Kansas Geological Survey, Current Research in Earth Sciences, Bulletin 235, pp. 55-72

Composition of rift-related igneous and sedimentary rocks of the Keweenawan Supergroup in the Poersch no. 1, OZ-1, Finn, and Friederich wells, northeastern Kansas

Robert L. Cullers¹ and Pieter Berendsen^2
1Kansas State University and ²Kansas Geological Survey

Abstract

A series of wells have been drilled up to a depth of 3,444 m (113,000 ft) into the midcontinent rift system (MRS) in northeastern Kansas. The age of a gabbro sill in the upper portion of the Precambrian rocks was determined to be 1,097.5 ± 3 Ma, and the magnetic polarity of the gabbro correlated to the lower Keweenawan rocks in the northern MRS (Van Schmus et al., 1990). The rocks below 2,259 m (7,411 ft) in the deepest well [Poersch no. 1, total depth 3,435 m (11,270 ft)] consist mostly of arkoses with subordinate amounts of shale, siltstone, and basalt. The rocks above 2,265 m (7,431 ft) in the Poersch well consist of basalt with minor siltstone, arkose, gabbro, and felsite. A proposed high-angle reverse fault could have juxtaposed the upper igneous rocks over the lower sedimentary rocks to produce a reversed stratigraphy. This would make the development of the southern MRS similar to that of the northern MRS. Thus, in the initial extensional phase of the MRS, broad subsidence coincided with abundant volcanism and little sediment production. Grabens formed in the later stages of rift development and were filled with abundant sedimentary rocks along with lesser volcanic rocks. The chemical characteristics of the basalts in the southern MRS are similar to those in the north. The southern basalts are subalkalic to alkalic and follow tholeiitic trends; a number of them are high-alumina basalts. Although there is a lot of scatter, Al₂O₃, Ni, and Cr concentrations decrease and Fe₂O₃, TiO₂, K₂O, rare earth elements, Ba, Hf, and Sc concentrations increase with decreasing Mg number. These trends are consistent with plagioclase, olivine, pyroxene, or spinel fractionation from primary basalts. One basalt could represent a primary magma because it has a high Mg number (0.68), high Ni (638 mg/kg) and Cr (233 mg/kg) concentrations, low incompatible element concentrations (e.g., La = 4.2 mg/kg), and a slight positive Eu anomaly. This possible primary magma could have formed by partial melting (20-25%) of an undepleted spinel peridotite at 30-40 km depth. Most basalts have not been contaminated by crustal rocks or silicic magmas. The mineralogy, chemical composition, and U-Pb geochronology of detrital zircons of the arkoses, siltstones, and shales are consistent with their derivation from the surrounding granitoid highlands with little or no input from the basalts. Even siltstones and arkoses within the mostly basaltic sequences are derived mostly from the granitoids, although the siltstones may have some input from the basalts (e.g., higher Ni concentration than the sandstones).

Report

Rocks associated with the midcontinent rift system (MRS) extend in the subsurface from south-central Kansas northward through Nebraska, Iowa, and Minnesota, where they are exposed in the Lake Superior region, northern Wisconsin, northern Michigan, and Ontario (fig. 1) (Dickas, 1984,1986; Van Schmus and Hinze, 1985). The surface rocks are a bimodal suite of mostly tholeiitic basalts and some alkaline basalts along with minor silicic rocks (Basaltic Volcanism Study Project, 1981; Van Schmus and Hinze, 1985). The alkaline basalts are most abundant in the lower portion of the section, and the tholeiites become more abundant up-section (Basaltic Volcanism Study Project, 1981). Subsidence of the basins resulted in deposition of a thick sequence of continental sedimentary rocks over the basalts (Morey and Green, 1982). The age range of igneous and sedimentary activity in the rift has been estimated to be 1,087-1,109 Ma (Davis and Sutcliffe, 1985; Palmer and Davis, 1987; Van Schmus et al., 1989, 1990; Davis and Paces, 1990; Van Schmus, 1992). The subsurface rocks are recognized by large gravity and magnetic anomalies, presumably reflecting the abundance of basalts along the rift (Van Schmus and Hinze, 1985).

Figure 1--The midcontinent rift system (shaded area) extends from the surface in Minnesota, northern Michigan, and northern Wisconsin into the subsurface in Iowa, Nebraska, and Kansas.

We have analyzed cuttings and cores from four wells drilled into the southern portion of the MRS in northeastern Kansas and have compared the composition of these rocks to the composition of similar rocks in the northern MRS. The Texaco Poersch no. 1 well (fig. 2) was drilled to a depth of 3,435 m (11,270ft) in Washington County, Kansas (Berendsen et al., 1988). In addition, a test hole, OZ-1, is located adjacent to the Poersch well. Most of the cuttings and all the core were obtained from the Poersch and OZ-1 wells. In addition, some cuttings were obtained from the Finn and Friederich wells, drilled to depths of 1,208 m (3,975 ft) and 1,350 m (4,440 ft), respectively, in the rift. Most of the discussion here centers on the results obtained from the Poersch well.

Figure 2--Location of the Poersch, Friederich, and Finn wells in northeastern Kansas.

Geology

The portion of the Poersch well below 7,430 ft (2,265 m) consists of mostly arkosic sandstone and conglomerate with minor siltstone, shale, and basalt (fig. 3). The Poersch well from 2,840 ft (866 m) to 7,430 ft (2,265 m) and the OZ-1 well are mostly basalt with minor siltstone, arkose, gabbro, and felsite (fig. 3). A U-Pb age on baddeleyite of 1,097.5 ± 3 Ma from gabbro in the upper portion of the Poersch well (Van Schmus et al., 1990) and a high-angle reverse fault interpreted from seismic (COCORP) data, which juxtaposes the upper igneous and lower sedimentary section (Woelk and Hinze, 1991), suggest that the basic igneous activity took place before the deposition of abundant arkosic basin fill from the surrounding highlands of the rift.

Figure 3--Schematic columnar section of the Poersch well [after Berendsen et al. (1988); depth in feet].

In the Friederich well a thick gabbro sill overlies basaltic rocks with minor interflow clastic sedimentary rocks. In the Finn well mostly arkosic sedimentary rocks overlie basaltic rocks near the bottom of the well. The stratigraphic succession in the Friederich well is similar to that in the upper portion of the Poersch well, probably because of the location of the wells in the same structural block (fig. 2). The rocks in the Finn well, however, cannot be correlated with those in the Friederich and Poersch wells because of structural complexity in the area (Berendsen et al., 1988).

Methods

Sampling procedure

The OZ-1 well was cored and completed in the same unit that tops the Precambrian section in the nearby Poersch well (fig. 3). This drilling allowed us to sample fresher portions of the cored interval. The Poersch no. 1 well was cored at selected intervals (Berendsen et al., 1988, appendix 3). The mafic rocks are all altered. The intensity of alteration is discussed in a later section. The interflow sedimentary rocks appear to be affected by the heat from the enclosing mafic rocks. The sedimentary rocks below 7,430 ft (2,265 m) are relatively unaltered and appear not to have been metamorphosed.

Cuttings were collected at 10-ft intervals in the well and forwarded to the Wichita Well Sample Library for washing and archiving. Our samples from the drill holes are unique in that they represent the deepest drill holes completed in the MRS in Kansas. The average sample weight was about 100 g, although some samples weighed up to about 500 g. The samples were handpicked using a binocular microscope to exclude any contaminants.

Analytical techniques

Cuttings from the Poersch no. 1, Friederich, and Finn wells were analyzed for major elements by atomic absorption at the Kansas Geological Survey; major elements of cores from the Poersch well and some cuttings were analyzed by atomic absorption at Kansas State University following the methods of Buckley and Cranston (1971).

Most trace elements, Fe₂O₃, and Na₂O were analyzed by neutron activation analysis at Kansas State University using the methods of Gordon et al. (1968) and Jacobs et al. (1977). Also Nb, Y, Rb, Sr, Zr, and Ni were analyzed by x-ray fluorescence at Kansas State University. Estimates of precision and values obtained on standard rocks using these methods can be found in the articles by Cullers et al. (1979, 1985).

Petrology

Details of the petrology of the Poersch well can be found in the report by Berendsen et al. (1988). The rocks are described briefly here so that the chemical composition of these rocks can be better interpreted. For example, some of the basalts are considerably altered so that elements such as Rb, Ba, and Sr have been mobilized (Wood et al., 1976).

The gabbro sill at the top of the section [OZ-1, 2,823-2,872 ft (860.5-875.4 m) depth; Poersch no. 1, 2,846-3,150 ft (867.5-960.1 m)] is 88 m (290 ft) thick. The gabbro is fine- to medium-grained and slightly serpentinized; it contains epidote, so it was undoubtedly altered. Five samples of the gabbro from core material were analyzed (table 1).

The predominantly basalt sequence from 3,150 to 4,140 ft (960.1-1,262 m) contains the least altered basalts, although they are deuterically and hydrothermally altered (Berendsen et al., 1988). Fibrous actinolite, chlorite, albite, and saussurite were observed in the altered zones. The amount of magnetite varies from a few percent to 20% in some basalts. The basalt cuttings analyzed at 3,720-3,730 ft (1,134-1,137 m) are the least altered rocks in this sequence. No sedimentary rocks occur in this interval.

Basalts in the sequence of rocks from 4,140 to 5,719 ft (1,262-1,743 m) are more hydrothermally and deuterically altered than those from 3,150 to 4,140 ft (960.1-1,262 m) (Berendsen et al., 1988). Some of the basalts are completely altered to albite and epidote, and we avoided analyzing these intervals (table 1). This interval also has arkoses and siltstones that alternate with the basalts. An arkose [5,396 ft (1,645 m)] and a siltstone [5,404 ft (1,647 m)] were sampled from a core adjacent to the basalt at 5,399 ft (1,646 m).

Table 1--Major element oxide (wt %) and trace element (mg/kg) concentrations of the igneous rocks in the OZ, Texaco no. 1 Poersch, Finn, and Friederich wells

The predominantly basalt sequence from 5,719 to 7,429 ft (1,743-2,264 m) depth is also commonly altered to epidote and albite (Berendsen et al., 1988). Again, we sampled four of the least altered basalts for analysis (table 1). One siltstone at 7,153 ft (2,180 m) was sampled from a core adjacent to the basalt at 7,159 ft (2,182 m).

The rocks at depths greater than 7,429 ft (2,264 m) consist mostly of arkosic sandstones and conglomerates. They appear to be a series of fining-upward sequences containing conglomerate or sandstone at the base with or without a siltstone at the top (Berendsen et al., 1988). The arkoses consist of clasts composed mostly of polycrystalline quartz and alkali feldspar (unaltered microcline, orthoclase, or perthite) with lesser plagioclase (slightly to very altered). The accessory minerals are micas, sphene, zircon, and opaque minerals. The minor siltstones contain more abundant hematite staining than do the coarser fractions (Berendsen et al., 1988). Four arkoses and one siltstone were analyzed in this sequence. Minor basalts and gabbros occur below 7,429 ft (2,264 m), except for 43-m-thick (141-ft-thick) and 72-m-thick (236-ft-thick) units at 8,680 and 10,622 ft (2,646 m and 3,238 m) depth, respectively. Six basalts or gabbros were sampled in this interval. Many of these mafic rocks are altered to some degree, containing chlorite, serpentine, epidote, carbonates, albite, pyrite, or chalcopyrite (Berendsen et al., 1988).

Geochemistry

Alteration

The chemical composition of the basalts and gabbros can be compared with that of rocks exposed along the MRS further to the north, even though most samples are altered. Again, however, we sampled the least altered samples for analysis. Also, elemental correlations in variation diagrams are similar to those expected for rocks controlled by igneous processes rather than for rocks affected by alteration.

Igneous rocks

The igneous rocks range from subalkalic to alkalic (fig. 4). They tend to follow tholeiitic trends toward iron enrichment in AFM diagrams (fig. 5). Most of these rocks plot in the basalt field in an alkali versus silica plot (fig. 6). In addition, although the database is small, the stratigraphically lower basalts [deeper than 7,159 ft (2,182 m)] in the Poersch well are all fairly rich in potassium (0.64-0-90 wt % K₂O). The basalts at less than 7,159 ft depth vary more in composition, ranging from 0.12 to 1.56 wt % K₂O. The chemical characteristics of the basalts and gabbros from the southern MRS are similar to those in the northern MRS (see table 3) (Green, 1972, 1982; Basaltic Volcanism Study Project, 1981; Weiblen, 1982).

Figure 4--Total alkali versus silica concentrations of the MRS basalts and gabbros [after Irvine and Barager (1971)]. Open circles, Poersch and OZ cores; open squares, Poersch cuttings; squares with dots, Poersch cutting of felsite; open triangles, Friederich cuttings; triangles with dots, Finn cuttings. The plotted results in this and other plots are recalculated to 100% without H₂O, as recommended by Irvine and Baragar (1971).

Figure 5--AFM diagram of the MRS rocks [after Irvine and Barager (1971)]. Symbols defined in fig. 4.

Figure 6--Most of the MRS samples plot in the basalt field of an alkali-silica plot [after Cox et al. (1979)]. Symbols defined in fig. 4.

Nine of 29 analyses of the basalts fall in the high-alumina field (greater than 16 wt % Al₂O₃). Most pre-Tertiary plateau basalts range from 13 to 15.5 wt % Al₂O₃ (Godlevskiy, 1959; Green, 1972; Ruegg, 1976; Basaltic Volcanism Study Project, 1981). In contrast, high-alumina basalts in the northern MRS make up a major part of some sequences (Green, 1982). Thus the basalts and gabbros from the southern MRS resemble those to the north.

Some major element correlations in the southern MRS basalts are similar to those in the northern MRS basalts. For example, TiO₂ increases and Al₂O₃ decreases with decreasing Mg number (fig. 7) [compare to Basaltic Volcanism Study Project (1981)].

Figure 7--(a) Variation diagram of alumina versus Mg number in the MRS rocks. (b) Variation diagram of TiO₂ versus Mg number. Symbols defined in fig. 4.

The compatible trace elements (Ni and Cr) decrease and the incompatible trace elements [rare earth elements (REEs), Ba, Hf, Ta, Sc] increase with decreasing Mg number. For example, the Ni concentration ranges from a high of 638 mg/kg at the highest Mg number of 68 to a low of 95-133 mg/kg at the lowest Mg numbers (fig. 8a); correspondingly, the Cr concentration ranges from a high of 233 mg/kg to a low of 79 mg/kg. This range in Ni and Cr concentrations is similar to that observed in basalts in the northern MRS (Corwall and Rose, 1957; Basaltic Volcanism Study Project, 1981; Klewin and Berg, 1991). The REE concentrations tend to increase with decreasing Mg number (fig. 8b). Although the tendency is not completely clear-cut, the Eu anomaly size tends to change from a positive anomaly at the higher Mg number and low REE concentrations to a more negative anomaly at the lower Mg number and high REE concentrations (fig. 9). The results are similar to those obtained for the Keweenawan reference suite (Basaltic Volcanism Study Project, 1981).

Figure 8--(a) Variation diagram of Ni (mg/kg) versus Mg number. (b) Variation diagram of La (mg/kg) versus Mg number. Symbols defined in fig. 4.

Figure 9--REE patterns of the mafic rocks in the Poersch, Friederich, and Finn wells obtained by normalizing to chondrites [chondrite values from Nakamura (1974)].

Sedimentary rocks

The arkoses, with their feldspar-quartz mineralogy have major element concentrations (table 2) that reflect their derivation from the nearby granitic highlands with little chemical weathering. The shales and siltstones contain higher TiO₂, Al₂O₃, Fe₂O₃, MgO, REE, Rb, Cs, Co, Sc, Cr, and Ni concentrations and lower SiO₂ concentrations than the arkoses. This probably reflects the higher clay and ferromagnesian mineral content and the lower quartz content of the shales and siltstones compared with the arkoses. For example, the REEs are concentrated in clay minerals relative to quartz or feldspar (Cullers et al., 1987; Cullers, 1988), thus explaining the higher REE content of the shales and siltstones compared with the arkoses.

Table 2--Major weight percent element oxide and trace element (mg/kg) concentrations of the sedimentary rocks in the Texaco no. 1 Poersch and the Finn Wells.

Figure 10--Chondrite-normalized REE patterns of sedimentary rocks in the Poersch and Finn wells.

Figure 11--(a) Elemental concentrations of arkoses normalized to the PAAS. (b) Elemental concentrations of shales and siltstones normalized to the PAAS. Values for PAAS from Taylor and McLennan (1985).

Discussion

Igneous rocks

The chemical characteristics of the southern MRS igneous rocks are similar to those of the northern MRS igneous rocks (table 3). We are cautious about putting much significance on quantitative modeling of the igneous processes because some of the scatter in the data could have resulted from fractionation of cuttings in the wells or from alteration. For example, at least four of the Poersch cuttings fall below the trend established by other samples in the Al₂O₃ versus Mg number plot (fig. 7a). Also, gabbro cuttings from Poersch 2860-70 were analyzed over the same interval as the nearby OZ core gabbros (table 1). Most elements are fairly similar in the two sets of samples, although the TiO₂ and Rb contents are much higher and the Al₂O₃ content is lower in the cuttings than in the corresponding core samples.

Table 3--Comparison of the range of elemental compositions of the basalts in the southern MRS to those in the north.

The best candidate for a primitive magma formed by direct melting of the mantle would be the basalt at 3,720 ft (1,134m) because it has high Mg number(O.68), high Ni (638 mg/kg) and Cr (233 mg/kg) concentrations, low incompatible element concentrations, low K₂O content (0.12 wt %), low TiO₂ content (1.3 wt %), low REE concentrations (e.g., La = 4.2 mg/kg), and a slight positive Eu anomaly [e.g., see criteria by Roeder and Emslie (1970), Cox (1980), Prestvik and Goles (1985), and Basaltic Volcanism Study Committee (1981)]. The REE patterns do not show depletion of light REEs relative to the heavy REES, as is seen in basalts from midocean ridges (fig. 9). Thus the primitive MRS basalt must have formed from mantle that was not previously melted (Basaltic Volcanism Study Committee, 1981; Green, 1982). These chemical characteristics are consistent with the primitive basalt at 3,720 ft (1,134 m) having been formed by partial melting (20-25%) of undepleted spinel peridotite at a depth of 30-40 km (Green and Ringwood, 1967; Basaltic Volcanism Study Committee, 1981; Green, 1982).

Basaltic magmas with the chemical characteristics of the 3,720-ft (1,134-m) basalt could have been the parents of the rest of the basalts with lower Mg number (Basaltic Volcanism Study Committee, 1981; Klewin and Berg, 1991). The less primitive basalts contain successively lower Al₂O₃, Ni, and Cr concentrations and higher Fe₂O₃, TiO₂, K₂O, REE, Ba, Hf, and Sc concentrations with decreasing Mg number than the 3,720-ft basalt. Also, the lower Mg number rocks tend to have more negative Eu anomalies than the higher Mg number rocks. In addition, Sr is anomalously depleted in Thompson (1982) plots (fig. 12). These trends are consistent with formation from a magma similar in composition to the basalt at 3,720 ft that crystallized plagioclase, olivine, pyroxene, or spinel (Phinney, 1970; Green, 1972, 1982; Annells, 1974; Weiblen and Morey, 1980; Basaltic Volcanism Study Project, 1981; Carlson and Hart, 1988; Klewin and Berg, 1991). Indeed, least-squares modeling of the average compositions of the basalts at different Mg numbers gives low residuals only if olivine, clinopyroxene, plagioclase, and spinel crystallize in a 0.24:0.20:0.53:0.03 ratio. Also, plagioclase crystallization results in lower Sr concentrations and more negative Eu anomalies in residual melts, 'Whereas olivine and pyroxene crystallization results in lower Ni and Cr concentrations in residual melts. The large scatter in the data in variation diagrams is certainly not consistent with the differentiated rocks coming from a single homogeneous magma to form the range of basaltic compositions. Rather, there must have been a series of magmas formed over a long period of time, perhaps at different depths, that fractionated in a similar fashion (Brannon et al., 1979; Klewin and Berg, 1991). A similar conclusion that no single model, such as partial melting of a single source followed by fractional crystallization, can produce the range of data of the northern MRS samples was reached by Klewin and Berg (1991).

Figure 12--Spidergrams of the southern MRS basalts and gabbros [normalized after Thompson (1982)].

The absence of negative Nb-Ta anomalies or the presence of small negative Nb-Ta anomalies in most basalts (fig. 12) suggests that there has been little or only modest crustal contamination of the basalts (Carlson and Hart, 1988; Mahoney, 1988; Cox and Hawkesworth, 1985). Contamination of the basalts by silicic crustal melts should produce enrichment in Ba, Th, La, and Ce and depletion of TiO₂ and P in Thompson plots (Thompson et al., 1983; Marsh, 1989; Klewin and Berg, 1991). These characteristics are not observed in the basalts or gabbros (fig. 12); thus contamination by silicic melts is unlikely. Also, fractional crystallization rather than contamination is the most significant process for the observed elemental variation among the MRS rocks, as illustrated in a plot of Ce/Yb versus Mg number [fig. 13; after Lightfoot et al. (1991)]. Only Poersch sample 7159 has a significantly higher Ce/Yb ratio (fig. 13) than the other basalts; it also has the most negative Nb-Ta anomaly (fig. 11b). Thus sample 7159 may have had the most crustal contamination of any of the samples. Isotopic studies done on the uppermost gabbro sill suggest that crustal contamination was insignificant in that intrusion (Van Schmus et al., 1990). This gabbro also had no.significant negative Nb-Ta anomaly, consistent with this conclusion.

Figure 13--Variation diagram of Ce/Yb ratio versus Mg number [after Lightfoot et al. (1991)]. The trends predicted by contamination of melt by silica-rich rocks or by fractional crystallization are indicated by the arrows. Overall, most samples lie along a slightly increasing Ce/Yb ratio with decreasing Mg number, as expected by fractional crystallization. One basalt (Poersch 7159), however, contains an anomalously high Ce/Yb ratio.

Sedimentary rocks

Sedimentary rocks of the northern MRS have been reasonably well studied (Catacosinos, 1981; Daniels, 1982; Kalliokoski, 1982; Merk and Jirsa, 1982; Morey and Ojakangas, 1982; Ojakangas and Morey, 1982; Elmore, 1984). They have been divided into three sequences (Weiblen and Morey, 1980; Green, 1982; Morey and Green, 1982; Welblen, 1982). The oldest sequence consists of a prevolcanic suite of mature quartz-rich sandstones, suggesting stable cratonic deposition (Ojakangas and Morey, 1982). The second sequence of immature red bed sedimentary rocks is less voluminous. It consists of coarse-grained arkosic conglomerates to siltstones containing volcanic rock fragments interbedded with volcanic rocks (Ojakangas and Morey, 1982; Merk and Jirsa, 1982). The third sequence is mostly postvolcanic rift basin sedimentary rocks that become increasingly more mature and contain more material from outside the rift higher in the sequence (Daniels, 1982; Morey and Ojakangas, 1982; Van Schmus and Hinze, 1985). These rocks consist of conglomerate, sandstone, siltstone, and shale.

The interflow and postvolcanic sedimentary rocks of the Poersch well can, in general, be correlated with those in the northern portion of the rift. Their petrologic and mineralogic composition indicates that they are synrift deposits (Berendsen et al., 1988). This is also consistent with the observation that samples from the upper gabbro sill in the Poersch and OZ wells have a stable magnetic remanence that demagnetizes from shallow reverse inclination to steep reverse inclination and thus may correlate with the lower part of the Keweenawan Supergroup of the Lake Superior region (Van Schmus et al., 1990). The petrography and chemical composition of the abundant arkoses and minor siltstones are consistent with derivation from granitoids of the uplifted blocks adjacent to the rift. Thus erosion of the granitoid highlands must have been rapid enough and the volume large enough to overwhelm the debris formed from the local basalts. This interpretation is consistent with provenance studies of arkosic sedimentary rocks in the lower portion of the Poersch no. 1 well using U-Pb geochronology of detrital zircon and monazite, as their ages are 1.7-1.8 Ga, 1.4-1.5 Ga, and 1.1-1.2 Ga (Mark Martin, personal communication, 1992). This implies that the sedimentary rocks were derived from the older granitoid plutonic and metamorphic rocks (1,800-1,100 Ma) that occur on either side of the rift (Denison et al., 1984).

The style of tectonic activity in the southern MRS is similar to that observed to the north if indeed a high-angle reverse fault thrusted the volcanic section in the Poersch well over the sedimentary section. In the northern MRS the early extensional phase of rifting was characterized by broad subsidence with mostly basaltic rocks and little or no sedimentary rocks, whereas in the later stages of rifting thermal collapse took place and thick sequences of clastic sedimentary rocks accumulated in the rift (Green, 1983; Behrendt et al., 1988; Cannon et al., 1989). North-northeast-trending high-angle normal and reverse faults are not uncommon along the trace of the rift in Kansas (Berendsen et al., 1989). Petrologic and mineralogic studies, including heavy mineral analysis on cuttings from available wells, indicate that rift development in northeastern Kansas followed a similar course to that in the northern portion of the MRS (Berendsen et al., 1989). Repeated movement on north-northeast-trending and northwest-trending faults, culminating in documented vertical displacement of several thousand feet during the Late Mississipplan and Early Pennsylvanian, resulted in complex tectonic patterns that are often difficult to interpret.

Summary

Based on the available data, it is reasonable to assume that the stratigraphic succession at the Poersch and possibly the nearby OZ wells is interrupted by a high-angle reverse fault, causing younger sedimentary rocks to be placed stratigraphically beneath older mafic volcanic and intrusive rocks. The following scenario for the formation of early Keweenawan rocks is thus hypothesized in the vicinity of the Poersch well. The initial opening of the rift was characterized by the outpouring of at least 4,000 ft (1,200 m) of basalt with minor felsite and interflow siltstone. This event was followed by deposition of a minimum of 4,000 ft of arkoses and siltstones, much of it derived from the rift shoulders. Seismic evidence indicates that a much thicker package of clastic sedimentary rocks was deposited in the subsiding basin (Serpa et al., 1984). Neither the tops nor the bottoms of the sections are exposed; thus the total thickness of the igneous and sedimentary sequence is unknown. A 265-ft-thick (80.8-m-thick) gabbro sill, dated at 1,097.5 ± 3 Ma, was intruded near the top of the volcanic sequence, and it is bounded by an unconformity separating it from the overlying Paleozoic sedimentary rocks.

The basalts and gabbros are subalkalic to alkalic and tend to follow tholeiitic trends, although a number of them are high-alumina basalts and thus have a chemical composition similar to mafic rocks in the MRS exposed to the north. The most primitive basalt could have formed by partial melting (20-25%) of a spinel peridotite at a depth of 30-40 km. Further fractionation of olivine, pyroxene, plagioclase, and spinel from magmas with a composition similar to that of the most primitive basalt could explain the chemical variation of the rest of the samples. The moderate scatter of the data in variation diagrams, however, precludes the differentiated rocks forming from a single magma. Rather, it is more likely that a series of undifferentiated magmas that were similar in composition formed at different depths. Fractional crystallization of similar minerals over a range of depths could have fractionated the primary magmas to produce the scatter in the observed variation diagrams.

Acknowledgments

We thank the Kansas Geological Survey for support of this research. We appreciate the cooperation of Karmie Galle, who analyzed some of our samples for major elements using atomic absorption, and Mike Magnuson, who provided part of the x-ray fluorescence results. We also thank the crew of the Kansas State nuclear reactor for irradiation of our samples and use of their counting equipment, Penny Cullers for preparation of some figures, and Lee Ann Davidson for the preparation of the tables.

References

Annells, R. G., 1974, Keweenawan volcanic rocks of Michipicoten Island, Lake Superior, Ontario, an eruptive center of Proterozoic age: Geological Survey of Canada, Bulletin 218, 141 p.

Basaltic Volcanism Study Project, 1981, Basaltic volcanism on the terrestrial planets: Pergamon Press, New York, 1,286 p.

Behrendt, J. C., Green, A. G., Cannon, W. F., Hutchinson, D. R., Lee, M. W., Milkereit, B., Agena, W. F., and Spencer, C., 1988, Crustal structure of the midcontinent rift system-results from GLIMPCE deep seismic reflection profiles: Geology, v. 16, p. 81-85

Berendsen, P., Blair, K. P., and Newell, K. D., 1989, Structural aspects of the midcontinent rift system in Kansas: Transactions, 1989 AAPG Midcontinent Section Meeting, 26 p.

Berendsen, P., Borchuding, R., Doveton, J., Gerhard, L., Newell, K. D., Steeples, D., and Watney, W. L., 1988, Texaco Poersch #1, Washington County, Kansas-preliminary geologic report of the pre-Phanerozoic rocks: Kansas Geological Survey, Open-File Report 88-22, 116 p.

Brannon, J. C., Haskin, L. A., and Green, J. C., 1979, Compositional differences in planetary flood basalts: Proceedings, Lunar and Planetary Science Conference X, p. 152-154

Buckley, D. E., and Cranston, R. E., 1971, Atomic absorption analyses of eighteen elements from a single decomposition of aluminosilicates: Chemical Geology, v. 7, p. 273-284

Cannon, W. F., Green, A. G., Hutchinson, D. R., Lee, M., Milkereit, B., Behrendt, J. C., Halls, H. C., Green, J. C., Dickas, A. B., Morey, G. B., Sutcliffe, R., and Spencer, C., 1989, The North American midcontinental rift beneath Lake Superior from GLIMPCE seismic reflection profiling: Tectonics, v. 8, p. 305-332

Carlson, R. W., and Hart, W. K., 1988, Flood basalt volcanism in the northwestern United States; in, Continental flood basalts, J. D. Macdougall, ed.: Kluwer Academic Publishers, Dordrecht, Netherlands, p. 35-61

Catacosinos, R. A., 1981, Origin and stratigraphic assessment of pre-Mount Simon clastics (Precambrian) of the Michigan basin: American Association of Petroleum Geologists Bulletin, v. 65, p. 1,617-1,620

Cornwall, H. R., and Rose, H. J., Jr., 1957, Minor elements in Keweenawan lavas, Michigan: Geochimica et Cosmochimica Acta, v. 12, p. 209-224

Cox, K. G., 1980, A model for flood basalt volcanism: Journal of Petrology, v. 21, p. 629-650

Cox, K. G., and Hawkesworth, C. J., 1985, Geochemical stratigraphy of Deccan Traps at Mahabaleshwar, Western Ghats, India, with implication for open system magmatic processes: Journal of Petrology, v. 26, p. 355-377

Cox, K. G., Bell, J. D., and Pankhurst, K G., 1979, The interpretation of igneous rocks: Allen and Unwin, London, 450 p.

Cullers, R. L., 1988, Mineralogical and chemical changes of soil and stream sediment formed by intense weathering of the Danberg granite, Georgia: Lithos, v. 21, p. 301-314

Cullers, R. L., Barrett, T., Carlson, R., and Robinson, B., 1987, Rare earth element and mineralogic changes in Holocene soil and stream sediment--a case study in the Wet Mountains, Colorado, USA: Chemical Geology, v. 63, p. 275-297

Cullers, R. L., Chaudhuri, S., Kilbane, N., and Koch, R. J., 1979, The rare earth contents in different size fractions and whole rocks from Pennsylvanian and Permian platform sediments from the midcontinent of the USA: Geochimica et Cosmochimica Acta, v. 43, p. 1,285-1,302

Cullers, R. L., Ramakrishnan, S., Berendsen, P., and Griffin, T., 1985, Geochemistry and petrography of lamproites, Late Cretaceous age, Woodson County, Kansas, USA: Geochimica et Cosmochimica Acta, v. 49, p. 1,383-1,402

Cullers, R. L., Chaudhuri, S., Arnold, B., Lee, M., and Wolf, C. W., 1975, Rare earth distributions in clay minerals and clay-sized fractions of the Lower Permian Havensville and Eskridge shales of Kansas and Oklahoma: Geochimica et Cosmochimica Acta, v. 39, p. 1,691-1,703

Daniels, P. A., Jr., 1982, Precambrian sedimentary rocks: Oronto Group, Michigan-Wisconsin; in, Geology and tectonics of the Lake Superior basin, R. J. Wold and W. J. Hinze, eds.: Geological Society of America, Memoir 156, p. 107-133

Davis, D. W., and Paces, J. B., 1990, Time resolution of geologic events on the Keweenawan Peninsula and implications for the development of the midcontinent rift system: Earth and Planetary Science Letters, v. 97, p. 54-64

Davis, D. W., and Sutcliffe, R. H., 1985, U-Pb ages from the Nipigon plate and northern Lake Superior: Geological Society of America Bulletin, v. 96, p. 1,572-1,579

Denison, R. E., Lidiak, E. G., Bickford, M. E., and Disvarsanyi, E. B., 1984, Geology and geochronology of Precambrian rocks in the central interior of the United States: U.S. Geological Survey, Professional Paper 1241-C, 20 p.

Dickas, A. B., 1984, Midcontinent rift system-Precambrian hydrocarbon target: Oil and Gas Journal, v. 70, p. 151-159

Dickas, A. B., 1986, Comparative Precambrian stratigraphy and structure along the midcontinent rift: American Association of Petroleum Geologists Bulletin, v. 70, p. 225-238

Elmore, R. D., 1984, The Copper Harbor Conglomerate--a Late Precambrian fining-upwards alluvial fan sequence in northern Michigan: Geological Society of America Bulletin, v. 95, p. 610-617

Godlevskiy, M. N., 1959, Trappi rudonosnze intruzil Noril'skogo raiona: Gosgeltekhizdat. Translated by Canada Translation Bureau

Gordon, G. E., Randle, K., Goles, G. G., Corliss, J. B., Beeson, M. H., Oxley, S. S., and Gram, O. E., 1968, Instrumental activation analysis of standard rocks with high resolution x-ray detectors: Geochimica et Cosmochimica Acta, v. 32, p. 369-396

Green, D. H., and Ringwood, A. E., 1967, The genesis of basaltic magmas: Contributions to Mineralogy and Petrology, v. 15, p. 103-190

Green, J. C., 1972, North Shore Volcanic Group; in, A centennial volume, P. K. Sims and G. B. Morey, eds.: Minnesota Geological Survey, p. 294-332

Green, J. C., 1982, Geology of the Keweenawan extrusive rocks; in, Geology and tectonics of the Lake Superior basin, R. J. Wold and W. J. Hinze, eds.: Geological Society of America, Memoir 156,p.47-56

Green, J. C., 1983, Geologic and geochemical evidence for the nature and development of the middle Proterozoic (Keweenawan) midcontinent rift of North America: Tectonophysics, v. 94, p. 413-437

Gromet, L. P., Dymek, R. F., Haskin, L. A., and Korotev, R. L., 1984, The "North American Shale composite"--its compilation, major, and trace element characteristics: Geochimica et Cosmochimica Acta, v. 48, p. 2,469-2,482

Irvine, T. N., and Baragar, W. R. A., 1971, A guide to the chemical classification of the common volcanic rocks: Canadian Journal of Earth Science, v. 8, p. 523-548

Jacobs, J. W., Korotev, R. L., Blanchard, D. P., and Haskin, L. A., 1977, A well-tested procedure for instrumental activation analysis of silicate rocks and minerals: Journal of Radioanalytical Chemistry, v. 40, p. 321

Kaltiokoski, J., 1982, Jacobsville Sandstone; in, Geology and tectonics of the Lake Superior basin, R. J. Wold and W. J. Hinze, eds.: Geological Society of America, Memoir 156, p. 147-155

Klewin, K. W., and Berg, J. H., 1991, Petrology of the Keweenawan Mamainse Point lavas, Ontario--petrogenesis and continental rift evolution: Journal of Geophysical Research, v. 96, p. 457-474

Lightfoot, P. C., Sutcliffe, R. H., and Doherty, W., 1991, Crustal contamination identified in Keweenawan Osler Group tholeiites, Ontario--a trace element perspective: Journal of Geology, v.99,p.739-760

Mahoney, J. J., 1988, Deccan Traps; in, Continental flood basalts, Macdougall, J. D., ed.: Kluwer Academic Publishers, Dordrecht, Netherlands, p. 151-194

Marsh, J. S., 1989, Geochemical constraints on coupled assimilation and fractional crystallization involving upper crustal composition and tholelitic magma: Earth and Planetary Science Letters, v. 92, p. 70-80

Merk, G. P., and Jlrsa, M. A., 1982, Provenance and tectonic significance of the Keweenawan interflow sedimentary rocks; in, Geology and tectonics of the Lake Superior basin, R. J. Wold and W. J. Hinze, eds.: Geological Society of America, Memoir 156, p. 97-106

Morey, G. B., and Green, J. C., 1982, Status of the Keweenawan as a stratigraphic unit in the Lake Superior region; in, Geology and tectonics of the Lake Superior basin, R. J. Wold and W. J. Hinze, eds.: Geological Society of America, Memoir 156, p. 15-25

Morey, G. B., and Ojakangas, R. W., 1982, Keweenawan sedimentary rocks of eastern Minnesota and northwestern Wisconsin; in, Geology and tectonics of the Lake Superior basin, R. J. Wold and W. J. Hinze, eds.: Geological Society of America, Memoir 156, p.135-146

Nakamura, N., 1974, Determination of REE, Ba, Fe, Mg, Na, and K in carbonaceous and ordinary chondrites: Geochimica et Cosmochimica Acta, v. 38, p. 757-775

Ojakangas, R. W., and Morey, G. B., 1982, Keweenawan prevolcanic quartz sandstones and related rocks of the Lake Superior region; in, Geology and tectonics of the Lake Superior basin, R. J. Wold and W. J. Hinze, eds.: Geological Society of America, Memoir 156, p. 85-96

Palmer, H. C., and Davis, D. W., 1987, Paleomagnetism and U-Pb geochronology of volcanic rocks from Michipicoten Island, Lake Superior, Canada--precise calibration of the Keweenawan polar wander track: Precambrian Research, v. 37, p. 157-171

Phinney, W. C., 1970, Chemical relations between Keweenawan lavas and the Duluth complex, Minnesota: Geological Society of American Bulletin, v. 81, p. 2, 487-2,696

Prestvik, T., and Goles, G. G., 1985, Comments on the petrogenesis and the tectonic setting of the Columbia River basalts: Earth and Planetary Science Letters, v. 72, p. 65-73

Roeder, P. L., and Emslie, R. F., 1970, Olivine-liquid equilibrium: Contributions to Mineralogy and Petrology, v. 29, p. 275-289

Ronov, A. B., Balashov, Y. A., Girin, Y. P., Bratishko, R. K., and Kazakov, G. A., 1974, Regularities of rare earth distribution in the sedimentary shell and in the crust of the earth: Sedimentology, v. 21, p. 171-193

Ruegg, N. R., 1976, Characteristicas de distrbuicao e teor de elementos principais em rockas basalticas da bacia do Parana: Bolivian Institute Geociencia, v. 7, p. 81-106

Serpa, L., Setzer, T., Farmer, H., Brown, L., Oliver, J., Kaufman, S., Sharp, J., and Steeples, D., 1984, Structure of the southern Keweenawan rift from COCORP surveys across the Midcontinent Geophysical Anomaly in northeastern Kansas: Tectonics, v. 3, p. 367-384

Taylor, S. R., and McLennan, S. M., 1985, The continental crust--its composition and evolution: Blackwell Scientific Publications, Oxford, England, 312 p.

Thompson, R. N., 1982, Magmatism of the British Tertiary volcanic province: Scottish Journal of Geology, v. 18, p. 49-107

Thompson, R. N., Morrison, M. A., Dickin, A. P., and Hendry, G. L., 1983, Continental flood basalts-arachnids rule OK?; in, Continental basalts and mantle xenoliths, C. J. Hawkesworth and M. J. Norry, eds.: Shiva Press, Cambridge, Massachusetts, p.158-185

Van Schmus, W. R., 1992, Tectonic setting of the midcontinent rift system: Tectonophysics, v. 213, p. 1-15

Van Schmus, W. R., and Hinze, W. J., 1985, The midcontinent rift system: Annual Reviews of Earth and Planetary Sciences, v. 13, p.345-383

Van Schmus, W. R., Martin, M. W., Sprowl, D. R., Gelssman, J., and Berendsen, P., 1990, Age, Nd and Pb isotopic composition, and magnetic polarity for subsurface samples of the 1,100 Ma midcontinent rift: Geological Society of America, Abstracts, v. 22, p. A174

Van Schmus, W. R., Bickford, M. E., Anderson, R. R., Shearer, C. K., Papike, J. J., and Nelson, B. K., 1989, Quimby, Iowa, scientific drill hole--definition of Precambrian crustal features in northwestern Iowa: Geology, v. 17, p. 536-539

Weiblen, P. W., 1982, Keweenawan intrusive rocks; in, Geology and tectonics of the Lake Superior basin, R. J. Wold and W. J. Hinze, eds.: Geological Society of America, Memoir 156, p. 57-82.

Weiblen, P. W., and Morey, G. B., 1980, A summary of the stratigraphy, petrology, and structure of the Duluth complex: American Journal of Science, v. 280-A, p. 88-133

Woelk, T. S., and Hinze, W. J., 1991, Model of the midcontinent rift system in northeastern Kansas: Geology, v. 19, p. 277-280

Wood, D. A., Gibson, I. L., and Thompson, R. N., 1976, Element mobility during zeolite facies metamorphism of the Tertiary basalts of eastern Iceland: Contributions to Mineralogy and Petrology, v. 55, p. 241-254

Complete Article

The complete article is available as an Acrobat PDF file: B235_pt4.pdf (308 kB). You will need the Acrobat PDF Reader, available free from Adobe, to read this report.

Kansas Geological Survey
Placed online Dec. 2012
http://www.kgs.ku.edu/Current/1993/Cullers/index.html
email:webadmin@kgs.ku.edu