Evolution of the Hydrothermal Fluids Inferred from the Occurrence and Isotope Characteristics of the Carbonate Minerals at the Pogo Gold Deposit, Alaska, USA

Abstract

Pogo is identified as a deep-seated, intrusion-related gold deposit. Carbonate minerals have a close spatial relationship to hydrothermal gold mineralization in all of its principal ore zones. The carbon and oxygen isotopic ratios of carbonate minerals (siderite, ankerite, and calcite) present within the deposit illustrate the isotopic evolution of the ore-forming fluid. The initial hydrothermal fluid phase is interpreted to be magmatic in origin. The fluid evolution was characterized by a gradual decrease in δ¹⁸O and a slight increase in δ¹³C with decreasing temperature. The dominant carbon-bearing species was CO₂, with methane introduced sporadically. Siderite is associated with early-stage mineralization and occurs with ankerite in main-stage ore assemblages. Calcite is recognized in the later stages of mineralization. Gold in the Pogo deposit occurs as native gold, Au-Bi-Te minerals, inclusions in sulfide minerals, or as “invisible gold”. The latter is found in pyrite, chalcopyrite, arsenopyrite, and quartz, based on ion microprobe analysis. The presence of invisible gold in these minerals has significant metallurgical implications for gold processing at the Pogo mine.

1. Introduction

The Pogo deposit is a high-grade gold deposit located in the arc-shaped Tintina Gold Province (TGP, Figure 1). The TGP extends over 1200 km across Alaska (USA) and the Yukon (Canada) and hosts a series of world-class gold deposits that are predominantly Early Cretaceous to Eocene in age (e.g., [1,2,3,4]). Pogo has been classified as an intrusion-related gold deposit [4,5,6,7,8], although some argue that its characteristics are more typical of orogenic gold systems [9,10,11,12].

The Pogo gold mine has operated continuously since 2006. Gold production for the financial year 2024 (ending 30 June 2024: FY24) totaled 8.7 tons [13]. Total (measured, indicated, and inferred) mineral resources and (proved and probable) ore reserves amount to 2.05 × 10⁷ tons at 10.1 g/t Au and 5.9 × 10⁶ tons at 8.0 g/t Au, respectively [13].

The low salinity and high CO₂ contents of intrusion-related gold systems differ from the characteristics of most other magmatic gold environments [12]. Analysis of ore-related carbonates can provide valuable insights into the nature of the hydrothermal mineralizing system. Since the carbon and oxygen isotopic ratios of carbonates are independently arranged, the trend of the two-isotope set can provide valuable information regarding potential ore precipitation mechanisms (e.g., [14]).

The Pogo ore assemblage includes carbonate minerals such as calcite, the ferroan dolomite–ankerite series, and siderite. Stable isotopes of the carbonate minerals can predict the genesis and evolution of the gold mineralizing fluids. The carbon and oxygen isotope ratios of carbonate minerals in this study were determined by isotope ratio mass spectrometry (IRMS). Understanding the nature of the Pogo hydrothermal system on the basis of isotope data can help to map the extent of the mineralization footprint and guide the discovery of new mining areas.

Gold in the Pogo deposit occurs as native gold, Au-Bi-Te minerals, inclusions in sulfide minerals, or as “invisible gold”. The invisible gold is further divided into solid solution in sulfides and gold nanoparticles (NPs; <100 nm grains, [15,16]). Microbeam techniques have been applied across a variety of deposits to investigate the presence of gold in sulfides (e.g., [17,18,19,20,21,22,23]). The application of the secondary ion mass spectrometry (SIMS) or ion microprobe technique to the Pogo deposit allows for high sensitivity microanalysis, making it possible to elucidate the nature of the gold in the ore-related minerals. These determinations have significant metallurgical implications for gold processing at the Pogo mine.

2. Geology and Mineralogy

The TGP is bound by the Tintina fault system to the north and the Denali fault system to the south (Figure 1). The Pogo deposit is located along the Goodpaster River, 145 km southeast of Fairbanks (inset in Figure 2). Gold-bearing quartz veins of the Pogo deposit are hosted by gneisses of the late Proterozoic to mid-Paleozoic Yukon–Tanana Terrane (e.g., [6,24]) and a suite of mid-Cretaceous felsic to intermediate plutons and dikes [25]. A simplified geologic map of the Pogo mine area is shown in Figure 2.

The Pogo deposit can be subdivided into five main mining areas: Liese, South Pogo, East Deep, North Zone, and Central Lodes (Figure 3a). Several peripheral mineral resources have also been identified at the Goodpaster and 4021 prospects. The Liese, and South Pogo vein systems are located south of the Liese Creek intrusive complex. Each area comprises a series of stacked ore bodies that dip shallowly (25° to 30°) to the northwest (Figure 3b, [6]). A mean ¹⁸⁷Re-¹⁸⁷Os molybdenite age of 104.2 ± 1.1 Ma was obtained for the Liese veins [26]. An ⁴⁰Ar/³⁹Ar age of 92.7 ± 0.3 Ma, obtained for a mafic dike that cuts the uppermost (L1) vein structure, constrains the minimum age of gold mineralization [7]. This age is comparable to the 91.2 ± 0.4 Ma and 91.7±0.4 Ma ⁴⁰Ar/³⁹Ar ages obtained for Liese Zone alteration assemblages [6].

The East Deep mining area contains two stacked quartz veins (E0.5 and E1) with similar orientations and mineral assemblages to those of the Liese veins. The East Deep veins are interpreted as extensions of the Liese vein system, now separated by the post-mineral Liese Creek intrusive complex [25]. Three principal NNW-striking, subvertical mineralized veins are mined at the North Zone (NZ1, NZ2, and NZ3 veins from west to east). The Central Lodes area is characterized by an extensive domain of sheeted vein arrays hosted in granitic orthogneiss.

Gold mineralization in all mining areas is associated with varying proportions of Fe-As-S minerals including pyrite, arsenopyrite, loellingite, and pyrrhotite, accompanied by minor bismuthinite and chalcopyrite [6]. Aggregates of pyrite or arsenopyrite occur in a matrix of quartz locally intergrown with Mg–Fe carbonate [7]. The Au–Bi assemblage, associated pyrite, arsenopyrite, and pyrrhotite, is typical of plutonic-related gold deposits [5]. Thirty-five intrusive rocks from the vicinity of the Liese zone were collected for major and trace element analyses [6]. They fall within the “I-type” field of Chappell and White [27] and belong to the “ilmenite series” of Ishihara [28]. These compositions are characteristic of intrusion-related gold mineralization in Central Alaska [6].

A detailed study of Pogo hydrothermal alteration assemblages by Smith et al. [6] included descriptions of carbonates relevant to this study. Very fine-grained silica with fine-grained, disseminated pyrite and/or arsenopyrite is typically associated with sericite + ferroan dolomite + chlorite ± quartz alteration, and the rocks are commonly fractured and sheared [6].

3. Materials and Methods

3.1. Samples

The carbonate-bearing samples used in this study were collected from the Liese, East Deep, and North Zone vein systems and from mineralized drill core samples in the general Pogo mining area (

Table 1. Summary table of the Pogo samples used in this study.

No.	Sample Name	Cal	Ank	Sid	Vein or Area, Location (or Drill Depth), Description	Coordinate				δ13C of
						X	Y	Z	REE	Graphite
1	2011032201				L1 Blk 10, 1760C6E, graphite schist	1813630	3820717	1777	●	●
2	2011032202		●		L1 Blk 10, 1760C6E, qtz vein with ankerite and sulfide	1813630	3820717	1777
3	2011032205		●		L1 Blk 08, 1558C9W, qtz vein with ankerite and sulfide	1811272	3820623	1579
4	2012080709	●			East Deep, 1300EXP, qtz vein with calcite and sulfide	1814935	3821196	1257
5	2012080711		●	○	North Zone, NZ02-1331, carbonate dissemination in altered intrusive	1812013	3822688	1318
6	2012081001	●	○		New Portal Bench, carbonate vein in weathered diorite (ground)	1814453	3821485	2090
7	2012081003		●		North Zone, drill core 12-685, 635′, qtz–ank vein	1814677	3823024	2081	●
8	2012081004	●			Drill core 12u130, 305.5′, thin qtz–cal vein with sulfide	1811122	3821419	887
9	2012081005	●			North Zone, drill core 12-685, 697′, qtz vein with calcite and sulfide	1814670	3823017	2020
10	2012081006	●			North Zone, drill core 12-598, (Loc 4021), 66′, calcite dissemination in altered rock	1823121	3814731	3108	●
11	2012081011		●		L1, Ankerite dissemination in altered intrusive	1811995	3822640	1318
12	2012081108		●		L1 BLK10, 1320, S3, qtz vein with ankerite and sulfide	1812948	3821164	1336	●
13	2014081806				X-Vein, North Zone, graphite schist with pyrite crystal	1812801	3822967	895	●	●
14	2014082206	●			East Deep, 1020Cut#2E, E1, qtz vein with calcite and sulfide	1815072	3821642	1050
15	2014082207	●			Estimated L1 extension, qtz–cal vein (ground)	1818057	3817256	822
16	2014082208	●	○		40 cm above 2014082207, qtz–cal vein (ground)	1818057	3817256	822
17	2014082210	○	●		Estimated L1 extension, qtz vein with brown carbonate (ground)	1818057	3817256	822
18	2014082211	●	○		Estimated L1 extension, qtz–cal vein (ground)	1818057	3817256	822
19	2014082212	●	○		Estimated L1 extension, carbonate veinlet (ground)	1818057	3817256	822
20	2014082213	●			Estimated L1 extension, just below 2014082207, qtz–cal vein (ground)	1818057	3817256	822
21	2014082215	●			Qtz–Cal vein (ground)	1816789	3818489	822
22	2014082216	●			Qtz–Cal vein with sulfide (ground)	1816789	3818489	822
23	2014082304-2				Burn, fracture-filling qtz vein with arsenopyrite and visible Au	1802710	3818930	511	●
24	M000002a	●			Hill 4021 prospecting area, drill core 13-610, qtz–cal vein in altered diorite	1823453	3815885	2807	●
25	M000004a	●			Hill 4021 prospecting area, drill core 13-610, qtz–cal vein in altered diorite	1823445	3815896	2784
26	M000013	●			East Deep East, drill core 13-678, calcite dissemination in EDE	1817768	3820455	155
27	M000015a	●			South Pogo, drill core 13-759, qtz–cal vein	1814301	3819357	1745
28	M000034a	●			NZ03, drill core 13EXP007, qtz–cal vein between NZ04-NZ03	1812626	3823140	134	●
29	M000068c			●	L1 NorthWest, drill core 12-691, qtz–sid vein with sulfide	1808261	3822132	101
30	10DL007A	●			East Deep East, 10-507, 841′, cal vein	1817712	3821658	1395
31	10DL007B	●	●		East Deep East, 10-507, 841′, qtz–cal–ank vein	1817712	3821658	1395
32	10DL008	●			East Deep, 10-508, 560′, crystalline calcite with rhodochrosite in a drusy cavity	1815203	3821563	1442
33	10DL012A	●			L1, 10-517, 371′, cal vein	1809548	3822370	1475
34	10DL012B	●	●		L1, 10-517, 371’, cal–ank vein	1809548	3822370	1475
35	11DL011A	●	●		Under the L1, 10u129, 108’, qtz–cal–ank vein with muscovite	1811935	3822586	1019
36	11DL011B		●		Under the L1, 10u129, 108′, qtz–ank vein	1811935	3822586	1019
37	11DL011DOS		●		Under the L1, 10u129, 108′, qtz–ank vein with DOS alterataion	1811935	3822586	1019
38	11DL012A	●	●		Under the L1, 10u129, 144′, cal–ank vein	1811935	3822596	1168
39	11DL012B	●	●		Under the L1, 10u129, 144′, qtz–cal–ank vein	1811935	3822596	1168
40	11DL012C	●	●		Under the L1, 10u129, 144′, cal–ank vein	1811935	3822596	1168
41	11DL025	●			Under the L3, 11u133, 120′, qtz–cal vein	1812390	3820875	-92
42	11DL028	●	●		Above the ED01, 11-573, 528′, qtz–cal–ank vein	1815763	3823111	1404
43	11-584A			●	East Deep area, 11-584, 980′, qtz–sid vein with chlorite and sulfide	1816137	3822155	1644
44	11-584B	●			East Deep area, 11-584, 548′, qtz–cal vein	1816140	3822098	2109
45	11-584C		●		East Deep area, 11-584, 430′, qtz–ank vein with sulfide	1816141	3822081	2240
46	11-584D	●			East Deep area, 11-584, 430′, cal vein	1816141	3822081	2240
47	11u201A	●	●		Between L1 and NZ02, 11u201, 60′, qtz–cal–ank vein	1811795	3823003	1038
48	14-716_278.2ft		○	●	Qtz–Sid vein	1813519	3818448	2226
49	14-742_474.6ft		●		SP-1a, qtz–ank vein with sulfide	1813107	3818234	1850
50	14-742_1027.2ft	●			SP-1b, qtz–cal vein	1813145	3818381	1319
51	14-752_1188.6ft			●	SP-1b, qtz–sid vein with muscovite	1812047	3817786	1252	●
52	14-765_589.9ft			●	SP-1a, qtz–sid vein with sulfide	1811963	3818573	1566
53	14-765_824.3ft			●	SP-1b, qtz–sid vein with sulfide	1812030	3818527	1346
54	14-771_876.1ft	●			SP-1b?, qtz–cal vein	1811607	3818969	1488
55	14-773_875.8ft		●		SP-1a?, qtz–ank vein with sulfide	1811862	3819396	1455
56	14-795_1242.8ft	●			X-Vein, qtz–cal vein	1812263	3823758	1132	●
57	14-795_1434.4ft		●		NZ04, qtz–ank vein with sericite	1812186	3823706	964
58	14-805_857.1ft	●			X-Vein, qtz–cal vein	1811981	3823809	1420
59	14-805_1008.9ft	●			NZ04, Qtz vein with calcite and sulfide	1811980	3823808	1419
60	14-805_1020ft	●			NZ04, qtz–cal vein	1811894	3823760	1292
61	15-882_316.5	●			Star, Silicified rock with calcite	1819343	3812334	2633
62	15-882_833.5	●			Star, qtz–cal vein	1819148	3812519	2192
63	15-884_500.5	●			Star, qtz–cal vein	1819394	3811999	2469
64	15-896_2500.4		●		East Deep, qtz–ank vein with sericite and sulfide	1814790	3823129	487
65	15-896_2581.3	●			East Deep, qtz–cal vein	1814801	3823084	420
66	15-900_156	●			East Deep, qtz–cal vein	1814985	3825571	2426
67	15-900_1116.8	●			East Deep, qtz–cal vein	1814639	3825093	1670
68	15-901_2500		●		East Deep, qtz–ank vein in altered host rock	1814636	3822901	615
69	15-926_1388	●			SP-1a, South Pogo, qtz–cal vein with disseminated sulfide	1814092	3819327	1334
70	15-926_1396		●		SP-1a, South Pogo, qtz–ank vein in altered host rock	1814093	3819328	1326
71	15-949_833.5	●			Northern extension of the Liese Zone, qtz–cal vein in altered host rock	1810795	3822859	963
72	15-949_988.4		●		Northern extension of the Liese Zone, qtz–ank vein in altered host rock	1810696	3822866	844
73	15EXP008_1177	●	●		Liese Zone, qtz–cal–ank vein in alterd host rock	1812500	3821393	464
74	15EXP043_389.7		●		East Deep, qtz–ank vein	1813646	3822714	404
75	15EXP043_582.9	●			East Deep, qtz–cal vein	1813773	3822786	278
76	555_stope access_X-Vein		●		X-Vein, North Zone, qtz–ank vein	1812723	3822854	544

DOS: dolomite–sericite alteration; qtz: quartz; cal: calcite; ank: ankerite; sid: siderite. The brief description and its X-Y-Z coordinate of the sample are listed in the table. Spatial reference: NAD 1983 UTM Zone 6N. The XRD analyses for carbonate identification are qualitatively indicated by a circle in the table. The solid circles indicate that the analysis described in the respective column was performed. No analysis has been done on open circles.

, Figure 4). The principal carbonate mineral in the Liese Zone is siderite (FeCO₃), with ferroan dolomite–ankerite series carbonates (Ca (Fe²⁺, Mg, Mn) (CO₃)₂) and calcite (CaCO₃) distributed more widely across the deposit.

One drill core sample was taken from the Liese (L1) orebody (Figure 3b) for SIMS analysis (DDH97-50, 486.0–485.3 feet). The sample consists of quartz and trace amounts of arsenopyrite, pyrite, and chalcopyrite. Two graphite-containing samples were taken from underground exposures of the Leise and North Zone veins (Table 1).

Photographs of representative carbonate-bearing samples and photomicrographs of their thin sections are shown in Figure 5.

3.2. Analytical Methods

3.2.1. X-Ray Diffraction (XRD) Analysis

Powder method X-ray diffraction (XRD) analysis was performed to determine the relative abundance of carbonate minerals in the collected samples. A Rotaflex (rotating anode) RINT2500 (Rigaku, Tokyo, Japan) instrument at the Geological Survey of Japan (GSJ), National Institute of Advanced Industrial Science and Technology (AIST), and the SmartLab (3 kW) (Rigaku, Tokyo, Japan) apparatus at Shizuoka University were used to conduct the analyses.

Carbonate minerals can be distinguished as calcite, ferroan dolomite–ankerite series carbonates, and siderite using XRD analysis. Since calcite (CaCO₃) and siderite (FeCO₃) contain one kind of metal cation, the d-spacing of the crystal lattice is constant, and the mineral can be clearly identified by XRD analysis. On the other hand, ferroan dolomite–ankerite series carbonates express the more complicated chemical formula of Ca (Fe²⁺, Mg, Mn) (CO₃)₂. A dolomite with an Fe/Mg molar ratio > 0.25 is classified as ankerite (e.g., [29]). Further identification of dolomite is made possible when XRD analyses are combined with XRF analysis to determine major element compositions.

3.2.2. Chemical Analysis Methods

Several carbonate-bearing rock samples were analyzed for gold by instrumental neutron activation analysis (INAA) and for major element compositions (Ca, Fe, Mg, and Mn) using X-ray fluorescence (XRF) after total fusion, and inductively coupled plasma mass spectrometry (ICP-MS) after near total digestion at Activation Laboratories Ltd. (Actlabs) in Canada. Concentrations of rare earth elements (REE) were obtained by ICP-MS after near total digestion. Samples analyzed for REE are listed in Table 1.

3.2.3. Isotope Ratio Mass Spectrometry (IRMS)

Carbon and oxygen isotopic ratios of carbonate minerals identified by XRD analysis were analyzed using a Finnigan MAT251 mass spectrometer (Thermo Finnigan, Bremen, Germany) at the GSJ, AIST, or using a Finnigan MAT250 mass spectrometer (Thermo Finnigan, Bremen, Germany) at Shizuoka University. At the GSJ (GSJ Method), carbon dioxide was directly liberated from mixtures of calcite and silicate minerals by treatment with 100% phosphoric acid for one hour in an evacuated reaction tube heated to 25 °C in a water bath [30]. Because silicate minerals are inactive with phosphoric acid, we were able to collect CO₂ samples from hydrothermal alteration zones, as well as from veins. Carbonate minerals besides calcite react slowly with phosphoric acid and require higher temperatures for the complete reaction. Ankerite (or ferroan dolomite) is treated at 50 °C for more than 12 h. When calcite and ankerite coexist in the same sample, an initial CO₂ sample is collected by reacting calcite with phosphoric acid at 25 °C for one hour, with a second CO₂ sample recovered by reacting ankerite with phosphoric acid at 50 °C for more than 12 h. CO₂ liberation from siderite is carried out using a silicon oil bath maintained at 100 °C for at least 12 h.

At Shizuoka University (Shizuoka Method; [31]), 0.15 mg aliquots of the powdered sample are placed into steel thimbles and dropped down one-by-one into phosphoric acid at 60 °C, in an online reaction chamber under vacuum conditions. The evolved gas is cryogenically purified to retain CO₂. The reaction chamber is connected to the inlet system of the mass spectrometer. The ankerite and siderite samples are treated offline at 100 °C for more than 12 h in a silicon oil bath.

Isotope ratios of the evolved CO₂ and machine standard CO₂ were alternately measured and are reported in standard δ notation in per mil (‰) relative to the Vienna Standard Mean Ocean Water (SMOW) for δ¹⁸O and relative to the Vienna Pee Dee Belemnite (PDB) for δ¹³C. The δ¹⁸O values of calcite, ankerite, and siderite are calculated from the isotope ratios of CO₂ using the oxygen isotope fractionation factors between CO₂ and carbonate minerals (for calcite: [32,33,34]; for ankerite and siderite: [35]) at the reaction temperatures. Using measurements on a laboratory working standard and the NBS 19, the measured isotope ratios were normalized to the limestone reference material [36]. Reproducibility was better than ±0.1‰ (2σ) for both the δ¹³C and δ¹⁸O values of calcite.

The two graphite-containing samples (Table 1) were sent to Actlabs for determination of δ¹³C values. XRD analysis confirmed that the rock did not contain any carbon-containing minerals, such as carbonates, other than graphite. The sample was combusted to CO₂, and the carbon isotope ratio of the CO₂ was measured using a mass spectrometer at Actlabs. The reproducibility was ±0.2‰ (2σ).

3.2.4. Secondary Ion Mass Spectrometry (SIMS)

Au and As concentrations were determined on a polished sample from the Leise (L1) ore body using an ims-1270 SIMS (Cameca, France) at the GSJ, AIST. A defocused Cs⁺ primary beam was restricted to 15 μm in diameter by a circular aperture to obtain a homogeneous primary beam (Köhler illumination) of about 0.1 nA. A square field aperture is introduced into the secondary ion optics, limiting the analyzed area of the sample surface to a central square measuring 3 × 3 μm to avoid the crater-edge effect.

The ¹⁹⁷Au⁻ and ⁷⁵As⁻ secondary ions, as well as the matrix ³⁴S⁻ as an internal reference, were detected under a total impact energy of 20 kV (a primary accelerating voltage of +10 kV and a secondary extraction voltage of −10 kV) using an electron multiplier (EM) in the analysis of sulfide minerals. A high mass resolving power of 3300 was used to eliminate the isobaric interference of ¹³³Cs³²S₂ on ¹⁹⁷Au (M/ΔM = 1728). The detection limits, which were computed as three times the standard deviation of the background noise, for the Au and As measurements were 15 ppb and 0.6 ppm, respectively, for an ³⁴S count of 3 × 10⁵ cps and assuming that the background noise has a Poisson distribution. The ¹⁹⁷Au⁻ and ²⁸Si⁻ secondary ions were measured in the analysis of quartz.

Quantitative SIMS analysis requires the use of matrix-matched external standards for each element because of matrix effects on the secondary ion intensities. Calibration of Au concentrations in pyrite and quartz was conducted using external standard samples of pyrite and quartz. These samples were implanted with ¹⁹⁷Au ions at a dosage of 1 × 10¹⁵ atoms/cm² with an implantation energy of 1.6 MeV by the Mining and Mineral Sciences Laboratories at the Canadian Center for Mineral and Energy Technology (CANMET-MMSL) using the 1.7 MeV Tandetron accelerator located at the interface Science Western laboratory at the University of Western Ontario, Canada. Calibration of Au concentrations in chalcopyrite was conducted using an external chalcopyrite standard that was implanted with ¹⁹⁷Au ions at a dosage of 1 × 10¹⁴ atoms/cm² with an implantation energy of 1.0 MeV at the same laboratory. Calibration of As concentrations in pyrite was performed using an external standard sample implanted with As ions at a dosage of 3 × 10¹⁵ atoms/cm² with an implantation energy of 0.4 MeV. After analyzing the implanted standard sample for depth profiling, the depth of the sputtered borehole was measured using a surface profiler to determine the relative sensitivity factor (RSF), which is then used to calculate the concentration of the targeted elements. Analytical details, including the RSFs for the Au and As measurements in pyrite and chalcopyrite, are found in the work of Morishita et al. [22,23]. The RSF for Au in quartz is 1.26 × 10²¹ atoms/cm³ in our laboratory. The key advantages of SIMS analysis are its high sensitivity and its depth-profiling capability.

4. Results

4.1. XRD Analysis

Qualitative results of the XRD analysis are listed in Table 1, with carbonate minerals classified as calcite, ankerite, or siderite. All of the ferroan dolomite–ankerite series minerals are listed as “ankerite” for the sake of convenience. Note that several samples were found to contain two kinds of carbonate mineral: siderite with ankerite and ankerite with calcite. The coexistence of siderite with calcite was not observed.

4.2. Chemical Analysis

The Ca, Fe, Mg, and Mn concentrations of the analyzed carbonate-bearing rock samples are provided in

Table 2. The Ca, Fe, Mg, Mn, and Au concentrations in the carbonate-bearing samples.

Sample Name	Ca	Fe	Mg	Mn	Au	Minerals Other than Carbonates
	wt %	wt %	wt %	wt %	ppb
2014082207	29.60	0.86	0.43	0.43	8	Qtz
M000002a	24.00	1.88	0.83	0.00	35	Qtz
10DL007A	38.13	0.46	0.22	0.28	13	Qtz
10DL008	36.56	1.21	0.17	2.17	<2	Qtz, Rho
11DL025	29.90	0.36	0.18	0.39	7	Qtz
11-584A	0.29	42.21	1.54	2.55	373	Qtz, Chl, Sulfide
11-584B	35.93	0.30	0.16	0.49	425	Qtz
11-584C	18.48	10.23	6.39	0.42	<2	Qtz
11-584D	37.90	0.42	0.17	0.70	<2	Qtz
14-765_589.9	0.40	27.80	1.18	0.75	4670	Qtz, Sulfide
14-765_824.3	0.52	26.20	2.05	2.11	503	Qtz, Sulfide
14-795_1434.4	6.31	5.99	2.66	0.11	503	Qtz, Ser
15-896_2500.4	10.60	17.40	3.43	0.81	154	Qtz, Ser, Sulfide

Qtz: quartz; Rho: rhodochrosite; Chl: chlorite; Ser: sericite. Many samples contain traces of unidentifiable clay minerals.

. The Au concentration and list of non-carbonate minerals present in each sample are also noted. The results of the REE chemical analysis are shown in

Table 3. REE and Au concentrations in samples from the Pogo deposit and the Burn deposit (southwest of the Pogo deposit).

Pogo Deposit	La	Ce	Pr	Nd	Sm	Eu	Gd	Tb	Dy	Ho	Er	Tm	Yb	Lu	Au
	ppm	ppm	ppm	ppm	ppm	ppm	ppm	ppm	ppm	ppm	ppm	ppm	ppm	ppm	ppb
2011032201	43.3	79.4	10.2	33.7	5.5	0.91	4.8	0.6	3.3	0.6	1.4	0.2	0.7	0.1	32
2012081003	50.9	101	12.1	41.7	7.6	1.83	7.1	1	5.9	1.1	2.8	0.4	2.3	0.4	167
2012081006	60.3	115	14.2	50.2	9.3	1.84	8.5	1.1	6	1.1	2.7	0.4	1.9	0.3	<2
2012081108	48.2	94.7	11.1	38.2	6.4	1.53	6	0.8	4.7	0.9	2.5	0.4	2	0.3	31
M000002a	14.2	28.3	3.5	14.3	3.1	0.79	3.8	0.6	4	0.9	2.6	0.4	2.1	0.4	35
M000034a	56	110	13.5	47	9	1.95	8.4	1.1	5.9	1	2.2	0.3	1.6	0.2	115
M000068c	7	14.6	1.7	6	1.3	0.75	1.5	0.2	1.8	0.4	1.4	0.3	1.7	0.3	64
11-584A	10.6	20.2	2.3	8.2	1.7	1	2	0.3	2.4	0.5	1.5	0.3	1.7	0.3	373
14-752_1188.6ft	52.3	101	11.9	40.8	6.9	1.21	5.8	0.7	3.5	0.5	1.2	0.1	0.8	0.1	30
14-795_1242.8ft	40.4	78.5	9.3	32.7	5.9	1.35	5.6	0.7	3.9	0.7	1.6	0.2	1	0.1	102
2014081806	48.6	104	13.8	48	8.3	1.49	7.4	0.9	4.9	0.8	1.8	0.2	0.9	0.1	77
Burn
2014082304	8.8	14.7	1.5	4.6	0.9	1.24	1	0.1	1	0.2	0.7	0.1	1	0.2	>30,000

. The locality and description of samples in Table 2 and Table 3 are provided in Table 1.

4.3. IRMS

The δ¹³C and δ¹⁸O values of the identified calcite, ankerite, and siderite minerals are listed in

Table 4. δ¹³C_PDB and δ¹⁸O_SMOW of calcite.

Sample Name	Description	δ13CPDB	δ18OSMOW
2012080709	Qtz vein with calcite and sulfide	−1.5	−5.3
2012081001	Carbonate vein in weathered diorite (ground)	−1.5	−3.2
2012081004	Thin qtz–cal vein with sulfide	−2.1	−2.8
2012081005	Qtz vein with calcite and sulfide	−0.8	−4.6
2012081006	Calcite dissemination in altered rock	−2.2	−2.6
2014082206	Qtz vein with calcite and sulfide	−0.2	−2.0
2014082207	Qtz–Cal vein (ground)	0.0	−2.3
2014082208	Qtz–Cal vein (ground)	−0.2	−2.1
2014082211	Qtz–Cal vein (ground)	−0.1	−1.8
2014082212	Carbonate veinlet (ground)	−0.3	−0.2
2014082213	Qtz–Cal vein (ground)	0.1	−2.3
2014082215	Qtz–Cal vein (ground)	−0.9	−0.2
2014082216	Qtz–Cal vein with sulfide (ground)	−3.3	2.9
M000002a	Qtz–Cal vein in altered diorite	−2.1	−2.2
M000004a	Qtz–Cal vein in altered diorite	−2.4	−1.5
M000013	Calcite dissemination	−0.9	−2.9
M000015a	Qtz–Cal vein	−1.3	−0.7
M000034a	Qtz–Cal vein	−1.0	−0.3
10DL007A *	Cal vein	−0.7	−5.4
10DL007B *	Qtz–Cal–Ank vein	−0.5	−3.2
10DL008 *	Crystalline calcite in a drusy cavity	1.4	2.8
10DL012A *	Cal vein	−1.1	−0.9
10DL012B *	Cal–Ank vein	−1.3	0.9
11DL011A *	Qtz–Cal–Ank vein with muscovite	−1.8	6.4
11DL012A *	Cal–Ank vein	−1.5	−2.6
11DL012B *	Qtz–Cal–Ank vein	−1.7	−0.4
11DL012C *	Cal–Ank vein	−1.3	−2.8
11DL025 *	Qtz–Cal vein	−1.1	−4.2
11DL028 *	Qtz–Cal–Ank vein	−1.0	−2.7
11-584B *	Qtz–Cal vein	−0.6	−5.0
11-584D *	Cal vein	−0.8	−3.6
11u201A *	Qtz–Cal–Ank vein	−3.4	9.3
14-742_1027.2ft	Qtz–Cal vein	−1.3	−3.2
14-771_876.1ft	Qtz–Cal vein	−1.6	−3.1
14-795_1242.8ft	Qtz–Cal vein	−0.7	1.8
14-805_857.1ft	Qtz–Cal vein	−0.6	−2.8
14-805_1008.9ft	Qtz vein with calcite and sulfide	−1.7	12.5
14-805_1020ft	Qtz–Cal vein	−1.5	6.1
15-882_316.5	Silicified rock with calcite	−2.5	6.9
15-882_833.5	Qtz–Cal vein	−1.3	−1.8
15-884_500.5	Qtz–Cal vein	−0.3	−0.4
15-896_2581.3	Qtz–Cal vein	−1.7	−5.6
15-900_156	Qtz–Cal vein	−0.8	−3.3
15-900_1116.8	Qtz–Cal vein	−0.7	0.0
15-926_1388	Qtz–Cal vein with disseminated sulfide	−0.9	−2.4
15-949_833.5	Qtz–Cal vein in altered host rock	−1.3	−1.8
15EXP008_1177	Qtz–Cal–Ank vein in alterd host rock	−1.7	−2.1
15EXP043_582.9	Qtz–Cal vein	−1.3	−4.2

* Analyzed using the GSJ Method. Otherwise, the Shizuoka Method was applied.

,

Table 5. δ¹³C_PDB and δ¹⁸O_SMOW of ankerite.

Sample Name	Description	δ13CPDB	δ18OSMOW
2011032202	Qtz vein with ankerite and sulfide	0.2	5.6
2011032205	Qtz vein with ankerite and sulfide	0.6	6.1
2012080711	Carbonate dissemination in altered intrusive	−2.9	4.5
2012081003a	12-685, 635′, qtz–ank vein	−1.8	−1.9
2012081011b	Ankerite dissemination in altered intrusive	−3.8	8.8
2012081108a	Qtz vein with ankerite and sulfide	−2.6	7.6
2014082210b	Qtz vein with brown carbonate	0.1	2.7
10DL007B *	Qtz–Cal–Ank vein	0.2	−3.7
10DL012B *	Cal–Ank vein	−0.9	1.9
11DL011A *	Qtz–Cal–Ank vein with muscovite	−1.8	7.3
11DL011B *	Qtz–Ank vein	−1.7	−2.1
11DL011DOS *	Qtz–Ank vein with DOS alteration	−1.5	0.1
11DL012A *	Cal–Ank vein	−0.9	−4.5
11DL012B *	Qtz–Cal–Ank vein	−1.6	−1.3
11DL012C *	Cal–Ank vein	−0.6	−4.9
11DL028 *	Qtz–Cal–Ank vein	−1.2	−3.4
11-584C *	Qtz–Ank vein	−1.3	11.1
11u201A *	Qtz–Cal–Ank vein	−4.0	10.5
14-742_474.6ft	Qtz–Ank vein with sulfide	−3.2	10.9
14-773_875.8ft	Qtz–Ank vein with sulfide	−2.8	10.9
14-795_1434.4ft	Qtz–Ank vein with sericite	−2.3	12.5
15-896_2500.4	Qtz–Ank vein with sericite and sulfide	−2.0	12.7
15-901_2500	Qtz–Ank vein in altered host rock	−2.0	−2.5
15-926_1396	Qtz–Ank vein in altered host rock	−0.1	1.9
15-949_988.4	Qtz–Ank vein in altered host rock	−2.0	1.8
15EXP008_1177	Qtz–Cal–Ank vein in altered host rock	−1.5	−3.7
15EXP043_389.7	Qtz–Ank vein	−1.8	−3.7
555_stope access_X-Vein	Qtz–Ank vein	−2.7	6.7

* Analyzed using the GSJ Method. Otherwise, the Shizuoka Method was applied.

and

Table 6. δ¹³C_PDB and δ¹⁸O_SMOW of siderite.

Sample Name	Description	δ13CPDB	δ18OSMOW
M000068c	Qtz–Sid vein with sulfide	−4.5	12.6
11-584A *	Qtz–Sid vein with chlorite and sulfide	−4.3	12.0
14-716_278.2ft	Qtz–Sid vein	−4.5	12.7
14-752_1188.6ft	Qtz–Sid vein with muscovite	−4.7	14.0
14-765_589.9ft	Qtz–Sid vein with sulfide	−3.3	14.0
14-765_824.3ft	Qtz–Sid vein with sulfide	−5.2	13.2

* Analyzed using the GSJ Method. Otherwise, the Shizuoka Method was applied.

, respectively. The sample locations and descriptions are provided in Table 1. There are 48, 28, and 6 isotope data for calcite, ankerite, and siderite, respectively. The δ¹³C values of graphite in samples 2011032201 and 2014081806 (Table 1) are −19.5‰ and −11.0‰, respectively.

4.4. SIMS

The results of the Au and As measurements are listed in

Table 7. Au and As measurements by SIMS.

Mineral	Core/Rim	Measurement No.	Locality No.	28Si	34S	As	Au	Au	As
				SIMS	SIMS	SIMS	SIMS	ppm	ppm
Pyrite	core	Au5*262.adp	M262		2.70 × 105	5.83	7.10× 10−2	0.06	195
Pyrite	core	Au5*263.adp	M263		2.40 × 105	10.3	2.54 × 10−1	0.22	389
Chalcopyrite	core	Au5*264.adp	M264		2.70 × 105	9.07	7.73 × 10−1	0.28
Chalcopyrite	rim	Au7*281.adp	M281	4.92 × 10−1	3.39 × 105	0.75	3.08	0.89
Arsenopyrite	core	Au5*265.adp	M265		8.62 × 104	1.11 × 104	2.64	6
Pyrite	core	Au6*267.adp	M267		2.47 × 105	8.25 × 10−1	1.05	0.89	30
Arsenopyrite	core	Au6*271.adp	M271		1.15 × 105	1.78 × 104	2.29	4
Qtz/Apy	boundary	Au6*272.adp	M272	1.19 × 104	7.03 × 104	1.07 × 104	2.00	16
Quartz	rim	Au6*273.adp	M273	2.11 × 104	0.53	4.92 × 10−1	1.87 × 10−1	1.38
Quartz	rim	Au6*274.adp	M274	1.63 × 104	1.03 × 103	6.85 × 102	9.82 × 10−1	9.36
Arsenopyrite	core	Au7*276.adp	M276	1.59	7.83 × 104	1.03 × 104	9.83 × 10−1	3
Quartz	core	Au7*277.adp	M277	5.39 × 104	9.49 × 10−1	7.31 × 10−1	1.81 × 10−1	0.52
Qtz/Qtz	boundary	Au7*279.adp	M279	9.37 × 104	2.27 × 103	1.36 × 102	1.44	2.38
Pyrite	core	Au7*280.adp	M280	1.60	2.56 × 105	3.33 × 10−1	1.57	1.29	12
Py/Cpy	boundary	Au7*282.adp	M282	9.29	3.13 × 105	9.46 × 10−1	3.05	1.5	27

The measurement Au7*279.adp is on a boundary between two quartz grains. Qtz: quartz; Apy: arsenopyrite; Cpy: chalcopyrite.

. The Au concentrations in pyrite, chalcopyrite, and quartz range from 0.06 to 1.29 ppm, 0.28 to 0.89 ppm, and 0.52 to 9.36 ppm, respectively. The Au concentration profile from each analysis is homogeneous in the depth direction, and Au is considered to be distributed homogeneously. The Au7*282.adp measurement was conducted on a grain boundary between pyrite and chalcopyrite. Although the RSF for Au in arsenopyrite is not available, the Au concentration in arsenopyrite is from 3 to 6 ppm when using the RSF value for Au in pyrite to calibrate the arsenopyrite SIMS data. Arsenopyrite values are shown to one significant digit, given that some errors due to matrix effects are expected. The As concentration in pyrite ranges from 12 to 389 ppm.

5. Discussion

5.1. Chemical Analyses and REE Patterns

The chemical formulae of the carbonate minerals analyzed in this study are shown in

Table 8. Chemical formulas of the analyzed carbonates.

	Chemical Formulas of Carbonates	Carbonate Name
2014082207	(Ca0.95Fe0.02Mg0.02Mn0.01)CO3	Calcite
M000002a	(Ca0.90Fe0.05Mg0.05)CO3	Calcite
10DL007A	(Ca0.98Fe0.01Mg0.01Mn0.01)CO3	Calcite
10DL008	(Ca0.93Fe0.02Mg0.01Mn0.04)CO3	Calcite
11DL025	(Ca0.97Fe0.01Mg0.01Mn0.01)CO3	Calcite
11-584A	(Ca0.01Fe0.87Mg0.07Mn0.05)CO3	Siderite
11-584B	(Ca0.98Fe0.01Mg0.01Mn0.01)CO3	Calcite
11-584C	(Ca0.50Fe0.20Mg0.29Mn0.01)CO3	Ankerite
11-584D	(Ca0.97Fe0.01Mg0.01Mn0.01)CO3	Calcite
14-765_589.9	(Ca0.02Fe0.87Mg0.09Mn0.02)CO3	Siderite
14-765_824.3	(Ca0.02Fe0.78Mg0.14Mn0.06)CO3	Siderite
14-795_1434.4	(Ca0.42Fe0.29Mg0.28Mn0.01)CO3	Ankerite
15-896_2500.4	(Ca0.36Fe0.43Mg0.19Mn0.02)CO3	Ankerite

The formula may deviate slightly from that of pure carbonate because. the carbonate-containing whole sample was analyzed (Table 2).

. The presence of quartz does not affect the cation concentrations of the carbonates, but some trace impurities do affect the analytical data (Table 2) and therefore, the chemical formula in Table 8. All of the ferroan dolomites are classified as ankerite, based on their Fe/Mg molar ratios (see Section 3.2.1).

The chondrite-normalized REE patterns of the analyzed Pogo samples are illustrated in Figure 6a. The deviation of the redox-sensitive Eu and Ce elements from a smooth REE pattern can provide important insights into the oxidation state of the hydrothermal mineralizing fluids. The valence of REE elements is usually trivalent, but Eu can take a bivalent state under a reducing environment, whereas Ce can take a tetravalent form in an oxidizing environment. If the hydrothermal fluid is depleted in Eu when bivalent Eu is incorporated into the surrounding rocks before precipitating the carbonate phase, negative Eu anomalies are found in the samples. Negative Eu anomalies are found in the REE patterns of most Pogo samples (Figure 6a), which indicates that Eu was bivalent under a reducing environment at the time of mineralization. The environment is similar to that of the average upper continental crust (Figure 6b). The REE pattern of sample M000002a is different from the others. This sample is from Hill 4021, a peripheral deposit located southeast of the Pogo vein systems (Figure 3a), and the hydrothermal properties may differ from those inferred for the main deposit area. Contrastingly, Figure 7 illustrates REE patterns with positive Eu anomalies. Samples M000068c and 11-584A contain siderite and were collected from the Leise vein system at the center of the deposit (Table 1). Sample 2014082304 was collected from an outcropping mineralized vein in the Burn exploration area, located across the Goodpaster River to the west of the Pogo mining area (Figure 4). These samples with positive Eu anomalies in the REE patterns signify the presence of a high-temperature hydrothermal component [37], and may have been located closer to the center of the magmatic hydrothermal system. Absorption of magmatic vapors, including SO₂, into deeply circulating hydrothermal fluids leads to the formation of highly acidic, oxidizing fluids [38].

5.2. SIMS Analysis

The locations of SIMS microanalyses on a Leise (L1) ore sample are shown on a reflected light photomicrograph in Figure 8. Au concentrations of 1.38 ppm (M273) and 9.36 ppm (M274) measured at a quartz grain margin are higher than those measured in the core of the grain (0.52 ppm, M277). Likewise, an Au concentration of 0.89 ppm measured at the rim of a chalcopyrite grain (M281) is higher than that measured in its core (0.28 ppm, M264). Au concentrations of 2.38 ppm and 16 ppm were measured at the boundary between two quartz grains (M279) and between quartz and arsenopyrite (M272), respectively. The higher Au concentrations at mineral grain boundaries might be due in part to matrix effects, but Au is generally thought to be concentrated on grain boundaries. The existence of Au located inside the quartz grains implies that the quartz precipitated from a hydrothermal mineralizing fluid.

5.3. Origin of the Ore-Forming Fluid

The Pogo deposit is defined as an intrusive-related gold system formed at a depth of six kilometers [3,4,24]. CO₂ is abundant in fluid inclusions in hydrothermal veins in intrusion-related Au systems [4]. McCoy et al. [2] and Thompson et al. [5] argue that such fluids are of magmatic origin. The homogenization temperatures of fluid inclusions in quartz from the Liese and North Zones of the Pogo deposit range between 290 and 470 °C, with a maximum recorded temperature of 580 °C [41]. Takaoka et al. [41] estimated a fluid δ¹⁸O range of 5.1 to 10.4‰, based on the homogenization temperatures and on δ¹⁸O measurements for vein quartz. The fluid was considered to originate from magmatic water or metamorphic water [41]. The homogenization temperatures of fluid inclusions in Au and sulfide-bearing quartz veins range between 308 and 470 °C, with trapping pressures estimated to be 1.7 to 2 kbar [6,42].

The δ¹³C and δ¹⁸O values of the carbonates collected from veins and hydrothermal alteration zones in the Pogo mining area are shown in Figure 9. The distribution of Pogo isotope data is different from that of a typical epithermal gold-bearing vein deposit. For example, the Kushikino deposit was formed from meteoric water (δ¹⁸O = −7‰, δ¹³O = −11‰; [43]), and the hydrothermal fluid of the high-grade Hishikari deposit is almost entirely meteoric in origin, with sporadic injections of magmatic fluid [44]. A fluid of magmatic origin does not indicate that it originates entirely from magma, but rather that it includes fluids in isotopic equilibrium with magma.

All the siderites in this study (Figure 9; Table 6) display high δ¹⁸O and low δ¹³C values relative to those of ankerites and calcites, and the δ¹³C and δ¹⁸O values fall within a relatively narrow range. The average δ¹³C and δ¹⁸O values of five siderites (excluding one outlier) are −4.7‰ and 12.9‰, respectively. We can thus formulate a working hypothesis that the siderite occurred at the earliest stage of the Pogo mineralization, followed by the precipitation of ankerite and calcite, based on their δ¹³C and δ¹⁸O values. Assuming that siderite is in equilibrium with a magmatic fluid at 500 °C, the oxygen isotope fractionation factor (10³lnα) between siderite and water is 2.88‰ [45], and the carbon isotope fractionation factor (10³lnα) between siderite and the dissolved carbon-bearing species of CO₂ is −0.15‰ [46,47]. The isotope fractionation factors were chosen with an emphasis on comparison between the three carbonate minerals, rather than by using the values from individual papers. While there may be some absolute error in this approach, it allows for a relatively consistent comparison.

The δ¹³C and δ¹⁸O values of the early-stage hydrothermal fluid using the isotope fractionation factors and average δ¹³C and δ¹⁸O values of the siderites (−4.7‰ and 12.9‰) are calculated as −4.8‰ and 10.0‰, respectively. The estimated δ¹⁸O value of the early-stage hydrothermal fluid is thus near the upper end of the range for primary magmatic water [48,49]. In the ilmenite-series intrusion-related Pogo gold system, the magmatic fluids that interacted with the surrounding rocks are likely to have had high δ¹⁸O values. The δ¹³C of magmatic CO₂ varies from −4.2 to −7.5‰ in mid-ocean ridge basalt (MORB) and from −2.8 to −6.7‰ in glasses from Hawaii [50]. The mode of the δ¹³C value of the Oka carbonatite lies at −5.0‰ [51]. The estimated δ¹³C value of dissolved inorganic carbon (DIC) in the early-stage hydrothermal fluid at Pogo (−4.8‰) is therefore consistent with that of magmatic water.

In summary, the results indicate that a magmatic fluid (δ¹⁸O = 10.0‰, δ¹³C = −4.8‰) was the initial ore-forming fluid for Pogo mineralization; however, the other carbonates (ankerite and calcite) exhibit lower δ¹⁸O values (Figure 9). Morishita [52] proposed a fluid evolution model for the 93 Ma Ohtani and Kaneuchi hypothermal tungsten–quartz vein deposits based on oxygen and carbon isotopic evidence. In the ore zones, a relatively high temperature magmatic fluid (δ¹⁸O = 12‰ at 600 °C) produced a greisen in the granodiorite host rock at a depth of 5 km. This alteration phase was later cut by mineralized veins at lower temperatures (δ¹⁸O = 12‰ at 500 °C). The decrease in temperature of the initial magmatic fluid from 600 °C to 500 °C was due to heat conduction into the surrounding rocks. An adiabatic expansion of the fluid and subsequent incorporation of interstitial water with a low δ¹⁸O value can cause a significant decrease in the δ¹⁸O value of the ore-forming fluid [52]. A similar mechanism is invoked to account for the decreasing δ¹⁸O values of the ore-forming fluids at Pogo.

5.4. Fluid Evolution in the Pogo Hydrothermal System

The earliest phase of the Pogo hydrothermal fluid system is assumed to be of magmatic origin. The δ¹³C and δ¹⁸O values are calculated as −4.8‰ and 10.0‰, respectively; see Section 5.3. Siderite-bearing samples with positive Eu anomalies in REE patterns (Figure 7) show different REE patterns than do the others (Figure 6), indicating a high-temperature hydrothermal fluid [37].

We assume that the end member of the counterpart fluid for mixing is meteoric water, which at the Pogo deposit has a δ¹⁸O value of −20‰ [53]. This value is consistent with that of meteoric water (−20‰) across southwestern Alaska [8]. The δ¹³C value of DIC in meteoric water (δ¹³C = 0‰) is taken as the δ¹³C value of DIC in sea water, modified locally from equilibration with atmospheric CO₂ in the high latitude ocean (e.g., [54,55,56]).

Figure 10 shows a mixing line between the magmatic fluid and meteoric water for the ore-forming hydrothermal fluid. The measured δ¹³C and δ¹⁸O values of siderite, ankerite, and calcite are superimposed on the simulation results (Figure 10). Most carbonates are in the range of the simulation results, although several ankerites and calcites display higher δ¹³C values than those in the simulation range. These anomalies are explained below.

Crustal contamination through assimilation occurs in magmas associated with intrusion-related gold systems [4]. Graphite schist in the Pogo deposit has a δ¹³C value of −19.5‰ (Sample 2011032201, Table 1). This ratio can be explained by considering that sedimentary carbon generated by photosynthesis in plants is depleted in the heavier isotope ¹³C. The carbon source is assumed to be organic matter. Graphitic schist sample 2014081806 has a δ¹³C value of −11.0‰. This value is higher than expected for organic carbon (Sample 2011032201, for example). The graphite in this sample might have originated from a partial reaction with hydrothermal carbonate. The CO₂/CH₄ ratios in fluid inclusions from the Liese Zone are interpreted to range between 1 and 49 [41,57]. The methane content in the Pogo fluid was typically only several percentage points, but occasionally increased up to 50%, corroborating a generally reducing environment for Pogo ore formation. The carbon isotope fractionation factors (10³lnα) between CO₂ and CH₄ at 500 °C and 400 °C are 15.3‰ and 19.2‰, respectively [58], which are consistent with the research findings of Bottinga [59] and Horita [60]. When CH₄ is produced in the hydrothermal fluid, the δ¹³C of CH₄ becomes significantly lower than the average δ¹³C of DIC in the fluid. As a result, the δ¹³C of CO₂ becomes higher than the average, which is also concluded in Ref. [52]. The carbon isotope fractionation between CO₂ and CH₄ is much larger than that between carbonates and CO₂, so carbonates with δ¹³C values higher than those of the simulation range are expected to occur (Figure 10). Therefore, a carbonate precipitated from a fluid containing CH₄ has a higher δ¹³C value than that precipitated from a fluid without CH₄. Two ankerite-bearing samples with higher δ¹³C values between 0 and 1‰ (Samples 2011032202 and 2011032205, Figure 10) are located near graphite schist, and it is assumed that the CH₄ content of the fluid was relatively high. A large euhedral calcite crystal (10DL008 in Table 1) with the highest δ¹³C value (Figure 10) is thought to have slowly crystallized from a low-temperature dilute fluid late in the mineralization history. Large variations in δ¹³C values (> 10‰) due to changes in the CO₂/CH₄ ratio have been observed in other intrusion-related mineral systems [52].

The simulation results are consistent with the working hypothesis proposed in Section 5.3. Siderite is first precipitated from a magmatic fluid (δ¹⁸O = 10‰, δ¹³C = −4.8‰) at 500 °C. The REE patterns with positive Eu anomalies (Figure 7) confirm the magmatic nature of the early-stage fluid. Ankerite was then precipitated during the gradual incorporation of surrounding fluid mixed with meteoric water-hosted DIC into the magmatic fluid. As a result, the δ¹⁸O and temperature of the fluid gradually decreased, while δ¹³C slightly increased. The dominant carbon-bearing species of DIC was CO₂, with CH₄ introduced sporadically. Calcite was precipitated with the further incorporation of surrounding meteoric fluid during the latter stages of mineralization. Pogo is defined as a deep-seated, intrusion-related gold deposit. However, the amount of magmatic fluid input appears unexpectedly small, with most of the Pogo mineralizing fluids interpreted to be of meteoric origin.

6. Concluding Remarks

Carbonate minerals have a close spatial relationship to hydrothermal gold mineralization in the Pogo ore deposit. Most Fe-bearing carbonates (siderite and ankerite) occur in association with hydrothermal alteration and sulfide minerals. Siderite is associated with early-stage gold mineralization and occurs with ankerite in main stage ore assemblages. Calcite is mainly recognized in the later mineralization stages.

The carbon and oxygen isotopic ratios of carbonate minerals (siderite, ankerite, and calcite) present within the deposit illustrate the isotopic evolution of the ore-forming fluid. The initial hydrothermal fluid phase is interpreted to be magmatic in origin. The fluid evolution was characterized by a gradual decrease in δ¹⁸O and a slight increase in δ¹³C with decreasing temperature. The general geologic environment of the Pogo deposit is assumed to be reducing at the time of ore formation, based on REE analyses and the presence of graphite. The dominant carbon-bearing species was CO₂, with methane generated sporadically, resulting in the localized crystallization of carbonates with higher δ¹³C values.

Invisible gold is found in pyrite, chalcopyrite, arsenopyrite, and quartz, based on SIMS or ion microprobe analysis. The presence of invisible gold in these minerals has significant metallurgical implications for gold processing at the Pogo mine.