocrysts and megacrysts (> 0.5 mm) that range in composition from An 85 to An94 are too calcic to be in equilibrium with the host glass. Microlite ( 82) clinopyroxenes are found as inclusions in plagioclase and as heavily resorbed, isolated phenocrysts. Phase equilibrium simulations with MELTs using variable pressure conditions and magmatic H2O, CaO, and Na2O compositions demonstrate that high anorthite plagioclase can crystallize at crustal pressures from a hydrated melt (0.5 < wt% H2O) containing low Na2O. Crystal size distributions (CSDs) for plagioclase phenocrysts and megacrysts show linear population densities, consistent with steady crystal nucleation and growth and are similar for samples from different ridge segments. Crystal residence times are less than several hundred years based on crystal growth rates of 10-10 mm/sec. Plagioclase size populations, textures, and chemical results are interpreted to reflect open-system processing in the production of plagioclase ultra-phyric basalts. Primitive magmas emplaced into the upper crust (~1-4 km) may become hydrated by reaction with older, hydrothermally altered country rock. A magma chamber with gradients in temperature and composition can produce crystals that display variations in textures and morphology, including zoning, resorption, dissolution, and skeletal growth features. However, the comparable CSD results from along the ridge axis suggest the overall thermal environment of crystal growth is broadly similar between ridge segments.
Variations in magma transport recorded by plagioclase ultraphyric basalts: Preliminary results from SWIR, Blanco and Juan de Fuca
AGU Fall Meeting Abstracts, Dec 1, 2011
ABSTRACT Plagioclase ultraphyric basalts (PUBs) have been sampled at slow to intermediate spreadi... more ABSTRACT Plagioclase ultraphyric basalts (PUBs) have been sampled at slow to intermediate spreading oceanic centers worldwide. PUBs contain >15% (often anorthitic) plagioclase phenocrysts, with plagioclase making up > 90% of the phenocryst mode. The petrogenesis of PUBs has traditionally been attributed to inclusion of plagioclase from the crystal mush below spreading ridges. However, the conditions under which this occurs and the process of crystal sorting remain an enigma. To build a more complete model for PUB formation, we compiled published major and trace element data from the literature and PetDB for MORBs containing >15% plagioclase phenocrysts. While there is a clear connection between spreading rate and the occurrence of PUBs (more common at slow and intermediate rate ridges), we find that PUBs have no preferred erupted glass compositions. Therefore, we conclude that the generation of PUBs is dominantly a physical process rather than related to a specific magma type. One of our primary goals is to use information from the population of plagioclase phenocrysts to understand PUB magma differentiation and transport processes. In situ major and trace element and 87Sr/86Sr isotopic data were collected for plagioclase and its host glass from samples at the Southwest Indian Ridge (SWIR), Blanco Transform, and the Juan de Fuca Ridge in order to understand the relationship between the phenocrysts and their host lava suite. The plagioclase megacrysts record contrasting magma storage and transport conditions at different spreading ridges. Crystals from a single sample are often isotopically distinct from the magma they reside in and may or may not be distinct from other crystals in the same sample. Lavas from the East Blanco depression contain plagioclase phenocrysts that are more radiogenic than their host glass. Additionally, plagioclase-hosted melt inclusions have chemical signals that are more evolved than the ambient glass. This demonstrates that the plagioclase megacrysts at Blanco crystallized from a more enriched magma and are xenocrystic with respect to their current host liquid. Conversely, at SWIR, at least one sample contains plagioclase phenocrysts that are much less radiogenic than their host glass. While another dredged sample from SWIR has plagioclase phenocrysts that are mostly in isotopic equilibrium with their host liquid. In all samples, in situ trace elements in plagioclase exhibit a wide range of distinctive chemical characteristics. This suggests diverse magma transport conditions prior to being aggregated. Our findings indicate that PUBs often do not crystallize from simple genetic suites, but instead reflect the amalgamation of diverse melts and complex storage systems that exist below mid-ocean ridges. The presence of PUBs may be a distinctive characteristic of specific magma transport conditions, however, the boundary conditions for their formation is as yet unclear, and the subject of continued work.
Potential Consequences of the Compositional Distribution of Trace Element Partitioning Experiments
AGUFM, Dec 1, 2018
What We Have Learned About the Existing Trace Element Partitioning data During the Population Phase of traceDs
AGU Fall Meeting Abstracts, Dec 16, 2015
Phase Equilibria of Plagioclase Ultraphyric Basalts: Constraints from High-Pressure Experiments
GSA Annual Meeting in Seattle, Washington, USA - 2017, 2017
Goldschmidt Abstracts, 2020
CO 2 and H 2 O in Plagioclase-Hosted Melt Inclusions from Ocean-Ridge Lavas: An Indicator of Crystallization in the Lower Oceanic Crust
AGU Fall Meeting Abstracts, Dec 1, 2019
Traceds: An Experimental Trace Element Partitioning Database
AGU Fall Meeting Abstracts, Dec 1, 2014
Experimental Trace Element Partitioning Databases: Characteristics and Challenges
AGUFM, Dec 1, 2011
ABSTRACT The primary goal of trace element partitioning studies is to create a database that can ... more ABSTRACT The primary goal of trace element partitioning studies is to create a database that can be used to develop models of how trace elements behave in natural geochemical systems. Over the past several years, we have developed two independent databases, the first as part of the GERM initiative and the second (traceds) as a preliminary effort to create a database of experimental partitioning results. The GERM database is comprised of summaries of experimental and phenocryst/host results. Not included are estimates of error, or results of individual experiments. Our new effort (traceds) includes analytical results only from experiments. We have begun with the clinopyroxene/melt data, including metadata on experimental method, analytical technique, phase composition, temperature, and pressure. What is missing is spatial information and details on error estimates; we have essentially no spatial information on where the analysis was performed within the experimental charge. We also have a range of error estimates, from those based on counting statistics to those based on the standard deviation among multiple determinations. Conflicting or absent database attributes are just a few of the challenges we have encountered in developing and attempting to exploit these data. User base feedback on the existing databases has focused on specific general characteristics: Data quality - the degree to which reproducibility, precision, equilibration of phases, and completeness of analysis can be estimated. Data relevance - are the data relevant to the system I wish to model (appropriate T, P, X, etc.). Methodologies -what numerical approaches are available to generate models for a specific system. In an attempt to evaluate the internal consistency of the database, we began by applying the existing numerical methods for predicting partitioning behavior in clinopyroxene saturated systems (e.g., Wood and Blundy, 2002). Ideally, a predictive model will result in a linear correlation between the experimental and calculated values. In addition, any dependent parameter that is unaccounted for in the model will result in a correlation of that parameter with the difference between the calculated and experimental values. We examined the database and the models with respect to a wide range of compositional and intensive variables. The results indicate that there are few coherent dependencies on model error. However, there is substantial scatter that cannot be explained by any known combination of parameters. The observed scatter combines analytical, experimental and model error. Based on our results, including the absence of any apparent dependencies, the internal error in predictive accuracy is approximately 3 times the cited analytical and experimental error (on the order of 50%). This means that our database is either much less precise than we believe, or that the database includes important dependencies that we cannot see through the apparent analytical and experimental error. Development of this common resource becomes increasingly important as both the experimental database and the level of expertise required to apply the numerical constraints increase in quantity and complexity.
Use of Otolith Microchemistry to Discriminate Oncorhynchus nerka of Resident and Anadromous Origin
Canadian Journal of Fisheries and Aquatic Sciences, 1994
We used a wavelength dispersive microprobe to describe patterns in otolith microchemistry of sock... more We used a wavelength dispersive microprobe to describe patterns in otolith microchemistry of sockeye salmon (Oncorhyncbus nerka). Patterns in Sr/Ca ratios in transects across otoliths were consistent with changes in environmental chemistry associated with life history. The patterns we observed were consistent with those described for other salmonids. We found that Sr/Ca ratios in otolith primordia of fish of known anadromous origin were significantly higher than those in otolith primordia of fish of known resident origin. However, variation among samples was associated with differences in chemistry of the freshwater environments. These differences could confound attempts to discriminate sympatric resident and anadromous forms in lakes where Sr/Ca ratios are high. Samples used to discriminate origin of emigrants from Redfish Lake, Idaho, confirmed the presence of both resident and anadromous fish, but the two groups were not clearly resolved. Otolith microchemistry has the potential to determine the parental origin of O. nerka and to describe life history patterns, but may not clearly identify all individuals in all lakes. More work is needed to determine the inherent variability among individuals and among populations and to determine the influence that migration, spawning, and incubation environments have on the chemistry of otoliths.
Goldschmidt Abstracts, 2020
We study melt inclusions to gain insight into the character and provenance of primitive magmas. O... more We study melt inclusions to gain insight into the character and provenance of primitive magmas. Over the past 30 years, our community has generated a growing body of knowledge based on melt inclusion analysis. The majority of that data comes from inclusions hosted in olivine. However, there are limits to what we can learn from olivine hosted inclusions. Olivine hosted inclusions can only be used to understand deep seated processes if they make it to the surface, and the significant density contrast with basaltic magmas raises the prospect of sampling bias. In addition, olivine deforms quickly (~30 min.) if heated at low pressure [1], creating issues with regards to the migration of volatiles within the inclusion. MI in plagioclase megacrysts provide a relatively untapped resource with regards to the character of primitive magmas. The reason for the emphasis on olivine is based on a number of assumptions. First, that olivine is the exclusive liquidus phase for most primitive magmas, and that plagioclase is not normally in equilibrium with high Mg magmas. Second, plagioclase "leaks", resulting in loss of volatiles from the inclusions. Third, diffusive equlibration of inclusion and host prevents us from obtaining an unmodified vision of the melt at the time of entrapment. We present evidence from plagioclase megacrysts and their inclusions documenting that plagioclase can be used, together with data from other phases and the suite of erupted lavas, to obtain a broader view of the character of magmas present at depth [2, 3]. There are issues that need to be dealt with concerning crystal relaxation if the inclusions are homogenized at low pressure. However, our data indicates that plagioclase is a more robust host than olivine, is an important phase early in the crystallization sequence and is more likely to be sampled from depth due to the lower density contrast with basaltic magma.
Petrogenesis of Luna 16 aluminous Mare basalts
Geophysical Research Letters, Nov 1, 1979
Application of Computer Simulation of Magmatic Processes to the Teaching of Petrology
Journal of Geological Education, 1993
A series of computer programs called COMAGMAT-3.0 has been developed to simulate basaltic magma d... more A series of computer programs called COMAGMAT-3.0 has been developed to simulate basaltic magma differentiation processes. The programs are designed for use by igneous penologists to show the chemical consequences of magma crystallization. This package is a powerful tool that can be used to help students better understand how pressure, oxygen fugacity, and initial melt composition effect the resultant liquid lines of descent for 10 major and 20 trace elements. The model calculates the mineral proportions and compositions of olivine, plagioclase, pyroxenes and iron-ore phases crystallizing at different pressure and redox conditions. The effect of each parameter (including mineral-melt distribution coefficients) can be investigated numerically and output to the screen or a printer. Moreover, new methods of igneous rock interpretation can be applied to magmatic systems based on the package. The COMAGMAT programs operate in a user-friendly environment with the built-in help, and graphics support makes it usef...
News and Short Contributions
Journal of Field Archaeology, 1978
Petrogenesis of Luna 16 aluminous Mare basalts
Geophysical Research Letters, 1979
The Flagstaff Lake Igneous Complex is a member of an array of Acadian plutons which extend from t... more The Flagstaff Lake Igneous Complex is a member of an array of Acadian plutons which extend from the Katahdin batholith in the northeast to New Hampshire in the southwest. It is characterized by four major rock types: (1) gabbro, (2) granite, (3) garnet tonalite, and (4) trondhjemite. Field, petrographic, and geochemical evidence support the hypothesis that these diverse rocks existed as contemporaneous liquids, and that extensive mixing occurred at the time of emplacement. The dominant mafic mineral pairs in the gabbros are olivine-clinopyroxene, two pyroxenes, and pyroxenehornblende. The high Fe, low Ni and Cr contents of most of the gab bro sampled imply that it underwent substantial fractional crystallization prior to and during emplacement. The granitic rocks of the Flagstaff Lake Igneous Complex are heterogeneous, ranging from two-mica granites with equal amounts (-30%) of orthoclase, plagioclase, and quartz, to mafic diorites. These K-feldspar-bearing silicic rocks are exposed...
Lithos, 2015
Abstract Snow Peak, Oregon, is a moderate size basaltic shield volcano (50–52 wt.% SiO2, > 7.4... more Abstract Snow Peak, Oregon, is a moderate size basaltic shield volcano (50–52 wt.% SiO2, > 7.4 km3) located within the forearc of the Cascadia subduction zone, ~ 50 km west of the current arc front. Herein we present new whole rock geochemistry, mineral chemistry and 11 new 40Ar/39Ar ages, together with petrologic modeling that allow us to constrain the timing and origin of volcanism. In contrast to previous K–Ar ages that suggested volcanism occurred at ~ 3 Ma, our new 40Ar/39Ar ages show that Snow Peak formed between 5.3 and 6 million years ago. The volcano lies unconformably on ~ 30 Ma volcanic rocks of the Western Cascades. Volcanism occurred over a total duration of 14 wt.% MgO. Snow Peak lavas have trace element characteristics transitional between the calc-alkaline basalt (CAB) and low-K tholeiite (LKT) primary magma types recognized throughout the Cascade Range, but are closer to LKT and are classified as such. Estimates based on phase equilibria models and plagioclase hygrometers suggest that the primary magmas contained moderate amounts of water (1.5–2 wt.%), consistent with LILE/HFSE ratios that are greater than MORB values. Snow Peak is part of a widespread suite of LKT magmas that erupted between 5–8 Ma throughout the central Oregon Cascade Range in response to intra-arc rifting, and Snow Peak shows that LKT magmatism at this time extended well into the forearc of the central Oregon Cascade Range. Overall LKT magmas of this age occur over a distance of ~ 120 km normal to the arc front. We speculate that the presence of Snow Peak may relate to the western extension of the Brothers Fault Zone into the Cascade forearc.
Geochemistry and petrogenesis of the Belmont diabase sheet, northern Culpeper Basin, Virginia
Geol. Soc. Am., Abstr. Programs; (United States), 1985
The Belmont diabase sheet (BDS) is one of numerous hypabyssal tholeiitic intrusions common to the... more The Belmont diabase sheet (BDS) is one of numerous hypabyssal tholeiitic intrusions common to the early Mesozoic rift valleys of the mid-Atlantic states. Petrographic, microprobe and whole rock major and trace element analyses in conjunction with limited field and geophysical evidence indicate that the BDS is a westward dipping, irregular basin-shaped intrusive sheet of high-TiO/sub 2/ quartz-normative type diabase. The bulk of the exposed portion of the BDS is fractionated relative to the chill zone composition. A semi-circular outcrop pattern of relatively high MgO rocks surrounds the main body of the BDS. These high-MgO rocks share a distinctive petrography characterized by large, anhedral OPX grains often mantled by inverted pigeonite and are interpreted as a basal cumulate zone. Drill core samples collected from the western portion of the sheet penetrate 500' of fractionated diabase without intersecting the high-MgO layer or a lower chill margin, however,subhedral grains of inverted pigeonite appear in the lowermost core samples. This suggests that the cumulate layer underlies the sheet of >500' depth on the W side and crops out up-dip to the E, N and S where the sheet margins are turned upward. Mass balance calculations indicate that the exposed high-MgO rocks may be derivedmore » from the chill margin composition by accumulation of 13% OPX + 2% CPX + 5% PLAG (AN52) and the remaining rocks of the BDS are derived through fractional crystallization of varying amounts of CPX, OPX and PLAG. Magmatic features characteristic of similar tholeiitic intrusions such as rhythmic layering and a Mg-olivine layer were not observed. However, examination of the BDS and other hypabyssal intrusions offers insight into the limitations and circumstances under which these features may be developed.« less
Simulation of igneous differentiation processes
Reviews in Mineralogy & Geochemistry, 1990
One of the primary goals of igneous petrology is the definition and evaluation of the roles of th... more One of the primary goals of igneous petrology is the definition and evaluation of the roles of the processes responsible for chemical differentiation. Quantitative simulation of these processes not only requires an understanding of the systematics of mineral-liquid equilibria in natural systems but also some understanding of how different processes interact. The programs presented here consist of empirically calibrated expressions of mineral-melt equilibria that are built around a thermodynamic framework. The framework of the programs (which are designed to model phase equilibria based igneous differentiation) have been published in earlier works (Nielsen and Dungan, 1983; Nielsen, 1985; Nielsen, 1988 a, b; Nielsen et al., 1988). This presentation is to show how modeling can be used to gain otherwise unobtainable insights into the mechanics of igneous differentiation, in particular, for processes that are difficult to simulate in the laboratory. In this chapter we will simulate the major differentiation processes that are characteristic of shallow igneous magma chambers. These processes include crystallization, recharge, assimilation .and eruption. In each example, we will examine how observations of natural systems can be converted into the mathematical constraints necessary for the construction of computer based models. Since our primary goal is to model natural igneous systems, the examples show how phase equilibria modeling can be used to help solve problems related to interpretation of the chemical and mineralogic diversity of suites of natural lavas.
[
Pyroxene-melt equilibria. [for lunar maria basalts
Plagioclase Ultra-Phyric Basalts (PUBs): Implications for the nature of the plumbing system at ultraslow-spreading ridges
AGU Fall Meeting Abstracts, Dec 1, 2010
ABSTRACT Plagioclase ultra-phyric pillow basalts (PUBs) of the Southwest Indian Ridge contain up ... more ABSTRACT Plagioclase ultra-phyric pillow basalts (PUBs) of the Southwest Indian Ridge contain up to 35% anorthitic plagioclase crystals. Plagioclase crystals exhibit compositional and textural heterogeneities reflecting a history of complex crystallization conditions. Phenocrysts and megacrysts (> 0.5 mm) that range in composition from An 85 to An94 are too calcic to be in equilibrium with the host glass. Microlite ( 82) clinopyroxenes are found as inclusions in plagioclase and as heavily resorbed, isolated phenocrysts. Phase equilibrium simulations with MELTs using variable pressure conditions and magmatic H2O, CaO, and Na2O compositions demonstrate that high anorthite plagioclase can crystallize at crustal pressures from a hydrated melt (0.5 < wt% H2O) containing low Na2O. Crystal size distributions (CSDs) for plagioclase phenocrysts and megacrysts show linear population densities, consistent with steady crystal nucleation and growth and are similar for samples from different ridge segments. Crystal residence times are less than several hundred years based on crystal growth rates of 10-10 mm/sec. Plagioclase size populations, textures, and chemical results are interpreted to reflect open-system processing in the production of plagioclase ultra-phyric basalts. Primitive magmas emplaced into the upper crust (~1-4 km) may become hydrated by reaction with older, hydrothermally altered country rock. A magma chamber with gradients in temperature and composition can produce crystals that display variations in textures and morphology, including zoning, resorption, dissolution, and skeletal growth features. However, the comparable CSD results from along the ridge axis suggest the overall thermal environment of crystal growth is broadly similar between ridge segments.
Variations in magma transport recorded by plagioclase ultraphyric basalts: Preliminary results from SWIR, Blanco and Juan de Fuca
AGU Fall Meeting Abstracts, Dec 1, 2011
ABSTRACT Plagioclase ultraphyric basalts (PUBs) have been sampled at slow to intermediate spreadi... more ABSTRACT Plagioclase ultraphyric basalts (PUBs) have been sampled at slow to intermediate spreading oceanic centers worldwide. PUBs contain >15% (often anorthitic) plagioclase phenocrysts, with plagioclase making up > 90% of the phenocryst mode. The petrogenesis of PUBs has traditionally been attributed to inclusion of plagioclase from the crystal mush below spreading ridges. However, the conditions under which this occurs and the process of crystal sorting remain an enigma. To build a more complete model for PUB formation, we compiled published major and trace element data from the literature and PetDB for MORBs containing >15% plagioclase phenocrysts. While there is a clear connection between spreading rate and the occurrence of PUBs (more common at slow and intermediate rate ridges), we find that PUBs have no preferred erupted glass compositions. Therefore, we conclude that the generation of PUBs is dominantly a physical process rather than related to a specific magma type. One of our primary goals is to use information from the population of plagioclase phenocrysts to understand PUB magma differentiation and transport processes. In situ major and trace element and 87Sr/86Sr isotopic data were collected for plagioclase and its host glass from samples at the Southwest Indian Ridge (SWIR), Blanco Transform, and the Juan de Fuca Ridge in order to understand the relationship between the phenocrysts and their host lava suite. The plagioclase megacrysts record contrasting magma storage and transport conditions at different spreading ridges. Crystals from a single sample are often isotopically distinct from the magma they reside in and may or may not be distinct from other crystals in the same sample. Lavas from the East Blanco depression contain plagioclase phenocrysts that are more radiogenic than their host glass. Additionally, plagioclase-hosted melt inclusions have chemical signals that are more evolved than the ambient glass. This demonstrates that the plagioclase megacrysts at Blanco crystallized from a more enriched magma and are xenocrystic with respect to their current host liquid. Conversely, at SWIR, at least one sample contains plagioclase phenocrysts that are much less radiogenic than their host glass. While another dredged sample from SWIR has plagioclase phenocrysts that are mostly in isotopic equilibrium with their host liquid. In all samples, in situ trace elements in plagioclase exhibit a wide range of distinctive chemical characteristics. This suggests diverse magma transport conditions prior to being aggregated. Our findings indicate that PUBs often do not crystallize from simple genetic suites, but instead reflect the amalgamation of diverse melts and complex storage systems that exist below mid-ocean ridges. The presence of PUBs may be a distinctive characteristic of specific magma transport conditions, however, the boundary conditions for their formation is as yet unclear, and the subject of continued work.
Potential Consequences of the Compositional Distribution of Trace Element Partitioning Experiments
AGUFM, Dec 1, 2018
What We Have Learned About the Existing Trace Element Partitioning data During the Population Phase of traceDs
AGU Fall Meeting Abstracts, Dec 16, 2015
Phase Equilibria of Plagioclase Ultraphyric Basalts: Constraints from High-Pressure Experiments
GSA Annual Meeting in Seattle, Washington, USA - 2017, 2017
Goldschmidt Abstracts, 2020
CO 2 and H 2 O in Plagioclase-Hosted Melt Inclusions from Ocean-Ridge Lavas: An Indicator of Crystallization in the Lower Oceanic Crust
AGU Fall Meeting Abstracts, Dec 1, 2019
Traceds: An Experimental Trace Element Partitioning Database
AGU Fall Meeting Abstracts, Dec 1, 2014
Experimental Trace Element Partitioning Databases: Characteristics and Challenges
AGUFM, Dec 1, 2011
ABSTRACT The primary goal of trace element partitioning studies is to create a database that can ... more ABSTRACT The primary goal of trace element partitioning studies is to create a database that can be used to develop models of how trace elements behave in natural geochemical systems. Over the past several years, we have developed two independent databases, the first as part of the GERM initiative and the second (traceds) as a preliminary effort to create a database of experimental partitioning results. The GERM database is comprised of summaries of experimental and phenocryst/host results. Not included are estimates of error, or results of individual experiments. Our new effort (traceds) includes analytical results only from experiments. We have begun with the clinopyroxene/melt data, including metadata on experimental method, analytical technique, phase composition, temperature, and pressure. What is missing is spatial information and details on error estimates; we have essentially no spatial information on where the analysis was performed within the experimental charge. We also have a range of error estimates, from those based on counting statistics to those based on the standard deviation among multiple determinations. Conflicting or absent database attributes are just a few of the challenges we have encountered in developing and attempting to exploit these data. User base feedback on the existing databases has focused on specific general characteristics: Data quality - the degree to which reproducibility, precision, equilibration of phases, and completeness of analysis can be estimated. Data relevance - are the data relevant to the system I wish to model (appropriate T, P, X, etc.). Methodologies -what numerical approaches are available to generate models for a specific system. In an attempt to evaluate the internal consistency of the database, we began by applying the existing numerical methods for predicting partitioning behavior in clinopyroxene saturated systems (e.g., Wood and Blundy, 2002). Ideally, a predictive model will result in a linear correlation between the experimental and calculated values. In addition, any dependent parameter that is unaccounted for in the model will result in a correlation of that parameter with the difference between the calculated and experimental values. We examined the database and the models with respect to a wide range of compositional and intensive variables. The results indicate that there are few coherent dependencies on model error. However, there is substantial scatter that cannot be explained by any known combination of parameters. The observed scatter combines analytical, experimental and model error. Based on our results, including the absence of any apparent dependencies, the internal error in predictive accuracy is approximately 3 times the cited analytical and experimental error (on the order of 50%). This means that our database is either much less precise than we believe, or that the database includes important dependencies that we cannot see through the apparent analytical and experimental error. Development of this common resource becomes increasingly important as both the experimental database and the level of expertise required to apply the numerical constraints increase in quantity and complexity.
Use of Otolith Microchemistry to Discriminate Oncorhynchus nerka of Resident and Anadromous Origin
Canadian Journal of Fisheries and Aquatic Sciences, 1994
We used a wavelength dispersive microprobe to describe patterns in otolith microchemistry of sock... more We used a wavelength dispersive microprobe to describe patterns in otolith microchemistry of sockeye salmon (Oncorhyncbus nerka). Patterns in Sr/Ca ratios in transects across otoliths were consistent with changes in environmental chemistry associated with life history. The patterns we observed were consistent with those described for other salmonids. We found that Sr/Ca ratios in otolith primordia of fish of known anadromous origin were significantly higher than those in otolith primordia of fish of known resident origin. However, variation among samples was associated with differences in chemistry of the freshwater environments. These differences could confound attempts to discriminate sympatric resident and anadromous forms in lakes where Sr/Ca ratios are high. Samples used to discriminate origin of emigrants from Redfish Lake, Idaho, confirmed the presence of both resident and anadromous fish, but the two groups were not clearly resolved. Otolith microchemistry has the potential to determine the parental origin of O. nerka and to describe life history patterns, but may not clearly identify all individuals in all lakes. More work is needed to determine the inherent variability among individuals and among populations and to determine the influence that migration, spawning, and incubation environments have on the chemistry of otoliths.
Goldschmidt Abstracts, 2020
We study melt inclusions to gain insight into the character and provenance of primitive magmas. O... more We study melt inclusions to gain insight into the character and provenance of primitive magmas. Over the past 30 years, our community has generated a growing body of knowledge based on melt inclusion analysis. The majority of that data comes from inclusions hosted in olivine. However, there are limits to what we can learn from olivine hosted inclusions. Olivine hosted inclusions can only be used to understand deep seated processes if they make it to the surface, and the significant density contrast with basaltic magmas raises the prospect of sampling bias. In addition, olivine deforms quickly (~30 min.) if heated at low pressure [1], creating issues with regards to the migration of volatiles within the inclusion. MI in plagioclase megacrysts provide a relatively untapped resource with regards to the character of primitive magmas. The reason for the emphasis on olivine is based on a number of assumptions. First, that olivine is the exclusive liquidus phase for most primitive magmas, and that plagioclase is not normally in equilibrium with high Mg magmas. Second, plagioclase "leaks", resulting in loss of volatiles from the inclusions. Third, diffusive equlibration of inclusion and host prevents us from obtaining an unmodified vision of the melt at the time of entrapment. We present evidence from plagioclase megacrysts and their inclusions documenting that plagioclase can be used, together with data from other phases and the suite of erupted lavas, to obtain a broader view of the character of magmas present at depth [2, 3]. There are issues that need to be dealt with concerning crystal relaxation if the inclusions are homogenized at low pressure. However, our data indicates that plagioclase is a more robust host than olivine, is an important phase early in the crystallization sequence and is more likely to be sampled from depth due to the lower density contrast with basaltic magma.
Petrogenesis of Luna 16 aluminous Mare basalts
Geophysical Research Letters, Nov 1, 1979