Economic Geology Vol. 84, 1989, pp. 410-424 Gold Mineralizationin the Okanagan Valley, Southern British Columbia: Fluid Inclusion and StableIsotope Studies X. ZHANG*, B. E. NESBITr, AND K. MUEHLENBACHS Department of Geology, University of Alberta,Edmonton, Alberta, CanadaT6G 2E3 Abstract Two styles of precious metalvein mineralization, epithermal and mesothermal, occur in the Okanagan Valley of southern British Columbia. The epithermal deposit (Dusty Mac) is composed of quartzstockwork mineralization in Eocene volcanic and volcaniclastic rocks. The mesothermal deposits (Fairview andOro Fino) consist of thick (1-10 m) quartz veins in Paleozoic and Mesozoic metamorphic rocks. Fluid inclusion and stable isotope studies indicate thattwo distinct hydrothermal fluids wereresponsible forthe mineralization events. At Dusty Mac, the epithermal fluid had a relativelylow temperature (230ø-250øC), extremely low salinity (<1 equiv wt % NaC1), and low $180 (-7 to -9%0) and $D(-133%0) values. At Fairview and OroFino, the mesothermal fluids had higher temperatures (280ø-330øC),significant CO• contents, and highersalinities (3-6 equivwt % NaC1). Stable isotope data show that the mesothermal fluids had low $D (-121 to -148%0) and $18C (-8.5%0) values, but high $180 (+4 to +6%0) values. The data indicate thatthe fluids involved in bothstyles of mineralization are180-shifted meteoric water, withshallow circulation responsible forthe epithermal deposits and deep circulation involved in the formation of the mesothermal deposits. Although both styles of mineralization occur in the same tectonic regime,it appears that they are not tem- porallyor genetically related to each other. Introduction A NUMBER of small precious andbase metaldeposits have been mined in the Okanagan Valley area of southern British Columbia since the end of the last century.The three main gold-producing districts are Fairview, Oro Fino, and Dusty Mac (Fig. 1). The Fairview camp includes three deposits: Fairview, Stemwinder, and MorningStar.They were minedat the beginning of this century for several years,re- opened in the 1930sfor a few years, andare currently beingreconsidered for production. The totalproduc- tion,to date,is 150,000 metrictons of ore thatyielded 510 kg of gold and 4,800 kg of silver (Hedley and Watson,1945). The Oro Fino campis located about 10 km northwest of Fairview and consists of two properties,Oro Fino and Twin Lake, which have mining histories similar to those atFairview. Recorded production was 24,000 metric tons of orethatyielded 270 kg of goldandminorsilver (Hedleyand Watson, 1945). The Dusty Mac mine operated from 1969 to 1976. Total production was 93,000 metric tons of ore,with recovery of 600 kg gold,10,500 kg of silver, and minor copper and lead(Schroeter and Panteleyev, 1986). There appears to be two distinctly different styles of preciousmetal mineralization in the Okanagan Valleyarea: mineralized quartz veins occurring in the Paleozoic andMesozoic metamorphic rocks (Fairview * Present address: Department of Earth andAtmospheric Sci- ences, PurdueUniversity, West Lafayette, Indiana47907. andOro Fino areas), andmineralized silicified zones occurring in Tertiary volcanic and volcaniclastic rocks (Dusty Macarea). Church (1973)suggested that min- eralization in the Dusty Mac deposit formed by the filling of dilation zones. Tempelman-Kluit (1984) proposed a meteoric water model for mineralization in the area. Nevertheless,the characteristics and or- igins of the ore-forming fluids andthe relationships between the two styles of mineralization have re- mained unclear. The purpose of the present study is to investigate thenature of thefluids responsible for precious metal mineralization in the area and to ex- aminethe relationships befweenthe two styles of mineralization. Fluid inclusion and stable isotope studies have been used to examinethe temperature- pressure environment of veinformation and the ori- gins of the ore-forming fluids for bothdeposit types. The results show that each deposit type was formed by a distinct type of fluid.The origins of bothtypes of deposits arerelated to the circulation of meteoric water, though in strikingly different geologic settings, which are similar to those described in Nesbitt et al. (1986). Geologic Setting There have been severalgeologicstudies con- ducted in the Okanagan Valleyarea. The earliest sys- tematic mapping ofthe area was performed by Cairnes (1940) andBostock (1941). Their workwas revised by Little (1961). Church (1973) studied the geology of Tertiary WhiteLake basin in theOkanagan Valley. 0361-0128/89/915/410-15 410
Transcript
Economic Geology Vol. 84, 1989, pp. 410-424
Gold Mineralization in the Okanagan Valley, Southern British Columbia: Fluid Inclusion and Stable Isotope Studies
X. ZHANG*, B. E. NESBITr, AND K. MUEHLENBACHS
Department of Geology, University of Alberta, Edmonton, Alberta, Canada T6G 2E3
Abstract
Two styles of precious metal vein mineralization, epithermal and mesothermal, occur in the Okanagan Valley of southern British Columbia. The epithermal deposit (Dusty Mac) is composed of quartz stockwork mineralization in Eocene volcanic and volcaniclastic rocks. The mesothermal deposits (Fairview and Oro Fino) consist of thick (1-10 m) quartz veins in Paleozoic and Mesozoic metamorphic rocks. Fluid inclusion and stable isotope studies indicate that two distinct hydrothermal fluids were responsible for the mineralization events. At Dusty Mac, the epithermal fluid had a relatively low temperature (230ø-250øC), extremely low salinity (<1 equiv wt % NaC1), and low $180 (-7 to -9%0) and $D (-133%0) values. At Fairview and Oro Fino, the mesothermal fluids had higher temperatures (280ø-330øC), significant CO• contents, and higher salinities (3-6 equiv wt % NaC1). Stable isotope data show that the mesothermal fluids had low $D (-121 to -148%0) and $18C (-8.5%0) values, but high $180 (+4 to +6%0) values. The data indicate that the fluids involved in both styles of mineralization are 180-shifted meteoric water, with shallow circulation responsible for the epithermal deposits and deep circulation involved in the formation of the mesothermal deposits. Although both styles of mineralization occur in the same tectonic regime, it appears that they are not tem- porally or genetically related to each other.
Introduction
A NUMBER of small precious and base metal deposits have been mined in the Okanagan Valley area of southern British Columbia since the end of the last
century. The three main gold-producing districts are Fairview, Oro Fino, and Dusty Mac (Fig. 1). The Fairview camp includes three deposits: Fairview, Stemwinder, and Morning Star. They were mined at the beginning of this century for several years, re- opened in the 1930s for a few years, and are currently being reconsidered for production. The total produc- tion, to date, is 150,000 metric tons of ore that yielded 510 kg of gold and 4,800 kg of silver (Hedley and Watson, 1945). The Oro Fino camp is located about 10 km northwest of Fairview and consists of two
properties, Oro Fino and Twin Lake, which have mining histories similar to those at Fairview. Recorded production was 24,000 metric tons of ore that yielded 270 kg of gold and minor silver (Hedley and Watson, 1945). The Dusty Mac mine operated from 1969 to 1976. Total production was 93,000 metric tons of ore, with recovery of 600 kg gold, 10,500 kg of silver, and minor copper and lead (Schroeter and Panteleyev, 1986).
There appears to be two distinctly different styles of precious metal mineralization in the Okanagan Valley area: mineralized quartz veins occurring in the Paleozoic and Mesozoic metamorphic rocks (Fairview
* Present address: Department of Earth and Atmospheric Sci- ences, Purdue University, West Lafayette, Indiana 47907.
and Oro Fino areas), and mineralized silicified zones occurring in Tertiary volcanic and volcaniclastic rocks (Dusty Mac area). Church (1973) suggested that min- eralization in the Dusty Mac deposit formed by the filling of dilation zones. Tempelman-Kluit (1984) proposed a meteoric water model for mineralization in the area. Nevertheless, the characteristics and or- igins of the ore-forming fluids and the relationships between the two styles of mineralization have re- mained unclear. The purpose of the present study is to investigate the nature of the fluids responsible for precious metal mineralization in the area and to ex- amine the relationships befween the two styles of mineralization. Fluid inclusion and stable isotope studies have been used to examine the temperature- pressure environment of vein formation and the ori- gins of the ore-forming fluids for both deposit types. The results show that each deposit type was formed by a distinct type of fluid. The origins of both types of deposits are related to the circulation of meteoric water, though in strikingly different geologic settings, which are similar to those described in Nesbitt et al. (1986).
Geologic Setting
There have been several geologic studies con- ducted in the Okanagan Valley area. The earliest sys- tematic mapping of the area was performed by Cairnes (1940) and Bostock (1941). Their work was revised by Little (1961). Church (1973) studied the geology of Tertiary White Lake basin in the Okanagan Valley.
0361-0128/89/915/410-15 410
Au MINERALIZATION STUDIES, OKANAGAN VALLEY, CANADA 411
-.":.• Tertiary volcanic ,& sedimentary rocks
*J-•J Jurassic Valhalla intrusions
::.•.._• J. urassi•: Nelson Intrusions
J•'•J Triassic Old Tom& Shoemaker Group
J• Carboniferous Kobau Group
•'.'.:• Shuswap Complex
/ Fault / Geological boundary •"•) Lake ß Mining District
0 15
km
FIG. 1. Simplified geologic map of Okanagan Valley, British Columbia. Data are from Little (1961), Church (1973), and Parrish et al. (1985). The locations of three major gold-producing camps (Dusty Mac, Oro Fino, and Fairview) are shown by solid circles.
The recent literature concerning this area includes geochronology of metamorphic and igneous rocks (Ross, 1974, 1981; Medford, 1975; Mathews, 1981; Parkinson, 1985), petrology of the plutonic rocks (Peto and Armstrong, 1976; Peto, 1979), and tecton- ics of the area (Tempelman-Kluit and Parkinson, 1986). The geology of mineral deposits in this area was described by Cockfield (1939), Hedley and Wat- son (1945), Church (1973), and Tempelman-Kluit (1984).
Regional geology
The Okanagan Valley is located near the boundary between the Omineca crystalline belt and Intermon- tane belt (Monger, 1984) and follows a gently west- dipping fault zone of Eocene age (Tempelman-Kluit and Parkinson, 1986), which extends for at least 100 km in an approximate north-south direction (Fig. 1). The Shuswap metamorphic complex, consisting mainly of gneisses, is restricted to the east side of the fault zone and is considered to be the oldest rock unit in this area with a possible age of Precambrian (Little, 1961). The Paleozoic and Mesozoic successions, which occur on the west side of the fault zone, are
also metamorphosed to varying degrees. Tertiary vol- canic and sedimentary rocks dip at comparatively low angles and mainly occur on the west side of the fault zone. The volcanic rocks, which are mostly andesitic, dacitic, and trachytic in composition, are similar to other Eocene volcanic rocks in British Columbia, Washington, and Montana (Church, 1973) and have K-Ar ages of about 50 m.y. (Parkinson, 1985). Two major groups of plutonic rocks, the Nelson and Val- halla complexes, intruded the Shuswap gneisses and the Paleozoic and Mesozoic metamorphic rocks. Geochronological studies by various authors (White et al., 1968; Medford, 1975; Peto and Armstrong, 1976; Parkinson, 1985) show that these rocks are Ju- rassic in age (140-180 m.y.).
An important structural feature of this area is a low- angle, north-trending, west-dipping, extensional fault zone which has been described by Tempelman-Kluit and Parkinson (1986). The upper plate of the fault zone contains Paleozoic to Tertiary formations and is broken by high-angle, northeast and easterly trending normal faults. The lower plate is composed mainly of Shuswap gneisses. The detachment shear zone be- tween the upper and lower plate is composed of my-
412 ZHANG, NESBITT, AND MUEHLENBACHS
lonitic gneiss. Large-scale (tens of kilometers) dis- placement between two plates occurred during the middle Eocene (Parkinson, 1985; Tempelman-Kluit and Parkinson, 1986).
Geology of the ore deposits
Fairview and Oro Fino camps: The geologic set- tings of the Fairview and Oro Fino camps are similar. Mineralization at the two sites is associated with
quartz veins and occurs in metamorphic rock units. At Fairview, the host units (Kobau Group) are a series of micaceous, chloritic or graphitic quartz schists of probable Carboniferous age (Cairnes, 1940; Little, 1961). At Oro Fino, the host rocks (Old Tom and Shoemaker Groups) are amphibolites and gneisses of Triassic age (Bostock, 1941; Little, 1961). Two plu- tonic units, the Fairview granodiorite and the Oliver granite, occur southwest and northeast, respectively, of the mineralized zone at Fairview (Figs. 1 and 2). The Oliver granite of the Valhalla intrusions has an age of 144 to 157 m.y. (White et al., 1968; Armstrong, unpub. data, cited in Parkinson, 1985). Although the Fairview granodiorite is regarded as a part of the older Nelson plutonic event, there are no published age dates to confirm this. Based on the observation that
the Fairview granodiorite possesses foliated textures and the Oliver granite does not, the Fairview grano- diorite appears to be pre- to synmetamorphic, while the Oliver granite is younger.
Numerous quartz veins, generally striking north- west and with varying widths and lengths, occur in the Fairview and Oro Fino area. The gold deposits are associated with one or more of the types of quartz veins. Three types of quartz veins are recognized:
A. Thick quartz veins. These veins are composed of massive white quartz with sporadic occurrences of sulfides and other minerals. They typically have a thickness greater than 1 m, up to as much as 9 to 10
FIG. 2. Geologic map of the Fairview camp. Data are from Little (1961), Parkinson (1985), and this study. Number = dip of angles of schistosity.
m. The veins usually parallel the regional schistosity. Veins of this type are the main ore-bearing unit.
B. Quartz veins with ribbon graphitic bands. These veins consist of white quartz with black graphitic bands producing a striking ribboned texture. The size of type B veins varies, but the width is usually less than 1 m. Like type A veins, these veins parallel the regional schistosity. Pyrite is sometimes abundant in this type of vein, but other sulfides are rarely ob- served. Type B veins are occasionally ore bearing.
C. Small quartz veins. Type C veins are relatively pure quartz veins. These veins are several centimeters to tens of centimeters thick. Unlike the other veins, these veins cut the schistosity at high angles. These veins are typically barren.
Type C veins generally postdate type A and B veins. In the Fairview deposit, the small quartz veins, type C, often cut type A and B veins. The temporal rela- tionship between type A and B veins is not clear, since they do not occur together in the field. However, they are probably contemporaneous, since both types of veins parallel the regional schistosity.
Sulfide minerals are irregularly distributed in the quartz veins as disseminated minerals, small knots, or small veins. It is estimated that the sulfide minerals
constitute less than 1 vol percent of the ore. Pyrite is the most abundant sulfide and is accompanied by small amounts of galena, sphalerite, and chalcopyrite, with traces ofbornite, tetrahedrite, argentitc, and boulan- gerite. Gold is reported to be principally associated with galena and sphalerite and not pyrite (Hedley and Watson, 1945). Several stages of mineral precipitation can be recognized. Stage 1 constitutes the period during which the main quartz vein and associated py- rite were precipitated. Stage 2A is the stage of pre- cipitation of chalcopyrite, sphalerite, pyrite, and mi- nor bornitc. During this substage, small veinlets of quartz + chalcopyrite _ pyrite _+ sphalerite cut through some stage 1 pyrite. Exsolution ofbornite in chalcopyrite indicates that bornitc belongs to this substage. Stage 2B is a stage of precipitation of galena, sphalerite, and minor chalcopyrite. During this sub- stage, veinlets of quartz + galena + sphalerite also cut stage 1 pyrite. In some cases, these veinlets dis- place stage 2A veinlets. Argentitc and boulangerite were precipitated during stage 3. Both minerals are closely related to, and in many cases replace, galena. The lack of visible gold makes it difficult to determine during which stage the gold precipitated, but if the gold is associated with the galena, it was probably deposited during stage 2B.
Silicification is the most obvious alteration feature
associated with the veins. The quartz in host rocks either increases in abundance or is recrystallized in the vicinity of the veins. Except for silicification, other styles of alteration, such as carbonatization, sericit- ization, and chloritization, are only recognized mi-
Au MINERALIZATION STUDIES, OKANAGAN VALLEY, CANADA 413
croscopically. Carbonatization and sericitization only occur within a few meters of the veins, but silicifi- cation may extend to a greater distance (10 m or more).
The age of the quartz veins has not been deter- mined by radiometric dating. Field geology suggests that they are postmetamorphic, since the veins were not subjected to synmetamorphic deformation.
Dusty Mac property: Dusty Mac is located at the eastern margin of the Tertiary White Lake basin (Fig. 1). The host rock of the ore is the White Lake For- mation, which is composed mainly of andesitic vol- caniclastics, lahars, and shales with thin seams of coal (Church, 1973). To the east, the White Lake For- mation overlies the Shuswap gneisses. A reverse fault system at Dusty Mac controls the location of miner- alization (Church, 1973). Brecciation and quartz veins are present in or adjacent to the fault.
The mineralized zone is a gently dipping lens of quartz breccia and veins with a varying admixture of crushed host rocks. The orebody is about 200 m long, 50 m wide, and 10 m thick (Church, 1973). Quartz breccia was apparent.•y a product of the crushing of earlier formed quartz veins. Numerous quartz veins, ranging from several millimeters to several centime- ters in width, cut the brecciated quartz, indicating that quartz veining continued syn- to postbrecciation. Precious metal mineralization is present in both brec- ciated and vein quartz. The metallic minerals, mainly pyrite with traces of chalcopyrite, bornitc, galena, sphalerite, and native silver, are disseminated throughout the brecciated quartz and small quartz veinlets. The major alteration features are silicifica- tion, chloritization, and carbonatization. Many of the brecciated rock fragments were strongly silicified. Chlorite forms as aggregates or small veinlets cutting through or replacing the brecciated quartz and rocks. Calcite and calcite-quartz veins cut through all the other phases.
Fluid Inclusion Studies
Techniques
Quartz samples from each deposit were selected for fluid inclusion studies. The criterion for selecting samples was that the quartz was paragenetically re- lated to mineralization or the quartz represented a particular stage of veining.
Generally, the fluid inclusions are very small (usu- ally 5-10 #m) in diameter. The criterion used for rec- ognition of primary inclusions was that the inclusions were distributed individually or in random clusters (Roedder, 1984). Chains of minute inclusions, which are usually less than 1 to 2 #m in diameter and are present along healed fractures, were considered to be secondary in origin.
Fluid inclusions were examined in thin (0.3-0.6 mm thick), doubly polished plates. Temperature de-
terminations were made using a Chaixmeca heating- freezing stage. The stage was calibrated using stan- dards provided by the manufacturer. Replicate mea- surements show that between the range of-60 ø and +30øC, the precision is _0.2øC. For higher temper- atures, homogenization temperatures of the inclusions were reproducible within _5øC, but some inclusions showed a larger range of _10øC.
Description of fluid inclusions
Three types of primary inclusions were recognized:
Type 1. CO2-H20 inclusions. These inclusions contain CO2 and an aqueous fluid. COz-HzO inclu- sions show two phases, COz liquid and HzO liquid, or three phases, CO2 liquid, COz vapor, and HzO liq- uid, at room temperature. For the two-phase COz- HzO inclusions, a third phase, CO2 vapor, forms when the inclusions are cooled to 5 ø to 15øC. The COz phase occupies variable percentages of the total vol- ume, ranging from 10 to 90 percent, but 20 to 30 percent is most common.
Type 2. COz inclusions. These inclusions contain relatively pure COz. At room temperature, the inclu- sions show one (COz liquid) or two (a CO2 vapor bub- ble enveloped by COz liquid) phases. A vapor phase would form when the one-phase COz inclusions were cooled. Unlike type 1 inclusions, no clathrate (COz. 5.75H•O) formed in type 2 inclusions when they were cooled, indicating that there is relatively little H20 present in these inclusions. It is probable that the COz inclusion is the end member of the COz- HzO inclusion system (type 1 inclusions), i.e., the pure COz phase, was a product of effervescence from the CO2-HzO fluid.
Type 3. Aqueous inclusions. This type of inclusion usually shows two phases (HzO liquid + HzO vapor) at room temperature. The vapor phase occupies less than 20 percent of the total inclusion volume. No daughter minerals were found in these inclusions.
Secondary fluid inclusions, classified as type 4 in- clusions, are present along the fractures and invariably filled by aqueous fluids. These inclusions are typically very small in size (usually less than 1 #m). Relative to primary inclusions, secondary inclusions are gen- erally not abundant, especially in Fairview and Oro Fino samples.
Temperature measurements
Five temperatures of phase transitions were de- termined: melting temperatures of solid COz (Tm½o,), clathrate (Tmc•athrate), and HeO (Tmice) and ho- mogenization temperatures of COe (Th½o,) and HeO- COe or HeO liquid-vapor (Th).
Melting temperatures of solid COe, clathrate, and ice were determined when the jagged meniscus sud- denly smoothed and disappeared. Homogenization of
414 ZHANG, NESBITT, AND MUEHLENBACHS
TABLE 1. Microthermometry Data for Fluid Inclusions in Quartz
CO2 vapor-liquid, H20 vapor-liquid, and H•O-CO• was determined when the boundary of two distinct fluid phases disappeared. Most of the CO• fluids ho- mogenized to a liquid phase. Thermal decrepitation was observed in some CO•-H20 inclusions at tem- peratures below homogenization, due to high internal pressures. Densities of CO• were derived from the equation of Angus et al. (1976), which gives data for liquid-vapor saturation of CO•. The homogenization temperatures are not corrected for pressure.
Most of the CO•-containing inclusions that were measured have melting temperatures ranging from -56.5 ø to -58.5øC (Table 1). A lowering of the melting temperature of solid CO• indicates the pres- ence of gas(es) other than CO2, probably CH4. Ac- cording to Swanenberg (1979), at the filling ratio CO•/(CO• + H•O) of 0.2 to 0.3 by volume, which is common in the type 1 inclusions from Fairview and Oro Fino, the solid CO• melting temperature of -57.5øC indicates a CH4 content of less than 5 per- cent in the CO2-CH4 system. Since most of the mea- surements of Tm,•,• are higher than -57.5øC, it is con- cluded that the •'O• phase in most of the inclusions is essentially pure CO•.
Salinity determinations were based on freezing point depressions of ice in the system of H•O-NaC1 (Potter et al., 1978) for aqueous inclusions and depressions of the melting point of clathrate in the system of H•O-CO2-NaC1 for CO•-H•O inclusions (Bozzo et al., 1973; Collins, 1979). Although the exact value of the melting temperature of CO• hydrate is questioned by some authors (e.g., Roedder, 1984), it nevertheless gives an estimate of the NaCI content. Since the CO• phase studied here is essentially pure, the influence of other gas(es) on the melting temper- ature of clathrate is small.
Fairview camp
In the Fairview camp, the dominant inclusion type is type 1--CO2-H•O inclusions. Only a few type 2 inclusions were noted. The average size of the CO•-
H•O inclusions is 5 to 10 tzm, with some as large as 20 tzm. A number of inclusions show variations in CO• contents ranging from 10 to 90 percent of the inclu- sion volume. This may indicate that trapping of het- erogeneous fluids has occurred (Ramboz et al., 1982; Roedder, 1984).
Samples from the Fairview property represent the three types of quartz veins. Among them, samples F- 1 through F-6, FA-1 through FA-6, and FA-30 are from type A veins; samples FA-18 through FA-24 represent type B veins; and FA-13-V and FA-14 rep- resent type C veins. Measurements show that CO2- HeO inclusions from the three types of veins have very similar properties (Table 2). Variance analysis on 267 homogenization temperatures of COe-HeO inclusions from the three types of quartz veins shows no significant differences among them. As a conse- quence, inclusions from the three types of quartz veins were considered statistically as a whole. Samples from the Morning Star and Stemwinder properties were randomly collected from various veins and waste dumps and may represent type A, B, or C veins.
The range of homogenization temperatures of COe- H20 inclusions are scattered considerably (Fig. 3), from 224 ø to 320øC for Stemwinder, 221 ø to 333øC for Morning Star, and 211 ø to 328øC for Fairview. In individual samples, most of the inclusions have a range of homogenitization temperatures of 50 ø to 80øC, with some as large as 100øC (e.g., 118øC for
TABLE 2. Comparison of Inclusions from Three Groups of Quartz, Fairview Deposit
B 3 40 210-320 271 23.5 7.5 C 2 52 220-328 276 24.6 7.8
416 ZHANG, NESBITT, AND MUEHLENBACHS
FA-18; 211ø-322øC for FA-4; 103øC for M-35). Three possible factors can be responsible for this be- havior: effervescence during trapping which caused heterogeneous trapping of the fluid, the presence of two or more generations of inclusions, and leaking of the inclusion after trapping. Since all the inclusions of this type that were measured were carefully se- lected and appear to be primary, the second possi- bility can be excluded. Leaking of the fluid inclusion may be partially responsible for this behavior, but most of the inclusions measured show little evidence
of fracturing, indicating that the contribution of leak- age to the large range of homogenization tempera- tures is probably limited. Thus, the most probable reason for the large range of homogenization tem- peratures is effervescence during trapping. This is consistent with the observation that variable CO2/ H20 ratios occur in these inclusions.
Figure 3 shows the distribution of homogenization temperatures of CO•-H•O inclusions from the three properties. The histograms indicate that CO•-H•O inclusions from the three properties have similar, nearly normal distributions of values for homogeni- zation temperatures, with modes and arithmetic av- erages between 270 ø and 285øC. It should be noted that even though the range of homogenization tem- peratures is large, most of the values are grouped in the interval of 260 ø to 310øC. It is thus reasonable
to take the range of 270 ø to 285øC as the estimate of the homogenization temperatures for the CO2-H•O inclusions.
[]type A quartz
[] type B quartz
45 a [•] type C quartz
3e Fairview / •3o // •\
•,o $
T øC
b
25 2o Morning Star /
•,o 5
•=o 200 220 240 2•0 28O 3OO 320 34O
T øC
Stemwinder /•
'-f o 200 220 240 2•o 28o 3o0 32o
T OC
FIG. 3. Histograms of homogenization temperature deter- minations on CO•-H•O inclusions from the Fairview camp. (a). Fairview deposit. Three types of quartz veins are included. The mean value of Th is 274øC and the mode is at 280øC. (b). Morning Star deposit. The vein types are not specified (see text). The mean value of Th is 282øC and the mode is at 285øC. (c). Stemwinder deposit. The vein types are not specified (see text). The mean value of T• is 270øC and the mode is at 275øC.
12 16 20 24 28
TOc
b
•[ Morning Star
o 12 16 20 24 26
TOc
c
TOc
FIG. 4. Histograms of homogenization temperatures of CO• phase in COa-HaO inclusions from the Fairview camp. (a). Fairview deposit. The mean value is 24.5øC and the mode is at 27.0øC. (b). Morning Star deposit. The mean value is 19.9øC and the mode is at 21.0øC. (c). Stemwinder deposit. The mean value is 22.9øC and the mode is at 25.0øC.
Two Type 3 aqueous inclusions have homogeni- zation temperatures at 275 ø and 313øC, respectively. These values are close to those estimated for CO•- H•O inclusions. Homogenization temperatures of type 4, secondary aqueous inclusions, range from 186 ø to 199øC.
Melting temperatures of clathrate in the CO•-H20 inclusions vary from 7.1 o to 8.9øC for Fairview, 6.6 ø to 9.2øC for Morning Star, and 7.5 ø to 9.0øC for Stemwinder. The arithmetic averages are 7.6 ø, 7.7 ø, and 8.5øC, respectively, which correspond to salini- ties of 3.0 to 4.7 equiv wt percent NaC1 equivalent.
Homogenization temperatures of the CO• phase in CO•-H•O inclusions are relatively widely distributed (Fig. 4) and the pattern of distribution is skewed to lower temperatures, which makes the modes 2 ø to 3øC higher than the arithmetic averages of 19.9 ø to 24.5øC. As described previously, most type 2 inclu- sions have relatively pure CO• compositions. Ho- mogenization temperatures of these inclusions range from 12.9 ø to 26.9øC, which are consistent with the homogenization temperatures of the CO2 phase in CO•-H•O inclusions, and result in a range of density estimates from 0.67 to 0.84 g/cm a, using the equation of Angus et al. (1976) for CO• in these inclusions. Considering that most of the measurements are in the range of 20 ø to 25øC, the density of the CO• is in the range of 0.71 to 0.77 g/cm a.
As suggested above, the entrapment of the fluids in CO2-H•O inclusions probably occurred during CO• effervescence (heterogeneous trapping). Thus, the homogenization temperatures of these inclusions represent their trapping temperature and no pressure correction for temperature is necessary (Roedder, 1984). It is possible to estimate the trapping pressure of the primary fluid inclusions using available data for
Au MINERALIZATION STUDIES, OKANAGAN VALLEY, CANADA 417
CO•. densities and trapping temperatures. As men- tioned above, CO•. inclusions are closely related to CO•.-H•.O inclusions and it is a good approximation to use the trapping temperature of CO•.-H•.O inclu- sions as that of the CO2 inclusions. Figure 5 shows that the pressure of entrapment is in the range of 980 to 1,250 bars, assuming that the trapping tempera- tures are 270 ø to 285øC and the CO•. densities are 0.71 to 0.77 g/cm a. It is important to have a relatively accurate estimate of the density of CO•. using this method. For the example above, a density difference of 0.05 g/cm a (5øC shift in homogenization temper- ature of CO•.) can produce a pressure shift of about 150 bars. The influence of errors in the estimates of
the trapping tempe.rature is much smaller (Fig. 5). The pressure of 980 to 1,250 bars corresponds to a depth of 3.5 to 4.5 km, assuming a lithostatic pressure gradient, or 10 to 13 km, using a hydrostatic pressure gradient (Haas, 1976).
Oro Fino camp
The samples used for study were collected from the veins near the collapsed shafts and waste dumps. Like the Fairview samples, CO•.-H•.O, type 1 inclu- sions are dominant. The inclusions are generally smaller than 10/•m in diameter. A few CO•. inclusions are present. At room temperature, the COe-H•.O in-
TEMPERATURE(•)
FIG. 5. P-T diagram for the CO2 system. Data are from Ken- nedy (1954). The closely spaced, hatched area represents the trapping temperature of 270 ø to 285øC and pressure of 980 to 1,250 bars for both CO2-H•O inclusions (homogenizing at 270 ø- 285øC) and CO• inclusions (homogenizing at 20ø-25øC, which corresponds to a CO• density of 0.71-0.77 g/cm a) from Fairview camp. The widely spaced hatched area represents the common trapping temperature of 294 ø to 308øC and pressure of 800 to 1,000 bars for CO• inclusions homogenizing at 28.5 ø _ 1.4øC, which corresponds to a CO2 density of 0.60 to 0.68/cm a, and CO•-H•O inclusions homogenizing at 294 ø to 308øC from Oro Fino camp.
clusions usually show three phases and the CO•. phase is about 20 to 30 vol percent of the inclusion. The vapor bubbles of aqueous inclusions occupy a similar percentage of the volume of the inclusion.
The range of homogenization temperatures for CO•.-H•.O inclusions varies from 264 ø to 349øC, with an arithmetic mean of 304øC. Except for sample O- 1, the samples show individual means from 294 ø to 308øC for CO•.-H•.O inclusions.
Melting temperatures of clathrates range from 6.3 ø to 7.9øC. The average of 6.9øC corresponds to a sa- linity of 5.9 equiv wt percent NaC1.
Homogenization temperatures of the CO•. phases range from 27.1 ø to 30.5øC, which corresponds to a density range for CO•. of 0.60 to 0.68 g/cm a (Angus et al., 1976).
Sample T-11 contains CO•.-H•.O, CO•, and sec- ondary aqueous fluid inclusions. Coexisting CO•. in- clusions and CO•.-H•.O can be used to estimate the pressure of trapping (Roedder, 1984). In sample T- 11, CO•. inclusions homogenize at 28.5øC (CO•. den- sity of 0.64 g/cm a) and CO•.-H•.O inclusions homog- enize at 294øC. This indicates a trapping temperature and pressure of approximately 294øC and 800 to 1,000 bars (Fig. 5). This pressure corresponds to a depth of 2.9 to 3.6 km, assuming a lithostatic pressure gradient, or 8.5 to 10.5 km, assuming a hydrostatic pressure gradient (Haas, 1976). The pressure esti- mated here is similar to but slightly lower than that estimated for Fairview.
Dusty Mac property
Samples were collected from the orebody and waste dumps. Both brecciated and vein quartz contain a small number of aqueous inclusions. The inclusions are usually small in size, with some up to 10 to 15/•m in diameter. The vapor bubble occupies only about 5 to 10 percent of the inclusion volume in samples from brecciated quartz (D-13, D-28, and D-29-B) and about 10 to 20 percent for samples from vein quartz (D-14-V, D-15, and D-29-V). No evidence of the presence of CO•. was observed in any inclusions from Dusty Mac. The results (Table 1) show that the prop- erties of the inclusions are slightly different between the two kinds of quartz.
Homogenization temperatures of primary inclu- sions range from 243 ø to 260øC with an average of 252øC for vein quartz and 221 ø to 240øC with an average of 232øC for brecciated quartz. This suggests either that the vein quartz was formed from a some- what hotter fluid than brecciated quartz or that the temperatures of the fluids were essentially the same but under different pressures during the formation of the quartz. Homogenization temperatures of a limited number of secondary inclusions range from 131 ø to 209øC with an average of 184øC.
Samples from the two types of quartz usually show
418 ZHANG, NESBITT, AND MUEHLENBACHS
a narrow range of -0.3 ø to -0.5øC for the melting temperatures of ice, which corresponds to salinities of 0.5 to 0.9 equiv wt percent NaC1. The only excep- tion is sample D-14-V, which has a lower melting temperature for ice, corresponding to a salinity of 2.6 equiv wt percent NaC1.
At 250øC, the saturated vapor pressure for a 2 wt percent NaC1 solution is 39.3 bars, which corresponds to a depth of 450 to 460 m, assuming a hydrostatic pressure gradient (Haas, 1971, 1976), or 140 m, as- suming a lithostatic pressure gradient. These pressure and depth estimates are minimum values since the fluid inclusions show no evidence of boiling. The cor- responding pressure correction for the homogeniza- tion temperatures is less than 10øC (Potter, 1977).
Stable Isotope Studies
Techniques
The oxygen, hydrogen, and carbon isotope com- positions of whole-rock samples and mineral separates from the deposits were analyzed to define the isotopic characteristics of hydrothermal alteration and the or- igins of the hydrothermal fluids. Except for some pure quartz samples, all quartz samples used for oxygen isotope study were treated with hydrochloric acid at room temperature for one day or more to remove carbonate mineral and then handpicked under a bin- ocular microscope. The purity of the mineral sepa- rates is better than 95 percent.
Hydrogen and carbon isotope determinations were made on fluids released by thermal decrepitation of fluid inclusions in quartz. Cryogenic techniques were used to separate CO2 from water. The CO2 was di- rectly analyzed for $•aC. The water samples were placed in glass, break-seal tubes and sent to the Global Geochemistry Company for $D analyses. Samples for fluid extraction were selected on the basis of the
highest proportion of primary inclusions relative to secondary inclusions. Even with careful selection, many secondary inclusions were present in the sam- ples used for the analyses; however, since the ore for- mation event is a multistage process, many of the sec- ondary inclusions probably host ore-related fluids. Nesbitt et al. (1987) present a more thorough dis- cussion of the reliability of the/iD and •aC deter- minations from extracted fluids.
Oxygen was released from quartz, silicates, and whole-rock samples by reaction with BrF5 (Clayton and Mayeda, 1963) and then converted to COe by reaction with hot carbon. The precision of isotopic analyses as indicated by replicate analyses on several quartz samples is _+0.2 per mil. The isotopic ratios are reported in the usual $ notation in per mil relative to SMOW for oxygen and hydrogen and PDB for car- bon (Craig, 1957, 1961).
Fairview and Oro Fino camps
Oxygen isotope results: Thirty-nine oxygen isotope analyses were made on vein quartz from five prop- erties. They show a range of $•80 values of 10.8 to 14.0 per mil (Table 3), but most of the values are in the range of 13 to 14 per mil for the Fairview deposits and 11.8 to 12.7 per mil for the Oro Fino deposits (Fig. 6).
At the Fairview property, five quartz samples (FA- 1, 2, 3, 4, and 8) from the orebody, type A veins have $•sO values of 12.9 to 13.9 per mil; the other five samples of type A quartz (F-l, 2, 4, 10, and 20) possess $•sO values of 11.8 to 13.8 per mil. Four samples from type B quartz (FAo18, 19, 23, and 24) have values of 13.1 to 13.5 per mil, and the $•sO values for three samples of type C quartz (FA-9, 13-V, and 14) range from 12.4 to 13.6 per mil. A sample (F-32) from a small quartz vein (10 mm wide) in micaceous quartz schist about 300 m northwest of the orebody has a •sO value of 12.3 per mil. The results show that oxygen isotope ratios among the three types of vein quartz are essentially identical.
At the Morning Star property, four quartz samples (M-2, 16, 29, and 35) from the mineralized vein have $•80 values of 13.4 to 14.0 per mil; three samples (MP-3, 5, and 6) collected from waste dumps have similar $•sO values of 13.3 to 13.9 per mil. Two sam- ples (M-3 and M-12) from a quartz vein (0.5-1.0 m wide) 100 m east of the shaft have similar $•sO values between 13.3 and 13.6 per mil. Sample M-8-V-Q is a quartz sample separated from a small quartz-calcite vein (5 mm wide) in chlorite schist about 250 m northeast of the shaft and has a $•sO value of 12.8 per mil. Five quartz samples (S-1, 3, 8, 12, and SP-6) from the Stemwinder property possess nearly identical oxygen isotope compositions with $•sO val- ues of 13.2 to 13.4 per mil.
The similarity of $•80 values and fluid inclusions among the different quartz veins throughout the area indicates that the hydrothermal fluids responsible for deposition of vein material were genetically related to each other. Given the observed $•sO values of 13 to 14 per mil for the quartz and a temperature of 280øC for its precipitation, the $•sO value of the hy- drothermal fluid was between 5.3 and 6.3 per mil, using the quartz-water fractionation curve of Clayton et al. (1972).
Three samples (O-1, 8, and 11) of vein quartz from the Oro Fino property have a narrow range of $•sO values of 11.8 to 12.5 per mil, and three samples (T-10, 11, and 12) of vein quartz from Twin Lake Property have similar $•sO values of 10.8 to 12.7 per mil (Table 3 and Fig. 6). Fluid inclusion studies have shown a formation temperature of 340øC for these veins. The calculated $•80 values of the fluids are from 4.7 to 6.6 per mil, which are similar to those from
Au MINERALIZATION STUDIES, OKANAGAN VALLEY, CANADA 419
TABLE 3. Isotopic Compositions of Samples from Fairview and Ore Fine Camps
Sample no. Mineral or rock b•SO,•mp•e •lSOfiui d bDttuid •13Cfiuid T(øC) •
(%0) FIG. 6. Histogram of vein quartz 6•SO values showing the dis-
tribution of 6•sO values for deposits of both epithermal (Dusty Mac) and mesothermal (Fairview and Oro Fino) styles of miner- alization. The 6•SO values are grouped at I to 3 per mil for epi- thermal quartz and 12 to 14 per mil for mesothermal quartz.
the Fairview area and indicate a genetic relationship between the ore fluids in the Oro Fino and Fairview areas.
A number of whole-rock and mineral separates from the metamorphosed host units have been ana- lyzed. Two silicified, micaceous, quartz schist samples (FA-13 and FA-17) from the Fairview ore zone yield •i•sO values of 11.7 and 14.1 per mil; three samples (M-5, M-6, and M-7) from the same rock unit but outside of the ore zone and with no alteration give higher •i•sO values of 13.0 to 16.7 per mil. The •i•SO values of three samples of chlorite schist from 250 to 500 m away the ore zone range from 5.7 to 8.3 per mil. The sample (M-8) with the lowest •i•SO value (5.7%0) is cut by a quartz-calcite vein and may be al- tered by hydrothermal fluids. The unaltered samples (M-5, M-6, M-7, M-9, and M-10) probably represent the general background values of the metamorphic unit.
Quartz separates from the micaceous quartz schists yield results which are similar to the results for the whole-rock analyses. The •i•SO values for quartz sep- arates, FA- 11-Q and FA- 13-Q, from altered units are 13.0 to 13.6 per mil, which are consistent with the •ilSo values of vein quartz. The •i•sO values for quartz separates, M-6-Q and M-7-Q, from unaltered rock are 15.1 to 16.6 per mil, which are 2 to 3 per mil higher than •i•SO values of vein quartz.
In sample M-7, coexisting quartz and biotite have a A value of 7.1 per mil. This yields a calculated equi- librium temperature of 425 ø __+ 25øC, using both the empirical, quartz-biotite fractionation factor of O'Neil and Ghent (1975) and the quartz-biotite equation of Javoy (1977). This temperature is reasonable given the local greenschist metamorphic grade. Using this temperature and oxygen isotope data for metamorphic quartz (M-6-Q and M-7-Q), the calculated •i•SO value of the corresponding metamorphic fluid is 11.6 to
13.1 per mil (Clayton et al., 1972), which is 6 to 7 per mil higher than that of the hydrothermal fluid.
Carbon and hydrogen isotope results Three samples extracted from fluid inclusions, two
from Fairview (FA-13-V and FA-24) and one from Morning Star (M-16), were used for carbon and hy- drogen isotope analyses. The results show a uniform •i•aC value of-8.5 __+ 0.4 per mil and somewhat scat- tered •iD values of-121 and -148 per mil.
Various authors (e.g., Hoefs and Morteani, 1979; Ohmoto and Rye, 1979; Kreulen, 1980; Hoefs, 1987) have suggested that there are three possible sources for CO2 in hydrothermal fluids: carbonate rocks with •i•3C values usually near zero, juvenile or homoge- nized crustal carbon with •i•SC values between -3 and -8 per mil, and organic carbon with •i•aC values typically lower than -20 per mil. Dissolution and de- carbonation of carbonate rocks should produce •i•sC values for CO2 in the hydrothermal fluid of zero per mil or greater (Shieh and Taylor, 1969; Ohmoto and Rye, 1979). It is therefore unlikely that carbonate was the principal carbon source for the hydrothermal fluids in the Fairview deposits. The lack of carbonate units in the stratigraphy in a large part of the Okan- agan Valley (Little, 1961) also indicates the improb- ability of a carbonate carbon source for the hydro- thermal fluid. Although the samples analyzed show lighter •i•sC values than that of juvenile carbon, we cannot totally exclude a juvenile origin. However, the observed carbon isotope data can be best explained by a reduced carbon source. The oxidation reaction (C + O2 -- CO2) and hydrolysis reaction (2H20 + 2C = CO2 -{- CH4) of reduced carbon could produce the •i•aC values observed in the Fairview deposits. The existence of graphite in the host-rock units is also strong evidence for a reduced carbon origin. The fluid inclusion study shows that a CH4 component may exist in the fluid, which may have been the product of hy- drolysis of graphite. Phase relations of the C-H-O sys- tem also show that, at 300øC and 1,000 bars, graphite- CO2-H20-CH4 may coexist in equilibrium (Holloway, 1984).
The $D values oœ the fluids œrom fluid inclusions
are similar to the local, present-day, meteoric water value (Fig. 7). Although only two values are reported (-121 and -148%0), they are unambiguously diag- nostic of meteoric water. The results of the •i•sO stud- ies, on the other hand, clearly indicate that the ore- forming fluids were strongly enriched in •SO relative to local meteoric water. Consequently, it is concluded that the hydrothermal fluids responsible for quartz veins and mineralization were derived, or mainly de- rived, from •SO-enriched, meteoric water. Studies of other Cordilleran mesothermal deposits (Nesbitt et al., 1986) as well as Korean mesothermal deposits
Au MINERALIZATION STUDIES, OKANAGAN VALLEY, CANADA 421
o /e SMOW -40 q:.•// •TER
/.,./o'•// .•// .•_.. [ MESOTHERMAL
-,•o '•.7//g.,?, l ,,-%, • FLUID / \ EPIT•ERMAL k,
-160 , /, , "%, -FLUID '< J -30 -20 -,o o +,o +20
6•8 0
FIG. 7. A (•]SO-(•D diagram shows that measured (•D values of both epithermal and mesothermal fluids are close to that of pres- ent-day meteoric water, whereas the (•tSo values are shifted-- slightly for epithermal fluids and substantially for mesothermal fluids. Shown are schematic (•tSo-(•D fields œor epithermal fluids and mesothermal fluids.
(Shelton et al., 1988) yield similar conclusions as to the origins of fluids responsible for mineralization.
Dusty Mac property
Quartz samples from the Dusty Mac property have a narrow range of b•80 values from 0.8 to 2.8 per mil (Table 4) and (Fig. 6). The results show little differ- ence between vein quartz (0.8-1.3%0) and brecciated quartz (1.6-2.8 %0). Using fluid inclusion temperatures and the quartz-water fractionation equation (Clayton et al., 1972), the calculated b•80 values of the quartz- forming fluids are -7.1 to -8.2 per mil for vein quartz and -7.4 to -8.9 per mil for brecciated quartz. These calculations show that the two types of quartz were equilibrated with isotopically similar fluids. Increasing the temperature by 20øC (pressure correction for homogenization temperature of 20 øC corresponds to a pressure of 250 bars), the $•sO values would increase by about 1.0 per mil for the fluids.
Two samples of fluid extracted from fluid inclusions from Dusty Mac have identical $D values of-133 per mil, which is nearly the same as the present-day me- teoric water value (Fig. 7). The oxygen and hydrogen isotope data can only be explained by the involvement of relatively pristine meteoric water in the ore-form- ing process.
Four whole-rock samples (D-3, D-4, D-9, and D- 14) from the vicinity of the Dusty Mac ore have $•sO values of-1.1 to -2.7 per mil, which are very similar to values for two samples (DA-3 and DB-24, -0.8 to -3.7%0) collected 500 m away from the ore. This in- dicates that oxygen isotope exchange between the low $]sO fluid and the rocks was established not only in the vicinity of the orebody but to a considerable dis- tance away from the ore. The widespread reduction in b]sO values in the rocks, -0.8 to -3.7 per mil com- pared with 6.5 _ I per mil values for "normal" an-
desite (Taylor, 1974), indicates that a large amount of meteoric water was involved in water-rock inter-
action. Making reasonable approximations of the $•80 values (initial rock -- 6.5%0, final rock = -2.0%0, and initial water -- -15%0) and temperature (T -- 250øC) and taking into account that the $•80 of rocks is equal to the $•sO of plagioclase (An3o) at equilibrium, the calculated water/rock ratio is between 1.1 and 1.7 by volume (Taylor, 1979).
Genetic Models for Mineralization
From a comparison of the isotopic characteristics of the ore-forming fluids (Fig. 7), two distinct min- eralizing processes can be recognized. At Dusty Mac, the fluids had extremely low salinities, low $•sO and bD values, and very low CO2 contents. At Fairview and Oro Fino, the fluids had low salinities and $D values but substantially higher •sO values and CO2 contents. Two general terms, epithermal (Dusty Mac) and mesothermal (Fairview and Oro Fino) are used here to describe these two styles of mineralization in the Okanagan Valley area.
Epithermal mineralization Numerous studies of Tertiary, epithermal, precious
metal deposits (Heald et al., 1987) have shown that in this type of deposit, fluid inclusions usually have homogenization temperatures of 200 ø to 300øC and salinities of 0 to 3 equiv wt percent NaC1. Stable iso- tope studies generally indicate the involvement of relatively pristine meteoric water in the formation of this type of deposit (O'Neil and Silberman, 1974; Field and Fifarek, 1985). The data from fluid inclu- sions and stable isotopes from Dusty Mac are consis- tent with those of Tertiary, epithermal, Au-Ag vein deposits described by Heald et al., (1987). It is gen- erally accepted that this type of deposit has formed at a relatively shallow depth (• 1 km, Heald et al., 1987). A genetic model involving the shallow circu-
TABLE 4. Oxygen and Hydrogen Isotope Compositions of Samples from Dusty Mac Property
Sample Mineral no. or rock •18Osample bl 8Ofluid •Dfluid T(øC) l
lation of meteoric water in the formation of epithermal deposit has been extensively used for such deposits and can be applied to Dusty Mac. As demonstrated above, a large amount of meteoric water flowed through the country rocks at Dusty Mac. A rough cal- culation indicates that only 0.5 ppb Au needed to have been removed from a rock volume of 0.5 km 3
to account for the total gold content of the Dusty Mac deposit. Meteoric water penetrated the volcanic pile through fractures and various fault systems. Heated by cooling volcanic and plutonic units, the tempera- ture of the fluids increased causing an increased sol- ubility of metals in the water. The fluids migrated upward and entered the local fault system, which be- came discharge channel(s). Fluid activity occurred at Dusty Mac in two discrete periods. The first period is related to the formation of the main silicification
zone; the second period is related to the brecciation and is responsible for the small quartz veins. Both of these units host mineralization.
Mesothermal mineralization
Many lode gold deposits occurring in Paleozoic or Mesozoic metamorphic rocks of the North American cordillera possess geological and geochemical char- acteristics which are similar to those of the Fairview
and Oro Fino deposits (Nesbitt et al., 1986). Various workers (Bohlke and Kistler, 1986; Taylor, 1987; Weir and Kerrick, 1987) have shown that the fluids responsible for the Mother Lode system of California were CO2 rich, with c•80 values of 6 to 16 per mil and c•D values of-50 to -80 per mil. Most of the authors also considered the fluids to have been de-
rived by metamorphism. A few researchers (Bohlke and Kistler, 1986) suggested that meteoric water had been involved in late-stage mineralization. However, some gold deposits in low- to intermediate-grade metamorphic rocks in Canadian cordillera show lower c•D (-80 to -160%o) values for the fluids (Nesbitt et al., 1986) than do the Mother Lode deposits. The de- creasing c•D values with increasing latitude of these ore-forming fluids is parallel to the c•D trend for me- teoric water. Based on this c•D pattern, Nesbitt et al. (1986) argue that deep circulation of the meteoric water was responsible for the formation of this type of deposit.
The present study of Fairview and Oro Fino shows that these deposits are similar in many aspects to the Mother Lode, California, gold deposits. However, the stable isotope results indicate that fluids responsible for mesothermal mineralization in the Okanagan area had a meteoric origin, or at least, the fluids were mainly derived from meteoric water. As indicated by the fluid inclusion study, the fluids descended to more than 10 km below the surface. This relatively long circulation path exposed the fluids to a large volume of rock producing low water/rock ratios. Isotopic ex-
change with the rock units would enrich the fluids in •80 but the •iD values would change relatively little, due to the limited hydrogen reservoir in the rocks.
An advantage of the model involving meteoric wa- ter is that it can best explain the large amounts of fluid involved in the ore formation process. A simple calculation provides an estimation of the mass of water involved in quartz precipitation for the main ore vein at Fairview. The solubility of SiO2 in water has been thoroughly studied (see Holland and Malinin, 1979). Assuming a reasonable thermal gradient (35øC/km), a hydrostatic pressure gradient, and a specific gravity of 2.5 g/cm 3 for quartz, the required volume of water for the formation of a vein of 200 X 200 X 5 m a (a size similar to that of Fairview main vein) should be 5 X 108 m a as the temperature decreases from 400 ø to 300øC (Holland and Malinin, 1979). Considering that many quartz veins exist in this area, an even larger quantity of water would have had to have been in- volved in their formation.
The model of dehydration during metamorphism for ore-forming fluids of the Mother Lode deposits is challenged by the fact that the age of mineralization is much younger than the last metamorphism in the area (Bohlke and Kistler, 1986; Weir and Kerrick, 1987). In Fairview and Oro Fino, the absolute age differences between mineralization and the last metamorphic event are not well established. But field observations at Fairview indicate that the quartz veins have not been subjected to metamorphism. In addi- tion, the oxygen isotope data on vein quartz and quartz separated from metamorphosed host rocks show different isotopic properties for the two fluids.
Templeman-Kluit (1984) first suggested that a ge- netic link may exist between the mesothermal deposits of the Fairview district and Eocene extensional fault- ing in the Okanagan area. Shown in Figure 8 is a sche- matic model of possible fluid flow in relation to ex- tensional tectonism and mesothermal mineralization. As extension is initiated, crustal thinning and detach- ment faulting bring the deep-seated, relatively high temperature lower plate upward to shallower levels causing an elevated geothermal gradient and a shal- lower brittle-ductile transition (Davis et al., 1986; Spencer and Welty, 1986). In the Okanagan area the
FIG. 8. Schematic mineralization model for mesothermal fluids, showing the extensional fault system and detachment zone as channels for deep circulation of meteoric water. The arrows in- dicate convecting water paths.
Au MINERALIZATION STUDIES, OKANAGAN VALLEY, CANADA 423
geothermal gradient during extension has been esti- mated to have been between 37 ø and 100øC/km (Mathews, 1981). This high geothermal gradient will tend to create hydrothermal convection in the normal fault zones of the upper plate.' Influx of cold water occurs on the flanks of the extensional zone and up- welling near the center (Reynolds and Lister, 1987). Consequently, mineralization will be largely confined to a zone near the central break in the upper plate. Such a distribution is observed in sites of mesothermal
mineralization in the Okanagan with mineralization from the Fairview district to the north along the edge of the upper block (Fig. 1).
The actual deposition of vein mineralization in the Fairview district was probably the result of cooling of the fluid, effervescence of COg and H•S, and sulfide precipitation as the fluids rose through the upper block. All of these processes would lead to a decreas- ing solubility of Au in the fluid and eventual saturation with respect to Au (Seward, 1984).
Conclusions
This study shows that the gold deposits in Okanagan Valley area can be divided into two groups: epithermal deposits represented by Dusty Mac, and mesothermal deposits represented by Fairview and Oro Fino. A detailed study of fluid inclusions and stable isotope ratios of the minerals and rocks reveals that:
1. The epithermal mineralization formed in a rel- atively low temperature (230 o_250 o C) and pressure environment. The fluids responsible for mineraliza- tion possessed very low salinities (0.5-0.9 equiv wt % NaCI) and had i5•sO = -8 to -9 per mil and iSD = -133 per mil.
2. The mesothermal mineralization formed in a
relatively high temperature (280ø-330øC) and high pressure (1 kbar or more) environment. The fluids responsible for mineralization were low in salinity (3- 6 equiv wt % NaCI) and had a high COg content. Stable isotope compositions of the fluids were •i•sO = 4 to 6 per mil, •iD = -121 to -148 per mil, and •i•aC = -8.5 per mil.
The data suggest that both ore-forming fluids were derived from meteoric water, with shallow circulation involved in the formation of the epithermal deposit and deep circulation in the mesothermal deposits. The relationships between these two styles of mineraliza- tion in the Okanagan Valley are not well understood. From the models of mineralization, we have noted that the shallow circulation of meteoric water is prob- ably closely related to igneous extrusion and/or in- trusion, but the deep circulation of meteoric water may be related to regional tectonics. Although both igneous activity and extensional faulting occurred in the early Tertiary in this area, much of the volcanic activity apparently postdated the extensional faulting,
since the volcanic rocks lie directly on exposed gneisses. Consequently, it appears that the two styles of mineralization are neither temporally nor geneti- cally related to each other.
Acknowledgments
Thanks are given to D. Rucker for his help during field work, D. Tempelman-Kluit for his initial intro- duction of the gold deposits in Okanagan Valley, and E. Toth for her lab assistance. P. Lhotka reviewed an earlier version of the manuscript. Financial support was provided by NSERC Strategic Grant #Gl107, and NSERC Operating Grant #A8003.
February 5, September 23, 1988 REFERENCES
Angus, S., Armstrong, B., deReuck, K. M., Altunin, V. V., Ga- derskil, O. G., Chapela, G. A., and Rowlinson, J. S., 1976, In- ternational thermodynamic tables of the fluid state, vol. 3, Car- bon dioxide: Oxford, Pergamon Press, 385 p.
B•hlke, J. K., and Kistler, R. W., 1986, Rb-Sr, K-Ar, and stable isotope evidence for the ages and sources of fluid components of gold-bearing quartz veins in the northern Sierra Nevada foothills metamorphic belt, California: ECON. GEOL., v. S 1, p. 296-322.
Bostock, H. S., 1941, Okanagan Falls, British Columbia: Canada Geol. Survey Map 627A.
Bozzo, A. T., Chen, H. S., Kass, J. R., and Barduhn, A. J., 1973, The properties of the hydrates of chlorine and carbon dioxide: Internat. Symposium Fresh Water from the Sea, 4th, Amaro- lassion, Greece, 1973, v. 3, p. 437-451.
Cairnes, C. E., 1940, Kettle River (west half), British Columbia: Canada Geol. Survey Map 538A.
Church, N. B., 1973, Geology of the White Lake basin: British Columbia Dept. Mines Petroleum Resources Bull., v. 61, 120 P.
Clayton, R. N., and Mayeda, T. K., 1963, The use of bromine pentafiuoride in the extraction of oxygen from oxides and sili- cates for isotopic analysis: Geochim. et Cosmochim. Acta, v. 27, p. 43-52.
Clayton, R. N., O'Neil, J. R., and Mayeda, T. K., 1972, Oxygen isotope exchange between quartz and water: Jour. Geophys. Research, v. 77, p. 3057-3067.
Cockfield, W. E., 1939, Lode-gold deposits of Fairview camp, camp McKinney and Videtie area, and the Divident Lakeview property near Osoyoos, B.C.: Canada Geol. Survey Mem., v. 179, 38 p.
Collins, P. L. F., 1979, Gas hydrates in CO2-bearing fluid inclu- sions and the use of freezing data for estimation of salinity: ECON. GEOL., v. 74, p. 1435-1444.
Craig, H., 1957, Isotopic standards for carbon and oxygen and correction factors for mass spectrometric analysis of carbon dioxide: Geochim. et Cosmochim. Acta, v. 12, p. 133-149.
-- 1961, Standard for reporting concentrations of deuterium and oxygen-18 in natural waters: Science, v. 133, p. 1833- 1834.
Davis, G. A., Lister, G. S., and Reynolds, S. J., 1986, Structural evolution of the Whipple and South Mountains shear zones, southwestern United States: Geology, v. 14, p. 7-10.
Field, C. W., and Fifarek, R. H., 1985, Light stable-isotope sys- tematics in the epithermal environment: Rev. Econ. Geology, v. 2, p. 99-128.
Haas, J. L., Jr., 1971, The effect of salinity on the maximum ther- mal gradient of a hydrothermal system at hydrostatic pressure: ECON. GEOL., v. 66, p. 940-946.
4 2 4 ZHANG, NESBITT, AND MUEHLENBACHS
-- 1976, Physical properties of the coexisting phases and ther- mochemical properties of the H20 component in boiling NaCI solutions: U.S. Geol. Survey Bull. 1421-A, 73 p.
Heald, P., Foley, N. K., and Hayba, D. O., 1987, Comparative anatomy of volcanic-hosted epithermal deposits: Acid-sulfate and adularia-sericite types: ECON. GEOL., v. 82, p. 1-26.
Hedley, M. S., and Watson, K. D., 1945, Lode-gold deposits, cen- tral southern British Columbia: British Columbia Dept. Mines Bull., no. 20, pt. 3, 27 p.
Hoefs, J., and Morteani, G., 1979, The carbon isotopic composition of fluid inclusions in Alpine fissure quartz from the western Tauern window (Tyrol, Austria): Neues Jahrb. Mineralogie Monatsh., 1979, no. 3, p. 123-134.
Holland, H. D., and Malinin, S. D., 1979, The solubility and oc- currence of non-ore minerals, in Barnes, H. L., ed., Geochem- istry of hydrothermal ore deposits: New York, Wiley Intersci., p. 461-508.
Holloway, J. R., 1984, Graphite-CH4-HaO-COa equilibria at low- grade metamorphic conditions: Geology, v. 12, p. 455-458.
Javoy, M., 1977, Stable isotopes and geothermometry: Geol. Soc. London Jour., v. 113, p. 609-636.
Kennedy, G. C., 1954, Pressure-volume-temperature relations in COa at elevated temperatures and pressures: Am. Jour. Sci., v. 252, p. 225-241.
Kreulen, R., 1980, COa-rich fluids during regional metamorphism on Naxos (Greece): Carbon isotopes and fluid inclusion: Am. Jour. Sci., v. 280, p. 745-771.
Little, H. W., 1961, Geology, Kettle River (west hal0, British Columbia: Canada Geol. Survey Map 15-1961.
Mathews, W. H., 1981, Early Cenozoic resetting of potassium- argon dates and geothermal history of north Okanagan area, British Columbia: Canadian Jour. Earth Sci., v. 18, p. 1316- 1319.
Medford, G. A., 1975, K-Ar and fission track geochronometry of an Eocene thermal event in the Kettle River (west hal o map area, southern British Columbia: Canadian Jour. Earth Sci., v. 12, p. 836-843.
Monger, J. W. H., 1984, Cordilleran tectonics: A Canadian per- spective: Geol. Soc. France Bull., v. 7, p. 255-278.
Nesbitt, B. E., Murowchick, J. B., and Muehlenbachs, K., 1986, Dual origins of lode gold deposits in the Canadian Cordillera: Geology, v. 14, p. 506-509.
Nesbitt, B. E., Muehlenbachs, K., and Murowchick, J. B., 1987, Dual origins of lode gold deposits in the Canadian Cordillera: Reply to comment by W. J. Pickthorn, R. J. Goldfarb, and D. L. Leach: Geology, v. 15, p. 472-473.
Ohmoto, H., and Rye, R. O., 1979, Isotopes of sulfur and carbon, in Barnes, H. L., ed., Geochemistry of hydrothermal ore de- posits: New York, Wiley Intersci., p. 509-567.
O'Neil, J. R., and Ghent, E. D., 1975, Stable isotope study of coexisting metamorphic minerals from the Esplanade Range, British Columbia: Geol. Soc. America Bull., v. 86, p. 1708- 1712.
O'Neil, J. R., and Silberman, M. L., 1974, Stable isotope relations in epithermal Au-Ag deposits: ECON. GEOL., v. 69, p. 902- 909.
Parkinson, D. L., 1985, U-Pb geochronometry and regional geol- ogy of the southern Okanagan Valley, British Columbia: The west boundary of a metamorphic core complex: Unpub. M.Sc. thesis, Vancouver, Univ. British Columbia, 149 p.
Parrish, R., Carr, S., and Parkinson, D. L., 1985, Metamorphic complexes and extensional tectonics, southern Shuswap Com- plex, southeastern British Columbia, in Tempelman-Kluit, D., ed., Field guides to geology and mineral deposits in the southern Canadian Cordillera: Geol. Soc. America, Cordilleran Sec. Mtg., Vancouver, British Columbia, 1985, p. 12-1-12-15.
Peto, P., 1979, On discerning magmatic populations in the batho-
lith west of Okanagan Lake: Chem. Geology, v. 26, p. 249- 265.
Peto, P., and Armstrong, R. C., 1976, Strontium isotope study of the composite batholiths between Princeton and Okanagan Lake: Canadian Jour. Earth Sci., v. 13, p. 1577-1583.
Potter, R. W., 1977, Pressure corrections for fluid-inclusion ho- mogenization temperatures based on the volumetric properties of the system NaCI-HaO: U.S. Geol. Survey Jour. Research, v. 5, p. 603-607.
Potter, R. W., II, Clynne, M. A., and Brown, D. L., 1978, Freezing point depression of aqueous sodium chloride solutions: ECON. GEOL., v. 73, p. 284-285.
Ramboz, C., Pichavant, M., and Weisbrod, A., 1982, Fluid im- miscibility in natural processes: Use and misuse of fluid inclusion data, II. Interpretation of fluid inclusions data in terms of im- miscibility: Chem. Geology, v. 37, p. 29-48.
Reynolds, S. J., and Lister, G. S., 1987, Structural aspects of fluid- rock interactions in detachment zones: Geology, v. 15, p. 362- 366.
Roedder, E., 1984, Fluid inclusions: Rev. Mineralogy, v. 12, 644 p.
Ross, J. V., 1974, A Tertiary thermal event in south-central British Columbia: Canadian Jour. Earth Sci., v. 11, p. 1116-1122.
-- 1981, A geodynamic model for some structures within and adjacent to the Okanagan Valley, southern B.C.: Canadian Jour. Earth Sci., v. 18, p. 1581-1598.
Schroeter, T. G., and Panteleyev, A., 1986, Gold in British Co- lumbia: British Columbia Ministry Energy, Mines Petroleum Resources, Prelim. Map no. 64.
Seward, T. M., 1984, The transport and deposition of gold in hydrothermal systems, in Foster, R. P., ed., Gold '82: Rotterdam, A. A. Balkema Pub., p. 165-181.
Shelton, K. L., So, C.-S., and Chang, J.-S., 1988, Gold-rich me- sothermal vein deposits of the Republic of Korea: Geochemical studies of the Jungwon gold area: ECON. GEOL., v. 83, p. 1221- 1237.
Shieh, Y. N., and Taylor, H. P., Jr., 1969, Oxygen and carbon isotope studies of contact metamorphism of carbonate rocks: Jour. Petrology, v. 10, p. 307-331.
Spencer, J. E., and Welty, J. W., 1986, Possible controls of base- and precious-metal mineralization associated with Tertiary de- tachment faults in the lower Colorado River trough, Arizona and California: Geology, v. 14, p. 195-198.
Swanenberg, H. E. C., 1979, Phase equilibria in carbonic systems and their application to freezing studies of fluid inclusions: Contr. Mineralogy Petrology, v. 68, p. 303-306.
Taylor, B. E., 1987, Origin and isotopic characteristics of Mother Lode hydrothermal fluids and gold deposits with comparison to Archcan analogues labs.], in Chater, A.M., ed., Abstracts to Gold '88: Toronto, Geol. Assoc. Canada, p. 148-150.
Taylor, H. P., Jr., 1974, The application of oxygen and hydrogen isotope studies to problems of hydrothermal alteration and ore deposition: ECON. GEOL., v. 69, p. 843-883.
-- 1979, Oxygen and hydrogen isotope relations in hydro- thermal mineral deposits, in Barnes, H. L., ed., Geochemistry of hydrothermal ore deposits: New York, Wiley Intersci., p. 236-277.
Tempelman-Kluit, D., 1984, Meteoric water model for gold veins in a detached terrane labs. I: Geol. Soc. America Abstracts with Programs, v. 16, p. 674.
Tempelman-Kluit, D., and Parkinson, D., 1986, Extension across the Eocene Okanagan crustal shear in southern British Colum- bia: Geology, v. 14, p. 318-321.
Weir, R. H., Jr., and Kerrick, D. M., 1987, Mineralogic, fluid inclusion, and stable isotope studies of several gold mines in Mother Lode, Tuolumne and Mariposa counties, California: ECON. GEOL., v. 82, p. 328-344.
White, W. H., Harakel, J. E., and Carter, N. C., 1968, Potassium- argon ages of some ore deposits in British Columbia: Canadian Mining Metall. Bull., v. 61, p. 1326-1334.
