GEOCHEMISTRY OF A VOLCANIC HYDROTHERMAL SYSTEM AT MOUNT SPURR, ALASKA

by

Laura Garchar A thesis submitted to the Faculty and the Board of Trustees of the Colorado School of Mines in partial fulfillment of the requirements for the degree of Masters of Science (Geology).

Golden, Colorado

Date

Signed: Laura Garchar

Signed: Richard F. Wendlandt, PhD Thesis Advisor

Golden, Colorado Date

Signed: John D. Humphrey, PhD Associate Professor and Department Head Department of Geology and Geological Engineering

ii ABSTRACT

Mount Spurr is an ice and snow-covered andesitic volcano located at the northern extent of the Aleutian arc in south central Alaska. Previous workers have identified a prospective geothermal resource on the volcano’s south side. This research aims to characterize more fully the hydrothermal system and builds on published geological mapping, geophysical surveys, soil sampling, exploration drilling, and water sampling.

Aqueous and mineralogical geochemical investigations at a field area on the south flank of Mt. Spurr focus on the possible extent of the hydrothermal system, its temperature at depth, the origin of waters, and location of fluid pathways in the system.

Three spatially distinct water compositions have been identified: 1) moderate temperature (˜50￿), acidic, bicarbonate-poor, δ18O and δD-enriched waters from Crater Lake; 2) low temperature (14-40￿), neutral, bicarbonate and Mg-rich, slightly δ18O and δD-enriched peripheral waters from Crater Canyon, Kid Canyon, and South Spurr; and 3) cold (<9￿), neutral, sulfate-rich, dilute/meteoric waters from Chaka Ridge and South Spurr. Geochemical modeling and stable hydrogen, oxygen, and carbon isotopes have confirmed that water chemistry cannot be explained solely by meteoric water-andesite interaction, or by mixing of Crater Lake and meteorically derived waters. Contributions to chemistry must come at least in part from a magmatic system. No alteration minerals were observed at the surface near springs, but kaolinite likely lies along the flow path of waters in the subsurface. Spring and seep waters are derived primarily from shallowly circulating meteoric water that has variously interacted with rock, incorporated condensed steam, in some cases mixed with andesitic water, and been diluted by rain and snowmelt overall. Montmorillonite-rich conglomerates and sandstones of the West Foreland Formation were encountered in exploration core holes located in the eastern field area. Post-depositional changes including the formation of montmorillonite from tuff, the presence of replacement-style calcite, as well as compaction and faulting processes are observed in hand samples and thin sections of 26-11 core. Stable carbon and oxygen isotopes of calcite samples are similar to published calcite isotopes from Cook Inlet sedimentary rocks. If all the calcite examined in this research formed from a fluid of the same isotopic composition, a 15‰ fractionation happened over ˜30￿. The calcite samples are interpreted to be of meteoric-dominated origin and to have formed at low temperature diagenetic conditions. The sedimentary rocks of the West Foreland Formation have low visual porosity, and no evidence of recent hydrothermal alteration has been observed.

Surface discharges do not seem to represent rapidly upwelling, deep, hydrothermal fluid. It appears that meteoric water doesn’t circulate to depth and that deep water doesn’t make it to the surface, except locally.

The transport of hydrothermal fluids to the surface seems to be controlled by faults. However, not all

iii mapped faults are conduits for deep fluid flow. The montmorillonite of the West Foreland Formation could be acting as a barrier to deep circulation of meteoric water, and could prevent upwelling of deeply sourced hydrothermal fluids. Montmorillonite could also be the source of resistivity anomalies in the field area.

Exploration drilling has not penetrated the entire thickness of the West Foreland Formation, and what lies below is unknown. If a fractured granite basement that contains hydrothermal fluids exists beneath the West

Foreland Formation, it could be a viable geothermal reservoir. Precise mapping of fault structures that act as conduits to fluid flow and deeper drilling, especially near Crater Canyon, will enhance the understanding of a possible geothermal resource at Mt. Spurr.

iv TABLE OF CONTENTS

ABSTRACT ...... iii

LIST OF FIGURES ...... viii

LIST OF TABLES ...... xiii

ACKNOWLEDGMENTS ...... xiv

CHAPTER 1 MOTIVATION ...... 1

1.1 Geothermal Resources of Alaska ...... 1

1.2 Geothermal Resource at Mt. Spurr ...... 5

CHAPTER 2 BACKGROUND ...... 6

2.1 Regional Geologic History ...... 6

2.1.1 Paleozoic ...... 8

2.1.2 Mesozoic ...... 8

2.1.3 Cenozoic ...... 9

2.1.4 Recent Faulting ...... 10

2.2 Eruptive History of Mt. Spurr ...... 11

2.3 Previous Geothermal Investigations ...... 12

CHAPTER 3 METHODS ...... 20

3.1 FieldSampling ...... 20

3.2 Analytical Chemistry ...... 23

3.3 Petrography of Drilled Core ...... 26

3.3.1 Clay Mineral Characterization ...... 26

3.3.2 Calcite Mineral Characterization ...... 27

CHAPTER 4 RESULTS ...... 30

4.1 Aqueous Geochemistry ...... 30

4.2 Petrography of Conglomerate and Sandstone Unit in Drilled Core ...... 30

4.3 Whole Rock and Clay Mineral Characterization ...... 33

v 4.4 Isotopes of Calcite in Veins, Rims, and Cements ...... 35

CHAPTER 5 INTERPRETATION: AQUEOUS AND GAS GEOCHEMISTRY ...... 46

5.1 Dissolved Chemical Species ...... 46

5.1.1 Cl-HCO3-SO4 ...... 48

5.1.2 Na-K-Mg ...... 50

5.1.3 Cl-Li-B ...... 52

5.1.4 Chloride vs. Enthalpy ...... 53

5.1.5 Silica Saturation ...... 54

5.2 Stable Isotopic Composition ...... 55

5.2.1 Origin of δ18O and δDinMt.SpurrWaters...... 56

5.2.1.1 Hot Water Fraction of Crater Canyon Discharges ...... 57

5.2.1.2 Elevation and Chloride Concentration Effects ...... 59

5.2.2 Origin of δ13CinMt.SpurrWaters ...... 60

5.2.3 Origin of δ34S in Mt. Spurr Waters ...... 61

5.3 Gas Geochemistry from Crater Peak Fumaroles ...... 63

5.3.1 N2-He-Ar ...... 63

5.3.2 CO2-CH4-N2 ...... 65

5.4 Geothermometry Calculations ...... 65

5.4.1 Solute Geothermometer Calculations ...... 65

5.4.2 Isotope Gethermometer Calculations ...... 67

5.4.3 Gas Geothermometer Calculations ...... 67

5.5 Summary of Aqueous and Gas Geochemistry ...... 69

CHAPTER 6 INTERPRETATIONS: MINERALS AND MINERAL RELATIONSHIPS IN DRILLED CORE ...... 73

6.1 Unit Identification ...... 73

6.2 Diagenetic Changes ...... 74

6.2.1 Ductile Deformation ...... 74

6.2.2 Authigenic Clay ...... 74

vi 6.2.2.1 Montmorillonite ...... 74

6.2.2.2 Chlorite ...... 75

6.2.2.3 Laumontite ...... 75

6.2.3 Calcite ...... 76

6.2.4 Porosity and Permeability ...... 77

CHAPTER 7 DISCUSSION ...... 80

7.1 Water-Rock Interaction ...... 80

7.1.1 Models Involving Crater Lake Water, Glacial Melt, and Andesite ...... 80

7.1.2 Models Involving Na-K-Ca-Mg Aluminosilicates and Water at Various Temperatures . . 84

7.2 Conceptual Cross-Sections ...... 89

7.2.1 Temperature Gradient ...... 92

7.3 Comparison to Producing Geothermal Systems in Similar Settings ...... 92

7.3.1 Southeast Asia ...... 93

7.3.2 Central America ...... 94

7.3.3 Summary ...... 95

CHAPTER 8 CONCLUSIONS AND FUTURE WORK ...... 96

8.1 Possible Geothermal System ...... 96

8.1.1 Extent of Hydrothermal System ...... 96

8.1.2 Origin of Waters ...... 96

8.1.3 FluidPathways...... 97

8.1.4 Deep temperature and Reservoir Characteristics ...... 97

8.2 Implications for Future Geothermal Exploration and Development ...... 97

REFERENCES CITED ...... 99

vii LIST OF FIGURES

Figure 1.1 Geothermal resources of Alaska are indicated by areas of red shading. Nearly all of the Aleutian arc volcanoes fall into this category. Data from U.S. Department of Energy, figure modified from INEL...... 2

Figure 1.2 Mt. Spurr is located in south-central Alaska in the Tordrillo Mountains on the northwest side of the Cook Inlet about 100km west of Anchorage. Regional faults and geographic features are labeled...... 3

Figure 1.3 Aerial photo of Mt. Spurr looking north. Crater Peak is a parasitic cone on the south side of the volcano. Kidazgeni and Crater glaciers issue from the volcano’s summit. Sampling locations for this research include Crater Canyon, Chaka Ridge, and a few places to the south of Mt. Spurr summit between Crater Peak and the Chakachatna River. Figure modified from Waythomas & Nye...... 4

Figure 2.1 Timeline showing the geologic evolution of the Cook Inlet Basin, as well as volcanic activity at Mt. Spurr and the timing of water sampling campaigns...... 6

Figure 2.2 Stratigraphic column of sediments in Cook Inlet basin, showing basement (Triassic and older), marine deposits (Jurassic-Cretaceous), and basin fill (Tertiary-present). The Eocene West Foreland Formation unconformably overlies the Chickaloon Formation, and is unconformably overlain by the Hemlock Conglomerate...... 7

Figure 2.3 Geologic map of Mt. Spurr by Nye et al. that shows andesitic volcanic deposits from Mt. Spurr and Crater Peak eruptions, as well as Alaska-Aleutian Range batholith granites, regional sedimentary units, Holocene deposits, and glaciers/perennial snow. . . . 13

Figure 2.4 Mercury and helium soil gas surveys at Mt. Spurr by Turner & Wescott...... 17

Figure 2.5 1D inversion-resistivity (MT) results at Mt. Spurr in plan view. The red lines represent the Ormat lease boundaries, and the black lines represent mapped faults. Locations of MT stations shown by small triangles in both maps. Conductive anomalies in ‘warmer’ purple-red colors. Figures from Martini et al...... 18

Figure 2.6 3D fence diagram of CSAMT data. Black areas are zones of resistivity ¡5￿m. Figure from Turner & Wescott. Transects are shown in map view by the blue lines in Figure2.7...... 19

Figure 2.7 Location of CSAMT survey lines (blue, from Turner & Wescott) overlayed on elevation contours and the geologic map of Nye et al.. Bold black lines represent faults from Martini et al.. The colored symbols are water sampling locations and are explained in thenextchapter...... 19

Figure 3.1 Sampling locations shown with faults by Martini et al., and geologic map by Nye et al...21

Figure 3.2 Sample sites in Crater Canyon ...... 22

Figure 3.3 Sample sites at Chaka Ridge ...... 24

Figure 3.4 Sampling sites in the South Spurr area...... 28

viii Figure 4.1 Core drilled over the summer of 2011 in hole 62-11 consists mainly of conglomerate and sandstone units. Some calcite mineralization is seen along fracture surfaces...... 36

Figure 4.2 Alteration in clasts and matrix of conglomerate unit in core hole 26-11...... 37

Figure 4.3 Chlorite replaces primary igneous minerals in sample “3434” conglomerate clast. Chlorite and calcite also fill a void that has been cut by a calcite vein. Photomicrograph taken in plane light...... 38

Figure 4.4 Masses of a brown clay mineral intergrown with calcite are common. a.) The clay mineral and calcite in conglomerate matrix and b.) a close up of the intergrown texture, both in sample “2630.” c.) Some of the original vesicular texture is preserved in this replacement from the conglomerate matrix in “2227.” d.) Calcite and the clay mineral in conglomerate matrix in sample “3419.” The shapes of these calcite and clay mineral masses also indicate compaction. All photomicrographs were taken with crossed polars. . 39

Figure 4.5 a.) Calcite fills voids around conglomerate clasts and b.) replaces primary textures of conglomerate clasts in sample “2630.” Calcite also occurs c.) as veins that cross-cut matrix grains in “3419”, and d.) in the pore space between matrix grains in “682”. All photomicrographs taken with crossed polars...... 40

Figure 4.6 a.) Epidote replaces plagioclase feldspar in conglomerate clasts in “682”, and b.) in matrix grains of “3557”. Both photomicrographs were taken with crossed polars. . . . . 40

Figure 4.7 Evidence of mechanical stress includes a.) and b.) fractured quartz grains with calcite in fractures from “3434” conglomerate matrix, and c. and d.) biotite grains that have been smashed around other grains in the matrix of “3557” and “2227” respectively. The quartz grains could indicate faulting, whereas the biotite grains could indicate compaction. All photomicrographs were taken with crossed polars...... 41

Figure 4.8 XRD reflections for sample “1865”, with 2Θ (degrees) along the x-axis, and intensity (counts per second) along the y-axis...... 42

Figure 4.9 XRD reflections for sample “1889”, with 2Θ (degrees) along the x-axis, and intensity (counts per second) along the y-axis...... 42

Figure 4.10 XRD reflections for sample “2061.5”, with 2Θ (degrees) along the x-axis, and intensity (counts per second) along the y-axis...... 43

Figure 4.11 XRD reflections for sample “2580.5”, with 2Θ (degrees) along the x-axis, and intensity (counts per second) along the y-axis...... 43

Figure 4.12 XRD reflections for sample “2583”, with 2Θ (degrees) along the x-axis, and intensity (counts per second) along the y-axis...... 44

Figure 4.13 XRD reflections for sample “3221”, with 2Θ (degrees) along the x-axis, and intensity (counts per second) along the y-axis...... 44

Figure 4.14 XRD reflections for sample “3434”, with 2Θ (degrees) along the x-axis, and intensity (counts per second) along the y-axis...... 45

Figure 4.15 XRD reflections for sample “3434” calcite and red mineral scraping, with 2Θ (degrees) along the x-axis, and intensity (counts per second) along the y-axis...... 45

ix Figure 5.1 A Cl-HCO3-SO4 diagram shows that most of the samples from Mt. Spurr have high sulfate and bicarbonate concentrations relative to chloride concentrations, which indicates that shallow processes are inflencing the chemistries of the samples. A linear trend from 1970 Crater Lake waters (orange diamonds) to Crater Canyon waters (red circles) suggests a possible mixing trend. The more sulfate-rich composition of the remaining samples could be due to the very dilute composition of samples or the addition of sulfate through steam heating. Samples CP92-1 and CP92-2 plot in the same location...... 49

Figure 5.2 Na-K-Mg diagram showing that waters from Mt. Spurr cluster in the Mg corner, as would be expected for cold groundwater that has poorly equilibrated with rock...... 51

Figure 5.3 Cl-Li-B diagram showing that Mt. Spurr waters can be segregated into two general groups: high Li waters that have obtained Cl and B by absorption of magmatic volatiles from a deep hydrothermal system, and low Li waters that do not represent deeply-derived fluids...... 52

Figure 5.4 Chloride vs. enthalpy figure modified from Fournier, showing the hypothetical composition of steam and effects of cold water mixing on a deep reservoir fluid...... 53

Figure 5.5 Enthalpy-chloride plot with linear trend between South Spurr and Crater Canyon samples...... 54

Figure 5.6 Sampling temperature vs. silica concentration diagram with the solubility curves for different silica phases. Chalcedony is probably the dominant form. Extrapolating observed silica concentrations to the chalcedony solubility curve yields temperature estimates higher than those measured in the field...... 55

Figure 5.7 Stable oxygen and hydrogen isotopes in Mt. Spurr waters. Black diamonds represent cold stream/snow waters from Motyka & Nye. South Spurr and Chaka Ridge samples plot close to the meteoric water line and the meteoric waters, while some Crater Canyon samples (red circles) show a slight enrichment in δ18O. A linear trend between Crater Canyon samples eventually intersects the range of “andesitic water” . The isotopic composition of Crater Canyon waters could be produced by mixing of meteoric water with “andesitic water”. The isotopic composition of Crater Lake waters (orange diamonds) does not lie on this trend, and likely reflects evaporation processes...... 56

Figure 5.8 Conceptual model of hydrothermal circulation between Crater Peak and thermal springs at Mt. Spurr. Residual magma at unknown depth expels steam, CO2,SO2, HCl and other magmatic gases into overlying hydrothermal layer that was formed by infiltration and accumulation of groundwater. Steam and gases rich in CO2 and H2S are driven off the boiling hydrothermal layer and feed fumaroles at Crater Peak and interact with shallow groundwaters to produce HCO3-SO4-Cl thermal springs at lower elevations. From Motyka & Nye...... 58

Figure 5.9 Stable isotopic composition of Spurr waters with respect to elevation. Crater Lake samples are more enriched than would be expected, suggesting the addition of light isotopes from rock dissolution and clay minerals, or andesitic water...... 59

Figure 5.10 Stable isotopic composition with respect to Cl concentration, suggesting mixing low-Cl, heavier isotopic composition meteoric water. The Cl concentrations of 1992 Crater Lake waters are lower than would be expected for a relatively enriched isotopic composition, suggesting the addition of light isotopes from rock dissolution and clay minerals, or andesitic water...... 60

x Figure 5.11 Stable C and S isotopes from Mt. Spurr water samples. Gray boxes represent ranges for C and S in the environment from Field & Fifarek ...... 62

Figure 5.12 N2-He-Ar ternary diagram for examining source of noble gases. “RM Spurr 2” plots near the composition of air but has a slightly elevated He content...... 64

Figure 5.13 CO2-CH4-N2 ternary diagram showing Crater Peak fumarole compositions plot in the field of low CH4 magmatic or oxidized gases as opposed to hydrothermal gases...... 64

Figure 5.14 Geothermometer calculations. The ranges represent variation in temperatures within the geothermometer, i.e. the “silica” range represents the results of amorphous silica, quartz steam loss, quartz no steam loss, chalcedony, and cristobalite equations. The temperatures estimated by the various geothermometers don’t agree, and this is because Mt. Spurr waters have lower sodium and silica, but higher calcium and magnesium than would be expected from a deep fluid...... 68

Figure 5.15 Gas geothermometers. The parameter “r” represents the gas/steam ratio of the indicated species. Figures produced using spreadsheets set up by Powell...... 70

Figure 6.1 Stable oxygen and carbon isotopes from veins and cements of the West Foreland Formation encountered in core hole 62-11 (this research) are plotted with calcite veins and cements sampled by Bruhn et al. in the Cook Inlet. Water samples and a hypothetical calcite composition in equilibrium with water samples are also plotted. . . . 78

Figure 7.1 Piper diagram showing surface discharge compositions (same symbols as previously described) as well as reaction paths (black lines) representing water-rock interaction between meteoric water (“945”) and andesite alteration products at a 1:25 ratio at a constant temperature of 0￿...... 81

Figure 7.2 Piper diagram showing surface discharge compositions (symbols same as previously described) as well as reaction paths (black lines) representing water-rock interaction between Crater Lake water and andesite alteration products at a 1:1 ratio along a reaction path that starts at 49￿ and ends at 0￿...... 82

Figure 7.3 Piper diagram showing the reaction path as the results of Figure 7.1 and Figure 7.2 are mixed with a ratio of 1:1 at a constant temperature of 0￿. The anion ternary reaction path is represented by a point at the location of 70AR-202...... 83

Figure 7.4 Diagrams for the Na2O-K2O-Al2O3-H2O system in the presence of silica from 0-100￿. Axes represent activity ratios...... 86

Figure 7.5 Diagrams for the Ca2O-K2O-Al2O3-H2O system in the presence of silica from 0-100˚C. Axes represent activity ratios...... 87

Figure 7.6 Diagrams for the MgO-Al2O3-H2O system in the presence of silica from 0-100˚C. Axes represent activity ratios...... 88

Figure 7.7 Location of conceptual cross-section lines (purple), overlayed on elevation contours and the geologic map by Nye et al.. Bold black lines are faults , and water sampling locations are shown with the same symbols...... 90

xi Figure 7.8 Conceptual N-S cross-section A-A’ of field area from Mt. Spurr summit to the Chakachatna River. Seeps in Crater Canyon are fault-controlled and represent shallowly circulating meteoric water that seems to be mixing with Crater Lake or andesitic water...... 91

Figure 7.9 Conceptual NW-SE cross-section B-B’ of field area from Mt. Spurr summit through Chaka Ridge. Seeps on Chaka Ridge are topographically controlled and represent shallow circulation of meteoric water through porous volcanics and minor incorporation of steam...... 91

Figure 7.10 Conceptual E-W cross-section C-C’ of field area from west of Crater Glacier to east of Chaka Ridge in the South Spurr area. Upwelling related to discharges at Crater and Kid Canyons could extend further south along faults. This figure represents a perspective where hydrothermal outflow is coming toward the viewer...... 92

Figure 7.11 Temperature gradients of Mt. Spurr, other volcanoes, and Cook Inlet sediments. Drilling at Mt. Spurr indicates a slightly higher temperature gradient than the Cook Inlet, but is lower than most developed geothermal systems. However, exploration drilling at Mt. Spurr has only gone to ˜1km depth and the constant influx of near-freezing surface waters likely overwhelms any thermal signature at shallow depths. . 93

xii LIST OF TABLES

Table 2.1 Summary of water sampling campaigns at Mt. Spurr. Asterisks denote samples that were not considered in this research because the data are questionable due to over-acidification of samples in the field. Samples from West Spurr are also not considered because they lie outside the field area of this research. References are as follows: [1] Martini et al. [2] Motyka & Nye [3] Neal et al. [4] Keith et al...... 15

Table 3.1 Summary of core samples and the analyses conducted on them. All samples are from core hole 26-11. TS=thin section, Clay, Calcite, FeOx= characterization of minerals by X-ray diffraction (XRD), and Iso=stable C and O isotopes of calcite...... 29

Table 4.1 Field Measurements. Flow rate estimated visually...... 30

Table 4.2 Water chemistry for samples at Mt. Spurr collected in this research. Concentrations of cations, anions, and major species are reported in mg/L, alkalinity in (mg/L as CaCO3), and conductivity in (￿mhos/cm). Isotopes are reported in units of per mil (‰). Oxygen and hydrogen isotopes are relative VSMOW, sulfur relative to VCDT, and carbon relative to VPDB. Green samples are located in the South Spurr field area, red samples are from Crater Canyon, and blue samples are from Chaka Ridge...... 31

Table 4.3 Published water chemistry at Mt. Spurr. Concentrations of cations, anions, and major species are reported in mg/L, alkalinity in (mg/L as CaCO3), and conductivity in (￿mhos/cm). Isotopes are reported in units of per mil (‰). Oxygen and hydrogen isotopes are relative to VSMOW, sulfur relative to VCDT, and carbon relative to VPDB. Green samples are located in the South Spurr field area, red samples are from Crater Canyon, blue samples are from Chaka Ridge, samples with asterisks are from Crater Lake, and black samples are from Kid Canyon. References are as follows: [1] [2] [3] [4] . 32

Table 4.4 X-Ray Diffraction Results ...... 34

Table 4.5 Stable oxygen and carbon isotopes of calcite samples from 62-11 core...... 35

Table 5.1 Hot water fraction calculations using δ18O values of Crater Canyon springs and seeps. . . . 58

Table 5.2 Residual gas composition (dry basis, mole %) from fumaroles at Crater Peak, 1982, at a sampling temperature of 94°C. The water mole percent of RM Spurr 1 and RM Spurr 2 was not determined, but is 97.9% for RM Spurr 3. From Motyka & Nye...... 63

Table 5.3 Aqueous geothermometers. Concentrations of chemical species are in mg/kg unless otherwise specified...... 66

Table 5.4 Summary of results and interpretations of aqueous and gas geochemistry...... 71

Table 7.1 Alteration minerals and amounts are based on likely SVC alteration products using SVC petrology by Nye & Turner...... 80

Table 7.2 “Meteoric” and “Crater Lake” waters that have reacted with SVC alteration minerals are derived from samples “945” and “70AR-202”, respectively. Basis species concentrations are reported in moles. Both samples also include a basis species of 1 free kg H2O...... 81

xiii ACKNOWLEDGMENTS

I would like to thank Ormat for providing the opportunity for me to work on this project, funding, and field access. Thanks to Brigette Martini for her initial involvement in this project and general advice.

Thanks to Lara Owens for reviewing my interpretations as well as drafts of this manuscript and related presentations. Thanks to Allison Payne, Matt Uddenberg, Robert Stephan, Brad Peters, and Manuel Barros for field assisstance.

I would also like to thank a number of people at the Colorado School of Mines that were part of my graduate experience. Ric Wendlant for general advice, help with alteration and clay mineral characterization, and for reviewing drafts of this manuscript and related presentations. Thanks to Stuart Simmons for help with interpretation of chemistry results and modeling advice, and for reviewing a draft of this manuscript. Additional thanks to Thomas Monecke for general advice and for reviewing a draft of this manuscript. I’d also like to acknowledge Alexis Navarre-Sitchler for modeling help, Elizabeth Easley for general advice, Marilyn Schwinger and Debbie Cockburn for administrative assistance, and Erich Hoover for helping with the manuscript formatting. Special thanks to John Skok for preparing thin sections for this research. And thanks to the Department of Geology and Geological Engineering as well the Rudy Epis Fellowship for additional financial support. Thanks to Simon Poulson at the Univeristy of Nevada, Reno, for general advice and lab assistance. Special thanks to Howard Spero and Natalie Caulk at the University of Davis, California for providing prompt calcite isotope results.

xiv CHAPTER 1

MOTIVATION

One state that could reap enormous benefits from the utilization of geothermal energy is Alaska. Alaskans consume the highest amount of energy per capita in the country, and have one of the highest rates nationwide per kilowatt hour (Alaska Energy Authority, 2009). Historically, nearly all of Alaska’s energy needs have been met by fossil fuels. The state has geographically scattered and isolated communities, many of which depend on diesel generators for electricity because of the lack of energy transmission and distribution infrastructure

(Alaska Energy Authority, 2009). Abundant oil resources mean that there has not been much motivation to diversify. But the unpredictability and fluctuations of oil prices, and a growing desire to be more energy independent have driven Alaska to start considering other energy sources. Besides oil and gas, Alaska is known to have geothermal, hydro, wind, biomass, and wave/tidal resources capable of producing electricity (Alaska Energy Authority, 2009).

1.1 Geothermal Resources of Alaska

It has been known at least since the early 1980’s that the state of Alaska has a number of proven or speculated geothermal resources (State Long Term Energy Plans 1980-1982, Turner & Wescott, 1986, (Mo- tyka et al., 1993)). However, due to the remote location of many potential resources and the aforementioned economic factors, these resources have remained largely untapped. In 2005, Alaska’s Golden Valley Electric Association launched the Sustainable Natural Alternative Power program (SNAP), which connects renew- able energy producers to membership-based purchasers (Golden Valley Electric Association, last checked 9/25/12). There are 49 producers who get paid about $0.30/kWh depending on how much energy is pro- duced and the amount of member contributions. The maximum incentive is $1.50/kWh (Golden Valley Electric Association, last checked 9/25/12; North Carolina State University, last checked 9/25/12). In 2008, the State of Alaska set up a state grant program overseen by the Alaska Energy Authority to assist utilities, power producers and local/tribal governments to conduct feasibility studies, reconnaissance studies, resource monitoring, and work related to design, permitting and construction of power-producing facilities. The leg- islation intends to provide $50 million to the program annually for 5 years (H.B. 152). Additionally, in 2010, the State of Alaska made residential renewable energy systems exempt from taxation (AS 29.45.050, 2010).

These recent economic and sociopolitical incentives have made utilization of geothermal resources in Alaska more favorable.

1 Figure 1.1: Geothermal resources of Alaska are indicated by areas of red shading. Nearly all of the Aleutian arc volcanoes fall into this category. Data from U.S. Department of Energy (2005), figure modified from INEL.

Alaska’s first and only electricity-producing geothermal power plant, Chena Hot Springs (Figure 1.1), came online in 2006 and produced 680kWe from 73° C fluid in 2009 (Alaska Energy Authority, 2009). Other candidates for geothermal development in Alaska include Pilgrim Hot Springs, Manley Hot Springs, Circle Hot Springs (which use their hot springs for space heating and spas), Makushin volcano on Unalaska island, Hot Springs valley on Akutan island, Baranof island, and Adak island (Figure 1.1) (U.S. Department of Energy, 2005). The Aleutian arc is characterized by active volcanic systems, shallow, magmatically heated rock, and deep fracture and fault systems. There are at least 56 sites in the Aleutian arc where surface expressions of hydrothermal systems have been identified (Motyka et al., 1993). The proximity of volcanoes on the west side of the Cook Inlet to population centers in south central Alaska makes them appealing targets for geothermal exploration. Mt. Spurr is the easternmost active volcano in the Aleutian arc and is located about 100km west of Anchorage (Figure 1.2) (Waythomas & Nye, 2002). Mt. Spurr is covered by about

67 cubic kilometers of perennial snow and ice, which is about 15 times more snow and ice than is present on Mount Rainier (Waythomas & Nye, 2002). Most of the area surrounding this volcano is uninhabited

2 wilderness, and its remote location limits recreational use of the area (Waythomas & Nye, 2002). Mt.

Spurr is considered a geothermal prospect because it’s close to poulation centers in the Cook Inlet, recent volcanic activity demonstrates a heat source, Eocene-younger regional faulting could provide permeability for a potential reservoir to form, and because there’s plenty of working fluid in the form of rain and snowmelt.

Figure 1.2: Mt. Spurr is located in south-central Alaska in the Tordrillo Mountains on the northwest side of the Cook Inlet about 100km west of Anchorage. Regional faults and geographic features are labeled.

The field area examined in this research lies immediately south of the summit of Mt. Spurr. The northern boundary of the study area is Crater Peak, the southern boundary is the Chakachatna River, the western boundary is Crater Glacier, and the eastern boundary is just east of Chaka Ridge (Figure 1.3). Mt. Spurr’s volcanic system has been widely studied by Alaska state agencies. Past and ongoing studies of gas emissions, ice melt, geologic history, seismic activity and geochemistry largely address the hazard of future eruptions.

3 Figure 1.3: Aerial photo of Mt. Spurr looking north. Crater Peak is a parasitic cone on the south side of the volcano. Kidazgeni and Crater glaciers issue from the volcano’s summit. Sampling locations for this research include Crater Canyon, Chaka Ridge, and a few places to the south of Mt. Spurr summit between Crater Peak and the Chakachatna River. Figure modified from Waythomas & Nye (2002).

4 1.2 Geothermal Resource at Mt. Spurr

A hydrothermal system, which is a system where cold water infiltrates to depth, is heated, and rises to the surface, needs to meet some requirements to be considered a geothermal energy resource. Ideally, a geothermal system has a high temperature heat source, ample amounts of a working fluid, and lasting permeability that allows the fluid to circulate. Theoretically, Mt. Spurr meets these requirements. One of the aims of this research is to explore the possibility that a hydrothermal system capable of being economically utilized for geothermal-based electricity production exists in the subsurface on the south flank of Mt. Spurr.

This research builds on studies by previous workers that include geologic mapping, geophysical surveys, soil sampling, and water sampling (past work is discussed in the following chapter, section2.3). The investigations of this research focus on the chemistry of surface water discharges and mineralogical relationships in drilled core obtained from a field area south of Mt. Spurr. Objectives include understanding the origin of waters, depth of circulation, possible processes that occur between depth and the surface, controls on fluid flow, as well as the deep temperature and spatial extent of the hydrothermal system.

5 CHAPTER 2

BACKGROUND

Geologic history and context of Mt. Spurr, as well as previous investigations of a geothermal system on the south flank of the volcano are discussed in this chapter.

2.1 Regional Geologic History

The geology of Alaska is the result of convergent plate boundary motion, the amalgamation of various terranes, and oblique shearing. Northward movement of the Pacific plate has acted as a conveyor belt that brings terranes from distant sites northward and adds them to southern Alaska (Waltham, 1995). Accretion of terranes started in the Paleozoic and continues to the present.

Figure 2.1: Timeline showing the geologic evolution of the Cook Inlet Basin, as well as volcanic activity at Mt. Spurr and the timing of water sampling campaigns.

The Peninsular terrane and subsequent igneous intrusions make up most of the inland geology just west and north of the Cook Inlet (Figure 2.1). The Alaska and Aleutian mountain belts have undergone repeated

6 deformation, intrusion, orogenic sedimentation, and volcanism in Mesozoic and Cenozoic times (Reed &

Lanphere, 1973). The Alaska-Aleutian Range batholith refers to the granitic rocks of these mountain belts.

Three intrusive epochs based on concordant K-Ar ages are recorded in plutonic (gabbro-granite) rocks that comprise the Alaska and Aleutian Ranges: 1) early-mid Jurassic (154-176 Ma), 2) late Cretaceous-early

Tertiary (58-83 Ma), and middle Tertiary (26-38 Ma) (Reed & Lanphere, 1973). Reed & Lanphere (1973) speculate that the three episodes of intrusion represent the products of three different convergent plate boundaries, the earliest of which may have been oceanic crust subducted beneath oceanic crust, while the two later episodes were derived from oceanic crust subducted beneath continental crust. The changing tectonic behavior of the Pacific plate could have caused these distinct subduction situations.

Figure 2.2: Stratigraphic column of sediments in Cook Inlet basin, showing basement (Triassic and older), marine deposits (Jurassic-Cretaceous), and basin fill (Tertiary-present). The Eocene West Foreland For- mation unconformably overlies the Chickaloon Formation, and is unconformably overlain by the Hemlock Conglomerate.

7 Active volcanoes near the Cook Inlet cut the Alaska-Aleutian Range Batholith. Sedimentary rocks in the Cook Inlet basin and surrounding margins can be divided into three stratigraphic groups: Triassic and older basement, Jurassic-Cretaceous marine deposits, and Tertiary basin fill (Figure 2.2).

2.1.1 Paleozoic

The oldest rocks of the western Cook Inlet area are Paleozoic carbonate and schistose rocks that are vari- ously metamorphosed up to epidote-amphibolite grade. The protoliths of these rocks are mafic-intermediate volcanics, limestone, and quartz-rich sedimentary rocks (Plafker et al., 1989). Overlying this sequence is the Peninsular terrane, which consists of metamorphosed limestone, chert, sandsone, and basaltic lava flows

(Magoon et al., 1975).

2.1.2 Mesozoic

During the Mesozoic, subduction of the east-moving Pacific plate beneath the North American plate formed the orogenic belt of the Western Cordillera (Canadian and American Rockies), which was added to the Laurentian shield. The divergent boundary of the Pacific Rise was overridden by the advancing North American plate, and motion of the Pacific plate changed to a northward direction (Waltham, 1995). Andesitic flows, breccias, and agglomerates of the Alaska-Aleutian Range batholith magma body that were vented to the surface comprise the early Jurassic Talkeetna Formation (Magoon et al., 1975). During the middle Jurassic, the Bruin Bay fault (Figure 1.2) may have initiated movement (Magoon et al., 1975). In middle-late Jurassic time, three more marine sequences were deposited: the Tuxedni Group, Chinitna Formation, and Naknek Formation, respectively. The Tuxedni Group consists of marine greywacke sandstone and siltstone (Magoon et al., 1975) that was deposited in a series of marine transgressions and regressions (Boss et al., 1976). This unit contains lithic fragments from the erosion of the Alaska Batholith (Magoon et al., 1975), and is thought to be the source rock of hydrocarbons found in the overlying lower Tertiary sediments (Magoon & Anders, 1992; Magoon et al., 1975). The Chinitna Formation is composed of dark gray siltstone with large concretions (Magoon et al., 1975) that is thought to represent a marine slope facies (Boss et al., 1976). This unit could also be a hydrocarbon source rock (Magoon et al., 1975). The Naknek Formation sediments include sandstone, conglomerate and mudstone from an igneous source, probably from the uplifted Talkeetna Formation (Trop et al., 2005). Naknek sediments are thought to be deposited in an environment that changed from a high-gradient alluvial fan system to high energy shallow marine setting (Boss et al., 1976). Next, extensive uplift and erosion of Jurassic and probably early Cretaceous deposits occurred, and there was a depositional hiatus (Boss et al., 1976; Magoon et al., 1975). Trop et al. (2005) attribute crustal scale shortening, uplift, and pluton exhumation with synorogenic sedimentation at this time

8 to the collision of the Wrangellia and Peninsular terranes with the former continental margin or amalgamation

of the terranes before collision.

In late Cretaceous time the Matanuska Formation, composed of marine shales and turbidites, was de-

posited in a shallow marine setting transitional to a deep marine setting (Boss et al., 1976; Magoon et al.,

1975). Another unconformity in the rock record records considerable deformation and erosion that has been

described as “compressional uplift followed by tensional downwarp” (Magoon et al., 1975). At this time, the

trough of the Cook Inlet Basin was formed (Magoon et al., 1975), with the Mesozoic and older sediments,

which were subsequently eroded, representing the floor.

2.1.3 Cenozoic

Tertiary sediments were deposited on top of Mesozoic and older sediments relatively rapidly by fluvial

processes under conditions of continuous uplift and erosion of the Alaska and Chugach mountain ranges (Boss et al., 1976). The main drainage direction seems to be to the southwest, and paleocurrent data also indicates the presence of cross-streams that flowed perpendicular to the main drainage (Hartman et al., 1972; Kremer & Stadnicky, 1985). The post-Cretaceous undulating erosional surface in the Cook Inlet Basin was covered with mudstone, siltstone, sandstone, minor conglomerate and coal of the Chickaloon Formation in Paleocene-lower Eocene time (Flores et al., 2004). The Chickaloon Formation is inconformably overlain by tuff, siltstone, sandstone, shale and polymictic conglomerate of the Paleocene-Eocene West Foreland formation (Figure 2.2, Figure 2.1), which is also known as the Wishbone Formation in the northern parts of the basin (Boss et al., 1976; Hartman et al., 1972). In Eocene-Oligocene time, regional strike-slip faulting and folding deformed this unit, and transpressional stress led to the creation of folds, faults, and eroded horst blocks and grabens with synorogenic fluvial deposits in the Cook Inlet Basin (Bruhn et al., 2000). At about 42 Ma, movement of the Pacific plate changed from northward to north-westward motion (Waltham, 1995). An unconformity separates the West Foreland Formation from the base of the Kenai Group sediments (Boss et al., 1976), which include the Hemlock Conglomerate, the Tyonek Formation, the Beluga Formation, and the Sterling Formation. The Hemlock Conglomerate was deposited in the Oligocene and is composed of pebbly sandstone, conglomerate, siltstone, and minor shale and coal (Calderwood & Frackler, 1972; Hartman et al., 1972; Kremer & Stadnicky, 1985). The sandstone makes up 50-70% of the unit composition (Boss et al., 1976). The top of the Hemlock Conglomerate grades conformably into the Tyonek Formation (Calderwood & Frackler, 1972).

The Tyonek Formation was deposited in Oligocene-Miocene time and contains massively bedded sand-

stones, conglomerate, interbedded shale, and coal beds (Calderwood & Frackler, 1972; Kremer & Stadnicky,

1985). The sandstone does not take up more than 25% of the total unit composition (Boss et al., 1976). The

9 top of the Tyonek Formation meets the base of the Beluga formation or the base of the Sterling formation in a mostly disconformable way (Calderwood & Frackler, 1972; Kremer & Stadnicky, 1985).

The late Miocene Beluga Formation is composed of intercalated silty-clayey sandstone, siltstone, shale, coal, ash, pebbly sandstone and conglomerate with abundant metamorphic clasts (Calderwood & Frackler,

1972; Kremer & Stadnicky, 1985). The Beluga Formation can be distinguished from the Tyonek by the absence of massive sandstone beds and smaller coal seams (Calderwood & Frackler, 1972). The overlying

Miocene-Pliocene Sterling Formation contains more plagioclase and volcanic rock fragments than the Beluga

Formation (Kremer & Stadnicky, 1985). The Sterling Formation also has thinner but more abundant coal beds (Kremer & Stadnicky, 1985). It is composed of thick partly silty-clayey sandstone, pebbly sandstone, minor conglomerate, siltstone, tuff, claystone and coal (Kremer & Stadnicky, 1985). The unit is at least 50% sandstone, often more than 75% sandstone by composition (Boss et al., 1976). Pliocene-Quaternary deformation in the Cook Inlet basin cannot be explained by the process of North American Plate-Pacific Plate subduction alone (Haeussler et al., 2000). It is believed that the Yakutat terrane collided with the subduction zone to the east of the Cook Inlet during late Miocene time, and added a right lateral component to the fault movement around the borders of the basin (Haeussler et al., 2000). Holocene movement along the Castle Mountain Fault (Figure 1.2) is thought to be a result of this dextral transpression (Haeussler et al., 2000).

2.1.4 Recent Faulting

The role of faulting on the continuing development and sedimentation of the Cook Inlet forearc basin and surrounding areas is likely significant. The depocenter of the Cook Inlet basin is bounded to the northwest and west by the Castle Mountain, Lake Clark, and Bruin Bay faults, and to the southeast by the Border Ranges Fault (Figure 1.2) (Montgomery & Barker, 2003). The Castle Mountain, Lake Clark, and Bruin Bay faults are all reverse faults with northwest side up and some component of oblique slip (Finzel et al., 2009). Displacement on the eastern part of the Castle Mountain Fault is up to 1.2km of reverse motion and 130km of right lateral separation since at least the Early Cenozoic, and scarps and historical activity suggest Holocene movement of 2.8-3mm/yr. The Lake Clark Fault is estimated to have 500m-1km of vertical offset and 5-26km of right lateral motion since the late Eocene. The Bruin Bay fault has left-lateral motion of 19-65km and vertical motion of at least 3km, but may have less vertical motion near the intersection with the Castle Mountain Fault. Significant motion on the Bruin Bay Fault may be as recent as Quaternary time

(Finzel et al., 2009).

The east-northeast trending Capps Glacier Fault (Figure 1.2) cuts through the study area to the east of Mt. Spurr and likely continues westward through the volcanic complex just to the south of Crater Peak

10 (Wilson et al., 2009). East of Mt. Spurr, the Capps Glacier Fault juxtaposes Cenozoic granitic rocks to

the northwest over gently deformed Cenozoic clastic rocks, apparently of the West Foreland Formation, to

the southeast (Figure 1.2). Geologic mapping and structural analysis by the Alaska Division of Geological

and Geophysical Survey indicates dextral-normal or transtensional fault movement is likely syntectonic with

the deposition of the West Foreland Formation (Gillis 2012, pers comm). Detrital zircon ages, sandstone

composition, and conglomerate clast composition from the West Foreland Formation outcrop suggests that

the sediment source is nearby Cretaceous metavolcanic wall rocks to the granitic intrusions, as well as the

intrusives, possibly from the footwall of the Capps Glacier Fault (Gillis 2012, pers comm). Displacement

along the Capps Glacier fault could be tens to hundreds of meters (Finzel et al., 2009). Folded Cenozoic

basin deposits to the southeast of the Capps Glacier Fault are oriented obliquely (20-30° counterclockwise) to the fault. Similar oblique orientations of cored anticlines in the Cook Inlet basin could imply that the Bruin Bay, Castle Mountain, and Capps Glacier faults and associated folds evolved in similar regional stress conditions (Finzel et al., 2009).

2.2 Eruptive History of Mt. Spurr

Mt. Spurr is part of the belt of Quaternary volcanoes that lies immediately north of the Aleutian trench, which is the convergent boundary where the Pacific Plate is forced beneath the North American Plate. Nye and Turner (1990) consider the Spurr Volcanic Complex (SVC) to be representative of eastern Aleutian volcanism, and note that it has a single, mature center that is located on the thickest crust of any Aleutian arc volcano. Additionally, Mt. Spurr is the easternmost active center of the Aleutian arc. SVC tephra and lava flow deposits record at least 35 eruptions in the last 6000 years (Riehle, 1985). Ancestral Mt. Spurr was composed of lava flows and pyroclastics dated around 255,000 to 18,000 years ago (Figure 2.1).

Ancestral Spurr lavas are mostly uniform 58-60% SiO2 andesite. This ancestral Spurr edifice was destroyed in a “Bezymianny-type” eruption about 5800 years ago that produced a horseshoe-shaped caldera, a debris avalanche partially overlain by ash-flow tuff, and two new vents: the current Mt. Spurr, which grew out of the collapsed summit region, and proto-Crater Peak, a parasitic cone on the south side of the edifice (Nye & Turner, 1990). The nature of magmatism changed abruptly after the caldera was formed. Two distinct magmas and mixtures of these two have been observed. The current Mount Spurr dome is made out of andesite that is more silicic than that of ancestral Spurr, but contains olivine and amphibole xenocrysts derived from more mafic magma. Lavas from proto-Crater Peak and Crater Peak are consistently more mafic than ancestral Spurr and contain characteristic pyroxenite clots that are not seen in ancestral Spurr samples

(Nye & Turner, 1990). Crater Peak has erupted in recent time, in 1953 and again in 1992. Both eruptions were Volcanian to sub-Plinian pyroclastic events (Waythomas & Nye, 2002) that blanketed populated areas

11 east and northeast of the volcano with as much as 3mm of sand-sized ash, interrupting airports, schools,

and businesses for a day or more (Neal et al., 1993). Resuspended ash was an intermittent problem for more

than a year later (Neal et al., 1993). Seismic activity as well as the formation of a crater lake and fumaroles

were observed at Mt. Spurr summit in 2004-2006 (Neal et al., 2008).

The Spurr Volcanic Complex lies on top of continental crust formed by amalgamation of accretionary

terranes including Silurian-early Jurassic limestone, basalt and clastic rocks, and Jurassic and/or Cretaceous

flysch (Figure 2.3). All of these rocks have been intruded by the Aleutian range batholith during Jurassic,

early Tertiary, and mid-Teriary times (Reed & Lanphere, 1973). Immediately to the south of Mt. Spurr,

the bedrock is granodiorite (62 and 69Ma), while to the north-northeast quartz monzonite intrudes Jurassic-

Cretaceous flysch and is faulted against the West Foreland Formation (Nye & Turner, 1990).

2.3 Previous Geothermal Investigations

Previous workers have demonstrated the presence of a heat source, a working fluid, and permeability at Mt. Spurr. Surface manifestations of an active hydrothermal system at Spurr summit cone include indications that subsurface heat has melted snow in a normally snow covered area, such as bare rock, subsidence caused by snowmelt, icicles in snow caves as well as transient seismic activity, crater lakes, and fumaroles. Other indicators of deep heat escaping to the surface in the field area include crater lake and fumarolic activity at Crater Peak, anomalous Hg and He in soil gas surveys (Figure 2.4), and warm springs and seeps in the valley immediately south of Crater Peak (Turner & Wescott, 1986). Ground- based magnetotelluric (MT) and gravity surveys in the study area (Figure 2.5) appear to indicate potential magmatic intrusions along deep fault structures. Shallower features interpreted as heavily altered volcanics and/or brine-filled units (Martini et al., 2011) from the MT data coincide with the low resistivity features identified in the controlled source audio magne- totelluric (CSAMT) surveys by Turner & Wescott (1986)(Figure 2.7, Figure 2.6). Recent eruptions (1953 and 1992) at Crater Peak and activity at Mt. Spurr summit (2004-2006) hint at the possible presence of a relatively shallow, still hot magma chamber. In fact, Keith et al. (1995) use

the extreme difference in Crater Lake SO4 and Cl contents to assert that the hydrothermal system south of Crater Peak was disturbed by the intrusion of magma leading to the 1992 eruption. Since attempts at dating pyroclastic rocks produce relatively young ages, this magma chamber likely still contains viscous to partially crystallized magma that could be a heat source for a geothermal system that is suitable for electricity generation (Turner & Wescott, 1986). Turner and Wescott (1986) suggest that geothermal waters could be located south of Crater Peak at about 60m depth in porous ancestral Spurr tephra beds that are sealed by impermeable Crater Peak lava flows.

12 Figure 2.3: Geologic map of Mt. Spurr by Nye et al. (2005) that shows andesitic volcanic deposits from Mt. Spurr and Crater Peak eruptions, as well as Alaska-Aleutian Range batholith granites, regional sedimentary units, Holocene deposits, and glaciers/perennial snow.

13 Contributions of water from melting of ever-present ice, snow, and glaciers are significant. Measurements

from 1971-2001 from a weather station at Beluga (a small tribal community on the west bank of the Cook

Inlet about 60km directly east of Mt. Spurr), show that typical ambient temperatures are 9-13°C, and that the area recieves about 0.6m of annual rainfall and 2m of annual snow. Assuming the snow is wet

and heavy and has a density of about 500kg/m3, the snow-water equivalent is 1000kg/m2. August and

September are the months of heaviest precipitation (Alaska Climate Research Center, last checked 9/25/12).

Previous workers have collected water samples from Crater Lake in Crater Peak, thermal springs in Crater

Canyon, discharge issuing from Kidazgeni glacier, as well as springs and seeps in Kid Canyon and Chaka

Ridge (Table 2.1).

Motyka and Nye (1993) suggest that Crater Lake waters have evolved from a shallow steam-heated hydrothermal layer that has subsequently boiled, and that thermal surface discharges in Crater Canyon represent Crater Lake waters that have percolated downslope. Neal et al. (2008) and Martini et al. (2011) speculate that Kid Canyon waters represent hydrothermal fluid derived from Crater Peak that has become entrained within basal glacial waters and transported to the surface. Elevated B and silica concentrations in springs and seeps seem to correlate with thermal manifestations in the study area (Martini et al., 2011). While some of the volcanic units mapped in the study area are likely quite porous, interlayered tephra and tuff that have decomposed to clay minerals could provide vertical barriers to fluid flow. Motyka and Nye (1993) suggest that Crater Peak lies on the extension of the Capps Glacier fault, which provides a zone of weakness that facilitates the upward migration of magma and penetration of meteoric waters to depth. In the study area, MT and gravity datasets support airborne (helicopter) magnetic survey data, LiDAR data, and surface mapping results to show three major structural trends: east-northeast, north-northeast, and northwest (Martini et al., 2011). Seismicity at Mt. Spurr in the general north-northeast trend has been observed during and after the 1992 Crater Peak eruption (Power et al., 1995). Rocks on the South Bench fault and North Bench faults (Figure 2.3) identified by Martini et al. (2011) show secondary argillic to advanced argillic alteration and slickenlines on fault surfaces. These faults are though to be related to the Capps Glacier Fault and regional Cook Inlet fault structures and could potentially serve as conduits to fluid flow. Three exploration core holes have been drilled in the study area: 62-2, 67-34, and 26-11. Drilling over the summer of 2010 in core hole 67-34 intersected 128m of volcaniclastic flow underlain by 168m of conglomerate and fine-grained sedimentary rocks (Ormat Nevada, Inc., 2012b). The volcaniclastic flow is presumably an ancestral Mt. Spurr deposit, and the conglomerate and sandstone units match the description for the West

Foreland Formation. Conventional core logging and spectral logging in the short-wave infrared (0.35-2.5￿m) using an ASD Fieldspec PRO portable spectrometer was performed. Spectral logging was targeted to areas

14 Table 2.1: Summary of water sampling campaigns at Mt. Spurr. Asterisks denote samples that were not considered in this research because the data are questionable due to over-acidification of samples in the field. Samples from West Spurr are also not considered because they lie outside the field area of this research. References are as follows: [1] Martini et al. (2011) [2] Motyka & Nye (1993) [3] Neal et al. (2008) [4] Keith et al. (1995)

Sample Name Year Location Notes Ref 154 10* 2010 Chaka Ridge Chaka Ridge -cold spring [1] SPS-1 10 2010 Chaka Ridge North Bench lower [1] 435 10* 2010 Chaka Ridge Chaka Ridge -cold spring [1] 222 10* 2010 Chaka Ridge Chaka Ridge -cold spring [1] SPS-2 10 2010 Chaka Ridge North Bench upper [1] 120 10* 2010 Chaka Ridge Chaka Ridge -cold seep [1] 67-34 2010 Chaka Ridge Core Hole 67-34 [1] Spurr-85 1989 Crater Canyon Crater Creek [2] UCC 2009 Crater Canyon Upper Crater Creek [3] LCC 2009 Crater Canyon Lower Crater Creek [3] SP001 10 2010 Crater Canyon Crate Warm Sp [1] 908071320 2010 Crater Canyon North CC2 [1] 908041626 2010 Crater Canyon North CC [1] 532 10* 2010 Crater Canyon Mid CC [1] 220 10* 2010 Crater Canyon Crater Glacier [1] 505 10* 2010 Crater Canyon Lower CC [1] 715 10* 2010 Crater Canyon Little Yellowstone [1] 70AR-201 1974 Crater Lake Spurr Crater Lake 1970 [4] 70AR-202 1974 Crater Lake Spurr Crater Lake 1970 [4] CP92-1 1996 Crater Lake Crater Lake 1992 [4] CP92-2 1996 Crater Lake Crater Lake 1992 [4] KID1 2010 Kid Canyon Kid Canyon Glacial outflow [1] SP013 2010 Kid Canyon Kid Canyon Glacier [1] SP014 2010 Kid Canyon Kid Canyon Glacier [1] SP030 2010 Kid Canyon Kid Canyon- base of pyroflow [1] SP031 2010 Kid Canyon Kid Canyon- base of pyroflow [1] Kid Glacier 2010 Kid Canyon Kid Glacier [1] 06SPCN001A 2006 South Spurr Kidazgeni Glacier Outflow [3] 945 10* 2010 South Spurr Kid Glacier [1] 555 10* 2010 South Spurr Acid Creek [1] SP002a 2010 South Spurr Acid Creek Upper [1] SP002b 2010 South Spurr Acid Creek Lower [1] 341 10* 2010 West Spurr Barrier Glacier [1] 005 10* 2010 West Spurr High Glacier [1] 025 10* 2010 West Spurr High Glacier [1] 750 10* 2010 West Spurr Straight Crk Glacier [1] 415 10* 2010 West Spurr Pothole Glacier [1] 450 10* 2010 West Spurr Harpoon Glacier [1]

15 of altered core. Data cube information obtained by the ASD was then processed using ENVI software to identify unique pixels and their spectra, which were matched to existing database mineral spectra, and visually confirmed. These logging methods have identified disseminated chlorite and montmorillonite in the conglomerate, and concentrated montmorillonite in the fine-grained sedimentary rocks (Ormat Nevada, Inc.,

2012b). Another core hole drilled in the summer of 2010 is 62-2, which intersected 268m of variously altered andesite with two fault zones underlain by 6m of volcaniclastic flow (Ormat Nevada, Inc., 2012b). These volcanic units are presumably ancestral Mt. Spurr deposits. Conventional core logging and spectral analysis using the same procedure have identified clay in fractures and fault zones as montmorillonite and kaolinite.

Core hole 26-11 was drilled during the summer of 2011. Core from 26-11 has been examined in detail in this study and will be described in the following sections.

16 Figure 2.4: Mercury and helium soil gas surveys at Mt. Spurr by Turner & Wescott (1986).

17 (a) 400m elevation 1D inversion-resistivity (MT) results at Mt. Spurr.

(b) 800m elevation 1D inversion-resistivity (MT) results at Mt. Spurr.

Figure 2.5: 1D inversion-resistivity (MT) results at Mt. Spurr in plan view. The red lines represent the Ormat lease boundaries, and the black lines represent mapped faults. Locations of MT stations shown by small triangles in both maps. Conductive anomalies in ‘warmer’ purple-red colors. Figures from Martini et al. (2011)

18 Figure 2.6: 3D fence diagram of CSAMT data. Black areas are zones of resistivity <5￿m. Figure from Turner & Wescott (1986). Transects are shown in map view by the blue lines in Figure 2.7.

Figure 2.7: Location of CSAMT survey lines (blue, from Turner & Wescott (1986)) overlayed on elevation contours and the geologic map of Nye et al. (2005). Bold black lines represent faults from Martini et al. (2011). The colored symbols are water sampling locations and are explained in the next chapter.

19 CHAPTER 3

METHODS

Waters from springs and seeps in the study area as well as sedimentary unit and clay mineral-rich samples from exploration core hole 26-11 were collected. Dissolved constituents and stable isotopes of water samples were analyzed. Thin sections were made from core samples, clay minerals were characterized by X-ray diffraction (XRD), and stable isotopes of calcite veins and cements were analyzed.

3.1 Field Sampling

Water and core samples were collected for this research during a field campaign 19 July-4 August 2011.

The locations of previous sampling campaigns and helicopter-based field accessibility guided the selection of spring sampling locations (Figure 3.1). In this and subsequent figures, orange diamonds represent samples from Crater Peak, red circles are samples from Crater Canyon, blue triangles are samples from Chaka Ridge, black crosses are samples from Kid Canyon, green squares are samples from scattered locations to the south of Mt. Spurr (South Spurr), and yellow pentagons with black dots represent exploration core holes. Some sites that have been previously studied (Table 2.1) were revisited in this research in an effort to record possible temporal variations or obtain more complete data sets. Additionally, some previously sampled locations were re-visited with the intent of collecting samples for this research were inaccessible (Kid Canyon, Crater Peak) or did not exhibit surface water flow (South Spurr) at the time of the field campaign. Sampling sites are described in Figure 3.2-Figure 3.4. The surface discharges that were sampled in Crater Canyon have variable character (Figure 3.2). The spring identified as “320” issues from the top of the west side of the canyon below Crater Glacier, and “CC Head” issues from the northernmost point of the canyon, also below Crater Glacier. Crater Creek starts below “CC Head” at the bottom of the canyon and receives contributions of water from other springs as well as water seeping from the bottom of the canyon. Sample “626” is a large algae-covered seep (˜15m tall) that issues from the canyon wall near the head of the canyon. “SP001” is also algae covered and issues from a protrusion in the west canyon wall, and “532” issues from the talus between “SP001” and Crater Creek. The yellow flowers near thermal surface discharges are Mimulus guttatus, or “Seep Monkey Flower,” a perennial plant that has a medium CaCO3 tolerance, can survive in temperatures as low as -38°C, lives in the pH range of 6-8, requires 25-61cm of precipitation, and is found throughout the western US, including near hot springs in Yellowstone (U.S. Department of Agriculture

Natural Resources Conservation Service, last checked 9/26/12). Surface discharges sampled at Chaka Ridge

(Figure 3.3) largely issue from talus where topography changes from relatively flat to steeply sloping (“154”,

20 Figure 3.1: Sampling locations shown with faults by Martini et al. (2011), and geologic map by Nye et al. (2005).

21 (a) “626”. Algae-covered seep issues from east side of canyon wall (b) “Crater Creek”. Stream is ˜2-4m wide and (c) “CC Head”. Spring issues from head of Crater below sloping horizontal strata. The wet part of the seep is about flows southward along bottom of canyon. Canyon, flows south down into canyon. “626” is ˜15m tall. the dark area on the right side of the canyon.

(d) “532”. Seep issues near bottom of canyon just above west side (e) “320”. Spring emerges near top of west side of (f) “SP001”. Seep covered in yellow Monkey Flow- of Crater Creek. canyon wall. ers and algae issues from ˜2m tall protrusion on east side of canyon wall and flows downslope into Crater Creek.

Figure 3.2: Sample sites in Crater Canyon

22 “SPS-2”, “222”, “Spring”, “435”). Sample “SPS-1” is probably heavily influenced by snowmelt. Surface

discharges from the South Spurr area (Figure 3.4) also vary in character. Sample “Stream” issues from slope

south of a relatively flat-lying alder forest on the north bank of the Chakachatna River. Sample “945” is

turbid glacial outflow, and “Canyon” issues from the bottom of a downward sloping canyon floor.

At each site, H2S, SiO2, alkalinity, temperature, pH, and conductivity were measured in the field. A Oakton Waterproof Multiparameter portable probe was used to measure pH, temperature, and conductivity.

The accuracy for pH is ±0.01 ph unit, temperature is 0.5￿, and conductivity is 1% full scale. Total alkalinity was measured by visual endpoint titration using a Hach Alkalinity Test Kit Model AL-DT. Hydrogen sulfide

was determined visually by color disc with a Hach H2S Color Disc Test Kit, Model HS-WR. Silica was measured visually by color disc with a Hach Silica Test Kit, Model SI-5. Approximate flow rate was estimated

by visual observation. Since the values for alkalinity, silica, H2S and flow rate were visually determined, some subjectivity is introduced into the resulting field values.

Water was extracted by syringe and filtered through a 0.45￿m membrane (to prevent algal growth, which

may remove Mg, NH3, and SO4) into a suite of bottles for laboratory geochemical analysis of anions, cations,

major components (SiO2, HCO3,SO4), dissolved gas (H2S), and stable isotopes. The samples for cation

analysis were acidified with 0.02N HNO3 to preserve the cation contents and prevent precipitation of trace metals. Representative samples of conglomerate and sandstone from core hole 26-11 were selected for petrographic analysis at the time of drilling based on the extent of alteration (fresh vs altered appearance) and proximity to fault zones. Samples for clay mineral analysis were also selected based on the occurrence of clay material and proximity to fault zones.

3.2 Analytical Chemistry

Anions, cations, and major components were analyzed by the Western Environmental Testing Laboratory (WETLab). Ammonia was analyzed with an ammonia selective electrode; pH was determined by an electro- metric method; bicarbonate, carbonate, hydroxide and alkalinity were determined by electrometric titration; phosphorous was determined by calorimetry; Cl, F, SO4, and NO3 were analyzed by ion chromotography; TDS was measured by gravimetric methods; electrical conductivity was determined by conductivity bridge methods; Si, Al, Ba, Be, B, Cd, Ca, Cr, Co, Cu, Fe, Pb, Li, Mg, Mn, Ni, K, Ag, Na, Sr, and Zn were analyzed by inductively coupled plasma-atomic emissions spectrometry (ICP-AES); Sb, As, Th, and Hg were analyzed

with an inductively coupled plasma-mass spectrometer (ICP-MS) according to standard and EPA methods.

Quality control procedures for the samples collected in this research implemented by WETLab include

running blanks every 10 samples to ensure that values are below detection limit, running lab control samples

23 (a) “154”. Looking northwest, ˜1m wide spring (b) “435”. Looking northwest, ˜0.5m spring issues (c) “SPS-2”. Looking north, ˜0.5m wide spring issues from issues from steep slope. from shallow slope. steep slope.

(d) “222”. Looking northwest, ˜0.5m wide (e) “Spring”. Looking southwest, seep issues (f) “SPS-1”. Looking west, spring issues from shallow slope. spring issues from steep slope. below outcrop and flows over cliff-forming rock and down slope. Rock outcrop ˜2m tall.

Figure 3.3: Sample sites at Chaka Ridge

24 (LCS) every 20 samples to ensure that the instruments are working properly, running duplicates of client

samples to ensure that results are precise, and running matrix spike (MS) samples to ensure that there are

no matrix effects. The results of these runs are within the acceptable lab parameters and have less than or

equal to 5% relative difference for all runs except a duplicate TDS for “Stream”, which is reported but does

not meet lab QC criteria for precision or accuracy.

Charge balance calculations were applied to the resulting data, and were less than 7% with a few excep-

tions. Low charge balance errors (<10%) ensure that the data collected in this research are of high quality

and thus can be interpreted with confidence. “Canyon” has a 45% error, and inspection of values compared

to the replicate reveals that the Na concentration of “Canyon” is an order of magnitude higher. Sample

“945”, “CC Head” and “435” also have high charge balance errors (100%, 19% and 13%, repectively) because

many of the constituents analyzed were below detection limit and therefore not included in the calculations.

For example, SO4 was the only species above the detection limit in “945.” Stable oxygen, hydrogen, and sulfur (from sulfate) isotope analyses were conducted at the University of Nevada, Reno. Oxygen-18 isotope analyses of water were performed using a Micromass MultiPrep prepara-

tion device interfaced to a Micromass IsoPrime stable isotope ratio mass spectometer, using the CO2 -H2O equilibration method of Epstein & Mayeda (1998). Hydrogen isotope analyses of water were performed using a Eurovector model 3028 elemental analyzer interfaced to a Micromass IsoPrime stable isotope ratio mass spectometer, using the method of Morrison et al. (2001). Oxygen-18 and deuterium results are reported in per mil (parts per thousand, ‰) with respect to Vienna Standard Mean Ocean Water (VSMOW) with an uncertainty of ±0.1” (1 standard deviation). Sulfate was precipitated from aqueous solution as BaSO4, following the method of Carmody et al. (1998). Sulfur isotope analyses were performed using a Eurovector model 3028 elemental analyzer interfaced to a Micromass IsoPrime stable isotope ratio mass spectometer, after the method of Giesemann et al. (1994); Grassineau et al. (2001). V2O5 was added to BaSO4 as a combustion aid. Sulfur-34 results are reported in per mil with respect to Vienna Canyon Diablo Troilite

(VCDT) with an uncertainty of ±0.2”. Stable carbon isotope analyses were done at the University of Arizona. Carbon-13 of dissolved inor- ganic carbon was measured on a continuous-flow gas-ratio mass spectrometer (ThermoQuest Finnigan Delta PlusXL) coupled with a Gasbench automated sampler (also manufactured by Finnigan). Samples were re- acted for >1 hour with phosphoric acid at room temperature in Exetainer vials previously flushed with He gas. Results are reported in per mil with respect to Vienna Pee Dee Belemnite (VPDB), with standardization

based on NBS-19 and NBS-18, and a precision of ±0.30‰ or better.

25 3.3 Petrography of Drilled Core

Samples from core hole 26-11 are named for the depth (in feet) from which they were recovered, and are

described in Table 3.1. Polished thin sections of eight core samples from core hole 26-11 were made and

observed at the Colorado School of Mines using transmitted optical microscopy. Six of the thin sections

are of conglomerate with a sandy matrix, and two are from coarse-grained sandstone intervals of the West

Foreland Formation.

3.3.1 Clay Mineral Characterization

Seven clay-rich samples were selected from the core of 26-11 for clay characterization by XRD at the

Colorado School of Mines (Table 3.1). Core fragments were scrubbed with a toothbrush under running water to eliminate the possible presence of drilling mud. All of the samples reacted to water by swelling. The core fragments were allowed to air-dry, then were crushed to a fine powder by five minutes of shaking in a ball mill. Powdered randomly oriented samples were scanned with a Scintag, Inc. XDS 2000 diffractometer

using Cu-Kα1 radiation at 40kV and 40mA from 3-40° 2Θ. The step size was 0.02, and the rate of scanning was 0.48 sec/step and 2.5 deg/min. Each sample was then mixed with water and centrifuged at 1000rpm for one minute to separate a <3￿m clay-sized fraction. The <3￿m clay-sized fraction was then deposited on a 0.45￿m membrane and transferred to a glass slide using the filter transfer method (Moore & Reynolds, 1997). These oriented clay-sized fraction samples were then scanned in the same manner previously described. Since smectite reflections were seen in the previous two scans, the oriented clay-sized fraction samples were put in a glass chamber and allowed to equilibrate with ethylene glycol, which effectively replaces water in the smectite

structure so that all unit layers of smectite have a uniform spacing of ˜16.9A.˚ These glycolated samples were also scanned immediately after removal from the glass chamber as previously described. Finally, three

samples were scanned in the 59-65° 2Θ range to reveal 060 reflections of clay minerals. In addition to the clay analyses, two mineral scrapings were also analyzed by XRD. These scrapings are both from sample “3434” and are of a white material in veins and a red coating that occurs with chlorite on a fracture surface. Both minerals were separately scraped from the core sample, crushed with a mortar and pestle, mixed with water, and applied to a glass slide with an eyedropper. The red coating showed calcite reflections, but otherwise did not have much of a response, so a heat treatment was applied to test the hypothesis that the red mineral was an amorphous iron oxyhydroxide that would increase in crystallinity

with heat. The red coating was heated in a silver capsule to 700￿ for three hours, and then mixed with

propanol and applied to a glass slide with an eyedropper. The sample was then scanned in the 10-59° 2Θ range to capture possible iron oxide reflections.

26 3.3.2 Calcite Mineral Characterization

Finally, 10 calcite samples (4 from veins, 4 from rinds rimming clasts, and 2 from matrix/cement) were

extracted from 26-11 core with a drill, and analyzed for stable carbon (δ13C) and oxygen (δ18O) isotopes at the University of California, Davis Stable Isotope Laboratory (Table 3.1). Some core intervals yielded

more than one type of calcite sample. The samples were pre-roasted under vacuum at 375°C to drive off any possible organic carbon contamination. The samples were then analyzed using an Elementar IsoPrime mass

spectrometer. Samples were reacted in 105% phosphoric acid at 90°C using a Gilson Multicarb Autosampler

system with individual acid injection vials. The CO2 gas generated from this reaction was analyzed in dual inlet mode on the mass spectrometer to obtain C and O isotope values. The δ13C results are reported in per mil with respect to VPDB and the δ18O results are reported in per mil with respect to VSMOW.

27 (a) “Stream”. Looking north. Spring is- (b) “945”. Glacial outflow, looking northwest. sues from topographic change along bank of Edge of glacer is about 2m thick. Chakachatna River.

(c) “Canyon”. Looking southwest, spring issues from canyon floor.

Figure 3.4: Sampling sites in the South Spurr area.

28 Table 3.1: Summary of core samples and the analyses conducted on them. All samples are from core hole 26-11. TS=thin section, Clay, Calcite, FeOx= characterization of minerals by X-ray diffraction (XRD), and Iso=stable C and O isotopes of calcite.

Sample Depth (m) Description Analysis Unaltered conglomerate. Matrix supported poorly 682 208 sorted rounded-subangular igneous clasts. Matrix is TS sand sized lithics with quartz and clay cement. Poorly sorted matrix supported conglomerate that looks slightly altered. Clasts are rimmed with dark 1638 499 TS, Iso hematite, and the matrix contains white calcite and clay. Conglomerate and sandstone. Matrix contains glassy 1865 568 Clay to massive green montmorillonite and calcite. Tan with slickenlines, dessicated clay, and white 1889 576 Clay fibrous calcite. Overall green color, brittle granular-massive 2061.5 628 Clay montmorillonite and calcite. Conglomerate. Matrix suppored poorly sorted gravel-pebble sized clasts in matrix of sand-sized 2227 679 TS, Iso clasts with silica and clay cement. Green chlorite veins and staining on interior. Massive green montmorillonite. Depth associated 2580.5 787 Clay with flowing water. Massive, compact, overall green colored 2583 787 Clay montmorillonite. Depth associated with fault zone. Altered conglomerate. Matrix has overall green color (montmorillonite) and thin white calcite veins 2630 802 TS, Iso cross-cut lithic clasts. Compact green montmorillonite on fracture surface has slickenlines. Altered conglomerate. Matrix-supported gravel-cobble sized lithic clasts in matrix of altered 3049 929 sand-sized clast with silica and clay cement. The TS, Iso matrix has green montmorillonite, and the clasts have white calcite rims. Green granular-massive brittle montmorillonite and Clay, 3221 982 calcite. Iso Coarse sand with a few rounded pebbles. Fracture along length of core is mineralized with white 3419 1042 TS, Iso crystalline calcite that shows slickenlines and patches of dark red FeOx staining. Friable conglomerate is falling apart along cobbles. TS, Heavily altered throughout, red FeOx staining along Clay, 3434 1047 cobble rims and calcite along broken surfaces. Calcite, Matrix has overall green color due to the presence of FeOx, montmorillonite. Some chloritization of clasts. Iso Sandy and friable, green montmorillonite in matrix 3557 1084 TS and white spots of calcite.

29 CHAPTER 4

RESULTS

The results of this study are summarized in this chapter and discussed in the following chapters.

4.1 Aqueous Geochemistry

Field measurements are summarized in Table 4.1. No H2S was detected, and all springs have close to neutral pH. In general, the sampling temperatures are not particularly high. Samples from Crater Canyon had the highest field temperatures, silica concentrations, and conductivity. Analytical results for waters collected in this research and published data are compiled in Table 4.2 and Table 4.3. Values below the detection limit have been omitted. The chemistry of all samples is examined in Chapter 5.

Table 4.1: Field Measurements. Flow rate estimated visually.

Sample Field Location Silica H2S Alkalinity Temp pH Cond Flow (mg/L) (mg/L) (mg/L) (°C) (￿S) (L/s) SP001 Crater Canyon 200 0 320 30.2 7.7 3200 <0.3 320 Crater Canyon 15 0 3 12.2 8 140 >0.3 626 Crater Canyon 100 0 461-491 22.3 8.3 2600 ˜0.9 CC Crater Canyon 7 1.8 6.9 6.3 20 >0.3 Head Crater Crater Canyon 160 357 14.5 7.9 1800 ˜1.2 Creek 532 Crater Canyon 180 404 22 7 2400 <0.3 Canyon South Spurr 100 0 161 3.8 640 ˜0.3 Stream South Spurr 100 0 14 9.1 220 ˜0.6 945 South Spurr 0 0 1.3 -0.1 10 >1.2 154 Chaka Ridge 0 0 4 0.2 7.9 40 <0.3 SPS-1 Chaka Ridge 0 0 6 0.1 8.3 40 >0.3 435 Chaka Ridge 10 0 2.2 3.3 8 40 <0.3 222 Chaka Ridge 15 0 4.7 0.8 7.9 45 >0.6 SPS-2 Chaka Ridge 16 0 7.2 0.3 7.5 60 >0.6 Spring Chaka Ridge 12 0 3.6 0.7 7.9 40 ˜0.6

4.2 Petrography of Conglomerate and Sandstone Unit in Drilled Core

Core hole 26-11 drilled through 15m of colluvium, 194m of debris flow, and 914m of a sedimentary unit before reaching a total depth of 1216m. It hit fault zones at depths of 576-577m, 616-617m, 786-789m and

908-983m. Water was encountered at 575m and 789m depth at <18 L/min. Water flow was brief, however,

and it was not possible to recover any samples. The highest measured temperature was 57￿ between 1036-1097m depth (Ormat Nevada, Inc., 2012a).

30 Table 4.2: Water chemistry for samples at Mt. Spurr collected in this research. Concentrations of cations, anions, and major species are reported in mg/L, alkalinity in (mg/L as CaCO3), and conductivity in (￿mhos/cm). Isotopes are reported in units of per mil (‰). Oxygen and hydrogen isotopes are relative VSMOW, sulfur relative to VCDT, and carbon relative to VPDB. Green samples are located in the South Spurr field area, red samples are from Crater Canyon, and blue samples are from Chaka Ridge.

18 34 13 Sample Temp pH Alk TDS Cond Na Ca K Li Mg Cl HCO3 SO4 B F SiO2 Al As Ba Fe Mn Sr H2S δ O δD δ S δ C (￿) Canyon 3.8 6.43 54 300 520 200 32 5.3 17 41 65 110 0.23 50 0.24 0 -18.5 -141.7 4.4 -10.5 Canyon 3.8 6.44 54 310 520 25 33 5.2 17 41 65 110 0.24 48 0.24 -18.6 -142.4 4.3 (replicate) Stream 9.1 6.79 11 140 200 5.6 20 1.4 4.6 2.5 13 56 28 0.013 0.11 0 -16.5 -126.2 10.2 -9.3 945 0 6.14 10 6.6 1.6 0.63 0.018 0.012 -20.9 -157.9 -20.8 CC Head 6.9 5.72 23 20 1.9 6.7 4.3 0.02 0 -20.1 -153.8 0.5 626 22.3 8.37 440 1200 1900 140 71 51 0.24 88 230 520 230 3.8 92 0.013 0.018 0.057 0.57 0 -18.5 -148.1 8.4 -3.0 626 22.3 8.33 440 1200 1800 150 76 52 0.25 89 210 530 220 4.1 0.46 98 0.013 0.02 0.062 -18.5 -147.4 9.0 (replicate) 320 12.2 5.89 1.4 74 120 3.3 11 1.1 2.3 1.8 43 0.13 18 0.092 0.011 0.021 0.016 <0.10 0 -19.6 -151.1 0.5 Crater Creek 14.5 7.88 340 830 1400 90 67 27 0.15 66 140 410 150 2.5 0.35 78 0.022 0.016 -19.5 -152.3 8.1 -3.2 SP001 30.2 7.5 280 1400 2400 210 97 45 0.42 80 470 340 240 5.4 130 0.022 0.044 0.032 0.34 0.92 0 -18.3 -147.3 -6.0 532 22 7.52 350 1100 1800 140 82 37 0.27 75 260 430 180 3.8 0.44 97 0.032 -19.1 -151.2 8.7 -5.0 154 0.2 6.24 2.2 20 34 1.1 2.5 0.66 0.59 2.6 9 7.9 435 3.3 6.24 1.5 27 32 0.89 2.8 1.8 9.7 9.3 -18.1 -137.2 -15.9 222 0.8 6.51 2.9 27 41 1.4 3.8 0.54 0.51 3.5 12 12 -18.6 -140 0.2 -6.5 SPS-1 0.1 6.47 3.2 11 37 1.4 3.4 0.67 3.9 8.9 11 0.022 -18.1 -136.7 0.6 -6.7 SPS-2 0.3 6.77 5.4 33 57 3.4 2.6 1.4 1.6 6.6 15 16 -18.8 -141.6 2.2 -8.3 Spring 0.7 6.51 2.7 33 37 1.2 3.3 0.55 3.3 10 11 -18.0 -135.1 8.5 -9.2

31 Table 4.3: Published water chemistry at Mt. Spurr. Concentrations of cations, anions, and major species are reported in mg/L, alkalinity in (mg/L as CaCO3), and conductivity in (￿mhos/cm). Isotopes are reported in units of per mil (‰). Oxygen and hydrogen isotopes are relative to VSMOW, sulfur relative to VCDT, and carbon relative to VPDB. Green samples are located in the South Spurr field area, red samples are from Crater Canyon, blue samples are from Chaka Ridge, samples with asterisks are from Crater Lake, and black samples are from Kid Canyon. References are as follows: [1] Motyka & Nye, 1993 [2] Keith et al., 1995 [3] Neal et al., 2008 [4] Ormat Nevada, Inc., 2012b

18 Sample Temp pH Alk TDS Cond Na Ca K Li Mg Cl HCO3 SO4 B F SiO2 Al As Ba Fe Mn Sr δ O δD Ref (￿) 945 10 8 1000 116 0 116 [4] Spurr Knob 6.9 7.7 12 2.2 2.4 4.2 19 27 0.14 36 [4] SP002a 18 4.9 38 47 3.2 6.4 1.5 2.8 7.2 49 8.3 0.055 6.6 0.069 [4] SP002b 3 3.7 260 660 17 96 7.6 46 1.8 370 0.56 75 130 0.011 0.55 130 2.8 0.78 [4] 06SPCN001A 12 8 116 122 0.54 [3] Crater Peak 40.2 6.4 1708 266 95 75 0.42 99 254 622 477 7.7 1 125 0.021 0.68 -16.7 -138 [1] Hot Springs Upper Crater 23 7.63 320 238 150 0.56 [3] Creek Lower Crater 27.3 8.05 370 493 260 0.4 [3] Creek SP001 10 41 7.04 310 1500 2100 210 100 45 0.4 79 470 320 240 5.7 0.61 130 0.12 0.03 0.051 0.46 0.62 0.91 [4] SPS-1 10 1 6.92 4.8 42 57 3.5 2.7 1.7 1.7 1 5.8 15 17 SPS-2 10 0 6.56 4.2 40 37 1.7 3.5 1 0 1 5.1 9.7 12 67-34 30 8.44 80 600 150 34 29 12 18 9.9 98 13 0.42 100 46 0.4 59 1.1 0.23 CP92-1* 49 2.48 56 184 3.51 0.034 40.5 53.2 0 1450 3.36 0.42 263 41.8 0.16 33.2 2.95 -7.2 -116.0 [2] CP92-2* 49 2.49 55.1 187 3.69 0.035 41.3 54.5 0 1490 3.18 0.42 312 50 0.26 31.6 2.94 -7.2 -113.5 [2] 70AR-201* 2.17 75 192 7.5 71 1180 0 442 24 [2] 70AR-202* 2.3 75 182 7.3 70 1100 0 315 24 [2] KID1 16 7.85 170 620 1000 89 55 12 0.16 35 140 200 130 0.2 0.52 97 18 0.0055 0.064 5.8 0.079 0.45 [4] SP013 1 7.54 30 200 230 36 80 5.2 9.8 25 36 40 0 150 100 0.24 13 0.34 0.77 [4] SP014 11 7.94 97 440 670 75 87 10 0.13 35 83 120 89 0.13 0.31 150 80 0.22 27 0.36 0.78 [4] SP030 16 6.92 200 540 900 72 51 7.2 0.18 52 92 250 110 0.22 0.35 64 0.066 0.011 0.38 [4] SP031 17 6.84 180 540 810 60 47 6.5 0.16 46 82 210 110 0.16 0.46 62 0.34 [4]

32 The sedimentary unit encountered is comprised primarily of conglomerate (82% of total core drilled) but is interbedded with sandstone (15%), which is sometimes gradationally associated with fine-grained shale or mudstone (3%) (Figure 4.1). For the most part, the conglomerate is highly competent. Zones of low competency rock that contain white-green montmorillonite and have slickenlines or calcite deposited along fracture surfaces (Figure 4.1) are interpreted as fault zones.

The conglomerate contains well rounded gravel-cobble sized volcanic (andesite 39%, dacite 7%, pumice

2%, basalt <1%), plutonic (granite 10%, seyenite 10%, diorite 9%, gabbro 2%), and sedimentary (chert

5%, shale 11%, mudstone 4%), as well as minor (<1%) metamorphic clasts that are poorly sorted and are usually matrix supported. The clasts range from being relatively fresh looking to being heavily weathered and friable, iron stained, chloritized (Figure 4.3), rimmed by hematite or calcite, displaying a color zonation from clast edge to center (Figure 4.2), having vugs filled with white montmorillonite, having crystals replaced by montmorillonite (Figure 4.4), being sheared or fractured (Figure 4.7), or being cross-cut by white calcite veins (Figure 4.5). The conglomerate matrix is composed of subangular fine-coarse grained lithic sandsone (plagioclase, quartz, lithic fragments) that is poorly-moderately well cemented by calcite, white-green montmorillonite, or silica. Calcite is ubiquitous and fills voids in sandy matrix, weathered clasts, and fractures that cut clasts (Figure 4.5). Epidote in conglomerate clasts is rare in thin sections, and appears to be formed by the alteration of plagioclase (Figure 4.6). Fragments with relict vesicular texture that is now replaced by montmorillonite and calcite are also found in conglomerate matrix and sandstone (Figure 4.4). The sandstone beds have a similar nature to the conglomerate matrix, but also contain biotite and sometimes black, solid residual hydrocarbons up to 3cm thick. Sandstone is deposited in massive beds, fining upward beds, with mud drapes, or is interbedded with laminated shale or massive mudstone. Deformed detrital grains (Figure 4.4, Figure 4.7) and fractured quartz grains (Figure 4.7) suggest compaction and mechanical crushing processes have occurred after deposition.

4.3 Whole Rock and Clay Mineral Characterization

XRD reflections in the 3-40° 2Θ range were collected for all 7 samples, and 3 samples were also scanned in the 59-63° 2Θ range (Figure 4.8-Figure 4.14). The different colored reflections in Figures 4.8-4.14 repre- sent scans done under different conditions. Black=whole rock, randomly oriented, blue=clay sized fraction

(oriented), red=glycolated clay-sized fraction, and green represents the 060 scan at higher 2Θ. All of the clay-rich samples (“1865” Figure 4.8, “1889” Figure 4.9, “2061.5” Figure 4.10, “2580.5” Figure 4.11, “2583”

Figure 4.12, “3221” Figure 4.13, and “3434” Figure 4.14) analyzed in this research are quite similar, except for sample “3434”.

33 Table 4.4: X-Ray Diffraction Results

Sample Depth Plagioclase Quartz Calcite Smectite Chlorite Kaolinite 060 060 (m) feldspar Scanned Scanned (Mont- (Chlorite) moril- lonite) Characteristic 3.32, 3.19, 4.27, 3.03 001 at 001 at 001 at 1.50-1.49 1.55-1.54 peaks 3.18 3.33 16.97 14.71, 7.16 (d-spacing) (gly) 002 at 7.10 1865 568.5 1889 575.8 2061.5 628.3 2080.5 634.1 2583 787.3 3221 981.8 3434 1046.7

34 Mineral constituents common to all the clay-rich samples include quartz, plagioclase feldspar, and discrete smectite (Table 4.4). Sample “3434” also contains discrete calcite, chlorite, and possibly also kaolinite. 060 reflections measured in three smectite-bearing samples show peaks in the 61.7-62.2° 2Θ (d-spacing 1.50-1.49) range, which corresponds to a dioctahedral phase of smectite. A dioctahedral smectite mineral could be montmorillonite or beidellite; montmorillonite is more common and is assumed to be the dominant phase for the purposes of this research. Sample “3434” has a reflection at 7.07A˚ that could represent chlorite (002), or kaolinite (001), or both. No other samples contained any evidence for illite, kaolinite, or mixed-layer clays.

Additionally, reflections for sample “3434” could be experiencing overlap between the montmorillonite (060) and a quartz peak. The other samples for which the (060) was measured don’t have significant quartz.

The vein scraping predominately contains calcite, with some minor gypsum (Figure 4.15). Reflections from the red mineral scraping include quartz, plagioclase feldspar, siderite, and minor geothite, magnetite, and hematite signatures. The peaks at 32.4° and 46.4° are attributed to siderite, and are slightly shifted from the reference, indicating the possibility that the siderite observed represents an intermediate composition in the solid solution between Fe and Mg.

4.4 Isotopes of Calcite in Veins, Rims, and Cements

Results from the calcite isotope analyses are summarized in Table 4.5. Samples 3049r, 3221c and 1889c contained less than 1% CaCO3 and had to be re-run with more sample material. Sample 1889c did not contain enough calcite to analyze even then, and no results were obtained.

Table 4.5: Stable oxygen and carbon isotopes of calcite samples from 62-11 core.

13 18 Sample Depth (m) Type δ CVPDB δ OVSMOW 1638r 499 Rim -8.12 12.54 2227m 679 Matrix -5.98 24.87 2227r 679 Rim -6.74 19.74 2227v 679 Vein -5.95 13.84 2630r 802 Rim -6.76 12.54 2630v 802 Vein -6.58 12.53 3049r 929 Rim -6.47 24.37 3221c 982 Cement -7.70 23.10 3419v 1042 Vein -9.12 13.31 3434v 1047 Vein -10.61 14.16

35 (a) Unaltered conglomerate from 206m depth. (b) Unaltered sandstone from 218m depth.

(c) Calcite along fracture surface at 1042m depth.

Figure 4.1: Core drilled over the summer of 2011 in hole 62-11 consists mainly of conglomerate and sandstone units. Some calcite mineralization is seen along fracture surfaces.

36 (a) Dark rimming around the edges of a clast at 245m (b) White rimming around the edges of a clast at depth. 671m depth.

(c) Calcite replacement of conglomerate matrix at 705m depth.

Figure 4.2: Alteration in clasts and matrix of conglomerate unit in core hole 26-11.

37 Figure 4.3: Chlorite replaces primary igneous minerals in sample “3434” conglomerate clast. Chlorite and calcite also fill a void that has been cut by a calcite vein. Photomicrograph taken in plane light.

38 Figure 4.4: Masses of a brown clay mineral intergrown with calcite are common. a.) The clay mineral and calcite in conglomerate matrix and b.) a close up of the intergrown texture, both in sample “2630.” c.) Some of the original vesicular texture is preserved in this replacement from the conglomerate matrix in “2227.” d.) Calcite and the clay mineral in conglomerate matrix in sample “3419.” The shapes of these calcite and clay mineral masses also indicate compaction. All photomicrographs were taken with crossed polars.

39 Figure 4.5: a.) Calcite fills voids around conglomerate clasts and b.) replaces primary textures of conglom- erate clasts in sample “2630.” Calcite also occurs c.) as veins that cross-cut matrix grains in “3419”, and d.) in the pore space between matrix grains in “682”. All photomicrographs taken with crossed polars.

Figure 4.6: a.) Epidote replaces plagioclase feldspar in conglomerate clasts in “682”, and b.) in matrix grains of “3557”. Both photomicrographs were taken with crossed polars.

40 Figure 4.7: Evidence of mechanical stress includes a.) and b.) fractured quartz grains with calcite in fractures from “3434” conglomerate matrix, and c. and d.) biotite grains that have been smashed around other grains in the matrix of “3557” and “2227” respectively. The quartz grains could indicate faulting, whereas the biotite grains could indicate compaction. All photomicrographs were taken with crossed polars.

41 Figure 4.8: XRD reflections for sample “1865”, with 2Θ (degrees) along the x-axis, and intensity (counts per second) along the y-axis.

Figure 4.9: XRD reflections for sample “1889”, with 2Θ (degrees) along the x-axis, and intensity (counts per second) along the y-axis.

42 Figure 4.10: XRD reflections for sample “2061.5”, with 2Θ (degrees) along the x-axis, and intensity (counts per second) along the y-axis.

Figure 4.11: XRD reflections for sample “2580.5”, with 2Θ (degrees) along the x-axis, and intensity (counts per second) along the y-axis.

43 Figure 4.12: XRD reflections for sample “2583”, with 2Θ (degrees) along the x-axis, and intensity (counts per second) along the y-axis.

Figure 4.13: XRD reflections for sample “3221”, with 2Θ (degrees) along the x-axis, and intensity (counts per second) along the y-axis.

44 Figure 4.14: XRD reflections for sample “3434”, with 2Θ (degrees) along the x-axis, and intensity (counts per second) along the y-axis.

Figure 4.15: XRD reflections for sample “3434” calcite and red mineral scraping, with 2Θ (degrees) along the x-axis, and intensity (counts per second) along the y-axis.

45 CHAPTER 5

INTERPRETATION: AQUEOUS AND GAS GEOCHEMISTRY

This chapter analyzes aqueous and gaseous geochemical data collected in this research as well as previous

work. A theoretical framework for interpreting dissolved constituents and isotopes in water samples, as well

as the composition of fumarole samples from Crater Peak, is presented.

5.1 Dissolved Chemical Species

Dissolved constituents commonly found in geothermal waters inlcude Cl, HCO3,SO4, F, I, Br, Na, K, Ca,

Mg, Rb, Cs, Li, Mn, Fe, Al, NH4, NH3, As, SiO2,B,CO2, and H2S. Differences in concentrations of these constituents in different geothermal systems are caused by differences in temperature, gas content, recharge source, magmatic input, rock type, water to rock ratio, duration of water-rock interaction, boiling and mixing (Simmons, 2008). Most of the cations previously mentioned are commonly found in rock-forming minerals, and their concentration in water is controlled by water-mineral equilibria and a preference for retaining the element in the mineral phase (Simmons, 2008). Some elements, such as Cl, B, Li, and Br, are more abundant in the liquid phase than in minerals, and are called “conservative” because they are conserved in the liquid phase and do not readily participate in reactions (Simmons, 2008). Most soluble constituents in geothermal fluids can be derived from water-rock interaction involving various types of igneous and sedimentary rocks, but magmatic sources for some constituents are also possible (Ellis & Mahon, 1964). Furthermore, an identical surface discharge composition can be produced by completely different processes (Ellis & Mahon, 1964). Context dictates the likelihood of certain processes being dominant over others. When deep geothermal well discharge chemistries have been compared to surface discharge chemistries, it has been shown that the concentrations of chemical constituents in underground reservoir systems that feed surface discharges are not the same as surface discharges (Ellis & Mahon, 1964). Nevertheless, surface discharge chemistry can sometimes be used to reconstruct possible reservor compositions. The concentrations of cations, anions, major species, and isotopic composition of a surficial water can reveal information about its recharge origin, mixing and boiling processes, possible interactions in the shallow subsurface, and equilibrium with rock. These factors are important for determining the suitability of applying geothermometry equations and understanding processes that affect fluid compositon. Most geochemical interpretation techniques can only be applied with confidence to specified types of fluids with limited compositional ranges (Giggenbach

& Goguel, 1989).

46 A fluid existing at depth may acquire chemical constituents by absorption of magmatic vapors and water-

rock interaction (Giggenbach, 1988). A hot fluid rising rapidly from depth will react with the wall rock during

ascent until it reaches a depth where a rapid decrease in pressure allows phase separation (boiling). In an

ideal, closed system, the fluid will travel to the surface and emerge as a two phase vapor and liquid discharge.

In reality, some of the vapor may be lost into the formation, or dissolved in shallow bodies of water, or the

fluid may mix with waters of a different source (Giggenbach & Stewart, 1982). Some chemical species, like

CO2, will preferentially separate into the vapor phase upon boiling, while others, like silica, will remain in the liquid phase. A slowly rising fluid will cool conductively and probably boil, whereas a very rapidly rising

fluid may not boil until it reaches the surface. If the ascent is rapid, water-rock reactions during ascent are minimal and the original deep chemistry will be preserved. If boiling, mixing, or water-rock interaction does occur, the chemistry of waters observed at the surface will not represent the original deep chemistry, but it will deviate from it in a predictable way (Giggenbach, 1988). In order to evaluate the characteristics of a deep fluid based on surface discharges, it is necessary to understand the processes that may be generating initial fluid chemistry as well as those that affect that chemistry from the fluid’s depth of origin to the surface. The importance of rock dissolution and equilibration in controlling water chemistry can be assessed by comparing geothermal discharges with two theoretical endmember compositions derived from 1) isochemical dissolution of crustal rock in acid fluids or 2) equilibration with thermodynamically stable mineral assem- blages resulting from isochemical recrystallization of rock (Giggenbach, 1988). Waters from hydrothermal systems are rarely in equilibrium with rock, and their composition is influenced by the initial fluid composi- tion as well as the kinetics of mineral dissolution and precipitation. Since the four major cations in crustal rocks and geothermal waters are Na, K, Mg, and Ca, the differing thermodynamic properties and relative concentrations of these cations can be used to assess the state of rock dissolution and water-rock equilibration (Giggenbach, 1988). It is important to be able to distinguish between water-rock interaction that has occured at depth versus water-rock interaction that has occured at or near the surface. If interaction has occured at depth, then the chemistry of the water can be used to back calculate original deep conditions. Conversely, if interaction has occured at shallow levels, the water chemistry represents only the shallow system, not the source. Crystalline minerals in rocks can liberate appreciable amounts of water-soluble constituents without displaying much evidence of hydrothermal alteration. A breakdown of crystal structure is not necessary for the liberation of constituents like Cl, F, B and NH4 because they are not only contained in silicate minerals, but also exist on the surfaces of grains, crystals, and microfissures in rock. However, all trace elements in rocks are not removed with the same ease, and elevated temperature can cause preferential extraction. Trace elements held within silicate minerals, like Li, Rb, and Cs, can only be released when the silicates decompose

47 or are recrystalized (Ellis & Mahon, 1964).

5.1.1 Cl-HCO3-SO4

The relative amounts of chloride, sulfate, and bicarbonate allow the characterization of water type,

source components, and influence of shallow processes (Figure 5.1). In a volcanic setting, these components

likely originate from the magma chamber, but CO2 and SO4 could also be derived air or organic sources, respectively. Linear trends on this type of ternary diagram represent mixing lines, and data plots at positions

that represent ratios, not total concentrations (Giggenbach & Goguel, 1989). Chloride in water is originally

derived from the incorporation of magmatic HCl gas into the water (Giggenbach & Goguel, 1989). Upon

ascent from depth to the surface, aqueous HCl is neutralized by the surrounding rock and can precipitate

as NaCl, or stay dissolved in water. High chloride water is thus interpreted to represent well-equilibrated fluid from an upflow zone, and deep geothermal well discharges have compositions similar to those of neutral

chloride springs (Giggenbach & Goguel, 1989). Sulfate is derived from oxidation of H2S vapors, and typically is associated with acidic waters and steam heating processes, or by mixing with sulfate-rich surface waters

(Giggenbach & Goguel, 1989). Sulfate originates from magmatic SO2 or H2S, and is typically low in deep

geothermal well discharges. Bicarbonate is derived from incorporation CO2 vapors, and the amount of HCO3

in a water sample depends on the pH and the pCO2 of the solution. High HCO3 waters are typically associated with the periphery of an upflow zone (Giggenbach & Goguel, 1989). The presence of high concentrations of sulfate and bicarbonate are thus associated with shallow interactions and processes that modify the original deep fluid chemistry. The water type most suitable for application of geochemical interpretation techniques is low sulfate neutral chloride water, which plots on the ternary near the Cl corner, or just below it along

the Cl-HCO3 edge in the “Mature Waters” field (Giggenbach & Goguel, 1989).

There is a large difference in Cl and SO4 contents between the Crater Lake samples (orange triangles) from 1970 (samples “70AR-201” and “70AR-202”), when the hydrothermal system should have been re- established after the 1953 eruption, and 1992 (samples “CP9201” and “CP92-2”), prior to eruption. Keith et al. (1995) suggested that the high Cl content in 1970 waters could be due to Cl remaining dissolved in evaporated water while sulfate mineral phases had precipitated. The high SO4 content in 1992 water is likely due to the increased sulfur degassing from rising magma in the Crater Peak vent and disproportionation

of magmatic SO2 to H2S and SO4 (Keith et al., 1995). Crater Lake was not accessible during the field campaign for this research, but if the hydrothermal system has been re-established in the 10 years since the

1992 eruption, which seems like a reasonable assumption, the composition of current Crater Lake water could be similar to the 1970 composition. A linear trend between the Crater Canyon samples can be extrapolated to the 1970 Crater Lake samples. Given that Crater Canyon samples are directly south and topographically

48 Figure 5.1: A Cl-HCO3-SO4 diagram shows that most of the samples from Mt. Spurr have high sulfate and bicarbonate concentrations relative to chloride concentrations, which indicates that shallow processes are inflencing the chemistries of the samples. A linear trend from 1970 Crater Lake waters (orange diamonds) to Crater Canyon waters (red circles) suggests a possible mixing trend. The more sulfate-rich composition of the remaining samples could be due to the very dilute composition of samples or the addition of sulfate through steam heating. Samples CP92-1 and CP92-2 plot in the same location.

49 lower than Crater Lake, it is possible that acidic Crater Lake waters simply percolate downhill and re-emerge

in Crater Canyon, interacting with volcanic rock to become more HCO3-rich. This model has been proposed by Motyka & Nye (1993), and will be revisited in light of the other data sets of this research. Sample “320”

and “CC Head” don’t plot with the other Crater Canyon samples, so these are interpreted to have undergone

different processes.

Kid Canyon samples (black crosses) plot near the thermal Crater Canyon samples (red circles). The

HCO3 in these samples could be derived from magmatic CO2 that has dissolved in cold water. Neal et al. (2008) and Martini et al. (2011) speculated that waters emerging from the mouth of Kid Glacier and those

in Kid Canyon could have a Crater Peak source. The samples from Chaka Ridge (blue triangles), except for

sample “67-34”, have Cl concentrations below the detection limit and thus plot in the Steam-Heated Waters

field. The sulfate in these dilute samples could be from mixing with an initially acid-sulfate type water that has been neutralized by water-rock interaction, or by contributions to gas seeps that may exist along the flow path. Sample “67-34” is the only water sampled from the subsurface, and was encountered in the exploration core hole of the same name. Near-equal relative contributions of all three anions suggest mixed

processes affect water chemistry, possibly incorporation of dissolved magmatic steam (CO2,H2S or SO4, and HCl) or interaction with an intermediate body of water containing these components. The composition of samples from South Spurr (green squares) occupies an intermediate position and could be produced by mixing between a fluid near that of Crater Canyon composition and a fluid near that of Chaka Ridge composition. Overall, anion concentrations of waters at Mt. Spurr indicate that they are not well-equilibrated fluid from an upflow zone, but rather represent waters peripheral to the main upflow of a potential hydrothermal system or shallowly circulating meteoric (rain, snowmelt) waters that have interacted with groundwater or rock before emerging at the surface.

5.1.2 Na-K-Mg

Relative concentrations of Mg, Na and K can be plotted to examine water-rock equilibration and rock leaching processes (Figure 5.2). The diagram is based on ionic solute geothermometers (discussed later in this section) that use the temperature dependence of Eqs 5.1 and 5.2 from Ellis & Mahon (1964) and Fournier & Truesdell (1973), respectively. These equations involve feldspar minerals of the full equilibrium assemblage expected to form after isochemical recystalization of an average crustal rock (Giggenbach & Goguel, 1989), which is interpreted as hypothetical “deep” rock that contains a geothermal reservoir.

K feldspar + Na+ = Na feldspar + K+ (5.1) − −

2.8K feldspar +1.6water + Mg2+ =0.8K mica +0.2chlorite +5.4silica +2K+ (5.2) − −

50 Because Na, K and Mg have non-linear thermodynamic correlations, and the concentrations of these species in geothermal waters typically vary by orders of magnitude, multipliers are necessary for ternary display. For processes that only vary in relative Mg concentrations, mixing can still be represented by a straight line (Giggenbach, 1988). Geothermal well discharges typically plot on the full equilibrium line (top curve inside the ternary diagram), and have very low Mg contents. Conversely, spring waters typically plot at lower temperatures not on the equilibrium line, and show an increase in Mg with decreasing temperature that is faster than aquisition of Na (Giggenbach, 1988). Data that plots near the Mg corner in the “Immature

Waters” field (basically groundwater) is not suitable for the application of K-Na and K-Mg geothermometers because it isn’t a deeply sourced fluid that would have preserved a chemical composition consistent with deep feldspar equilibrum, whereas waters that plot in the “Partial Equilibrium” field should yield realistic

K-Mg temperatures because the equilibrium assumption is met (Giggenbach & Goguel, 1989).

Figure 5.2: Na-K-Mg diagram showing that waters from Mt. Spurr cluster in the Mg corner, as would be expected for cold groundwater that has poorly equilibrated with rock.

The samples from this and previous research plot in the Mg corner, in the “immature waters” field (Figure 5.2). The sample that has the highest K and Na concentration relative to Mg is “Crater Peak Hot Springs.” This clustering in the Mg corner suggests influence of groundwater, which typically has higher

Mg concentrations than deep geothermal wells. Magnesium from a deep reservoir re-equilibrates relatively quickly upon ascent as temperature drops and alteration minerals precipitate, leaving the remaining fluid relatively depleted in Mg (Giggenbach, 1988). Figure 5.2 indicates that the surface waters at Mt. Spurr are

51 not influenced by equilibration with presumably deep, feldspar-bearing rocks.

5.1.3 Cl-Li-B

A Cl-Li-B ternary diagram can be used to examine the possible origin of Cl and B in geothermal waters

(Figure 5.3). Chloride and boron, which are conservative constituents, are likely derived from incorporation

magmatic volatiles (HCl and H3BO3); Li must come from rock dissolution (Giggenbach & Goguel, 1989).

Lithium is inert for the most part and remains in solution. At high temperatures, HCl and H3BO3 are volatile and highly mobile in steam, and could be absorbed into fluid to generate an acid brine capable of leaching rock and releasing Li from silicate mineral structures into solution. As temperature decreases, both acids are neutralized by wall rocks. Some Cl can be removed by the conversion of HCl to NaCl, KCl, or

MgCl2, but the remainder is dissolved and stays in solution. Boron will remain in gas form upon ascent and can be carried in the vapor pase even at low temperatures (Giggenbach & Goguel, 1989).

Figure 5.3: Cl-Li-B diagram showing that Mt. Spurr waters can be segregated into two general groups: high Li waters that have obtained Cl and B by absorption of magmatic volatiles from a deep hydrothermal system, and low Li waters that do not represent deeply-derived fluids.

The thermal samples from Crater Canyon, 1992 Crater Lake waters (orange diamonds), “Canyon” (green

square near top of diagram) and “KID1” (black plus sign) have higher Cl and B concentrations relative

to Li, suggesting that processes of absorption of magmatic volatiles are more important that processes of

rock dissolution for these samples. The 1992 Crater Lake samples have the most B, while thermal Crater

52 Canyon samples have less B but more Cl, and “KID1” has the highest Cl:B ratio. The remaining Mt. Spurr waters plot near the field of igneous rocks, indicating reactions with igneous rock are the dominant process controlling chemistry.

5.1.4 Chloride vs. Enthalpy

Chloride concentrations can be used in conjunction with temperature data to speculate about boiling and mixing reactions (Fournier, 1979). In developed geothermal fields, where enthalpy and chloride composition of a deep fluid can be measured directly, it has been shown that enthalpy decreases as chloride concentra- tion increases, and that this is a linear function of boiling (Figure 5.5). Measured water compositions fall underneath a linear trend (boiling line) from the original reservoir composition to boiling water at 100°C at positions that reveal possible mixing relationships with a groundwater endmember that has zero Cl and enthalpy content (Henley & Ellis, 1983).

Figure 5.4: Chloride vs. enthalpy figure modified from Fournier (1991), showing the hypothetical composition of steam and effects of cold water mixing on a deep reservoir fluid.

Enthalpy was calculated for the samples of this study using steam tables with the measured surface temperature and a pressure of 1 bar. Results are plotted in Figure 5.5. A generally linear trend can be seen from South Spurr (green squares) to Kid Canyon (black plus signs) to Crater Canyon (red circles) waters, between some of the same samples that showed mixing trends in Figure 5.1. This mixing line hypothetically continues to higher chloride and enthalpy values that will eventually intersect a point on the boiling line that represents a composition between the (unknown) original reservoir fluid and that same fluid after steam loss from boiling (Figure 5.4). This positively sloping linear trend mimics the cold water mixing trend on

Figure 5.4, and provides evidence of cold water mixing at Mt. Spurr. Crater Canyon samples are the least

53 diluted and South Spurr samples are the most diluted by cold water. Crater Lake (orange diamonds) and

Chaka Ridge waters (blue triangles) do not fall on this mixing trend, indicating that they are not affected

by the same processes. Crater Lake and Chaka Ridge compositions could be produced by mixing with a

different cold water endmember, or by different amounts of steam loss prior to or after dilution.

Figure 5.5: Enthalpy-chloride plot with linear trend between South Spurr and Crater Canyon samples.

5.1.5 Silica Saturation

The concentration of dissolved silica is controlled by the solubility of different silica phases such as chalcedony, quartz, cristobalite, opal, amorphous silica, and volcanic glass. When two or more silica phases are in contact with a given solution, the most soluble phase controls silica until that phase completely dissolves, is converted to another more stable phase, or is taken out of solution by alteration products or precipitates. Silica precipitation is minor upon rapid ascent from depth, but can occur as temperature falls during conductive cooling, or when hot fluid mixes with colder water Fournier (1985). Silica solubility fields in Figure 5.6 were calculated using the silica geothermometers of Fournier (1981) (Table 5.3). Silica contributions to this system are likely largely from volcanic glass. The waters with the highest silica concentrations are the Crater Lake waters (red circles) (Figure 5.6). Since Crater Lake waters are acidic, the silica could be derived from leaching of volcanic rocks and rock dissolution processes. Of the remaining Mt. Spurr samples, Kid Canyon (black plus signs) and Crater Canyon (red circles) generally have

54 Figure 5.6: Sampling temperature vs. silica concentration diagram with the solubility curves for different silica phases. Chalcedony is probably the dominant form. Extrapolating observed silica concentrations to the chalcedony solubility curve yields temperature estimates higher than those measured in the field. more silica than South Spurr (green squares) and Chaka Ridge (blue triangles). These samples likely derive silica from interaction with rock, and decreasing amounts of silica are likely due to dilution or interaction at a lower temperature. The relatively low field temperatures of all samples from Mt. Spurr discharges indicate that chalcedony is probably the controlling silica phase. Most samples are saturated or supersaturated with respect to chalcedony, excluding “945” and “CC Head”. Extrapolating observed silica concentrations to the chalcedony solubility curve yields temperatures higher than those measured in the field, suggesting that Mt. Spurr samples have equilibrated with silica at warmer temperatures before discharge. The chalcedony equilibration temperatures range from ˜25-135￿ for water south of Crater Peak, and up to ˜195￿ for Crater Lake waters. Since some dilution is certainly happening here, the chalcedony temperature estimates represent a minimum temperature of last equilibration.

5.2 Stable Isotopic Composition

Stable isotopes of H and O can be used to determine the recharge origin of a geothermal fluid, while stable isotopes of C and S can be used to determine the origin of carbon and sulfur. Carbon and sulfur isotopic data is unlikely to provide unique interpretations to geochemical problems, and should be augmented with additional evidence from other sources. Since identical concentrations of chemical constituents can be

55 generated by different processes, having corroborating information about where the fluid comes from is

important for understanding dissolved constituents and fluid evolution.

5.2.1 Origin of δ18O and δD in Mt. Spurr Waters

The global isotopic composition of water can be related to processes of evaporation, condensation, evap- otranspiration, and decrease in temperature as one moves north or south of the equator. The extremely depleted values of H and O isotopes at Mt. Spurr (Table 4.2, Figure 5.7) are likely due to high latitude and cold temperature effects. Most Mt. Spurr waters plot near the meteoric water line, indicating that recharge to the system is predominately by rainfall. Samples from Crater Canyon (red circles) are enriched in δ18O and δD and seem to form a positive linear trend.

Figure 5.7: Stable oxygen and hydrogen isotopes in Mt. Spurr waters. Black diamonds represent cold stream/snow waters from Motyka & Nye (1993). South Spurr and Chaka Ridge samples plot close to the meteoric water line and the meteoric waters, while some Crater Canyon samples (red circles) show a slight enrichment in δ18O. A linear trend between Crater Canyon samples eventually intersects the range of “andesitic water” (Giggenbach, 1992). The isotopic composition of Crater Canyon waters could be produced by mixing of meteoric water with “andesitic water”. The isotopic composition of Crater Lake waters (orange diamonds) does not lie on this trend, and likely reflects evaporation processes.

Several processes may be responsible for deviations from the meteoric water line: water-rock interaction, boiling, evaporation, and/or mixing. Rocks typically have more enriched δ18O signatures than meteoric

56 water, so isotopic exchange between water and rock will enrich the water and deplete the rock. Since oxygen

is generally more common in rock-forming minerals than hydrogen, water-rock interaction enriches δ18O more than δD, typically producing a horizontal shift to the right of the meteoric water line (Craig, 1963).

The average δ18O composition for SVC lavas is 7.25‰ (Nye & Turner, 1990), and water-rock interaction could be enriching Spurr waters. Boiling would produce a concave up curved trend (Giggenbach & Stewart,

1982), and is probably not important because the observed trend is more linear than curved. Evaporation could cause enrichment in both δ18O and δD, and likely influences the isotopic composition of Crater Lake, since the pool has a surface area exposed to the atmosphere and is not flowing. Evaporation is probably not

an important process affecting the isotopic composition of the springs and seeps since the water is flowing, the

air is relatively humid, and the field temperatures of surface discharges were relatively low. Mixing is another

way to simultaneously enrich δ18O and δD, and mixing of meteoric water with an enriched endmember, such

as andesitic water (δ18O=10‰ and δD=-20‰ Giggenbach (1992)), could cause the shift observed in the Crater Canyon samples.

5.2.1.1 Hot Water Fraction of Crater Canyon Discharges

Motyka & Nye (1993) present a conceptual model (Figure 5.8) of fluid flow at Mt. Spurr in which Crater Lake waters percolate downslope to form Crater Canyon waters. The isotopic data collected in this research suggests that andesitic water (as opposed to exclusively Crater Lake and meteoric waters) could play an important role in the final observed spring chemistry, because the enrichment trend of the Crater Canyon samples does not line up with the isotopic composition of Crater Lake waters. Fractions of Crater Lake water and andesitic water for samples in Crater Canyon were calculated with δ δ x = spring − meteoric (5.3) δ δ hot − meteoric where x is the hot water (i.e. Crater Lake or “andesitic” water) fraction, and δspring,δmeteoric,δhot rep- resent the δ18O and δD values of spring, meteoric and hot waters (Giggenbach, 1992). According to δ18O calculations using Eq.5.3, all samples except “Crater Peak Hot Springs” are at least 80% meteoric water (Table 5.1). It is apparent from these calculations that meteoric water contributions to the isotopic composition of Crater Canyon samples are quite large, and that “CC Head” likely represents water melting directly from Crater Glacier. If the evaporation effects of Crater Lake waters could be quantified, it might be possible to better estimate the composition of water percolating downslope. But even if unevaporated Crater Lake water fell on the same enrichment trend as Crater Lake samples, the enrichment trend could still be extrapolated

57 Figure 5.8: Conceptual model of hydrothermal circulation between Crater Peak and thermal springs at Mt. Spurr. Residual magma at unknown depth expels steam, CO2,SO2, HCl and other magmatic gases into overlying hydrothermal layer that was formed by infiltration and accumulation of groundwater. Steam and gases rich in CO2 and H2S are driven off the boiling hydrothermal layer and feed fumaroles at Crater Peak and interact with shallow groundwaters to produce HCO3-SO4-Cl thermal springs at lower elevations. From Motyka & Nye (1993).

Table 5.1: Hot water fraction calculations using δ18O values of Crater Canyon springs and seeps.

Sample % Crater %“Andesitic” % Meteoric Lake water water water Crater Creek 3.91 1.67 94.43 532 7.03 3.00 89.97 626 11.72 5.00 83.28 626R 11.72 5.00 83.28 SP001 13.28 5.67 81.05 Crater Peak 25.78 11.00 63.22 Hot Springs CC Head 0 0 100 320 3.12 1.33 95.54 Calculations made with δ18O=-7.2 and δD=-115 for Crater Lake, δ18O=10 and δD=-20 for andesitic water using Eq. 5.3.

58 to andesitic water.

5.2.1.2 Elevation and Chloride Concentration Effects

The average meteoric water values of δ18O and δD at high elevations and inland from the ocean will be more depleted relative to SMOW than waters at lower elevations and on the coast. A negatively sloped

linear trend between isotopic composition and elevation (Figure 5.9) is observed in the Chaka Ridge and

South Spurr samples. This is the expected result for meteoric water at both field areas. Crater Canyon

samples are near the same elevation, but have more variable isotopic composition. The Crater Lake samples

are more enriched in δ18O and δD than would be expected for their elevation. Deviations from expected meteoric trends at Crater Lake and Crater Canyon suggest that light isotopes are being added to Crater

Lake and the most isotopically enriched Crater Canyon waters, possibly from rock dissolution, exchange

with clay minerals, or addition of andesitic water.

Figure 5.9: Stable isotopic composition of Spurr waters with respect to elevation. Crater Lake samples are more enriched than would be expected, suggesting the addition of light isotopes from rock dissolution and clay minerals, or andesitic water.

Chloride concentration with respect to δ18O and δD is also shown in Figure 5.10. High Cl contents can be interpreted to indicate relatively deeply circulating water that has had minimal near-surface modification, and a deeply circulating water that is well-equilibrated with rock is expected to be enriched in δ18O. A positive linear correlation between Cl concentration and δ18O would be expected for deep geothermal fluid

59 from an upflow zone. The lack of a positive trend among water samples from Crater Canyon (red circles)

and South Spurr (green squares) suggests that the concentration of Cl or isotopic composition is being

modified by mixing with shallow waters with lower Cl concentrations and/or heavier isotopic compositions.

The Cl concentrations of 1992 Crater Lake waters are lower than would be expected for a relatively enriched

isotopic composition. This also suggests that Crater Lake waters are incorporating light isotopes from other

waters or minerals, possibly from rock dissolution, interaction with clay alteration minerals, or the addition

of andesitic water.

Figure 5.10: Stable isotopic composition with respect to Cl concentration, suggesting mixing low-Cl, heavier isotopic composition meteoric water. The Cl concentrations of 1992 Crater Lake waters are lower than would be expected for a relatively enriched isotopic composition, suggesting the addition of light isotopes from rock dissolution and clay minerals, or andesitic water.

5.2.2 Origin of δ13C in Mt. Spurr Waters

Stable C and S isotopes of Mt. Spurr samples and published values are shown in Figure 5.11. The distribution of 13C in high temperature geologic environments is quite variable, and can be influenced by equilibrium or kinetic redox reactions, disparate sources of carbon, or differing proportions of compositionally

distinct carbon in samples. Data are reported in terms of total carbon, where 13C could be composed of

13 13 C-enriched oxidized (i.e. carbonates, CO2) and C-depleted reduced (i.e. organic, CH4) forms of carbon. Mantle derived igneous rocks have a δ13C value near -5‰. The sources of carbon in hydrothermal systems can

60 be complex because the C could come from biogenic or magmatic sources, or could represent C remobilized

through dissolution of alteration minerals such as calcite (Field & Fifarek, 1985).

The δ13C values of waters at Mt. Spurr range from -20.8‰ to -3‰ and are plotted with the isotopic composition of some biogenic and igneous signature ranges for reference in Figure 5.11(a). Most samples

from Crater Canyon cluster around -5‰, suggesting magmatic influence, and this is supported by their proximity to Crater Peak. Samples “222,” “SPS-1,” and “SPS-2” from Chaka Ridge also have δ13C values

near -5‰, but magmatic carbon is likely transported from its source along faults to these sample locations,

or the -5‰ value could be a coincidence. Samples “945” and “435” have more negative δ13C values, which likely indicates contributions of carbon from biogenic or hydrocarbon compounds. The remaining samples

are somewhere in between, and may represent δ13C contributions from mixed sources.

5.2.3 Origin of δ34S in Mt. Spurr Waters

Values of δ34S are reported in terms of total sulfur, where 34S could be composed of 34S-enriched oxidized (i.e. sulfate) and 34S-depleted reduced (i.e. sulfide) forms of sulfur. Sedimentary sulfides have a large range of

δ34S values (-50 to 50‰) that have been experimentally attributed to kinetic fractionations that accompany

34 the reduction of SO4 to H2S. Variations in δ S in either sulfates or sulfides from an individual ore deposit rarely have a range of more than 5-7‰. Magmatic sulfur has a value in the vicinity of 0‰. Like carbon, the isotopic composition of sulfur can be influenced by the dissolution of alteration minerals, for example pyrite. The sulfates of many geothermal and volcanic-hosted environments are of supergene origin, and have

aquired distinctive 34S-depleted sulfate sulfur by near surface oxidation of 34S-depleted hypogene sulfide-

sulfur. Values of δ34S in paragenetic suflide sequences has been observed to become increasingly more depleted in δ34S relative to the δ34S of the total sulfur in the system. (Field & Fifarek, 1985)

The δ34S values of waters at Mt. Spurr range from 0.2‰ to 10.2‰ (Figure 5.11(b)). The samples that have the lowest, most magmatically influenced values are “CC Head,” and “320,” from Crater Canyon, and “222,” “SPS-1” and “SPS-2” from Chaka Ridge. All of these samples are dilute waters without an obvious

connection to a source of magmatic S. Another cluster of values lies at 8-9‰, and the sample with the

34 highest δ S value is “Stream.” The S isotope results must be examined in light of the fact that SO2 and

H2S degassing from the magma body at Mt. Spurr is not well expressed at the surface–the only substantial sulfur-bearing gas flux at Mt. Spurr occurs during eruptions, the rest is scrubbed out by intermediate water bodies (Doukas & Gerlach, 1995). This knowledge, in addition to low Cl/SO4 ratios in waters at Mt. Spurr suggest that it is quite likely that the sulfur isotopes are experiencing some δ34S fractionation due to oxidation of H2S. Therefore, the isotopic signatures obtained do not reflect the isotopic signature of an original deep source but have been affected by shallow fractionation.

61 (a) δ13CvaluesforCraterCanyonclusteraround-5‰,suggestingamagmaticcarboninfluence.Thespread seen in the other samples could be due to contributions of carbon from other sources, such as organics, or from air contamination.

(b) δ34Svaluesarelikelyinfluencedbyshallowfractionationeffects.MtSpurrhasahistoryofsulfate scrubbing except during eruptions, so oxidation of hydrogen sulfide could be causing sulfur-34 fractionation that masks the original deep signature.

Figure 5.11: Stable C and S isotopes from Mt. Spurr water samples. Gray boxes represent ranges for C and S in the environment from Field & Fifarek (1985)

62 5.3 Gas Geochemistry from Crater Peak Fumaroles

Crater Peak was not accessible during the field campaign of this research, and no new samples were

collected. Published gas analyses from fumaroles at Crater Peak are tabulated in Table 5.2 and plotted in

Figure 5.12 and Figure 5.13.

Table 5.2: Residual gas composition (dry basis, mole %) from fumaroles at Crater Peak, 1982, at a sampling temperature of 94°C. The water mole percent of RM Spurr 1 and RM Spurr 2 was not determined, but is 97.9% for RM Spurr 3. From Motyka & Nye (1993).

RM Spurr 1 RM Spurr 2 RM Spurr 3

CO2 96.1 95.3 94.7 H2S 0.18 nd 0.77 SO2 <0.1 nd nd H2 nd 0.38 0.32 CH4 0.022 0.046 0.034 N2 2.93 3.24 4.09 O2 0.04 0.049 0.042 Ar nd 0.040 0.055 He nd 0.0005 bd δ13C -12.4 nd -14.2 R/Ra nd 6.6 nd 13 nd= not determined, bd= below detection limit, δ CfromCO2with respect to PDB

Noble gases are not reactive and can therefore provide information about the source of the gases, whereas

reactive gasses (H2O, CO2,H2S, NH3,H2,N2,CH4) can provide information about subsurface conditions and chemical equilibria (Nicholson, 1993).

5.3.1 N2-He-Ar

An inital classification of gas samples in terms of their source components, mixing relationships, and possible air contamination can be displayed on a N2-He-Ar ternary diagram. Noble gasses (Ar, He, Kr, Ne, and Xe) are found in the atmosphere and may be introduced into a geothermal system through meteoric recharge. He and Ar can also be produced by radiogenic decay of U and Th, respectively, and may enter geothermal fluid by water-rock interaction, magmatic input, or leakage of gases from the mantle (Nicholson,

1993). The source of He is generally attributed to radiogenic decay, and 4He content increases with increasing residence time of gas in the crust (Giggenbach & Goguel, 1989). The 3He to 4He ratio, usually expressed

3He 3He as R/Ra, or 4He sample/ 4He atmosphere, varies by rock type and is ˜6.1 for subcontinental lithospheric mantle and 8 for mid ocean ridge basalts (MORB) (Gautheron & Moreira, 2002). Argon can come from magmatic

vapors or radioactive sources, but these contributions are very small relative to the total Ar content, and

the source of Ar is thus attributed to meteoric water (Giggenbach & Goguel, 1989). The source of N2 in magmatic systems is somewhat uncertain, but thought to be derived from decomposition of organic matter

63 Figure 5.12: N2-He-Ar ternary diagram for examining source of noble gases. “RM Spurr 2” plots near the composition of air but has a slightly elevated He content.

Figure 5.13: CO2-CH4-N2 ternary diagram showing Crater Peak fumarole compositions plot in the field of low CH4 magmatic or oxidized gases as opposed to hydrothermal gases.

64 that has contacted a magmatic intrusion, directly released from mantle, or may be transported by subducted

slab. Air-saturated groundwater has a N2/Ar ratio of 38, while the same ratio for air is 84 (Giggenbach & Goguel, 1989).

Data from the single fumarole sample that had measured concentrations of all three components is

shown in Figure 5.12. “RM Spurr 3” plots near air (N2/Ar=84), suggesting air contamination of the sample. However, a He content slightly elevated above air could be caused by crustal influence, which is supported by an R/Ra value of 6.6 (the typical value for subcontinental lithospheric mantle).

5.3.2 CO2-CH4-N2

Relative concentrations of carbon dioxide, methane, and nitrogen can also be used to examine the type of

discharge. Volcano-related magmatically influenced gases typically have low CH4 contents, while hydrother- mal gases typically have more CH4 (Giggenbach et al., 1990). The source of N2 is largely from meteoric

water, and CO2 could come from magmatic degassing or dissolution of carbonate minerals. The fumarole samples from Crater Peak have low relative methane concentrations and plot near the N2 corner, indicating magmatic or oxidized type discharges (Figure 5.13).

5.4 Geothermometry Calculations

The chemistry of surface water and gas discharges records the state of last equilibration depending on the kinetics of the particular species that is being exchanged. Empirical cation solute geothermometers are based on the concentrations of Na, K, Mg, Ca Li, and silica (Table 5.3). Likewise, gas geothermometers are

based on concentrations of Ar, H2,H2S, CO2, and CH4 (Figure 5.14). Geothermometry calculations can provide insight into the original, deep composition of a fluid if the surface discharge composition and likely subsurface processes are known.

5.4.1 Solute Geothermometer Calculations

Published solute geothermometers are summarized in Table 5.3. In the case of a rapidly ascending, minimally modified fluid, the concentrations of dissolved constituents from depth are preserved upon ascent to the surface and can be used to estimate the temperature of the deep hydrothermal system. The Na-K geothermometer equations are based on ion exchange between co-existing alkali feldspars according to Eq. 5.1. At temperatures <120°C, reactions with other minerals such as clays become impor- tant. The rate of Na-K re-equilibration is slower than that of K-Mg and silica dissolution/precipitation, so

the Na/K geothermometer can preserve equilibrium conditions from deeper depths at higher temperatures

(Nicholson, 1993). Since the geothermometer is based on a ratio, it is not affected by dilution and boiling,

unless Na or K is added with the diluting water (Nicholson, 1993). The silica geothermometers are based on

65 Table 5.3: Aqueous geothermometers. Concentrations of chemical species are in mg/kg unless otherwise specified.

Geothermometer Equation Temp (￿) Ref Quartz (no T oC = 1309 273 50-250 [1] 5.19 log(SiO2) − steam loss) − Quartz (max T oC = 1522 273 100-250 [1] 5.75 log(SiO2) − steam loss) − o 1032 Chalcedony T C = 4.69 log(SiO ) 273 50-250 [1] − 2 − Amorphous T oC = 731 273 50-250 [1] 4.52 log(SiO2) − Silica − T oC = 1674 273 Na-K-Ca Na 4 √Ca >180 [1] log( K )+ 3 log K +2.06 +2.47 − o 1217 T C = Na 273 >120 [2] 1￿.75+log￿ ( K￿ ) − ￿ o 856 T C = Na 273 >120 [3] 0.8573+log( K ) − o 1390 T C = Na 273 >120 [4] Na/K log( K )+1.750 − o 933 T C = Na 273 25-250 [5] log( K )+0.993 − o 1319 T C = Na 273 250-350 [5] log( K )+1.699 − o 883 T C = Na 273 >120 [6] log( K )+0.780 − o 1178 T C = Na 273 >120 [7] log( K )+1.470 − o 1000 T C = Li 273 Li/Na log( Na )+0.389 − [8] Li and Na concentrations in molal o 1590 Li/Na T C = Li 273 [9] log( Na )+0.779 − T oC = 2200 273 Li/Mg log Li +5.470 50-300 [10] Mg2 − o 4410 T C = 2 273 K/Mg 14￿ log￿ K 50-300 [4] − Mg − o 2.88 106 18 T C = ·18 273.15 1000+δ ￿ OSO￿ δ OH2O-SO4 1000 ln 4 +4.1 − [11] 1000+δ18O ￿ H2O References are: [1]Fournier (1981) [2] Fournier (1979)[3] Truesdell￿ (1976) [4] Giggenbach￿ (1988)[5] Arnorsson (1983) [6] Tonani (1980) [7] Nieva & Nieva (1987) [8] (Fouillac & Michard, 1981) [9] (Kharaka et al., 1982) [10] (Kharaka & Mariner, 1989) [11] Truesdell & Hulston (1980) [12] Nicholson (1993)

66 silica solubility over various temperature ranges. A Na-K-Ca geothermometer was derived to provide more

accurate temperature estimates for high Ca waters where Ca, Na and K compete in ion-exchange reactions

with silicate minerals (Nicholson, 1993). These calculations are affected by pCO2, precipitation of calcite with loss of CO2 upon boiling, mixing with CO2-rich groundwater, and ion-exchange reactions that involve Mg (Nicholson, 1993). There is also an optional Mg correction that may be applied to the Na-K-Ca geother- mometer. Low Na/Li values seem to correspond to the hottest parts of a geothermal field (Nicholson, 1993).

Li seems to be more temperature dependent than Na, and is likely derived from ion-exchange reactions at depth (Nicholson, 1993). Exchange reactions with Mg are rapid at low temperatures, and the Mg content of geothermal waters typically decreases as temperature increases (Giggenbach, 1988). The oxygen-18 isotope geothermometer assumes relatively long residence times of geothermal fluids in which isotopic equilibrium between sulfate and water is established (Nicholson, 1993). The water-sulfate oxygen isotopic signature can be affected by boiling, dilution/mixing, oxidation of H2S, and dissolution of sulfate-bearing evaporite minerals or sea water (Nicholson, 1993). Solute geothermometer calculations are represented graphically in Figure 5.14, and show little agreement. Na/K and Li/Mg geothermometers predict high equilibrium temperatures, while K/Mg and quartz predict low equilibrium temperatures. Na-K-Ca and Na/Li predictions tend to fall somewhere in between the extremes. Surface discharges have lower silica and Na, but higher Ca and Mg than would be expected from a deep fluid. Poor agreement between solute geothermometers (Figure 5.14) coupled with indications of shallow processes including mixing (Figure 5.1, Figure 5.5, Figure 5.6, Figure 5.7, Table 5.1) and interaction with rock at low temperatures (Figure 5.2, Figure 5.3) indicate that in general, Spurr surface discharges represent shallow fluid that is heavily diluted with meteoric water.

5.4.2 Isotope Gethermometer Calculations

The ￿18O geothermometer predicts temperatures 30-85°C for Mt. Spurr samples. These low tempera- tures suggest that the most important process affecting the calculations is likely mixing with dilute meteoric

water, which causes temperature estimates to be lower than if dilution was not happening. Low Cl/SO4

ratios in most Mt. Spurr waters also indicate that H2S oxidation is occuring, and sulfate from condensed ox- idized volcanic gases typically has unusually light (more negative) δ18O, which similarly causes temperature estimates to be lower than if oxidation was not occurring. (Nicholson, 1993).

5.4.3 Gas Geothermometer Calculations

Temperatures of gas equilibration can be visualized graphically through Y-T type plots (Figure 5.15),

which require the specification of steam fraction where the gas equilibrated, or through plots that use gas

67 Figure 5.14: Geothermometer calculations. The ranges represent variation in temperatures within the geothermometer, i.e. the “silica” range represents the results of amorphous silica, quartz steam loss, quartz no steam loss, chalcedony, and cristobalite equations. The temperatures estimated by the various geother- mometers don’t agree, and this is because Mt. Spurr waters have lower sodium and silica, but higher calcium and magnesium than would be expected from a deep fluid.

68 ratios (Powell & Cumming, 2010). Y-T type geothermometers can be affected by processes that modify

fumarole gas concentrations, including reservoir boiling and near surface condensation. Ratio plots fix

hydrogen fugacity by assuming a single redox state for the system and by using Ar as a proxy for pH2O, however Ar is sensitive to air contamination (Powell, 2000).

Concentration ratios for Crater Peak fumaroles yield a temperature near 250°C with a reservoir that does not show evidence of two-phase conditions in the subsurface (Figure 5.15(a)). However, Ar error could shift the fumarole samples into the two-phase region. The Y-T plots (Figure 5.15(b), Figure 5.15(c),

Figure 5.15(d)) yield equilibration temperatures from 200-250°C and equilibrated reservoir steam fractions of 0.1-1.0%. These temperatures are slightly higher than those calculated by solute geothermometers, and may reflect the difference in proximity to a heat source at Crater Peak to locations further afield.

5.5 Summary of Aqueous and Gas Geochemistry

Results and interpretations from the diagrams discussed in this section are summarized in Table 5.4. Anion, cation, stable isotope, silica, enthalpy and gas contents are reported in relative amounts (Figure 5.1- Figure 5.15 ), see Table 4.2 or Table 4.3 for absolute concentrations. Waters discharging from the south side of Mt. Spurr represent circulating waters of original meteoric origin that have variously interacted with rock, incorporated dissolved steam, in some cases mixed with andesitic water, and been heavily diluted by rain and snowmelt. They may be described by three general compositions based on aqueous geochemistry: 1.) Crater Lake waters, 2.) peripheral waters, and 3.) dilute/meteoric waters. Crater Lake waters seem to be the waters with the most direct haudraulic connection to magmatic heat and magmatically influenced chemical constituents (Figure 5.1, Figure 5.3, Figure 5.9, Figure 5.10). Discharges at Crater Canyon and Kid Canyon seem to represent peripheral waters that show components of magmatic influence (Figure 5.3, Figure 5.3, Figure 5.7, Figure 5.11(a)) and may be derived from the Crater Peak system, but have not communicated directly with a deep reservoir. In these locations, fluid flow seems to be controlled by faults that channel water to the surface, perhaps some distance away from the upflow zone. Samples from South Spurr locations seem to represent increasingly diluted waters (Figure 5.1, Figure 5.5, Figure 5.6) that may have have been channeled farther from the source over longer time periods. The spatial variability of South Spurr waters makes it hard to attribute a localized process to the formation of observed chemistry that applies to all South Spurr samples. Finally, dilute waters, like those at Chaka Ridge, have low concentrations of chemical constituents and no apparent direct hydraulic connection with magmatic processes. Structural channeling of deep magmatically influenced fluid to Chaka

Ridge could be possible, but this process does not seem to be well expressed at the surface. Chaka Ridge lies below the glacier that covers the summit of Mt. Spurr, and it is possible that the waters sampled there

69 (a) Ratio-type diagram for the estimation of equilibration temperature between CO2 (b) Y-T type diagram for the estimation of steam-reservoir equlibrium tem- peratures. The horizontal axis represents carbon dioxide-methane equlib- and H2 . Samples that plot between the “equilibrated liquid” and “equilibrated vapor” rium (CO2+4H2=CH4+2H2O Fischer-Tropsch reaction), while the vertical lines show evidence of two phase conditions in the subsurface (Powell & Cumming, 2010). axis represents H2-H2S equilibrium in the presence of magnetite and pyrite (3FeS2+2H2+4H2O=Fe3O4+6H2S and FeS2+H2=FeS+H2S) (Powell, 2000).

(c) Y-T type diagram for the estimation of steam-reservoir equlibrium tem- (d) Y-T type diagram for the estimation of steam-reservoir equlibrium tem- peratures. The horizontal axis represents carbon dioxide-methane equlib- peratures. The horizontal axis represents carbon dioxide-methane equlibrium rium (CO2+4H2=CH4+2H2O Fischer-Tropsch reaction), while the vertical axis (CO2+4H2=CH4+2H2O Fischer-Tropsch reaction), while the vertical axis repre- represents calcite-carbon dioxide equilibrium (CaCO3+K-mica=CaAl2-silicate+K- sents iron sulfide-hydrogen sulfide equilibrium (FeS2+FeO+2H2O=Fe2O3+2H2S). feldspar+CO2)(Powell,2000).

Figure 5.15: Gas geothermometers. The parameter “r” represents the gas/steam ratio of the indicated species. Figures produced using spreadsheets set up by Powell (2000).

70 Table 5.4: Summary of results and interpretations of aqueous and gas geochemistry.

Temp Anion Cation Stable Gas of Last Samples Con- Con- Silica Enthalpy Interpretation Isotopes Content Equil tent tent (C ) Close to MWL. Meteoric/dilute water 320, CC Head, 945, Variable δ13C High High influenced by interaction 945 10, 154, 435, 222, likely influenced Very Silica: SO4, Mg, Low with rock, possibly SPS-1, SPS-1 10 SPS-2, by organics or low 0-50 low B high Li steam heating. Meteoric APS2 10 Spring air origin. contamination. Peripheral to 06SPCN001A, 636, upflow/outflow, waters Crater Creek, SP001, Slightly δ18O influenced by interaction SP001 10, 532, CPHS, High High enriched. In Mid- Silica: with rock, diluted by UCC, LCC, 67-34, HCO3, Mg, low High range of range 50-150 metoeric water. Possibly KID1, SP013, SP014, low B Li magmatic δ13C. meteoric water mixing SP030, SP031, SP002a, with andesitic water or SP002b Crater Lake water. Close to MWL. Outflow/dilute waters High Variable δ13C influenced by interaction Canyon, Stream, Spurr Sulfate, Mg, likely influenced Mid- Silica: Low with rock, dilution by Knob low B variable by organics or range 25-125 meteoric water. Li air Meteoric origin. contamination. Upflow, acidic water that has incorporated N /Ar 2 magmatic volatiles. similar to Silica: 70AR-201, 70AR-202, Magmatic origin. High High air, but 50-200, CP9201, CP92-2, RM Enriched in Very Possible air Cl, high Mg, low High R/Ra close Fu- Spurr 1, RM Spurr 2, δ18O and δD high contamination or crustal B Li to crust. maroles: RM Spurr 3 influence to water Low 200-250 samples. One phase methane. reservoir in subsurface, low steam fractions.

71 predominately represent shallowly circulating glacial runoff and rain (Figure 5.7) that has interacted with the volcanic units and possibly absorbed some steam from a degassing magma chamber (Figure 5.1).

While it is desirable to deduce a reservoir temperature from the aqueous and gas geochemistry, it is important to keep in mind that the surface discharges in the field area are not representative of a deep reservoir fluid, and that there are many processes that complicate temperature estimation. The most useful information that can be taken away from the geothermometer calculations is the interpretation of shallow processes that are affecting the concentrations of chemical constituents. Working out an original deep chemistry form surface discharges at Mt. Spurr is problematic. However, the silica estimations are probably the most reliable, and represent a minimum estimate, since dilution is certainly happening.

72 CHAPTER 6

INTERPRETATIONS: MINERALS AND MINERAL RELATIONSHIPS IN DRILLED CORE

This chapter focuses on understanding the lithology and post-depositional changes of drilled core from core hole 26-11, focusing on indications of hydrothermal alteration that may be related to a potential deep geothermal reservoir south of Mt. Spurr. The conglomerate and sandstone unit encountered in 26-11 is interpreted to represent the West Foreland Formation. Evidence for the weathering of volcanic glass, compaction, and faulting is examined, and processes for the ubiquitous presence of calcite are suggested.

6.1 Unit Identification

The overall appearances of samples used in this study are consistent with previous general lithological descriptions of one of the Tertiary sedimentary units that outcrops just to the east of Mt. Spurr and also occurs in the Cook Inlet: the West Foreland Formation (Calderwood & Frackler, 1972; Finzel et al., 2009; McLean, 1986; Ormat Nevada, Inc., 2012a). The type section for the West Foreland Formation occurs on the west side of the Cook Inlet about 60 km southeast of Mt. Spurr between about 3.2km-3.5km depth (Calderwood & Frackler, 1972). The lithology described includes gray-green micaceous partly carbonaceous claystone, interbedded coal seams, dark-gray micaceous shale, beds of “green, waxy claystone,” gray-white sandstone, tuffaceous ash layers, and conglomerate with green pebbles of “probable volcanic origin in finer sand matrix” (Calderwood & Frackler, 1972). Pebbles sometimes occurring in the sandstone are “composed of volcanic material”, and pore space has been filled with “clay and volcanic ash” (Calderwood & Frackler, 1972). The unit is about 50% lithic sandstone, and sandstone beds ranging in thickness from 3-10m (Kremer & Stadnicky, 1985). Sandstone grains are fine-medium sized, angular-subangular shaped, poorly sorted, and are surrounded in a clay matrix or calcite cement (McLean, 1986). The average composition of the sandstone is Q23F30L47, and commonly observed minerals include hornblende, biotite, epidote, opaque minerals, and volcanic rock fragments (McLean, 1986). West Foreland Formation clast studies by the Alaska Division of Geologic and Geophysical Survey (DGGS) on outcrop just south of the Capps Glacier indicate a bulk clast composition of 79-87% extrusive volcanic clasts and 12-15% plutonic clasts. Minor greenstone (<6%) and volcaniclastic (<1%) clasts were also present (Finzel et al., 2009). The predominately volcanic composition of the clasts indicates a source to the north and west in the Alaska Range, which is composed the Alaska

Batholith, as opposed to a source in the south and east in the Chugach Mountains, which are made of metamorphic rocks (Hartman et al., 1972). This unit may be more conglomerate-rich (or sandstone-poor) near Mt. Spurr than in the Cook Inlet basin because it would have been closer to the source.

73 6.2 Diagenetic Changes

Post-depositional changes seen in the West Foreland Formation core from 26-11 appear to be related to diagenetic processes (as opposed to hydrothermal alteration), and match published descriptions of Cook

Inlet sediments that have undergone diagenesis (Bolm & McCulloh, 1986; Bruhn et al., 2000; McLean, 1986).

Post-depositional conversion of volcanic glass to clay minerals, compaction, faulting, dissolution of biotite, and infiltration of calcite have been documented in Kenai Group sediments in the Cook Inlet basin (Hayes et al., 1976). Diagenetic changes in sandstones encountered in the COST No. 1 well, a deep (3776m) well drilled in the Cook Inlet that penetrates the West Foreland through Naknek formations, mostly affect the volcanic rock fragments and plagioclase feldspars (McLean, 1986), and are attributed to high temperatures from deep burial or high geothermal gradients (Bolm & McCulloh, 1986). Lower Cretaceous and younger sedimentary rocks all contain authigenic clay minerals and calcite cement, and lower Tertiary samples contain extensive amounts of clay matrix and/or calcite cement (McLean, 1986).

6.2.1 Ductile Deformation

Volcanic fragments in COST no. 1 sandstones have commonly undergone ductile formation from com- paction (Bolm & McCulloh, 1986). Detrital grains in sandstone in core hole 26-11 have been deformed plastically around each other so that there is very little pore space remaining (Figure 4.4, Figure 4.7), suggesting compaction during deep burial. In some cases, quartz grains are deformed in a brittle manner (Figure 4.7), and this fracturing could indicate mechanical crushing due to fault movement. Fractured quartz grains from sample “3434” (Figure 4.7) were observed at a depth associated with slickenlines on the core.

6.2.2 Authigenic Clay

Authigenic clay minerals, including montorillonite and chlorite as discreet phases or mixed layers, increase with depth in COST no. 1. The abundance of volcanic rock fragments is positively correlated with clay mineral occurrence, and the clay minerals are thought to be formed by chemical breakdown of rock fragments (Bolm & McCulloh, 1986). The montmorillonite encountered in core hole 26-11 appears to have formed under relatively low temperature conditions. Only limited conclusions can be reached from the results of this research, and future clay-mineral sampling campaigns from deeper drilling would aid in the interpretation of clay mineral evolution with depth.

6.2.2.1 Montmorillonite

Montmorillonite-rich zones coincide with fault zones in core hole 26-11, but also occur in stratigraphic layers. The montmorillonite in the West Foreland Formation is likely derived from alteration of volcanic glass.

74 Alteration of volcanic glass to form clay minerals could be driven by decomposition over time and circulation

of connate/ground waters, or could be driven by injections of hydrothermal fluid of unknown composition.

Without additional corroborating evidence, it is not possible to distinguish which process controlls alteration

at Mt. Spurr based on the data collected in this research. However, dioctahedral montmorillonite is not

stable above 150°C (Monecke et al., 2007), so the clay minerals observed must have been formed under relatively low temperature conditions.

6.2.2.2 Chlorite

Chlorite in sample “3434” occurs along fracture surfaces, and could indicate the presence of high-

temperature fluid movement through fractures.

Procedures outlined in Moore & Reynolds (1997) were used to determine the stoichiometric formula for

the chlorite in “3434” based on the XRD reflections. There are slighly more Fe atoms in the octahedral sheet, and the total number of Fe atoms in 6 octahedral sites is calculated to be 1.72, which yields the

formula (Mg,Al)1.72-6Fe4(Si,Al)4O10(OH)8. Given a Fe value of 1.72 atoms per formula unit, calculations from Battaglia (1999) were used to estimate the number of Mg and Al atoms per formula unit, yielding values of Mg=0.71 and Al=0.32. Background is indistiguishable from possible hkl reflections in the chlorite, so the determination of chlorite polytype was not possible.

The chlorite formula can be used as a geothermometer, if the amount of AlIV is known. Two geother- mometers were used in this research (Eq. 6.1 from Cathelineau (1988) and 6.2 from Kranidoitis & MacLean (1987)) to obtain estimates of formation temperatures.

T (oC)= 61.92 + 321.98(Al ) (6.1) − IV

o T ( C) = 106(AlIV ) + 18 (6.2)

Assuming that AlIV=0.32, as previously calculated, application of a chlorite geothermometer from Kranidoitis and MacLean (1987) yields a temperature of formation of 104￿, while the chlorite geothermometer of Cathineau (1988) yields a temperature of 41￿. Both geothermometers are empirical, but the equation of

Kranidoitis & MacLean (1987) attempts to incorporate an Fe correction factor (AlIV = AlIV observed + − Fe 0.7 Fe+Mg). These temperature estimations could also be so different because of error propagation in the chlorite composition, since the composition was determined graphically based on the XRD reflection patterns.

6.2.2.3 Laumontite

Laumontite is a hydrated calcium-aluminium silicate mineral of the zeolite group that can be formed by a hydrolysis reaction that involves air or magmatically-derived CO2 and feldspar, as shown in Eq. 6.3, from

75 Giggenbach (1984). This reaction occurs at temperatures <200￿.

laumontite + K feldspar + CO = calcite + K mica +4SiO +3H O (6.3) − 2 − 2 2 Laumontite was not encountered in core hole 26-11 but has been documented in COST no. 1 at depths

>2134m. In COST no. 1, laumontite post-dates authigenic clay minerals and is thought to form from the release of Ca and Al from albitization of plagioclase and breakdown of volcanic rock fragments (Bolm &

McCulloh, 1986). Bruhn et al. (2000) attribute laumontite and zeolite mineral formation to the interaction of connate Na-Ca-Cl brine in Mesozoic and older sediments with the surrounding sedimentary rocks. The formation of laumontite is thought to be related to an episode of thermal alteration that culminated in late Mesozoic-early Cenozoic time and left a regional imprint throughout the Cook Inlet basin (Bolm &

McCulloh, 1986).

6.2.3 Calcite

Calcite in 26-11 could have been formed by a hydrothermal fluid that has experienced loss of CO2 upon boiling or a pressure drop, or by hydrolysis of calcium-aluminium-silicate minerals. Hydrolysis is favored in rock dominated environments where slow moving fluid reacts with rock in sub-boiling conditions (Simmons & Christenson, 1994), which is a likely scenario for the West Foreland Formation studied in this research. At higher temperatures (>200￿), a number of reactions can contribute to the formation of calcite, including Eqs. 6.4, 6.5, and 6.6, as well as the non-stoichiometric process of volcanic glass conversion to calcite (Giggenbach, 1984). In a fluid-dominated environment, platy calcite is favored in near-vertical channels where rising fluids boil (Simmons & Christenson, 1994).

anorthite + K feldspar + CO + H O = calcite + K mica +2SiO (6.4) − 2 2 − 2

0.5clinozoisite +0.75K feldspar + CO +0.5H O = calcite +0.75K mica +1.5SiO (6.5) − 2 2 − 2

wairakite + K feldspar + CO = calcite + K mica +4SiO + H O (6.6) − 2 − 2 2 At Broadlands-Ohaaki, New Zealand, calcite replacement is confined to pumice clasts and glass shards, and is associated with smectite, mixed layer illite-smectite, kaolinite, mordenite, cristobalite, and quartz at temperatures below 170￿ (Simmons & Christenson, 1994). Above 170￿, calcite replaces primary plagioclase and volcanic glass, and is associated with quartz, illite, adularia, chlorite, and pyrite (Simmons & Christenson, 1994). Calcite replacement of pumice and volcanic glass, as well as the presence of smectite is observed in the West Foreland Formation.

76 Both sparry and microsparry carbonate cement fills intergranular spaces in COST no. 1, and calcite

precipitation in the core hole is attributed to post-depositional changes in temperature, pressure, or pore-

fluid composition (Bolm & McCulloh, 1986). Bruhn et al. (2000) attribute calcite formation in the Cook

Inlet basin sediments to the reaction of Na-HCO3 connate fluids in Tertiary sediments with surrounding

rock, to the mixing of upward migrating Na-Ca-Cl brine with the Na-HCO3 fluids, and to fluctuations in

pCO2 during faulting. The calcite isotope results of this research are plotted with published data from Bruhn et al. (2000) in

Figure 6.1. The calcite samples from the West Foreland Formation encountered in core hole 62-11 have an isotopic composition that overlaps that of the Cook Inlet calcites in δ13C, and overlaps or is slightly enriched in δ18O. The Mt. Spurr water and West Foreland Formation calcite δ13C values are within similar ranges of ˜4-12‰. Carbon-13 values that are less than -15‰ can be considered outliers, as these samples are likely recieving carbon contributions from an organic source such as methane. If all the calcite examined in this research formed from a fluid of the same isotopic composition, and carbon is assumed to fractionate very little with temperature, a 15‰ δ18O fractionation could have occurred over a temperature difference of about 30￿. Since the δ18O and δ13C values for the Mt. Spurr waters are known, Eq. 6.7 from Faure (1986), can be used to calculate the isotopic composition of calcite expected to be in equilibrium with Mt. Spurr springs and seeps.

6 18 18 3 2.78 10 δ Ocalcite δ Owater 10 ln αcalcite water = • (6.7) − • − T 2 This calculation yields a calcite isotopic composition that overlaps with the calcites from the West Foreland Formation and the Cook Inlet (Figure 6.1). This suggests that the calcite in the West Foreland Formation could have formed from water of meteoric-dominated origin at relatively cold temperatures under diagenetic conditions. Bruhn et al. (2000) speculate that the calcites of the Cook Inlet reflect equilibrium

with waters at 80°C. In summary, the temperature of which both the Cook Inlet and West Foreland Formation calcites reflects

is interpreted to be less than 100°C, which is lower than one would expect if a hydrothermal system is influencing these calcites.

6.2.4 Porosity and Permeability

Sandstones of the West Foreland Formation have documented porosities ranging from 10-20% (Kremer &

Stadnicky, 1985; Magoon et al., 1979; McLean, 1986). The samples of the West Foreland Formation studied

in this research have very little porosity, either because calcite or calcite and clay minerals fill all void spaces

77 Figure 6.1: Stable oxygen and carbon isotopes from veins and cements of the West Foreland Formation encountered in core hole 62-11 (this research) are plotted with calcite veins and cements sampled by Bruhn et al. (2000) in the Cook Inlet. Water samples and a hypothetical calcite composition in equilibrium with water samples are also plotted.

78 in conglomerate sections, or because the grains are plastically and mechanically deformed so that all pore space is filled in sandstone sections. Sandstones in COST no. 1 at <2088m depth commonly have visible intergranular porosity where the pore spaces are lined with authigenic clay, whereas deeper samples have only fracture porosity. Porosity decreases with depth due to more abundant authigenic clay and compacted grains (Bolm & McCulloh, 1986).

Bruhn et al. (2000) attribute fluid migration from depth to the surface in Cook Inlet sediments to

1.) high fluid pressure caused by compaction, volumetric strain related to deformation, dynamo-thermal metamorphism, alteration of organic-rich rock and hydrocarbons, or glacial loading, and 2.) faulting, which allows connate waters trapped in Mesozoic sediments to migrate upward into Tertiary sediments. High pressure fluid encountered while drilling in Cook Inlet anticline structures in lower Teritary and Mesozoic sediments occurs in permeable zones that are sealed on either side by zeolite or carbonate minerals, which channel fluid flow parallel to bedding in anticlines (Bruhn et al., 2000). Flowing water in core hole 26-11 was only encountered in fault zones, and the flow was generally short- lived (Ormat Nevada, Inc., 2012a). Calcite and clay minerals seem to play an important role in reducing pore space in core hole 26-11, although there does not appear to be much change in the character of either mineral with depth, at least in the 1215.5m that were sampled. It seems that water flows through the West Foreland Formation only along faults, and that there is very little formation porosity and permeability in this unit in general due to the infilling of pore spaces by diagenetic clay minerals and calcite. However, not all mapped faults show evidence of transmitting deep fluid.

79 CHAPTER 7

DISCUSSION

This chapter discusses the findings of this research in an attempt to relate details to a conceputal model of what is occuring in the subsurface. Stable hydrogen, oxygen, and carbon isotopes suggest that contributions to chemistry must come at least in part from a magmatic system, but a hydrothermal system is not well expressed at the surface.

7.1 Water-Rock Interaction

The water and core data sets examined in this research do not appear to be related. Both datasets reflect low temperature conditions, but there is not enough data to tell if the springs and seeps have interacted directly with the West Foreland Formation sediments. The possibilities of sampled waters interacting with SVC rocks and resulting chemistry is explored by geochemical modeling using Geochemist’s Workbench software (GWB).

7.1.1 Models Involving Crater Lake Water, Glacial Melt, and Andesite

Three models examine possible water-rock interaction and mixing scenarios. The parameters used are defined in Table 7.1 and Table 7.2. The water-rock interaction scenarios involve 1) the mixing of meteoric water with SVC andesites, 2) the mixing of Crater Lake water with SVC andesites, and 3) the mixing of the two resulting waters. Since water will equilibrate with the alteration products of the andesite rather than the original mafic minerals, the alteration products of SVC andesites are included in the models. Representative water samples were selected from the dataset (“945” to represent meteoric water and “70AR-202” for Crater

Lake water), speciated at the sampling temperature with the Spec8 program (balance on Cl-), and reacted once at the sampling temperature with SVC alteration products using the React program. The results of those models are shown in Figure 7.1-Figure 7.3.

Table 7.1: Alteration minerals and amounts are based on likely SVC alteration products using SVC petrology by Nye & Turner (1990).

SVC Alteration Mineral Amount (g) Amrphˆsilica 8.75 Calcite 27.62 Beidellit-Ca 33.63 Chrysotile 8.79 Chamosite-7A 21.21

80 Table 7.2: “Meteoric” and “Crater Lake” waters that have reacted with SVC alteration minerals are derived from samples “945” and “70AR-202”, respectively. Basis species concentrations are reported in moles. Both samples also include a basis species of 1 free kg H2O.

Basis Species Meteoric Water Crater Lake Water Temp (￿) 0 49 H+ 1.45E-05 6.78E-03 Ca++ 1.21E-05 4.56E-03 Na+ 2.11E-05 3.28E-03 K+ 1.24E-05 1.88E-04 Cl- 4.01E-05 1.86E-02 HCO3- 1.59E-05 1.65E-05 SO4– 1.62E-05 3.30E-03 Al+++ 7.29E-07 1.68E-06 Mn++ 2.12E-07 9.15E-08 Fe++ 3.13E-07 1.80E-07 Mg++ 2.00E-06 2.89E-03 F- 5.11E-06 5.29E-06 Li+ 1.40E-05 1.45E-05 B(OH)3 3.90E-04 SiO2(aq) 1.02E-05 3.51E-06

Figure 7.1: Piper diagram showing surface discharge compositions (same symbols as previously described) as well as reaction paths (black lines) representing water-rock interaction between meteoric water (“945”) and andesite alteration products at a 1:25 ratio at a constant temperature of 0￿.

81 Figure 7.2: Piper diagram showing surface discharge compositions (symbols same as previously described) as well as reaction paths (black lines) representing water-rock interaction between Crater Lake water and andesite alteration products at a 1:1 ratio along a reaction path that starts at 49￿ and ends at 0￿.

82 Figure 7.3: Piper diagram showing the reaction path as the results of Figure 7.1 and Figure 7.2 are mixed with a ratio of 1:1 at a constant temperature of 0￿. The anion ternary reaction path is represented by a point at the location of 70AR-202.

83 The interaction of meteoric water with andesite could have produced a trend seen in the anion ternary

that relates “945” to Crater Canyon samples “CPHS”, “626”, “626R”, and “Crater Creek,” as well as a

trend in the cation ternary that relates “945” to Chaka Ridge samples “Spring”, “SPS-1”, “435”, and Crater

Canyon sample “CC Head” (Figure 7.1). However, the combined cation and anion chemistry of surface

discharges at Mt. Spurr, shown in the quadrilateral diagram of Figure 7.1, is not well described by the

reaction path. Chaka Ridge surface discharges and cold waters in Crater Canyon are interpreted to be

the best example of meteoric water-rock interaction, based on the observation that these springs and seeps

emerge from pumice and talus slopes (Figure 3.3), appear to represent shallowly circulating meteoric water

(Chapter 5), and the reaction paths in this first geochemical model suggest that the cation content of these

samples is controlled by interaction between cold water and andesitic rock.

The interaction of Crater Lake water with andesite could have produced a trend seen in the anion ternary (Figure 7.2) that lies just below Crater Canyon samples “SP001”, “532”, “SP031” and “SP030”, but does predict the cation composition of any Mt. Spurr surface discharges very well. Additionally, the reaction path on the quadrilateral diagram of Figure 7.2 does not intersect the majority of the samples. In summary, Crater Lake-andesite interaction does not describe the observed chemistry in Crater Canyon or other field areas very well, but it’s possible that some of the anions in samples in Crater Canyon could be influenced by the products of Crater Lake-andesite interaction. The waters developed in the previous water-rock interaction models were used in exploring the possibil- ity of a mixing scenario where Crater Lake waters percolate through andesitic lava flows and are diluted by meteoric water that has also interacted with andesite, as proposed by Motyka & Nye (1993) (Figure 5.8). The calculated reaction paths, shown by black lines in Figure 7.3, could account for the cation composi- tions of “222” and “320”, both cold dilute waers, but don’t yield a composition that matches other Crater Canyon samples. The anion composition does not change. The reaction path in the quadrilateral diagram of Figure 7.3 follows a trend between “70AR-202” and “320”, which is a cold sample from Crater Canyon. In general, these water-rock interaction and water mixing models poorly predict the observed surface water chemistry at Mt. Spurr, suggesting that other processes are influencing the springs and seeps.

7.1.2 Models Involving Na-K-Ca-Mg Aluminosilicates and Water at Various Temperatures

No alteration minerals were observed at the surface near springs and seeps at the time of sampling. Additional geochemical modeling of water samples is utilized to predict the alteration minerals that might

+ + exist in the shallow subsurface along the reaction paths of flowing water. The activities of SiO2, Na ,K , Ca2+,Mg2+ are used to examine the state of equilibrium between geothermal waters and expected alteration minerals. This is analagous to using the aforementioned cation geothermometers, but the state of equilibrium

84 or non-equilibrium can be displayed graphically. Some error is inherent in the pH calculations of fluids at

high temperatures since the dissolved gas composition of deep fluids is unknown, thus temperatures >100￿ were not considered. The default thermodynamic dataset in GWB was implemented for activity calculations

(LLNL Thermodynamic Database (1986) based on the dataset of T. Wolery et al.: data file data0.3245r46,

reformatted to Geochemist’s Workbench (1994) as the default thermo.dat database) using the Debye-H¨uckel

activity model. The calculations of GWB have to be viewed considering that thermodynamic data for many

alteration minerals, including clay minerals, is sparse and poorly understood. Montmorillonite is not included

in the GWB dataset, so beidellite is considered the equivalent for the purposes of this modelling.

Water samples were speciated using the Spec8 program and results were compiled into a GSS program

+ + 2+ 2+ + spreadsheet. Activities for SiO2, Na ,K ,Ca ,Mg and H concentrations of samples were calcu- lated using the GSS program. Theoretical activity-activity diagrams that depict mineral stability fields in chemically restricted systems were constructed using the Act2 program for Na2O-K2O-Al2O3-H2O, Ca2O-

3+ K2O-Al2O3-H2O, and MgO-Al2O3-H2O systems using kaolinite (swapped with Al ) as the diagram species,

in the presence of water and quartz (swapped with SiO2(aq)), at temperatures from 0-100￿ (Figure 7.4, Figure 7.5, and Figure 7.6). The axes in these diagrams represent the activity ratios of K+/H+, Na+/K+,

2+ +2 2+ +2 Ca /H ,Mg /H , and SiO2, respectively. Water samples that plot within the stability fields of given minerals signify equilibrium with the mineral at specified temperature and pressure conditions. Activity scatter data was imported from the GSS spreadsheet and overlayed on the activity-activity diagrams using the Act2 program. Samples “67-34”, “KID1” and “626R” cluster together in all the diagrams and have the most positive (largest number) cation activity, while the remaining samples have lower cantion activities. Samples from Crater Canyon, with the exception of “320” have similar silica activities between -2.5 and -3, South Spurr samples have silica activities between -3 and -3.5, and Chaka Ridge samples have silica activities between -3.5 and -4. Samples plot in the kaolinite field in all three cation scenarios considered, therefore kaolinite is interpreted to be a likely component of subsurface alteration. A pink-orange clay encountered in altered andesites of exploration core hole 62-2 and volcaniclastic flows of exploration core hole 67-34 is speculated to be kaolinite (Ormat Nevada, Inc., 2012b). Thus it seems the relatively low-temperature fluids sampled at Mt. Spurr could have caused the kaolinitic alteration observed in volcanic rocks encountered in exploration core holes. The spread of samples in between stability fields for gibbsite, kaolinnite, and pyrophyllite in Figure 7.6 depends on silica activity, which is influenced by pH. The cluster of “67-34”, “KID1” and “626R” lies in the muscovite or saponite fields, indicating a distinct chemistry that could have interacted with different

(possibly higher temperature) alteration products than the majority of samples.

85 Figure 7.4: Diagrams for the Na2O-K2O-Al2O3-H2O system in the presence of silica from 0-100￿. Axes represent activity ratios.

86 Figure 7.5: Diagrams for the Ca2O-K2O-Al2O3-H2O system in the presence of silica from 0-100˚C. Axes represent activity ratios.

87 Figure 7.6: Diagrams for the MgO-Al2O3-H2O system in the presence of silica from 0-100˚C. Axes represent activity ratios.

88 Overall, the activity-activity diagrams examine the state of equilibrium between water samples and

alteration minerals, and indicate that Mt. Spurr springs and seeps are not in chemical equilibrium with

minerals used as the basis for geothermometer calculations. Compared to similar activity-activity diagrams

by Fournier (1981) of the Broadlands geothermal system at 260°C, it is clear that the surface discharges from M.t Spurr are not similar to a geothermal reservoir fluid.

7.2 Conceptual Cross-Sections

Building on the work of Turner & Wescott (1986), Motyka & Nye (1993) (Figure 5.8), and Martini

et al. (2011), three conceptual cross-sections of a possible hydrothermal system at Mt. Spurr are proposed

(Figure 7.7-Figure 7.10). Cross-section A-A’ ( Figure 7.8) runs roughly north-south from Mt. Spurr summit

to the Chakachatna river, B-B’ (Figure 7.9) runs southeast from Mt. Spurr summit along Chaka Ridge

through the exploration core holes, and C-C’ (Figure 7.10) runs roughly east-west through the South Spurr area. Simplified surface geology by Nye et al. (2005), known and inferred faults by Martini et al. (2011), core hole stratigraphy (Ormat Nevada, Inc., 2012a,b), areas of low resistivity (Figure 2.6) from geophysical surveys (Martini et al., 2011; Turner & Wescott, 1986), and spatial relationships and processes interpreted from water chemistry have been included. The thicknesses of lithological units, placement of a magma body, and orientations of inferred structures are not well constrained. The West Foreland Formation thickness is drawn to scale according to Cook Inlet records, but is unknown in the study area since exploration drilling has not penetrated the uint. Arrows represent fluid pathways and a dot pattern represents boiling hydrothermal fluids. For the most part, the waters observed at the surface represent shallowly circulating meteoric water

that has interacted with SVC andesites and incorporated some volcanic steam (CO2,H2S). I agree with the interpretation of Motyka & Nye (1993) that some Crater Lake water could percolate downslope, be diluted by meteoric water, and re-emerge as springs and seeps in Crater Canyon. But I also think Crater Canyon surface discharges have a chemistry that is distinct from that of Crater Lake, and that deeply penetrating faults, such as the Capps Glacier or Crater Canyon faults, could be channeling some deep fluid or magmatic components to the surface. I have interpreted the montmorillonite-rich West Foreland Formation as a seal that prevents surface waters from circulating deep enough to be influenced by magmatic processes, and also prevents deep hydrothermal fluid from upwelling to the surface, except locally along faults. If granite does exist below the West Foreland Formation, and if it is fractured or otherwise capable of storing hot fluid, this fractured granite could be a viable geothermal reservoir. The conceptual cross-sections of Figure 7.8-

Figure 7.10 aim to show a hydrothermal system that exists in the field area below the sediments of the

West Foreland Formation. This hydrothermal system is heated by a magma body, recharged by meteoric

89 Figure 7.7: Location of conceptual cross-section lines (purple), overlayed on elevation contours and the geologic map by Nye et al. (2005). Bold black lines are faults (Martini et al., 2011), and water sampling locations are shown with the same symbols.

90 Figure 7.8: Conceptual N-S cross-section A-A’ of field area from Mt. Spurr summit to the Chakachatna River. Seeps in Crater Canyon are fault-controlled and represent shallowly circulating meteoric water that seems to be mixing with Crater Lake or andesitic water.

Figure 7.9: Conceptual NW-SE cross-section B-B’ of field area from Mt. Spurr summit through Chaka Ridge. Seeps on Chaka Ridge are topographically controlled and represent shallow circulation of meteoric water through porous volcanics and minor incorporation of steam.

91 Figure 7.10: Conceptual E-W cross-section C-C’ of field area from west of Crater Glacier to east of Chaka Ridge in the South Spurr area. Upwelling related to discharges at Crater and Kid Canyons could extend further south along faults. This figure represents a perspective where hydrothermal outflow is coming toward the viewer. water, and has fluid pathways along deeply penetrating faults that cut through volcanic units and the West Foreland, extending to the depth of the reservoir.

7.2.1 Temperature Gradient

While the temperatures encountered in core holes at Mt. Spurr are not as high as subsurface waters in most developed geothermal systems, they are slightly higher than background in the Cook Inlet (Figure 7.11). Similar situations of nearly isothermal temperature profiles were encountered with geothermal exploration drilling at Newberry and Makushin, which are both volcanoes that formed along convergent plate bound- aries with extensive snow and ice cover. With increasing depth, tepmperature gradients at Newberry and Makushin jump suddenly to higher temperatures. The isothermal trend is likely caused by mixing with cold meteoric water, and the sudden jumps could occur when thermal water influence overtakes cold influence, possibly due to structures or increasing proximity to the heat source. At Mt. Spurr, extensive dilution by near-freezing temperature water likely overwhelms any possible warm upwelling signature at shallow depths, and a sudden spike in temperature could occur once the clay cap of the West Foreland Formation is penetrated.

7.3 Comparison to Producing Geothermal Systems in Similar Settings

Exploited geothermal systems that are related to recent andesitic volcanism and occur on convergent plate boundaries are compared to the potential geothermal system at Mt. Spurr. These producing fields could help predict what might be expected at Mt. Spurr in possible future exploration and production scenarios.

92 Figure 7.11: Temperature gradients of Mt. Spurr, other volcanoes, and Cook Inlet sediments. Drilling at Mt. Spurr indicates a slightly higher temperature gradient than the Cook Inlet, but is lower than most developed geothermal systems. However, exploration drilling at Mt. Spurr has only gone to ˜1km depth and the constant influx of near-freezing surface waters likely overwhelms any thermal signature at shallow depths.

7.3.1 Southeast Asia

Geothermal systems in the Philippines, including Palinpinon, Tongonan, Bacon-Manito, Mak-Ban, and Tiwi, are adjacent to calc-alkaline andesitic stratovolcanoes that have been active in the late Pliocene to Recent time (Reyes, 1990). The stratigraphy of many of these systems consists of andesite, andesitic hyalo- clastites and tuff breccias, shallow marine-bathyal carbonaceous fossiliferous calcisiltite intercalated with volcanics, and quartz monzodiorite-diorite intrusives, and cross-cutting dikes (Reyes, 1990). Neutral pH al- teration is divided into 4 zones based on key clay minerals and temperature: smectite, transition, illite, and biotite. Epidote and amphibole are also identified as subzones (Reyes, 1990). The smectite zone is <180￿, and commonly associated minerals include opal, cristobalite, tridymite, laumontite, calcite, dolomite, and pyrite. The appearance of mixed layer illite-smectite, albite, adularia, hyalophane, sphene, anhydrite, and ru-

93 tile marks the beginning of the transition zone (180-230￿) (Reyes, 1990). The illite zone (230-320￿) begins with the appearance of illite, wairakite, and chalcopyrite. The biotite (>270￿), epidote (>250￿), and am- phibole (>280￿) zones coincide with biotite, prehnite, amphibole, sphalerite, and galena alteration(Reyes, 1990).

Indonesian geothermal fields, including Awibengkok and Wayang Windu, are associated with recent

(<1Ma) andesitic volcanism related to the Sunda Arc and commonly underlie sector collapses (Bogie et al.,

2008). Basement consists of shallow marine carbonates and epiclastic sediments, that are overlain by an- desitic piles and interbedded ash deposits (Stimac et al., 2008). Wayang Windu has a hot neutral pH liquid reservoir that is overlain by a perched vapor-dominated two-phase reservoir. Ash beds at Wayang

Windu originally consisted of very fine-grained glass shards and titanomagnetite that have altered to cal- cium smectite and interlayered illite-smectite, and are usually found at temperatures <200￿ (Bogie et al., 2008). Hydrothermal alteration at Wayang Windu is strongly developed in pyroclastic deposits and can be separated into four spatial zones: above the conductor, in the conductor, in vapor-dominated reservoirs, and in deep liquid reservoirs (Bogie et al., 2008). Opal, cristobalite, kaolinite, alunite, and rare sulfur alteration at shallow depths is associated with perched steam-heated groundwater aquifers. The conductor is com- posed of smectite, quartz, chlorite, calcite, pyrite, and zeolites, and has been overprinted in some parts by kaolinite, calcite, anhydrite, and quartz (Bogie et al., 2008). With increasing depth, the smectite transitions to mixed layer illite-smectite and then illite. In the deep liquid reservoir, wairakite, prehnite, epidote, and less commonly adularia are common as alteration and vein minerals (Bogie et al., 2008). At Awibengkok, the geothermal system is capped by a zone of smectite alteration with pyrite, hematite, calcite, anhydrite, and zeolites that formed at temperatures <180￿ that transitions to mixed-layer smectite-illite, chlorite, calcite, pyrite, titanite, and quartz (Stimac et al., 2008). At deeper depths, epidote, illite, quartz, chlorite and minor albite, adularia, calcite, wairakite, pyrite, anhydrite and titanite alteration define an assemblage that corresponds to 240-270￿ (Stimac et al., 2008).

7.3.2 Central America

The Miravalles geothermal field in Costa Rica is associated with Quarternary-Pleistocene andesitic-dacitic volcanic features of the Guanacaste Cordillera, which was formed by the subduction of the Cocos Plate beneath the Caribbean Plate (Gherardi et al., 2002). The area contains four main fault systems. Fluid from producing wells has a neutral pH (6.5-8.3), high chloride (2688-3493 mg/kg) content, relatively low and similar sulfate and bicarbonate contents (21-63 and 5-80 mg/kg, respectively), high Na (1571-2075 mg/kg), moderate K (193-251 mg/kg), low Ca (39-91 mg/kg), and trace Mg (average 0.08 mg/kg) (Gherardi et al., 2002). However, most surface discharges at Miravalles are little affected by the direct contribution of

94 deep reservoir waters. At shallow levels, many small cold-thermal aquifers receive only steam, gases, and some heat from depth (Gherardi et al., 2002). Hydrothermal alteration in the Miravalles reservoir includes chlorite, sphene, pyrite, albite, k-spar, quartz, sericite, laumontite, wairakite, calcite, anhydrite, and epidote replacement of primary igneous minerals (Rochelle et al., 1989).

7.3.3 Summary

Several similarities between geothermal systems in the Aleutian Arc, Southeast Asia, and Central Amer- ica emerge: shallow alteration produced by neutral pH, relatively low temperature waters that includes smectite, carbonates and zeolites. Based on observations of the producing geothermal fields in similar set- tings, deeper drilling at Mt. Spurr could encounter an illite-smectite transition and zeolites (laumontite has been documented in West Foreland Formation sediments of the Cook Inlet), as well as hydrothermal epidote, quartz and adularia. These minerals are indicators of higher temperature reactions and if encountered would be encouraging suggestions of a nearby hydrothermal system.

95 CHAPTER 8

CONCLUSIONS AND FUTURE WORK

This chapter summarizes the findings of this research, suggests directions for future exploration efforts, and presents some considerations for future development.

8.1 Possible Geothermal System

Mt. Spurr’s position on an active convergent plate boundary provides a reliable heat source; active regional faults could provide deeply penetrating fractures; piles of porous volcanics could provide permeability

(although volcanic rocks are reactive and deformable, and can theoretically also have very low permeability when altered); abundant rain and snowmelt could provide ample working fluid. Theoretically, a geothermal reservoir could be present in subsurface volcanic units capped by tephra, or in a possible fractured granite basement that is capped by the West Foreland Formation.

8.1.1 Extent of Hydrothermal System

The most obvious expression of a hydrothermal system at Mt. Spurr is on the south flank below Crater Peak, in Crater Canyon. That is the location of thermal springs and seeps, and waters that have the most evidence of deeply sourced magmatic contributions. No alteration was observed at the surface near any of the discharging waters in the field area, but exploration drilling has encountered a clay mineral speculated to be kaolinite in andesitic rocks in the eastern field area at Chaka Ridge, as well as montmorillonite and calcite south of Chaka Ridge. Fluid flow has occurred at shallow depths in the past, but the alteration observed could have been produced by relatively low temperature waters. It seems that the entire field area examined in this research could lie above a hydrothermal system. It’s not clear exactly how far the hydrothermal system could extend beyond the field area, or whether any compartmentalization exists at depth. Areas on the other sides of Mt. Spurr have not been explored for geothermal potential, and this is likely due to the large volume of ice and snow that covers the volcano and obstructs observation, as well as a lack of evidence for thermal manifestations anywhere but on the south flank.

8.1.2 Origin of Waters

The concentrations of dissolved constituents and isotopic composition of surface discharges at Mt. Spurr indicate that these waters do not directly represent potential geothermal reservoir fluid. For the most part, the waters observed at the surface represent shallowly circulating meteorically-derived water that has

96 interacted with SVC andesites and incorporated some volcanic steam (CO2,H2S). Some evidence of deep circulation or upwelling hydrothermal fluid can be found in peripheral and Crater Lake waters, but these have been heavily diluted with meteoric and shallow groundwater. Waters in Crater Canyon do show evidence of magmatically-sourced contributions, maybe from deep fluid that has travelled to the surface along the

Capps Glacier or Crater Canyon faults.

8.1.3 Fluid Pathways

The hydrologic framework and extent of a possible geothermal reservoir are influenced by subsurface lithology, which is only known to ˜1km depth in the study area. It appears that deep hydrothermal fluid does not reach the surface, except locally along deeply penetrating faults. The presence of a clay cap would effectively prevent communication between depth and the surface, and this is supported by the very dilute meteorically-dominated composition of surface discharges.

8.1.4 Deep temperature and Reservoir Characteristics

Temperature and deep reservoir characteristics are difficult to predict because the limited dataset that is available relies heavily on surface discharges, and most surface discharges are highly modified and/or not deeply sourced. Deeper drilling could confirm or disprove the presence of a fractured granite reservoir below the West Foreland Formation and will enhance the understanding of the hydrothermal system at Mt. Spurr. However, future drilling will have to be prepared to go through at least 900m of swelling clay.

8.2 Implications for Future Geothermal Exploration and Development

Conventional binary systems can produce electricity from water as low as 74￿ (Kagel, 2008). This minimum temperature threshold is nearly met by many of the last equilibration temperatures of the shallowly circulating waters, and likely would be met by any deeper fluid that may exist. Future studies of the hydrothermal system at Mt. Spurr would benefit from a more complete understanding of the stratigraphy of the study area, especially what lies beneath the West Foreland Formation, and an understanding of faults in terms of their fluid transport capabilities. Possible development plans will have to consider business and economic incentives, remoteness of the area, weather, and the possibility of future volcanic eruptions affecting or destroying any existing infrastructure.. Besides reservoir temperature, other considerations in potential development and power plant design include the presence of non-condensable gasses as well as the corrosiveness and scaling potential of the geothermal

fluid. Minimal amounts of gas flux at the surface of Mt. Spurr, even preceding volcanic eruptions, suggest that most SO2 is scrubbed by hydrolysis reactions (Doukas & Gerlach, 1995), while rising CO2 could be incorporated into calcite mineralization, or dissolved in shallow aquifers that subsequently become rich in

97 bicarbonate. The near-neutral pH of spring and seep samples indicates low corrosion potential, but the presence of elevated concentrations of silica, sulfate, and bicarbonate could indicate silica, carbonate, or sulfide mineral scaling potential. Additionally, incentives must exist for development to proceed-to develop geothermal energy and to develop geothermal energy in Alaska.

98 REFERENCES CITED

Alaska Climate Research Center. last checked 9/25/12. Alaska Climatology. http://climate.gi.alaska.edu/Climate/index.html, last checked 9/25/12.

Alaska Energy Authority. 2009. Alaska energy: a first step toward energy independence. Alaska Center for Energy and Power.

Arnorsson, S. 1983. Chemical equilibria in Icelandic geothermal systems. Implications for chemical geother- mometry investigations. Geothermics, 12, 119–128.

Battaglia, S. 1999. Applying X-ray geothermometer diffraction to a chlorite. Clays and Clay Minerals, 47(1), 54–63.

Bogie, I., Kusumah, Y. I., & Wisnandary, M. C. 2008. Overview of the Wayang Windu geothermal field, West Java, Indonesia. Geothermics, 37, 347–365.

Bolm, J. G., & McCulloh, T. H. 1986. Geologic Studies of the Lower Cook Inlet COST No. 1 Well, Alaska Outer Continental Shelf. Bulletin 1596. USGS. Chap. Sandstone Diagenesis.

Boss, R. F., Lennon, R. B., & Wilson, B. W. 1976. North American Oil and Gas Fields. Vol. 24. AAPG Memoir. Chap. Middle Ground Shoal Oil Field, Alaska, pages 1–22.

Bruhn, R. L., Parry, W. T., & Bunds, M. P. 2000. Tectonics, fluid migration and fluid pressure in a deformed forearc basin, Cook Inlet, Alaska. GSA Bulletin, 112(4), 550–563.

Calderwood, K. W., & Frackler, W. C. 1972. Proposed stratigraphic nomenclature for Kenai Group, Cook Inlet Basin, Alaska. AAPG Bulletin, 56(4), 739–754.

Carmody, R. W., Plummer, L. N., Busenberg, E., & Coplen, T. B. 1998. Methods for collection of dissolved sulfate and sulfide and analysis of their sulfur isotopic composition. USGS-OFR 97-234, 91pp.

Cathelineau, M. 1988. Cation site occupancy in chlorites and illites as a function of temperature. Clay Minerals, 23, 471–485.

Craig, H. 1963. Nuclear Geology in Geothermal Areas. Pisa, Italy: Consiglio Nazionale delle Ricerche, Laboratorio de Geologia Nucleare. Chap. The isotopic geochemistry of water and carbon in geothermal areas, pages 17–53.

Doukas, M. P., & Gerlach, T. M. 1995. Sulfur dioxide scrubbing during the 1992 eruptions of Crater Peak, Mount Spurr volcano, Alaska. USGS Bulletin 2139.

Ellis, A. J., & Mahon, W. A. J. 1964. Natural hydrothermal systems and experimental hot-water/rock interactions. Geochimica et Cosmochimica Acta, 28, 1323–1357.

Epstein, S., & Mayeda, T. 1998. Variation of 18O content of waters from natural sources. Geochimica et Cosmochimica Acta, 4, 213–224.

Faure, G. 1986. Principles of Isotope Geology. 2 edn. New York, NY: Wiley.

99 Field, C. W., & Fifarek, R. H. 1985. Geology and Geochemistry of Hydrothermal Systems.Reviewsin Economic Geology, vol. 2. Society of Economic Geologists. Chap. Light stable isotope systematics in the epithermal environment.

Finzel, E. S., Gillis, R. J., Ridgway, K. A., & Lepain, D. L. 2009. Preliminary evaluation of basin margin exhumation and provenance of Cenozoic strata, Chuitna and Beluga rivers area, Cook Inlet forearc basin, Alaska. ADGGS Preliminary Interpretive Report 2009-4.

Flores, R. M., Stricker, G. D., & Kinney, S. A. 2004. Alaska coal geology, resoures and coalbed methane potential. USGS DDS-77.

Fouillac, C., & Michard, G. 1981. Sodium/lithium ratio in water applied to geothermometry of geothermal reservoirs. Geothermics, 5, 41–50.

Fournier, R. O. 1979. Geochemical and hydrological considerations and the use of enthalpy-chloride diagrams in the prediction of underground conditions in hot-spring systems. Journal of Volcanology and Geothermal Research, 5, 1–16.

Fournier, R. O. 1981. Geothermal Systems: Principles and Case Histories. New York, NY: Wiley. Chap. Application of water geochemistry to geothermal exploration and reservoir engineering.

Fournier, R. O. 1985. Geology and Geochemistry of Hydrothermal Systems. Reviews in Economic Geology, vol. 2. Society of Economic Geologists. Chap. The behavior of silica in hydrothermal solutions.

Fournier, R. O. 1991. Application of Geochemistry in Geothermal Reservoir Development. UNITAR. Chap. Water geothermometers applied to geothermal energy, pages 37–69.

Fournier, R. O., & Truesdell, A. H. 1973. An empirical Na-K-Ca-geothermometer for natural waters. Geochimica et Cosmochimica Acta, 37, 1255–1275.

Gautheron, C., & Moreira, M. 2002. Helium signature of the subcontinental lithospheric mantle. Earth and Planetary Science Letters, 199, 39–47.

Gherardi, F., Panichi, C., Yock, A., & Gerardo-Abaya, J. 2002. Geochemistry of the surface and deep fluids of the Miravalles volcano geothermal system (Costa Rica). Geothermics, 31, 91–128.

Giesemann, A., Jager, H. J., Norman, A. L., Krouse, H. P., & Brand, W. A. 1994. On-line sulfur-isotope determination using an elemental analyzer coupled to a mass spectrometer. Analytical Chemistry, 66, 2816–2819.

Giggenbach, W. F. 1984. Mass transfer in hydrothermal alteration systems–a conceptual approach. Geochim- ica et Cosmochimica Acta, 48, 2693–2711.

Giggenbach, W. F. 1988. Geothermal solute equilibria. Derivation of Na-K-Mg-Ca geoindicators. Geochimica et Cosmochimica Acta, 52, 2749–2765.

Giggenbach, W. F. 1992. Isotopic shifts in waters from geothermal and volcanic systems along convergent plate boundaries and their origin. Earth and Planetary Science Letters, 113, 495–510.

Giggenbach, W.F., & Goguel, R. L. 1989. Collection and analysis of geothermal and volcanic water and gas dishcarges. New Zealand Department of Scientific and Industrial Research Report.

Giggenbach, W.F., & Stewart, M. 1982. Processes controlling the isotopic composition of steam and water discharges from steam vents and steam-heated pools in geothermal areas. Geothermics, 11, 71–80.

100 Giggenbach, W.F., Garcia, N., Londono, A., Rodriguez, V., Rojas, N., & Calvache, V. 1990. The chemistry of fumarolic vapor and thermal-spring discharges from the Nevado del Ruiz volcanic-magmatic hydrothermal system, Colombia. Journal of Volcanology and Geothermal Research, 42, 13–39.

Golden Valley Electric Association. last checked 9/25/12. SNAP Program. http://www.gvea.com/resources/snap, last checked 9/25/12.

Grassineau, N. V., Mattey, D. P., & Lowry, D. 2001. Sulfur isotope analysis of sulfide and sulfate minerals by continuous flow-isotope ratio mass spectrometry. Analytical Chemistry, 73, 220–225.

Haeussler, P. J., Bruhn, R. L., & Pratt, T. L. 2000. Potential seismic hazards and tectonics of the upper Cook Inlet basin, Alaska, based on analysis of Pliocene and younger deformation. GSA Bulletin, 112(9), 1414–1429.

Hartman, D. C., Pessel, G. H., & McGee, D. L. 1972. Kenai Group of the Cook Inlet Basin. ADGGS Open File Report.

Hayes, J. B., Harms, J. C., & Wilson, T. 1976. Symposium Proceedings, Recent and ancient sedimentary environments in Alaska. Anchorage, AK: Alaska Geological Society. Chap. Contrasts between braided and meandering stream deposits, Beluga and Sterling Formations (Tertiary), Cook Inlet, Alaska.

Henley, R. W., & Ellis, A. J. 1983. Geothermal systems ancient and modern: a geochemical review. Earth- Science Reviews, 19, 1–50.

Kagel, A. 2008. The state of geothermal technology part II: surface technology. GEA Almanac.

Keith, T. E. C., Thompson, M. T., & McGimsey, R. G. 1995. Chemistry of Crater Lake waters prior to the 1992 eruptions of Crater Peak, Mount Spurr volcano, Alaska. USGS Bulletin 2139.

Kharaka, Y. K., & Mariner, R. H. 1989. Thermal history of sedimentary basins: methods and case histories. New York, NY: Springer-Verlag. Chap. Chemical geothermometers and their application to formation waters from sedimentary basins.

Kharaka, Y. K., Lico, M. S., & Law, L. M. 1982. Chemical geothermometers applied to formation waters, Gulf of Mexico and California basins. AAPG Bulletin, 66, 588.

Kranidoitis, P., & MacLean, W. H. 1987. Systematics of chlorite alteration at the Phelps Dodge Massive Sulfide Deposit, Matagmi, Quebec. Economic Geology, 82, 1898–1911.

Kremer, M. C., & Stadnicky, G. 1985. Guide to the Geology of the Kenai Peninsula, Alaska. Alaska Geological Society. Chap. Tertiary stratigraphy of the Kenai Peninsula-Cook Inlet region.

Magoon, L. B., & Anders, D. E. 1992. Biological Markers in Sediments and Petroleum. Englewood Cliffs, NJ: Prentice Hall. Chap. Oil-to-source rock correlation using carbon-isotopic data and biological marker compounds, Cook Inlet-Alaska Peninsula, Alaska.

Magoon, L. B., Hampton, M. A., Sable, E. G., Smith, R. A., & Chemlik, F. B. 1975. Hydrocarbon potential, geologic hazards and the technology, time-frame and infrastructure for exploration and development of the Lower Cook Inlet, Alaska: a preliminary report. USGS Open File Report 75-549.

Magoon, L. B., Bouma, A. H., Fisher, M. A., Hampton, M. A., Scott, E. W., & Wilson, C. L. 1979. Resource report for proposed OCS Sale No. 60, Lower Cook Inlet-Shelikof Strait. USGS Open File Report 79-600.

101 Martini, B. A., Lide, C., Owens, L., Walsh, P., Delwiche, B., & Payne, A. 2011. Geothermal resource definition at Mt. Spurr, Alaska. GRC Transactions, 35, 897–904.

McLean, H. 1986. Geologic Studies of the Lower Cook Inlet COST No. 1 Well, Alaska Outer Continental Shelf. Bulletin 1596. USGS. Chap. Sandstone Petrography.

Monecke, T., Giorgetti, G., Scholtysek, O., Kleeberg, R., Gotze, J., Hannington, M. D., & Petersen, S. 2007. Textural and mineralogical changes associated with the incipient hydrothermal alteration of glassy dacite at the submarine PACMANUS hydrothermal system, eastern Manus Basin. Journal of Volcanology and Geothermal Research, 160, 23–41.

Montgomery, S. L., & Barker, C. E. 2003. Coalbed methane, Cook Inlet, south-central Alaska: A potential giant gas resource. AAPG Bulletin, 87(1), 1–13.

Moore, D. M., & Reynolds, R. C. 1997. X-Ray Diffraction and the Identification and Analysis of Clay Minerals. 2 edn. New York, NY: Oxford University Press.

Morrison, J., Brockwell, T., Merren, T., Fourel, F., & Phillips, A. M. 2001. On-line high-precision stable hydrogen isotopic analyses on nanoliter water samples. Analytical Chemistry, 73, 3570–3575.

Motyka, R. J., & Nye, C. J. 1993. Short Notes on Alaskan Geology. Professional Report 113. Fairbanks, AK: ADGGS. Chap. Fumarolic gas chemistry (1982) and thermal spring water chemistry, Crater Peak (1985), Mount Spurr, Alaska.

Motyka, R. J., Liss, S. A., Nye, C. J., & Moorman, M. A. 1993. Geothermal resources of the Aleutian Arc. ADGGS Professional Report 114.

Neal, C. A., McGimsey, R. G., Doukas, M. P., & Ellersieck, I. 1993. Photographs of the 1992 Eruptions of Crater Peak, Spurr Volcano, Alaska. USGS Open File Report 93-707.

Neal, C. A., McGimsey, R. G., Dixon, J. P., Manevich, A., & Rybin, A. 2008. 2006 Volcanic Activity in Alaska, Kamchatka, and the Kurile Islands: Summary of Events and Response of the Alaska Volcano Observatory. USGS Scientific Investigations Report 5214.

Nicholson, K. 1993. Geothermal Fluids. Heidelberg, Germany: Springer-Verlag.

Nieva, D., & Nieva, R. 1987. Developments in geothermal energy in Mexico, part 12-A cationic composition geothermometer for prospection of geothermal resources. Heat recovery systems and CHP, 7, 243–258.

North Carolina State University. last checked 9/25/12. Alaska Incentives/Policies for Renewables and Efficiency. http://www.dsireusa.org/incentives/index.cfm?getRE= 1?re=undefined&ee=1&spv=0&st=0&srp=1&state=AK, last checked 9/25/12.

Nye, C. J., & Turner, D. L. 1990. Petrology, geochemistry, and age of the Spurr volcanic complex, eastern Aleutian arc. Bulletin of Volcanology, 205–226.

Nye, C. J., Waythomas, C. F., & Layer, P. W. 2005. Geologic Map of Mount Spurr, Alaska. Alaska Division of Geological and Geophysical Surveys and Alaska Volcano Obseratory Report of Investigations 2005-xxx.

Ormat Nevada, Inc. 2012a. Drill records, logs, reports from Mt. Spurr Geothermal Exploration; Well 26-11; Tyonek Quadrangle, Alaska. ADGGS Geologic Minerals Center Data Report (pending).

Ormat Nevada, Inc. 2012b. Drill records, logs, reports from Mt. Spurr Geothermal Exploration; Wells 62-2 and 67-34; Tyonek Quadrangle, Alaska. ADGGS Geologic Minerals Center Data Report (pending).

102 Plafker, G., Nokleberg, W. J., & Lull, J. S. 1989. Bedrock geology and tectonic evolution of the Wrangellia, Peninsular, and Chugach terranes along the Trans-Alaska Crustal Transect in the Chugach Mountains and southern Copper River Basin, Alaska. Journal of Geophysical Research, 94(B4), 4255–4295.

Powell, T. 2000. A review of exploration gas geothermometry. Proceedings: Twentyy-Fifth Workshop on Geothermal Reservoir Engineering, Stanford University SGP-TR-165.

Powell, T., & Cumming, W. 2010. Spreadsheets for geothermal water and gas chemistry. Proceedings: Thirty-Fifth Workshop on Geothermal Reservoir Engineering, Stanford University SGP-TR-188.

Power, J. A., Jolly, A. D., Page, R. A., & McNutt, S. R. 1995. Seismicity and forecasting of the 1992 eruptions of Crater Peak vent, Mount Spurr volcano, Alaska: an overview. USGS Bulletin 2139, 149–159.

Reed, B. L., & Lanphere, M. A. 1973. Alaska-Aleutian Range Batholith: geochronology, chemistry and relation to circum-pacific plutonism. GSA Bulletin, 84, 2583–2610.

Reyes, A. G. 1990. Petrology of Philippine geothermal systems and the application of alteration mineralogy to their assessment. Journal of Volcanology and Geothermal Research, 43, 279–309.

Riehle, J. R. 1985. A reconnaisance of the major Holocene tephra deposits in the Upper Cook Inlet region, Alaska. Journal of Volcanology and Geothermal Research, 37–74.

Rochelle, C. A., Milodowski, A. E., Savage, D., & Corella, M. 1989. Secondary mineral growth in fractures in the Miravalles geothermal system, Costa Rica. Geothermics, 18(1/2), 279–286.

Simmons, S. F. 2008. Introduction to Geochemical Aspects of Geothermal Systems. Class notes. Chap. Overview of the chemistry of geothermal fluids.

Simmons, S. F., & Christenson, B. W. 1994. Origins of calcite in a boiling geothermal system. American Journal of Science, 294, 361–400.

Stimac, J., Nordquist, G., Suminar, A., & Sirad-Azwar, L. 2008. An overview of the Awibengkok geothermal system, Indonesia. Geothermics, 37, 300–331.

Tonani, F. 1980. Some remarks on the application of geochemical techniques in geothermal exploration. Proc. Adv. Eur. Geoth. Res., Second symposium, Strasbourg, 428–443.

Trop, J. M., Szuch, D. A., Rioux, M., & Blodgett, R. B. 2005. Sedimentology and provenance of the Upper Jurassic Nanek Formation, Talkeetna Mountains, Alaska: Bearings on the accretionary tectonic history of the Wrangellia composite terrane. GSA Bulletin, 117(5/6), 570–588.

Truesdell, A. H. 1976. Summary of Section III. Geochemical techniques in exploration. Proceedings 2nd UN Symposium on the development and use of geothermal resources, San Francisco, 1975, 1, liii–lxxix.

Truesdell, A. H., & Hulston, J. R. 1980. Isotopic evidence on environments of geothermal systems: in Handbook of Environmental Isotope Geochemistry.Elsevier.

Turner, D. L., & Wescott, E. M. 1986. Geothermal energy resource investigations at Mt. Spurr, Alaska. UAG R-308.

U.S. Department of Agriculture Natural Resources Conservation Service. last checked 9/26/12. Plants Profile: Mimulus guttatus DC. seep monkeyflower. http://plants.usda.gov/java/profile?symbol=migu.

U.S. Department of Energy. 2005. Alaska. Geothermal Technologies Program.

103 Waltham, T. 1995. A guide to the geology of Alaska and Yukon. Proceedings of the Geologists’ Association, 313–332.

Waythomas, C. F., & Nye, C. J. 2002. Preliminary volcano-hazard assessment for Mount Spurr volcano, Alaska. USGS Open File Report 01-482.

Wilson, F. H., Hults, C. P., Schmoll, H. R., Haeussler, P. J., Schmidt, J. M., Yehle, L. A., & Labay, K. A. 2009. Preliminary Geologic Map of the Cook Inlet Region, Alaska. USGS Open File Report 2009-1108.

104