Hydrothermal Fluids and Diagenesis of Mississippian Carbonates: Implications for Regional Mineralization in Western Kansas, U.S.A

Abstract

Hydrothermal fluids altered Mississippian (Osagian) carbonates in the Rebecca K. Bounds (RKB) core in western Kansas, U.S.A. Carbonate mineralization is similar to that associated with Mississippian valley type (MVT) mineralization. The RKB core displays fractures, vugs, channels, and breccias filled with saddle dolomite and blocky calcite cements. Homogenization temperature indicates that dolomite (65 to 126 °C, 18.4 to 23 wt. % NaCl) and calcite (67 to 101 °C, 13.2 to 22.4 wt. % NaCl) cements were precipitated by hot, saline fluids. These data are consistent with previous studies on the southern midcontinent. Carbon and oxygen isotope values for dolomite (δ¹³C 0.15 to 2.08‰, δ¹⁸O −6.44 to −4.66‰) and calcite (δ¹³C −1.01 to 1.79‰, δ¹⁸O −9.44 to −8.69‰) indicate multiple pulses of fluids likely sourced from basins to the south and west. Strontium isotopes data (0.7088812 to 0.7094432 in dolomite and 0.7089503 to 0.7111501 in calcite) indicate fluid interaction with granitic basement or basement-derived siliciclastics. These results are consistent with mixing of upwelling Ordovician-sourced fluids and Permian evaporitic brines, transported by advective and/or vertical migration. Although sulfide minerals were not observed in this study, earlier reports in western Kansas document sphalerite linked to hydrothermal brines in underlying strata. This study highlights the potential for MVT mineralization in the Mississippian of western Kansas.

1. Introduction

Hydrothermal fluid migration and subsequent alteration of Paleozoic-age strata are an important part of the complex diagenetic history on the North American midcontinent. The fluid sources and drivers are mainly related to tectonic (Marathon-Ouachita, Ancestral Rocky Mountains, Sevier, and Laramide orogenies) and non-tectonic (post-orogenic hydrothermal fluid flow) drivers [1,2]. The results presented here align with those reported in previous studies in Kansas as well as in Oklahoma and Missouri [1,3,4,5,6,7,8,9,10,11,12,13]. For the purposes of this study, hydrothermal fluid refers to fluid interpreted to be at a temperature exceeding the ambient burial temperature of the surrounding host rock [14].

This study specifically investigates carbonate cements filling vugs, fractures, channels, and breccias in Mississippian (Osagian) limestone in the Rebecca K. Bounds (RKB) core (API: 15-071-20446), Greeley County, Kansas (T18S R42W, Sec. 17 SE NE SE NE) (38°29′1.30″ N, 100°54′28.96″ W) (Figure 1), which is located in a complex geologic setting. The research helps in understanding the late-stage diagenetic history of the Mississippian strata of the western High Plains (western Kansas). This study builds on new data and previous comprehensive studies that examined Paleozoic rocks in south-central and eastern Kansas. Mississippian rocks of the midcontinent not only have produced significant oil and gas [15] but also host economically important metals such as lead, zinc, and copper, as well as other critical minerals in the Tri-State Mississippi valley-type (MVT) mineral district (Missouri, Oklahoma, and Kansas) [16]. Data collected for this study are compared with the data collected in the underlying Arbuckle Group (Lower Ordovician) as observed in the nearby Patterson core (T22S R38W, Sec. 25, SE SW SE SE) (38°6′11.74″ N, 100°51′13.30″ W), which also is mineralized and contains minor Zn-Pb mineralization (Figure 1) [17].

The questions addressed in this research are as follows: (a) what are the origins and pathways of the hydrothermal fluids responsible for late diagenetic cementation in Mississippian strata of western Kansas; (b) what is the relative and absolute timing of fluid flow and cementation; and (c) what is the likelihood of economic MVT sulfide mineralization associated with these late diagenetic fluids? Several methods are applied to address these questions, including thin-section petrography, cathodoluminescence (CL) petrography, fluid inclusion microthermometry, and isotope geochemistry.

2. Geological Setting

2.1. Stratigraphy

Erosion of the Precambrian basement created a highly irregular surface across the region. On this uneven topography, Cambrian-Ordovician strata, including the Arbuckle Group, were deposited (Figure 2) [18]. The Arbuckle Group is mainly composed of carbonate rocks in Kansas; however, it is absent in northeastern and northwestern Kansas [18,19]. The thickness of the Arbuckle in Kansas increases from north to south up to 423.67 m (1390 ft) [20]. By the end of the Early Ordovician, a major sea-level drop exposed the mid-continent. This resulted in intense weathering and erosion, forming a regional unconformity overlying the carbonate platforms [21]. The Middle Ordovician section in Kansas includes the Simpson Group and Viola limestone, consisting of sandstone, shale, and carbonates [22,23]. In the Hugoton embayment, the Simpson becomes thinner toward the north and west [23]. The Viola limestone is absent on uplifts and arches but thickens to over 60.96 m (200 ft) near the Oklahoma border [23]. Upper Ordovician rocks include the Maquoketa shale, which is mostly confined to the northern Kansas basin. It is up to 47.24 m (155 ft) thick in the northeast and thins to less than 12.19 m (40 ft) southward, but is absent in southwest Kansas [23].

In central Kansas, the Silurian and Devonian intervals include the Hunton limestone and overlying Misener sandstone (Figure 2). Silurian strata dominate the intervals, while the Devonian sequence is more restricted toward the northeast [25]. The Hunton Group consists of dolomitic limestone with interbedded chert, becoming thinner toward the southwest due to both overlap and erosion. The Misener is a thin sandstone that includes shale and limestone and was likely deposited on an irregular surface shaped by erosion before Mississippian deposition [25].

The Mississippian Subsystem includes rocks deposited approximately 359 to 323 million years ago throughout Kansas. During the Mississippian Subperiod, Kansas was located in low latitudes and covered by a shallow tropical to subtropical epeiric sea [26,27] (Figure 3). These rocks are exposed at the surface only in the southeastern part of the state [15]. From east to west, Mississippian rocks lie deeper in the subsurface, reaching depths of more than a thousand meters in central and western Kansas [15]. Structural highs of the Central Kansas and Nemaha uplifts are largely missing Mississippian rocks due to erosion, but hydrocarbon production is common on their flanks [15]. Mississippian rocks are also thinned or absent over anticlines and thickened in synclines and basins [15].

During deposition, Kansas was located near 20° south latitude in a tropical to subtropical belt [15,27]. Repeated sea-level changes led to alternating layers of partially silicified marine mudstones, skeletal wackestones, packstones, and grainstones [15]. By the end of the period, only the southern region remained submerged [15]. In south-central Kansas, limestone underwent weathering and dissolution during low sea levels, leaving behind a porous chert-rich zone [15] (Figure 1). Mississippian rocks in Kansas are unconformably overlain by Pennsylvanian cyclothems, including basal conglomerates, shales, sandstones, and minor carbonates. This contact is marked by truncation, erosion, and the development of a cherty zone, which grades upward into Pennsylvanian deposits [15].

2.2. Structure

The local structural geology is affected by regional tectonics. The 1.1Ga Midcontinent Rift System runs from northeast to south-central Kansas into Oklahoma and is parallel to the Nemaha uplift and the Humboldt fault zone to the east [29,30]. The Nemaha uplift itself is a series of transpressional structures, extending from southeastern Nebraska to eastern Kansas into north-central Oklahoma [30,31,32,33]. A series of northwest–southeast trending faults in west-central Kansas forms the structurally elevated Central Kansas uplift [30].

Additional tectonic influence is exerted by the Late Paleozoic formation of the Southern Oklahoma Aulacogen to the south [34,35]. Similarly, the local structure in western Kansas is affected by the Pennsylvanian formation of the Anadarko Basin, a foreland “super basin” to uplifts in the Southern Oklahoma Aulacogen (Amarillo Arch), with southward-migrating depocenters and a southern fold-thrust belt that trends NW–SE [36,37,38]. The Anadarko basin sediments thin to the NW into the Hugoton Embayment, which contains NW-striking normal faults that downthrow to the SW [39,40]. Lastly, the study area is affected by Mesozoic and Cenozoic tectonics of the Laramide Orogeny to the west [41].

3. Methods

Eighteen samples, representing mineralized and unmineralized rock, from RKB subsurface core in Greeley County, Kansas, spanning depths of 487.68 m to 1036.32 m (1600 ft to 3400 ft) [42], were prepared as polished thin sections for petrographic analysis, double-polished thick sections for chips or wafers for fluid inclusion study, and powder samples for isotope measurements. The targeted interval analyzed for this study is Mississippian (Osagian), between 1659.63 m and 1686.76 m (5445 ft to 5534 ft) [42] (Figure 2). Petrographic studies were conducted using an Olympus-BX53 microscope. Cathodoluminescence (CL) imaging was conducted using a cold cathode CITL MK5-1 apparatus mounted on an Olympus-BX51 microscope equipped with 4× and 10× long focal distance objective lenses and a low-light, digital camera system. Limestone fabrics and textures are classified according to Dunham [43], and dolomite according to Sibley and Gregg [44]. Porosity types are described according to the classification of Choquette and Pray [45]. Fluid inclusion (FI) microthermometry was conducted using a U.S. Geological Survey gas-flow heating/freezing stage to obtain homogenization temperature (T_h) and salinity of fluids included in calcite and dolomite cements. The fluid inclusion stage was operated at temperature ranges of −190 °C to +600 °C with the precision of ±0.1 °C for freezing and ±0.1 °C for heating measurements. No pressure corrections were made on T_h values; thus, these represent minimum trapping temperatures and not-filling temperatures of the included fluids. Although this is a limitation, the measured T_h values are significantly higher than ambient burial temperatures, so a pressure correction is not required to distinguish between diagenetic and hydrothermal fluid origins. Ultraviolet (UV) epifluorescence petrography was used to assess the existence of petroleum inclusions. Fluid inclusion assemblages (FIAs) were identified using the criteria of Goldstein and Reynolds [46]. Salinity, in weight % NaCl equivalent (eq), of included fluids was calculated from final melting temperature (T_mice) using the Bodnar equation [47]. Carbon and oxygen (δ¹³C, δ¹⁸O) isotope analyses were obtained on calcite and dolomite samples at the W. M. Keck Paleoenvironmental and Environmental Stable Isotope Laboratory at the University of Kansas using an isotope ratio mass spectrometer. Carbon and oxygen isotope analyses were performed on powdered carbonate samples weighing between 0.09 and 1.60 mg. Samples were reacted with phosphoric acid under vacuum at 70 °C on a Thermo Kiel IV Carbonate Device to generate CO₂, which was analyzed using a Thermo MAT 253 (Lawrence, Kansas) dual inlet isotope ratio mass spectrometer. Values are reported in δ¹³C and δ¹⁸O relative to the VPDB standard, with calibration to NBS-19 and NBS-18 and analytical precision better than ±0.05‰. Calculations for calcite–water and dolomite–water fractionation (δ¹⁸O_water) were made using O’Neil et al. [48] and Friedman and O’Neil [49] fractionation equations. Radiogenic strontium (Sr) isotope ratios (⁸⁷Sr/⁸⁶Sr) were determined using a thermal ionization mass spectrometer in the Isotope Geochemistry Laboratory (IGL) at the University of Kansas. The internal and external analytical precision for ⁸⁷Sr/⁸⁶Sr ratios was better than ±3 ppm (2σ), based on repeated measurements of the NBS 987 standard.

4. Results

4.1. Petrography

Figure 4a shows examples of the RKB core described for this study. The core study focused on targeted intervals containing late-stage carbonate cements, including calcites and dolomites filling vugs, channels, fractures, and breccias (Figure 4b). The targeted intervals span the Osagian stage (Mississippian), from its top at 1659.9 m (5446 ft) to its base at 1685.5 m (5530 ft).

Host limestone in the core interval studied here includes bioturbated mudstone-wackestone, echinoderm-rich packstone to grainstone, and bryozoan-rich wackestone, packstone, and grainstone [42]. Skeletal fragments include primarily crinoids with lesser amounts of bryozoans, bivalves, and ostracods; for full petrographic descriptions of the limestones, see Mohammadi et al. [42]. Other features were observed, including chert-replaced burrows, cross bedding, and abraded clasts, indicating variable energy conditions such as deposition in a shallow subtidal marine setting ranging from low to high energy (Figure 4b). Crinoidal grainstones are characterized by syntaxial calcite cement filling intergrain and small vug porosity (Figure 5a,b). Syntaxial and other early diagenetic calcite cements in Mississippian limestones on the midcontinent are discussed by Mohammadi et al. [10]. Additionally, very fine planar dolomite crystals frequently replace mudstones (Figure 5a,b), creating thin dolomite beds a few centimeters thick. The origin of similar Mississippian replacement dolomite is discussed by Mohammadi et al. [11].

Occasionally, skeletal fragments are partially or completely replaced by chert and occasionally contain authigenic quartz filling intraparticle porosity (Figure 5c). Fractures and channels frequently are filled by chert, authigenic quartz, and calcite cements (Figure 5d,e). Type 1b stylolites, solution seams, and fitted fabrics [50] (Figure 5e) are observed throughout the limestone in the cored section and are observed to crosscut chert- and calcite-filled fractures, channels (Figure 5e), and breccias.

Three different generations of dolomite are identified, associated with large vugs (Figure 4b), fractures, channels (Figure 4b and Figure 5c), and breccias (Figure 5f). These include the following: (1) Medium to coarse crystalline planar dolomite replacing host limestone (Figure 5f). This dolomite is not to be confused with the finer crystalline dolomite replacing mudstone described above. (2) 1st stage saddle dolomite exhibiting bright-red to dull compositional zonation under CL. And (3) 2nd stage saddle dolomite displaying several diffuse dull to dark red zones under CL (Figure 6a,b). Saddle dolomite cement is followed paragenetically by blocky calcite cement displaying bright orange to orange-yellow CL and lacking zonation (Figure 6). Fractures and channels filled by 1st stage saddle dolomite cement occasionally crosscut fractures and channels filled by chert and authigenic quartz (Figure 5d and Figure 6c,d).

4.2. Fluid Inclusions

One- and two-phase (liquid and vapor) fluid inclusions were observed in 2nd stage saddle dolomite and calcite cements filling vugs, fractures, channels, and breccia in the core intervals studied (Figure 7a,b). In addition, one-phase secondary liquid petroleum (oil) inclusions were observed in the 2nd stage saddle dolomite cement under UV epifluorescence (Figure 7c,d). Primary inclusions were distinguished from secondary inclusions following Roedder (1984) [51]. Homogenization temperature (T_h) (minimum trapping temperature) and last ice melting temperature (T_mice) were measured from primary inclusions in carbonate cements in the RKB core samples (

Table 1. The sample IDs indicate the depth of the intervals per foot from which the samples were collected (1 ft = 0.305 m) (modified from [42]).

Sample ID	Mineral	Assemblage		Th (°C)	NaCl Eq	Description
RKB-5453	Dolomite	1	106	−16.5	19.8	Primary
			106	−16.5	19.8
			106	−16.5	19.8
			106	−16.5	19.8
			106	−16.5	19.8
		2	93	−15.2	18.8	Primary
			93	−15.2	18.8
			93	−15.2	18.8
			93	−15.2	18.8
			93	−15.2	18.8
			93	−15.2	18.8
			93	−15.2	18.8
		3	92	−15.2	18.8	Primary
			92	−15.2	18.8
		4	126	−19.6	22.1
			126	−19.6	22.1
			126	−19.6	22.1
		-	109	−14.7	18.4	Primary
RKB-5474.9	Dolomite	1	107	−18	21.0	Primary
			107	−18	21.0
			107	−18	21.0
		-	65	-		Primary
		-	107	−18	21.0	Primary
		-	111	−20	22.4	Primary
		-	121	−20	22.4	Primary
RKB-5495	Dolomite	1	116	−20	22.4	Primary
			116	−20	22.4
			116	−20	22.4
			116	−20	22.4
			116	−20	22.4
		-	104	−20.9	23.0	Primary
		2	124	-		Primary
			124	-
		3	104	-		Primary
			104	-
		-	-	−20.9	23.0	Primary
RKB-5505.6	Calcite	-	79	−18.2	21.1	Primary
		-	67	−20	22.4	Primary
		-	76	−20	22.4	Primary
		-	85	−18.2	21.1	Primary
		-	85	−18.9	21.6	Primary
		-	79	−18.2	21.1	Primary
		-	79	−23.6	24.7	Primary
		1	79	−18.2	21.1	Primary
			79	−18.2	21.1
			79	−18.2	21.1
			79	−18.2	21.1
			79	−20	22.4
RKB-5525	Calcite	1	78	−13.5	17.3	Primary
			78	−13.5	17.3
			78	−13.5	17.3
		2	78	-		Primary
			78	-
			78	-
			78	-
		-	76	-		Primary
		-	80	−9.3	13.2	Primary
		-	101	−19.8	22.2	Primary

, Figure 8). Saddle dolomite cements (2nd stage) have T_h assemblages with values ranging from 65 to 126 °C and salinity ranging from 18.4 to 23 wt. % NaCl eq. Fluid inclusions in calcite cements display T_h assemblage values that range from 67 to 101 °C and salinity from 13.2 to 22.4 wt. % NaCl eq (Table 1). Some 2nd-stage saddle dolomite assemblage values slightly overlap with calcite cement; however, they typically have higher T_h values than calcite. These T_h values are similar to those measured in saddle dolomites in the underlying Arbuckle Group (Lower Ordovician) in the Patterson core [17]. However, calcite temperatures observed in the Arbuckle section in the underlying Patterson core have lower T_h values than calcites in the overlying Mississippian rocks. Possible primary and secondary oil inclusions were also observed in blocky calcite cements.

4.3. Isotope Geochemistry

Stable δ¹⁸O and δ¹³C isotope analyses were obtained on dolomite and calcite cements filling vugs, channels, and breccias in the RKB core. Most of the saddle dolomite samples used for isotope analysis were probably 2nd stage, as it is volumetrically more common in the large vugs, channels, and breccias that were sampled than 1st stage saddle dolomite. Saddle dolomite cements have δ¹⁸O and δ¹³C ranges from −6.44 to −4.66‰ and 0.15 to 2.08‰, respectively. Ranges for δ¹⁸O and δ¹³C in calcite cements were −9.44 to −8.69‰ and −1.01 to 1.79‰, respectively (

Table 2. Carbon and oxygen isotope data in calcite and dolomite filling vugs and channels in the RKB core (API: 15-071-20446), modified from Mohammadi et al. [42]. The sample ID indicates the depth in feet of the interval collected (1 ft = 0.305 m). All values are expressed relative to the VPDB standard.

Sample ID	Mineral and Open Space Type	δ13C‰ VPDB	δ18O‰ VPDB	δ18O‰ VSMOW	δ18O‰ Water VSMOW *
RKB-5453	channel-filling saddle dolomite	2.08	−6.14	24.58	2.58 to 6.08
RKB-5458.5	channel-filling saddle dolomite	1.98	−5.67	25.06
RKB-5474.9	vug-filling saddle dolomite	1.90	−5.64	25.10	1.4 to 6.1
RKB-5495	vug-filling saddle dolomite	1.46	−6.50	24.21	3.71 to 5.71
RKB-5505.6	channel-filling calcite	−0.25	−9.08	21.55	0.05 to 2.55
RKB-5525	vug-filling calcite	1.79	−9.44	21.18	1.18 to 4.18
RKB-5533.4 a	channel-filling saddle dolomite	0.15	−6.14	24.58
RKB-5533.4 b	channel-filling calcite	−0.96	−8.86	21.78
RKB-5533.6 a	channel-filling saddle dolomite	0.57	−4.66	26.11
RKB-5533.6 b	channel-filling calcite	−1.01	−8.69	21.95

* The calculations for water were obtained by using ranges of fluid inclusion homogenization temperatures obtained for samples analyzed for δ¹⁸O‰ and fractionation equations from O’Neil et al. [48] and Friedman and O’Neil [49]. Homogenization temperatures are uncorrected for pressure and represent minimum filling temperatures. Therefore, water values should be taken as estimates. Superscripts “a” and “b” indicate that different minerals were analyzed from the same sample.

, Figure 9). These values display a similar trend as observed in the underlying Patterson core in the Arbuckle [17] and Mississippian calcite cements studied by Ritter and Goldstein [52] (Figure 9).

Radiogenic Sr ratios (⁸⁷Sr/⁸⁶Sr) for saddle dolomite filling channels, vugs, and breccias range from 0.7088812 to 0.7094432, and for calcite range from 0.7089503 to 0.7111501 (

Table 3. Strontium isotope data of late diagenetic carbonate cements from RKB core [42].

Sample ID	Mineral and Open Space Type	87Sr/86Sr (Exp Corr) *
RKB-5453	Channel-filling saddle dolomite	0.7088812
RKB-5474.9	Vug-filling saddle dolomite	0.7090068
RKB-5495	Vug-filling saddle dolomite	0.7094432
RKB-5505.6	Channel-filling calcite	0.7089503
RKB-5525	Vug-filling calcite	0.7111501

* (exp corr) is exponential correction. The sample IDs indicate the depth (in feet) of the interval in which the sample was collected (1 ft = 0.305 m).

, Figure 10). Mississippian Sr values measured for this study are largely less radiogenic compared to the values observed in underlying Arbuckle rocks in the Patterson core [17]. In addition, Mississippian values in this study overlap the Mississippian values measured by Mohammadi et al. [10,11] in the Mississippian section of the Ozark-Cherokee Platform and north-central Oklahoma.

5. Discussion

5.1. Paragenesis in the Mississippian of Western Kansas

Figure 9 shows a hypothesis for the relative timing of the diagenetic events observed in the RKB core. The early diagenetic features observed in this study involve early stabilization and lithification of the hosting limestone, composed primarily of skeletal grainstones, wackestones, and mudstones, soon after deposition. This includes growth of early intergrain, syntaxial, and vug-filling calcite cements (Figure 5a). Early secondary porosity filled by this cementation includes fractures and vugs. Early diagenetic cementation in Mississippian strata is discussed in detail by Mohammadi et al. [10,11] and is not the focus of this paper. Early to intermediate diagenetic fracture and channel fillings include calcite, chert, and authigenic quartz, with the quartz likely precipitated during intermediate diagenesis as the fractures postdate early syntaxial and vug-filling calcite cementation (Figure 5c–e and Figure 11).

The replacement of skeletal fragments by chert likely occurred at the same time as, or prior to, filling of channels by chert and quartz during the intermediate stage of diagenesis (Figure 5c and Figure 9). Thick accumulations of silicious sponge deposits have been observed in shelf margins in Mississippian (Osagian) strata in Kansas [27,54,55,56,57]. Franseen [27] suggested that a regional upwelling of silica-rich water sourced by sponge spicules is a likely source of silica in the Mississippian carbonates.

The dolomite replacing mudstone observed in this study is similar to the Mississippian replacement dolomite described by Mohammadi et al. [11]. They suggested, based on petrography and isotope geochemistry, that the replacement dolomites formed under marine phreatic conditions associated with deep circulation of Mississippian seawater.

Formation of fractures and vugs began during the early stages of diagenesis and persisted through to the late stage (Figure 11). Channels, likely resulting from solution widening of fractures, developed during intermediate to late diagenesis. The timing of the development of this secondary porosity is indicated by its filling by intermediate diagenetic chert, calcite, and authigenic quartz, as well as late diagenetic saddle dolomite. Breccias formed only during the final stage of diagenesis and are observed to be filled only by saddle dolomite and late blocky calcite cements. Early small vug porosity is filled by early calcite cement, whereas late diagenetic vugs are filled by saddle dolomite and blocky calcite cement. Development of stylolites, resulting from pressure solution [58], occurred during intermediate extending to late diagenesis as evidenced by the solution of skeletal grains and mud matrix as well as crosscutting relationships observed with fractures, channels (Figure 5e), and breccias (Figure 11).

Continued development of fractures, vugs, channels, and breccia during late diagenesis is associated with hydrothermal fluid migration through the Mississippian rocks (Figure 8; Table 1). The large open spaces likely developed as a result of dissolution by hydrothermal fluids alternating with dolomite- and calcite-precipitating fluids. The late diagenetic open spaces were filled by saddle dolomites, both 1st and 2nd stage, followed by late calcite cements (Figure 5f and Figure 6). Saddle dolomite cements are believed to precipitate from warm to hot aqueous fluids after burial [59,60] (Figure 6a,b). Fluid inclusion data (Figure 8; Table 1) also indicate that the saddle dolomites precipitated from hot and saline fluids. A second stage of replacement dolomitization (RD-2) of limestone occurred along with possible recrystallization of earlier dolomite and is associated with late diagenetic open space filling saddle dolomite (Figure 5f). Following the precipitation of saddle dolomite, blocky calcite cement filled the remaining vug, channel, and breccia open space (Figure 5c and Figure 6a,b).

Petroleum migration through the section is evidenced by secondary oil inclusions observed in 2nd stage saddle dolomite (Figure 7c,d). Observation of possible primary and secondary petroleum inclusions in blocky calcite cements adds further evidence for the timing of petroleum migration. The paragenetic timing of petroleum migration observed in this study is consistent with the timing observed in Mississippian strata in the Tri-State Mineral District, on the Cherokee-Ozark Platform, and in north-central Oklahoma [10,11].

5.2. Composition of Late Diagenetic Fluids

Fluid inclusion homogenization temperatures recorded in the Cambrian-Ordovician (Arbuckle Group) of south-central Kansas range approximately between 90 and 131 °C [61]. These temperatures are similar to the temperatures recorded in saddle dolomite cement in the Mississippian of western Kansas (this study), where homogenization temperatures of fluid inclusion assemblages range from 65 to 126 °C with salinities ranging from ~18 to 23 wt. % NaCl eq. These are also similar to fluid inclusions measured in saddle dolomites in Mississippian strata on the Cherokee-Ozark Platform [10]. Blocky calcite cement assemblages measured in the Mississippian of western Kansas, in contrast, are cooler (~67 to 100 °C) than those measured in late diagenetic calcite cement on the Cherokee-Ozark Platform and overlap with Mississippian calcite cements measured on the Cherokee-Ozark Platform and in north-central Oklahoma. None of the salinities of fluid inclusions measured in this study in calcite (13–23 wt. % NaCl eq.) indicate the presence of dilute fluids such as were observed in Mississippian rocks in north-central Oklahoma or on the Cherokee-Ozark Platform [10,11].

In the Mississippian (Osagian) in western Kansas, fluid inclusion temperatures measured in open-space-filling saddle dolomites (92 to 126 °C) overlap with the lower range of temperatures measured in the underlying Ordovician (Arbuckle Group) in the nearby Patterson core (53 to 266 °C) (Figure 1 and Figure 8) [17]. Hydrothermal fluids that precipitated saddle dolomite in the Mississippian RKB core have higher salinity compared with the underlying Ordovician rocks (Figure 8). This suggests a different fluid origin with different pulses of advective fluids in the Mississippian and Ordovician strata. Alternatively, and more likely, higher salinities as well as lower temperatures in the Mississippian fluid inclusions may be due to mixing of upwelling fluids sourced in the Ordovician with descending, cooler, evaporitic fluids sourced from overlying Permian evaporites [9].

Open-space-filling calcite cement δ¹⁸O values from the RKB core range from −9.44 to −8.69‰ with δ¹³C values ranging from −1.01 to 1.79‰ (Table 2; Figure 9), consistent with warm basinal fluids and possible sulfate reduction. These values are comparable to the late stage of calcite cementation in eastern Kansas [52], on the Cherokee-Ozark platform [42], and in the Tri-State MVT district [10]. Calcite and dolomite carbon and oxygen isotope values from the underlying Arbuckle Group, in the nearby Patterson core, are consistent with precipitation by hot fluids undergoing sulfate reduction (Figure 9) [17].

Sr isotope ratios measured in late-stage saddle dolomite and blocky calcite cement in the RKB core overlap with radiogenic Sr ratios measured in late diagenetic calcite cements in Mississippian rocks on the Cherokee-Ozark Platform [10] and are significantly higher than the Sr ratios measured for calcite cements in the Mississippian of north-central Oklahoma [11]. These values are more radiogenic than what would be expected for carbonates precipitated in equilibrium with the Mississippian seawater (Table 3, Figure 10). Although the Sr isotope ratios measured in this study do not reach the highest values of those measured in carbonate cements in Lower Ordovician (Arbuckle Group) carbonates in the nearby Patterson core (Figure 1), they fall at the higher end of what would be expected for a system influenced by Arbuckle seawater and overlap with values that would be imparted by fluids that have been modified by granitic basement rocks or sedimentary rocks derived from the granitic basement.

Taken together, these observations suggest that the fluid responsible for precipitating the saddle dolomite and blocky calcite cement in the Mississippian section in the RKB core was not derived from Mississippian seawater but rather was introduced from an external source, as is evidenced by the fluid inclusion and isotope data presented here. Based on the fluid inclusion and isotope data obtained for late diagenetic carbonate cements studied in the RKB core, it is evident that the conduit of hydrothermal fluids was not localized in place, but flow involved upward migration of fluids along faults and fractures. Fluids migrated laterally from the basin to the south and/or west, or a combination of both. The higher salinities and lower temperature of included fluids observed in saddle dolomite and blocky calcite cement are consistent with mixing of these fluids with cooler and more saline fluids descending from overlying Permian strata.

5.3. Fluid Sources and Timing

Multiple pulses of basinal fluids likely affected Phanerozoic rocks on the Midcontinent of North America, including the Mississippian of western Kansas [2,61,62,63,64]. Possibly a gravity-driven northward fluid flow system transported hot basinal brines northward out of the Anadarko Basin during late Paleozoic tectonic activity to the south [65]. Goldstein et al. [9] suggest that the basinal fluids migrated through the entire stratigraphic package and are associated with the uplift of the Ouachita and Ancestral Rocky Mountains during late Pennsylvanian and Permian time. Mohammadi et al. [2], using U-Pb analysis of calcite cements in the same core studied by Goldstein et al. [9], observed two different pulses of fluid, the first related to the late Pennsylvanian-Permian tectonic events and a second dated to the late Miocene to Pliocene. Published radiometric dates [2] suggest that, in addition to the late Pennsylvanian, Permian, and Neogene events, basinal brine migration on the Midcontinent can be dated to rifting in the Gulf of Mexico to the south during the Early Jurassic and to the Sevier and Laramide Orogenies during the Cretaceous to Paleocene [2].

It is speculated here that the complex late diagenetic carbonate cement history observed in this study is the result of multiple pulses and a variety of sources of non-resident fluids that influenced Mississippian rocks in western Kansas. We further speculate that hydrothermal fluid flow in the study area also may have involved vertical fluid pumping from lower Ordovician into Mississippian strata, related to fault reactivation and post-orogenic events [2].

5.4. Implications for MVT-Style Mineralization in Western Kansas

Although galena, sphalerite, and other metal sulfides were not found in the examined Mississippian samples of western Kansas, this study shows that the open-space-filling carbonate cements were precipitated by basinal fluids that are very similar to fluids associated with known MVT mineralization, such as the large Tri-State MVT district, in Mississippian rocks on the Cherokee-Ozark Platform [10]. Evans [3] observed sphalerite mineralization in the Ordovician (Arbuckle Group) of western Kansas and suggested the possibility of economic MVT mineralization in the region. Coveney [4,66] reported fluid inclusion homogenization temperatures about 80 to >150 °C in Ordovician sphalerites in three oil wells in western Kansas, which he attributed to regional advective fluid flow. Further research is certainly warranted, including examination of additional cores and/or well cuttings that may identify sulfide mineralization in the Mississippian of western Kansas.

6. Conclusions

This study investigates the origin and timing of dolomite and calcite cementation precipitated by hydrothermal fluids during late diagenesis in Mississippian carbonates in the Rebecca K. Bounds (RKB) core, Greeley County, western Kansas, U.S.A.

Fluid inclusion homogenization temperatures and salinities are comparable with those measured in the Arbuckle Group and in the Mississippian on the Cherokee-Ozark Platform, in north-central Oklahoma, and south-central Kansas. These data, along with radiogenic strontium isotope signatures, indicate that the fluids were not locally sourced and had interacted with the granitic basement or basement-derived siliciclastics. The evidence suggests possible mixing of upwelling Ordovician-derived fluids with descending Permian evaporitic brines. Fluid movement involved upward migration along faults and fractures, lateral migration from the basin, or a combination of both.

Saddle dolomite, precipitated by warm saline fluids, predates blocky calcite cements precipitated by cooler saline brines. Petroleum migration occurred after saddle dolomite precipitation and is recorded by secondary oil inclusions in saddle dolomite and (possibly) primary and secondary oil inclusions in blocky calcite cement. Isotopic evidence (C, O, Sr) and previously published radiometric dating indicate that fluid flow affecting the RKB core is likely associated with post-orogenic and fault reactivation events.

Although galena, sphalerite, or other sulfide minerals were not identified in the Mississippian rocks in this study, the hydrothermal fluids that precipitated saddle dolomite and blocky calcite share similar temperature, salinity, and isotopic characteristics with basinal fluids responsible for MVT mineralization in the Tri-State district on the Cherokee-Ozark Platform. Previous studies identified sphalerite in the underlying Arbuckle Group with fluid inclusions characteristics of MVT mineralization. The absence of sulfides in the RKB core does not preclude their occurrence in Mississippian strata elsewhere in western Kansas. Additional subsurface sampling is required to evaluate this potential more fully.