This study is focused on the geology and geochemistry of Midcontinent Rift-related intrusive and extrusive rocks present in northern Ontario, Canada. The study focuses on three sections in order to elucidate the geochemical characteristics of Midcontinent Rift-related rocks and investigate geochemical evolutionary signatures through time. The Coubran Lake basalts, the rocks of the Logan Basin and the intrusive rocks of the Nipigon Embayment were all included in this study to provide a sound spatial representation of Midcontinent Rift (MCR) intrusive and extrusive rocks present within Canada.

The Coubran Lake basalts within the Coldwell Complex, have provided new representation of Midcontinent Rift-related volcanism along the northeast shore of Lake Superior. Physical features, including proximity to surrounding syenites, as well as alteration features, suggest that the Coubran Lake basalts represent a pre-existing sequence into which the Coldwell Complex intruded. Geochemical evidence has linked the Coubran Lake basalt to the Two Duck Lake gabbro (TDLG) suggesting the basalts may represent the volcanic expression of the TDLG. Trace element and Nd isotope data suggest the Coubran Lake basalts are akin to the basalt type I composition, linking the unit to the Lower Siemens Creek volcanics, the basal units in the Ely's Peak and Grand Portage areas of the North Shore Volcanics, and to the lower suite of the Osler Group. The geochemical data presented here is consistent with the available paleomagnetic data suggesting the basalts were erupted early in the history of the Midcontinent Rift.

A re-evaluation of 2397 spatially referenced samples with associated whole-rock geochemistry has yielded previously unrecognized variation within the Midcontinent Rift-related Nipigon sills of the Nipigon Embayment. The whole rock major and trace element geochemistry of 796 Nipigon sill samples was re-examined to investigate the origin of the sills. This investigation revealed three distinct Nipigon sill types based on Th/Yb_pm ratios and ?_Nd1100 values: 395 Nipigon type I sill samples (Th/Yb_pm = 1.97 to 3.4; ε_Nd(t=1100Ma) = -0.5 to -1.5), 171 Nipigon type II sill samples (Th/Yb_pm = 3.4 to 5.0; ε_Nd(t=1100Ma) =  -1.5 to -3.0), and 55 Nipigon type III sill samples (Th/Yb_pm = 5.0 to 6.5; ε_Nd(t=1100Ma) = -5.0 to -7.0). Spatially, the distribution of the more contaminated Nipigon sill types II and III suggest centres which are inferred to represent zones of more contamination.

Field work in the Logan Basin has delineated the following timing requence between the various units present in the Logan Basin: the Riverdale sill, the Devon Volcanics, the Logan sills, the Pigeon River dykes, the Cloud River dykes, the Mount Mollie dyke and the Crystal Lake gabbro. Geochemical source characteristics support this timing relationship as the Riverdale sill, the Devon volcanic and the Logan sills are geochemically similar to the ultramafic intrusions of the Nipigon Embayment which were emplaced early in MCR history. The three dyke sets of the Logan Basin have been determined to be emplaced in the following order: Pigeon River dykes, the Cloud River dykes and the Mount Mollie dyke. All three dyke sets display similar source characteristics yet show stronger crustal contamination signatures (i.e., higher Th/Yb_pm and more negative ε_Nd(t=1100Ma) ) through time consistent with each dyke set having spent more time in the magma chamber through the duration of the emplacement sequence.

Based on the plots of Th/Yb_pm versus Nb/Th_pm and Th/Yb_pm versus Nb/Yb_pm, previously unrecognized relationships between units have allowed for inferences into the timing and geochemical evolution of the Midcontinent Rift. The progression of magmatism for both the Logan Basin and the Nipigon Embayment suggests that source characteristics became more depleted through time. Evidence of a link between the magmatism of the Logan Basin and the Nipigon Embayment is also indicated by the relationship between the Riverdale sill, the Devon Volcanics and the Logan sills with the mafic and ultramafic units of the Nipigon Embayment (excluding the Nipigon sills). Furthermore, the spatial link between the two magmatically endowed areas is bridged as the dykes of Sibley Peninsula showing evidence to be potential feeders to Nipigon sills.

For more details about this thesis contact Dr. Peter Hollings

A copy of the thesis can be downloaded here

The Thunder Bay Silver District is composed of two curvilinear groups of veins, the Mainland and Island Belts.  The veins can be mineralogically divided into the barren, silver-bearing, and 5-element association veins.

During the present study, fluid inclusion, sphalerite composition, and sulfur isotopic composition were investigated, as well as some aspects of the mineralogy of the veins.  Fluid inclusions in quartz, calcite, and fluorite indicate temperatures ranging from 70E-450EC during vein deposition could be found in veins of both geographical and all mineralogical groups.  Fluid inclusions in sphalerite indicate ore minerals were deposited between 80E-120EC at all locations, 1-30 equiv. wt. % NaCl.  The dominant salt in solution was CaCl₂.

Fe content in sphalerite indicates a log a₈₂ of -18 to -25 during ore deposition at all deposits except Spar Island .  At Spar Island copper sulfide mineralogy suggests log a₈₂ of -14 to -18.  Sphalerite often contained trace quantities of Cd.

Sulfur isotopic composition of sulfides and barite ranged from -9.7 to +12.2%, with two anomalous values (near +30%) from supergene samples.  Paired data indicated isotopic non-equilibrium during deposition sulfides.

The consistency of fluid inclusion, sulfur activity, and sulfur isotopic data suggest both depositional environment and hydrothermal fluid source were similar for all veins in the Thunder Bay district.  Heterogeneity of the deposits, particularly the presence or absence of Ag, Ni, Co and As may be due to the evolution of the fluids between their inception and entry into the depositional environment.

The source of the fluids is unknown, although their original sulfur isotopic composition is inferred to be fairly low (+4.7%), suggesting a magmatic source of sulfur is possible. 

A copy of the thesis can be found here

The Beardmore-Geraldton greenstone belt is located along the boundary between the Wabigoon and Quetico subprovinces in the Superior province of the Canadian Shield.  Historical gold mines and current gold exploration camps are located throughout the Beardmore-Geraldton greenstone belt and adjacent Wabigoon subprovince in moderately to steeply dipping ductile to brittle-ductile shear zones in many types of lithologies, with various metamorphic grades and alteration styles.  The shear zones typically strike east-west along the boundary of the Beardmore-Geraldton greenstone belt with the Wabigoon subprovince to the north as well as along the boundary with the Quetico subprovince to the south.

Field mapping and petrography, including microstructural analysis, were used to characterize the gold mineralization and investigate the control on gold mineralization in the Elmhirst, Castlewood, Pagwachuan, and Milestone properties in Greenstone, Ontario.  Two property-scale and seven trench-scale maps were produced from field mapping and one hundred twenty-three transmitted and reflected light thin sections were prepared from samples collected from these properties.  Lithology, structure, microstructure, metamorphic grade, alteration, and subprovince were considered as possible controlling factors on gold mineralization.

In the study area gold mineralization is hosted in relatively competent lithologies including schistose to mylonitic metamorphosed granodiorite, syenite, and pegmatite; chlorite-albite schist, schistose amphibolite, and garnet-hornblende-biotite schist.  The regional-scale structures that host gold mineralization are ductile to brittle-ductile shear zones that have a penetrative, schistose foliation and rarely a mylonitic fabric.  The property-scale structures that host gold mineralization are folded, mylonitized, and boudinaged quartz veins; shear-related early, late, and complex folds; folded fault breccia, and pressure shadows of rigid minerals within schistose to mylonitic fabric.  Gold mineralization is located in areas of micro-folds, micro-boudins, micro-fault breccia, undulatory extinction, and grain size reduction along irregular grain boundaries, subgrain boundaries, folded healed fractures, and in pressure shadows within areas of strain heterogeneity.  Gold mineralization is located in both areas of greenschist facies and amphibolite facies metamorphism in the study area.  The styles of alteration also vary within the study area with sericite, hematite, ankerite, calcite, and unaltered samples all hosting gold mineralization.  Both the Wabigoon and Quetico subprovinces host gold mineralization in similar structures and microstructures as the Beardmore-Geraldton greenstone belt although it is within different lithologies, metamorphic grades, and alteration styles.    

Similar structure and microstructure are always present in areas of gold mineralization in the study area while lithology, alteration, metamorphic grade, and subprovince vary.  Gold mineralization within stable prograde and retrograde metamorphic minerals indicates that gold mineralization was synchronous with ductile to brittle-ductile deformation and metamorphism.  Further gold exploration in the study area should focus on structure and microstructure as indicators for gold mineralization.

A copy of the thesis can be downloaded here

The Shakespeare intrusion, which hosts the Shakespeare Ni-Cu-PGE sulfide deposit, is part of the 2.2 Ga Nipissing Gabbro magmatic event, a large mafic-ultramafic suite located in the southern Superior and Southern Province. The Shakespeare intrusion intrudes the 2.2-2.45 Ga Huronian Supergroup and is located in the northern part of the Southern Province, only a few km from the Archean- Proterozoic unconformity which separates the Superior Province from the Southern Province and is adjacent to the ~ 2450 Ma East Bull Lake and Agnew Lake intrusions.

The Shakespeare deposit is the largest Ni-Cu-PGE deposit in the Nipissing Large Igneous Province (LIP). Field mapping and drilling have provided a detailed stratigraphic understanding of the deposit and provide the framework for geochemical models of sulfide segregation related to assimilation. The U-enriched rocks that interacted with the intrusion during ascent through the crust provide an excellent opportunity to gauge the role of assimilation during the formation of the Shakespeare deposit while providing further evidence that crustal sulfur is necessary in order to generate economic concentrations of magmatic sulfides.

A detailed study of three drill holes representing poor, average, and above average intercepts of Ni-Cu-PGE sulfides was performed  in order to establish the mechanism by which an immiscible sulfide liquid was produced and also to establish the contamination history of the Shakespeare deposit. Various geochemical techniques were employed, including SEM-EDS, ICP-MS, ICP-AES, total sulfur, sulfur isotope, and XRF analysis. During the course of the SEM­-EDS study primary magmatic textures within hollingworthite-irarsite-platarsite were observed, suggesting limited remobilization during metamorphism. Rheniite, cerium-rich allanite, and various tellurides and selenides were also observed.

δ³⁴S analysis yielded values between 0.01‰ and 2.38‰, averaging 1.14‰.  These values are indicative of an unfractionated sulfur source, suggesting that the source of sulfur in the Shakespeare deposit was either derived from the mantle or an unfractionated crustal source, such as the quartz pebble conglomerates of the Matinenda Formation. Sulfur: selenium ratios yielded values between 1245 to 3271, averaging  1810. The average value is significantly lower than accepted mantle values (2850-4000) and may point to an assimilant rich in selenium.

Spider diagrams of mantle normalized data from the Shakespeare intrusion indicate that all of the rocks of the Shakespeare intrusion display a geochemical profile which is characteristic of magmas that have undergone significant degrees of crustal contamination. Negative Nb and Ti anomalies, enrichment in HFSE relative to MREE, and enrichment in LREE  relative to HREE indicate that ascending  Shakespeare magmas assimilated significant quantities of upper crustal material.

The results of the study show that the Shakespeare intrusion is highly enriched in U, Th, and LREE, relative to other Nipissing intrusions. The most likely explanation for the enrichment is that the Shakespeare intrusion interacted with the pyritic quartz pebble conglomerates of the Matinenda Formation, possibly near the past producing Agnew Lake uranium mine. Geochemical and textural evidence suggests that magma mixing was the dominant mechanism responsible for the precipitation of an immiscible sulfide liquid. R factor calculations and sulfide metal concentrations suggest a second immiscible sulfide liquid precipitated from later mixing with the PGE- and base metal-poor biotite quartz diorite magmas which ponded in the roof zone of the intrusion.

The results of this study have major implications for Ni-Cu-PGE exploration in the Nipissing LIP and indicate that any economically significant concentrations of magmatic sulfides will be found in gabbroic rocks which have assimilated significant amounts of U- and Th-rich crustal material from the Matinenda Formation, resulting in discernable differences in the U-contents of potential host intrusions versus intrusions that are not likely to host a Ni-Cu­ PGE deposit.

A copy of the thesis can be downloaded here

The 1108 Ma Thunder mafic to ultramafic intrusion is a small, 800 x 100 x 500 m, Cu- PGE (platinum group element) mineralized body, located on the outskirts of Thunder Bay, Ontario, and associated with the early magmatic stages of the Midcontinent Rift (MCR) based on geochemical similarities to mafic and ultramafic rocks of the Nipigon Embayment and an unpublished ²⁰⁷Pb/²⁰⁶Pb zircon age of 1108.0 ± 1.0 Ma.  The Thunder intrusion is similar to the other known mineralized early-rift MCR intrusions; however, it is the only known occurrence hosted by the metavolcanic and metasedimentary assemblages of the Archean Shebandowan greenstone belt.  Major textural and geochemical differences can be used to subdivide the intrusion into a lower mafic to ultramafic basal unit and an upper gabbroic unit; the similar trace and rare earth element ratios of the two units suggest a single magmatic pulse that has undergone subsequent fractional crystallization and related cumulate phase layering.  The estimated parental composition of the Thunder intrusion has a mg# (MgO/(MgO + 0.9*FeO^T)) of 57 which represents a more evolved magma than other early-rift mafic to ultramafic  intrusions.  This may indicate the involvement of multiple staging chambers during the accent of the parent magma.

Trace and rare earth element patterns are consistent with an ocean island basalt source but with indications of contamination such as high Th concentrations and the presence of a negative Nb anomaly.  The ԐNd_T values from the intrusion range between - 0.7 and + 1.0, with no trends indicative of wall rock participation, whereas the ⁸⁷Sr/⁸⁶Sr_i ratios range from 0.70288 to 0.70611 and trend towards wall rock values of 0.70712 and 0.70873.  The weak correlation between ԐNd_T and ⁸⁷Sr/⁸⁶Sr_i is also a feature of the Nipigon Sills where it has been interpreted to be due to shallow-level crustal contamination whereas plots of MgO and SiO₂ versus ԐNd_T indicate contamination at depth by an older crustal source. 

Ni-Cu-PGE sulphide mineralization (20 m of 0.22 wt. % Cu, 0.06 wt. % Ni, 0.25 ppm Pt and 0.29 ppm Pd) is hosted by feldspathic peridotite in the lower mafic to ultramafic unit adjacent to the footwall of the Thunder intrusion.  Sulphides typically occur from 1 - 5 modal %, rarely up to 30 modal %, with textures ranging from medium- to fine-grained disseminated, globular and rarely net-textured.  Pyrrhotite, chalcopyrite and rare pentlandite with common secondary marcasite-pyrite replacement occur along with trace michenerite, kotulskite, merenskyite, sperrylite, hessite, electrum and argentian pentlandite.  Whole rock geochemical data displays fractionated Ni-Cu-PGE patterns with depletion of iridium-platinum group elements (IPGE) relative to platinum-platinum group elements (PPGE). 

Sulphide δ³⁴S values from the Thunder intrusion range from -2.0 to +3.8 ‰ and are similar to values for the metavolcanic host rock that range from -3.1 to +2.3 ‰.  Two samples from the basal mineralization zone sulphides yield Δ³³S values of 0.066 and 0.122 ‰ and one sample from the metavolcanic wall rock yields 0.149 ‰.  The δ³⁴S and Δ³³S values for the Thunder intrusion fall within range of typical upper mantle.  The sulphur source appears to be of mantle origin, however, the assimilation of crustal sulphur is a possibility but hard to distinguish as the wall rock S isotope and S/Se^T signature is similar to that of upper mantle. 

A copy of the thesis can be downloaded here

Paleosecular magnetic changes through a 1.5m section of post-glacial lacustrine deposits are strongly influenced by mineralogy and differential compaction.  The sediments chiefly comprise clay and, in the lower one third part of the section, rhythmites which vary from ~2mm to ~5cm in thickness. 125 paleomagnetic specimens were collected in total from the 1.5m section. Anisotropy of magnetic susceptibility and anisotropy of anhysteretic remanent magnetization identify a magnetic fabric with maximum susceptibility parallel to the bedding plane of the sediments and with a preferred northwest-southeast axis, probably indicating paleo-current alignment. Minimum susceptibility represents the pole to bedding, due to grain alignment.  

Incremental acquisition and demagnetization of isothermal and anhysteretic magnetizations and the orthogonal three-axis test indicate that the sediments contain two magnetic phases with different coercivities. Magnetic hysteresis measurements (clays n=226; silts n=37) show that clay is dominated by single domain magnetite and hematite (means of M_s=58.47±9.22Am², M_rs=17.12±27.22Am², H_c=21.09±7.69mT, H_cr=62.04±4.09mT) whereas silt is dominated by pseudo-single domain and single domain magnetite (means of M_s=681.0±395.9Am², M_rs=163.2±84.75Am², H_c=26.07±2.94mT, H_cr=56.08±3.17mT). The silt is dominated by magnetite, whereas the clay carries both hematite and magnetite.

Incremental alternating field demagnetization of clay, in 10 to 17 stages, successfully isolated three distinct remanence components, with different directions. A weak and spurious component that is due to handling magnetization, for example, was removed in early demagnetization steps. The two remaining components, isolated by further demagnetization, have geological significance. Mineral magnetic properties as well as differential compaction are responsible for the different directions of the two geologically significant magnetizations in the clay. One has intermediate coercivities (20 - 40mT) with a Fisher mean declination = 357.4° and Fisher mean inclination = 62.0°, approaching the present Earth's magnetic field (declination=356.7° and inclination=74.7°). The other has harder coercivities (>40mT) and represents a stable geological magnetization. The silt has a single component of magnetization carried by magnetite. Laboratory re-deposition of clay in large (125cm³) cubes accurately records the direction of the laboratory magnetic field; thus major inclination shallowing in natural sediments occurred subsequent to acquisition of original DRM.

A magnetostratigraphic record of the hard coercivity component over the 1.5 meter section reveals paleosecular variation of the Earth's magnetic field, with some complications due to compaction. A single westward drifting loop is recorded in the sediments, representing a time-period of circa 600 - 2400 years, with declinations varying from 199.8° to 119.9° (Fisher mean=005.2°) and inclinations varying from -2.4° to 77.4° (Fisher mean=50.9°). In comparison the inclination of the geocentric axial dipole (GAD) for the study location is +66°.

The Coldwell Alkaline Complex was emplaced as a Midcontinent Rift-related intrusion into Archean-aged rocks of the Schreiber-Hemlo Greenstone belt, centered around the town of Marathon, Ontario. Although the complex consists of predominantly alkaline rocks, numerous PGE-(Cu) mineralized gabbroic intrusions also comprise a substantial portion of the complex and are of considerable economic interest. The PGE-(Cu) mineralized sub-ophitic to ophitic gabbros within the Marathon deposit of the Eastern Gabbro Suite have been studied in detail by numerous workers, leading to many interpretations of genetic models for the emplacement of the gabbros and subsequent mineralization.

This study focuses on characterizing two trenches in the Boyer Lake district, located along the northeastern border of the complex that consist of predominantly PGE-(Cu) mineralized coarse-grained to pegmatitic sub-ophitic olivine gabbros as well as many smaller ultramafic intrusions and pyroxenites that are in contact with Archean footwall rocks. Of particular interest, is an area located on Trench 2 that consists of olivine- and plagioclase-rich olivine gabbros, and does not exhibit the typical sub-ophitic texture of olivine gabbros elsewhere on the trenches as well as in the Marathon deposit, rather poikilitic olivines that commonly contain thin partial rims of pyroxene. Additionally, this zone is characterized by enriched PGE mineralization, a chalcopyrite  bornite sulphide assemblage and geochemical signatures that are all characteristic features of the W-Horizon, which is a known lens of high-grade PGE mineralization within the Marathon deposit. Mineral chemistry data in conjunction with Sm-Nd and Sr isotope analyses suggest that the PGE-enriched olivine- and plagioclase-rich zone consists of very primitive intrusions, and likely represents some of the earliest, if not earliest emplaced rocks comprising the study area in contrast to more pyroxene-rich and less mineralized sub-ophitic gabbros. The chaotic nature of the intrusions, variable geochemistry, PGE grades and mineralization styles suggests that the genesis of the Boyer Showing may best fit with a previously hypothesized dynamic magma conduit model that has been postulated for the genesis of the gabbros of the Marathon deposit.

Boreal forest soils represent globally significant stores of carbon, and to a lesser degree nitrogen, yet there is little consensus about how they are altered by natural and anthropogenic disturbances. Disturbance by wildfire and forest harvesting is widespread, yet there are few investigations comparing the impacts of these disturbances on soil nutrient cycling, and even fewer on compound disturbance, such as salvage logging. A chronosequence approach was used to investigate the impacts of full-tree harvesting, wildfire, and compound disturbance i.e. post-fire salvage logging on soil carbon and nitrogen stores, pH and carbon to nitrogen ratios overtime. Stable carbon and nitrogen depth profiles were examined to identify the dominant mechanisms driving the observed trends following disturbance. The findings suggest that full-tree harvesting and wildfire impact carbon and nitrogen stores in fundamentally different ways, with stores being reduced after wildfire and increased after harvest. However, the differences observed were confined primarily to the organic and upper mineral soil, and the changes diminished with time. Results from this study suggest that impacts of post-fire salvage logging on carbon and nitrogen stores resembles that of fire, but that recovery is much slower making the practice potentially more harmful than harvesting and wildfire alone. Additional exploration into the long-term effects of compound disturbance on carbon and nitrogen stores is needed if salvage logging is to continue to become a justifiable practice. The driving mechanism of ¹³C enrichment with depth was inconclusive, with both mixing and kinetic fractionation during humification appearing to be equally dominant at these sites. Rayleigh distillation analysis indicated no differences in overall depth enrichment among the sites. Similarly, examining ¹⁵N enrichment profiles did not contribute to developing a process-based understanding of nitrogen cycling in these soils. This research demonstrates a need for further investigation into the usefulness of stable isotope depth profiles to indicate mechanisms responsible for storage change on a variety of soil types and textures following disturbance.

A kimberlite indicator minerals suite from above the DO18 kimberlite has been analyzed both texturally and chemically resulting in identified kimberlitic olivine, garnet, clinopyroxene, spinel, and ilmenite. Through a detailed glacial analysis of the Lac de Gras region containing DO18, the indicator suite is not limited to an origin from DO18. Northern kimberlite pipes have indications of kimberlite indicator trains following a southward ice flow direction, towards DO18. Textural evidence from the grains supports kimberlitic material that underwent minimal transport, as the grains are dominantly euhedral. Through major element chemical analysis, all minerals indicate an origin from a diamondiferous kimberlitic grains source. This suite is likely derived from a single kimberlite as the total chemical variation is limited for each mineral phase and along compositional trends. All indicators are a reflection of source from within the subcontinental lithospheric mantle beneath the Lac de Gras region, defined by both peridotitic and eclogitic compositions. Of the suite, garnet and olivine strongly indicate metasomatism to the subcontinental lithospheric mantle through increased titanium content and average forsterite value of 92, this corresponds to the rapid low angle Farallon plate subduction, dehydration, and partial melting. 

Musselwhite mine is situated in the North Caribou Lake greenstone belt within the North Caribou Terrane of the Superior Province, located approximately 500 kilometers north of Thunder Bay, Ontario.

The focus of this study was to classify the biotite at Musselwhite mine between the different economic zones of interest. Biotite from ten sampleswas chosen to be examined using petrography and quantitative mineral analysis with the scanning electron microscope (SEM). The biotite from Musselwhite mine is hosted in mafic metavolcanic rocks, classified as schists. The majority of the samples are comprised of biotite, hornblende, plagioclase, chlorite and calcite with minor amounts of tourmaline, apatite and quartz. Ilmenite is also present with minor amounts of pyrite, pyrrhotite, pentlandite and arsenopyrite. The biotite in the zones of interest show variation in iron content, with biotite plotting in the biotite and phlogopite zones on a metamorphic biotite graph that compare Fe/Fe+Mg and Al_Tot values.  

The samples examined have experienced amphibolite facies metamorphism and retrograde metamorphism due to the presence hornblende and chlorite, respectively.