THE TIMING AND STRUCTURAL EVOLUTION OF THE DONLIN CREEK GOLD DEPOSIT, SOUTHWEST ALASKA by KENNETH DANIEL MACNEIL B.Sc. (Advanced Major), St. Francis Xavier University, 2001 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Geological Sciences) The University of British Columbia (Vancouver) September 2009 © Kenneth Daniel MacNeil ii ABSTRACT Lithologic and structural data from more than 220 structurally oriented diamond drill holes are used to delineate the subsurface geology, 3-D geometry and to constrain the structural evolution of the >30 million ounce Donlin Creek gold deposit. The geology of the Donlin Creek gold deposit is characterized by north-northeast – south- southwest shortening that produced structural fabrics similar to those formed in thin- skinned fold and thrust belts. This deformation consists of dominantly southwest dipping thrust faults, subordinate northeast dipping back thrusts, and associated thrust ramp anticlines in the Kuskokwim Group sedimentary rocks. The fold and thrust style deformation is locally overprinted by northeast trending, low amplitude open folds interpreted to be related to regional dextral strike-slip tectonics. Low displacement, north-northeast striking, steeply southeast dipping normal faults cut the thrust faults and both generations of folds. The normal faults, northeast trending folds, and older fold and thrust geometry are intruded by a porphyritic dyke and sill swarm. Bedding discordant intrusions are emplaced along the north-northeast striking, steeply southwest dipping normal faults. Bedding concordant intrusions are most common in the folded sedimentary rocks in the hanging walls of thrust faults. The intrusions are cut by tensional gold being quartz ± carbonate + sulphide veins, which represent the youngest structural fabric. The distribution of veins is dictated by the rheology of the host rocks. For example, the brittle intrusions are the best host to gold mineralization; however veins are also present in coarse grained sedimentary rocks. U-Pb and 40Ar/39Ar geochronology provide absolute ages for deposition of the Kuskokwim sediments (~ 88 Ma), emplacement of the felsic intrusions (~71 Ma) and timing of gold mineralization (~ 71 Ma). These data indicate approximately 14.9 million years between initial sedimentation, emplacement of intrusions, and precipitation of gold mineralization. Geochronology constrains the timing of fold and thrust style deformation, and overprinting north-northeast trending folds to between ~ 88 Ma and ~ 71 Ma. The available geologic and geochronologic data also indicates that intrusion of the post-kinematic Late Cretaceous dyke swarm was closely followed in time by deposition of >30 Moz of gold. iii TABLE OF CONTENTS ABSTRACT................................................................................................................................................. ii TABLE OF CONTENTS.............................................................................................................................iii LIST OF TABLES........................................................................................................................................v LIST OF FIGURES.....................................................................................................................................vi ACKNOWLEDGEMENTS ..........................................................................................................................x CO-AUTHORSHIP STATEMENT ..............................................................................................................xi CHAPTER 1: INTRODUCTION .................................................................................................................. 1 1.1 GENERAL INTRODUCTION...................................................................................................... 1 1.2 LOCATION................................................................................................................................. 1 1.3 OBJECTIVE ............................................................................................................................... 3 1.4 THESIS PRESENTATION ......................................................................................................... 4 1.5 METHODOLOGY....................................................................................................................... 4 1.6 GEOLOGY OF THE KUSKOKWIM MOUNTAINS MINERAL BELT......................................... 16 1.7 TECTONIC EVOLUTION OF SOUTHWESTERN ALASKA..................................................... 17 1.8 GEOLOGY OF THE KUSKOKWIM MOUNTAINS.................................................................... 19 1.8.1 Igneous History .................................................................................................................... 22 1.9 GEOLOGY OF THE DONLIN CREEK PROPERTY................................................................. 24 1.9.1 Kuskokwim Group................................................................................................................ 24 1.9.2 Igneous Rocks ..................................................................................................................... 25 1.9.3 Timing of Igneous Activity .................................................................................................... 26 1.9.4 Structural Setting of Donlin Creek ....................................................................................... 26 1.9.5 Mineralization....................................................................................................................... 29 1.9.6 Alteration.............................................................................................................................. 29 1.10 CLASSIFICATION OF THE DONLIN CREEK GOLD DEPOSIT .............................................................. 34 REFERENCES...................................................................................................................................... 39 CHAPTER 2: GEOCHRONOLOGY OF THE DONLIN CREEK GOLD DEPOSIT, SOUTHWEST ALASKA.................................................................................................................................................... 43 2.1 INTRODUCTION...................................................................................................................... 43 2.2 REGIONAL GEOLOGY AND TECTONICS.............................................................................. 43 2.3 DONLIN CREEK PROPERTY GEOLOGY............................................................................... 46 2.3.1 Sedimentary Rocks.............................................................................................................. 46 2.3.2 Igneous Rocks ..................................................................................................................... 49 2.3.3 Structural Fabric................................................................................................................... 49 2.3.4 Mineralization....................................................................................................................... 56 2.3.5 Alteration.............................................................................................................................. 58 2.4 PREVIOUS RADIOMETRIC DATING ...................................................................................... 60 2.4.1 TIMING OF IGNEOUS ACTIVITY ....................................................................................... 60 2.4.2 TIMING OF HYDROTHERMAL ACTIVITY.......................................................................... 60 2.5 METHODOLOGY..................................................................................................................... 63 2.5.1 Shortwave Infrared Analysis ................................................................................................ 63 2.5.2 U-Pb GEOCHRONOLOGY.................................................................................................. 66 2.5.3 40Ar/39Ar GEOCHRONOLOGY ............................................................................................ 70 2.6 DATA........................................................................................................................................ 73 2.6.1 SHORTWAVE INFRARED SAMPLE DESCRIPTIONS AND DATA ................................... 73 2.6.2 SHORTWAVE INFRARED INTERPRETATION.................................................................. 78 2.7 RADIOMETRIC DATING RESULTS ........................................................................................ 78 2.7.1 238U/206Pb (SHRIMP-RG) Zircon Geochronology Samples and Results ............................. 78 2.7.2 40Ar/39Ar Geochronology Samples and Results................................................................... 81 iv 2.8 ANALYSIS................................................................................................................................ 85 2.8.1 SHORTWAVE INFRARED RESULTS................................................................................. 85 2.8.2 40Ar/39Ar Geochronology ...................................................................................................... 86 2.8.3 U/Pb (SHRIMP-RG) Zircon Geochronology ........................................................................ 87 2.9 DISCUSSION........................................................................................................................... 87 2.10 CONCLUSION ......................................................................................................................... 89 REFERENCES...................................................................................................................................... 90 CHAPTER 3: STRUCTURAL GEOLOGY AND GEOMETRY OF THE DONLIN CREEK GOLD DEPOSIT ................................................................................................................................................... 93 3.1 GENERAL INTRODUCTION.................................................................................................... 93 3.2 REGIONAL GEOLOGIC SETTING .......................................................................................... 93 3.3 LOCAL GEOLOGIC SETTING................................................................................................. 96 3.4 OUTLINE.................................................................................................................................. 99 3.5 GEOLOGY OF THE MAIN RESOURCE AREA...................................................................... 114 3.6 STRUCTURAL FABRICS OF THE MAIN RESOURCE AREA............................................... 117 3.6.1 Pre-Intrusion Structural Fabrics of the Main Resource Area............................................. 118 3.6.2 Syn-Intrusion Structural Fabrics of the Main Resource Area ............................................ 125 3.6.3 Post-Intrusion Structural Fabrics of the Main Resource Area ........................................... 130 3.6.4 Trench Geology ................................................................................................................. 143 3.6.5 Lewis Gravel Pit Geology .................................................................................................. 146 3.7 INTERPRETATION OF STRUCTURAL FABRICS ................................................................ 150 3.8 GEOMETRY MAIN RESOURCE AREA................................................................................. 152 3.8.1 Fold and Thrust Geometry................................................................................................. 152 3.8.2 Intrusive Geometry............................................................................................................. 154 3.8.3 Post Intrusion Deformation ................................................................................................ 156 3.9 SUMMARY AND INTERPRETATION OF STRUCTURAL FABRICS..................................... 157 3.10 THE EVOLUTION OF THE DONLIN CREEK GOLD DEPOSIT ............................................. 159 3.10.1 Regional Structural and Tectonic Fabrics ..................................................................... 159 3.10.2 Regional Synthesis........................................................................................................ 161 3.10.3 Regional Tectonic Model ............................................................................................... 162 3.11 CONCLUSION ....................................................................................................................... 165 REFERENCES.................................................................................................................................... 168 CHAPTER 4: SUMMARY ....................................................................................................................... 172 4.1 CONCLUSIONS..................................................................................................................... 172 4.2 CRITIC OF RESULTS ............................................................................................................ 173 4.3 EXPLORATION IMPLICATIONS ........................................................................................... 174 4.4 RECCOMEDATIONS FOR FUTURE WORK......................................................................... 175 REFERENCES.................................................................................................................................... 177 APPENDIX I: ZIRCON SPOT LOCATIONS FOR SHIRMP ANALYSES............................................... 178 A-I APPENDIX I: ZIRCON SPOT LOCATIONS FOR SHIRMP ANALYSES............................................. 178 A-II APPENDIX II: REVERSE ISOCHRON PLOTS FOR 40AR-39AR GEOCHRONOLOGY........................... 201 v LIST OF TABLES Chapter 1 TABLE 1.1. FOLDING IN THE KUSKOKWIM BASIN. ........................................................................................... 20 TABLE 1.2. THE CHARACTERISTICS OF IGNEOUS ROCKS AT DONLIN CREEK. .................................................. 27 TABLE 1.3. KEY CHARACTERISTICS OF DONLIN CREEK GOLD MINERALIZATION................................................ 30 TABLE 1.4. SUMMARIZED KEY CHARACTERISTICS OF THE DONLIN CREEK GOLD DEPOSIT ................................ 35 Chapter 2 TABLE 2.1. U-PB SHRIMP-RG ZIRCON ANALYTICAL DATA FOR SAMPLES IN THE STUDY AREA. ........................ 67 TABLE 2.2. 40AR/39AR RESULTS FROM ALTERATION MINERALS ASSOCIATED WITH GOLD MINERALIZATION AT DONLIN CREEK. .................................................................................................................................. 71 vi LIST OF FIGURES Chapter 1 FIGURE 1.1. TERRANE MAP OF THE WESTERN MARGIN OF NORTH AMERICA...................................................... 2 FIGURE 1.2. GEOLOGY OF THE KUSKOKWIM MOUNTAINS AND LOCATION OF THE ~ 70 MA MINERAL DEPOSITS. . . 5 FIGURE 1.3. DONLIN CREEK PROPERTY SURFACE GEOLOGY MAP INTERPRETED FROM DRILL CORE INTERCEPTS. 7 FIGURE 1.4. DRILL HOLE AND TRENCH LOCATION MAP FOR THE MAIN RESOURCE AREA OF THE DONLIN CREEK PROPERTY............................................................................................................................................ 9 FIGURE 1.5. THE DISTRIBUTION OF DRILL HOLE ORIENTATIONS, AND THE DISTRIBUTION OF PLANAR STRUCTURAL MEASUREMENTS FOR THE DONLIN CREEK GOLD DEPOSIT ...................................................................... 12 FIGURE 1.6. THE FREQUENCY DISTRIBUTION OF ALPHA ANGLES FOR BEDDING, VEINS, FAULTS AND FRACTURES COMPARED TO THEIR EXPECTED NORMAL DISTRIBUTION FOR THE DONLIN CREEK GOLD DEPOSIT ............ 13 FIGURE 1.7. FOLD PATTERNS (AXIAL TRACES) IN THE KUSKOKWIM MOUNTAINS MINERAL BELT. ...................... 21 FIGURE 1.8. DONLIN CREEK IGNEOUS AND SEDIMENTARY ROCKS, EXAMPLES AND TEXTURES.......................... 23 FIGURE 1.9. SIMPLIFIED CLAY ALTERATION AT DONLIN CREEK....................................................................... 32 Chapter 2 FIGURE 2.1. DONLIN CREEK ROCK TYPES..................................................................................................... 47 FIGURE 2.2. GEOLOGY OF THE MAIN RESOURCE AREA OF DONLIN CREEK..................................................... 50 FIGURE 2.3. CROSS-SECTIONS DRAWN PERPENDICULAR TO BEDDING ............................................................ 52 FIGURE 2.4. CROSS SECTIONS DRAWN PERPENDICULAR TO DISCORDANT IGNEOUS ROCKS. ............................ 54 FIGURE 2.5. DONLIN CREEK VEIN TYPES ...................................................................................................... 57 FIGURE 2.6. LASER STEP HEATING 40AR/39AR AGE SPECTRA FOR ALTERATION ASSOCIATED WITH GOLD MINERALIZATION AT THE DONLIN CREEK PROPERTY. ............................................................................. 82 Chapter 3 FIGURE 3.1. SIMPLIFIED GEOLOGY OF THE DONLIN CREEK. ........................................................................... 97 FIGURE 3.2. INTERPRETED SURFACE GEOLOGY OF THE MAIN RESOURCE AREA. .......................................... 100 FIGURE 3.3. CROSS-SECTION 541700 ....................................................................................................... 102 FIGURE 3.4. CROSS-SECTION 687850 ....................................................................................................... 104 FIGURE 3.5. CROSS-SECTION 6979800. .................................................................................................... 106 vii FIGURE 3.6. CROSS-SECTION 539600 ....................................................................................................... 108 FIGURE 3.7. CROSS-SECTION 540100 ....................................................................................................... 110 FIGURE 3.8. CROSS-SECTION 541300 ....................................................................................................... 112 FIGURE 3.9. NORTH-NORTHEAST-TRENDING, STEEPLY-DIPPING NORMAL FAULT THAT OFFSETS A CONCORDANT INTRUSION. ....................................................................................................................................... 116 FIGURE 3.10. QUARTZ VEIN EXAMPLES....................................................................................................... 119 FIGURE 3.11. BEDDING, FAULT AND CONTACT PHOTOGRAPHS AT DONLIN CREEK. ........................................ 120 FIGURE 3.12. A) POLES TO BEDDING PLANES AND B) CONTOURED POLES TO BEDDING FOR THE DONLIN CREEK PROPERTY ........................................................................................................................................ 122 FIGURE 3.13. POLES TO BEDDING PLANES SHOWN FOR EACH PROSPECT IN THE MAIN RESOURCE AREA.. ..... 123 FIGURE 3.14. A) POLES TO FAULT PLANES PLOTTED FOR ALL FAULTS, B) FAULTS HOSTED IN SEDIMENTARY ROCKS AND C) FAULTS HOSTED IN IGNEOUS ROCKS............................................................................. 126 FIGURE 3.15. POLES TO FAULT PLANES FOR THE MAIN RESOURCE AREA. .................................................... 127 FIGURE 3.16. SLICKENLINES FOR THE MAIN RESOURCE AREA AND AIRSTRIP. ................................................ 129 FIGURE 3.17. A) POLES AND B) CONTOURED POLES TO IGNEOUS CONTACTS FROM ORIENTED DRILL CORE ACROSS THE DONLIN CREEK PROPERTY. ............................................................................................ 131 FIGURE 3.18. STRUCTURAL GEOLOGY OF THE DONLIN CREEK AIRSTRIP ...................................................... 132 FIGURE 3.19. A) POLES AND B) CONTOURED POLES TO BEDDING PLANES FOR THE OUTCROPS ALONG THE DONLIN CREEK AIRSTRIP................................................................................................................... 134 FIGURE 3.20. POLES TO VEINS CLASSIFIED BY COMPOSITION FOR THE DONLIN CREEK GOLD DEPOSIT ........... 136 FIGURE 3.21. POLES TO VEINS (UNDIFFERENTIATED) SHOWN BY ROCK TYPE AND CLASSIFIED ACCORDING TO GOLD GRADE.. ................................................................................................................................... 137 FIGURE 3.22. THE RELATIONSHIP BETWEEN VEIN ORIENTATION AND HOST ROCK COMPOSITION. .................... 139 FIGURE 3.23. A) POLES TO VEINS (UNDIFFERENTIATED) FOR THE DONLIN CREEK PROPERTY........................ 140 FIGURE 3.24. POLES TO VEINS (UNDIFFERENTIATED) BY PROSPECT AREA. ................................................... 141 FIGURE 3.25. STRUCTURAL DATA COMPILED FROM HISTORIC TRENCH MAPPING............................................ 144 FIGURE 3.26. GEOLOGY OF THE LEWIS HILL GRAVEL PIT ............................................................................ 147 FIGURE 3.27. FAULT GOUGE ON THE DIP SLOPE OF CONCORDANT INTRUSIVE BODY IN THE LEWIS HILL GRAVEL PIT. GOUGE CONTAINS SOME SEDIMENTARY ROCK FRAGMENTS (AT PENCIL TIP) ................................... 149 FIGURE 3.28. STEEPLY DIPPING VEINS AND FRACTURES WITH ASSOCIATED ALTERATION HALOS IN THE LEWIS HILL GRAVEL PIT................................................................................................................................ 149 viii FIGURE 3.29. GOLD MINERALIZATION FOR THE MAIN RESOURCE AREA. ....................................................... 153 FIGURE 3.30. SCHEMATIC BLOCK DIAGRAM SHOWING THE POSSIBLE RELATIONSHIP BETWEEN FAULTS AND VEINS IN AN EXTENSIONAL ENVIRONMENT. .................................................................................................... 160 FIGURE 3.31. PLATE RECONSTRUCTIONS FOR 60 MA REPRODUCED FROM A) ENGBRETSON ET AL. (1985), B) BRADLEY ET AL. (1993) AND C) MILLER ET AL. (2002). FIGURE ADAPTED FROM MILLER ET AL. (2002).. 163 FIGURE 3.32. TECTONIC RECONSTRUCTION FOR THE PACIFIC MARGIN OF NORTH AMERICA BETWEEN 84 AND 52 MA. .................................................................................................................................................. 166 Appendix I FIGURE A1.1. ZIRCON SPOT LOCATIONS FOR SHIRMP SAMPLE DC06-1646.. ............................................. 179 FIGURE A1.2. ZIRCON SPOT LOCATIONS FOR SHIRMP SAMPLE DC06-1646. .............................................. 180 FIGURE A1.3. ZIRCON SPOT LOCATIONS FOR SHIRMP SAMPLE DC06-1646. .............................................. 181 FIGURE A1.4. ZIRCON SPOT LOCATIONS FOR SHIRMP SAMPLE DC06-1646. .............................................. 182 FIGURE A1.5. ZIRCON SPOT LOCATIONS FOR SHIRMP SAMPLE GRAVEL PIT................................................ 183 FIGURE A1.6. ZIRCON SPOT LOCATIONS FOR SHIRMP SAMPLE GRAVEL PIT................................................ 184 FIGURE A1.7. ZIRCON SPOT LOCATIONS FOR SHIRMP SAMPLE GRAVEL PIT................................................ 185 FIGURE A1.8. ZIRCON SPOT LOCATIONS FOR SHIRMP SAMPLE GRAVEL PIT................................................ 186 FIGURE A1.9. ZIRCON SPOT LOCATIONS FOR SHIRMP SAMPLE GRAVEL PIT................................................ 187 FIGURE A1.10. ZIRCON SPOT LOCATIONS FOR SHIRMP SAMPLE DC07-1415 ............................................. 188 FIGURE A1.11. ZIRCON SPOT LOCATIONS FOR SHIRMP SAMPLE DC07-1415. ............................................ 189 FIGURE A1.12. ZIRCON SPOT LOCATIONS FOR SHIRMP SAMPLE DC07-1415. ............................................ 190 FIGURE A1.13. ZIRCON SPOT LOCATIONS FOR SHIRMP SAMPLE DC07-1415. ............................................ 191 FIGURE A1.14. ZIRCON SPOT LOCATIONS FOR SHIRMP SAMPLE DC06-1280. ............................................ 192 FIGURE A1.15. ZIRCON SPOT LOCATIONS FOR SHIRMP SAMPLE DC06-1280. ............................................ 193 FIGURE A1.16. ZIRCON SPOT LOCATIONS FOR SHIRMP SAMPLE DC06-1280. ............................................ 194 FIGURE A1.17. ZIRCON SPOT LOCATIONS FOR SHIRMP SAMPLE DC06-1280. ............................................ 195 FIGURE A1.18. ZIRCON SPOT LOCATIONS FOR SHIRMP SAMPLE DC06-1280. ............................................ 196 FIGURE A1.19. ZIRCON SPOT LOCATIONS FOR SHIRMP SAMPLE DC06-1280. ............................................ 197 FIGURE A1.20. ZIRCON SPOT LOCATIONS FOR SHIRMP SAMPLE DC06-1285. ............................................ 198 ix FIGURE A1.21. ZIRCON SPOT LOCATIONS FOR SHIRMP SAMPLE DC06-1285. ............................................ 199 FIGURE A1.22. ZIRCON SPOT LOCATIONS FOR SHIRMP SAMPLE DC06-1285. ............................................ 200 Appendix II FIGURE A2.1. 40AR-39AR PLATEAU AND INVERSE ISOCHRON AGE DETERMINATIONS FOR SAMPLE DC06-1251-AR. ......................................................................................................................... 201 FIGURE A2.2. 40AR-39AR PLATEAU AND INVERSE ISOCHRON AGE DETERMINATIONS FOR SAMPLE DC06-1284-AR. ......................................................................................................................... 203 FIGURE A2.3. 40AR-39AR PLATEAU AND INVERSE ISOCHRON AGE DETERMINATIONS FOR SAMPLE DC06-1290_1-AR...................................................................................................................... 205 FIGURE A2.4. 40AR-39AR PLATEAU AND INVERSE ISOCHRON AGE DETERMINATIONS FOR SAMPLE DC06-1280_2-AR...................................................................................................................... 207 FIGURE A2.5. 40AR-39AR PLATEAU AND INVERSE ISOCHRON AGE DETERMINATIONS FOR SAMPLE DC06-1415_1-AR...................................................................................................................... 209 FIGURE A2.6. 40AR-39AR PLATEAU AND INVERSE ISOCHRON AGE DETERMINATIONS FOR SAMPLE DC06-1415_2-AR...................................................................................................................... 213 x ACKNOWLEDGEMENTS This project was funded by Barrick Gold Corp. and would not have been possible without their financial and logistical support. Francois Robert, Paul Dobak, Rich Harris, Rob Brown, Glenn Asch and Rob Krcmarov, thank you for your advice, support and patience. Additional funding for this project was provided by a personal SEG research grant. I am indebted to my academic supervisor, Dick Tosdal. His guidance, patients, and sage advice have been invaluable throughout this project. Thank you for taking the time to edit my manuscripts, and to teach me about fold and thrust belts. I want to thank Adam Simmons for showing me the ropes on the SHRIMP-RG, and for taking the time to discuss U-Pb and 40Ar/39Ar geochronology. Adrian Newton thanks very much for teaching me to use the Terraspec® and interpret the results. I would also like to thank Ken Hickey and Jim Mortensen for their edits and constructive criticism. xi CO-AUTHORSHIP STATEMENT The four chapters of this dissertation are written as manuscripts for publication in refereed journals. Each chapter represents primarily my own initiative, except where noted below. I plan to submit each of the manuscripts with my supervisor, Richard Tosdal as a co-author. Richard Tosdal participated in all the stages of this research project included project design, establishing field protocol and manuscript editing. Chapter 2 Geochronology of the Donlin Creek gold deposit, southwest Alaska Authors: Daniel MacNeil, Richard Tosdal, Joseph L. Wooden, Thomas D. Ullrich Richard Tosdal provided research guidance and contributed to the editing of this chapter. Joeseph Wooden contributed to the work at the SHRIMP-RG lab, including interpretation of the data for 5 samples. Thomas Ullrich provided 40Ar/39Ar Geochronology for 6 samples and assisted with the interpretation of the data. They will all provide revisions of the manuscript before publication. Chapter 3 Structural geology and geometry of the Donlin Creek gold deposit. Authors: Daniel MacNeil, Richard Tosdal. Richard Tosdal provided research guidance, contributed to the editing of this chapter and donated a considerable amount of time in field discussing the various geologic challenges presented in this chapter. 1 1 CHAPTER 1: INTRODUCTION 1.1 GENERAL INTRODUCTION The > 30 million ounce Donlin Creek deposit in the Kuskokwim Mountains of southwestern Alaska (Fig 1.1) has characteristics of epithermal, reduced granite, Carlin style and orogenic gold deposits. It has been classified as alkalic-rock-related (Bundtzen and Miller, 1997), low sulphidation epithermal (Ebert et al, 2000a), shallow reduced intrusion related or high level intrusion related (Ebert et al, 2002; Hart et al., 2002), and orogenic gold mineralization (Goldfarb et al., 2002). The structural evolution of this gold deposit and host rocks as well as the timing of those events with respect to the geologic history is critical to any of these models. Donlin Creek is a high tonnage, low grade gold deposit that has measured and indicated resource potential of 371.7 Mt @ 2.46 g/t Au (29.38 Moz Au) with an additional inferred resource of 46.5 Mt @ 2.46 g/t Au (3.46 Moz Au) (NovaGold Resources inc. press release February 7, 2008). Gold mineralization is refractory and is associated with arsenopyrite. The highest gold grades are located at intersections between mineralized fracture zones and granite porphyry intrusions Due to the lack of high quality bedrock exposure the subsurface geometry of the greater than 30 Moz Donlin Creek gold deposit has not been formally documented. Previous studies have inferred the subsurface geometry of Donlin Creek by examining the distribution of porphyry intrusions which occur as dykes and sills across the entire property. This study places the intrusive rocks in the context of the geology of their host sedimentary rocks and for the first time presents an interpretation of the 3- dimensional geometry of Donlin Creek that integrates surface and subsurface lithologic and structural data with alteration and gold distribution. The study constrains the timing of deposition of the sedimentary rocks and the age of both the intrusions and gold mineralization. 1.2 LOCATION The property is located along the western margin of the Kuskokwim Mountains approximately 480 kilometers northwest of Anchorage (Fig. 1.1). Crooked Creek, the 3 closest village to the project, approximately 20 kilometers to the south along the Kuskokwim River, the closest navigable waterway to the project (Fig. 1.2). The Donlin Creek property has a 1,650 meter gravel airstrip for access and 150-person camp located on American Ridge (Fig 1.3). Surface and subsurface rights of the 109 km2 Donlin Creek property are owned by regional native corporations. The Calista Corporation holds the subsurface rights and the Kuskokwim Corporation holds the surface rights. The Donlin Creek project is jointly controlled by Barrick Gold Corporation and Nova Gold Resources Inc. The project is currently managed by a limited liability company, the Donlin Creek LLC. Calista Corporation has the right to earn up to 15 percent by making pro-rata share payments of the project’s capital costs. The Kuskokwim Mountains are characterized by low topographic relief. Elevations range from 150-460 meters above sea level. Ridges are typically rounded and are dominated by alpine tundra vegetation. Outcrop on the property is rare. Rubble crop accounts for approximately 5% of the surface area of the property. Hillsides in this region are covered by soil solifluction lobes. Lowlands are forested with black spruce, tamarack, alder, birch, larch, muskeg, permafrost and stunted black spruce are typical of lowest elevations and poorly drained areas. 1.3 OBJECTIVE This project seeks to define the 3 dimensional geologic architecture of the main resources area of the Donlin Creek gold deposit. It delineates the 3-D geometry of the igneous bodies and identifies the structural and lithologic controls on their emplacement into the host sedimentary rocks. The intensity and distribution of gold mineralization is examined in this context. Several key questions about the geology of the gold deposit guide the research. These are: 1. What is the structural evolution of Donlin Creek? 2. What is the 3-dimensional geometry of Donlin Creek? 3. What are the controls on dyke and sill emplacement? 4 4. What are the controls on the distribution of veins? 1.4 THESIS PRESENTATION This thesis is presented in manuscript format, in accordance with the University of British Columbia guidelines. Following an introductory chapter, the results are presented as two individual research papers. Chapter 1 summarizes the regional, camp and local scale geology of the deposit. It presents structural, alteration and geochronological research done in previous studies. Gold mineralization is discussed in the context of the various gold ore deposit models that have been proposed. Chapter 2 establishes the timing of key geologic events and presents U-Pb and 40Ar-39Ar geochronology data. This data constrains the age of deposition of the host sedimentary rocks, the timing of intrusion and the timing of gold mineralization. Chapter 3 contains a detailed structural analysis, and provides a 3-dimensional interpretation of the subsurface geometry on a series of 6 intersecting cross-sections. This chapter discusses the evolution of Donlin Creek and places the deposit in the context of the tectonic evolution of southwest Alaska. Chapter 4 discusses the key results of this study and contains recommendations for future research. 1.5 METHODOLOGY This thesis utilizes data from a number of sources, including modern and historic exploration drilling databases, 3-dimensional models, drill core logging, and surface mapping. Two key structural geology databases were compiled from oriented drill core and surface exploration trench maps. These databases form a critical basis of this study because the Donlin Creek property contains very few intact bedrock exposures. Where present, the majority of the exposure on the property consists of rubble and frost-heaved and rotated blocks. A few important outcrops are located along the Donlin Airstrip and in the Lewis Road gravel pit, located in the Akivik Prospect (Figs. 1.3 and 1.4). These databases, data validation processes, and the graphical products generated from these them are described below. 5 Figure 1.2. Geology of the Kuskokwim Mountains and location of the ~ 70 Ma mineral deposits. Dark grey stars represent prospects with an epithermal affinity (Hg, As, Sb geochemical signature), light grey stars show prospects with a magmatic affinity (Bi, Bo, W geochemical signature). Modified after Ebert et. al. (2002). 6 Structurally Oriented Diamond Drill Core Data The Donlin Creek (Barrick Gold Corp) oriented drill core structural database was evaluated at the beginning of this project. As a first-pass evaluation, vein orientations were examined and compared to a high quality database of surface vein measurements complied by Ebert et al. (2003a). Systematic and non-systematic errors are present in this database. The most obvious source of error in this database consists of a 180º rotation of some of the planar fabrics. In general, this error correlates to the year that the data was collected. Since the vein orientations in this database were calculated from the original alpha, beta and gamma measurements, it is likely that this error of is a result of changing the location of the reference line that angles around the core axis are measured from (i.e., top of the drill core or the bottom of the drill core). Non-systematic errors are also present and produce a very high degree of scatter. These errors yield small circle distributions about 000º/90º when fabrics are plotted on stereonets. Since it is not possible to determine the sources of all of the inconsistencies in this database it has been abandoned. A systematic compilation of historic drill hole data was done in order to reconstitute the structural database. This compilation includes Westgold, Placer Dome, Nova Gold, BGC Engineering and Barrick Gold Corp structural databases. The most complete and accurate dataset was obtained from the BGC Engineering. The BGC Engineering database was validated (see below) and a large percentage of these data were used as a starting position for constructing a complete, high quality structural database for the Donlin Creek deposit. Individual drill holes that could be validated from the databases were added. This compilation effort produced a total 332 structurally oriented drill holes and includes 27,895 structural measurements. Structural Data from Surface Trenching Data Structural data from 42 historic surface trenches were compiled from paper maps found at the Donlin Creek Exploration Camp. This compilation yielded a high quality structural database consisting of 2,470 structural measurements. This dataset is significant because it consists of data collected from exposed rocks at the surface which can be used to validate the subsurface database (see Chapter 3). 7 Figure 1.3. Donlin Creek property surface geology map interpreted from drill core intercepts due to lack of outcrop. This map shows prospect locations, the airstrip location and the location of the Main Resource Area. Modified after Szumigala et al. (2000). 8 Data Validation Structural data obtained from the drill hole, and historic trench databases were validated in several ways. First the data was first compared to established vein orientations collected form surface trenches (e.g., Ebert et. al., 2003a). The veins were plotted on a stereonet to ensure that they conform to surface measurements. Next, the orientation all other structural fabrics were examined in order to check for spurious measurements. This was done visually by examining the distribution of each fabric on a stereonet. Also, planar fabrics were plotted as though the represent linear fabrics (Fig. 1.5). Errors associated with calculating planar and linear attitudes from alpha, beta and gamma angles can be detected by plotting each structural measurement as though they represent the trend and plunge of a line. Errors commonly appear as small circle distributions about the most common drill hole orientation. The distribution of points for all the structural measurements suggests that the alpha, beta, and gamma angles are calculated correctly (Fig. 1.5). If spurious data was found, the entire drill hole was omitted from the database. Next, the two databases were checked for bias. The two most common biases found in large drill hole databases include sampling bias and drill hole orientation bias. The majority of the historic structural data used in this project has been collected by measuring alpha, beta, and gamma angles from planar and linear fabrics in oriented drill core. These measurements were later converted to planes and lines using trigonometry. This method of data collection is very fast but often results in a measurement bias because low angles to the drill core axis are difficult to measure. Alpha angle distributions for fault, fracture, and vein measurements are under sampled at very low angles relative their expected normal distributions (Fig. 1.6). Bedding appears to be under sampled only at very high alpha angles. There is minimal drill hole orientation bias in the databases. Drill hole bias would appear on Figure 1.5 as wide zones of data exclusion at high angles to the most common drill hole orientation. Cross-Sections and Surface Geology Map A series of 6 intersecting cross-sections are presented (see Chapter 3). These cross-sections define the 3-dimensional geometry of the Donlin Creek deposit. Cross- sections were generated from drill hole data obtained from Barrick Gold Corp. using 9 Figure 1.4. Drill hole and trench location map for the Main Resource Area of the Donlin Creek Property (see Figure 1.3). Structural oriented drill holes are shown by large red circles, non-oriented drill hole locations are shown by small grey circles. Trenches are shown by black lines. 10 11 Geosoft Target v.7.0.1. Structural data (this study) have been overlain on the drilling data. For presentation purposes the cross-sections have been modified using ArcGIS v.9.2 and Adobe Illustrator CS2 v. 12.0. Where possible, surface bedding measurements have been added to the cross-sections in order to further validate the subsurface structural measurements. The location of each cross-section has been selected in order to: 1) incorporate the highest density of structural data; 2) show the true dips of bedding and bedding concordant intrusions and; 3) portray the true dip of the discordant intrusions. an interpreted surface geology map (see Chapter 3) was constructed using data and observations from the trench mapping campaigns and subsurface mapping, including previous studies (e.g., Szumigala, 2000; Ebert et al., 2000; Piekenbrock and Petsel, 2003) and 3-D models created by Westgold, Placer Dome, Nova Gold Resources Inc., and most recently Barrick Gold Corp. Lithologic and structural data is projected to the surface from depths ranging between 0 and 10 meters depending upon the depth of overburden, weathering and oxidation. The composition of sedimentary rocks is constrained from drill core intercepts, and is considered an approximation of the dominant lithology in that area. A limitation to this step is imposed by a need during exploration to simplify mixed sedimentary rocks by assigning the dominant rock to a particular interval. Nonetheless, the identified rock type is considered to be reasonably accurate. Data Conventions, Structural Analysis, Domain Analysis The following conventions apply to all structural data presented this study. Planar structural measurements are presented according to the strike azimuth/dip using the right hand rule convention (e.g., 300º/60º). Linear structural measurements are presented using the plunge toward trend convention (e.g., 60ºÆ300º). Stereograms use a Schmidt equal area stereonet. The lack of outcrop at Donlin Creek precludes the possibility of observing the relationships between structural fabrics in outcrop, and places an important limitation on this study. The structural data presented in this study includes measurements compiled from existing trench mapping and structural measurements compiled from structurally oriented drill core. Due to the very large number of structural 12 Figure 1.5. The distribution of drill hole orientations for the Donlin Creek gold deposit, (A) and the distribution of planar structural measurements (undifferentiated) plotted as they represent linear fabrics, (B). Figure 2.3A shows the distribution of drill holes (steep to moderately plunging points) and the distribution of trenches (sub-horizontal plunging lines). The lack of wide zones of data exclusion at high angles to the average drill hole orientation on Figure 2.3B indicates that there is not a significant drill hole orientation bias in this data. This plot also shows that there are no systematic alpha, beta and gamma angle measurement or conversion problems which commonly plot as conical distributions about the average drill hole. B A Avg. drill hole orientation 13 Figure 1.6. The frequency distribution of alpha angles for bedding, veins, faults and fractures compared to their expected normal distribution for the Donlin Creek gold Deposit, Southwest Alaska. These plots demonstrate that structures with very low and very high angles to the core axis are under sampled with the exception of bedding which is under sampled at very high angles to the core axis 14 Vein Alpha Angles vs Expected Normal Distribution 0 2 4 6 8 10 12 0 10 20 30 40 50 60 70 80 90 Alpha Angle (degrees) F r e q u e n c y % Bedding Alpha Angles vs Expected Normal Distribution 0 2 4 6 8 10 12 0 10 20 30 40 50 60 70 80 90 Alpha Angle (degrees) F r e q u e n c y % Fault Alpha Angles vs Expected Normal Distribution 0 2 4 6 8 10 12 0 10 20 30 40 50 60 70 80 90 Alpha Angle (degrees) F r e q u e n c y % Fracture Alpha Angles vs Expected Normal Distribution 0 2 4 6 8 10 12 0 10 20 30 40 50 60 70 80 90 Alpha Angle (degrees) F r e q u e n c y % 15 measurements for each structural fabric, and due to high degree of scatter common in datasets of this size, a statistical approach is used to determine the average orientation of a given fabric. This study assumes that the average orientation determined by the method outlined below is representative of the true average orientation of that fabric. The point counts presented in this chapter use a Gaussian counting method such that the counting circle area is equal to 1% of the area of the stereonet hemisphere, where the kurtosis, k, is equal to 100. This is similar to the more traditional fixed circle Schmidt counting method (Robin and Jowett, 1986). This method was chosen because the weight of a given sample diminishes with angular distance from the counting station. Practically, this reduces the influence of widely scattered data and produces smooth contours about individual point maxima. Contour plots are presented in multiples of the standard deviation (S) above the expected uniform distribution for the total number of data points (N) over the spherical area of the stereonet. The weighting function used in this method is the Fischer Function: w = exp[k(cos(t)-1)] where t is the angular distance and k is the kurtosis. The expected count is calculated: E=N/k. The variance is the integral (W*E)2/hemisphere area and the standard deviation is the square root of the variance (Robin and Jowett, 1986). The uniform distribution of historic prospects across the southern portion of the Donlin Creek property provided natural boundaries for identifying distinct structural domains (Fig. 1.4). Domain boundaries were identified where historic exploration drilling programs encountered significant gold mineralization or dramatic changes in the subsurface geology. Structural fabrics were subdivided in 3-dimensions according to the prospect location that they were measured in. These structures were then plotted on individual stereonets and overlain on each of the domains (see Chapter 3). This methodology provides an opportunity to examine how structural fabrics change across the Donlin Creek property and identifies zones that share the same subsurface geometry (See Chapter 3). Data Presentation Oriented drill core data is examined in plan and cross-section (see Chapter 3) in order to understand the subsurface geometry. Data collected from oriented diamond drill core is compared to data collected from historic surface trenching, (most trenches are now collapsed or reclaimed) as a means of quality control. Structural data is 16 presented on highly detailed geologic cross-sections; however it not possible to draw balanced cross-sections because continuous marker horizon(s) have not yet been identified. Northeast trending cross-sections are drawn perpendicular to the average bedding orientation and portray the true dip of the sedimentary and concordant igneous rocks. West-northwest trending cross sections are perpendicular to the average discordant contact orientation showing true dip of the igneous bodies. This is the first time this has been done for the Donlin Creek deposit. This study offers a series of scenarios explaining the evolution of Donlin Creek and places the deposit in the context of the tectonic evolution of southwest Alaska. Finally, this study presents geochronology data that constrains the timing of sedimentation of the host rocks; intrusion and gold mineralization (see Chapter 3). All samples, with the exception of an ash layer collected from the Lewis Road Gravel Pit, were obtained from drill core. This data places absolute age constraints on the deposition of the Kuskokwim Group sedimentary rocks, the age of intrusion of the voluminous quartz feldspar porphyry intrusions and finally, the age of gold deposition. This age data also broadly constrains the deformation history of Donlin Creek. 1.6 GEOLOGY OF THE KUSKOKWIM MOUNTAINS MINERAL BELT The Donlin Creek gold deposit is located in the Kuskokwim Mountains Mineral Belt of southwest Alaska (Fig 1.2). This northeast trending belt of precious metal enriched polymetallic mineral deposits is spatially associated with Late Cretaceous to early Tertiary subduction-related igneous complexes (Decker et al., 1994; Bundtzen and Miller, 1997; Nokleberg et al., 2000). Historic production from the Kuskokwim Mountains Mineral Belt consists of 3.22 Moz of gold, 412,000 oz of silver and 1,377,412 kg of mercury. Antimony and tungsten have also been produced. These commodities have been extracted from deposits scattered over an area approximately 550 km long and 350 km wide (Fig. 1.2) and are associated with four types of Late Cretaceous-early Tertiary igneous complexes: (1) alkali-calcic, comagmatic volcanic- plutonic complexes; (2) calc-alkaline, meta-aluminous reduced plutons; (3) peraluminous granite-porphyry sills and dykes; and (4) subaerial bimodal volcanic rocks (Bundtzen and Miller, 1997). Five main types of deposits are known: (1) plutonic- 17 hosted copper-gold polymetallic stockwork, skarn, and vein mineralization; (2) peraluminous granite porphyry-hosted gold polymetallic bodies; (3) plutonic-related, boron-enriched silver-tin polymetallic breccia pipes and replacement bodies; (4) gold and silver epithermal veins; and (5) gold, polymetallic heavy mineral placer deposits. Based upon 10 deposits genetically related to Late Cretaceous-early Tertiary intrusions, including the Donlin Creek deposit, the Kuskokwim Mountains Mineral Belt is estimated to contain a minimum inferred reserve of approximately 35.23 Moz gold, 6.46 Moz silver, 12,160 metric tons of tin and 28,088 metric tons of copper (Bundtzen and Miller, 1997). Miller et al., (2002) propose that polymetallic gold mineralization in the Kuskokwim Mineral Belt formed during three main time intervals, ~ 70, ~60 and ~ 30 Ma, each of which is coeval with a major period of strike slip faulting in southwest Alaska. Bundtzen and Miller (1997) suggest that lode deposits represent geologically and spatially related, vertically zoned hydrothermal systems that formed in response to Late Cretaceous to Early Tertiary plutonisim and volcanism generated by north-directed subduction of the Kula plate. Differences in the geology, structural style and metal tenor of the hydrothermal systems are thought to be due to the current outcrop level of exposure of the vertically zoned hydrothermal systems (Bundtzen and Miller, 1997). 1.7 TECTONIC EVOLUTION OF SOUTHWESTERN ALASKA The Kuskokwim Group is a backarc continental margin basin fill assemblage that formed during a period of continental margin-parallel arc formation throughout the circum-Pacific (Nokleberg et al, 2000). Initial sedimentation of the Kuskokwim Basin occurred shortly after the lithostratigraphic basement terranes of southwest Alaska reached their current position and configuration in Late Cretaceous (Albian) time (Decker et al, 1994). Sedimentation began immediately after a change from sinistral to dextral transpression, a response to a change in the obliquity of convergence between the Kula oceanic plate and the Cretaceous North American continental margin (Decker et al., 1994, Bundtzen and Miller, 1997; Nokleberg et al., 2000; Miller et al., 2002). Throughout the Late Cretaceous and early Tertiary, crustal thinning and extension about a northeast – southwest axis resulted in an elevated geothermal gradient which facilitated the development of the various hydrothermal systems preserved today 18 (Scholl et al., 1994; Redfield et al., 2007). Tectonic transport in southern Alaska has been accommodated along major transform boundaries including the Denali, Kaltag, and Nixon Fork continental scale fault zones (Bundtzen and Miller, 1997; Miller et al., 2002; Scholl et al., 1994). Ore bearing hydrothermal systems in the Kuskokwim Mountains are focused along secondary splays to major transform faults throughout southwestern Alaska (Bundtzen and Miller, 1997). Arc related igneous rocks of Late Cretaceous and earliest Tertiary age (45 to 80 Ma form a margin-parallel magmatic belt in the Kuskokwim Mountains and represent a swath of subduction related magmatism produced by subduction of the Kula-Farallon oceanic ridge between ~80 Ma to ~ 60 Ma (Nokleberg et al., 2000). Collectively these rocks are interpreted to represent remnants of the Kluane Arc that have undergone dextral-slip displacement since Late Cretaceous time (Nokleberg et al., 2000). Paleomagnetic data for the Paleocene (~ 60-56 Ma) show a counterclockwise rotation of the Pacific oceanic plate (Lonsdale, 1988). This rotation is interpreted to be related to compression between Eurasia and North America (Plafker and Berg, 1994) and is used to support an oroclinal bending model for central and southwestern Alaska during this time (e.g. Carey, 1955; Coe et al., 1989). Redfield et al., (2007) offer an alternative model to explain the Paleocene paleomagnetic data. They suggest that the configuration of present day southwest Alaska can be explained by escape tectonics. According to their model, terranes are forced northward by oblique subduction under the margin of western North America. These terranes move northward as a crustal raft or orogenic float along major faults until they encounter the rigid Brooks Range at the apex of the Alaska Orocline. The Brooks Range acts as a backstop preventing northward motion which results in terrane extrusion or tectonic escape westward toward the open face. By the early Miocene, the Kula plate was completely subducted (Nokleberg et al., 2000) and the leading edge of the Pacific Plate began to subduct (Nokleberg et al., 2000). Dextral transpression continued in southwest Alaska due to oblique convergence of the newly subducting Pacific plate and continues today (Nokelberg et al., 1999; Redfield et al., 2007). The Denali-Farewell and Iditarod-Nixon Fork faults are major northeast-trending faults that transect the Kuskokwim Mountains Mineral Belt (Fig. 1.1). These faults have 19 been shown to have experienced significant displacement since the Late Cretaceous (Miller et al., 2002). They show that the Iditarod-Nixon Fork fault has undergone a minimum of ~ 90 km right lateral offset. The Denali-Farewell fault segment in southwestern Alaska has a total dextral displacement of 134 km since Late Cretaceous (Albian) time. Steeply dipping faults are common within the Kuskokwim Mountains and typically form parallel and conjugate to these larger regional structures. Miller et al., (2002) divided Iditarod sub-basin into five fold domains based upon the orientation of fold axial traces defined by regional mapping (Fig. 1.7). These domains include: 1) East-trending fold axes in the northeast; 2) northeast-trending fold axes in the northwest between the Dishna River and Nixon Fork faults; 3) north-northeast- trending en echelon folds to the northwest adjacent to the Iditarod-Nixon Fork fault; 4) east-trending fold axes in the south-central part of the basin; and 5) north-trending fold axes in between splays in the Denali Fault (Miller et al., 2002). The timing of movement on these faults is established using fossil evidence and pre- and post-folding intrusions (Miller et al., 2002). In general the folds pre-date the emplacement of the volcanic-plutonic complexes. East trending folds in domain 1 are constrained to between ~ 91 Ma and 87 Ma using fossil evidence (Table 1.1). North-northeast trending folds in domain 2 contain an interbedded 77 Ma volcanic ash layer, indicating the age of the ash layer is a minimum age for folding (Table 1.1). North-Northeast trending folds in Domain 3 are between ~ 93 Ma and 89 Ma based upon fossil evidence (Table 1.1). Campanian to Maastrichtian (83.5-65 Ma) were recovered from east- trending folds in Domain 4. Finally, north-trending folds in Domain 5 are truncated by the Chuilnuk Pluton indicating that they are at least 68.5 Ma. (Table 1.1). 1.8 GEOLOGY OF THE KUSKOKWIM MOUNTAINS The Kuskokwim Mountains are underlain by the > 10 km thick Upper Cretaceous Kuskokwim Group that consists of coarse- to fine-grained turbiditic sandstone and shale with minor conglomerate (Decker et al., 1994). Minor interbedded andesitic tuff and flows are present near the top of the Kuskokwim Group (Miller and Bundtzen, 1994). Fossil evidence constrains the age of the Kuskokwim Group to Cenomanian to Campanian or Maastrichtian (~100 Ma to ~ 65 Ma) (Elder and Box, 1992; Miller et al., 20 Table 1.1. Folding in the Kuskokwim Basin. Orientation Domain Age Contsraint(s) Interpretation/Comments East 1 Strata contains Turonian to Early Coniacian (~91- 87 Ma) fossil ages N-S Shortening Northeast 2 77 Ma interbedded tuff. Unclear, fold axes trend are parallel to basin margin. North-Northeast 3 Folded strata contains Tournian (93.5-89 Ma bivalves Related to dextral displacement East 4 Youngest fossils recovered are Campanian to Maastrichtian (83.5-65 Ma) N-S Shortening - folds refolded about N-S axis. North 5 Folds truncated by the 68.5 Chuilnuk Pluton Related to dextral displacement 21 Figure 1.7. Fold patterns (axial traces) in the Kuskokwim Mountains Mineral Belt. Modified after Miller et al. (2002). 22 2002). An interbedded tuff near the top of the Kuskokwim Group yielded a 77 Ma biotite K/Ar age (Miller and Bundtzen, 1994). These data constrain the Kuskokwim Group to between about 95-77 Ma (Miller et al., 2002). Ash layers interbedded with the Kuskokwim sedimentary rocks at Donlin Creek yielded ages between 87.4 ±1.1 Ma and 88.9 ± 1.1 Ma (Chapter 3). The local host rocks at Donlin Creek represent the older parts of the Kuskokwim Group. The Iditarod-Nixon Fork and the Denali-Farewell fault systems divide the Kuskokwim Group into three geographic areas, including the central Kuskokwim basin, the Iditarod basin to the northwest and the Nushagak basin to the southeast (Decker et al., 1994) (Fig. 1.2). The Kuskokwim Group is interpreted to have been deposited in a northeast trending, strike slip basin that subsided between a series of amalgamated terranes (Decker et al., 1994). These terranes include Mesozoic oceanic arc assemblages dominated by marine volcanic and sedimentary rocks (Togiak and Goodnews terranes), a Paleozoic passive margin sequence (Nixon Fork Terrane) consisting of platform and deep water carbonate rocks and Proterozoic continental basement rocks consisting of amphibolite and metaplutonic rocks (Kilbuck terrane) (Decker et al., 1994). 1.8.1 Igneous History Igneous activity in the Kuskokwim Mountains began during the final stages of Late Cretaceous sedimentation in the Kuskokwim Basin and continued into the Tertiary (Miller and Bundtzen, 1994; Decker et al., 1994). Volcano-plutonic complexes in the Kuskokwim Basin are as large as 650 km2, form distinct topographic highs and intrude and overlie Kuskokwim Group sedimentary rocks (Miller and Bundtzen, 1994). Volcanic components of these complexes consist of intermediate tuffs and flows that range in age from 76 to 63 Ma (Bundtzen and Miller, 1994). Associated plutons are calc-alkaline in composition, range from monzonite to granodiorite and yield ages between 71 and 66 Ma (Decker et al., 1994; Miller and Bundtzen, 1994; Bundtzen and Miller, 1997). Subaerial volcanic tuffs, flows and domes are regionally extensive and dominantly andesitic, but locally include dacitic, rhyolitic and basaltic compositions and range in age from 71 Ma to 54 Ma (Miller et al., 2002). Although these rocks are similar 23 Figure 1.8. Donlin Creek igneous and sedimentary rocks, examples and textures. Mafic Dyke Fine Grained P h Crystalline P h Lathe Rich P h Aphanitic Flow Banded P h Channel Deposit Cross bedded quartz sands in h l Convolute Bedding in Siltstone Soft sediment deformation (greywacke and siltstone) Inoceremus Fossil in h lBioturbated mudstone 24 in age to the volcano-plutonic complexes and display broadly calc-alkaline trends, associated plutonic components have not yet been identified in outcrop (Miller et al., 2002). Late Cretaceous to Tertiary, hypabyssal, felsic to intermediate, aphanitic to porphyritic dykes, sills and stocks occur throughout the Kuskokwim region. These units are distinctly peraluminous in composition and commonly contain garnet phenocrysts indicative of crustal melts or extensive assimilation of crustal rocks (Miller and Bundtzen, 1994). Late Cretaceous to early Tertiary intermediate to mafic dykes constitute the fourth type of igneous activity in the Kuskokwim region. These bodies are typically less than 3 meters wide and are ubiquitously altered to chlorite, calcite and silica (Miller et al., 2002). 1.9 GEOLOGY OF THE DONLIN CREEK PROPERTY Donlin Creek is underlain by an 8.5 kilometer long, 2.5 kilometer wide quartz- feldspar phyric granite porphyry dyke and sill swarm hosted by lithic sandstone, siltstone and shale of the Late Cretaceous Kuskokwim Group (Fig. 1.3). In general, the stratigraphy strikes southeast and dips moderately (35º - 50º) toward the southwest. Dykes at Donlin Creek strike northeast and dip moderately to steeply southeast. Sills at Donlin Creek trend ~ 305ºand dip shallowly to moderately toward the south- southwest and north-northeast. 1.9.1 Kuskokwim Group Lithic Sandstone (Greywacke) Lithic sandstone varies from light to dark grey to black in colour. Weathered surfaces are commonly rusty. This unit is clast supported and is fine-grained to very coarse-grained. Grains are typically angular to subangular and are poorly to moderately well sorted. Individual clasts consist of metamorphic, igneous and sedimentary lithic fragments. Fossils are uncommon; however, inoceramus hash, brackish water pelecypods, and dicot leaf debris have been observed (Bundtzen, 2004). Lithic sandstones are commonly carbonaceous (locally carbonaceous), have a locally calcareous matrix and contain trace to 0.5 percent framboidal pyrite. Beds are generally upright; however, overturned beds are locally present. Greywacke is typically 25 massively bedded with individual beds ranging from centimeters to hundreds of meters in thickness. Soft sedimentary deformation structures including dewatering channels and convolute bedding are abundant. Common graded bedding, cross bedding, bioturbation and channel structures indicate that these rocks are upright (Fig. 1.8). Rare, very thinly bedded (1 to 5 cm), very well sorted, well rounded white quartz sands layers are present. Basal greywacke contacts are gradational with siltstone and mudstone. Thicker accumulations of lithic sandstone form resistant ridges. The greywacke at Donlin Creek contains sedimentary structures consistent with deposition in a basin margin sedimentary environment. Siltstone and mudstone horizons are very dark grey to black in colour. Weathered surfaces are similar in colour however, localized horizons of increased sulphide abundance locally give these rocks a rusty appearance. These rocks are typically upright; however, overturned beds are locally present. Siltstone is present in greater volumes than mudstone. Siltstone and mudstone intervals are highly carbonaceous, locally contain high concentrations of framboidal pyrite and commonly contain inoceramus hash, brackish water pelecypods, and dicot leaf debris (Bundtzen, 2004). Common soft sedimentary deformation structures consist of convolute bedding, dewatering channels, ripple marks and flame structures. The latter indicates that these rocks are upright. Interbedded ash layers are numerous within mudstone and siltstone horizons. Ash layers are typically light to medium dark grey in colour and range in thickness from millimeters to a few tens of centimeters. Contacts between the ash layers and their host sedimentary rocks are typically very sharp; however, rare instances of gradational contacts are preserved. Basal siltstone and mudstone contacts are typically very sharp. Sedimentary structures in the siltstones and mudstones at Donlin Creek are consistent with deposition in a basin margin sedimentary environment. 1.9.2 Igneous Rocks Igneous rocks at Donlin Creek form a bimodal dyke and sill swarm. Discordant igneous rocks (dykes) are generally a few tens of meters thick and maintain uniform thicknesses over hundreds of meters of strike length. Bedding concordant intrusions 26 (sills) are characterized by irregular bodies that pinch and swell and are highly discontinuous along strike. The igneous rocks are subdivided based upon composition and texture and include a mafic igneous phase and a quartz feldspar porphyry phase (Table 1.2). The quartz feldspar porphyries are subdivided into 5 sub-phases based upon the size, shape and distribution of phenocrysts as well as key characteristics of the groundmass (Piekenbrock and Petzel, 2002). These categories include, from oldest to youngest: mafic dykes, fine-grained porphyry, crystalline (crowded) porphyry, aphanitic porphyry, lath-rich porphyry, and blue porphyry based upon cross-cutting relations (Table 1.2). Locally, however, each of the igneous phases cross-cut one another. The crystalline porphyry and the fine-grain porphyry are the most common igneous units on the property (Piekenbrock and Petzel, 2002). 1.9.3 Timing of Igneous Activity Bundtzen and Miller, (1994) reported K/Ar dates between 70.9 ± 2.1 and 65.1 ± 2.0 Ma for magmatic biotite and muscovite phenocrysts collected from 3 unaltered dykes located a few kilometers east and west of the Donlin Creek deposit. Szumigala et al., (1999) reported 40Ar/39Ar ages from biotite in felsic and mafic igneous rocks from the Queen prospect. These returned biotite plateau ages of 70.3 ± 0.2 Ma for a felsic porphyry dyke and 72.6 ± 0.9 Ma for a mafic dyke sample (Szumigala et al., 2000). Szumigala et al., (2000) also reported a whole rock age for a mafic dyke of 74.4 ± 0.8 Ma. Ebert et al., (2003a) reported a U-Pb age of 69.2 ± 0.5 from a felsic dyke in the Lewis Prospect. They also obtained a U-Pb age of 69.9 ± 0.2 for an unaltered quartz feldspar porphyry north of the Dome Prospect (Ophir Prospect). Goldfarb et al., (2004) obtained a U-Pb age of 66.5 ± 0.5 Ma for this same unit. U-Pb ages obtained from quartz feldspar porphyry intrusions in the North and South Lewis prospects (this study) returned ages of 71.7 ± 0.7 Ma and 71.4 ± 0.7 Ma (see Chapter 2). 1.9.4 Structural Setting of Donlin Creek Donlin Creek is situated between the Holitna segment of the Denali Fault to the south and the Iditarod-Nixon Fork Fault system to the north (Figs. 1.1 and 1.2). The Holitna segment of the Denali Fault has ~134 km of dextral offset since the Late 27 Table 1.2. The characteristics of igneous rocks at Donlin Creek. The igneous rocks are subdivided based upon composition, texture and key characteristics of the ground mass. Rock Phenocryst Phenocryst Distinguishing Name Code Composition Colour Description Size Groundmass Features Blue Porphyry (Youngest) RDXB Alkali Granite to Granite Grey-blue Inequigranular, porphyritic, holocrystalline, coarse to medium grained quartz, feldspar and minor biotite. 2-10 millimeters Fine grained, idiomorphic, euhedral quartz, feldspar and biotite with clots and disseminations of graphite and pyrite. Characteristic blue-grey colour attributed to high concentrations of disseminations and clots of graphite and sulphides in the groundmass and in feldspar phenocrysts. Aphanitic Flow Banded Porphyry RDA Alkali Granite to Granite White to dark grey Inequigranular, porphyritic, holocrystalline medium grained to aphanitic 1-3 millimeters Fine grained, idiomorphic, microgranular quartz and feldspar. Inhomogeneous with respect to phenocryst distribution. Salt and pepper colour, well developed chilled margins that are commonly flow banded.. Lathe Rich Porphyry RDXL Alkali Granite to Granite tan to dark grey Inequigranular, porphyritic, holocrystalline, quartz feldspar with minor biotite 1-4 millimeters and 5-10 millimeters. fine grained, idiomorphic, euhedral quartz, feldspar and biotite Phenocrysts are characteristically tabular with a bimodal size distribution. Crystalline (Crowded Porphyry) RDX Alkali Granite to Granite light grey to dark grey Inequigranular, porphyritic, holocrystalline, quartz feldspar with minor biotite 1-10 millimeters Phaneritic to aphanitic Phenocrysts are densely packed or crowded. Volumetrically most abundant phase. Occasional chilled margins. Fine Grained Porphyry RDF Alkali Granite to Granite white to dark grey Inequigranular, porphyritic, holocrystalline, quartz feldspar with minor biotite 1-5 millimeters idiomorphic, euhedral, fine-grained Earliest phase, commonly has well developed chilled margins. Mafic Intrusive Rocks (Oldest) MD Andesite to Quartz Andesite Grey to green Inequigranular, porphyritic 1-3 millimeters medium grained to aphanitic, idiomorphic, euhedral quartz, feldspar and biotite. Oldest intrusive phase on the property, thin, typically < 3 meters in thickness and highly discontinuous. 28 Cretaceous (Miller et al., 2002). Miller et al. (2002) constrained movement on the Nixon Fork fault to >90 km of right lateral displacement since about 58 Ma but suggested, based upon displacement of Cretaceous conglomerate, that this offset may be as old as ~ 90 Ma. Donlin Creek lies within a structurally complex area located ~25 km southeast of the Iditarod-Nixon Fork Fault (Fig 1.2). Proximal to regional faults in this area are north trending en echelon folds, interpreted to be related to regional dextral strike-slip faulting (Miller et al., 2002). Away from the large faults east-trending folds are characteristic; these do not have an obvious connection to regional strike-slip tectonics (Fig. 1.7). The Donlin Creek fault marks the boundary between these two contrasting fold domains in this part of the Kuskokwim Mineral Belt (Figs. 1.2, 1.4 and 1.7). This fault, like many of the larger faults in the Kuskokwim Mountains, is a northeast trending fault situated just to the north of the Donlin Creek deposit. Previous studies (e.g. Szumigala et al., 1999; Miller et al., 2000; Piekenbrock and Petsel, 2003; Ebert et al., 2003a, b) have outlined a general structural history for the Donlin Creek property. This work is summarized below. Approximately east-trending folds are the oldest recognized structures. These folds lack a penetrative axial planar cleavage, and are associated with north-vergent thrust faults. Thrust faults are generally bedding plane parallel and are responsible for the dominant southwest dips (30-50º) observed in the Kuskokwim Group stratigraphy (O’Dea and Bartch, 1997; Piekenbrock and Petsel, 2003). At approximately 70 Ma, multiple north-northeast- striking discordant intrusions and west-northwest-trending bedding concordant intrusions are emplaced into the Kuskokwim Group (O’Dea and Bartch, 1997). Northeast- and northwest-striking high angle faults cut the intrusions, Kuskokwim Group host sedimentary rocks, and thrust faults in the Lewis and ACMA zones. Elsewhere, low amplitude north-northeast-trending folds affect the Kuskokwim Group. These only locally crop out, and their district scale significance is unknown (see Chapter 3). No cleavage or other evidence of dynamic recrystallization is present, suggesting relatively shallow depth of formation for the folds and thrust faults. Gold-bearing, north-northeast-striking, high-angle extensional fractures crosscut all igneous rocks and all generations of faults on the property (O’Dea and Bartch, 1997; Ebert et al., 2002a). The veins are most common in the brittle igneous units but are 29 also found in the coarse -grained lithic sandstones (O’Dea and Bartch, 1997; Goldfarb et al., 2004) (see Chapter 3). Ebert et al., (2003b) documented evidence for left and right-lateral strike-slip, oblique-slip and dip-slip separation on this youngest extensional event. 1.9.5 Mineralization Gold mineralization occurs throughout the entire 8.5 kilometer long Donlin Creek dyke and sill swarm but is best developed in the within the Aurora, 400, Vortex, Akivik, ACMA, South Lewis, Lewis and Dome prospects (Fig. 1.3). Mineralization at Donlin Creek is divided into two styles (Table 1.3). These include an early Dome-Duqum style gold mineralization and a later ACMA-Lewis style of gold mineralization (Table 1.3). Dome-Duqum style mineralization occurs in small zones and consists of early quartz veins with both sheeted and stockwork morphologies hosted in the intrusive phases and in hornfelsed sedimentary rocks (Szumigala et al., 1999; Ebert et al., 2002b). Quartz veins contain minor arsenopyrite, pyrite, chalcopyrite and pyrrhotite ± sphalerite and have an Au, Ag, Cu, Zn, Bi ± Te trace element signature. Dome-Duqum style mineralization contains free gold in the form of electrum (Ebert et al., 2003b). The younger ACMA-Lewis style gold mineralization occurs as vein and disseminated style gold mineralization. The published resource estimates for Donlin Creek refer exclusively to ACMA-Lewis style gold mineralization. This style of gold mineralization consists of Au-As-Sb-Hg bearing quartz-ankerite-dolomite veins with minor pyrite, arsenopyrite ± stibnite. Alteration halos associated with veins contain lower grade gold mineralization. Gold closely correlate with arsenopyrite content. Individual alteration halos are typically small (millimeters to meters scale) but become much wider where vein density is sufficient for alterations halos to link together to form broad zones of continuous gold mineralization. ACMA-Lewis style mineralization is refractory, with gold contained within arsenopyrite and to a lesser degree pyrite (McCoy et al., 1998; Goldfarb et al., 2004). 1.9.6 Alteration Alteration is widespread across the Donlin Creek property. In general, the felsic igneous rocks have been affected by varying intensities of illite, ± kaolinite, quartz and 30 Table 1.3. Key characteristics of Donlin Creek gold mineralization. Dome Duqum 2.17 ppm 0.63 Au, Ag, Cu, Zn, Bi Arsenopyrite, pyrite, chalcopyrite, pyrrhotite, Fe- rich sphalerite, trace electrum, native bismuth, bismuth tellurides and selenides. Sheeted veins and stockworks Quartz Ribbon veins, sulphides occur in disseminations and clots in veins, vein margins and occasionally replace mafic sites. intense silicification, abundant quartz veins and contain trace to 3 percent sulphides mainly arsenopyrite, pyrite, and chalcopyrite 3.1 to 14.6 wt percent CO2, salinities up to 39 wt% 235 - 450ºC 1.5 - 3 Km ACMA-LEWIS 3.00 1.18 Au, Ag, As, Sb, Hg Pyrite, arsenopyrite, stibnite, native arsenic, marcasite, realgar orpiment, cinnibar Sheeted vein swarms and disseminated haloes around veins. Quartz +/- carbonate suggary veins, open space filling, vuggy silica, vuggy carbonate, bladed carbonate. illite, sericite ± kaolinite ± chlorite/smectite plus quartz- carbonate-pyrite 2.2 to 6.7 wt percent CO2, have an average of 2.5 wt percent NaCl equiv. and formed at 1570 +/- 400 meters depth 130 - 300ºC 1-2 Km (lithostatic) Depth ConstraintsTextures Alteration Ore Fluid Temperature Constraints Mineralization Type Avg. Grade Au:Ag Geochem. Signature Ore and Related Minerals Morphology Vein Gangue 31 pyrite. Two separate alteration systems have been recognized. An early, high temperature event is best developed at the Dome and Duqum prospects (Dome- Duqum style) and a later (ACMA-Lewis style) lower temperature event associated with the main gold resource (Ebert et al., 2003b) (Table 1.3). Figure 1.9 shows the simplified distribution of clay alteration mineralogy across the Donlin Creek property. ACMA-Lewis Style Alteration ACMA-Lewis style alteration is associated with veinlets which occur across the property and are associated with both mineralized and unmineralized rocks. Gold mineralized rocks contain quartz-carbonate-pyrite-arsenopyrite veinlets and generally correspond with zones of intense hydrothermal illite alteration (Ebert et al., 2003b). Barren 1-2 mm wide pyrite veins cut and are themselves locally cut by mineralized veinlets. Barren white and clear quartz veinlets and white calcite-ankerite veinlets with no sulphides post-date mineralized veins (Ebert et al., 2003b). Pyrite is typically widespread in all altered rocks (0.5 to 2%) but generally more abundant (1-4%) in mineralized zones (Ebert et al., 2003b). Porphyritic felsic intrusive rocks in and around mineralized zones are pervasively altered to illite, sericite ± kaolinite ± chlorite/smectite plus quartz-carbonate-pyrite (Ebert et al., 2003b). Illite is much more common than sericite in all altered zones. In contrast, the Kuskokwim Group sedimentary rocks are not susceptible to pervasive alteration. Instead, a similar alteration assemblage in these sedimentary rocks is restricted to fractures adjacent to strongly altered igneous units. Feldspar phenocrysts within the felsic igneous rocks are strongly altered to illite ± kaolinite. Biotite phenocrysts are altered to sericite, and also commonly partially replaced by pyrite ± arsenopyrite. The fine grained matrix of the porphyry is typically altered to illite ± sericite, quartz, carbonate, pyrite ± very fine grained graphite (Ebert et al., 2003b). It is possible that the fine grained graphite present in this system, particularly in the Blue Porphyry intrusions, is a relict mineral phase and therefore may not be related to alteration by the hydrothermal system. A second lower temperature alteration style is detectible peripheral to the mineralized zones. This alteration consists of chlorite/smectite with lesser kaolinite, illite and quartz. Ebert et al., (2003b) interpret this to be a product of a laterally decreasing temperature away from the most intense parts of the hydrothermal system, 32 Figure 1.9. Simplified clay alteration at Donlin Creek modified from Ebert et al. (2003b). The blue dotted line represents the approximate distribution of quartz veining and moderate to strong illite clay alteration. 33 suggesting that clay alteration is directly associated with the mineralizing event (Fig. 1.9). Gold is locally strongly developed in areas with illite only and illite + kaolinite alteration (Ebert et al., 2003b). Dome-Duqum Style Alteration Dome-Duqum zones are characterized by intense silicification, abundant quartz veins and contain trace to 3 percent sulphides mainly arsenopyrite, pyrite, and chalcopyrite (Ebert et al., 2003b). Ebert et al. (2003b) document two alteration events including, (1) an older alteration pattern associated with Dome-Duqum veins that is now largely destroyed; and (2) a late low-temperature clay alteration overprint. Alteration surrounding Dome- Duqum style veins within the felsic porphyry intrusions currently consists of illite with minor phengite, monmorillonite, kaolinte, interlayered chlorite/smectite, calcite, pyrite, minor remnant chlorite, and minor apatite (Ebert et al., 2003b). Calcite is the dominant carbonate mineral within the Dome-Duqum areas in contrast to the ACMA-Lewis alteration where pure carbonate phases are rare and Fe, Mg, and Mn bearing carbonate phase dominate (Ebert et al., 2003b). Fluid inclusion studies on samples from Donlin Creek have been done by Dunne (1993), Szumigala et al., (1999), Ebert et al., (2003b) and Goldfarb et al., (2004) support the observation that there were two distinct overprinting hydrothermal systems active at Donlin Creek. Dunne (1993) examined 4 samples obtained from the Lewis and Far Side prospects. Fluid inclusions in this material yielded homogenization temperatures of 210º to 260ºC and contained multiple solid daughter phases (Dunne, 1993). Szumigala et al. (1999) summarized internal company studies by Roberts, (1993) and Reynolds, (1996). Secondary fluid inclusions examined from igneous quartz phenocrysts in altered intrusive rocks from the Rochelieu, Queen, South Lewis, Lewis, and Dome prospects. With the exception of the Dome Prospect, all fluid inclusions were reported as gold related, dilute and trapped between ~ 275 and 300 ºC. Dome inclusions were estimated to have been trapped at temperatures between 400 to 450 ºC, are hypersaline and are believed to have been trapped close to a magmatic source. Ebert et al., (2003c) collected 16 mineralized samples from the ACMA, Lewis, Queen, Snow, Far Side, Placer Pit and Dome prospects. Ebert et al. (2003b) conclude that ACMA-Lewis veins have homogenization temperatures between 130º and 240ºC, 34 contain 2.2 to 6.7 wt percent CO2, have an average of 2.5 wt percent NaCl equiv. and formed at 1570 ± 400 meters depth. Dome-Duqum inclusions yield homogenization temperatures that range between 260º and 410ºC, and have between 3.1 to 14.6 wt percent CO2. Ebert et al. (2003b) report that Dome veins formed at depths between 1500 and 3000 meters. Goldfarb et al. (2004) selected 20 samples for fluid inclusion work from the South Lewis, Lewis, Queen and Dome prospects based upon high gold grades and/or ore-related sulphide mineralogy. They concluded that homogenization temperatures between 275º and 300ºC were representative of minimum trapping temperatures for the fluids that deposited gold-rich arsenopyrite at Donlin Creek (ACMA-Lewis style mineralization). Trapping pressures of the ore-forming minerals were estimated by Goldfarb et al. (2004) to be between 1 and 2 kilometers, assuming lithostatic pressures. The Dome prospect yielded homogenization temperatures between 235º and 450 ºC based upon vapor rich inclusions that contained equal amounts of CO2 and CH4. Taken together these data indicate that older Dome-Duqum mineralization and alteration formed between 1.5 and 3 kilometers depth at temperatures between 235 ºC and 450 ºC in a high CO2, highly saline environment. Lewis-ACMA style veins and alteration formed between 1 and 2 kilometers depth at temperatures between 130 ºC and 300 ºC under relatively low salinity, low CO2 conditions. 1.10 Classification of the Donlin Creek Gold Deposit There is no unique genetic classification for the Donlin Creek gold deposit. It has been classified as granite porphyry-hosted gold polymetallic (Bundtzen and Miller, 1997), orogenic or intrusion related (Goldfarb et al., 2004), low-sulphidation epithermal (Ebert et al., 2003a), reduced porphyry to sub-epithermal Au-As-Sb-Hg (Ebert et al., 2003a), and distal or high level epizonal intrusion related (Hart et al., 2002). Table 1.4 contains a list of deposit characteristics for Donlin Creek and a list of key characteristics for each of the gold deposit models that have been proposed for the deposit. The ore forming fluids at Donlin Creek were relatively low salinity, CO2 rich and were likely produced by devolatilization of sedimentary rock-dominated terranes at 35 Table 1.4. Summarized key characteristics of the Donlin Creek gold deposit and summary of key characteristics of the ore deposit models that have been proposed for it. 36 Key Characteristics Donlin Creek Low Sulphidation Epithermal Orogenic Intrusion Related Alkaline Porphyry Au related Age ~ 70 Ma Most are Cretaceous or younger. Archean and phanerozoic. Three prolific time periods: > 3.0 Ga, Ca. 2.8-2.55 and 2.1-1.8 Ga Mostly Phanerozoic but a few proterozoic and Archean examples exist. Precambrain and younger however most Cretaceous and younger. Tectonic Setting Compressional Continental Margin Back Arc Basin undergoing active strike slip Convergent plate margins, extensional zones (including rifts) within magmatic arcs. Compressional: accretionary orogens with strong association with felsic magmas. Accretionary to collisional/post collisional: Back arcs, foreland fold and thrust, and magmatic arcs. Magmatic belts in continental tectonic sedding well inboard of the subduction zone and hosted by reduced metasedimentary rocks. Magmatic arcs at convergent plate margins with some component of strike slip (trench advance and trench retreat). Structural Setting Ore is focused in low order dilational jogs realted to regional dextral strike slip motion on larger faults. Form in all structural settings but are most common in strike-slip faulting, transtensional and extensional stress regiems that are regional in scale. Focused along second or third order structures with spatial relation to large-scale trans-continental structures Reginal and local stress regiems direct fluid flow. No unique structural setting required. Local areas of extension in shallow-level plutons near or along a major fault zone. Structural Control Vein distribution is controlled by rheology. Disseminated mineralization controlled by permeability barriers (lithology). Au bearing, NE trending (020/70) extensional quartz and quartz carbonate veins Combination of lithologic (rheology and permeability). Fluid pathways from: dilatent zones, fault intersections, flexures and fault jogs in strike slip settings, tensional veins and sheeted veins. Normal and listric faults in extensional environments and reverse faults in compressional settings. Brittle faults to ductile shear zones, reverse to strike slip to oblique slip motion, fracture arrays, stockwork vein, foliated zones, fold hinges. Sheeted veins, stockwork veins, brittle and ductile structures are common (not necessarily in the same deposit. Ductile and/or brittle sheeted veins, concentric and radial mineralized fractures, stockworks, breccias. Favourable Host Rocks Porphyritic alkali granite to granite, andesite to quartz andesites and lesser greywacke, shale and siltstone. Dominantly associated with calc-alkalic or alkalic volcanic rocks. Andesite-rhyodacite, biomadal rhyolite-basalt, Alkalic rocks. Strong association between gold and greenstone belts. Spatial association of felsic magmas and marine rocks. metalluminous sub alkalic intrusions of intermediate to felsic compositions that lie near the boundary between illmenite and magnetite series. Regional magmatic suites that may include I- and S-types, transitional I- and S-types, and subalkalic (to locally alkalic), metaaluminous (to locally peraluminous) reduced intrusions. Exclusively in I-type magnetite series intrusive suites. From in calc-alkaline to alkaline rocks phenocryst assemblages consisting of one or more of the minerals quartz, K feldspar, plagioclase, hornblende, biotite (rarely pyroxene and olivine) in a fine grained matrix. Ore related minerals Pyrite, arsenopyrite, stibnite, native arsnic, realgar Cinnabar, stibnite pyrite/marcasite, arsenopyrite, Au-Ag selenides, Se sulphosalts, pyrrhotite, Fe-rich sphalerite, galena, chalcopyrite, tetrahedrite/tennanite, enargite/luzonite, covellite Pyrite, arsenopyrite, sphalerite, chalcopyrite, pyrrhotite Low sulphide content (< 5 Vol%), reduced mineral assemblages including arsenopyrite, pyrite, pyrrhotite. Hematite and magnetite are absent. Chalcopyrite, pyrite (potassic zones), rare molybdenite Organic Material Yes Yes Yes Yes No Vein Mineralogy Quartz +/- carbonate + sulphides Quartz, quartz carbonate Quartz, Quartz+Carbonate+/-sericite Quartz, quartz carbonate, typically low sulphide content. Quartz, feldspar, tourmaline, magnetite, hemetite. Alteration Mineralogy Illite, sericite, kaolinite Clays: illite, smectite, kaolinite +/- adularia Fe sulphide-carbonate-sericite +/- albite Fracture controlled with rare pervasive alteration including albite or K-feldspar, sericitic, silicic, greisens, calc-silicate, and/or advanced argillic. Gold rich alkaline porphyry alteration consists of potassic with peripheral propylitic but can include early calcic-sodic or variably developed hydrolitic and/or advanced argillic, especially in lithocaps when preserved. Geochemical Signature Ag-As-Au-Hg-Sb Au, Ag, As, Sb, Hg Au, As, Sb, Hg, W, Mo (Bi, Te) Au with elevated Bi, W, As, Mo, and/or Sb. Proximal and distal intrusion related deposits include: Sb and Hg., Au, Cu, Mo, W, Temperature/Depth 130-300 ºC for ACMA-Lewis mineralization, 1-2 Km 180-280ºC, 100 meters to between 800 and 1500 meters, metamorphism is not considered an important factor. 300 +/- 50 at 1-3 kbars, 2-20 km depth syn to post peak metamorphism <200 to 600 degrees at 0.5 to 3 Kbars. High temperatures > 500ºC at 1-3 km from the surface. Fluid Chemistry Main resource: 2.2 to 6.7 wt percent CO2, have an average of 2.5 wt percent NaCl equiv. Near neutral pH, dilute. Systems dominated by meteoric waters with significant magmatic component. NaCl, CO2, H2S are principal species. Low salinity mixed aqueous-carbonic fluid with low to moderate CO2 Low temperature near solidis carbonic hydrothermal hydrothermal fluids. Hydrous , CO2 rich relatively sulphur rich fluids with variable salinity. References: Szumigala et al. (2000); Ebert et al. (2003a, b); Goldfarb et al. (2004) Corbett, 2002; Cooke and Simmons, 2000; Hedenquist et al. 2000 Grooves et al. 1998, Goldfarb et al. 2001 Lang and Baker, 2001. Tosdal and Richards, (2001); Sillitoe, 2002; Jensen and Barton, 2000, Hitzman, 2003. 37 depth or represent a gas-rich magmatic fluid (Ebert et al., 2003a; Goldfarb et al., 2004). The features that Donlin Creek shares with orogenic gold deposits are as follows. The deposit formed in a collisional margin during a period of active strike slip tectonics. The deposit has an associated placer gold deposit, a Au/Ag ratio of ~ 1, a Ag-As-Au-Sb-Hg geochemical signature, contains organic material, and is associated with a sericite- carbonate-pyrite alteration assemblage (Goldfarb et al., 2004). Furthermore, ore fluids contain significant quantities of CO2 and isotopically heavy oxygen (Ebert et al., 2000; Goldfarb et al., 2004). The range of temperatures (130-300 ºC) and depths of ore formation at Donlin Creek (1-2 Km) are consistent with those of intrusion related gold deposits (Goldfarb et al., 2004). Mineralization at Dome is associated with Bi and Te, elements commonly associated with intrusion related gold deposits (e.g., Baker and Lang, 2001). Ebert et al., (2003b) show that ore forming fluids at the Dome-Duqum and ACMA-Lewis areas contain similar ranges of sulphur, lead and carbon isotopes. Based upon this data they propose that these styles of mineralization are genetically related. Similar fluid inclusions found in these two zones support this interpretation. A genetic link between the Dome-Duqum and ACMA-Lewis styles of mineralization supports an intrusion related gold model for this deposit. According to this model, hotter, CO2 rich Bi, Te bearing ore fluids at Dome-Duqum would have formed proximal to a buried causative intrusion. The cooler, As, Sb, Hg bearing fluids found at ACMA-Lewis formed distally to the causative intrusion and therefore be part of the proximal intrusion related deposit sub-class (e.g., Baker and Lang, 2001). The strongly developed hornfels at Dome supports an intrusion proximal interpretation for Dome-Duqum style mineralization. Continuous quartz and quartz+carbonated vein hosted gold mineralization cuts the intrusive rocks at Donlin Creek suggesting that the intrusions were cool enough to accommodate brittle fractures at the time of mineralization. This may indicate that the intrusions themselves are not the magmatic source for the gold ore fluids. However, a potential causative intrusion is not exposed at Donlin Creek, making it difficult to prove conclusively that Donlin fits the intrusion related gold model. Donlin Creek most closely fits a low sulphidation epithermal gold deposit model. The age, tectonic setting, and structural style are each permissive for such a model. Donlin Creek contains quartz ± carbonate sulphide veins, contain typical epithermal 38 textures including banding, open space filling and bladed carbonate, a feature that is commonly sited as evidence for low temperature boiling in epithermal gold deposits. Donlin Creek contains a typical epithermal geochemical signature (Au-As-Sb-Hg) and contains a pyrite-arsenopyrite-stibinte-native arsenic-realgar sulphide assemblage. Donlin Creek fits into the alkalic-type sub-class of epithermal deposits defined by Richards (1995). This epithermal deposit model can form deeper (up to 2 km depth) and ore related fluids typically have a larger magmatic fluid component and an elevated CO2 content (Richards, 1995). This deposit class contains significant breccias and includes fluorite and vanadium-micas such as roscoelite which are notably absent at Donlin Creek. On major difference between alkalic-type epithermal deposits is that they typically form from oxidized magmas (e.g., Jenson, 1995; Barton, 2000), whereas the magmas associated with Donlin Creek are strongly reduced. Ebert et al. (2003b) observed that Donlin Creek shares characteristics with low sulphidation epithermal gold deposits but proposed that Donlin Creek be classified as a reduced sub-epithermal or reduced epithermal gold deposit based upon its association with reduced magmas. 39 REFERENCES Bundtzen, T.K., 2004, Assessment of calcium carbonate resource potential near the Donlin Creek Project, Iditarod A-5 quadrangle, southwest Alaska: Internal technical report prepared by Pacific Rim Consultants Inc. Bundtzen, T.K. and Miller, M.L., 1997, Precious metals associated with Late Cretaceous-early Tertiary igneous rocks of southwestern Alaska. In Goldfarb, R.J., and Miller, M.L., Eds. Mineral Deposits of Alaska. Economic Geology, Monograph 9, p. 242-286. Bundtzen, T.K., Miller, M.L., Laird, G.M., and Bull, K.E., 1992, Geology and mineral resources of the Iditarod Mining District, Iditarod B-4 and eastern B-5 quadrangles, southwest Alaska: Alaska Division of Geological and Geophysical Surveys, Professional Report 97, 46p., 2 plates. Carey, S.W., 1955, The orocline concept in geotectonics: Part 1. Papers and Proceedings of the Royal Society of Tasmania, v. 89, p. 255-288. Coe, R.S., Globerman, B.R., Plumley, P.R., and Thrupp, G.A., 1989, Rotation of central and southern Alaska in the early Tertiary: oroclinal bending by megakinking? In Kissel, C., and Laj, C., eds. Paleomagnetic Rotations and Continental Deformation. NATO-ASI Series. Boston, Kluwer Academic, p. 327-339. Cooke, D.R., and Simmons, S.F., 2000, Characteristics and genesis of epithermal gold deposits: Reviews in Economic Geology, v. 13, p. 221-244. Corbet, G., 2002, Epithermal Gold for Explorationists: Aig Journal – Applied geoscientific practice and research, paper 2002-01 p. 1-26. Decker, J., Bergman, S.C., Blodgett, R.B., Box, S.E., Bundtzen, T.K., Clough, J.G., Coonrad, W.L., Gilbert, W.G., Miller M.L., Murphy, J.M., Robinson M.S., and Wallace W.K., 1994, Geology of southwestern Alaska, in Plafker, G., and Berg, H.C., eds. The Geology of Alaska. Geological Society of America. DNAG Series. G-1. p. 285-310. Ebert S., Tosdal, R., Goldfarb, R, Dodd, S., Petsel, S., Mortensen, J., and Gabites, J., 2003a, The 25 million once Donlin Creek Gold Deposit, Southwest Alaska: a possible Link between reduced porphyry Au and sub-epithermal Au-As- Sb-Hg mineralization in Regional Geologic Framework and deposit specific exploration models for intrusion- related gold mineralization, Yukon and Alaska. [CD-ROM]. Edited by Shane Ebert (2003): MDRU Special Publication 3. The Mineral Deposit Research Group, University of British Columbia. Ebert, S., Dodd, S., Miller, L., and Petsel, S., 2003b, The Donlin Creek Au-As- Sb-Hg deposit, southwestern Alaska Ebert, S., Baker, T., and Spenser, R.J., 2003b, Fluid inclusion studies at the Donlin Creek gold deposit, Alaska, possible evidence for reduced porphyry-Au to sub-epithermal transition mineralization in Regional Geologic Framework and deposit specific exploration models for intrusion-related gold 40 mineralization, Yukon and Alaska. [CD-ROM]. Edited by Shane Ebert (2003): MDRU Special Publication 3. The Mineral Deposit Research Group, University of British Columbia. Ebert, S., Baker, T., and Spenser, R.J., 2003c, Fluid inclusion studies at the Donlin Creek gold deposit, Alaska, possible evidence for reduced porphyry-Au to sub- epithermal transition mineralization: in Regional Geologic Framework and Deposit Specific Exploration Models for Intrusion-Related Gold Mineralization, Yukon and Alaska. [CD-ROM]. Edited by Shane Ebert (2003): MDRU Special Publication 3. The Mineral Deposit Research Group, University of British Columbia, p. Ebert, S., Miller, L., Petsel, S., Kowalczyk, P., Tucker, T.L., and Smith, M.T., 2000 Geology, mineralization, and exploration at the Donlin Creek project, southwestern Alaska: British Columbia and Yukon Chamber of Mines Special Volume 2, p. 99-114. Engebretson, D.C., 1987. Reconstructions, plate interactions, and trajectories of oceanic and continental plates in the Pacific Basin, in Monger J.W.H and Francheteau, J., eds. Circum-Pacific Orogenic Belts and Evolution of the Pacific Ocean Basin. Geodynamics Series Volume 18, International Lithosphere Program Contribution, p. 19-27. Elder, W.P., and Box, S.E., 1992, Late Cretaceous inoceramid bivalves of the Kuskokwim Basin, Southwestern Alaska, and their implications for basin evolution. Memoir of the Paleontological Society, Vol. 26, Supplement to Vol. 66 no. 2 of the Journal of Paleontology, p. 1-39. Goldfarb, R.J., Ayuso, R., Miller, M.L., Ebert, S.W., Marsh, E.E., Petsel, S.A., Miller, L.D., Bradley, D.B., Johnson, C., and McClelland, W., 2004, The Late Cretaceous Donlin Creek gold deposit, southwestern Alaska: Controls on epizonal ore formation. Economic Geology Vol. 99, pp. 643-647. Goldfarb, R.J., Groves, D.I., and Gardoll, S., 2001, Orogenic gold and geologic time—a global synthesis: Ore Geology Reviews, v. 18, p. 1–75. Groves, D.I., Goldfarb, R.J., Gebre-Mariam, M., Hagemann, S.G., and Robert, F., 1998, Orogenic gold deposits: A proposed classification in the context of their crustal distribution and relationship to other gold deposit types: Ore Geology Reviews, v. 13, p. 7–27. Haeussler, P.J., Bradley, D.C., Wells, R.E., and Miller, M.L., Life and death of the Resurrection Plate: evidence for its existence and subduction in Paleocene-Eocene time: Geological Society of America Bulletin, vol. 15, p.867-880. Hart, C.J.R., McCoy, D.T., Goldfarb, R.J., Smith, M., Robert, P., Hulstein,R., Bakke, A.A., and Bundtzen, T.K., 2002, Geology, exploration, and discovery in the Tintina gold province, Alaska and Yukon: Society of Economic Geologists Special Publication 9, p. 241–274. 41 Jenson, E.P., and Barton, M.D., 2000, Gold deposits related to alkaline magmatism: Reviews in Economic Geology, v. 13, p. 279-314. Lang, J.R., and Baker, T., 2001, Intrusion-related gold systems: Then present level of understanding: Mineralium Deposita, v. 36, p. 477–489. Lonsdale, P., 1988, Paleogene history of the Kula plate—off- shore evidence and onshore implications: Geological Society of America Bulletin, v. 100, p. 755-766. Miller, M.L., and Bundtzen T.K., 1994, Generalized geologic map of the Iditarod quadrangle, Alaska, showing potassium-argon, major oxide, trace element, fossil, paleocurrent and archaeological sample localities. U.S. Geological Survey. Miscellaneous Field Study, MF-2219-A, 48 p., scale, 1:250,000. Miller M.L., Bradley, D.C., Bundtzen, T.K., and McCelland, W., 2002, late Cretaceous through Cenozoic strike-slip tectonics of southwestern Alaska. The Journal of Geology, Vol. 110, p. 247-270. Moll-Stalcup, E.J., 1994, Latest Cretaceous and Cenozoic magmatism in mainland Alaska: Geologic Society of America, Geology of North America, v. G-1, p. 589-620. Nokleberg, W.J., Parfenov, L.M., Monger, J.H., Norton, I.O., Khanchuk, A.I., Stone, D.B., Scotese, C.R., Scholl, D.W., and Fujita, K., 2000, Phanerozoic Tectonic Evolution of the Circum-North Pacific: U.S. Geological Survey, Professional Paper 1625. O’Dea, M., and Bartch, M., 1997, Structural controls on gold mineralization at the Donlin Creek deposit, Southwest Alaska. Etheridge Henley Williams consultant report for Placer Dome Exploration Inc. September 1997. Piekenbrock, J.R., and Petsel, S.A., 2003, Geology and interpretation of the Donlin Creek gold deposits: Juneau, Alaska, NovaGold Resources Inc., Unpublished Report, 58 p. Redfield T.F, Scholl, D.W., Fitzgerald P.G., and Beck M.E. Jr., 2007, Escape tectonics and the extrusion of Alaska: Past, present, and future: Geology, v. 35 no. 11, p. 1039- 1042. Roberts, Paul, 1993, Report on trench mapping and sampling, Donlin Creek Project: unpublished Teck Exploration Ltd. report, 37 p., 4 sheets. Sillitoe, R.H., 2001, Some metallogenic features of gold and copper deposits related to alkaline rocks and consequences for exploration: Mineralium Deposita, v. 37, p. 4-13. Scholl, D.W., Stevenson, A.J., Mueller, S., Geist, E.L., Vallier, T.L., and Engebretson, D.C., 1994, Regional-scale strain partitioning leading to escape tectonism and formation of offshore arc-trench systems, Alaska-Aleutian-Bering Sea region: Geological Society of America Abstracts with Programs, v. 26, p. 136. 42 Szumigala, D.J., Dodd, S.P., and Arribas, A., Jr., 2000, Geology and gold mineralization of the Donlin Creek prospects, southwestern Alaska, in Wiltse, M.A., eds., Short Notes on Alaska Geology, Professional Report 119, State of Alaska Department of Natural Resources Division of Geological and Geophysical Surveys, pp. 91-115. Tosdal, R.M, and Richards, J.P., 2001, Magmatic and structural controls on the development of porphyry Cu ± Mo ± Au deposits: Society of Economic Geologists Reviews, v. 14, p. 157-181. 43 2 Chapter 2: GEOCHRONOLOGY OF THE DONLIN CREEK GOLD DEPOSIT, SOUTHWEST ALASKA 2.1 INTRODUCTION The Donlin Creek deposit is a Late Cretaceous, >30 million ounce, dominantly intrusion hosted gold deposit located in the Kuskokwim Basin of southwest Alaska (Fig 1.1). The oldest structures at Donlin Creek formed during a north-northeast – south- southwest shortening event that produced a series of northward-directed thrust faults, subordinate north-dipping back thrusts, and related folds. North-east-trending folds 1interpreted to be related to strike-slip motion on the northeast-trending, dextral Nixon Fork Fault over print the fold and thrust style deformation (Fig. 1.7). Mafic intrusions, possibly as old as ~ 74 Ma (Szumigala et al. 2000) were emplaced into the Kuskokwim Group, followed by a second more voluminous set of felsic intrusions. Discordant intrusions are parallel to, and cut by, a set of north-northeast-striking, moderately southeast-dipping normal faults. The majority of the gold mineralization is hosted by steeply-dipping, northeast-striking extensional veins in the intrusions. Geochronology presented in this chapter constrains the timing of deposition of the Kuskokwim Group, as well as emplacement of the granite porphyry dykes and sills and the age of gold mineralization. The timing of these events has important implications to the tectonic evolution of southwestern Alaska. This chapter contains a review of the regional, and property scale geology of Donlin Creek, and summarizes the geochronology done in previous studies. Short wave infrared analysis (SWIR) was used to define the alteration mineralogy of the samples submitted for 40Ar/39Ar geochronology. New 40Ar/39Ar and SHRIMP U-Pb geochronology constrain the timing of the main structural fabrics at Donlin Creek. 2.2 REGIONAL GEOLOGY AND TECTONICS The Donlin Creek gold deposit is located in the Kuskokwim Mountains of southwestern Alaska (Fig. 1.1). The Kuskokwim Mountains are underlain by a Late 1 A version of this chapter will be submitted to a refereed journal for publication. MacNeil, D., Tosdal, R., Wooden, J., and Ullrich, T., Geochronology of the Donlin Creek gold deposit, southwest Alaska. 44 Cretaceous turbidite basin-fill assemblage called the Kuskokwim Group. The Kuskokwim Group forms an important overlap assemblage in southwestern Alaska and likely formed a marine embayment that stitched together Paleozoic arc and pericratonic continental margin terranes by Albian time (Decker et al., 1995). The Kuskokwim Group is intruded and overlain by Late Cretaceous to early Tertiary subduction related plutonic and volcanic rocks. These rocks are collectively called the Kuskokwim Mountains Magmatic Belt (Moll-Stalcup, 1994) and represent one of a series of north- northeast trending magmatic belts found in southwestern Alaska. The Kuskokwim Mountains Magmatic Belt consists of calc-alkaline to alkaline, basaltic to rhyolitic volcanic fields, isolated calc-alkaline stocks, felsic to mafic dyke swarms, and sub- alkalic to alkalic volcanic-plutonic complexes (Moll-Stalcup, 1994) (Fig. 1.2). This belt is interpreted to represent an unusually broad swath of dominantly subduction related arc magmatism; a result of subduction of the Kula and/or Resurrection plate during the Late Cretaceous and Early Tertiary (Moll-Stalcup, 1994; Miller et al., 2002). These magmas have undergone significant amounts of fractionation, and show evidence of crustal interaction (Moll-Stalcup, 1994; Bundtzen and Miller, 1994 Goldfarb et al., 2002). Peraluminous, locally garnet-bearing hypabyssal granite porphyry dykes, sills and stocks constitute melted crust (Moll-Stalcup, 1994; Miller and Bundtzen, 1997). Southwestern Alaska is cut by two major, northeast-trending faults which have undergone considerable offset since ~ 100 Ma (Miller et al., 2002) (Fig. 1.1). The Farewell-Denali fault system to the south has accommodated ~ 90 kilometers of dextral offset, and the Iditarod-Nixon Fault system to the north has accommodated ~ 134 kilometers of dextral offset (Blodgett and Clough, 1985; Miller et al., 2002). Smaller faults and fault splays are oriented parallel to major faults, and are commonly conjugate (Miller and Bundtzen, 1998). The Kuskokwim Group is affected by at least two phases of folding that pre-date the ~ 76 to 63 Ma volcanic-plutonic complexes (Moll-Stalcup, 1984; Decker et al, 1995; Bundtzen and Miller, 1997; Bundtzen et al., 1999; Miller et al., 2002) (Fig 1.7). Open to isoclinal folds have northeast trending axial traces, amplitudes between 1 and 3 kilometers, and are locally refolded by a broad fold generation with wavelengths of approximately 25 kilometers, 1-2 kilometer amplitudes and ~ east-west trending axial traces (Miller et al., 2002) (Fig. 1.7). Bundtzen and Miller, (2002) observed that the northeast-trending folds are spatially associated and 45 kinematically compatible to major dextral strike-slip faults, including the Nixon Fork and Denali faults (Fig. 1.7). The ~ east – west oriented folds indicate north-south compression and are restricted to the Kuskokwim sedimentary rocks between the Denali and Nixon Fork faults (Fig. 1.7). These folds are not easily reconciled with strike-slip tectonics. Miller et al., (2002) suggested that these folds are result of partitioning deformation at the regional scale where rocks adjacent to major strike slip faults form compatible northeast trending folds, and rocks between the faults experience ~ north-south compression. The Donlin Creek gold deposit is part of a larger Late Cretaceous to early Tertiary belt of hydrothermal mineralization called the Kuskokwim Mountains Mineral Belt (Bundtzen and Miller, 1997). The Kuskokwim Mountains Mineral Belt is a northeast-trending belt of precious metal enriched, polymetallic mineral deposits, spatially associated with Late Cretaceous to early Tertiary magmatism (Decker et al., 1994; Bundtzen and Miller, 1997; Nokleberg et al., 2000) (Fig. 1.2). Historic production from the Kuskokwim Mountains Mineral Belt consists of 3.22 Moz of gold, 412,000 oz of silver, and 1,377,412 kg of mercury. Antimony and tungsten have also been produced from this approximately 550 km long and 350 km wide mineral belt (Fig. 1.2) (Bundtzen and Miller, 1997). Mineral deposits are associated with four types Late Cretaceous-early Tertiary igneous complexes: (1) alkali-calcic, comagmatic volcanic- plutonic complexes; (2) calc-alkaline, meta-aluminous reduced plutons; (3) peraluminous granite-porphyry sills and dykes; and (4) subaerial bimodal volcanic rocks (Bundtzen and Miller, 1997). Five main types of deposits are known: (1) plutonic- hosted copper-gold polymetallic stockwork, skarn, and vein mineralization; (2) peraluminous granite porphyry-hosted gold polymetallic bodies; (3) plutonic-related, boron-enriched silver-tin polymetallic breccia pipes and replacement bodies; (4) gold and silver epithermal veins; and (5) gold, polymetallic heavy mineral placer deposits (Buntzen and Miller, 1997). Based upon 10 deposits genetically related to Late Cretaceous-early Tertiary intrusions, including the Donlin Creek deposit, this belt is estimated to contain a minimum inferred reserve of approximately 35.23 Moz gold, 6.46 Moz silver, 12,160 metric tons of tin and 28,088 metric tons of copper (Bundtzen and Miller, 1997). Miller et al. (2002) propose that polymetallic gold mineralization in the Kuskokwim Mineral Belt formed during three main time intervals, ~ 70, ~58 and ~ 30 46 Ma, that are coeval with strike slip faulting magmatism. Bundtzen and Miller (1997) suggest that lode deposits represent geologically and spatially related, vertically zoned hydrothermal systems that formed in response to Late Cretaceous to early Tertiary plutonisim and volcanism, generated by north-directed subduction. They attribute the differences in the geology, structural style, and metal tenor of the hydrothermal systems to the current outcrop level of exposure of the hydrothermal systems (Bundtzen and Miller, 1997). 2.3 DONLIN CREEK PROPERTY GEOLOGY The structural geology presented herein provides a context for the geochronology (this Chapter). A complete description of the subsurface geometry and a systematic structural analysis is provided in Chapter 3. Simplified cross-sections are included in this Chapter in order to illustrate the relative timing of key structural fabrics and to demonstrate their relationships to the rock units at Donlin Creek. The 8.5 kilometer long by 2.5 kilometer wide Donlin Creek property geology is dominated by a ~74 Ma (Szumigala et al., 2000) to ~71 Ma (this study) bimodal dyke swarm that intrudes fine- to coarse-grained, thin to massively bedded, light grey to dark grey, lithic sandstones (greywacke), thin to thickly bedded dark grey to black carbonaceous siltstone and shale, and minor interbedded conglomerates of the Late Cretaceous Kuskokwim Group (Fig. 1.3). 2.3.1 Sedimentary Rocks The sedimentary rocks generally strike east to southeast and dip moderately (35º - 50º) toward the southwest. Dark grey to black, poorly to moderately sorted, angular to sub-angular lithic sandstone (greywacke) is the most abundant sedimentary rock. Individual clasts consist of metamorphic, igneous and sedimentary lithic fragments (Szumigala et al., 2000). Thick intervals of siltstone and mudstone are present on the property, but these units more commonly appear as thick interbeds in lithic sandstone. Siltstone and mudstone ranges in colour from very dark grey to black. The sedimentary rocks are highly carbonaceous, locally contain high concentrations of framboidal pyrite, and commonly contain inoceramus hash, brackish water pelecypods, 47 Figure 2.1. Donlin Creek rock types: A) Flame structures at greywacke – siltstone contact; B) channel structures in greywacke; C) inoceremus hash; D) soft sedimentary deformation in interbedded siltstone and mudstone; E) interlayered volcanic ash (F) mafic dyke; G) fine grained porphyry; H) crystalline porphyry; I) lath rich porphyry; J) aphanitic porphyry; K) blue porphyry. 48 49 and dicot leaf debris (Bundtzen, 2004). Common soft sedimentary deformation structures consist of convolute bedding, dewatering channels and flame structures (Fig 2.1). The latter indicates that these rocks are upright. Interbedded ash layers are numerous within mudstone and siltstone horizons (Fig. 2.1). Ash layers are typically light to medium dark grey in colour, and range in thickness from millimeters to a few tens of centimeters. Contacts between the ash layers and their host sedimentary rocks are typically very sharp; however, instances of gradational contacts are preserved. 2.3.2 Igneous Rocks The intrusive rocks form discordant north-northeast-striking, moderately to steeply-dipping tabular intrusions, and moderately to steeply southwest-dipping, bedding concordant intrusions. Two compositions are present. The oldest intrusions are thin, volumetrically minor (<3 meters thick), highly discontinuous, ~ 74 Ma porphyritic andesite to quartz andesites (Szumigala et al., 1999) (Fig. 2.1). A younger, ~ 71 Ma (this study), texturally variable quartz feldspar porphyry is the dominant igneous rock on the property (Figs. 2.1F-K and 2.2). The quartz feldspar porphyry intrusions are subdivided into 5 phases based upon the size, shape, and distribution of phenocrysts, as well as key characteristics of the groundmass (Piekenbrock and Petsel, 2002). These categories, from oldest to youngest (based on cross cutting relationships), include: fine-grained porphyry, crystalline (crowded) porphyry, aphanitic porphyry, lath rich porphyry, and blue porphyry (Fig 2.1F-K). The discordant intrusions are commonly continuous along strike, reach strike lengths in excess of 1 kilometer, and thickness up to 100 meters (Fig. 2.2, 2.4A,B). Concordant felsic intrusions are typically discontinuous, and rarely exceed 200 meters in length (Figs. 2.2, 2.3A,B). 2.3.3 Structural Fabric The pre-intrusion geology of Donlin Creek is controlled by early, bedding parallel, north-northeast and south-southwest-dipping thrust faults, and related folds (see Chapter 3)(Figs. 2.2, 2.3A, B). Thrust faults are developed in, or adjacent to, contacts between fine, and coarse grained sedimentary rocks (Figs. 2.2, 2.3A, B). Thrust fault anticlines are commonly developed in the hanging wall of thrust faults (Figs. 2.3A, B). These upright, tight to open folds have sub-horizontal fold axes, east- 50 Figure 2.2. Geology of the Main Resource Area of Donlin Creek. Cross sections lines are defined by red lines. Prospect boundaries are shown in grey. Northeast trending cross-sections are perpendicular to bedding and bedding concordant igneous rocks. East-southeast trending sections are perpendicular to the bedding discordant igneous rocks. 51 52 Figure 2.3. Cross-sections drawn perpendicular to bedding through the Main Resource Area of Donlin Creek showing the relationship between bedding concordant and discordant igneous rocks hosted in the fold and thrust geometry of the Kuskokwim Group sedimentary rocks. A) Cross-section through the South Lewis and Queen prospects; B) Cross-section through the ACMA and Akivik prospects. 53 54 Figure 2.4. Cross sections drawn perpendicular to discordant igneous rocks portraying true dips for A) the South Lewis, and B) and Queen prospects. 55 56 southeast-trending, nearly vertical axial planes, and no penetrative axial planar cleavage. Thrust faults separate greywacke rich sequences from those that contain an abundance of siltstone and shale (Figs. 2.3A, B). Discordant, north-northeast- striking,northeast-diping back thrusts, related thrust ramp anticlines, and overprinting north-northeast-trending normal faults (Figs. 2.2, 2.3A, B). Discordant, north-northeast- steeply south-southeast-dipping, tabular intrusions are most common intrusive morphology (Fig. 2.2). They represent the dominant structural fabric on the property, and are interpreted to have been emplaced along the north-northeast-trending normal faults (Fig. 2.2, 2.4A, B). The intrusions are also cut by the normal faults, implying that extension continued after the intrusions had cooled enough to host brittle structures. Discordant intrusions are commonly localized in the brittle, coarse grained sedimentary rocks (Figs. 2.2, 2.4A, B). Concordant igneous bodies are common in fine grained sedimentary rocks (Figs. 2.3A, B). Intrusions exploit sedimentary rock contacts and bedding planes in siltstone and shale. Discordant intrusions become concordant in fine grained sedimentary rocks (Figs. 2.3A, B). This is particularly common where intrusions are focused along thrust faults and exploit folded bedding contacts in thrust ramp anticlines (Figs 2.3A, B) The fold and thrust dominated, pre-intrusion geology of Donlin Creek strongly affected the emplacement of the intrusions and explains much of their current geometry (Figs. 2.2, 2.3A, B). Gold mineralization at Donlin Creek is focused within, and adjacent to, a northeast-trending, steeply southeast-dipping set of extensional veins that cut igneous and coarse grained sedimentary rocks (O’Dea and Bartch, 1997; Szumigala et al., 2000; Ebert et al., 2000; Piekenbrock and Petsel, 2003). The highest gold grades at Donlin Creek occur where these mineralized extensional fractures intersect northeast- trending fault zones (Szumigala et al., 2000). 2.3.4 Mineralization Veins are intimately associated with gold mineralization at Donlin Creek, and are divided into 4 main types (V1-V4) according to their dominant infill mineralogy (Fig 2.5). The relative age of the veins at Donlin Creek is based upon cross cutting relationships. The oldest veins on the property consist of thin, sulphide dominated veins (V1), which range in thickness from 0.5 millimeters to 5 millimeters (Fig. 2.5A). They are 57 Figure 2.5. Donlin Creek vein types: A) V1 pyrite ±- arsenopyrite in a altered crystalline porphyry; B) V2 quartz ± sulphides (pyrite, arsenopyrite) ± carbonate in an altered crystalline porphyry; C) V3 quartz + realgar ± arsenopyrite +/- native arsenic ± carbonate in an aphanitic porphyry ; D) V3 quartz + native arsenic + stibnite; E) V4 barren carbonate veins in a coarse grained lithic sandstone. 58 discontinuous, and commonly associated with a broader halo of ankerite alteration and disseminated sulphides. Quartz-sulphide veins (V2) cross-cut older sulphide veins, and range from < 2 millimeters to 10 centimeters in width (Fig. 3.5B). They are commonly discontinuous, composed of variable amounts of quartz, pyrite and arsenopyrite, and have an ankerite + illite dominated alteration halo, with very finely disseminated pyrite, and acicular arsenopyrite (Fig. 2.5B). Thicker (5 millimeters to 15 centimeters), continuous, sulphide-bearing quartz veins (V3) occur across the Main Resource Area (Fig. 2.5, C, D). They contain variable amounts of pyrite, arsenopyrite, native arsenic, and stibnite, and locally contain significant quantities of calcite, orpiment, and realgar (Fig. 2.5, C, and D). These veins typically do not have broad alteration halos, but locally have thick (up to 1 centimeter) arsenopyrite selvages. The youngest veins on the property are carbonate-only veins (V4) (Fig. 2.5E). These veins range from millimeters to tens of centimeters. Vein compositions vary between carbonate, ankerite and dolomite, and are locally gold bearing. On average, the veins at Donlin Creek have the same strike, but dip more steeply compared to the normal faults and discordant intrusions. This relationship is commonly observed where veins and normal faults are kinematically linked in extensional environments. 2.3.5 Alteration Alteration in the Main Resource Area is closely associated with veins, veinlets and disseminated sulphides, but also occurs in unmineralized rocks (Ebert et al., 2003a). Gold bearing quartz-carbonate-pyrite-arsenopyrite veinlets typically coincide with the most intense zones of hydrothermal illite alteration (Ebert et al., 2003a). Gold mineralized granite porphyry intrusions are pervasively altered to illite ± kaolinite ± chlorite/smectite plus quartz-carbonate-pyrite (Ebert et al., 2003a). Similar alteration is present in greywacke and shale, but in limited extent and restricted to fractures (Ebert et al., 2003a). Feldspar phenocrysts in the porphyry intrusions are strongly to intensely altered to illite ± kaolinite. Biotite phenocrysts are altered to sericite and partially replaced by sulphides (pyrite ± arsenopyrite) (Ebert et al., 2003a). Fine grained porphyry matrix (e.g., aphanitic porphyry intrusions) is altered to a mixture 59 Figure 2.6. Drill hole DC06-1251 showing the relationship between lithology, alteration and gold mineralization. This figure demonstrates the close relationship between ammonium bearing illite and gold mineralization within the porphyry intrusion host rocks. 60 of illite ± sericite, silica, carbonate, and commonly contains disseminated pyrite. Very fine-grained graphite is present within altered intrusions, particularly in the blue porphyry. The best gold grades are spatially associated with intensely illite altered porphyry intrusions (Fig. 2.6). The alteration mineralogy is determined using a TerraSpec® SWIR spectrometer from Analytical Spectral Devices, Inc. (see below). This is discussed in more detail in the discussion section of this manuscript. 2.4 PREVIOUS RADIOMETRIC DATING 2.4.1 TIMING OF IGNEOUS ACTIVITY Magmatic biotite and muscovite phenocrysts collected from three unaltered porphyry dykes, located a few kilometers east and west of the Donlin Creek deposit produced K-Ar ages between 70.9 ± 2.1 and 65.1± 2.0 Ma (Miller and Bundtzen, 1994) (Fig. 3.6). Goldfarb et al. (2002) obtained an inferred U-Pb age of 69.2 ± 0.5 Ma, based upon two strong abraded, concordant to nearly concordant zircons, obtained from a porphyry intrusion in the North Lewis Prospect (Fig. 2.7). Igneous rock 40Ar/39Ar dates collected by Szumigala et al. (2000) sampled a felsic porphyry intrusion from the Queen Prospect which yielded biotite plateau ages of 70.3 ± 0.2 Ma and 72.2 ± 0.9 Ma (Fig. 2.7). A whole-rock age of 74.4 ± 0.8 was obtained for a mafic dyke from the Queen Prospect (Szumigala et al., 2000) (Fig. 2.7). It is not certain whether these ages reflect cooling through the mica closer temperatures or true crystallization ages of these rocks. U-Pb zircon geochronology has also been done at the Dome Prospect (Fig 2.1). Goldfarb et al. (2004) reported a U-Pb age 66.5 ± 0.5 Ma from two concordant zircon fractions obtained from an intrusion at the Dome prospect. Ebert et al. (2003b) reported a U-Pb age of 69.9 ± 0.5 Ma for an unaltered quartz feldspar porphyry intrusion approximately 1 kilometer north of Dome. 2.4.2 TIMING OF HYDROTHERMAL ACTIVITY The timing of mineralization is constrained by 40Ar/39Ar dating of hydrothermal sericite (e.g., Gray et al., 1997 and Szumigala et al., 2000) (Fig. 2.7). Grey et al. 1997 61 Figure 2.7. Geochronology for the Donlin Creek Property. Samples collected during this study are shown in red. Thick black lines bracket the age of intrusion (including analytical errors) based upon 2 weighted mean 207Pb corrected 206Pb/238U SHRIMP ages. 62 63 reported a 40Ar/39Ar age of 69.5 ± 1.1 Ma for gold mineralization obtained from hydrothermal sericite at the Snow Prospect (Fig 2.7). Szumigala et al. (2000) obtained 40Ar/39Ar dates on sericite that range between 73.6 ± 0.6 Ma and 67.8 ± 0.3 Ma at the Queen and Lewis prospects (Fig. 2.7). Szumigala et al. (2000), reported 40Ar/39Ar dates from sericite associated with gold mineralization at the Dome Prospect. These ages were 68.0 ± 1.0 Ma and 65.1 ± 0.9 Ma, and contradict field observations, which indicate that Dome mineralization predates mineralization found in the Main Resource Area (Ebert et al., 2003a) (Fig 2.7). Goldfarb et al. (2004) suggested that hydrothermal micas at the Dome Prospect were reset based on their 65.5 ± 0.5 Ma intrusion age 2.5 METHODOLOGY 2.5.1 Shortwave Infrared Analysis Short wave infrared (SWIR) analysis allows rapid identification of minerals. It is particularly useful for clay alteration minerals that would otherwise require costly and time-consuming analysis such as X-ray diffraction methods. SWIR analysis was performed on clay powder samples extracted from hydrothermally altered feldspar sites in porphyritic intrusive rocks (Fig. 2.8). Clays were extracted from feldspar sites using a stainless steel dental pick, placed in aluminum foil, and later transferred to a High Intensity Source Probe- (MUG-LIGHT) analyzer-compatible petri dish for analysis. Spectra were collected using a TerraSpec® SWIR spectrometer from Analytical Spectral Devices, Inc. (ASD) in order to confirm the alteration mineralogy associated with gold mineralization in advance of 40Ar/39Ar geochronology (see below). The Terraspec is a spectrometer that measures infrared radiation between 1300-2500 µm wavelengths for geological materials. The MUG-LIGHT source was used for the analysis, and has a beam diameter of 30 millimeters. Light interacting with molecular bonds in the mineral powder produces characteristic spectral responses. Absorption wavelengths (features) of minerals indicate their presence in a given sample. Hydrous minerals (OH-, H2O, CO3-2, NH4+, AlOH, FeOH and MgOH), are commonly formed during hydrothermal alteration, and produce key absorption features that are used for mineral identification (e.g. Clark et al., 1990; Thompson et al., 1999). 64 Figure 2.8. Short Wave Infrared spectra collected using TerraSpec® SWIR spectrometer: A) Sample DC06-1251-AR: Crystalline Ammonium Illite; B) Sample DC06-1281-1-AR: Crystalline Ammonium Illite; C) Sample DC06-1281-2: Crystalline Ammonium Illite; D) Sample DC06-1415-1-AR: Crystalline Ammonium Illite; E) Sample DC06-1415-2-AR: Crystalline Ammonium Illite; F) Sample DC06- 1284-AR: Crystalline Ammonium Muscovite. 65 66 A white reference standard (100% reflectance across the entire spectrum) was used to calibrate the spectrometer ever hour. Multiple measurements were collected for each powder sample in order to check for sample heterogeneity. The Spectral Geologist (TSG) Software Professional 2007 (TSG, 2007), and SpecMin Pro software (SII, 2005) were used to aid in interpreting mineral spectra by comparing them to established spectra reference libraries. 2.5.2 U-Pb GEOCHRONOLOGY Sample preparation was undertaken at The University of British Columbia. A total of 0.5 – 20 kg of each sample was pulverized using a jaw crusher and Bico disk mill. Isolation of zircon involved standard mineral separation techniques, including the use of a Wilfley table, heavy liquids (methylene iodide) and a Frantz Isodynamic Separator. Prior to Sensitive High-Resolution Ion Microprobe – Reverse Geometry (SHRIMP-RG) analysis, the handpicked zircons from each rock sample were mounted in epoxy resin and polished to expose zircon cores. Following photography in reflected light (RL), the zircons were then imaged using a cathodoluminescence (CL) detector on the JEOL JSM 5600 scanning electron microscope (SEM) in SUMAC. For selected samples lacking CL zonation, back-scattered imagery was employed (BSE) using the same SEM. The CL images were further enhanced for contrast and detail using Adobe PhotoshopTM. Each mount was then gold-coated and placed in the sample chamber of the SHRIMP-RG (Sensitive High Resolution Ion Microprobe–Reverse Geometry) at SUMAC. For each analysis, an 8 nA 1602-primary ion beam was rastered across the grain for 2 minutes, to remove the gold coat and surface contamination. Positive secondary ions were then collected by excavating an approximately 1 mm deep flat- floored, rounded to elliptical pit (~20 – 30 mm diameter). For each analysis, six scans of peaks at 90Zr216O, 204Pb, 206Pb, 207Pb, 238U, 232Th16O and 238U16O were collected. Beam tuning and centering was done using the 238U16O peak, with maximum count times for 206Pb and 207Pb of 20 seconds. Data were referenced to the SUMAC R33 (419 Ma) internal standard, which were analyzed repeatedly during the analysis period. Uranium concentrations were obtained by comparison with zircon standard MAD (4200 ppm). Common Pb correction is based on the average crustal earth of Stacey and 67 Table 2.1. U-Pb SHRIMP-RG zircon analytical data for samples in the study area. Sample; 206Pb 1 U Th 232Th 206Pb 206Pb 2 1σ 207U 1σ Apparent Age (Ma, 1σ) Spot Name % (ppm) (ppm) 238U (ppm) 238U % 206Pb 2 % 206Pb238U 3 1646-1 -0.53 198 145 0.76 2.22 0.0131 1.38 0.0514 5.00 83.31 1.2 1646-2 -1.67 78 32 0.42 0.95 0.0144 2.15 0.0470 8.02 90.96 1.8 1646-3 -4.93 77 33 0.45 0.89 0.0142 2.88 0.0535 7.85 85.96 1.8 1646-4 2.64 69 29 0.44 0.82 0.0136 2.37 0.0530 7.27 88.79 1.8 1646-5 -3.85 118 62 0.54 1.43 0.0146 2.34 0.0524 8.00 89.31 1.5 1646-6 -0.79 146 61 0.43 1.66 0.0134 1.45 0.0485 5.99 85.05 1.2 1646-7 0.00 133 50 0.39 1.52 0.0133 1.36 0.0511 5.59 84.70 1.2 1646-8 -1.12 69 23 0.35 0.81 0.0139 1.97 0.0451 8.55 88.34 1.7 1646-9 1.92 56 18 0.32 0.69 0.0139 2.26 0.0673 7.28 88.58 1.9 1646-10 -1.36 154 75 0.50 1.73 0.0132 1.48 0.0582 5.39 82.51 1.2 1646-11 -3.15 63 32 0.53 0.77 0.0146 2.45 0.0434 8.59 91.06 1.8 1646-12 -2.50 85 42 0.51 1.00 0.0140 2.09 0.0483 7.36 87.70 1.5 1646-13 -2.51 60 23 0.40 0.69 0.0137 2.31 0.0484 8.78 85.41 1.8 GP-1 1.39 178 118 0.68 2.05 0.0132 1.40 0.0508 5.12 85.10 1.1 GP-2 1.85 76 36 0.49 0.91 0.0136 2.02 0.0477 7.55 88.74 1.6 GP-3 1.07 73 22 0.32 0.89 0.0141 2.08 0.0485 7.64 90.95 1.7 GP-4 -0.14 80 26 0.34 0.93 0.0135 1.74 0.0475 7.71 86.56 1.5 GP-5 -6.57 41 29 0.74 0.46 0.0141 5.35 0.0461 11.83 84.84 2.1 GP-6 0.06 611 387 0.65 7.31 0.0139 0.63 0.0488 2.62 89.10 0.6 GP-7 -1.20 115 38 0.34 1.41 0.0145 1.54 0.0489 6.08 91.78 1.3 GP-8 0.20 86 29 0.34 1.02 0.0137 1.72 0.0466 8.32 88.11 1.6 GP-9 0.84 86 24 0.29 1.00 0.0135 1.75 0.0529 9.85 86.50 1.5 GP-10 1.01 375 310 0.86 4.48 0.0138 1.01 0.0510 3.24 88.77 0.7 GP-11 -0.16 327 129 0.41 8.59 0.0307 0.65 0.0507 2.42 194.18 1.3 1415-1 0.98 151 57 0.39 1.74 0.0133 1.52 0.0361 13.68 86.57 1.3 1415-2 0.00 89 34 0.39 1.10 0.0145 1.94 0.0485 8.19 92.44 1.8 1415-3 0.00 89 53 0.62 1.06 0.0139 1.85 0.0566 7.58 88.30 1.7 1415-4 -6.28 59 23 0.40 0.71 0.0148 3.76 0.0920 25.29 89.71 2.1 1415-5 0.00 69 26 0.39 0.85 0.0144 2.14 0.0554 8.21 91.56 2.0 1415-6 2.01 103 33 0.33 1.18 0.0131 1.71 0.0362 21.10 84.78 1.3 1415-7 0.00 183 129 0.73 2.16 0.0137 1.21 0.0479 5.14 87.75 1.1 1415-8 0.00 101 42 0.42 1.20 0.0137 1.52 0.0574 5.86 86.93 1.4 1415-9 0.00 159 76 0.49 1.86 0.0136 1.26 0.0505 5.12 87.04 1.1 1415-10 0.00 52 25 0.50 0.63 0.0142 2.12 0.0456 9.33 91.19 2.0 1280-1 0.62 225 79 0.36 2.12 0.0109 1.42 0.0466 9.35 69.97 1.0 1280-2 0.17 957 144 0.16 9.20 0.0112 0.67 0.0451 3.73 71.82 0.5 1280-3 0.54 221 79 0.37 2.18 0.0115 1.37 0.0485 8.41 73.31 1.0 1280-4 2.09 154 48 0.32 1.53 0.0113 1.93 0.0365 26.21 73.66 1.2 1280-5 0.76 365 58 0.16 3.48 0.0110 1.17 0.0410 11.70 71.18 0.8 1280-6 1.24 225 127 0.58 2.20 0.0112 1.51 0.0374 15.79 72.99 1.0 1280-7 0.00 224 39 0.18 2.15 0.0112 1.39 0.0476 5.92 71.63 1.0 1280-8 1.13 137 48 0.37 1.26 0.0106 1.94 0.0371 18.98 68.86 1.3 1280-9 1.42 140 45 0.33 1.25 0.0103 1.94 0.0388 18.27 66.71 1.2 1280-10 0.00 96 32 0.34 0.95 0.0114 2.08 0.0501 8.73 73.00 1.6 1280-11 0.98 249 101 0.42 2.45 0.0113 1.38 0.0431 11.43 72.87 1.0 1280-12 0.00 469 113 0.25 4.52 0.0112 0.96 0.0477 4.16 71.83 0.7 1280-13 2.88 120 47 0.40 1.16 0.0109 2.41 0.0280 46.75 71.80 1.4 1280-14 0.04 287 392 1.41 43.20 0.1752 0.57 0.0759 1.41 1038.04 5.9 1280-15 0.16 795 259 0.34 7.54 0.0110 0.75 0.0474 3.69 70.70 0.5 1280-16 0.27 931 88 0.10 8.52 0.0106 0.79 0.0438 4.68 68.37 0.5 1280-17 2.57 180 78 0.45 1.64 0.0103 2.04 0.0419 26.06 66.57 1.1 DC06-1646 (Volcanic Ash) Gravel Pit (Volcanic Ash) DC06-1280 (Crystaline Porphyry) DC07-1415 (Volcainc Ash) 68 Sample; 206Pb 1 U Th 232Th 206Pb 206Pb 2 1σ 207U 1σ Apparent Age (Ma, 1σ) Spot Name % (ppm) (ppm) 238U (ppm) 238U % 206Pb 2 % 206Pb238U 3 DC06-1285 (Fine Grained Porphyry) 1285-1 0.00 273 53 0.20 2.60 0.0111 1.21 0.0506 7.09 70.76 0.9 1285-2 0.15 693 115 0.17 6.79 0.0114 0.74 0.0457 3.69 73.18 0.5 1285-3 3.56 102 26 0.26 1.00 0.0110 2.60 73.72 1.5 1285-4 0.00 720 140 0.20 6.99 0.0113 0.85 0.0498 3.03 72.25 0.6 1285-5 0.00 727 73 0.10 6.98 0.0112 0.76 0.0492 3.28 71.49 0.6 1285-6 0.00 196 53 0.28 1.92 0.0114 1.45 0.0504 5.95 72.73 1.1 1285-7 0.30 363 191 0.54 6.12 0.0195 0.83 0.0464 4.37 125.10 1.0 1285-8 0.87 331 60 0.19 3.12 0.0109 1.19 0.0375 11.46 70.67 0.8 1285-9 -0.59 629 113 0.19 5.94 0.0111 0.85 0.0519 5.53 70.55 0.6 1285-10 -4.27 64 15 0.25 0.64 0.0122 3.34 0.0822 22.36 74.68 1.9 1285-11 1.54 66 22 0.35 0.63 0.0110 2.56 0.0413 23.64 71.09 1.7 1285-12 0.88 203 32 0.16 1.92 0.0109 1.47 0.0385 12.11 70.50 1.0 1 Common Lead 2 Atomic ratios of radiogenic Pb 3 206Pb/238U age using 207Pb to correct for common lead 69 Figure 2.9. Stacked spectra for all 40Ar/39Ar analyses. All samples are hydrothermal, NH4+ bearing crystalline illite except for DC06-1284-AR which is hydrothermal, NH4+ bearing crystalline muscovite. Muscovite, on this diagram is discernable from illite by a slight shift in the large negative feature at 2200 nanometers and by its relative flat profile between 400-1200 and 1600 to 1900 nanometers. 70 Kramers (1975). Data reduction was undertaken using SQUID 1.02, and ages calculated using Isoplot 3.0 (Ludwig, 2003). SHRIMP-RG analyses were undertaken in conjunction with Dr. Joe Wooden and Dr. Frank Mazdab of SUMAC. U- Pb results are shown in Table 3.1 and Figure 3.8. A complete data table is presented in Figure 2.1. 2.5.3 40Ar/39Ar GEOCHRONOLOGY Clay samples were extracted from the highly altered feldspar sites of hypabyssal intrusive rocks at Donlin Creek using a clean, stainless steel dental pick. The clay material was then analyzed with a TerraSpec® SWIR spectrometer from Analytical Spectral Devices, Inc. (see below) in order to determine the alteration mineralogy of the recovered the clay powder. Approximately 10 mg of clay powder, with grain sizes between 0.25 and 0.75 microns, was collected and subsequently washed in deionized water, rinsed and then air-dried at room temperature. A hand magnet was passed over the samples to remove any magnetic minerals. The samples were wrapped in aluminum foil with similar-aged samples and neutronflux monitors (Fish Canyon Tuff sanidine; 28.02 Ma (Renne et al., 1998). The samples were irradiated at the McMaster Nuclear Reactor in Hamilton, Ont., for 56 MWH, with a neutron flux of approximately 3x1016 neutrons/cm2. Analyses (n = 54) of 18 neutron flux monitor positions produced uncertainties of <0.5% in the J value. The samples were analyzed at the Noble Gas Laboratory, Pacific Centre for Isotopic and Geochemical Research, University of British Columbia, Vancouver, BC, Canada. The separates were step-heated at incrementally higher powers in the defocused beam of a 10-W CO2 laser (New Wave Research MIR10) until fused. The gas evolved from each step was analyzed using a VG5400 mass spectrometer equipped with an ion-counting electron multiplier. All measurements were corrected for total system blank, mass spectrometer sensitivity, mass discrimination, radioactive decay during and subsequent to irradiation, as well as interfering argon from atmospheric contamination and the irradiation of calcium, Chlorite and potassium (isotope production ratios: (40Ar/39Ar)K = 0.0302, (37Ar/39Ar)Ca = 1416.4306, (36Ar/39Ar)Ca = 0.3952, Ca/K = 1.83 (37ArCa/39ArK). The plateau and correlation ages were calculated using Isoplot ver.3.09 (Ludwig, 2003). Errors are quoted at the 2-sigma 71 Table 2.2. 40Ar/39Ar results from alteration minerals associated with gold mineralization at Donlin Creek. Sample number and properties1 ; Laser Power % Age 40Ar/39Ar 38Ar/39Ar 37Ar/39Ar 36Ar/39Ar Ca/K2 Cl/K3 %40Ar atm4 f 39Ar5 40Ar*/39ArK6 2 61.84 ± 1.25 14.095 ± 0.007 0.094 ± 0.034 0.062 ± 0.07 0.016 ± 0.038 0.237 0.018 32.36 1.73 9.292 ± 0.191 2.1 68.66 ± 1.05 13.282 ± 0.010 0.061 ± 0.038 0.072 ± 0.034 0.010 ± 0.044 0.278 0.010 20.17 1.82 10.336 ± 0.162 2.2 71.77 ± 0.57 12.755 ± 0.006 0.046 ± 0.036 0.095 ± 0.025 0.006 ± 0.032 0.365 0.007 13.95 3.63 10.814 ± 0.087 2.3 73.21 ± 0.70 12.485 ± 0.009 0.035 ± 0.024 0.099 ± 0.022 0.005 ± 0.024 0.378 0.005 10.76 6.99 11.036 ± 0.107 2.4 75.13 ± 0.75 12.542 ± 0.007 0.030 ± 0.030 0.079 ± 0.016 0.004 ± 0.068 0.302 0.003 8.74 6.25 11.330 ± 0.116 2.5 74.23 ± 0.57 12.309 ± 0.006 0.025 ± 0.028 0.074 ± 0.037 0.004 ± 0.054 0.283 0.002 8.10 5.87 11.191 ± 0.088 2.6 73.41 ± 1.06 11.989 ± 0.014 0.022 ± 0.045 0.076 ± 0.026 0.003 ± 0.041 0.292 0.002 6.90 8.58 11.066 ± 0.164 2.7 74.07 ± 0.52 11.966 ± 0.005 0.017 ± 0.026 0.065 ± 0.028 0.003 ± 0.070 0.247 0.001 5.57 5.40 11.168 ± 0.080 2.8 73.34 ± 1.44 11.808 ± 0.019 0.016 ± 0.052 0.074 ± 0.032 0.002 ± 0.072 0.283 0.000 5.52 8.04 11.056 ± 0.222 2.9 72.57 ± 1.16 11.647 ± 0.016 0.016 ± 0.053 0.078 ± 0.026 0.002 ± 0.041 0.302 0.000 5.41 12.05 10.936 ± 0.178 3 72.10 ± 1.06 11.576 ± 0.014 0.014 ± 0.037 0.090 ± 0.020 0.002 ± 0.040 0.346 0.000 5.43 11.22 10.864 ± 0.163 3.2 72.25 ± 0.99 11.55 ± 0.0140 0.014 ± 0.041 0.120 ± 0.022 0.002 ± 0.032 0.465 0.000 5.12 15.03 10.887 ± 0.152 3.3 72.97 ± 0.48 11.533 ± 0.006 0.013 ± 0.050 0.115 ± 0.019 0.002 ± 0.054 0.451 0.000 3.98 13.38 10.998 ± 0.073 Tot./Avg. 11.856 ± 0.002 0.022 ± 0.006 0.192 ± 0.002 0.003 ± 0.007 0.351 0.004 100.00 10.965 ± 0.024 2 64.11 ± 1.90 16.264 ± 0.005 0.024 ± 0.118 0.051 ± 0.031 0.022 ± 0.044 0.193 0.001 39.41 2.10 9.631 ± 0.29 2.2 72.43 ± 0.80 14.514 ± 0.005 0.018 ± 0.046 0.034 ± 0.038 0.012 ± 0.031 0.126 0.000 24.05 6.31 10.906 ± 0.124 2.4 71.75 ± 0.55 12.424 ± 0.006 0.015 ± 0.042 0.019 ± 0.040 0.005 ± 0.028 0.074 0.000 12.43 14.55 10.801 ± 0.085 2.5 72.32 ± 0.46 11.340 ± 0.006 0.014 ± 0.026 0.019 ± 0.031 0.001 ± 0.072 0.071 0.000 3.20 13.29 10.889 ± 0.07 2.6 72.81 ± 0.62 11.279 ± 0.008 0.013 ± 0.047 0.014 ± 0.020 0.001 ± 0.130 0.054 0.000 1.93 11.18 10.965 ± 0.095 2.7 72.48 ± 1.11 11.169 ± 0.015 0.013 ± 0.049 0.011 ± 0.072 0.001 ± 0.034 0.041 0.000 1.60 17.83 10.914 ± 0.170 2.8 72.55 ± 0.59 11.178 ± 0.008 0.013 ± 0.032 0.005 ± 0.110 0.001 ± 0.074 0.021 0.000 1.54 15.40 10.924 ± 0.091 2.9 72.95 ± 0.57 11.386 ± 0.007 0.013 ± 0.039 0.005 ± 0.084 0.001 ± 0.083 0.017 0.000 2.59 9.95 10.987 ± 0.087 3 72.28 ± 0.74 11.513 ± 0.010 0.015 ± 0.035 0.009 ± 0.063 0.002 ± 0.062 0.033 0.000 4.08 5.21 10.884 ± 0.113 3.2 72.51 ± 0.55 11.610 ± 0.006 0.014 ± 0.099 0.006 ± 0.094 0.002 ± 0.085 0.021 0.000 4.33 4.18 10.919 ± 0.084 Tot./Avg. 11.689 ± 0.002 0.014 ± 0.008 0.029 ± 0.004 0.003 ± 0.009 0.053 0.000 100.00 10.880 ± 0.021 2 28.50 ± 3.38 15.237 ± 0.017 0.024 ± 0.216 0.004 ± 1.105 0.036 ± 0.049 0.017 0.001 69.21 0.40 4.244 ± 0.507 2.1 41.05 ± 1.99 11.462 ± 0.011 0.021 ± 0.143 0.006 ± 0.451 0.017 ± 0.057 0.026 0.001 41.66 0.60 6.134 ± 0.301 2.2 72.96 ± 0.90 13.205 ± 0.007 0.014 ± 0.120 0.005 ± 0.310 0.007 ± 0.051 0.018 0.000 14.41 1.75 11.000 ± 0.139 2.3 78.87 ± 0.45 12.808 ± 0.005 0.014 ± 0.048 0.004 ± 0.114 0.003 ± 0.044 0.017 0.000 5.96 5.41 11.911 ± 0.070 2.4 74.65 ± 0.51 11.820 ± 0.005 0.014 ± 0.074 0.003 ± 0.155 0.002 ± 0.108 0.013 0.000 3.32 4.05 11.260 ± 0.078 2.5 73.58 ± 0.39 11.392 ± 0.005 0.012 ± 0.031 0.004 ± 0.112 0.001 ± 0.058 0.014 0.000 1.90 13.28 11.095 ± 0.059 2.6 72.19 ± 0.66 11.056 ± 0.009 0.013 ± 0.032 0.004 ± 0.084 0.000 ± 0.109 0.015 0.000 0.96 18.64 10.882 ± 0.101 2.7 72.23 ± 1.11 11.070 ± 0.016 0.011 ± 0.034 0.003 ± 0.090 0.001 ± 0.098 0.013 -0.001 1.11 27.49 10.887 ± 0.171 2.8 72.73 ± 0.35 11.213 ± 0.005 0.013 ± 0.036 0.003 ± 0.050 0.001 ± 0.059 0.012 0.000 1.61 18.65 10.965 ± 0.053 3 72.23 ± 0.40 11.342 ± 0.005 0.013 ± 0.052 0.004 ± 0.063 0.001 ± 0.060 0.014 0.000 3.03 7.39 10.887 ± 0.062 3.2 72.63 ± 0.80 11.695 ± 0.008 0.012 ± 0.077 0.004 ± 0.354 0.002 ± 0.118 0.015 0.000 4.18 2.35 10.949 ± 0.123 Tot./Avg. 11.283 ± 0.002 0.013 ± 0.007 0.008 ± 0.009 0.001 ± 0.012 0.014 0.000 100.00 10.947 ± 0.027 2 59.32 ± 1.34 16.535 ± 0.008 0.021 ± 0.101 0.007 ± 0.310 0.025 ± 0.026 0.027 0.001 44.40 0.91 8.902 ± 0.204 2.1 71.98 ± 1.04 13.717 ± 0.009 0.018 ± 0.097 0.010 ± 0.129 0.009 ± 0.043 0.036 0.001 18.71 1.25 10.84 ± 0.160 2.2 70.38 ± 0.45 11.619 ± 0.005 0.014 ± 0.047 0.008 ± 0.067 0.003 ± 0.047 0.029 0.000 7.83 5.17 10.595 ± 0.069 2.3 70.10 ± 0.55 11.269 ± 0.006 0.013 ± 0.074 0.009 ± 0.055 0.002 ± 0.086 0.034 0.000 5.14 4.01 10.552 ± 0.084 2.4 70.88 ± 1.23 11.074 ± 0.017 0.013 ± 0.032 0.006 ± 0.051 0.001 ± 0.052 0.023 0.000 2.93 10.76 10.672 ± 0.189 2.5 70.62 ± 0.57 10.921 ± 0.008 0.013 ± 0.045 0.004 ± 0.049 0.001 ± 0.110 0.015 0.000 1.90 9.88 10.632 ± 0.088 2.6 71.08 ± 0.38 10.980 ± 0.005 0.012 ± 0.050 0.004 ± 0.096 0.001 ± 0.088 0.014 0.000 1.75 9.10 10.702 ± 0.058 2.7 70.97 ± 0.34 10.943 ± 0.004 0.012 ± 0.047 0.003 ± 0.055 0.001 ± 0.107 0.010 0.000 1.62 9.71 10.684 ± 0.052 2.8 71.33 ± 0.33 10.969 ± 0.004 0.013 ± 0.065 0.002 ± 0.109 0.001 ± 0.101 0.008 0.000 1.43 12.67 10.74 ± 0.0510 2.9 71.01 ± 0.33 10.924 ± 0.004 0.012 ± 0.050 0.002 ± 0.235 0.001 ± 0.105 0.006 0.000 1.38 9.83 10.691 ± 0.050 3 70.92 ± 0.33 10.912 ± 0.004 0.013 ± 0.036 0.002 ± 0.096 0.001 ± 0.102 0.007 0.000 1.30 7.70 10.677 ± 0.051 3.2 71.11 ± 0.35 10.930 ± 0.005 0.012 ± 0.028 0.003 ± 0.128 0.001 ± 0.072 0.009 0.000 1.22 8.28 10.706 ± 0.054 3.4 71.38 ± 0.32 10.958 ± 0.004 0.012 ± 0.028 0.004 ± 0.043 0.001 ± 0.075 0.014 0.000 1.20 10.74 10.748 ± 0.049 Tot./Avg. 11.027 ± 0.001 0.013 ± 0.007 0.008 ± 0.006 0.001 ± 0.011 0.014 0.000 100.00 10.671 ± 0.014 2 69.19 ± 1.10 14.064 ± 0.008 0.021 ± 0.052 0.004 ± 0.135 0.012 ± 0.040 0.015 0.001 24.63 2.66 10.409 ± 0.169 2.1 72.38 ± 0.62 12.127 ± 0.006 0.016 ± 0.099 0.003 ± 0.219 0.004 ± 0.060 0.013 0.000 8.91 4.48 10.898 ± 0.096 2.2 71.93 ± 0.47 11.504 ± 0.006 0.014 ± 0.036 0.004 ± 0.117 0.002 ± 0.033 0.016 0.000 4.98 8.33 10.83 ± 0.073 2.3 72.14 ± 0.65 11.301 ± 0.009 0.013 ± 0.028 0.004 ± 0.032 0.001 ± 0.063 0.014 0.000 3.15 12.24 10.862 ± 0.100 2.4 71.53 ± 0.77 11.132 ± 0.011 0.013 ± 0.047 0.004 ± 0.062 0.001 ± 0.072 0.014 0.000 2.48 10.91 10.768 ± 0.118 2.5 71.40 ± 0.52 11.035 ± 0.007 0.013 ± 0.027 0.004 ± 0.043 0.001 ± 0.043 0.015 0.000 2.02 22.14 10.748 ± 0.080 2.6 71.78 ± 0.37 11.157 ± 0.005 0.013 ± 0.040 0.004 ± 0.068 0.001 ± 0.044 0.013 0.000 2.48 15.54 10.806 ± 0.057 2.7 72.08 ± 0.41 11.253 ± 0.005 0.013 ± 0.050 0.004 ± 0.104 0.001 ± 0.065 0.016 0.000 2.73 9.68 10.853 ± 0.062 2.8 71.91 ± 0.51 11.315 ± 0.005 0.014 ± 0.089 0.005 ± 0.041 0.002 ± 0.118 0.017 0.000 3.14 5.49 10.826 ± 0.078 3 71.85 ± 0.37 11.169 ± 0.005 0.013 ± 0.042 0.008 ± 0.052 0.001 ± 0.092 0.028 0.000 2.24 8.53 10.817 ± 0.057 Tot./Avg. 11.250 ± 0.001 0.013 ± 0.007 0.009 ± 0.006 0.001 ± 0.010 0.016 0.000 100.00 10.798 ± 0.015 DC06-1280-2-AR, hydrothermal illite; J7 = 0.003755 ± 0.000012 (2σ), Volume 39ArK8 = 902.66, Integrated Date = 70.88 ± 0.29 Isotope Ratios Atmospheric DC06-1280-1-AR, hydrothermal illite; J7 = 0.003752 ± 0.000010 (2σ), Volume 39ArK 8 = 700.8, Integrated Date = 72.62 ± 0.40 DC06-1284-AR, hydrothermal Muscovite J7 = 0.003756 ± 0.000012 (2σ), Volume 39ArK 8 = 578.26, Integrated Date = 72.26 ± 0.36 DC06-1251-AR, hydrothermal illite; J7 = 0.003753 ± 0.000010 (2σ), Volume 39ArK8 = 716.52, Integrated Date = 72.75 ± 0.36 DC06-1415-1-AR, hydrothermal illite; J7 = 0.003756 ± 0.000012 (2σ), Volume 39ArK8 = 699.4, Integrated Date = 71.72 ± 0.30 72 Sample number and properties1 ; Laser Power % Age 40Ar/39Ar 38Ar/39Ar 37Ar/39Ar 36Ar/39Ar Ca/K2 Cl/K3 %40Ar atm4 f 39Ar5 40Ar*/39ArK6 2 65.75 ± 1.87 15.429 ± 0.009 0.018 ± 0.146 0.004 ± 0.363 0.019 ± 0.049 0.019 0.000 31.00 1.18 9.901 ± 0.287 2.1 70.24 ± 0.87 13.447 ± 0.006 0.017 ± 0.099 0.003 ± 0.386 0.010 ± 0.040 0.012 0.000 18.60 3.11 10.59 ± 0.134 2.2 71.61 ± 0.78 12.725 ± 0.007 0.015 ± 0.078 0.002 ± 0.373 0.007 ± 0.050 0.009 0.000 12.76 4.05 10.8 ± 0.120 2.3 70.92 ± 0.61 12.028 ± 0.006 0.013 ± 0.073 0.005 ± 0.106 0.004 ± 0.049 0.020 0.000 9.36 6.51 10.694 ± 0.094 2.4 71.90 ± 0.72 11.961 ± 0.008 0.013 ± 0.073 0.002 ± 0.261 0.004 ± 0.064 0.009 0.000 6.79 4.28 10.845 ± 0.111 2.5 71.31 ± 0.40 11.573 ± 0.005 0.014 ± 0.036 0.003 ± 0.085 0.003 ± 0.039 0.011 0.000 5.96 12.80 10.754 ± 0.062 2.6 71.87 ± 0.51 11.453 ± 0.006 0.013 ± 0.038 0.003 ± 0.093 0.002 ± 0.060 0.011 0.000 4.24 13.52 10.841 ± 0.078 2.7 71.82 ± 0.38 11.389 ± 0.005 0.013 ± 0.032 0.003 ± 0.129 0.002 ± 0.036 0.011 0.000 3.58 10.83 10.833 ± 0.059 2.8 72.16 ± 0.43 11.397 ± 0.005 0.013 ± 0.067 0.003 ± 0.093 0.002 ± 0.071 0.011 0.000 3.29 12.27 10.886 ± 0.067 2.9 72.22 ± 0.40 11.413 ± 0.005 0.013 ± 0.048 0.003 ± 0.073 0.002 ± 0.054 0.013 0.000 3.44 13.62 10.894 ± 0.062 3 72.11 ± 0.55 11.346 ± 0.006 0.012 ± 0.081 0.003 ± 0.065 0.002 ± 0.123 0.013 -0.001 2.60 8.85 10.877 ± 0.085 3.1 71.62 ± 0.48 11.244 ± 0.006 0.013 ± 0.087 0.004 ± 0.094 0.001 ± 0.045 0.018 0.000 2.39 8.97 10.803 ± 0.073 Tot./Avg. 11.689 ± 0.002 0.014 ± 0.008 0.0290±0.0043 0.0026±0.0085 0.053 0.000 100 10.880 ± 0.021 Isotope Ratios Atmospheric DC06-1415-2-AR, hydrothermal illite; J7 = 0.003749 ± 0.000010 (2σ), Volume 39ArK8 = 484.09, Integrated Date = 71.68 ± 0.25 73 (95% confidence) level and are propagated from all sources except mass spectrometer sensitivity and age of the flux monitor. The best fit statistically-justified plateau and plateau age were selected based on the following criteria: 1. Three or more contiguous steps comprising more than 50% of the 39Ar; 2. Probability of fit of the weighted mean age greater than 5%; 3. Slope of the error-weighted line through the plateau ages equals zero at 5% confidence; 4. Ages of the two steps on either side of a plateau must not have non-zero slopes with the same sign (at 1.8σ, six or more steps only). Results are presented in Table 2.2 and Figures 2.10A-F. Steps that deviate from the plateau can be explained in a number of ways. Deviations can be attributed to mineral inclusions, dirty samples (multiple phases), fluid inclusions, degrees of alteration, argon loss and/or addition at crystal growth boundaries. 2.6 DATA 2.6.1 SHORTWAVE INFRARED SAMPLE DESCRIPTIONS AND DATA The shortwave infrared data presented in this section defines the clay alteration mineral(s) that are most closely related to gold mineralization. In addition to the samples described below, spectra for 127 porphyry intrusion SWIR samples have been interpreted for drill hole DC06-1251, in order to determine the alteration minerals associated with gold mineralization (Fig. 2.6). Illite shows an excellent correlation with gold mineralization. Specifically NH4+ bearing illite is coincident with high gold grades (Fig. 2.6). This data is important to the interpretation of the 40Ar/39Ar geochronology presented later in this chapter. The SWIR samples and ASD data are described individually. These descriptions are followed by an interpretation in the next section. Sample DC06-1251-AR, (541722.7 mE, 6878617.8 mN, 223.2 m Elev) This sample was collected from drill core at a depth between 79.80 and 80.15 meters. DC06-1251-AR is a powered clay sample collected from intensely hydrothermally altered feldspar sites in a crowded crystalline porphyry intrusive rock. 74 Figure 2.10. Weighted Mean 207Pb corrected 206Pb/238U age plots for volcanic ash and felsic intrusive rocks from the Donlin Creek Property. Error bars are reported at 2σ values. Grey point data is rejected due to either interpreted lead loss or inheritance. “Point data” represent individually ablated and ionized points from separate zircons on a grain mount. Zircon is the mineral analyzed for all samples. A) Sample DC07-1646 consisted of volcanic ash and yielded an age of 87.08 ± 1.1 Ma; B) Sample GP (volcanic ash) returned an age of 88.08 ± 0.9 Ma; C) Sample 1415 (Volcanic ash) yielded an age of 87.4 ± 1.2 Ma; D) Sample DC06- 1280 (Crystalline Porphyry) returned an age of 71.6 ± 0.6 Ma; E) Sample DC06- 1285 (Fine Grained Porphyry) produced an age of 71.4 ± 0.7 Ma. 75 The clay material was collected with a stainless steel dental pick and ranges in size from 0.25 µm to 0.75 µm and is white with a reddish iridescence. The sample was collected from an interval containing 1.04 g/t Au associated with quartz + sulphide veins. Veins contain between 1% and 3% pyrite and trace (0.01%) to 1% very fine grained arsenopyrite. Disseminated sulphides (~ 2% pyrite and trace arsenopyrite) occur between vein intervals in alteration halos. The SWIR spectrum for this sample has an excellent signal to noise ratio and is considered to be of high quality (Fig. 2.8A). This sample contains major features at 1.410 µm and 2.197 µm. Large features are present at 1.905 µm, 2.107 µm, 2.343 µm and 2.434 µm. A notable doublet feature exists at 2.000 µm. These features are consistent with illite. Sample DC06-1280-1-AR, (541319.4 mE, 6878989 mN, 154.9 m Elev) This sample was collected from drill core at a depth between 207.7 and 207.83 meters. The clay material ranges in size from 0.25 µm to 0.75 µm, and is white with a reddish iridescence. DC06-1280-AR consists of illite bearing clay obtained from intensely hydrothermally altered feldspar phenocrysts, set in an aphanitic groundmass of an aphanitic porphyry intrusive (RDA). The interval that the sample was collected from contains 3.98 g/t Au associated with quartz + carbonate + sulphide veins. The veins contains up to ~ 3% pyrite and 2% arsenopyrite. Alteration halos around these veins contain 3% disseminated pyrite and ~ 1% disseminated arsenopyrite. The SWIR spectrum for this sample has an excellent signal to noise ratio, and is considered to be of high quality (Fig. 2.8B). This sample contains major features at 1.410 µm and 2.197 µm. Large features are present at 1.905 µm, 2.107 µm, 2.343 µm and 2.434 µm. A notable doublet feature exists at 2.000 µm. These features indicate that this mineral is illite. 76 Sample DC06-1280-2-AR, (541276.5 mE, 6879021.1 mN, 54.2 m Elev) This sample was collected from drill core at a depth between 322.0 and 322.1 meters. The clay material ranges in size from 0.25 µm to 0.75 µm and is white with a reddish iridescence. Sample DC06-1280-2-AR is composed of clay collected from intensely hydrothermally altered feldspar phenocrysts associated with gold mineralization in a blue porphyry intrusion (RDXB). The interval that the sample was collected from contains 1.19 g/t Au. This unit consists of dominantly disseminated sulphide mineralization. It contains ~2% disseminated pyrite and trace to ~0.5% disseminated arsenopyrite. The SWIR spectrum for this sample has an excellent signal to noise ratio, and is considered to be of high quality (Fig. 2.8C). This sample contains major features at 1.410 µm and 2.197 µm. Large features are present at 1.905 µm, 2.107 µm, 2.343 µm and 2.434 µm. A notable doublet feature exists at 2.000 µm. The features on this spectrum indicate that the alteration mineral is illite. Sample DC06-1415-1-AR, (539632.3 mE, 6879020.3 mN, -57.3 m Elev) Hydrothermal Illite This sample was collected from drill core at a depth between 231.04 and 231.18 meters. The clay material ranges in size from 0.25 µm to 0.75 µm and is white with a reddish iridescence. It consists of clay collected from intensely hydrothermally altered feldspar phenocrysts in a mineralized interval of a lath rich porphyry intrusion (RDXL). The interval that the sample was collected from contains 1.50 g/t Au. Quartz sulphide and sulphide only veins occur in this interval. Quartz veins contain ~ 2% pyrite and ~0.5% arsenopyrite. Sulphide only veins are thin and can be discontinuous. These veins are composed of pyrite with trace amounts of arsenopyrite. Disseminated sulphides (~0.5% pyrite and trace (~0.01%) arsenopyrite is present in alteration halos associated with these veins. The SWIR spectrum for this sample has an excellent signal to noise ratio, and is considered to be of high quality (Fig. 2.8D). This sample contains major features at 1.409 µm and 2.196 µm. Large features are present at 1.905 µm, 2.105 µm, 2.343 µm 77 and 2.434 µm. A notable doublet feature exists at 2.000 µm. Together, the features on this spectrum indicate that the alteration mineral is illite. Sample DC06-1415-2-AR, (539608.7 mE, 6879033.8 mN, -110.6 m Elev) This sample was collected from drill core at a depth between 231.04 and 231.18 meters. The clay material ranges in size from 0.25 µm to 0.75 µm and is white with a reddish iridescence. The clay material was collected from intensely hydrothermally altered feldspar phenocrysts in a strongly gold mineralized lath rich porphyry intrusion (RDXL). The interval that the sample was collected contains 7.44 g/t Au. Mineralization consists of sulphide and quartz + sulphide veins and associated gold bearing disseminated sulphide mineralization. Quartz veins contain ~ %5 pyrite and 1% arsenopyrite. Sulphide veins consist of pyrite and contain trace amounts of arsenopyrite. Disseminated sulphides are composed of ~3% pyrite and trace to ~0.25% arsenopyrite. The SWIR spectrum for this sample has an excellent signal to noise ratio, and is considered to be of high quality (Fig. 2.8E). This sample contains major features at 1.410 µm and 2.200 µm. Large features are present at 1.903 µm, 2.104 µm, 2.348 µm and 2.434 µm. A notable doublet feature exists at 2.000 µm. These features indicate that the alteration mineral is illite. Sample DC06-1284-AR, (542159.4 mE, 6879514.4 mN, -2.6 m Elev) This sample was collected from drill core at a depth between 404.00 and 406.00 meters. The clay material ranges in size from 0.25 µm to 0.75 µm and is white with a reddish iridescence. The clay was collected from intensely hydrothermally altered feldspar phenocrysts in a strongly mineralized lath rich porphyry intrusion. The interval that the sample was collected contains 9.66 g/t Au. Mineralization is present in the form of sulphide bearing quartz and quartz + carbonate veins and associated disseminated sulphides. Veins contain between ~2% and ~3% pyrite and trace to 1% arsenopyrite. Disseminated style mineralization contains ~ 2% pyrite and trace to 0.75% arsenopyrite. The SWIR spectrum for this sample has an excellent signal to noise ratio, and is considered to be of high quality (Fig. 2.8F). This sample contains major features at 78 1.411 µm and 2.204 µm. Large features are present at 1.904 µm, 2.106 µm, 2.346 µm and 2.446 µm. A notable doublet feature exists at 2.000 µm. This spectrum is consistent with muscovite. 2.6.2 SHORTWAVE INFRARED INTERPRETATION Illite contains major features that, on average, lie between 1.406 µm – 1.412 µm and 2.198 µm – 2.208 µm, large features between 1.908 µm and 1.912 µm and medium features between 2.344 µm - 2.354 µm and 2.436 µm – 2.444 µm (SII, 2005). Muscovite contains a very similar distribution of features across the wavelengths detected by the Terraspec. Muscovite contains major features that range between 2.196 µm and 2.212 µm, large features between 1.400 µm and1.414 µm, and medium features between the ranges 2.344 µm – 2.352 µm and 2.438 µm – 2.446 µm. Samples DC06-1251, DC06-1280-1-AR, DC06-1280-2-AR, DC06-1415-1-AR and, DC06-1415-2-AR contain major and large features and medium features that fall within the range of illite. Sample DC06-1284-AR contains features indicative of muscovite. The slight shift in the spectrum of sample DC06-1284-AR relative to the other samples and the relatively flat profile between wavelengths 1.600 µm and 1.750 µm further supports the interpretation that this mineral is muscovite (Fig. 2.8H, 2.9). All of the samples analyzed contain a high depth absorption feature centered at between 2.020 µm – 2.120 µm, the characteristic absorption wavelength of NH4+ (SI, 2005). All of the samples are interpreted to contain NH4+. The small features at wavelengths 2.11 µm and 2.450 µm suggest that all of the samples are crystalline, and thus are more likely to have formed from a hydrothermal system rather than weathering processes (Fig. 2.9) (SI, 2005). The similarity between each of the minerals is obvious when the profiles are stacked (Fig. 2.9). 2.7 RADIOMETRIC DATING RESULTS 2.7.1 238U/206Pb (SHRIMP-RG) Zircon Geochronology Samples and Results Sample locations, descriptions and spot counts for 238U/206Pb (SHRIMP-RG) zircon geochronology are presented below. The results by spot analysis are listed in Table 2.1. Weighted mean averages for spot analyses are shown on Figure 2.10. 79 Cathodoluminescence and plane polarized light microphotograph images of zircon grains showing individual spot locations are presented in Appendix I. Sample DC06-1646 (542080.4mE, 6879258mN, 226.9m RL) Ash: This sample represents an interbedded grey, platy, 65-cm thick ash layer collected from drill hole DC06-1646 (Table 2.1A). This sample was collected from between two layers of siltstone, and is interpreted to have been deposited during a period of quiescence during sedimentation of the Kuskokwim Group. This sample produced abundant, fine grained (50-150 µm, long axis), clear colourless, stubby to prismatic zircon crystals. Relatively few of the crystals were broken or abraded suggesting that they were not significantly reworked. The zircons typically do not have complex internal zoning; however, some of the crystals have overgrown cores. The cores and overgrowths, where measured, produced consistent age estimates. Spot samples 1646-1 and 1646-7 were omitted due to high concentrations of uranium (Fig. 2.10A, Table 2.2). Spot sample 1646-10 was omitted because it produced a discordant age (Fig. 2.10A, Table 2.2). The remaining 10 spot analyses on individual crystals from this sample yield a weighted mean 207Pb corrected 206Pb/238U age of 88.8 ± 1.1 Ma. This is interpreted to reflect the age of deposition for this part of the Kuskokwim Group sedimentary rocks at Donlin Creek. Sample GP, (540737 mE, 6879016 mN, 230.0 m Elev) Ash: Sample GP was collected from the Lewis Road Gravel Pit where an exposed, discontinuous 15-25 centimeter wide, white to dark, hydrothermally altered, volcanic ash layer is exposed on the southwestern high-wall. This sample was deposited in an interlayered siltstone, shale and greywacke sequence. This sample produced a total of 10 prismatic to stubby, clear and colourless zircon grains that ranged in size between 50 µm and 100 µm microns along their long axes. All of the zircons contain obvious growth zoning, and two contain indications of complex cores. The zircons do not show any indication of significant reworking. Spot analyses GP-1 and GP-11 were omitted from the age estimate due to high concentrations of uranium (Fig. 2.10B, Table 2.2). Spot GP-1 is interpreted to have undergone post crystallization lead-loss, and spot GP-11 may reflect an inherited 80 component (Fig. 2.10B). The remaining 9 analyses yield a weighted mean, 207Pb corrected 206Pb/238U age of 88.8 ± 0.9 Ma. This is interpreted to reflect the depositional age of the Kuskokwim Group sedimentary rocks at Donlin Creek. Sample 1415, (539588 mE, 6879045 mN, 155.8 m Elev) Ash: Sample 1415 was collected from drill core taken from the 400 Prospect. This sample is a grey to light grey, hydrothermally altered volcanic ash sample that is interlayered with Kuskokwim Group siltstone. It is interpreted to have been deposited during a period of quiescence. This sample contained numerous (> 200) stubby to prismatic, clear, and colourless zircons that range in size from 50 µm to 150 µm. The zircons are intact and show limited evidence of reworking. Individual zircon grains contain distinct cores and rims and growth zones. Spot analysis 1415-6 was omitted because it produced a discordant age (Fig. 2.10C, Table 2.2). Spot analyses 1415-2, 1415-5 and 1415-10 are interpreted to contain an inherited component, and are not included in the statistical analysis (Fig. 2.10, Table 2.2). The remaining 6 analysis yield a weighted mean, 207Pb corrected 206Pb/238U age of 87.4 ±1.2 Ma, which represents the depositional age for this part of the Kuskokwim Group sediments at Donlin Creek. Sample 1280, (541375.2 mE, 6878951.3 mN, 275.4 m Elev) Crystalline Porphyry: Sample 1280 was collected from drill core from the South Lewis prospect near the eastern boundary of the 400 Prospect. This sample is a quartz feldspar porphyry intrusive (crowded crystalline porphyry), and represents the most voluminous igneous rock on the property. Zircons obtained from this sample range from stubby to prismatic, clear, colourless, and range in length from 50 µm to 400 µm (long axis). The zircon grains contain distinct cores and rims with obvious growth zones. Spot analyses 1280-9, 1280-16 and 1280-17 were omitted from the statistical age estimate (Fig. 2.10D, Table 2.2). Spot analysis 1280-9 has discordant 206Pb/238U and 207Pb/206 ages (Fig. 2.10D, Table 2.2). Spot analysis 1280-16 has an anomalously high concentration of uranium (Fig. 2.10D, Table 2.2). These values are interpreted to be the result of lead loss. Sample 1280-17 produced a discordant age and is interpreted to be an inherited grain 81 (Fig. 2.10D, Table 2.2). The remaining 13 analyses produce a weighted mean, 207Pb corrected 206Pb/238U age of 71.6 ± 0.6 Ma, which is interpreted to reflect the crystallization age for this rock. Sample 1285, (542080.4 mE, 6879258 mN, 226.9 m Elev) Fine Grain Porphyry: Sample 1285 is taken from drill core from the North Lewis Prospect. This sample is quartz feldspar porphyry intrusive (fine grain porphyry). Cross-cutting relations indicate this is the oldest texturally distinct intrusive phase on the Donlin Creek property. Zircons obtained from this sample range from stubby to prismatic, clear, colourless, and range in length from 100 µm to 400 µm (long axis). The zircon grains contain distinct cores and rims with well developed growth zones. Spot analyses 1285- 2 and 1285-7 were omitted from the statistical age estimate. Spot analysis 1280-7 contained an anomalously high 232Th/238U, yielded a much older age (125.1 ± 1) and is interpreted to be an inherited grain (Fig. 2.10E, Table 2.2). Spot analysis 1285-2 was omitted from the estimate because incorporating it increased the mean weighted statistical deviation (MSWD) (Fig. 2.10E, Table 2.2). The remaining 10 analyses yield a weighted mean, 207Pb corrected 206Pb/238U age of 71.4 ± 0.7 Ma, reflecting the crystallization age for this rock. 2.7.2 40Ar/39Ar Geochronology Samples and Results Descriptions for the 6 samples below are given in Section 2.6.1, and are also summarized in Table 2.1B. The results for argon analyses below are presented in Figure 2.11 A-F and Table 2.3. Sample DC06-1251-AR, (541722.7 mE, 6878617.8 mN, 223.2 m Elev) Hydrothermal Illite Sample DC06-1251-AR consists of hydrothermal, ammonium bearing illite (see below) obtained from an intensely altered feldspar sites in a crowded crystalline porphyry intrusive from the South Lewis Prospect (Fig. 2.11A, Table 2.3). Laser step-heating of illite from sample DC06-151-AR produced a weighted mean plateau age of 72.75 ± 0.52 Ma between steps 9 and 13, accounting for 59.7% of the 39Ar released (Fig. 82 Figure 2.6. Laser step heating 40Ar/39Ar age spectra for alteration associated with gold mineralization at the Donlin Creek Property. Accepted plateau step boxes are filled (grey) and open plateau steps are rejected. Error boxes are reported at 2σ. Inverse isochron plots are in Appendix II. Sample A) DC06-1251- AR (illite) produced a plateau age of 72.75 ± 0.52 Ma; B) Sample DC06-1284-AR (muscovite) returned a plateau age of 72.43 ± 0.42 Ma; C) Sample DC06-1280-1- AR (illite) produced a plateau age of 72.47 ± 0.43 Ma; D) Sample DC06-1280-2-AR (illite) yielded a plateau age of 71.09 ± 0.38 Ma; E) Sample DC06-1415-AR (illite) returned a plateau age of 71.88 ± 0.43 Ma; F) Sample DC06-1415-2-AR (illite) produced a plateau age of 71.97 ± 0.52 Ma. 83 2.11A, Table 1.3). The integrated age (72.75 ± 0.36) shows excellent overlap with the plateau age. Some disturbance is discernable in the low temperature steps suggesting that steps 1 through 3 experienced argon loss, and steps 4 though 8 contained excess argon (Fig. 2.11A, Table 2.3). The interpreted age for DC06-1251-AR is the plateau age of 72.75 ± 0.52 Ma. Sample DC06-1280-1-AR, (541319.4 mE, 6878989 mN, 154.9 m Elev) Hydrothermal Illite Sample DC06-1280-1-AR consists of hydrothermal, ammonium bearing illite (see below) obtained from an intensely altered, gold mineralized (3.98 g/t Au) interval of aphanitic porphyry intrusion, located in the Vortex Prospect (Fig. 2.11B, Table 2.3). This sample yielded a weighted mean plateau age of 72.47±0.43 Ma between steps 7 through 11, accounting for 74.5% of the 39Ar released (Fig. 2.11B, Table 2.3). Lower temperature steps 1 though 6 are erratic. Steps 1 and two suggest argon loss, which may be explained by the presence of additional alteration mineral phases (e.g., kaolinite). Excess argon (steps 4 though 6) may indicate that some relict feldspar is present in the sample, which would yield slightly older ages (Fig. 2.10B, Table 2.3). A well defined plateau exists for higher temperature steps 7 to 11. The integrated age for this sample (72.62 ± 0.40) is within analytical error of the plateau age; however lower temperature disturbances indicate that the integrated age is suspect. The interpreted age for sample DC06-1280-1-AR is the plateau age of 72.47 ± 0.43 Ma. Sample DC06-1280-2-AR, (541276.5 mE, 6879021.1 mN, 54.2 m Elev) Hydrothermal Illite Sample DC06-1280-2-AR consists of hydrothermal ammonium bearing illite (see below) obtained from a gold mineralized (1.19 g/t Au), hydrothermally altered blue porphyry (RDXB) from the Vortex Prospect. Laser step-heating of illite in sample DC06-1280-2-AR produced a weighted mean plateau age of 71.09 ± 0.38 Ma between steps 5 through 13 accounting for 88.7% of the 39Ar. Low temperature steps (1, 3 and 4) exhibit signs of argon loss (Fig. 2.11C, Table 2.3). This low temperature disturbance may reflect the presence of additional alteration mineral phases such as kaolinite. Step 2 yielded a slightly older age, and may indicate that some unaltered feldspar is present in the sample (Fig. 2.11C, Table 2.3). This sample yielded an integrated age of 70.88 ± 84 0.29 which is within the margin of error of the plateau age. The plateau age is the interpreted age for this sample, the low temperature disturbances indicate the integrated age a minimum estimate for the age of gold mineralization. Sample DC06-1415-1-AR, (539632.3 mE, 6879020.3 mN, -57.3 m Elev) Hydrothermal Illite Sample DC06-1415 consists of hydrothermal ammonium bearing illite (see below) obtained from a gold mineralized (1.50 g/t Au) interval of lath rich porphyry (RDXL) intrusion from the 400 prospect. Laser step-heating of this sample produced a weighted mean plateau age of 71.88 ± 0.39 Ma between steps 2 and 10, accounting for 97.3% of the 39Ar (Fig. 2.11D, Table 2.3). There was argon loss at the lowest temperature (step 1) which may indicate the presence of additional alteration mineral species in this sample (Fig. 2.11D, Table 2.3). This sample produced an integrated age of 71.72 ± 0.30 Ma which is slightly younger than the plateau age due to argon loss in step 1. The preferred age for this sample is the plateau age of 71.88 ± 0.39 Ma. Sample DC06-1415-2-AR, (539608.7 mE, 6879033.8 mN, -110.6 m Elev) Hydrothermal Illite Sample DC06-1415-2-AR consists of hydrothermal ammonium bearing illite (see below) obtained from an intensely gold mineralized (7.44 g/t Au) interval of lath rich porphyry (RDXL) intrusion from the 400 prospect. Laser step-heating of this illite sample produced a weighted mean plateau age of 71.97 ± 0.40 Ma between steps 7 and 12, accounting for 62.7% of the 39Ar (Fig. 2.11E, Table 2.3). This sample shows disturbances at low temperature (steps 1 though 6), and in the final step (step 13) (Fig. 2.11E, Table 2.3). These disturbances indicate argon loss and may be explained by the presence of an additional alteration mineral species in the sample. Sample DC06- 1415-2-AR produced an integrated age of 71.68 ± 0.25 which is slightly lower that the plateau age due to disturbances indicated by argon loss. The preferred age for this sample is the weighted mean plateau age of 71.97 ± 0.40 Ma. 85 Sample DC06-1284, (542159.4 mE, 6879514.4 mN, -2.6 m Elev) Hydrothermal Muscovite Sample DC06-1284-AR consists of hydrothermal ammonium bearing muscovite (see below) obtained from an intensely gold mineralized (9.66 g/t Au) interval of lath rich porphyry (RDXL) intrusion from the Queen prospect. Laser step heating of this muscovite sample produced a weighted mean plateau age of 72.43 ± 0.42 Ma between steps 2 and 9, accounting for 93.7% of the 39Ar (Fig. 2.11F, Table 2.3). This sample exhibits argon loss at step 1 which may indicate that there are additional alteration minerals in the sample (Fig. 2.11F, Table 2.3). Step 10 was omitted from the weighted mean plateau age because it produces a larger weighted mean standard deviation. This sample produced an integrated age 72.26 ± 0.36 which overlaps with the plateau age but has a young bias due to argon loss in step 1. The preferred age for DC06- 1284-AR is the plateau age: 72.43 ± 0.42 Ma. 2.8 ANALYSIS 2.8.1 SHORTWAVE INFRARED RESULTS Alteration analysis demonstrates a clear spatial correlation between gold mineralization and illite alteration at Donlin Creek (Fig. 2.6). Specifically, NH4+ bearing illite samples closely correspond to strongly mineralized intervals in porphyry intrusions at Donlin Creek (Fig. 2.9). The sedimentary rocks are not included in this analysis because they are very dark grey to black in colour and do not produce interpretable spectral using the Terraspec. SWIR analysis indicates that samples DC06-1251, DC06- 1280-1-AR, DC06-1280-2-AR, DC06-1415-1-AR and, DC06-1415-2-AR are crystalline, ammonium bearing illite and indicates that ages obtained from these samples reflect the timing of gold mineralization. Sample DC06-1284-AR is crystalline ammonium bearing muscovite. The distribution of muscovite alteration associated with gold mineralization is not as well constrained however the sample was obtained from an intensely gold mineralized zone (9.66 g/t Au). The close association with gold mineralization and the presence of ammonium in this sample suggests that the age reflects the timing of gold mineralization. 86 2.8.2 40Ar/39Ar Geochronology Previous studies constrain the timing of hydrothermal activity at Donlin Creek to between 67.8 ± 0.3 Ma and 73.6 ± 0.6 Ma using samples from the Lewis, Queen and Snow prospects (Grey et al., 1996; Szumigala et al., 2000). Given the early temperature step disturbances in 40Ar/39Ar results from this study, age estimates from 40K/40Ar geochronology are considered unreliable. The 40K/40Ar age estimates from Miller Bundtzen, (1994) are therefore ignored. Each of the previous 40Ar/39Ar analyses cite sericite (very fine grained white mica) as the alteration mineralogy that was dated. These studies do not report where the samples were collected with respect to gold mineralization and do not identify the alteration mineral that is analyzed. Analyses obtained using 39Ar/40Ar geochronology in this study are spatially and genetically linked to gold mineralization (Fig 2.6). All of the samples except DC06- 1284-AR consist of crystalline, hydrothermal NH4+ bearing illite. The ages obtained from the samples fall within analytical error of one another other except DC06-1280-2- AR (71.09 ± 0.38 Ma) which is youngest age obtained. The sample (DC06-1280-2-AR) was obtained from a high-grade gold interval (9.66 g/t Au) of blue porphyry (RDXB), the youngest (cross-cutting relations) texturally distinct intrusive phase. This date may represent the minimum age of gold mineralization at Donlin Creek. Sample DC06- 1284-AR consists of crystalline, hydrothermal NH4+ bearing muscovite that was collected from a high grade gold zone. This sample yielded an age of 72.43 ± 0.42 and is within analytical error of all of the illite samples except DC06-1280-AR. This result, which is similar to those from the illite altered samples, is interpreted to reflect the age of gold mineralization. Previous studies conclude that hydrothermal activity related to gold mineralization at Donlin Creek occurred between 73.6 ± 0.6 Ma and 67.8 ± 0.3 Ma (Gray et al., 1997; Szumigala et al., 2000). The 40Ar/39Ar dates obtained from crystalline NH4+ bearing illite and muscovite are genetically and spatially associated with gold mineralization indicate that gold mineralization occurred between 73.27 Ma and 70.71 Ma. Sample DC06-1280-2-AR was collected from a gold mineralized interval in the youngest texturally distinct intrusive phase and likely represents the lower age limit of gold mineralization. 87 2.8.3 U/Pb (SHRIMP-RG) Zircon Geochronology Ash layers interbedded with siltstone and shale beds collected across the Main Resource Area yield U-Pb zircon ages between 87.4 ± 1.1 Ma and 88.9 ± 1.1 Ma. This age range is interpreted to represent the depositional age of the Kuskokwim Group sediments at Donlin Creek. Intrusive rocks, including fine grained porphyry (RDF), the oldest texturally distinct phase by cross cutting relations, and crowded crystalline porphyry (RDX), the most voluminous textural phase, yielded ages between 71.7 ± 0.7 Ma and 71.4 ± 0.7 Ma. These ages are interpreted to represent cooling ages for the felsic intrusions. These ages do not overlap with the age determination obtained by Goldfarb et al. (2002). The inheritance, degrees of lead loss and subtle to strongly discordant zircon analyses associated their analysis suggests that their result (69.2 ± 0.5 Ma) underestimates the age of intrusion at the Lewis Prospect. The U/Pb zircon geochronology presented herein places tight constrains on the age of the emplacement of the felsic intrusions, and indicates that there is approximately 14.9 million years between deposition of the Kuskokwim Group sediments now underlying Donlin Creek, and emplacement of the felsic intrusions. 2.9 DISCUSSION According to two weighted mean, 207Pb corrected 206Pb/238U ages, the intrusions at Donlin Creek were emplaced at ~ 71 Ma. A single muscovite, and 5 illite plateau ages indicate that gold mineralization is coincident with intrusion at ~ 71 Ma. Previous studies (e.g., Bundtzen and Miller, 1994; Grey et al., 1997; Szumigala et al., 2000; Ebert et al., 2003a and Goldfarb et al., 2004) concluded that gold mineralization at Donlin Creek post-dates the intrusions. Younger ages obtained for gold mineralization in the Main Resource Area are inconsistent the results obtained in this study, and are interpreted to represent minimum ages for gold mineralization. The U-Pb zircon dates of 69.2 ± 0.5 and 69.9 ± 0.5 Ma obtained by Goldfarb et al. (2004) and Ebert et al. (2003b) show indications of inheritance and/or post-crystallization lead loss, and are therefore interpreted to represent minimum ages for intrusion. Szumigala et al. (2000) obtained an 40Ar/39Ar biotite plateau age of 70.3 ± 0.3 Ma from the Queen Prospect. This age is not within analytical error of the U-Pb intrusion ages determined in this 88 study. Since it is not possible to determine whether their ages represent a true crystallization age, or cooling through the mica closure temperature it is considered a minimum estimate for the age for intrusion. Cross-cutting relations indicate that gold mineralization at the Dome Prospect pre-dates mineralization in the Main Resource Area. Therefore, age constraints on gold mineralization from the Main Resource Area represent a minimum age for gold mineralization at the Dome Prospect. Szumigala et al., (2000) obtained sericite 40Ar/39Ar plateau ages between 68.0 ± 1.0 Ma and 65.1 ± 0.9 Ma from the Dome prospect. These ages are not consistent with the cross-cutting relationships or with age constraints on mineralization in the Main Resource Area. The results from the U-Pb and 40Ar/39Ar geochronology constrain place absolute ages on some of the main structural fabrics at Donlin Creek. The pre- intrusion structural evolution of Donlin Creek is dominated by north-northeast – south- southwest shortening which produced a fold and thrust geometry in the Kuskokwim Group sedimentary rocks (Figs. 2.2 - 2.4). A second, north-northwest-trending, shallowly-plunging (24ºÆ219º) fold generation, likely associated with dextral motion on the Nixon Fork faults overprints the fold and thrust geometry (see Chapter 3). Since the intrusions are not folded, these folds are interpreted to have formed prior to intrusion. The ~ 88 Ma depositional age of the Kuskokwim Group sediments provides a maximum age limit for deformation caused by north-northeast – south-southwest shortening. Shortening may have continued to least to 70.7 Ma, the youngest age (including analytical error) obtained for the felsic intrusive rocks. The intrusive rocks are not offset by thrust faults, indicating that the fold and thrust related deformation occurred prior to intrusion (see Chapter 3). Discordant, north-northeast-striking, steeply dipping dykes, and moderately southwest-dipping sills are emplaced into the folded and thrusted strata and along north-northeast-trending, low displacement normal faults. The intrusions are also cut by these normal faults indicating that these faults formed prior to intrusion and continued at least until the intrusions cooled enough to accommodate brittle faults (see Chapter 3). Gold mineralization is spatially associated with these extensional faults and their damage zones. Hydrothermal illite and sericite ages, obtained using 39Ar/40Ar geochronology, indicate that gold mineralization is temporally related to the emplacement of the intrusions and is interpreted to overlap in 89 time with north-northeast trending extensional faults. Gold-bearing veins, on average, strike parallel to the faults and intrusions but have steeper average dips and can be kinematically related according the simple normal fault model proposed by Anderson (1951) where steeply dipping veins form in the hanging walls of normal faults (see Chapter 3). 2.10 CONCLUSION Absolute age determinations, including U-Pb and 40Ar/39Ar constrain the timing of the structural fabrics observed at Donlin Creek. The timing and orientation of these fabrics are important to the structural evolution of the deposit and to the tectonic evolution of southwestern Alaska. Fossil evidence constrains the age of the Kuskokwim Group to between ca. 95-77 Ma (Miller et al., 2002). The absolute ages for deposition of the Kuskokwim sediments (~ 88 Ma), felsic intrusion (~71 Ma) and gold mineralization (~ 71 Ma) at Donlin Creek indicate a gap of approximately 14.9 million years between initial sedimentation, emplacement of intrusions, and gold mineralization. The ages constrain the timing of fold and thrust style deformation, and the overprinting north-northeast-trending folds to between ~ 88 Ma and ~ 71 Ma. The available geologic and geochronologic data also indicates that intrusion of the post- kinematic Late Cretaceous dyke swarm was closely followed in time by deposition of >30 Moz of gold. 90 REFERENCES Anderson, 1951, The Dynamics of Faulting and Dyke Emplacement with Applications to Brittan: Edinburgh. Oliver and Boyd, White Plains, NY, 1951. Bundtzen, T.K., 2004, Assessment of calcium carbonate resource potential Near Donlin Creek Project, Iditarod A-5 Quadrangle, Southwest Alaska: Barrick Gold Corp. Internal Technical Report prepared by Pacific Rim Consultants Inc. Bundtzen, T.K. and Miller, M.L., 1997. Precious metals associated with Late Cretaceous-early Tertiary igneous rocks of southwestern Alaska, In Goldfarb, R.J., and Miller, M.L., Eds. Mineral Deposits of Alaska. Economic Geology, Monograph 9, p. 242-286. Blodgett, R. B., and Clough, J. G., 1985, The Nixon Fork terrane; Part of an in place peninsular extension of the North American Paleozoic continent: Geological Society of America Abstracts with Programs, v. 17 , p. 342. Butler, R.H.W., 1987, Thrust sequences. Journal of the Geological Society, London, Vol. 144, p. 619-634. Clark, R.N., King, T, Klejwa, M., Swayze, G.A., and Vergo, N., 1990, High spectral resolution reflectance spectroscopy of minerals: Journal Geophysical Research, v. 95 (B-8), p. 12653-12680. Decker, J., Bergman, S.C., Blodgett, R.B., Box, S.E., Bundtzen, T.K., Clough, J.G., Coonrad, W.L., Gilbert, W.G., Miller M.L., Murphy, J.M., Robinson M.S., and Wallace W.K., 1994. Geology of southwestern Alaska, in Plafker, G., and Berg, H.C., eds. The Geology of Alaska. Geological Society of America. DNAG Series. G-1. p. 285-310. Ebert, S., Tosdal, R., Goldfarb, R, Dodd, S., Petsel, S., Mortensen, J., and Gabites, J., 2003a, The 25 million once Donlin Creek Gold Deposit, Southwest Alaska: a possible Link between reduced porphyry Au and sub-epithermal Au-As- Sb-Hg mineralization in Regional Geologic Framework and deposit specific exploration models for intrusion- related gold mineralization, Yukon and Alaska. [CD-ROM]. Edited by Shane Ebert (2003): MDRU Special Publication 3. The Mineral Deposit Research Group, University of British Columbia. Ebert, S., Baker, T., and Spenser, R.J., 2003c, Fluid inclusion studies at the Donlin Creek gold deposit, Alaska, possible evidence for reduced porphyry-Au to sub- epithermal transition mineralization in Regional Geologic Framework and deposit specific exploration models for intrusion-related gold mineralization, Yukon and Alaska. [CD-ROM]. Edited by Shane Ebert (2003): MDRU Special Publication 3. The Mineral Deposit Research Group, University of British Columbia. 91 Ebert, S., Miller, L., Petsel, S., Kowalczyk, P., Tucker, T.L., and Smith, M.T., 2000, Geology, mineralization, and exploration at the Donlin Creek project, southwestern Alaska: British Columbia and Yukon Chamber of Mines Special Volume 2, p. 99-114. Goldfarb, R.J., Ayuso, R., Miller, M.L., Ebert, S.W., Marsh, E.E., Petsel, S.A., Miller, L.D., Bradley, D.B., Johnson, C., and McClelland, W., 2004, The Late Cretaceous Donlin Creek Gold Deposit, Southwestern Alaska: Controls on Epizonal Ore Formation. Economic Geology v. 99, p. 643-647. Gray, J.E., Gent, C.A., Snee, L.W., and Wilson, F.H., 1997, Epithermal mercury- antimony and gold-bearing vein lodes of southwestern Alaska, Economic Geology Monograph 9, p. 287-305. Hamblin, W.K., 1965, The origin of “reverse drag” on the downthrown side of normal faults. GSA Bulletin. Vol. 76 No. 10, pp. 1145-1164. Ludwig, K.R 2003. Isoplot 3.09: A Geochronological Toolkit for Microsoft Excel. Berkeley Geochronology Center, Special Publication No. 4, 70 p. Miller, M.L., and Bundtzen T.K., 1994, Generalized geologic map of the Iditarod quadrangle, Alaska, showing potassium-argon, major oxide, trace element, fossil, paleocurrent and archaeological sample localities. U.S. Geological Survey. Miscellaneous Field Study, MF-2219-A, 48 p., scale, 1:250,000. Miller M.L., Bradley, D.C., Bundtzen, T.K., and McCelland, W., 2002, late Cretaceous through Cenozoic Strike-Slip Tectonics of Southwestern Alaska. The Journal of Geology, v. 110, p. 247-270. Moll-Stalcup, E.J., 1994. Latest Cretaceous and Cenozoic magmatism in mainland Alaska: Geologic Society of America, Geology of North America, v. G-1, p. 589-620. Moll-Stalcup, E., and Arth, J.G., 1989, The nature of the curst in the Yukon-Koyukuk province as inferred from the chemical and isotopic composition of five Late Cretaceous to early Tertiary volcanic fields in Western Alaska: Journal of Geophysical Research, B, Solid Earth and Planets, v. 94, p. 15,989-16,020. O’Dea, M., and Bartch, M., 1997, Structural controls on gold mineralization at the Donlin Creek deposit, Southwest Alaska. Etheridge Henley Williams consultant report for Placer Dome Exploration Inc. September 1997. Piekenbrock, J.R., and Petsel, S.A., 2003. Geology and Interpretation of the Donlin Creek Gold Deposit. In-house Report for NovaGold Resources, April 2003 Nokleberg, W.J., Parfenov, L.M., Monger, J.H., Norton, I.O., Khanchuk, A.I., Stone, D.B., Scotese, C.R., Scholl, D.W., and Fujita, K., 2000, Phanerozoic Tectonic Evolution of the Circum-North Pacific. U.S. Geological Survey, Professional Paper 1625. 92 Patton, W.W., Jr., Box, S.E., Moll-Stalcup, E.J., and Miller, T.P., 1994, Geology o f west-central Alaska, in Plafker, George, and Berg, H.C., eds., The Geology of Alaska, the Geology of North America: Boulder, Colorado, Geological Society of America, v. G- 1, p. 241-269. Petsel, S.A., 2001, ACMA Orientation Data Review – Donlin Creek Project. in-house Report for Nova Gold Resources. February 2002. Piekenbrock, J.R., and Petsel, S.A., 2003, Geology and interpretation of the Donlin Creek gold deposits: Juneau, Alaska, NovaGold Resources Inc., Unpublished Report, 58 p. Renne, P.R., Swisher III, C.C., Deino, A.L., Karner, D.B., Owens, T., DePaolo, D.J., 1998. Intercalibration of standards, absolute ages and uncertainties in 40Ar/39Ar dating. Chemical Geology v.145 (1–2), p. 117–152 SpecMin Pro, 2005. Spectral International Inc. Software Release: 4.0, P.O. Box 1027, Arvada, CO. Szumigala, D.J., 1996, Gold mineralization related to Cretaceous-Tertiary magmatism in west-central Alaska: A geochemical model and prospecting guide for the Kuskokwim region: Geological Society of Nevada Ore Deposits of the American Cordillera Symposium, Reno-Sparks , Nevada, April 10-13, 1995, Proceedings, p. 1317-1340. Szumigala, D.J., Dodd, S.P., and Arribas, A., Jr., 2000, Geology and gold mineralization of the Donlin Creek prospects, southwestern Alaska, in Wiltse, M.A., eds., Short Notes on Alaska Geology, Professional Report 119, State of Alaska Department of Natural Resources Division of Geological and Geophysical Surveys, p. 91-115. The Spectral Geologist (TSG). 2002, Commonwealth Scientific and Industrial Research Organization (CSIRO), Distributed by AuSpec Spectral Analysis Software: Release 5.0. Sydney Australia. Thompson, H. A., Parks, G. A. & Brown, Jr. G. E., 1999, Clays & Clay Minerals v. 47, p. 425-438. 93 3 CHAPTER 3: STRUCTURAL GEOLOGY AND GEOMETRY OF THE DONLIN CREEK GOLD DEPOSIT 3.1 GENERAL INTRODUCTION 2The Donlin Creek gold deposit is underlain by a ~ 71 Ma porphyritic bimodal dyke and sill swarm that intrudes a tilted and shortened greywacke-siltstone-shale sedimentary rock sequence. The geometry of the intrusive rocks at Donlin Creek is controlled by the oldest folds and bedding parallel thrust faults in the host sedimentary rocks of the Kuskokwim Group and a series north-northeast-striking, steeply southwest- dipping normal faults. Pre-intrusion shortening of the sedimentary rocks produced approximately east-trending thrust faults and related folds that are similar in style to other sedimentary rock dominated thin skin fold and thrust belts such as the Sevier fold and thrust belt of the western United States. Previous studies (e.g., Szumigala et al., 2000; Ebert et al., 2003a, c; Piekenbrock and Petsel, 2003; Goldfarb et al., 2004) described the general geology of the Donlin Creek. This chapter contains a summary of the regional, local and deposit scale geology, as well as a structural analysis based mainly on drill core data, and structural data from mapping historic trenches. This information is incorporated into a 3-dimensional geologic interpretation that places the geologic evolution of the Donlin Creek gold deposit in the context of the tectonic evolution of southwestern Alaska. 3.2 REGIONAL GEOLOGIC SETTING Southwestern Alaska is underlain by pericratonic passive margin sequences of Paleozoic age, exotic continental margin parallel arc terranes, and metamorphosed, Early Proterozoic sedimentary and plutonic basement rocks (Decker et al., 1994; Nokleberg et al., 2000) (Fig. 1.1). The geology of southwestern Alaska is controlled by major continental scale strike-slip faults including the north-northeast-trending Nixon- Fork fault (88-94 kilometers net dextral displacement) to the north and the Farewell- Denali fault (134 kilometers net dextral displacement) to the south of Donlin Creek 2 2 A version of this chapter will be submitted to a refereed journal for publication. MacNeil, D., and Tosdal, R., Structural geology and geometry of the Donlin Creek gold deposit. 94 (Figs. 1.1 and 1.2) (Miller et al., 2002). These faults have undergone significant displacement, and are inferred to have been accompanied at ~ 70 Ma, ~58 Ma and ~ 30 Ma by major periods of magmatism and mineralization (Miller et al., 2002). The Kuskokwim Group, which hosts the > 30 million ounce Donlin Creek gold deposit, is a post-accretionary, Upper Cretaceous overlap sequence, consisting of approximately 12 kilometers of interbedded lithofeldspathic sandstone, siltstone and shale deposited unconformably on the Paleozoic and older rocks (Decker et al., 1994) (Fig 1.2). Initial sedimentation of the Kuskokwim Group sedimentary rocks is interpreted to have occurred in an active strike-slip fault controlled marine embayment in Albian time (Decker, 1994; Miller et al., 2002). Late Cretaceous to early Tertiary plutonic and volcanic rocks intrude and/or overlie all younger rocks in this part of southwestern Alaska (Miller and Bundtzen, 1994). Widespread igneous activity in the Kuskokwim Mountains began during the final stages of Late Cretaceous sedimentation of Kuskokwim Basin, and continued into the Tertiary (Miller and Bundtzen, 1994; Decker et al., 1994). Volcanic-plutonic complexes in the Kuskokwim Basin form distinct topographic highs, and intrude and overlie Kuskokwim Group sedimentary rocks (Miller and Bundtzen, 1994). Volcanic components of these complexes consist of intermediate composition tuffs and flows that range in age from 76 to 63 Ma (Moll et al.,1981; Miller and Bundtzen, 1994; Decker et al., 1994; Bundtzen et al., 1999). Associated plutons are calc-alkaline, range from monzonite to granodiorite, and yield ages between 71 and 66 Ma (Decker et al., 1994; Bundtzen, 1994; Bundtzen and Miller, 1997). Subaerial volcanic tuffs, flows and domes are regionally extensive and dominantly andesitic in composition, but locally include dacitic, rhyolitic and basaltic components, and range in age from 71 Ma to 54 Ma (Miller et al., 2002). Although these rocks are similar in age to the volcanic-plutonic complexes and display broadly calc-alkaline trends, there are no known associated plutonic components (Miller et al., 2002). Late Cretaceous to Tertiary hypabyssal felsic to intermediate, aphanitic to porphyritic dykes, sills, and stocks occur throughout the Kuskokwim region. These units are peraluminous in composition and commonly contain garnet phenocrysts indicative of crustal melts or magma that has assimilated significant sedimentary material during ascent (Miller and Bundtzen, 1994). Late Cretaceous to early Tertiary intermediate to mafic dykes constitutes the fourth type of 95 igneous activity in the Kuskokwim region. These bodies are typically less than 3 meters wide and are ubiquitously altered to chlorite, calcite and silica (Miller and Bundtzen, 1994). Folds in the Kuskokwim basin pre-date the volcanic-plutonic complexes which range from about 76 to 63 Ma (Moll et al., 1981; Miller and Bundtzen 1994; Decker et al., 1995; Bundtzen et al., 1999; Miller et al., 2002) (Fig. 1.7). At the scale of the Kuskokwim Basin, north- and north-northeast trending fold traces can be related to dextral strike slip motion along major faults (Miller et al., 2002). Approximately east- west oriented fold traces are not consistent with dextral strike-slip tectonics. Miller et al. (2002) suggested that these folds could be related to oblique subduction under the North American plate margin, whereby deformation was partitioned at the regional scale. According to their model, dextral motion and related folding occurred in areas near Nixon Fork and Denali fault zones and north-south compression of the Kuskokwim Group occurred in the intervening areas between the faults (Fig. 1.7). Miller et al. (2002) demonstrate that dextral strike slip motion continued throughout the Tertiary to the present day. Paleomagnetic data show that between 66 and 44 Ma, southwestern Alaska rotated 44º +/- 11º counter clockwise (Coe et al., 1989). Although it is clear that strike slip tectonics continued to control the geology of southwest Alaska, the mechanism driving this is not known. Two models capable of explaining this rotation involve oroclinal bending. Grantz (1966) and Coe et al., (1989) support a model commonly called ‘megakinking’. They argued that that flexural slip and oroclinal bending explains the degree of rotation and dextral fault motion on the present day western limb of the orocline. Scholl et al. (1994) and Redfield et al. (1997) demonstrated that crustal shortening in eastern Alaska and westward-directed tectonic escape could explain oroclinal bending and the dextral strike-slip fault history of southwestern Alaska. Uncertainties related to the timing of oroclinal bending, poorly constrained movement on major faults and with the configuration and location of ridge subduction make it difficult to evaluate these models. 96 3.3 LOCAL GEOLOGIC SETTING The 8.5 kilometer long by 2.5 kilometer wide Donlin Creek property geology is underlain by a bimodal dyke swarm that intrudes fine- to coarse-grained, thin to massively bedded light grey to dark grey lithic sandstones (greywacke), thin to thickly bedded dark grey to black carbonaceous siltstone and shale, and minor interbedded conglomerates of the Late Cretaceous Kuskokwim Group (Fig. 3.1). The sedimentary rocks generally strike east to southeast and dip moderately (35º - 50º) toward the southwest. The intrusive rocks form concordant and discordant intrusive bodies and are divided into two types based upon composition. The oldest intrusive rocks on the property are thin (<3 meters thick) volumetrically minor, highly discontinuous ~ 74 Ma (Szumigala et al., 2000) porphyritic andesite to quartz andesite. A younger, ~ 71 Ma (this study; see Chapter 2), texturally variable quartz feldspar porphyry is the dominant igneous rock unit on the property. The discordant intrusions have strike lengths in excess of 1 kilometer, and reach thickness up to 100 meters. Concordant felsic igneous units are typically discontinuous and rarely exceed 200 meters in length (Fig. 3.2). Previous studies by O’Dea and Bartch (1997); Szumigala (2000), and Piekenbrock and Petsel (2003) identified many of the structural elements required to understand the structural evolution of Donlin Creek. These studies divide the structural history of Donlin Creek into two main deformation events. The oldest deformation event has been related to either: 1) north-northeast – south-southwest compression; or 2) block rotation related to extension about this same axis. All previous studies at Donlin Creek favour the former scenario where the oldest deformation is characterized by imbrication of the Kuskokwim Group along south-dipping thrust faults which have tilted the sedimentary rocks toward the southwest. North-northeast-trending, predominantly east-dipping faults cut the igneous bodies, older folds and thrust faults. O’Dea and Bartch (1997) described a bedding-parallel fault fabric in some of the concordant igneous bodies, leading them to conclude that the igneous rocks were emplaced late or after this deformation event. The second major deformation event at Donlin Creek proposed by O’Dea and Bartch (1997), and later by Piekenbrock and Petzel (2003) is marked by east-northeast – west-northwest extension. A series of 97 Figure 3.1. Simplified geology of the Donlin Creek property, prospect locations, locations and the Main Resource Area. Gold distribution in the Main Resource area (black) is shown using a 1 g/t Au cut off. 98 representing tensile veins formed during ~ north-northeast – south-southwest compression (O’Dea and Bartch, 1997; Szumigala et al., 2000). 99 north-northeast trending faults, interpreted to be normal faults, cut the intrusive rocks and typically have minor displacement (centimeters to a few meters). These studies concluded that the north-northeast dominant, steeply dipping mineralized extensional fractures and normal faults at Donlin Creek are part the shortening episode, 3.4 OUTLINE The intrusions at Donlin Creek are the main host for gold bearing veins and are therefore the most carefully mapped rock units on the property. The continuity and textural variability of the intrusive phases provided an early understanding the distribution of igneous rocks. Modeling the igneous units has identified important aspects of the geometry of this deposit. For example, modeling identified key locations where the dominant intrusive morphology is characterized by dykes (Lewis area) and areas dominated by sills (ACMA area) (Fig. 3.2). It also demonstrated that southwest- dipping sills become progressively steeper with depth in the ACMA Prospect, thereby defining an apparent fold geometry. Drill holes designed to test recumbent fold geometry revealed short intervals of overturned sedimentary rocks but did not intersect a second fold limb at depth. This study contributes to the geology of Donlin Creek by linking the geometry of the sedimentary rocks and their structural fabrics to the distribution and morphology of the porphyry intrusions and gold mineralization. Structural fabrics are used to constrain the relative timing of events and provide the framework that is necessary to understand the structural evolution of the Donlin Creek in the context of the tectonic history of southern Alaska. Absolute age dating, including U-Pb and 40Ar/39Ar geochronology are constrain the age of deposition of the Kuskokwim Group, the age of intrusion and finally the timing of gold mineralization (see Chapter 2). A surface geology map and a series of 6 intersecting cross-sections presented in this chapter (Figs. 3.2 and 3.3-3.8) illustrate key geologic elements of the Main Resource Area. The cross-sections are not balanced cross-sections because a continuous marker horizon has not yet been identified. Northeast-trending cross- sections are drawn perpendicular to the average bedding orientation, and portray the true dip of the sedimentary rocks and concordant igneous intrusions (Figs. 3.3 and 3.6- 100 Figure 3.2. Interpreted surface geology of the Main Resource Area. This map has been compiled from multiple sources including near surface drill data, surface trench mapping, and subsurface mapping including cross sections and 3 dimensional models by Placer Dome, Nova Gold Resources Inc. and, Barrick Gold Corp. 101 A 102 Figure 3.3. Cross-section 541700 illustrating the interpreted geology through the South Lewis, North Lewis and Queen Prospects. This cross section is perpendicular to bedding and shows the true dip of the sedimentary rocks. The positions of two thrust ramp anticlines are marked “A” on this figure. The “C” marks the location of a syncline. 103 104 Figure 3.4. Cross-section 687850 through the South Lewis, Vortex and Akivik Prospects. This cross section faces north-northeast and shows the true dip of the discordant igneous rocks and apparent dips for the sedimentary rocks. The igneous rocks in this section represent tabular, north-northeast striking intrusions. 105 106 Figure 3.5. Cross-section 6979800 through the Queen Prospect. This cross section is parallel to bedding (apparent dips) and perpendicular to the igneous rock (true dips). The igneous rocks shown in this section are tabular north- northeast striking intrusions. 107 108 Figure 3.6. Cross-section 539600 through the 400 Prospect. This cross section is perpendicular to bedding and shows the true dip of the sedimentary rocks. The “T” marks the position where discordant igneous rocks exploit bedding planes in folded sedimentary rocks and become concordant intrusions. 109 110 Figure 3.7. Cross-section 540100 through the ACMA and eastern edge of the 400 Prospect. This cross section is perpendicular to bedding and portrays the true dip of the sedimentary rocks. Thrust faults are numbered according to their relative timing. The back thrust is indicated by thrust fault 3 and is intruded by dykes which form a north-northeast dipping complex separating two anticlinal closures. 111 112 Figure 3.8. Cross-section 541300 through the ACMA and Akivik prospects. This cross section is perpendicular to bedding and portrays the true dip of the sedimentary rocks. Thrust faults are numbered (1-4) according to their relative timing. The letter “T” marks an important transition from discordant intrusions to concordant intrusions. 113 114 3.8). West-northwest-trending cross-sections are perpendicular to the average discordant contact orientation showing the true dip of the igneous bodies (Figs. 3.4- 3.5). Each cross-section are divided into two parts. Structural and lithologic data is presented at the same scale as the interpreted geology. Geologic interpretations incorporate all of the surface and subsurface geologic data. This is the first time this has been done at for the Donlin Creek deposit and permits the construction of a series of scenarios explaining the evolution of Donlin. 3.5 GEOLOGY OF THE MAIN RESOURCE AREA This study includes data for the entire Donlin Creek property but focuses on the Main Resource Area (Fig. 3.1). The area of interest is defined by the surface projection of gold mineralization using a 1g/t Au cut-off (as of November 01, 2008). This area includes the Aurora, 400, ACMA, Akivik, Vortex, Richelieu, South Lewis, Lewis and Queen prospects (Fig. 3.1). An interpreted surface geology map (Fig. 3.2) was constructed using data and observations from trench mapping campaigns and subsurface mapping, including previous studies (Szumigala, 2000; Ebert et al., 2000; Piekenbrock and Petsel, 2003) and 3-D models created by Westgold, Placer Dome, Nova Gold Resources Inc., and most recently Barrick Gold Corp. Lithologic and structural data is projected to the surface from depths ranging between 0 and 10 meters depending upon the depth of overburden, weathering and oxidation. The composition of sedimentary rocks is constrained from drill core intercepts, and is considered an approximation of the dominant lithology in that area. A limitation to this step is imposed by a need during exploration to simplify mixed sedimentary rocks by assigning the dominant rock to a particular interval of drill core. Nonetheless, the identified rock type is considered to be reasonably accurate. The geology of Donlin Creek is controlled by early bedding-parallel north- northeast- and south-southwest-dipping thrust faults (see Figs 3.3 and 3.6-3.8). Thrust faults are developed in or adjacent to contacts between fine- and coarse-grained sedimentary rocks. These faults typically dip moderately to steeply toward the southwest; however, out of sequence thrusts or back-thrusts are present and dip 115 toward the northeast (Figs. 3.7-3.8). Anticlines are common in the hanging walls of thrust faults (Figs. 3.2-3.3). These upright, tight to open folds have sub-horizontal fold axes and east-southeast-trending nearly vertical axial planes with no associated penetrative cleavage. Two important zones of thrust fault and thrust fault related folds occur in the Main Resource Area. The first zone trends through the southern margin of the Aurora, 400, ACMA, and East ACMA prospects (Fig. 3.2 and 3.6-3.8). The second zone passes through the Akivik, Vortex and South Lewis prospects (Figs. 3.2-3.3). In both of these areas, thrust faults separate accumulations of greywacke from siltstone and shale. Discordant, north-northeast-trending steeply south-southeast-dipping tabular intrusions are the dominant intrusive morphology (Figs. 3.2, 3.4-3.5). These intrusions are localized in the brittle, coarse grained sedimentary rocks, and are characteristic of the North Lewis and Queen prospects (Figs. 3.2 and 3.3). Concordant igneous bodies are localized in fine grained sedimentary rocks where they intrude along sedimentary rock contacts and bedding planes in siltstone and shale. Concordant intrusions are best developed in the South Lewis, Akivik and 400 prospects (Figs. 3.2 and 3.3) as well as the southern portions of the ACMA, 400 and Aurora prospects (Figs. 3.2 and 3.4- 3.6-3.8). North-northeast-trending faults that dip steeply to the southeast cut the intrusions and bedding-parallel faults. These north-northeast-trending faults are marked by fault gouge and represent low displacement (up to a few meters) normal faults which are most commonly observed in drill core, but have also been mapped in historic surface trenches. These faults do not significantly offset the geology at the scale of the Main Resource Area. They can be observed locally in outcrops along the Donlin Creek airstrip (see below). These faults have also been documented in the North and South Lewis prospects where they offset bedding concordant intrusions (Fig. 3.9). Gold at Donlin Creek is localized within and adjacent to a northeast trending, steeply southeast-dipping set of tensional veins that cut igneous and coarse-grained sedimentary rocks (O’Dea and Bartch, 1997; Szumigala et al., 2000; ; Ebert et al., 2000; Piekenbrock and Petsel, 2003). These veins are best developed within the 116 Figure 3.9. North-northeast-trending, steeply-dipping normal fault that offsets a concordant intrusion photographed in the South Lewis area. Adapted from O’Dea and Bartch, (1997). Intrusion Intrusion Intrusion Sedimentary Rock Sedimentary Rock 117 intrusions, but locally extend into the sedimentary rocks, particularly in the coarse grained sedimentary rocks. Veins are present within the faults and peripheral damage zones of north-northeast-trending normal faults. The highest gold grades at Donlin Creek are localized where mineralized veins are coincident with northeast-trending fault zones (Szumigala et al., 2000). The four main types of veins (V1-V4) are classified according to their dominant infill mineralogy (Fig. 3.10). The relative age of the veins is based upon observed cross-cutting relationships. The oldest veins in the Main Resource Area are sulphide dominated veins (V1) that range in thickness from 0.5 millimeters to 5 millimeters (Fig. 3.10A). These veins are discontinuous, and are commonly associated with an alteration envelope composed of ankerite and disseminated sulphides. Quartz- sulphide veins (V2) cross cut older sulphide veins (Fig. 3.10B, C, D). These veins range in width from < 2 millimeters to 10 centimeters. They are commonly discontinuous, and composed of variable amounts of quartz, pyrite and arsenopyrite. Ankerite + illite dominate the alteration halo with very finely disseminated pyrite and acicular arsenopyrite. A second more continuous sulphide bearing quartz vein set (V3) occurs across the Main Resource Area. These veins range from 5 millimeters to 15 centimeters. They contain variable amounts of pyrite, arsenopyrite, native arsenic, stibnite and locally contain significant quantities of calcite, orpiment and realgar (Fig. 3.10E, F). These veins typically do not have broad alteration halos, but locally have thick (up to 1 centimeter) arsenopyrite selvages. The youngest veins on the property are carbonate-only veins (V4) that range from millimeters to tens of centimeters in width. Vein compositions vary between calcite, ankerite and dolomite, and are locally auriferous (Fig. 3.10G). 3.6 STRUCTURAL FABRICS OF THE MAIN RESOURCE AREA The analysis below is based upon 2,646 structural measurements compiled from historic paper trench geology maps and 30,374 structural measurements compiled from oriented drill core (Fig. 1.7). Structural data collected from historic trench maps provides an independent check on the quality of the subsurface structural data, and therefore are not combined into a single structural database. Structural fabrics are 118 presented depending upon whether they are pre- syn- or post-intrusion. Structural fabrics are first examined for the entire property, and then in the context of a domain analysis whereby structures are examined according to the prospect zones in which they are located. Conclusions made from these analyses constrain the geology on a series of six intersecting cross sections (Figs. 3.2-3.8). These geologic cross sections are designed to illustrate the subsurface geology and three dimensional geometry of the Donlin Creek gold deposit (see below). The timing of structural fabrics presented below is bracketed by the depositional age of the Kuskokwim Group sedimentary rocks (~87 Ma), the age of the intrusions (~71Ma), and the age of the gold bearing veins (~71 Ma). The geochronology is discussed in Chapter 2. 3.6.1 Pre-Intrusion Structural Fabrics of the Main Resource Area The oldest structural fabrics within the Kuskokwim Group control the morphology of the gold-bearing intrusions, which are the most intense gold mineralized host rock. Gold is concentrated in extensional fractures and alteration halos in the intrusive rocks. Therefore, understanding the morphology and the controls on the emplacement of the intrusions is important to the economic geology of Donlin Creek. Bedding Bedding generally dips to the southwest; however, northeast-dipping beds are locally present (Fig. 3.11A and 3.12). The poles to 3,954 bedding measurements from drill core across the entire Donlin Creek property are generally distributed along a great circle girdle (girdle axis: 05ºÆ121º) with a single point maximum corresponding to the average bedding plane 125º/42º (Fig. 3.12). A domain analysis based upon prospect locations in the Main Resource Area shows the distribution of folds. Folds occur in corridors across the Main Resource area (Fig 3.2). For example, folds are present in the subsurface of the Queen prospect, and in the most southerly prospects including the Aurora, Akivik, ACMA, South Lewis and East ACMA prospects (Figs 3.12 and 3.13). Great circle patterns are notably absent in the North Lewis, Richelieu and Aurora prospects (Fig. 3.13), indicating generally homoclinal southwest dips. The orientation and distribution of folds indicate that they are an important component of the subsurface geometry and are important to the structural evolution of this deposit. Fold 119 Figure 3.10. Quartz vein examples; (A) V1 Sulphide (pyrite, arsenopyrite) veins; (B) V2 Quartz + Sulphide vein with bands perpendicular to vein walls; (C) V2 Crack and seal quartz + sulphide vein; (D) thin V2 quartz + sulphide veins in outcrop; (E) V3 Realgar veins after V1 sulphide veins; (F) V3 Quartz + native arsenic vein; (G) V2 open space filling veins. A B C D E F G 120 Figure 3.11. Bedding, fault and contact photographs at Donlin Creek. A) Southwest-dipping, interbedded siltstone, shale and greywacke from the North Lewis prospect. Oxidized bedding-parallel fault (between the arrows). B) Bedding-parallel faults in drill core highlighted by red brackets. C) Southwest- dipping sill contact in a gravel pit located in the Akivik-ACMA prospect areas. D) Irregular dyke contact from the Akivik-ACMA area; the red arrow points in the direction of strike (northeast); the black lines are parallel to bedding. 121 122 Figure 3.12. A) Poles to bedding planes and B) contoured poles to bedding for the Donlin Creek property. The poles to bedding planes form a great-circle girdle (girdle axis: 5ºÆ121º) and have a point maximum corresponding to the average bedding plane 125º/42º. 123 Figure 3.13. Poles to bedding planes shown for each prospect in the Main Resource Area. The poles to bedding planes form great circle girdles in the 400, ACMA, Akivik, and East ACMA prospects. Great circle patters in the 400, Akivik, ACMA and East ACMA prospects are consistent with folded sedimentary rocks. Great circle patterns are notably absent in the other prospects indicating generally homoclinal southwest dips. 124 125 traces shown on Figure 3.2 are interpreted based upon this data; however, their exact positions were refined based upon cross section interpretation. Brittle, bedding parallel faults are the oldest and most common fault fabric at Donlin Creek (Figs. 3.11, 3.14). Across the Donlin Creek property, these faults have an average orientation of 143º/34º. The fault zones range in thickness from millimeters to tens of meters and are filled by fault gouge that varies in colour from black or orange to white depending upon the intensity of overprinting hydrothermal alteration (Fig. 3.11A, B). With the exception of the Vortex, North Lewis and South Lewis prospect areas, the distributions of poles to fault planes in the Main Resource Area are very similar to the distribution of poles to bedding planes (compare Figs. 3.13, 3.15). The ACMA fault and bedding domain stereograms in particular correlate very well. Southwest-dipping bedding parallel faults are typical across the entire area whereas northeast-dipping faults are more common in or adjacent to shale rich areas including the 400, ACMA, Akivik, East ACMA and South Lewis prospects (Fig. 3.15). Slickenline data at Donlin Creek is difficult to interpret. Faults across the property commonly contain hydrothermally altered gouge or are highly polished and rarely display a measureable slip vector. Slickenlines observed in drill core, generally plunge sub-parallel to the dip plane of faults and along bedding (Fig. 3.16). Slickenlines data is scattered but mostly plunges down dip. In many case more than one direction is evident. 3.6.2 Syn-Intrusion Structural Fabrics of the Main Resource Area Intrusive Contacts The orientations of intrusions, particularly the discordant intrusions at Donlin Creek, are constrained by surface trenching programs and subsurface geologic modeling. At the scale of the Main Resource Area, the discordant intrusions appear as tabular bodies that are continuous over hundreds of meters. Likewise, the concordant igneous rocks generally mimic bedding at the scale of the Main Resource Area. In detail, however, contact relationships for the igneous rocks range from sharp planar contacts to highly irregular contacts. The poles to 1,125 subsurface igneous- sedimentary rock contacts are characterized by a high degree of scatter (high variance 126 Figure 3.14. A) Poles to fault planes plotted for all faults, B) faults hosted in sedimentary rocks and C) faults hosted in igneous rocks. The poles to faults hosted in igneous rocks are contoured at a different interval because there are fewer measurements and a higher degree of scatter in the dataset. The igneous rocks contain a fault fabric that is parallel to bedding and a second fault fabric that strikes north-northeast and dips moderately to the southeast. 127 Figure 3.15. Poles to fault planes for the Main Resource area. An older bedding parallel fault fabric is present across the Main Resource Area (Aurora, 400, ACMA, East ACMA, Richelieu). A north-northeast striking, steeply southeast dipping fault fabric characterizes the Vortex, South Lewis, and North Lewis prospects which contain a significant volume of north-northeast trending discordant north-northeast striking intrusive rocks. 128 129 Figure 3.16. Slickenlines for the main resource area and airstrip. The average fault and bedding orientations are plotted as plans on this diagram. In general slickenlines plunge down dip with respect to bedding and faults. 130 and low kurtosis). When contoured, point maxima are discernible that are consistent with discordant and concordant igneous contact orientations observed in historic surface trenching (Fig. 3.17A). Two point maxima are present in the contoured data (Fig. 3.17B). The first point maximum corresponds to the average contact orientation 127º/34º, and is consistent with what is expected for bedding parallel concordant igneous rocks. The second point maximum strikes north-northeast, dips moderately toward the southeast, and has the average orientation 030º/56º (Fig. 3.17B). The average orientation of the discordant igneous contacts is similar to a family of north- northeast-trending faults which suggests that the two may be related (compare Figs. 3.14A,C and 3.17). 3.6.3 Post-Intrusion Structural Fabrics of the Main Resource Area Faults and Folds – The Donlin Creek Airstrip Outcrops along the Donlin Creek airstrip provide additional information about the pre-intrusion property geology (Fig. 3.18). The Donlin Creek airstrip is underlain by Kuskokwim Group sedimentary rocks that are dominated by broad open folds with wavelengths between 5 and 20 meters and amplitudes between 1 and 2 meters (Fig. 3.18). The folds have no associated axial planar cleavage. Folds are cut by north- northeast-trending faults similar in orientation to normal faults identified through trench mapping. Locally, bedding appears to be rotated into the plane of some of these faults which indicate that these folds are at least in part affected by the normal faults (Fig. 3.18). Poles to bedding data from the airstrip outcrops form a point maximum that correspond to the average bedding orientation 118º/16º (Fig. 3.19). This is consistent in strike but shallower in dip than the average property-wide bedding data (Fig. 3.13). Poles to bedding along the airstrip (Fig. 3.19) define a great circle girdle (girdle axis 21º Æ211) oriented ~90º to the great circle girdle defined by poles to bedding on property and prospect scale stereograms (compare Figs. 3.13 and 3.19). Two generations of faults are present along the airstrip. Older, bedding parallel faults are offset by a younger dominantly north-northeast-trending, moderately to steeply southeast- and northwest-dipping faults (Fig. 3.18). Slickenlines plunge down 131 Figure 3.17. A) poles and B) contoured poles to igneous contacts from oriented drill core across the Donlin Creek property. This data has a high degree of scatter due to the curviplanar intrusive margins at the scale of drill core. Pole point maxima interpreted from the contoured data are consistent with surface mapping. Bedding parallel concordant contacts have the average orientation 127º/34º and bedding discordant contacts have the average orientation 030º/56º. 132 Figure 3.18. Structural geology of the Donlin Creek Airstrip modified from O’Dea and Bartch, 1997. 133 134 Figure 3.19. A) poles and B) contoured poles to bedding planes for the outcrops along the Donlin Creek Airstrip. A B 135 dip on north-northeast-trending fault planes and are consistent with either west- northwest – east-southeast compression (reverse faults) or tension (normal faults).Intrusions exposed in outcrops in the Lewis Hill gravel pit show no evidence of this fold pattern, supporting an interpretation that these folds pre-date the intrusions. The poles to igneous contacts contain a high degree of scatter but do contain any indication of folds of this orientation. Veins The average orientation for the four vein types (V1-V4) are similar (Fig. 3.20). The differences in average orientation for each of the vein types likely reflects the relatively low number of V1, V2 and V3 vein samples (Fig. 3.20). Poles to veins collected from sedimentary and igneous rocks have been assigned a gold grade according to the assay composite sample interval in which they occur (Fig. 3.21). Assay intervals at Donlin Creek are typically 2 meter composite samples; however, individual intervals can range from 10 centimeters to 6 meters in length. Although some variability is present, veins in mineralized sedimentary and igneous rocks contain similar point maxima (Fig. 3.21). Together, the analyses indicate there are no favourable orientations for gold mineralization within the range of vein orientations at Donlin Creek (Fig. 3.21). Average vein orientations show a slight variation in orientation depending on the composition of the host rocks in which they are found (Fig. 3.22). The average strike azimuth ranges in orientation from 006º to 034º and dip angle ranges between 64º and 79º (Fig. 3.22). Veins within the sedimentary rocks strike more northerly and have shallower dip angles than the veins found in each of the igneous rocks, with the exception of the fine grained porphyry unit which has a small sample population (N=96). The variability in strike and dip across the host rock types is explained by diffraction across contacts and between contrasting rheology during fracture propagation. The poles to veins (all types) across the property form a point maximum corresponding to the average vein orientation 024º/71º (Fig 3.23). This orientation is consistent across the Main Resource Area (compare Figs. 3.23, 3.24). The veins at Donlin Creek are interpreted to have formed during the same deformation event. 136 Figure 3.20. Poles to veins classified by composition for the Donlin Creek gold deposit. Although there are fewer recorded V1 and V3 compositions the average vein orientations are similar suggesting that all of the veins were emplaced in the same stress regime. 137 Figure 3.21. Poles to veins (undifferentiated) divided by host sedimentary rocks (A) and host igneous rocks (B) classified by gold grade of the interval the veins are found in. The average orientation of low grade gold veins is the same as for high grade gold veins indicating that veins have similar average orientations between igneous and sedimentary rocks and that there is no preferential orientation within the range of vein orientations that is favourable for gold mineralization. 138 139 Figure 3.22. The relationship between vein orientation and host rock is examined in more detail. The poles to vein data in igneous rocks strike more easterly are characterized by steeper dips compared to those in sedimentary rocks. This is interpreted to reflect diffraction during initial fracture propagation across lithologic contacts. 140 Figure 3.23. A) Poles to veins (undifferentiated) for the Donlin Creek Property, B) contoured poles to veins (undifferentiated). 141 Figure 3.24. Poles to veins (undifferentiated) by prospect area. The vein orientation at Donlin Creek is consistent across the Main Resource Area. Veins at Donlin Creek do not vary in orientation across the Main Resource Area. 142 143 Post Intrusion Faults Post intrusion faults dip steeply northeast and northwest and show minor offsets. (Fig. 3.14). The poles to faults that cut sedimentary rocks show a different distribution compared to those that cut igneous rocks (compare Figs. 3.14B,C). Bedding-parallel faults are the most common fault orientation in sedimentary rocks (Fig. 3.14B). Poles to these planes form a point maximum that corresponds to the average fault orientation 143º/33º. Fault fabrics that cut igneous rocks show a high degree of scatter and therefore have been contoured separately (Fig. 3.14C). Two point maxima are discernable in the igneous rocks. As with the sedimentary rocks, a bedding parallel faults fabric is present (Fig. 3.14C). A second point maximum is present in this data corresponding to the average fault orientation 023º/53º. Faults consistent with this orientation have been observed along the Donlin Creek airstrip, and in historic exploration trenches (e.g., Fig. 3.9). 3.6.4 Trench Geology Detailed trench mapping completed between 1996 and 2003 is concentrated in the North and South Lewis prospects, and confirms the work of O’Dea and Bartch (1997). A total of 2,646 surface structural measurements have been compiled from 49 trench maps. Structural data from trench mapping closely agrees with subsurface structural measurements collected from drill core and confirms the reliability of the subsurface structural data (Fig. 3.25). For example, poles to bedding planes in the trench dataset form a point maximum that corresponds to the average bedding plane 121º/40º, which is very similar to the average bedding plane obtained from the subsurface data which has an average bedding plane of 125º/42º (compare Figs 3.25 and 3.12). Similarly, poles to veins measured in surface trenches form a point maxima corresponding to the average plane 017º/76º which is consistent with the average vein orientation of 024º/71º obtained the subsurface data (compare Figs. 3.23 and 3.25). Fault data from trench mapping is classified differently than fault data obtained from oriented drill core. The surface trench maps classify faults as either shears (N= 598) or normal faults (N= 399) (Fig. 3.25). Shears show two well defined point maxima that correspond to the average shear orientations 110º/70º and 016º/76º. The poles to faults classified as normal faults show a similar but more diffuse distribution. Normal fault point maxima 144 Figure 3.25. Structural data compiled from historic trench mapping. This data closely compares to subsurface structural measurements collected from oriented drill core. 145 146 correspond to the average fault planes at 135º/35º and 021º/60º (Fig. 3.25). In general, these two fault classifications are similar in strike, but the faults classified as shears dip more steeply than the normal faults. The subsurface dataset does not include a normal fault classification; hence a direct comparison between the two can not be performed. Combined fault data (shears and normal fault classifications) from the trench mapping dip more steeply on average, than undifferentiated average fault data from oriented drill core (Figs. 3.25 and 3.14C). The poles to faults measured in surface trenches form maxima corresponding to the average orientations 017º/76º and 110º/70º, which are similar in strike but significantly steeper than maxima observed in the subsurface dataset, which have the average orientations 023º/53º and 143º/33º. This discrepancy may be a result of sampling bias. Steeper dipping planes are easier to measure and therefore are commonly oversampled in outcrop. Folds similar in style to those observed along the Donlin airstrip are documented in surface trenches. The presence of folds outside of the airstrip area suggests that they are evenly distributed across the property and that they are an important component of the pre-intrusion structural history of Donlin Creek. 3.6.5 Lewis Gravel Pit Geology The geology of the gravel pit on the Lewis Hill contains structural fabrics consistent with those developed in typical fold and thrust belts. Sedimentary rocks are in faulted contact with a south-dipping, concordant, crowded crystalline porphyry (RDX) intrusion (Fig. 3.26). South-dipping thrust faults cut the sedimentary rocks and contain abundant light grey to white fault gouge. Thrust faults in the gravel pit strike between 085º and 100º and dip between 66º and 70º. The exposed contact of the porphyry intrusion (Fig. 3.26, domain 1) contains pods of strongly clay altered fault gouge (Fig. 3.26, domain 6 and Fig. 3.27) suggesting that it was either emplaced along a south dipping thrust fault or the contact was reactivated causing slip post-intrusions. The sedimentary rocks between the thrust faults consist of siltstone and thinly interbedded greywacke. A series of tight to isoclinal folds occur adjacent to the south-west-dipping intrusive body and are associated with the thrust faults (Fig. 3.26, domain 3). These folds become more open toward the southern high wall where a well developed fault ramp anticline is exposed in the hanging wall of a thrust fault (Fig. 3.26, domain 4) The 147 Figure 3.26. Geology of the Lewis Hill Gravel Pit. Key geological elements include: 1) Concordant crystalline porphyry intrusion; 2) Thrust fault contact with Kuskokwim Group sedimentary rocks; 3) Isoclinally folded fine grained sediments; 4) Thrust ramp anticline developed above a southwest dipping thrust fault; 5) Location of mineralized veins (see Fig. 3.28); 6) Location of altered fault gouge (see Fig. 3.27) on dip slope of intrusion. 148 149 Figure 3.27. Isolated pods of clay altered fault gouge (inside dotted line) on dip slope of concordant intrusive body in the Lewis Hill gravel pit. Gouge contains some sedimentary rock fragments (at pencil tip). This gouge is interpreted to indicate that this intrusion was emplaced along a pre-existing thrust fault. Figure 3.28. Steeply dipping veins (arrows) and fractures with associated alteration halos in the Lewis Hill gravel pit. 150 folds in the gravel pit trend between 080º and 100º and plunge between 33º and 65º. The intrusion is cut by steeply-dipping quartz veins (Figs. 3.26, domain 5 and Fig 3.30). The quartz veins strike between 004 and 036 and dip between 75º and 90º. These veins are open space filling veins and although the outcrop is oxidized it is possible to discern clay alteration halos related to the individual veins. The alteration haloes contain very fine disseminations of pyrite and arsenopyrite. 3.7 INTERPRETATION OF STRUCTURAL FABRICS Bedding on average dips moderately to steeply toward the southwest; however, northeast-dipping stratigraphy is locally present on the property (Figs. 3.12 and 3.13). Bedding parallel thrust faults, formed during north-northeast – south-southwest shortening are the most common fault fabric at Donlin Creek (Figs. 3.14 and 3.15). Northward imbrication of the sedimentary rocks along these faults explains the moderate to steep southwest dips to bedding. The great circle distribution of poles to bedding planes across the Main Resource Area show corridors of tight, upright folds with subhorizontal fold axes. The orientation of bedding parallel faults (thrust faults) and their spatial relationship to folds supports the interpretation that these are fault related folds (thrust ramp anticlines) that formed in the hanging walls of thrust faults. The lack of penetrative axial planar cleavage implies that this shortening event occurred at a shallow level in the crust as it was not accompanied by dynamic recrystallization. North-northeast-trending normal faults form a point maximum that corresponds to the average fault plane 023º/53º. Slickenlines associated with these faults most commonly plunge down dip and are generally consistent with a normal fault geometry. North-northeast-trending normal faults cut all of the intrusions and appear along the contacts of discordant intrusions. The poles to the contacts of the discordant igneous rocks have an average contact orientation of 030º/56º which is very similar to the average orientation of the north-northeast trending faults (023º/53º) and implies that normal faults may pre-date as well as post-date the emplacement of the intrusions. It is possible that faulting and intrusion are contemporaneous. 151 Shallowly southwest-plunging folds along the airstrip and in historic trenches in the North and South Lewis prospects are oriented approximately at 90 degrees to those that formed during north-northeast – south-southwest compression (compare Figs. 3.12, 3.19). Folds are cut by north-northeast-trending faults and also appear to terminate against them. Whether there is, or is not, a kinematic linkage between the folds and north-northeast-trending faults is unclear. The folds are parallel to a family of regional folds that are related to strike slip faulting along major faults including the Nixon Fork fault to the north and the Denali Fault to the south (Fig. 1.5) (Miller et al., 2002). If the folds are related to the normal faults, it is possible that these folds are reverse drag or roll over folds (e.g., Hamblin, 1965). If these folds are related to reverse faults then they are typical of folds formed in the hanging wall of reverse faults (e.g., Ramsay, 1967). Alternatively, and more likely, the north-northeast striking faults post-date these folds and have locally affected the folds by tightening or steepening fold limbs. The youngest structural fabric at Donlin Creek is a north-northeast-trending moderately- to steeply-dipping vein swarm that hosts gold mineralization. North- northeast striking, moderately- to steeply-dipping extensional veins are spatially associated with north-northeast-trending faults and their adjacent damage zones. Veins, on average, strike parallel to these faults but dip more steeply. Open space filling, crack and seal textures are common. Coxcomb textured veins are present, particularly in carbonate and quartz-carbonate bearing veins, indicating that they formed in open space. Hydrothermal illite associated with gold mineralization was dated using Ar-Ar geochronology (this study – see Chapter 2) and constrains the age of gold mineralization to ~ 71 Ma. The distribution of veins is controlled mainly by host rock rheology. Veins are best developed in the igneous rocks. Those that form in sedimentary rocks are typically less continuous. Veins in coarse grained sedimentary rocks locally have good continuity; however, those in siltstone and shale quickly terminate in horsetail structures. The timing of the vein array is constrained by the absolute and relative age of the intrusive bodies (~71 Ma see Chapter 2). Disseminated pyrite and arsenopyrite is present in fault gouge across the property suggesting the veins formed during or after the formation of the north-northeast-trending faults. Studies by Piekenbrock and 152 Petsel (2003) demonstrate that the highest grades and most continuous gold mineralization occur where steeply-dipping northeast-striking veins occur in zones of high fault density, particularly along northeast-trending faults. This control is evident in gold assay data for the Main Resource Area (Fig. 3.29). The best gold grade are concentrated where disseminated gold mineralization haloes link separate vein swarms either along favourable host rocks or structural fabrics (Fig. 3.29). 3.8 GEOMETRY MAIN RESOURCE AREA The geometry of the Main Resource Area is controlled by the pre-intrusion geology and structural fabrics of the Kuskokwim Group sedimentary rocks. Two principal geometric elements are present: a fold and thrust geometry and the geometry of the younger intrusive. These elements are evident on a series of six intersecting cross-sections (Figs 3.2-3.8). 3.8.1 Fold and Thrust Geometry The pre-intrusion structural geometry of the sedimentary rocks is typical of thin skinned fold and thrust belts (e.g., Butler, 1987). Thrust faults present throughout the Main Resource Area generally strike ~ 125º and dip moderately to steeply (~50º) toward the southwest. Thrust fault locations are identified in the subsurface dip domains by abrupt changes in bedding attitudes. Thrust fault ramps are consistently shorter than thrust fault flats. The relative timing of motion on the thrust faults is shown for thrust faults in the ACMA area (Figs. 3.7-3.8). Thrust fault ramp anticlines lie in the hanging wall of thrust faults. (Figs. 3.2, 3.3 and 3.6-3.8). Fault ramp anticlines are open to tight folds with subhorizontal fold axes. The axial planes of these folds strike approximately parallel (~125º) to the associated thrust faults which they are associated with (Figs. 3.2, 3.3 and 3.6-3.8). Back thrusts (e.g., Butler, 1987) or out-of-sequence thrusts (e.g. Coward, 1984) are present at Donlin Creek and control the geology at depth in the 400, ACMA and East ACMA prospects (Figs. 3.6-3.8). The back thrust fault on these sections dips moderately to the northeast and accounts for the orientations of bedding planes in this area. For example, bedding near the surface dips shallowly toward the southwest, closer to the fault bedding is rotated progressively more steeply (toward the 153 Figure 3.29. Gold mineralization for the Main Resource Area. Mineralization is shown schematically on geology in the inset figure. Gold assay values are shown on a shaded relief grid (50 meter vertical exaggeration) and have been calculated using a 10 meter cell size and a search distance of 90 meters. 154 southwest) into the plane of the fault (Fig. 3.8). The sense of slip is responsible for the fold pattern in the hanging wall and explains the short intervals of overturned stratigraphy at depth (Figs. 3.6 and 3.8). The syncline developed in the siltstone and shale rich horizons near the surface of Figures 3.2-3.3 and 3.7-3.8 is likely related to the back thrust geometry. Synclines, being less common than anticlines in fold and thrust geometries, can form above blind thrusts, producing an anticline-syncline pair above the advancing fault tip (e.g., Boyer and Elliott, 1981). Synclines in fold and thrust sequences may also form in back thrust geometries such as the one drawn on Figures 3.6 to 3.8. If this process continues, a break-back thrust develops, and inhibits propagation of the individual thrust (Butler, 1987). The geometry presented in these sections suggests this process characterizes shortening in this part of the Donlin Creek Property. Thrust faults, ramp anticlines, triangle zones, and the large back thrust and syncline (Figs. 3.3 and 3.6-3.8) are continuous over 1.5 kilometers of geology in the southern portion of the Main Resource Area. Together these features are evidence that the pre-intrusion rocks underwent significant shortening during north-northeast – south-southwest compression. Cross-cutting relations between these structures and the younger intrusive rocks demonstrate that the fold and thrust architecture developed prior to porphyry intrusion (see below). Within the shortened terrane, the rheology of the sedimentary rocks strongly affects the style of deformation. The highly competent greywacke form rigid blocks that accommodate strain by brittle fracture, whereas interbedded siltstone and shale, which are less competent, accommodate shortening by folding. 3.8.2 Intrusive Geometry The distribution of the texturally distinct phases of porphyry intrusions across the Main Resource Area is relatively uniform (Fig. 3.2). North-northeast trending discordant igneous rocks are the most common morphology and account for most of the igneous rock by volume. The intrusions are generally continuous and do not show significant offset at the scale of the Main Resource Area (Fig. 3.2). Local offsets of concordant intrusions, on the scale of a few meters are mapped along northeast striking, steeply southeast-dipping faults. In drill core, sedimentary rocks adjacent to 155 the thicker discordant intrusions commonly display local changes in bedding attitudes. This occurs at distances as far away 10 meters from an intrusive body. Local bedding changes that are spatially associated with intrusions may represent minor folds generated during the emplacement of the intrusions. It is also possible that these folds pre-date the emplacement of the intrusions and may be related to either the formation of the fold and thrust geometry or to older north-northeast trending normal faults (see below). The morphology of the intrusions is strongly influenced by the pre-existing fold and thrust geometry and by the rheology of the Kuskokwim Group sedimentary rocks. For example, the North Lewis and Queen prospects are dominated by massive greywacke beds with minor siltstone and shale (Figs. 3.2, 3.4-3.5). This relatively brittle greywacke hosts abundant north-northeast striking, steeply southeast dipping discordant tabular intrusions (Figs. 3.4-3.5). These intrusions have the average orientation of 030º/56º. In contrast, the 400, ACMA and East ACMA prospects contain a relatively high volume of siltstone and shale, which is deformed mainly by folding rather than brittle failure (Figs. 3.7-3.8). Intrusions hosted by these rocks are dominantly concordant and have the average orientation 127º/34º (Figs. 3.2 and 3.6- 3.8). Discordant intrusions transition to concordant bodies in two locations in the Main Resource Area. At depth, in the ACMA prospect, where discordant igneous rocks intrude fine grained sedimentary rocks, they exploit bedding planes and attain steeply dipping concordant morphologies (Figs. 3.2, 3.6-3.8). The transition from discordant to concordant intrusions is mapped in two locations in the Main Resource Area. In the southern portion of the Aurora, 400, ACMA and East ACMA prospects, intrusions are interpreted to have been emplaced along older thrust faults (Fig. 3.2 and 3.6-3.8). The discordant intrusions become concordant where they encounter fine grained sedimentary rocks. For example, a thrust ramp anticline is developed in the siltstone and shale rich hanging wall of the 400 and ACMA prospects (see “A” Fig. 3.2 and “T” on Figs 3.6-3.8). Intrusions are focused along this fault and have exploited bedding planes and bedding contacts. The intrusions appear folded because they occupy an older fold geometry which developed in the hanging wall of a north dipping, out of sequence back thrust (Figs. 3.6-3.8). In the South Lewis, Vortex and Akivik prospects, 156 discordant tabular intrusions are hosted in sedimentary rocks dominated by coarse grained greywacke (Figs. 3.4-3.5). 3.8.3 Post Intrusion Deformation Cross-cutting relations show that north-northeast-trending, steeply dipping faults at least locally cut the intrusions. These faults also cut the low wavelength, low amplitude, shallowly southwest-plunging, north-northeast-striking folds in the sedimentary rocks along the airstrip, and in the North and South Lewis prospects. Although distribution of these faults and folds is poorly constrained, the continuity of the intrusions and distribution of bedding orientations on ~ east – west sections suggest that these structures do not significantly affect the geometry at the scale of the Main Resource Area. The intrusive rocks contain brittle faults that parallel bedding parallel faults formed during north-northeast – south-southwest compression. These faults range between 1-5 centimeters, show limited offset (few centimeters to few tens of centimeters), and contain fault gouge. O’Dea and Bartch (1997) observed these sedimentary layer parallel faults in the concordant intrusions and suggested that they constrain the timing of intrusion to syn- to post north-northeast – south-southwest compression. The poles to faults cutting intrusive rocks show that a post emplacement bedding parallel fault fabric is present and supports their conclusion (Fig. 2.14C). An alternative interpretation for this fault fabric is that faults originated as cooling joints in concordant igneous rocks that have been subsequently reactivated. If the intrusive rocks were emplaced during north-northeast – south-southwest compression, the intrusions should contain textural evidence such as ductile deformation fabrics. The intrusions do not contain ductile fabrics consistent with a syn-kinematic interpretation for their emplacement. Extensional fractures that host gold mineralization cut all other fabrics on the property. Vein textures including open space filling, coxcomb and crack and seal textures suggest that these are tensile cracks that formed before and during the hydrothermal event responsible for gold mineralization. Although the veins are kinematically compatible with the north-northeast trending extensional faults, there is insufficient evidence linking the two fabrics except for their geometric similarity. These 157 extensional fractures are best developed in the brittle igneous rocks where they show the best continuity. 3.9 SUMMARY AND INTERPRETATION OF STRUCTURAL FABRICS The depositional age of the Kuskokwim Group sedimentary rocks, determined from interbedded ash layers (see Chapter 2), and the age of the felsic igneous rocks (see Chapter 2) constrain the age of fold and thrust development and related structural fabrics to between ~ 88 and ~ 71 Ma. Illite alteration associated with gold mineralization indicates that intrusion and mineralization occurred at ~ 71 Ma (see Chapter 2). The structural elements important to understanding the structural evolution of the > 30 million ounce Donlin Creek Gold deposit are as follows. A north-northeast – south-southwest shortening event produced a series of northward-directed thrust faults with subordinate north-dipping back thrusts and related folds. The timing of this event is uncertain; however, the depositional age of the Kuskokwim Group sediments (~ 88 Ma) provides a maximum age. The northeast-trending folds are likely related to strike slip motion on the Nixon Fork Fault and over print the fold and thrust style deformation. The timing of this is also uncertain. Most of the folds recognized regionally predate emplacement of the volcanic-plutonic complexes (~ 76 – 66 Ma) (Miller et al., 2002). Mafic intrusions, possibly as old as ~ 74 Ma (Szumigala et al., 2000) were emplaced into the Kuskokwim Group and were followed by a second, more voluminous, set of felsic intrusions at ~ 71 Ma (see Chapter 2). The discordant intrusions are parallel to, and cut by, a set of north-northeast-striking, moderately southeast-dipping normal faults (O’Dea and Bartch, 1997; this study). Gold mineralization is precipitated in steeply dipping, northeast-striking fractures at ~71 Ma (see Chapter 2). The similar age of intrusion and gold mineralization suggests that these events are related. Veins hosting gold mineralization have a spatial association with and, are kinematically compatible to, the north-northeast-trending normal faults and their damage zones. Gold veins contain textures that indicate they formed in open fractures (see above). The veins, on average, strike parallel to the north-northeast-trending normal faults, but have steeper average dips. Although it is not possible to genetically link the veins and normal faults, 158 their geometries are consistent with veins that form in the hanging wall of normal faults (e.g., Anderson, 1951). The Anderson (1951) model is based on the premise that the principal tectonic stresses are either parallel or perpendicular to the earth’s surface at shallow depths and fault orientations and slip directions are primarily determined by these stress axes. In any given stress state, two orientations are favoured for the formation of fault surfaces, and will intersect along a line parallel to the intermediate principal stress (σ2). Faults are inclined approximately 30º to the maximum compressive stress axis (σ1). Since the principal stress axes are mutually perpendicular, fault surfaces are symmetrically inclined to the minimum principal stress (σ3). Near the surface of the earth, normal faults form when the maximum compressive stress, (σ1) is vertical, and σ2 and σ3 are horizontal, in this case, normal faults propagate along a plane inclined approximately 60º to the earth’s surface. The geometries of the faults and veins at Donlin Creek agree with the Anderson (1951) model (Fig. 3.30). The extensional faults at Donlin Creek on average strike 023º and dip 53º. Extensional veins, on average, strike 024º and dip 71º (Fig. 3.30). In summary, the absolute ages and cross cutting relations presented of the structural fabrics at Donlin Creek indicate the following sequence of events: 1. Deposition of the Kuskokwim Group by 88 Ma. 2. North-northeast shortening forming a fold and thrust belt (younger than ~ 88 Ma). 3. Formation of north-northeast-trending folds related to strike slip motion on the nearby Nixon Fork Fault (< ~ 88 Ma). 4. Formation of north-northeast striking steeply dipping normal faults (< ~ 88 Ma to at least 71 Ma. 5. Widespread intrusion at ~71 Ma utilizing pre-existing fault fabrics. 6. Formation of an auriferous hydrothermal system at ~71 Ma. An alternative scenario is possible whereby north-northeast-trending structural fabrics may have originated as tensional faults related to north-northeast – south- southwest compression (e.g., O’Dea and Bartch, 1997; Goldfarb et al., 2002). According to their model intrusion along the tensional fractures was facilitated by west- 159 northwest – east-southeast extension, as indicated by the normal sense of displacement. Similarly, veins are tensional and emplaced under north-northeast – south-southwest compression when the intrusions cooled sufficiently enough to host brittle fractures. This model has two main shortcomings, including: 1) it requires sinistral motion on regional faults such as the Nixon Fork and Denali, for which there is currently no evidence; and 2) there is no evidence of ductile deformation in the intrusions to support a syn-kinematic emplacement. 3.10 THE EVOLUTION OF THE DONLIN CREEK GOLD DEPOSIT 3.10.1 Regional Structural and Tectonic Fabrics Margin parallel, dextral strike-slip faults including the Nixon Fork and Denali, have been active in southwestern Alaska since ~100 Ma (Miller et al., 2002). Miller et al. (2002) demonstrated that strike-slip tectonism was active during initial basin sedimentation, induration, and subsequent folding of the Kuskokwim Basin. Strike-slip faulting was accompanied by magmatism and ore deposition at ~70, ~58 and ~30 Ma (Miller et al., 2002). Donlin Creek is situated between the Denali and Nixon Fork faults where folds in the Kuskokwim Group sedimentary rocks are consistent with ~ north-south compression (Fig. 1.5). There is also evidence for ~ north-south shortening at the ~ 71 Ma Red Devil Mercury deposit (Fig 1.2) (MacKevett and Berg, 1963). At the northern margin of the Kuskokwim Basin the ~ 71Ma Nixon Fork gold skarn contains structural fabrics consistent with dextral strike-slip motion along the Nixon Fork Fault (Miller et al., 2002) (Fig. 1.2). South of Donlin Creek on the Denali Fault, the ~71 Ma Au-Bi-W bearing Forty-seven Creek prospect contains structures that are also consistent with dextral strike-slip (Goldfarb et al., 2004 and references therein). Goldfarb et al. (2002) suggested that the complex series of structures at Donlin could be related to the interaction between coast-parallel, northward-directed terrane transport along the Canadian margin to the southeast, and approximately orthogonal active subduction under southwestern Alaska. Miller et al. (2002) suggested that the mixed signals across the Kuskokwim Basin could be explained by partitioning deformation at the regional scale (e.g., Dewey, 1980) in response to oblique subduction under the North 160 \ Figure 3.30. Schematic block diagram showing the possible relationship between faults and veins in an extensional environment. 161 American margin. Oblique subduction is consistent with plate motions during the Late Cretaceous as defined by Engebretsen et al. (1985) which show that the dextral motion on the major northeast trending faults would have been driven at least in part by the obliquely subducting Kula and/or Resurrection plate (e.g. Haeussler et al., 2003). Structural fabrics at Donlin Creek are not consistent with strike-slip tectonics and can not be adequately explained by partitioning regional deformation as suggested by Miller et al., (2002). This process may explain the presence of folds that appear kinematically unrelated at different locations in the Kuskokwim Basin, does not account for their development at the same location as is the case at Donlin Creek. 3.10.2 Regional Synthesis Miller et al., (2002) show compelling evidence to support active dextral motion on major faults since at least ~ 83 Ma but the driving mechanism is not clear. Tectonic models for southwestern Alaska are difficult to evaluate because there is a lack of consensus about plate configurations at this time (Fig. 3.31). For example, three contrasting models have been proposed for the identity of the subducting oceanic plate (or plates) and location an actively subducting spreading center at 60 Ma (e.g. Engebretson et al., 1985; Bradley et al., 1993; Miller et al., 2002) (Fig. 3.31). The approximately north-south shortening observed at Donlin Creek is present regionally, over a distance of ~ 250 kilometers, between the Denali and Nixon Fork Faults (Fig. 1.5). The distribution of these fold orientations (e.g. Miller al., 2002) suggests that they are part of regional deformation event, and are therefore important to the tectonic evolution of southern Alaska. The evolution of a back-arc basin is related to the dynamics of the related subduction system that they are part of. Compression in a back-arc is commonly attributed to the behavior of the subducting slab with respect to the upper plate in the subduction system (Heuret and Lallemand, 2006). The approximately east-west-trending folds can not be kinematically linked to dextral strike slip tectonics but are consistent with compression related to oblique subduction under southwest Alaska in the Late Cretaceous (Engebretson et al., 1985). At ~ 71 Ma igneous activity in the Kuskokwim Basin consisted of subaerial volcanism, emplacement of volcanic-plutonic complexes and a hypabyssal granite porphyry dyke swarm (Bundtzen and Miller, 1997, this study). The intrusions are commonly elliptical, 162 north-northeast-trending igneous bodies. The granite porphyry dykes, present throughout the Kuskokwim display this same structural control, and are interpreted here to have been emplaced along steeply-dipping extensional faults. Hydrothermal activity at ~ 71 Ma was spatially associated with intrusions throughout Kuskokwim Basin, and has produced the Donlin Creek, Vinasale, Nixon Fork and Forty-seven Creek gold deposits, and epithermal mercury deposits such as Red Devil (Bundtzen and Miller, 1997) (Fig. 1.2). The emplacement of north-northeast-striking, steeply southeast- dipping granite porphyry dykes across the Kuskokwim Basin, and the north-northeast structural control on many of the ~ 71 Ma mineral deposits support the interpretation that Kuskokwim basin was undergoing active north-northwest – east-southeast extension during this time. 3.10.3 Regional Tectonic Model There is evidence at Donlin Creek for three Late Cretaceous deformation events including, 1) north-northeast – south-southwest compression, forming fold and thrust style deformation after ~ 88 Ma; 2) north-northeast-trending folds interpreted to have formed by strain-partitioned shortening across north-northeast-trending regional faults during active dextral strike slip tectonics and; 3) east-southeast – west-northwest extension at ~ 71 Ma. The north-northeast – south-southwest shortening and east- southeast – west-northwest extension events are not easily reconciled with Late Cretaceous dextral strike slip tectonics. A possible mechanism for the evolution of southwestern Alaska based upon structural fabrics at Donlin Creek and throughout the Kuskokwim Basin is presented below. Redfield et al., (2007) postulate that escape tectonics, coupled with oblique subduction during Eocene time can explain the paleomagnetic variability, and apparent block rotations across southwest Alaska (e.g., Stein and Freymueller, 2002). Redfield et al. (2007) suggested that Eocene terranes along the Canadian Cordillera and Alaska moved northward as a crustal raft or orogenic float (e.g., Oldow et al., 1990). These crustal blocks ascended the margin of western North America until they encountered the apex of the modern Alaska orocline which acted like a backstop, preventing northward motion, and forcing the terranes west-southwestward toward the Aleutian and Bearing sea subduction zones (Fig 1.1). A similar scenario is plausible for the Late Cretaceous evolution of southwestern 163 Figure 3.31. Plate reconstructions for 60 Ma reproduced from A) Engbretson et al. (1985), B) Bradley et al. (1993) and C) Miller et al. (2002). Figure adapted from Miller et al. (2002). 164 165 Alaska (Fig. 2.32). Dextral oblique subduction under southwestern Alaska during the Late Cretaceous is interpreted to be responsible for the ~ east-west folds across the Kuskokwim Basin, and the fold and thrust geometry at Donlin Creek (Fig. 3.32). Similar to Eocene tectonics, oblique subduction along the western margin of North America during the Late Cretaceous could have driven terrane transport northward (Fig. 3.32). Terranes are forced westward at the intersection of the Tintina and Kaltag faults (Fig. 3.32). Westward regional extension toward the free face is possible after the terranes pass the apex of the curved Tintina and Kaltag faults (Fig. 3.32). The Kuskokwim Basin is interpreted to have occupied a favorable position for extension by ~ 71 Ma when extensional fabrics including normal faulting and intrusion and tensional gold veins are formed (Fig. 3.32). Alaska experienced ~ 44º of counterclockwise rotation during the early Tertiary (Coe et al., 1989). A rotation of this magnitude rotates the structure into the geometry observed today (Fig. 3.32C). 3.11 CONCLUSION The structural evolution of Donlin Creek is characterized by north-northeast – south- southwest shortening that produced fold and thrust style deformation after ~ 88 Ma. This produced a series of dominantly southwest-dipping thrust faults, subordinate northeast dipping back thrusts and associated thrust ramp anticlines in the Kuskokwim Group sedimentary rocks. Fold and thrust style deformation is locally overprinted by northeast-trending, low amplitude, open folds interpreted to have formed between ~ 88 and 71 Ma, related to dextral strike slip on the Nixon Fork Fault (see Chapter 2). Low displacement north-northeast striking, steeply southeast-dipping normal faults cut the thrust faults and both generations of folds. The normal faults and the older fold and thrust geometry is exploited by a ~ 71 Ma porphyritic dyke and sill swarm (see Chapter 2). Bedding discordant intrusions are emplaced along north-northeast-striking steeply- dipping normal faults. Sills are most common where intrusions invade folded fine- grained sedimentary rocks in the hanging walls of thrust faults. The intrusions are cut by ~ 71 Ma tensional gold bearing quartz and quartz + carbonate + sulphides veins which define the youngest structural fabric recognized (see Chapter 2). The distribution of veins is dictated by the rheology of the host rocks. The intrusions are the 166 Figure 3.32. Tectonic reconstruction for the pacific margin of North America between 84 and 52 Ma modified from Nokleberg et al. (2000). A) Oblique subduction causes shortening that produces fold and thrust style deformation at Donlin Creek and similarly oriented folds regionally. B) Terrane extrusion facilitates extension toward the open face; C) Post oroclinal bending bringing thrust faults (and folds) into their current configurations. 167 main host to gold mineralization; however, veins are also present in coarse grained sedimentary rocks. 168 REFERENCES Anderson, 1951, The Dynamics of Faulting and Dyke Emplacement with Applications to Brittan: Edinburgh. Oliver and Boyd, White Plains, NY, 1951. Boyer, S. E. and Elliott, D. 1982, Thrust systems: Bulletin of the American Association of Petroleum Geologists 66, p. 1196-1230. Bradley, D. C.; Haeussler, P. J.; and Kusky, T. M. 1993, Timing of early Tertiary ridge subduction in southern Alaska: U.S. Geol. Survey. Bulleton. 2068, p. 163–177. Bundtzen, T.K. and Miller, M.L., 1997, Precious metals associated with Late Cretaceous-early Tertiary igneous rocks of southwestern Alaska, In Goldfarb, R.J., and Miller, M.L., Eds. Mineral Deposits of Alaska. Economic Geology, Monograph 9, p. 242-286. Bundtzen, T. K., Harris, E. E., Miller, M. L., Layer, P. W., and Laird, G. M., 1999, Geology of the Sleetmute C-7, C-8, D-7, and D-8 quadrangles, Horn Mountains, southwestern Alaska. Alaska Div. Geol. Geophys. Surv. Rep. Investig. 98-12, 38 p., scale, 1 : 63,360. Butler, R.H.W., 1987, Thrust Sequences: Journal of the Geological Society, London, v. 144, p. 619-634. Coe, R.S., Globerman, B.R., Plumley, P.R., and Thrupp, G.A.., 1989, Rotation of central and southern Alaska in the early Tertiary: oroclinal bending by megakinking?: in Kissel, C., and Laj, C., eds. Paleomagnetic rotations and continental deformation: NATO-ASI Series, Boston, Kluwer Academic, p. 327-339. Coward, M. P., 1984, The strain and textural history of thin-skinned tectonic zones: examples from the Assynt region of the Moine thrust zone: Journal of Structural Geology v. 6, p. 89-99. Decker, J., Bergman, S.C., Blodgett, R.B., Box, S.E., Bundtzen, T.K., Clough, J.G., Coonrad, W.L., Gilbert, W.G., Miller M.L., Murphy, J.M., Robinson M.S., and Wallace W.K., 1994, Geology of southwestern Alaska, in Plafker, G., and Berg, H.C., eds. The Geology of Alaska. Geological Society of America. DNAG Series. G-1. p. 285-310. Ebert S., Tosdal, R., Goldfarb, R, Dodd, S., Petsel, S., Mortensen, J., and Gabites, J., 2003a, The 25 million once Donlin Creek Gold Deposit, Southwest Alaska: a possible Link between reduced porphyry Au and sub-epithermal Au-As- Sb-Hg mineralization in Regional Geologic Framework and deposit specific exploration models for intrusion- related gold mineralization, Yukon and Alaska. [CD-ROM]. Edited by Shane Ebert (2003): MDRU Special Publication 3. The Mineral Deposit Research Group, University of British Columbia. 169 Ebert, S., Dodd, S., Miller, L., and Petsel, S., 2003b, The Donlin Creek Au-As- Sb-Hg deposit, southwestern Alaska Ebert, S., Baker, T., and Spenser, R.J., 2003b, Fluid inclusion studies at the Donlin Creek gold deposit, Alaska, possible evidence for reduced porphyry-Au to sub-epithermal transition mineralization in Regional Geologic Framework and deposit specific exploration models for intrusion-related gold mineralization, Yukon and Alaska. [CD-ROM]. Edited by Shane Ebert (2003): MDRU Special Publication 3. The Mineral Deposit Research Group, University of British Columbia. Ebert, S., Baker, T., and Spenser, R.J., 2003c, Fluid inclusion studies at the Donlin Creek gold deposit, Alaska, possible evidence for reduced porphyry-Au to sub- epithermal transition mineralization: in Regional Geologic Framework and Deposit Specific Exploration Models for Intrusion-Related Gold Mineralization, Yukon and Alaska. [CD-ROM]. Edited by Shane Ebert (2003): MDRU Special Publication 3. The Mineral Deposit Research Group, University of British Columbia, p. Ebert, S., Miller, L., Petsel, S., Kowalczyk, P., Tucker, T.L., and Smith, M.T., 2000, Geology, mineralization, and exploration at the Donlin Creek project, southwestern Alaska: British Columbia and Yukon Chamber of Mines Special Volume 2, p. 99-114. Engebretson, D.C., 1987, Reconstructions, plate interactions, and trajectories of oceanic and continental plates in the Pacific Basin, in Monger J.W.H and Francheteau, J., eds. Circum-Pacific Orogenic Belts and Evolution of the Pacific Ocean Basin: Geodynamics Series Volume 18, International Lithosphere Program Contribution, p. 19-27. Goldfarb, R.J., Ayuso, R., Miller, M.L., Ebert, S.W., Marsh, E.E., Petsel, S.A., Miller, L.D., Bradley, D.B., Johnson, C., and McClelland, W., 2004. The Late Cretaceous Donlin Creek gold deposit, Southwestern Alaska: Controls on epizonal ore formation. Economic Geology v. 99, pp. 643-647. Grantz, A., 1966, Strike-slip faults in Alaska: U.S. Geol. Surv. Open-File Rep. 66-53, 82 p. Hamblin, W.K., 1965, The origin of “reverse drag” on the downthrown side of normal faults. GSA Bulletin. v. 76 No. 10, pp. 1145-1164. Haeussler, P. J.; Bradley, D. C.; Miller, M. L.; and Wells, R., 2000, Life and death of the Resurrection Plate: evidence for an additional plate in the NE Pacific in Paleocene- Eocene time. Geological. Society of America. Abstract. 32: A382. Heuret, A., Lallemand, S., 2005, Plate motions, slab dynamics and back-arc deformation, The Physics of earth and Planetary Interiors, v. 149, pp. 31-51. MacKevett, E. M., Jr., and Berg, H. C. 1963, Geology of the Red Devil quicksilver mine, Alaska. U.S. Geol. Surv. Bull. 1142-G, 16 p. 170 Miller M.L., and Bundtzen, T.K., 1994, Generalized geologic map of the Iditarod quadrangle, Alaska, showing potassium-argon, major oxide, trace element, fossil, paleocurrent, and archaeological sample localities: U.S. Geological Survey. Miscellaneous Field Study Map MF-2219-A, 48 p., scale, 1: 250,000. Miller M.L., Bradley, D.C., Bundtzen, T.K., and McCelland, W., 2002, Late Cretaceous through Cenozoic Strike-Slip Tectonics of Southwestern Alaska: The Journal of Geology, v. 110, p. 247-270. Moll-Stalcup, E.J., 1994, Latest Cretaceous and Cenozoic magmatism in mainland Alaska: Geologic Society of America, Geology of North America, v. G-1, p. 589-620. Moll, E.J., Silberman, M.L., and Patton, W.W., Jr. 1981. Chemistry, mineralogy, and K- Ar ages of igneous and metamorphic rocks of the Medfra quadrangle, Alaska: U.S. Geological Survey Open File Report 80-811-C, 19 p. 2 Sheets, Scale 1:250,000. Oldow, J.S., Bally, A.W., and Lallemant, H.G., 1990, Transpression, orogenic float, and lithospheric balance: Geology, v. 18, no.10, p. 991–994. O’Dea, M., and Bartch, M., 1997, Structural controls on gold mineralization at the Donlin Creek deposit, Southwest Alaska: Etheridge Henley Williams consultant report for Placer Dome Exploration Inc. September 1997. Nokleberg, W.J., Parfenov, L.M., Monger, J.H., Norton, I.O., Khanchuk, A.I., Stone, D.B., Scotese, C.R., Scholl, D.W., and Fujita, K., 2000, Phanerozoic Tectonic Evolution of the Circum-North Pacific. U.S. Geological Survey, Professional Paper 1625 p. Patton, W.W., Jr., Box, S.E., Moll-Stalcup, E.J., and Miller, T.P., 1994, Geology of west-central Alaska, in Plafker, George, and Berg, H.C., eds., The Geology of Alaska, the Geology of North America: Boulder, Colorado, Geological Society of America, v. G- 1, p. 241-269. Petsel, S.A., 2001, ACMA Orientation Data Review – Donlin Creek Project: In-house Report for Nova Gold Resources. February 2002. Piekenbrock, J.R., and Petsel, S.A., 2003, Geology and interpretation of the Donlin Creek gold deposits: Juneau, Alaska, NovaGold Resources Inc., Unpublished Technical Report, 58 p. Ramsay, J. G. 1967, Folding and Fracturing of Rocks. McGraw Hill, New York. Redfield T.F, Scholl, D.W., Fitzgerald P.G., and Beck M.E. Jr., 2007, Escape tectonics and the extrusion of Alaska: Past, present, and future: Geology, v. 35, no. 11, p. 1039- 1042. 171 Robin, P.Y.F., Jowett, E.C., 1986, Computerized density contouring and the statistical evaluation of orientation data using counting circles and continuous weighting functions: Tectonophysics, v. 121, pp. 207-223. Scholl, D.W., Stevenson, A.J., Mueller, S., Geist, E.L., Vallier, T.L., and Engebretson, D.C., 1994, Regional-scale strain partitioning leading to escape tectonism and formation of offshore arc-trench systems, Alaska-Aleutian-Bering Sea region: Geological Society of America Abstracts with Programs, v. 26, p. 136. Scholl, D. W.; Stevenson, A. J.; Mueller, S.; Geist, E.; Engebretson, D. C.; and Vallier, T. L. 1992, Exploring the notion that southeast-Asian-type escape tectonics and trench clogging are involved in regional-scale deformation of Alaska and the formation of the Aleutian-Bering Sea region, in Southeast Asia structure, tectonics and magmatism: Texas A&M University, Geodynamics Research Institute Symposium, Program and Abstracts, p. 57–61. Stein, S., and Freymueller, J.T., 2002, Plate boundary zones: American Geophysical Union Geodynamics Series, v. 30, p. 425. Szumigala, D.J., 1996, Gold mineralization related to Cretaceous-Tertiary magmatism in west-central Alaska: A geochemical model and prospecting guide for the Kuskokwim region: Geological Society of Nevada Ore Deposits of the American Cordillera Symposium, Reno-Sparks, Nevada, April 10-13, 1995, Proceedings, p. 1317-1340. Szumigala, D.J., Dodd, S.P., and Arribas, A., Jr., 2000, Geology and gold mineralization of the Donlin Creek prospects, southwestern Alaska, in Wiltse, M.A., eds., Short Notes on Alaska Geology, Professional Report 119, State of Alaska Department of Natural Resources Division of Geological and Geophysical Surveys, p. 91-115. 172 4 CHAPTER 4: SUMMARY 4.1 CONCLUSIONS The structural analyses, 3-dimensional geologic interpretations, and geochronology presented herein define the geological framework for the Donlin Creek gold deposit located in the Kuskokwim Mountains, southwest Alaska. This is the first study to provide a systematic structural analysis, and a 3-D geologic interpretation for this >30 million ounce gold deposit. A surface geology map, and a series of 6 intersecting cross-sections, based on structurally oriented diamond drill core, illustrates the 3-D geometry (Figs. 3.2-3.8). Absolute age dating, including U-Pb and 40Ar/39Ar help constrain the timing of the structural fabrics at Donlin Creek which are important to the structural evolution of the deposit, and to the tectonic evolution of southwestern Alaska. This thesis makes two significant contributions to the geology of Donlin Creek. It defines the geology and structural fabrics of the Kuskokwim Group sedimentary rocks, which host the gold mineralized granite porphyry intrusions. The geology of the sedimentary rocks explains the distribution of the intrusions, particularly the bedding concordant intrusions which were previously thought to have been folded. Previous studies (Bundtzen and Miller, 1994; Szumigala et al., 2000; Ebert et al., 2000; Goldfarb et al., 2004) concluded that mineralization post-dates intrusion. The available geologic and geochronologic data indicates that intrusion of the post-kinematic Late Cretaceous dyke swarm was closely followed in time by deposition of >30 Moz of gold The absolute ages for deposition of the Kuskokwim sediments (~ 88 Ma), felsic intrusion (~71 Ma) and gold mineralization (~ 71 Ma) at Donlin Creek indicate a gap of approximately 14.9 million years between initial sedimentation and emplacement of intrusions which were closely followed by gold mineralization at Donlin Creek. These ages constrain the timing of fold and thrust style deformation, and overprinting north- northeast trending folds to between ~ 88 Ma and ~ 71 Ma. A major conclusion of this study is that gold mineralization is temporally related to the emplacement of the granite porphyry intrusions. The structural evolution of Donlin Creek is characterized by north-northeast – south-southwest shortening that produced fold and thrust style deformation after ~ 88 173 Ma. This produced a series of dominantly southwest-dipping thrust faults, subordinate northeast-dipping back thrusts, and associated thrust ramp anticlines in the Kuskokwim Group sedimentary rocks. Fold and thrust style deformation is locally overprinted by northeast trending, low amplitude, open folds interpreted to have formed between ~ 88 and 71 Ma which are likely related to dextral strike-slip on the Nixon Fork fault. Low displacement north-northeast-striking, steeply southeast-dipping normal faults cut the thrust faults and both generations of folds. The normal faults, and the older fold and thrust geometry is exploited by a ~ 71 Ma porphyritic dyke and sill swarm. Bedding discordant intrusions are emplaced along north-northeast-striking steeply-dipping normal faults. Bedding concordant intrusions are most common where they invade folded sedimentary rocks in the hanging walls of thrust faults. The intrusions are cut by ~ 71 Ma tensional gold bearing quartz ± carbonate + sulphide veins which are the youngest structural fabric. The distribution of veins is dictated by the rheology of the host rocks. For example, the intrusions are the best host to gold mineralization; however veins are also present in coarse grained sedimentary rocks. The greywacke, siltstone and mudstone sedimentary rocks at Donlin Creek contain primary sedimentary structures including, cross-bedding, ripple marks, channel structures, and thinly interbedded quartz sands consistent with deposition in a basin marginal sedimentary environment. Fossil evidence (e.g., Bundtzen, 2004) supports this interpretation. 4.2 CRITIC OF RESULTS The results obtained from this study represent the current level of understanding of the structural geology and 3 dimensional subsurface geology of Donlin Creek. This study constrains the timing of sedimentation, emplacement of intrusions and precipitation of gold mineralization. A major limitation of this study is the lack of outcrop exposure. The orientations of the structural fabrics documented in this study are mainly constrained by stereonet point maxima obtained from oriented diamond drill core. Structural fabrics (e.g., igneous contacts) with few structural measurements are correlated with surface outcrops in order to increase confidence in their average orientations. In general, the structural fabrics defined in this study are based upon 174 large sample populations which contain robust point maxima and reflect the average orientation of a given fabric. An increase in the density of structural data obtained through infill oriented diamond drilling or open pit development is not likely to significantly change the subsurface geometry at the scale it is presented in this study (see Chapter 3). The distribution of north-northeast trending low amplitude folds observed in outcrop along the Donlin Creek airstrip is poorly constrained. Although similar folds have been recognized in historic trench mapping in the Main Resource Area there is insufficient bedrock exposure to determine whether these folds are related to faults or are related to north-northwest – south-southeast regional shortening (see Chapter 3). Determining the cause of these folds may change the relative timing and the structural evolution of Donlin Creek as it relates to this poorly constrained deformation event. The timing of sedimentation has been constrained using 3 different volcanic ash horizons within the Kuskokwim Group sedimentary rocks obtained from the Main Resource Area of the Donlin Creek deposit. These samples yield SHRIMP U-Pb zircon ages between 87.4 ± 1.1 Ma and 88.9 ± 1.1 Ma and represent a robust age determination for sediment deposition near Donlin Creek. The results of two SHRIMP U-Pb zircon ages range between 71.7 ± 0.7 Ma and 71.4 ± 0.7 Ma and represent high quality cooling ages for felsic intrusions at Donlin Creek. These results demonstrate that previous age determinations (e.g., Goldfarb et al., 2000) represent minimum estimates due to inheritance and/or post-crystallization lead loss. Finally, the timing of gold mineralization is very well constrained using Ar-Ar geochronology (see Chapter 2). A total of 6 separate analysis yield age determinations between 71.09 ± 0.38 and 72.43 ± 0.42. Together these results indicate that emplacement of the intrusions and the precipitation of gold mineralization occurred approximately 15 million years after sedimentation. The timing of intrusion and gold deposition is indistinguishable according to the analytical methods used. 4.3 EXPLORATION IMPLICATIONS Gold mineralization is closely associated with granite porphyry intrusions. The most significant thickness of granite porphyry intrusions at Donlin Creek occur where 175 bedding discordant and bedding concordant intrusions occur together. This is common in areas where north-northeast striking normal faults cut fault ramp anticlines, and have been intruded by granite porphyry dykes along normal faults and sills where intrusions are emplaced into folded sedimentary rocks. The controls on the emplacement of intrusions can be used to predict their extension along strike and at depth. This is particularly important for the bedding concordant intrusions which commonly change orientation. Regionally, the presence of fold and thrust fabrics should be considered a favourable exploration criteria in areas known to contain granite porphyry intrusions and evidence of hydrothermal activity. Thrust faults constitute planar weaknesses which can locally increase the volume of potentially gold bearing host rocks. The association of NH4+ illite and/or muscovite in granite porphyry intrusions and gold mineralization is an effective prospecting tool in the Kuskokwim Mountains. Relatively inexpensive SWIR technology can be used to identify these alteration mineral species. 4.4 RECOMMENDATIONS FOR FUTURE WORK The geologic architecture presented in this study represents a major step in understanding the geology of Donlin Creek, and provides insight into the tectonic evolution of southwest Alaska. An important limitation to this study is that a consistent marker horizon could not be identified in the Main Resource Area. A detailed sedimentary study, including stratigraphic analysis and lithogeochemistry for the Kuskokwim Group sedimentary rocks may produce a marker horizon that could be used to verify the geologic interpretations provided in this study. A marker horizon could be used to quantify the degree of shortening at Donlin Creek by allowing the construction of balanced cross-sections. Future geochronological and geological studies should aim to better understand the Dome Prospect and its possible relationship to the mineralization in the Main Resource Area. Goldfarb et al. (2004) obtained a U-Pb date on an intrusion in the Dome Prospect that yielded an age of 66.5 ± 0.5 Ma. They suggested that prolonged magmatism on the northern portion of the Donlin Creek property could explain younger 176 ages for gold mineralization, which contradict cross-cutting field relations which indicate that Dome mineralization is older than mineralization in the Main Resource area. The U-Pb analyses provided by Goldfarb et al. (2002) contain inheritance and lead-loss and represent minimum estimates for emplacement of the felsic intrusions. Currently there is insufficient evidence to support intrusions younger than ~ 71 Ma at Donlin Creek. A systematic geochemical analysis of the veins on the Donlin Creek property may provide additional insights into the genetic classification of Donlin Creek and the potential relationship between mineralization observed in the Main Resource Area and the Dome Prospect. 177 REFERENCES Ebert, S., Miller, L., Petsel, S., Kowalczyk, P., Tucker, T.L., and Smith, M.T., 2000, Geology, mineralization, and exploration at the Donlin Creek project, southwestern Alaska: British Columbia and Yukon Chamber of Mines Special Volume 2, p. 99-114. Bundtzen, T.K. and Miller, M.L., 1997. Precious metals associated with Late Cretaceous-early Tertiary igneous rocks of southwestern Alaska. In Goldfarb, R.J., and Miller, M.L., Eds. Mineral Deposits of Alaska. Economic Geology, Monograph 9, p. 242-286. Bundtzen, T.K., 2004, Assessment of Calcium Carbonate Resource Potential Near Donlin Creek Project, Iditarod A-5 Quadrangle, Southwest Alaska: Internal Technical Report prepared by Pacific Rim Consultants Inc. 65 p. Goldfarb, R.J., Ayuso, R., Miller, M.L., Ebert, S.W., Marsh, E.E., Petsel, S.A., Miller, L.D., Bradley, D.B., Johnson, C., and McClelland, W., 2004. The Late Cretaceous Donlin Creek Gold Deposit, Southwestern Alaska: Controls on Epizonal Ore Formation. Economic Geology v. 99, p. 643-647 Miller, M.L., and Bundtzen T.K., 1994, Generalized geologic map of the Iditarod quadrangle, Alaska, showing potassium-argon, major oxide, trace element, fossil, paleocurrent and archaeological sample localities. U.S. Geological Survey. Miscellaneous Field Study, MF-2219-A, 48 p., scale, 1:250,000. Szumigala, D.J., Dodd, S.P., and Arribas, A., Jr., 2000, Geology and gold mineralization of the Donlin Creek prospects, southwestern Alaska, in Wiltse, M.A., eds., Short Notes on Alaska Geology, Professional Report 119, State of Alaska Department of Natural Resources Division of Geological and Geophysical Surveys, p. 91-115. 178 5 APPENDIX I: Zircon Spot locations for SHIRMP Analyses 5.1 APPENDIX I: Zircon Spot locations for SHIRMP Analyses 179 Figure A1.1. Zircon spot locations for SHIRMP sample DC06-1646. Spot locations are shown at the same scale in cathodoluminescence (top) and plane light (bottom) photomicrographs. 1646-1 1646-2 1646-3 1646-1 1646-3 1646-4 1646-5 1646-2 180 Figure A1.2. Zircon spot locations for SHIRMP sample DC06- 1646. Spot locations are shown at the same scale in cathodoluminescence (top) and plane light (bottom) photomicrographs. 1646-4 1646-6 1646-7 1646-8 1646-4 1646-5 1646-7 1646-8 1646-9 1646-6 181 Figure A1.3. Zircon spot locations for SHIRMP sample DC06-1646. Spot locations are shown at the same scale in cathodoluminescence (top) and plane light (bottom) photomicrographs. 1646-10 1646-11 1646-12 1646-10 1646-11 182 Figure A1.4. Zircon spot locations for SHIRMP sample DC06-1646. Spot locations are shown at the same scale in cathodoluminescence (top) and plane light (bottom) photomicrographs. 1646-10 1646-11 1646-12 1646-13 1646-12 1646-13 183 Figure A1.5. Zircon spot locations for SHIRMP sample Gravel Pit. Spot locations are shown at the same scale in cathodoluminescence (top) and plane light (bottom) photomicrographs. GP-2 GP-1 GP-1 GP-2 GP-3 GP-4 184 Figure A1.6. Zircon spot locations for SHIRMP sample Gravel Pit. Spot locations are shown at the same scale in cathodoluminescence (top) and plane light (bottom) photomicrographs. GP-1 GP-2 GP-3 GP-4 GP-4 GP-3 185 Figure A1.7. Zircon spot locations for SHIRMP sample Gravel Pit. Spot locations are shown at the same scale in cathodoluminescence (top) and plane light (bottom) photomicrographs. GP-5 GP-6 GP-7 GP-4 GP-3 GP-5 GP-6 GP-7 186 Figure A1.8. Zircon spot locations for SHIRMP sample Gravel Pit. Spot locations are shown at the same scale in cathodoluminescence (top) and plane light (bottom) photomicrographs. GP-9 GP-10 GP-8 GP-9 GP-8 187 Figure A1.9. Zircon spot locations for SHIRMP sample Gravel Pit. Spot locations are shown at the same scale in cathodoluminescence (top) and plane light (bottom) photomicrographs. GP-10 GP-11 GP-10 GP-11 188 Figure A1.10. Zircon spot locations for SHIRMP sample DC07-1415. Spot locations are shown at the same scale in cathodoluminescence (top) and plane light (bottom) photomicrographs. 1415-2 1415-3 1415-4 1415-5 1415-1 1415-1 1415-2 1415-3 1415-4 1415-5 189 = Figure A1.11. Zircon spot locations for SHIRMP sample DC07- 1415. Spot locations are shown at the same scale in cathodoluminescence (top) and plane light (bottom) photomicrographs. 1415-6 1415-7 1415-8 1415-9 1415-6 1415-7 1415-8 1415-9 190 Figure A1.12. Zircon spot locations for SHIRMP sample DC07- 1415. Spot locations are shown at the same scale in cathodoluminescence (top) and plane light (bottom) photomicrographs. 1415-10 1415-10 191 Figure A1.13. Zircon spot locations for SHIRMP sample DC07-1415. Spot locations are shown at the same scale in cathodoluminescence (top) and plane light (bottom) photomicrographs. 1280-1 1280-2 1280-3 1280-1 1280-2 1280-3 1280-4 1280-5 192 Figure A1.14. Zircon spot locations for SHIRMP sample DC06- 1280. Spot locations are shown at the same scale in cathodoluminescence (top) and plane light (bottom) photomicrographs. 1280-4 1280-5 1280-6 1280-7 1280-8 1280-9 1280-1 1280-2 1280-3 1280-4 1280-5 1280-6 1280-17 193 = Figure A1.15. Zircon spot locations for SHIRMP sample DC06- 1280. Spot locations are shown at the same scale in cathodoluminescence (top) and plane light (bottom) photomicrographs. 1280-4 1280-5 1280-6 1280-7 1280-8 1280-9 1280-9 1280-8 1280-7 1280-6 1280-17 1280-17 194 Figure A1.16. Zircon spot locations for SHIRMP sample DC06- 1280. Spot locations are shown at the same scale in cathodoluminescence (top) and plane light (bottom) photomicrographs. 1280-10 1280-11 1280-6 1280-7 1280-8 1280-9 1280-12 1280-13 1280-14 1280-10 1280-11 1280-17 195 Figure A1.17. Zircon spot locations for SHIRMP sample DC06- 1280. Spot locations are shown at the same scale in cathodoluminescence (top) and plane light (bottom) photomicrographs. 1280-12 1280-13 1280-14 1280-10 1280-11 1280-12 1280-13 1280-14 196 Figure A1.18. Zircon spot locations for SHIRMP sample DC06-1280. Spot locations are shown at the same scale in cathodoluminescence (top) and plane light (bottom) photomicrographs. 1280-14 1280-15 1280-13 1280-14 1280-15 1280-14 197 Figure A1.19. Zircon spot locations for SHIRMP sample DC06- 1280. Spot locations are shown at the same scale in cathodoluminescence (top) and plane light (bottom) photomicrographs. 1280-16 1280-16 198 Figure A1.20. Zircon spot locations for SHIRMP sample DC06-1285. Spot locations are shown at the same scale in cathodoluminescence (top) and plane light (bottom) photomicrographs. 1285-1 1285-2 1285-3 1285-4 1285-5 1285-6 1285-1 1285-2 1285-3 1285-4 1285-5 1285-6 1285-5 1285-5 199 Figure A1.21. Zircon spot locations for SHIRMP sample DC06-1285. Spot locations are shown at the same scale in cathodoluminescence (top) and plane light (bottom) photomicrographs. 1285-1 1285-2 1285-3 1285-4 1285-5 1285-6 1285-7 1285-8 1285-7 1285-8 1285-9 1285-10 1285-11 200 Figure A1.22. Zircon spot locations for SHIRMP sample DC06-1285. Spot locations are shown at the same scale in cathodoluminescence (top) and plane light (bottom) photomicrographs. 1285-12 1285-11 1285-10 1285-12 201 5.2 APPENDIX II: Reverse Isochron Plots for 40Ar-39Ar Geochronology 202 Figure A2.1. 40Ar-39Ar plateau and inverse isochron age determinations for sample DC06-1251-AR. Sample DC06-1251-AR consists of hydrothermal ammonium illite associated with gold mineralization. 203 204 Figure A2.2. 40Ar-39Ar plateau and inverse isochron age determinations for sample DC06-1284-AR. This sample consists hydrothermal ammonium bearing muscovite associated with gold mineralization. 205 206 Figure A2.3. 40Ar-39Ar plateau and inverse isochron age determinations for sample DC06-1280-AR. This sample consists of hydrothermal ammonium illite associated with gold mineralization. 207 208 Figure A2.4. 40Ar-39Ar plateau and inverse isochron age determinations for sample DC06-1280-2-AR. This sample consists of ammonium bearing illite associated with gold mineralization. 209 210 Figure A2.5. 40Ar-39Ar plateau and inverse isochron age determinations for sample DC06-1415-1-AR. This sample consists of ammonium illite associated with gold mineralization. 211 212 Figure A2.6. 40Ar-39Ar plateau and inverse isochron age determinations for sample DC06-1415-2-AR. This sample consists of ammonium illite associated with gold mineralization. 213