Paleoclimatic inferences from paleopedology and ... - GSA Journals

Paleoclimatic inferences from paleopedology and magnetism of the Permian Maroon Formation loessite, Colorado, USA

Kristy L. Tramp† Department of Earth Science, Rice University, Houston, Texas 77005, USA

G.S. (Lynn) Soreghan‡ R. Douglas Elmore§ School of Geology and Geophysics, University of Oklahoma, Norman, Oklahoma 73019, USA

ABSTRACT The Maroon Formation in the eastern Eagle basin (Colorado) consists of .700 m of lithified loess with .200 paleosols interpreted as Protosols and Argillisols on the basis of field, petrographic and geochemical data. Additionally, magnetic susceptibility aids assessment of the intensity of pedogenesis. Bulk magnetic susceptibility (xb) through the section repeatedly fluctuates between low values (average 3.51 3 1028 6 1.59 3 1028 m3/kg) in parent loessite and higher values (average 5.70 3 1028 6 2.70 3 1028 m3/kg) in paleosols. Moreover, magnetic susceptibility positively correlates with abundance of clay-sized material as well as Al2O3 and K2O and effectively distinguishes Protosols and Argillisols. Lowtemperature demagnetization indicates the presence of ultra-fine-grained magnetite. The integration of geochemical, petrographic, and rock magnetic data suggest that changes in magnetic susceptibility reflect pedogenesis and relate primarily to climate- and time-dependent pedogenic production and concentration of ultra-finegrained magnetite. The Maroon Formation loess and associated soils accumulated in an overall arid system, documented in part by formation of incipiently formed paleosols developed into Argillisols by eolian clay and carbonate additions rather than by in situ clay formation. However, the paleosols showing bulk magnetic susceptibility values of greater than 200 xb document a highE-mail: [email protected]. E-mail: [email protected]. § E-mail: [email protected]. † ‡

frequency (1042105 yr) fluctuation between arid times of loess accumulation and slightly wetter times of reduced silt influx and resultant pedogenesis. This fluctuation likely reflects glacial-interglacial climate shifts that operated in low-latitude Pangea during icehouse conditions. These results suggest that climate-related magnetic susceptibility variations within loess successions can be preserved and useful in very ancient (pre– Pliocene–Pleistocene) sequences. Keywords: magnetic susceptibility, loess, paleosols, Eagle basin, Ancestral Rocky Mountains. INTRODUCTION The late Paleozoic represents the bestdocumented pre-Quaternary icehouse, and upper Paleozoic strata provide an excellent opportunity to study climate behavior and evolution during this time. Continental glaciation over Gondwana is well documented in the preserved glacial strata, and coeval strata of low-latitude Euramerica consist commonly of cyclothems that record repeated glacioeustatic fluctuations (e.g., Wright and Vanstone, 2001; Crowell, 1978). This glacioeustatic signature records relatively highfrequency (105 yr; Heckel, 1986) climate change during the late Paleozoic icehouse; less well documented is the continental record of low-latitude climate change. In this paper, we focus on the purely continental record preserved in the Maroon Formation loessite of northwestern Colorado (western equatorial Pangea), which accumulated in Early Permian time, during waning

phases of the late Paleozoic icehouse. Loess rivals lake deposits in its remarkable stratigraphic completeness, and so offers an excellent opportunity for studying high-resolution continental climate. Moreover, magnetic susceptibility variations in recent (Pliocene–Pleistocene) loess successions have been increasingly recognized as a robust paleoclimate proxy. Heller and Liu (1984) first documented bulk magnetic susceptibility (xb) lows in parent loess and highs in paleosols throughout a 150 m Pliocene– Pleistocene sequence of the Chinese Loess Plateau. Kukla et al. (1988) subsequently correlated these xb patterns to the deep-sea oxygen isotope record, thereby directly linking loess deposition and global ice volume for the Quaternary icehouse. Many subsequent studies have used xb as a paleoclimate proxy in Plio–Pleistocene loess-hosted paleosol sequences (e.g., Kukla and An, 1989; Beget et al., 1990; Wang et al., 1990; An et al., 1991; Feng et al., 1994). Although the origin of the xb signature in these Pliocene–Pleistocene loess deposits is still debated, numerous studies have documented differences in magnetic mineralogy and magnetic grain size between parent loess and paleosol, and workers infer pedogenic production of ferrimagnetic minerals during soil-forming climate phases (e.g., Maher, 1986; Maher and Taylor, 1988; Maher and Thompson, 1991; Zheng et al., 1990; Liu et al., 1992; Verosub et al., 1993; Evans and Heller, 1994; Singer et al., 1996). Magnetic susceptibility is particularly useful for differentiating unweathered loess from pedogenically altered loess, especially in incipiently developed paleosols where macroscopic char-

GSA Bulletin; May/June 2004; v. 116; no. 5/6; p. 671–686; doi: 10.1130/B25354.1; 9 figures; Data Repository item 2004077.

For permission to copy, contact [email protected] q 2004 Geological Society of America

671

TRAMP et al.

Figure 1. Paleogeography of the late Paleozoic Eagle basin. Insert places the Eagle basin in modern coordinates. The Maroon Formation loessite was deposited downwind of Maroon Formation eolian sand-sheet deposits (Johnson, 1989). The 700 m section of Maroon Formation loessite studied begins above a fault-contact base and is overlain by the State Bridge Formation. The diagonal line shows the approximate paleolatitude. A–A9 line locates cross section (Fig. 2). Modified from Johnson (1987a, 1987b, 1989) and Johnson et al. (1992). acteristics (color change, clay enrichment) may not be readily apparent (Maher et al., 1994). Rock-magnetic and geochemical studies of loess sequences have contributed substantially to our understanding of climate change in the Pliocene–Pleistocene (e.g., Bronger and Heinkele, 1990; Kemp et al., 1995; Feng, 1997), but have not generally been applied to ancient loess systems. Soreghan et al. (1997) established the presence of ultra-fine-grained magnetite in high xb paleosols over a small section of the upper Paleozoic Maroon Formation loessite, and Soreghan et al. (2002a) documented pedogenic-magnetic relationships within the Pennsylvanian–Permian lower Cutler beds (Rico and Halgaito Formations of southwestern Utah). Here, we integrate sedimentologic, geochemical, and rock-magnetic analyses to evaluate the preservation and origin of the xb signature throughout the Maroon Formation loessite. GEOLOGIC SETTING The Eagle basin (northwest Colorado) formed near the paleoequator (;5–108N) as

672

part of the Ancestral Rocky Mountains system in late Paleozoic time (Kluth and Coney, 1981; Fig. 1 herein). Pennsylvanian and Lower Permian strata of the Eagle basin consist of marine shale, carbonate, and evaporite, and continental clastic rocks that together record a marine- to terrestrial transition from Pennsylvanian to Early Permian time (Mallory, 1972; DeVoto, 1980; Johnson et al., 1992; Fig. 2 herein). The marine-paralic strata contain numerous meter-scale, inferred glacio-eustatic, sequences (Driese and Dott 1984; DeVoto et al., 1986; Johnson et al., 1988, 1992) that record linked eustatic-climatic fluctuations that operated in late Paleozoic time (Cecil 1990; Johnson et al., 1992; Soreghan, 1994; Miller et al., 1996; Rankey, 1997). Within the central Eagle basin, for example, cyclic alternations of eolian and fluvial strata in the Maroon Formation have been linked to fluctuations between more arid glacial and more humid interglacial climate phases (Johnson et al., 1988). Downwind of these eolian sand sheets of the Maroon Formation, a thick sequence of loess with intercalated paleosols collected on the (paleo) northern side of the Sawatch uplift (Johnson 1987a, 1989; Fig. 2 herein).

The Maroon Formation is nonfossiliferous, but has been presumed to be Middle Pennsylvanian (Desmoinesian) to Early Permian (Wolfcampian) in age on the basis of constraints imposed by enclosing strata (Tweto and Lovering, 1977; Johnson, 1987b). To refine the dating in the study section, we attempted to use palynology, but recovery was extremely poor, consisting of only scattered and abraded Middle Pennsylvanian Lycospora and Densosporites that were reworked (C. Eble, University of Kentucky, 2000, written commun. concerning a Maroon Formation palynological study). Preliminary detrital-zircon analyses of the Maroon loessite (Soreghan et al., 2002b, and unpublished data, 2003), however, revealed the local presence of very young zircons that indicate the age of the unit to be no older than earliest Permian (Wolfcampian). METHODS We described and sampled 700 m of a continuous section of the Maroon Formation loessite where it is exposed near Basalt, Colorado (Fig. 1). Paleosols were identified and de-

Geological Society of America Bulletin, May/June 2004

PALEOCLIMATIC INFERENCES, PERMIAN MAROON FORMATION LOESSITE, COLORADO, USA

Figure 2. (A) Diagrammatic cross section and (B) stratigraphic column of upper Paleozoic Eagle basin strata in the vicinity of the western Sawatch uplift (see Fig. 1). Approximate thicknesses of strata are shown. Cross section location (A–A9) is shown in Figure 1. Figures are modified from Mallory (1960, 1972), DeVoto (1972, 1980); Johnson (1987a, 1987b), and Johnson et al. (1993).

scribed following the methods of Retallack (1988), and samples for petrographic, geochemical, and rock-magnetic analyses were collected every 0.5 to 1.0 m in massive reddishorange (10R 6/6 to 10R 7/4) siltstone (inferred loessite) and at 0.05 to 0.10 m intervals in reddish-brown (10R 4/6) mudstone/siltstone (inferred paleosol) intervals. Approximately 1500 paleomagnetic sample cores over the 700 m section (2.5 cm diameter) were drilled with a portable gas-powered and water-cooled drill. Hand samples (100–500 g) for palynological study were collected from the base, middle, and top of 50 loessite-paleosol profiles spaced throughout the 700 m section. Selected cores were oriented by using a Brunton and inclinometer for rock-magnetic analyses. Paleosols were qualitatively rated by using presence and intensity of macroscopic pedogenic features (cf. Harden, 1982), and a subset of nine paleosol profiles reflecting the range of variation present were selected for detailed petrographic and geochemical analysis. These profiles were then described and ranked on the basis of micromorphologic features observed in thin section. Thin sections were stained for carbonates (Dickson, 1966) and etched with hydrofluoric acid to distinguish feldspar from quartz. Point counts (400 points) of thin sections through selected profiles were performed to assess modal proportions of quartz, feldspar, opaque minerals, mica, heavy minerals, fine-grained matrix, and carbonate. Grains ,4 mm were counted as clay sized; and grains .4 mm were identified as quartz, feldspar, dolomite, calcite, mica, and opaque or heavy min-

erals. Carbonate grain-size classification follows Folk (1962). A loessite and a paleosol sample from both a weakly and well-developed paleosol were analyzed by using a scanning electron microscope (SEM) to assess clay texture and morphology. To evaluate clay type and approximate content, three paleosol profiles from the base, middle, and top of the section were selected for X-ray diffraction (XRD) analysis. Bulk-rock powders and clay extracts were analyzed to assess clay mineralogy and approximate content. Fifty-two samples from nine profiles were powdered and analyzed (by a commercial laboratory) with whole-rock Xray fluorescence spectrometry (XRF) to determine proportions of major element oxides (SiO2, Al2O3, Fe2O3, MgO, CaO, K2O) and selected trace elements (Zr, Ti). Major element oxides were normalized to ZrO2 in order to facilitate comparisons among paleosol profiles, as well as assess inputs of allochthonous dust during pedogenesis. All cores were cut to standard length, and weight-normalized bulk magnetic susceptibility (xb) was measured on a Sapphire SI-2 instrument. An average of three measurements with errors at least an order of magnitude less than the measured value was recorded for all samples. Details of the paleomagnetic and some rock-magnetic analyses and results are in Appendix DR1.1 1 GSA Data Repository item 2004077, Appendix DR1, Figures DR1–DR2, and Tables DR1–DR4, is available on the Web at http://www.geosociety.org/ pubs/ft2004.htm. Requests may also be sent to [email protected].

Hysteresis properties for samples from a loessite-paleosol (Argillisol) couplet were determined by using a ‘‘Micromag’’ alternating gradient force magnetometer at the Institute for Rock Magnetism, University of Minnesota, to test the results of earlier studies (Soreghan et al., 1997; Cogoini et al., 2000). The paramagnetic susceptibility (xp) was recorded from the high-field slopes of the hysteresis curves. Maximum field used was 2 T. The ferrimagnetic susceptibility (xf) was calculated by subtracting xp from xb. Normalized xf was calculated by dividing the xf by the roomtemperature saturation magnetization determined from the hysteresis loops. Differences in this parameter reflect concentrationindependent variation in the proportions of ultra-fine-grained material (Hunt et al., 1995). The thermal decay patterns of lowtemperature saturation isothermal remanence magnetizations (SIRM) were recorded on a Magnetic Property Measurement System (MPMS) at the Institute for Rock Magnetism, University of Minnesota, for samples from another representative loessite-paleosol couplet. These data were used to estimate the relative percent of superparamagnetic (SP) grains, which is the difference in remanence between 50 K and 300 K, after subtracting the drop in remanence at the Verwey transition (Hunt et al., 1995; Mike Jackson, 2001, personal commun.). This relative amount should only be considered approximate because hematite is present in the samples and contributes to the remanence.


673

TRAMP et al.

674



Petrography and Geochemistry of the Loessite

DATA AND INTERPRETATIONS Field Sedimentology Johnson (1989) interpreted the Maroon Formation in the study location as loessite (lithified loess) with interbedded paleosols. The following brief summary of the field sedimentology is based heavily on Johnson (1989) and supplemented with our observations. Most of the section consists of reddishorange (10R 6/6 to 10R 7/4), massive siltstone beds of variable thickness (a few centimeters to several meters). This thickness variation might reflect variations in rate of silt influx during loess accumulation or brief shifts in depositional conditions. Commonly, reddishbrown (10R 4/6) siltstone units exhibit gradational lower contacts and sharp upper contacts with reddish-orange siltstone. The reddish-brown siltstone units locally fine upward from siltstone to mudstone and commonly exhibit subangular blocky structure, subvertical downwardly bifurcating tubules (#3 cm diameter and 1 m length), and local low-chroma mottles. The reddish-brown siltstone units are interpreted as paleosols on the basis of the presence of pedogenic structure, increased clay (relative to parent loessite), root traces, and reduction spots. Mudcracks, small horizontal traces, and rare semispherical impressions (raindrop imprints, cf. Robb, 1992) also occur locally. Although loessite with interbedded paleosols constitutes most of the 700-m-thick section, the lower 200 m includes (1) relatively common occurrences of mudstone units (up to 2 m thick) that drape gentle paleotopography and are interpreted as ephemeral pond deposits, and (2) upwardly fining intervals consisting locally of a basal intraclastic conglomerate overlain by sandstone and siltstone and interpreted as fluvial-channel deposits. Further, siltstone units in the uppermost 100 m of the section locally exhibit lateral-accretion bedding, recording fluvial reworking of loess.

The parent material (loessite) for each paleosol consists of compositionally uniform reddish-orange (10R 6/6 to 10R 7/4) siltstone (Fig. 3A) with a modal analysis (n 5 8) of 55.7% 6 5.1% (s) quartz, 27.3% 6 2.4% feldspar, ;2.2% 6 2.8% accessory phases (mica, opaque and heavy minerals), 3.0% 6 6.3% dolomite, and 11.8% 6 0.8% finegrained matrix. Bulk-rock geochemical data for several Maroon Formation loessite samples (Table DR1 [see footnote 1]) also indicate a relatively homogeneous composition. XRD analysis shows that clay-mineral content is at threshold detection levels (,10%) and consists of illite and subordinate chlorite. Locally, fine-grained (,4 mm) material (hereafter referred to as ‘‘matrix’’) occurs as grain rims that thicken in grain indentations (Fig. 3B) similar to features interpreted in other eolian systems as inherited clay coats (Wilson, 1992). Cements within the loessite include dolomite and, locally (in the lower 200 m), calcite, in addition to silica, minor potassium feldspar, and hematite. Carbonate ranges from microspar to spar (,10 to 40 mm) and is typically disseminated. Paleopedology Although outwardly similar in appearance, paleosols display subtle variations in texture, color, structure, and mottling intensity. In paleosols, peds are preserved by remnant cutans that define ped surfaces (Retallack, 1988; Wright, 1992; Mack, 1997); the thickness of intervals exhibiting color change and pedogenic structure (0.3 to 1.0 m) likely reflects the degree of pedogenic development. Irregular, downwardly bifurcating tubules define root traces and are commonly preserved in paleosols along with diffuse, drab-colored mottles or haloes associated with micro–reducing environments around former plant roots (Retallack, 1988; Wright, 1992; Mack, 1997).

On the basis of presence and intensity of such features, paleosols were macroscopically rated by using a point system (cf. Bilzi and Ciolkosz, 1977; Harden, 1982; Bown and Kraus, 1987; Retallack, 1988; Table DR2 [see footnote 1]), wherein most Maroon Formation paleosols rank between 1 and 10 points (Table DR3 [see footnote 1]). To further refine rankings and classification, we analyzed micromorphology, mineralogy, and major oxide geochemistry for the nine selected paleosol profiles. Micromorphology was assessed following the techniques of Brewer (1976), Bullock et al. (1985), Wright (1991), Retallack (1990, 1993), and Fitzpatrick (1984). Pedogenic development was rated by using identification of key micromorphologic pedogenic features including presence of agglomeroplasmic texture, sepic clay, channel argillans, and carbonate fabric. Matrix-to-grain ratios, clay alignment, and clay segregation (along peds) generally increase with paleosol maturation (Brewer, 1976; Retallack, 1990; Fitzpatrick, 1984). By using these methods, we recognize three distinct types of paleosols in the Maroon Formation loessite, as detailed below and summarized in Table DR3. Weakly Developed Paleosols Macro- and Micromorphology Weakly developed paleosols locally exhibit melanization (darkening), slight pedogenic structure, and rare root traces, justifying 1 to 5 points on our rating scale (Table DR3 [see footnote 1]). These paleosols exhibit no macroscopic horizonation, although slight enrichments in carbonate (dolomite) and/or finegrained matrix are detectable in petrographic and geochemical data (Figs. 4A, 4B). Petrographic analysis indicates that dolomite microspar to spar (,10 to 40 mm) is disseminated and reaches proportions of up to 20% at profile tops (Fig. 3C). Fine-grained matrix generally occurs as clay coats and reaches proportions of up to 20% near paleosol tops (Fig. 3D). Granular to intertextic crys-

Figure 3. Outcrop photographs of Maroon Formation loessite and photomicrographs of loessite and paleosol microfabrics. (A) View of vertical section of Maroon Formation loessite and interbedded paleosols (dark bands denoted with arrows). (B) Section fifteen meter thick section of interbedded loessite and paleosols (denoted by arrows) of study section. (C) Typical loessite fabric. Width of field of view represents 1.44 mm; cross-polarized light. (D) Loessite grains with dust rims that are commonly thicker in grain indentations. Width represents 0.72 mm; plane-polarized light. (E) Outcrop view of weakly developed paleosol (Protosol) denoted with P. Image height represents ;1.5 m. (F) Disseminated dolomite cement in a carbonate-rich zone. Width represents 0.72 mm; cross-polarized light. (G) Protosol clay-rich zone where clay infills cracks, but does not constitute an appreciable amount of the fabric. Width represents 1.44 mm; plane-polarized light. (H) Disseminated dolomite cement in carbonate-rich zone in an Argillisol. Width represents 0.72 mm; crosspolarized light. (Caption continued on p. 677.)

N


675

TRAMP et al.

676



tic fabric prevails where dolomite is most abundant, whereas an intertextic argillasepic to insepic fabric characterizes upper reaches of these paleosols (Fig. 3D). Geochemistry Weakly developed paleosols exhibit flat geochemical patterns relative to those exhibited by other paleosol types (Fig. 4). Note that major oxides are normalized to ZrO2 to avoid the problem of closure and to aid evaluation of dust additions to the soil profile (cf. Reheis, 1990; Reheis et al., 1995). We follow many other soil studies in using ZrO2 (contained in zircon) as an index factor, owing to its resistance to weathering. (For brevity, Zr-normalized oxides are hereafter abbreviated to their elements.) In weakly developed paleosols, the range of variation of Ti and Al is within 1s of the variation in average parent loessite (Tables DR1, DR4a [see footnote 1]). Paleosol Interpretation and Classification On the basis of their very incipient development marked in part by preservation of primary sedimentary structures and lack of distinctive horizonation, these paleosols are interpreted as Protosols following the classification of Mack et al. (1993). The nearly flat trends in major oxides for Protosols (Figs. 4A, 4B) reflect weak pedogenesis. The upward decrease of Si reflects dilution by clay and carbonate in the upper parts of the solum. The Mg and Ca trends track dolomite content (Figs. 4A, 4B). Values of Al and K are proxies for clay minerals and also mimic trends in clay-sized material. Zr and Ti are extremely immobile and should become enriched as labile minerals are leached during pedogenesis. Slight deviations from the parent material for Al and Ti suggest minor additions of allochthonous clay- and titanium-rich dust to Protosols (cf. Reheis, 1990; Maynard, 1992; Mason and Jacobs, 1998).

parent loessite into the paleosol and typically exhibit moderately developed pedogenic structure and common root traces (;2 to 3 cm diameter and up to 0.5 m long). These paleosols range from 2 to 7 points on the macroscopic rating scale (Table DR2 [see footnote 1]). Petrographic observations indicate that these paleosols have a dolomite enrichment commonly coincidental with a clay enrichment. Within the carbonate-rich zone, disseminated dolomite ranges from microspar to spar (,10 to 40 mm) and typically reaches peak proportions at or near the preserved paleosol top (Figs. 3E, 4). Rarely, euhedral gypsum crystals with rhombic dolomite inclusions occur in these zones (Fig. 3F). The clay-enriched zone exhibits agglomeroplasmic to porphyroskelic fabric with mosepic to skelsepic ferriargillans (sepic clay; Fig. 3G). Crystic plasma microfabric in an intertextic fabric with common crystallaria are present locally. (Fig. 3H). Clay forms cutans that coat and bridge grains (cf. Walker et al., 1978; Matlack et al., 1989; Malicse and Mazzullo, 1996). In addition, channel argillans are preserved as elongate clay-rich segregations (Fig. 3I; cf. Brewer, 1976; Fitzpatrick, 1984). Geochemistry Geochemical trends for moderately developed paleosols are generally more pronounced than those exhibited by the Protosols (Figs. 4C to 4E). In paleosols with dolomite as the carbonate cement, Mg and Ca values peak immediately below or in conjunction with the K peak. Trends in Al, Ti, and K are similar to those in Protosols with dolomite as the carbonate cement. In moderately developed paleosols, Zr-normalized Al and Ti variations range from 1.6s to 4s relative to variation exhibited in average parent loessite, significantly above values in Protosols (Tables DR1, DR4b [see footnote 1]).

Moderately Developed Paleosols Macro- and Micromorphology Moderately developed paleosols display a gradational melanization (up to 0.5 m) from

Paleosol Interpretation and Classification Moderately developed paleosols display stronger pedogenesis than Protosols, as evidenced by a thicker melanized zone, more

strongly developed pedogenic structure, and common root traces. The significant range (2 to 7 points) in macroscopic rating suggests that even relatively well developed paleosols can exhibit weakly developed macroscopic features, perhaps as a result of postpedogenic truncation (cf. Retallack, 1981, 1988, 1990; Bronger and Heinkele, 1990; Mack et al., 1993). Moderately developed paleosols contain on average 10% more carbonate and clay than Protosols, consistent with increased pedogenesis. However, the occurrence of the dolomite in a discrete layer could reflect either a pedogenic or eolian origin (see discussion). Clay-enriched zones of moderately developed paleosols contain up to 40% clay in the form of grain bridges, grain argillans, and elongate clay segregations or channel argillans. This clay morphology suggests an illuvial origin for the clay (cf. Walker et al., 1978; Matlack et al., 1989; Malicse and Mazzullo, 1996). Presence of sepic clay and channel argillans further suggests stronger pedogenesis than that exhibited in the Protosols (Brewer, 1976; Retallack, 1990; Fitzpatrick 1993). On the basis of macro- and microscopic evidence for moderately intense pedogenesis characterized dominantly by clay-enrichment, these paleosols are interpreted as moderately developed Argillisols according to the classification of Mack et al. (1993). Major oxide trends of Si, Al, Fe, Mg, Ca, and K are similar to those exhibited in Protosols, but show greater deviation from parent material, reflecting an increased degree of pedogenesis. Zr-normalized major element trends are more pronounced than in Protosols; the Al, Fe, and Ti trends indicate greater additions of allochthonous dust to the moderately developed Argillisols. (Reheis, 1990; Maynard, 1992; Mason and Jacobs, 1998). Well-Developed Paleosols Macro- and Micromorphology Well-developed paleosols display melanization over a zone up to 1.0 m thick and have strong pedogenic structure, significant clay, local reduction haloes, and common root trac-

Figure 3. (Caption continued from p. 675.) (I) Euhedral gypsum crystals with dolomite inclusions. Width represents 3.6 mm; crosspolarized light. (J) Argillisol Bt horizon with sepic clay composing an appreciable amount of the fabric. Width represents 0.72 mm; cross-polarized light. (K) Crystallaria composed of dolomite highlighted by rectangle. Width represents 1.44 mm; cross-polarized light. (L) Elongate clay segregations (channel argillans) in Bt of Argillisol. Width represents 3.6 mm; cross-polarized light. (M) Outcrop view of well-developed Argillisol with pencil for scale (pencil circled in white and white line equal to pencil length). (N) Abundant disseminated dolomite in Argillisol Bt horizon. Width represents 1.44 mm; cross-polarized light. (O) Reduced zone with skeletal grains, clay, and dolomite cement surrounding micritic dolomite center. Width represents 3.6 mm; plane-polarized light. (P) Prismatic gypsum laths associated with the carbonate-rich zone of well-developed Argillisol. Width represents 1.44 mm; plane-polarized light.

N


677

TRAMP et al.

678



es (up to 0.5 m long and 2–3 cm diameter) that collectively justify paleosol ratings of 7 to 11 points (Table DR2 [see footnote 1]). These paleosols have a thick (up to 1.5 m) dolomite-rich zone (30%–80%), characterized typically by disseminated microsparitic dolomite and significant (up to 50%) clay (Fig. 3J). Micromorphologic features include rare dolomicritic micronodules surrounded by a concentric reduced zone (Fig. 3K), common channel ferriargillans, and rare clusters of small (0.25 mm long) euhedral gypsum crystals (Fig. 3L). Minor amounts of illite and subordinate kaolinite occur in the clay fraction. SEM analysis reveals clay coats around grains and minor amounts of needle-shaped (inferred authigenic) clay (Brewer, 1976; Fitzpatrick, 1984; Wright, 1991). Geochemistry Major oxide geochemical trends (Fig. 4F) for Si and Fe are similar to those of Protosols and moderately developed Argillisols, but show a greater degree of up-profile change, as evidenced by significant up-profile shifts in Al, K, Mg, and Ca. Variations in Ti and Al for well-developed paleosols greatly exceed (by 5s) the variations exhibited in the parent material (Tables DR1, DR4c [see footnote 1]). Paleosol Interpretation and Classification In the field, well-developed paleosols display a thick (up to 1 m) zone of melanization, intense ped development, common root traces, and a clay-enriched horizon. Although dolomite concentration increases markedly upprofile, it is typically disseminated and occurs as nodules only rarely, suggesting that it is likely eolian rather than pedogenically derived (see later discussion). SEM analysis indicates pervasive grain coats within clay-enriched zones, reflecting clay illuviation, but only minor authigenic clay. Well-developed paleosols in which the presence of an illuvial clay (Bt) horizon with aligned clay is the most prominent feature are interpreted as well-developed Argillisols (Mack et al., 1993). Major oxide geochemical trends in well-

developed Argillisols show greater deviation from parent-material values (Si, Al, Fe, Mg, Ca, K) relative to Protosols and moderately developed Argillisols, confirming the interpretation of more intense pedogenesis in the welldeveloped Argillisols. Maximum Al, K and Mg, Ca values in discrete clay- and carbonateenriched zones, respectively, support this interpretation. Zr-normalized major element trends are more pronounced than those exhibited in Protosols and moderately developed Argillisols, with the Al, Fe, and Ti trends indicating greater additions of allochthonous dust to well-developed Argillisols. Although moderately developed Argillisols and welldeveloped Argillisols belong to the same paleosol class and could be combined, petrographic and geochemical analyses distinguish the gradational nature of Argillisol development (Reheis, 1990; Maynard, 1992; Mason and Jacobs, 1998). Potential Controls on Clay and Carbonate Enrichment Formation of clay cutans and accumulation of argillic horizons typically occur by illuviation of clays originating from either weathering of unstable phases or allochthonous (eolian) input (Walker et al., 1978; Matlack et al., 1989; Retallack, 1990; Buurman et al., 1998). The Maroon Formation paleosols contain substantial labile phases (e.g., plagioclase), and XRD analysis indicates ,10% clay-mineral (illite, chlorite, kaolinite) content. Additionally, geochemical trends of Ti/Zr and Al/Zr confirm addition of allochthonous clay to the upper reaches of paleosol profiles. Accordingly, we infer that most of the clay-sized material in these paleosols was transported to the profile during pedogenesis; such allochthonous additions of eolian dust to arid-region soils is well recognized (Yaalon and Ganor, 1973; Rabenhorst et al., 1984; McFadden et al., 1987; Vine, 1987; Chadwick and Davis, 1990; Reheis and Kihl, 1995; Mason and Jacobs, 1998). Further, the Maroon Formation paleosols display clay coats, and SEM analy-

sis reveals predominantly illuviated clay within paleosols and detrital clay coating of grains in parent loessite, indicating that the clay is predominantly detrital and was translocated down-profile. Carbonate in a loessite-hosted paleosol can be of phreatic, pedogenic, or eolian origin. Phreatic carbonate typically precipitates within the most porous and permeable facies, consists of intergranular spar, and is not intimately associated with pedogenic features or horizons (e.g., Tandon and Narayan, 1981; Machette, 1985; El-Sayed et al., 1991; Wright, 1992; Pimental et al., 1996; Mack et al., 2000). The carbonate in the study section is disseminated microsparitic to sparitic dolomite, concentrated in a discrete zone at or near the top of a paleosol profile, and commonly covaries with clay content (Bt horizons). Further, the carbonate remains disseminated and rarely forms nodules, despite concentrations of up to 80%. These attributes are inconsistent with either phreatic or pedogenic carbonate (cf. Gile et al., 1966; Wright, 1991; Mack et al., 2000) and instead suggest the carbonate was derived in the same manner as the clay, via eolian influx. The regional setting of the evaporite Eagle basin would have provided abundant sources of carbonate, as well as the arid conditions that presumably precluded formation of pedogenic carbonate nodules (see later discussion). Rock Magnetism Magnetic Susceptibility Patterns Magnetic susceptibility values increase from parent loessite to paleosols and peak at or near paleosol tops (Figs. 4, 5, 6). To statistically compare the xb difference between loessite and paleosol, we subdivided the population on the basis of macroscopic descriptions that identified samples as either parent loessite or pedogenically modified loessite. Two subsets include parent loessite (n 5 523) and all pedogenically modified loessite (n 5 734). The sample set representing parent loessite (n 5 523) has a mean xb value of 3.51 3

Figure 4. Stratigraphic columns of Protosols and Argillisols with major oxide geochemical values (weight percent oxide normalized to ZrO2) and point-count data (volume percentage). Abbreviations: Qtz—quartz; Fld—feldspar; Dolo—dolomite; MS—magnetic susceptibility (in m3/kg). Protosols and moderately developed Argillisol profiles are scaled to the same range. The x-axis values are as follows: Qtz, 0 to 65%; Fld, 0 to 32%; Clay, 5 to 50%; Dolo, 0 to 80%; SiO2/ZrO2, 1080 to 3415; Al2O3/ZrO2, 90 to 275; FeO/ZrO2, 0 to 125; MgO/ZrO2, 0 to 130; CaO/ZrO2, 0 to 170; K2O/ZrO2, 30 to 110; TiO2/ZrO2, 0 to 15; and magnetic susceptibility, 2 to 12 m3/kg. Geochemical values for well-developed Argillisols are scaled separately to SiO2/ZrO2; Al2O3/ZrO2, 90 to 900; FeO/ZrO2, 20 to 350; MgO/ ZrO2, 0 to 225; CaO/ZrO2, 0 to 170; K2O/ZrO2, 30 to 320; TiO2/ZrO2, 0 to 50. (A and B) Protosols. (C, D, and E) Moderately developed Argillisols. (F) A well-developed Argillisol. Zones of increased carbonate (‘‘carb’’) and clay are labeled and shaded gray. Error bars are less than symbol size for geochemical values and magnetic susceptibility values. Point-count values greater than 10% are significant.

N


679

Figure 5. Magnetic susceptibility (x)of the studied section; bottom of section is on lower left, and top of section is on upper right. Magnetic susceptibility exhibits high values in paleosols and low values in parent-material loessite. Stars indicate paleosol profiles chosen for a detailed study. Note that chosen profiles span entire section. Large gray box indicates section described by Soreghan et al. (1997). Error bars are less than symbol size.

TRAMP et al.

680



Figure 6. Selected stratigraphic interval demonstrating that high magnetic susceptibility values correlate with paleosols and low magnetic susceptibility values correlate with loessite. Section begins on lower left and ends on upper right. Note section break near the middle. Boxed profiles also have been studied for major oxides, petrographically determined mineral contents and clay proportions, and micromorphology. 1028 m3/kg 6 1.59 3 1028 m3/kg, whereas the subset of all pedogenically modified loessite (n 5 734) has an average xb value of 5.70 3 1028 m3/kg 6 2.70 3 1028 m3/kg. Despite our oversimplified subdivisions, a statistical measure of the difference between two means from large (n $ 30) independent samples (analysis of variance) demonstrates that the loessite and paleosol subsets exhibit statistically separate xb values within a 99% confidence limit. On average, xb values are ;1.5 times greater in paleosols relative to loessite. Modern loess and paleosol xb variations exhibit similar statistical differences, albeit the absolute intensities can be up to an order of magnitude higher. For example, Heller and Liu (1984) reported means of 2.37 3 1027 m3/kg 6 0.95 3 1027 m3/kg for 159 paleosol samples and 1.00 3 1027 m3/kg 6 0.46 3 1027 m3/kg for 225 loess samples in their study of magnetic susceptibility in the Chinese Loess Plateau. The average xb value for parent loessite from the subset of nine selected paleosol profiles is 2.50 3 1028 m3/kg 6 0.57 3 1028 m3/kg. For all Protosols in this subset, the highest xb values range from 4 to 5 3 1028 m3/kg (n 5 4; Fig. 4), which is low relative to xb values from moderately developed Argillisols in the subset (5 to 11.5 3 1028 m3/kg; n 5 3; Fig. 5) and well-developed Argillisols in the subset (10 to 12 3 1028 m3/kg; n 5 2; Fig. 6). An average value of 5.70 3

1028 m3/kg 6 2.70 3 1028 m3/kg for all paleosols suggests that the majority are Protosols. Field and microscopic observations support this conclusion. In the study section, xb values do not sharply return to low (loessite) values immediately above paleosol tops. Commonly, loess immediately overlying a paleosol displays a slight red-brown hue and has low to moderate xb that gradually decreases by 10 cm above the subjacent top. We interpret this gradational change to represent a slightly weathered zone in the base of the superjacent loess, similar to several Quaternary loess-paleosol sequences, attributed to a combination of declining availability of moisture coupled with an increase in the rate of loess accumulation (e.g., Leonard, 1951; Thorp et al., 1951; Follmer, 1983; Kukla and An, 1989; Johnson and Feng, 1993; Kemp, 1995; Kemp et al., 1995). Magnetic Minerals and Paleomagnetism As previously described, xb increases from parent loessite into paleosol, peaking at or near paleosol tops (Figs. 4, 5, 6). A representative profile originally described in Soreghan et al. (1997) was selected for additional analysis to test the results from previous studies and determine the origin of the xb increase (Fig. 7). Both xp and xf increase from the loessite into the paleosol (Fig. 7). Normalized xf (Fig. 7) is also higher for the paleosol than the loessite, which is consistent with a greater

amount of ultra-fine-grained ferrimagnetic material in the paleosol (Hunt et al., 1995). Hysteresis curves for both paleosols and loessite samples are wasp-waisted (Soreghan et al., 1997), suggesting a mixture of both highand low-coercivity minerals, and/or a mixture of grain sizes. A more pronounced constriction of the waist in the paleosol samples (e.g., Soreghan et al., 1997) is consistent with a greater contribution of ultra-fine-grained magnetite in the paleosol than in the loessite. These results are consistent with the data from another Maroon Formation loessite-paleosol profile reported in Cogoini et al. (2000). Patterns of thermal demagnetization of an SIRM imparted at 19 K with a 2 T field were recorded for a well-developed Argillisol to test the conclusion of our previous studies (Soreghan et al., 1997; Cogoini et al., 2000) on the Maroon Formation that superparamagnetic (SP) magnetite is a major contributor to the xb signature. Thermal demagnetization of low-temperature SIRM for samples from this Maroon Formation Argillisol exhibit a greater relative drop in remanence from 50 to 300 K for the paleosol samples compared to the more loessitic samples (Fig. 8). This result suggests a greater abundance of ultra-fine-grained magnetic material, presumably magnetite, in paleosols relative to loessite. The presence of the Verwey transition at ;110 K in the low-temperature curves is indicative of mostly multidomain (md) magne-


681

TRAMP et al.

tite and is found in most loessite and paleosol samples. Paleomagnetic analysis (Appendix DR1 [see footnote 1]) indicates that alternating field (AF) demagnetization did not remove a linear component of magnetization. The magnitude of AF decay in paleosols (18%, standard deviation 5 13.5%) is similar although perhaps slightly higher than that in loessite (14%, standard deviation 5 13.9%). Thermal demagnetization removed a magnetization, interpreted as an early chemical remanent magnetization (CRM), with southeast declination and shallow inclination. The pole position for the CRM (458N, and 1178E) falls near the late Paleozoic part of the apparent polar wander path (Appendix DR1). Isothermal remanent magnetization (IRM) acquisition and triaxial thermal decay curves indicate that the remanence is dominated by hematite, although some magnetite and maghemite may be present (Appendix DR1). Integration of Magnetic Susceptibility with Sedimentology and Geochemistry To investigate pedogenesis as a possible source for the xb signature, sedimentological and geochemical data were compared with peak xb values of paleosols. Magnetic susceptibility vs. macroscopic paleosol rating shows a moderate positive relationship with an r2 value of 0.64 (Fig. 9A), whereas an r2 value of 0.81 implies a good correlation between paleosol peak xb values and pedogenic micromorphologic rating (Fig. 9B), indicating that xb closely tracks pedogenesis. Furthermore, cross-plotted geochemical and mineralogical data (Figs. 9C, 9D, 9E) demonstrate that xb positively correlates with Al and K (r2 5 0.78, 0.67 respectively) and negatively correlates with quartz (r2 5 0.60). The positive correlation of xb with geochemical proxies for clay (Al, K) suggests either a causal relationship between xb and clay minerals or a coincidental relationship wherein xb is associated with the clay (discussed further below). DISCUSSION Origin of the Magnetic Susceptibility Signature The higher normalized xf values in paleosols relative to loessite are consistent with the low-temperature x data, which suggest that fine superparamagnetic (SP) magnetite contributes to the xb values in paleosols (e.g., Hunt et al., 1995). The percentage of AF decay is apparently slightly higher in paleosols

682

Figure 7. A selected profile from the Maroon Formation showing three columns from left to right: (1) stratigraphic column; (2) bulk magnetic susceptibility xb (diamonds), paramagnetic susceptibility xp (squares), and ferrimagnetic susceptibility xf (triangles); Data for two of the samples were originally presented in Soreghan et al. (1997). Abbreviations: Qtz—quartz; Fld—feldspar; Dolo—dolomite; MS—magnetic susceptibility (in m3/kg).

Figure 8. Thermal demagnetization patterns of low-temperature SIRM for loessitic and paleosol samples. The SIRM were acquired at 19 K. Diamonds—well-developed paleosol; squares—loessite. The slight drop at about 110–120 K in both curves is the Verwey transition. relative to loessite, which could mean that there is more detrital (remanence-carrying) magnetite in the former. However, the percentages of AF decay are similar overall, and they are not statistically distinct between paleosol and loessite. In addition, the higher SIRM at room temperature for loessite relative to paleosol is not consistent with more remanence-carrying magnetite in paleosols compared to loessite. Accordingly, the evidence for more remanence-carrying magnetite in paleosols compared to parent loessite is equivocal. The small increase of xp from the

loessite into paleosol may be caused by clays, which occur in greater abundance in the paleosols relative to loessite, although clay content is everywhere less than 10%. Alternatively, the high-field x characteristics of hematite (Collinson, 1983) could indicate that hematite constitutes a part of the overall increase in xp from loessite into paleosol. The amount, however, is difficult to determine. Our study demonstrates a positive correlation of xb with clay-sized material K and Al bearing minerals, suggesting the magnetite resides mainly in clay-rich zones. Potential in-



Figure 9. Magnetic susceptibility compared to (A) sum of macroscopic paleosol rating (color, pedogenic structure, texture, mottling, and traces), (B) sum of pedogenic micromorphologic features (texture, sepic clay, channel argillans, dolomite nodules, and dolomicrite), (C) mineral proportion (quartz), and (D, E) major oxide weight percents (Al, K) cross-plotted with magnetic susceptibility. Major oxide weight percents were obtained by XRF analysis and mineral proportions were obtained by point counting. terpretations for the xb signature include (1) formation of ferrimagnetic minerals during clay transformations accompanying burial diagenesis, (2) detrital contribution of ferrimagnetic minerals during pedogenesis, and/or (3) in situ pedogenic production and concentration of ultra-fine-grained magnetite. Several studies have documented a connection between clay diagenesis caused by burial processes (Katz et al., 1998, 2000) or fluids (Woods et al., 2000, 2002) and magnetite authigenesis with an associated CRM. The increase in clay and the trends in rock-magnetic parameters (relative SP content, etc.) between

loessite and paleosols are potentially consistent with this model. Analyses of Nuccio et al. (1989) indicate that burial of the Maroon Formation was minimal (,1 km) in the late Paleozoic but accelerated so that the formation reached significant depth and temperature (5 km and .120 8C) in the Cretaceous, sufficient for illitization of original clay minerals (e.g., Curtis, 1985; Retallack, 1990, 1991; Wright, 1992). Although a late Paleozoic CRM that resides in hematite is present in our samples, AF demagnetization does not remove a linear component of magnetization of any age. A low-coercivity CRM residing in magnetite is,

therefore, not present, and this fact suggests that the clay-diagenesis model is not applicable to our data. The age and the hematite constituent of the CRM that is present is also inconsistent with the burial diagenetic model. Despite these arguments, we cannot eliminate the possibility that some SP magnetite was created by clay diagenesis. Other studies have suggested a detrital origin for the xb signature. Reynolds et al. (2001), for example, showed that the incorporation of detrital magnetite with surficial deposits of arid lands produced moderately high values of xb. If the origin of the magnetite in


683

TRAMP et al.

Maroon Formation paleosols is detrital, then xb should track evidence for dust additions (i.e., trends in Zr-normalized Al and Ti; see previous discussion). Crossplots of xb with proxies for dust additions exhibit relatively weak correlations; however, r2 values are 0.38 for Al/Zr and 0.56 for Ti/Zr. These facts, together with the equivocal evidence for any significant difference in amount of remanencecarrying magnetite between loessite and paleosols, suggest that an additional mechanism is clearly needed to explain the xb signature. Superparamagnetic (SP) magnetite forms pedogenically in Quaternary soils worldwide (Maher, 1986; Maher and Taylor, 1988; Maher and Thompson, 1991; Maher et al., 1994), and its concentration commonly peaks in clay-rich zones (e.g., Maher, 1998; Grimley et al., 1998). Clay enrichment and magnetite production appear to occur as intrinsic pedogenic processes (Maher and Taylor, 1988; Maher and Thompson, 1991). We therefore infer that pedogenic formation of SP magnetite likely explains most (.50%) of the magnetic susceptibility signature recorded in the Maroon Formation loessite. We cannot exclude the possibility that some SP hematite may also contribute to the magnetic susceptibility signature. Pedogenic Enhancement of Magnetic Susceptibility: A Climate or Time Control? Our analyses indicate that magnetic susceptibility tracks pedogenesis in the Maroon Formation loessite, which is analogous to the findings of Soreghan et al. (2002a) for the Pennsylvanian–Permian loessite of the Lower Cutler beds (Utah). Given the homogeneous parent material and the unlikelihood of systematic variation in vegetation and topography, time and climate are the key controls on pedogenesis in this system. Paleosols can achieve different levels of development (e.g., Protosol vs. Argillisol) given constant climatic conditions but variable durations of pedogenesis or differing climatic conditions but similar time. A key issue is the relative contribution of time and climate on pedogenesis in this system. If we assume relatively constant rates of dust (clay and carbonate) influx during pedogenesis, the allochthonous dust content of the Maroon Formation paleosols provides a semiquantitative measure of pedogenic duration. A greater incorporation of allochthonous dust corresponds to longer durations of pedogenesis. Advanced pedogenic development (e.g., Argillisols compared to Protosols) is reflected partially by the greater deviation up-

684

profile of Zr-normalized Al and Ti values and is consistent with development over a longer period of time. For Cenozoic paleosol sequences, Maher (1986) noted that semiarid soils characterized by wet and dry alterations undergo optimal conditions for pedogenic production of ferrimagnetic magnetite. Several lines of evidence suggest a relatively arid climate for deposition of the Maroon Formation loess. Johnson (1987b) and Johnson et al. (1988) inferred that the Maroon Formation loess accumulated downwind of extensive eolian sands that formed as a result of the regionally arid climate of the evaporitic Eagle basin. Further, several attributes of the loessite confirm relatively arid conditions, e.g., (1) relatively shallow depths (10 to 30 cm) to carbonateenriched zones and lack of nodule development (Gile et al., 1966; Machette, 1985; Retallack, 1994) indicating little remobilization and leaching of carbonate down-profile, (2) gypsum locally associated with carbonateenriched zones (Fitzpatrick, 1984, 1993; Wright, 1991), and (3) deeply penetrating root traces that suggest well-drained conditions (Retallack, 1988). The lack of carbonate nodules is particularly noteworthy given the abundant carbonate content within many of the paleosols; the absence of nodules, together with the rare gypsum, suggests mean annual precipitation (MAP) values insufficient to remobilize the available carbonate, likely as minimal as 300 cm (Retallack, 1991; Birkeland, 1999). At first glance, such low MAP values seem inconsistent with the occurrence of Argillisols in the Maroon Formation loessite. Typically, formation of Argillisols and their associated illuvial clay (Bt) horizons is linked to relatively wet climates (e.g., Mack and James, 1994). Bt horizons, however, do develop in arid climates as a result of atmospheric dust influx (Birkeland, 1999), and we infer that the Maroon Formation paleosols reflect relatively arid conditions, albeit conditions slightly wetter than those that prevailed during times of loess accumulation. Within many Cenozoic loess sequences, the xb signature not only distinguishes paleosols from parent loess but also acts as a paleoclimate proxy owing to pedogenic concentration of fine-grained magnetite during relatively humid periods (e.g., Kukla and An, 1989; Beget et al., 1990; Wang et al., 1990; An et al., 1991; Feng et al., 1994; Maher, 1998). Similar to Cenozoic loess sections, magnetic susceptibility in the Maroon Formation loessite accurately distinguishes interbedded paleosols from parent loessite, quantitatively tracks character and degree of pedogenesis, and therefore reflects both cli-

mate and duration of soil-forming intervals. We infer that the meter-scale Maroon Formation paleosols and associated xb fluctuations record relatively high-frequency (i.e., 1042105 yr), glacial-interglacial fluctuations that operated in western equatorial Pangea during the waning stages of the late Paleozoic icehouse, possibly analogous to the Pliocene–Pleistocene paleosol sequences of the Chinese Loess Plateau and elsewhere. CONCLUSIONS Paleosols within the Maroon Formation loessite (latest Pennsylvanian–Early Permian) include Protosols and Argillisols. The xb signature throughout the Maroon Formation loessite consistently exhibits low values in loessite and significantly higher values (1.5 times greater on average) in paleosols. Magnetic susceptibility not only distinguishes loessite from paleosol, but also tracks degree of pedogenesis such that the Argillisols exhibit the highest xb values. We infer that the xb signature is primarily attributable to climate and time-dependent pedogenic production and concentration of ultra-fine-grained magnetite. The Maroon Formation loess and associated soils accumulated in an overall arid system, documented in part by formation of Argillisols attributable to eolian clay and carbonate influx. However, the paleosols showing bulk magnetic susceptibility values of greater than 200 xb document a high-frequency (1042105 yr) fluctuation between arid times of loess accumulation and slightly wetter times of pedogenesis. We link this fluctuation to glacialinterglacial climate shifts that operated in low-latitude Pangea during waning stages of the late Paleozoic icehouse. These results indicate that climate-related magnetic susceptibility variations within loess successions can be preserved and are useful in the study of very ancient (pre–Pliocene–Pleistocene) sequences. ACKNOWLEDGMENTS This paper forms part of K. Tramp’s M.S. thesis, funded in part by the National Science Foundation (EAR-9805130 and EAR-0001052), and supervised by G.S. Soreghan and R.D. Elmore. We thank GSA Bulletin reviewers J. Quade, M. Jackson, E. Bestland, J. Geissman, and R. Molina-Garza for constructive comments and suggestions on an earlier version of the manuscript and J. Mason, M. Reheis, G. Mack, and G. Retallack for helpful advice on various aspects of the study. Analytical and technical support at the University of Oklahoma was provided by G. Morgan, R. Conlon, B. Weaver, D. Simpson, G. Miller, R. Furley, and S. Seals. We thank M. Cogoini for preliminary field and laboratory analyses, draft suggestions, and paleomagne-


PALEOCLIMATIC INFERENCES, PERMIAN MAROON FORMATION LOESSITE, COLORADO, USA tism technique instruction, C. Eble (Kentucky Geological Survey) for palynological processing and interpretation, M. Blum (Louisiana State University) for insightful comments, and S. Johnson for providing discussions on the Maroon Formation loessite. We appreciate the efforts of M. Hamilton (Geological Survey of Canada), P. Link, and M. Fanning (Australian National University) for processing results of preliminary detrital zircon data.

REFERENCES CITED An, Z., Kukla, G.J., Porter, S.C., and Xiao, J., 1991, Magnetic susceptibility evidence of monsoon variation on the Loess Plateau of central China during the last 130,000 years: Quaternary Research, v. 36, p. 29–36. Beget, J.E., Stone, D.B., and Hawkins, D.B., 1990, Paleoclimatic forcing of magnetic susceptibility variations in Alaskan loess during the late Quaternary: Geology, v. 18, p. 40–43. Bilzi, A.F., and Ciolkosz, E.J., 1977, A field morphology rating scale for evaluating pedological development: Soil Science, v. 124, p. 45–48. Birkeland, P.W., 1999, Soils and geomorphology: New York, Oxford University Press, 430 p. Bown, T.M., and Kraus, M.J., 1987, Integration of channel and floodplain suites: I. Developmental sequence and lateral relations of alluvial paleosols: Journal of Sedimentary Petrology, v. 57, p. 587–601. Brewer, R., 1976, The fabric and mineral analysis of soils: New York, Krieger Publications, 482 p. Bronger, A., and Heinkele, Th., 1990, Mineralogical and clay mineralogical aspects of loess research: Quaternary International, v. 7/8, p. 37–51. Bullock, P., Fedoroff, N., Johgerius, A., Stoops, G., and Tursina, T., 1985, Handbook for soil thin section description: Albrighton, UK, Waine Research Publications, 152 p. Buurman, P., Jongmans, A.G., and PiPujol, M.D., 1998, Clay illuviation and mechanical clay infiltration—Is there a difference, in Follmer, L.R., Johnson, D.L., and Catt, J.A., eds., Revisitation of concepts in paleopedology, Transactions of the Second International Symposium on Paleopedology: Oxford, Pergamon, v. 51–52, p. 66–69. Cecil, C.B., 1990, Paleoclimate controls on stratigraphic repetition of chemical and siliciclastic rocks: Geology, v. 18, p. 533–536. Chadwick, O.A., and Davis, J.O., 1990, Soil-forming intervals caused by eolian sediment pulses in the Lahontan basin, northwestern Nevada: Geology, v. 18, p. 243–246. Cogoini, M., Elmore, R.D., Soreghan, G.S., and Lewchuk, M., 2000, Contrasting rock-magnetic characteristics of two upper Paleozoic loessite/paleosol profiles: Physics and Chemistry of the Earth, v. 26, p. 905–910. Collinson, D.W., 1983, Methods in paleomagnetism and rock magnetism: London, Chapman and Hall, 500 p. Crowell, J.C., 1978, Gondwanan glaciation, cyclothems, continental positioning, and climate change: American Journal of Science, v. 278, p. 1345–1372. Curtis, C.D., 1985, Clay mineral precipitation and transformation during burial diagenesis: Royal Society of London Philosophical Transactions, Ser. A, v. 315, p. 91–105. DeVoto, R.H., 1972, Pennsylvanian and Permian stratigraphy and tectonism in central Colorado: Colorado School of Mines Quarterly, v. 67, p. 139–185. DeVoto, R.H., 1980, Pennsylvanian stratigraphy and history of Colorado, in Kent, H.C., and Porter, K.W., eds., Colorado geology: Denver, Colorado, Rocky Mountain Association of Geologists, p. 71–101. DeVoto, R., Bartleson, B.L., Schenk, C.J., and Waechter, N.B., 1986, Late Paleozoic stratigraphy and syndepositional tectonism, northwestern Colorado, in Stone, D.S., ed., New interpretations of northwest Colorado geology: Denver, Colorado, Rocky Mountain Association of Geologists, p. 37–49. Dickson, J.A.D., 1966, Carbonate identification and genesis

as revealed by staining: Journal of Sedimentary Petrology, v. 36, p. 491–505. Driese, S.G., and Dott, R.H., Jr., 1984, Model for sandstonecarbonate ‘‘cyclothems’’ based on upper member of Morgan Formation (Middle Pennsylvanian) of northern Utah and Colorado: American Association of Petroleum Geologists Bulletin, v. 68, p. 574–597. El-Sayed, M.I., Fairchild, I.J., and Spiro, B., 1991, Kuwaiti dolocrete: Petrology, geochemistry and groundwater origin: Sedimentary Geology, v. 73, p. 59–75. Evans, M.E., and Heller, F., 1994, Magnetic enhancement and palaeoclimate: Study of a loess/palaeosol couplet across the Loess Plateau of China: Geophysical Journal International, v. 117, p. 257–264. Feng, Z.D., 1997, Geochemical characteristics of a loesssoil sequence in central Kansas: Soil Science Society of America Journal, v. 61, p. 531–541. Feng, Z.D., Johnson, W.C., Lu, Y., and Ward, P.A., 1994, Climatic signals from loess-soil sequences in the central Great Plains, USA: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 110, p. 345–358. Fitzpatrick, E.A., 1993, Soil microscopy and micromorphology: New York, John Wiley and Sons, 304 p. Fitzpatrick, E.A., 1984, Micromorphology of soils: London, Chapman and Hall, 433 p. Folk, R.L., 1962, Spectral subdivision of limestone types, in Ham, W.E., ed., Classification of carbonate rocks: American Association of Petroleum Geologists Memoir 1, p. 62–84. Follmer, L.R., 1983, Sangamon and Wisconsin pedogenesis in the Midwestern United States, in Wright, H.E., Jr., ed., Late-Quaternary environments of the United States: Minneapolis, University of Minnesota Press, v. 1, p. 138–144. Gile, L.H., Peterseon, F.F., and Grossman, R.B., 1966, Morphological and genetic sequence of carbonate accumulation in desert soils: Soil Science, v. 101, p. 347–360. Grimley, D.A., Follmer, L.R., and McKay, E.D., 1998, Magnetic susceptibility and mineral zonations controlled by provenance in loess along the Illinois and central Mississippi river valleys: Quaternary Research, v. 49, p. 24–36. Harden, J.W., 1982, A quantitative index of soil development from field descriptions: Examples from a chronosequence in central California: Geoderma, v. 28, p. 1–28. Heckel, P.H., 1986, Sea-level curve for Pennsylvanian eustatic marine transgressive-regressive depositional cycles along Midcontinent outcrop belt, North America: Geology, v. 14, p. 330–334. Heller, F., and Liu, T.S., 1984, Magnetism of Chinese loess deposits: Geophysical Journal, v. 77, p. 125–141. Hunt, C.P., Banerjee, S.K., Han, J., Solheid, P.A., Oches, E., Sun, W., and Liu, T.S., 1995, Rock-magnetic proxies of climate change in the loess-palaeosol sequences of the western Loess Plateau of China: Geophysical Journal International, v. 123, p. 232–244. Johnson, S.Y., 1987a, Stratigraphic and sedimentologic studies of late Paleozoic strata in the Eagle basin and northern Aspen subbasin, northwest Colorado: U.S. Geological Survey Open-File Report 87–286, 82 p. Johnson, S.Y., 1987b, Sedimentology and paleogeographic significance of six fluvial sandstone bodies in the Maroon Formation, Eagle basin, northwest Colorado: U.S. Geological Survey Bulletin 1787-A, 18 p. Johnson, S.Y., 1989, Significance of loessite in the Maroon Formation (Middle Pennsylvanian to Lower Permian), Eagle basin, Northwest Colorado: Journal of Sedimentary Petrology, v. 59, p. 782–791. Johnson, W.C., and Feng, Z., 1993, Barton County Sanitary Landfill, 3.5 miles north and 3.5 miles east of Great Bend, Kansas: Soils developed in loess, in Johnson, W.C., ed., Second International Paleopedology Symposium: Kansas Geological Survey Open-File Report 93–30, p. 8–1–8–22. Johnson, S.Y., Schenk, C.J., and Karachewski, J.A., 1988, Pennsylvanian and Permian depositional systems and cycles in the Eagle basin, northwest Colorado, in Holden, G.S., ed., Geological Society of America Field trip Guidebook: Colorado School of Mines Professional Contribution 12, p. 156–175.

Johnson, S.Y., Chan, M.A., and Konopka, E.A., 1992, Pennsylvanian and Early Permian paleogeography of the Uinta-Piceance basin region, northwestern Colorado and northeastern Utah: U.S. Geological Survey Bulletin 1787-CC, 35 p. Katz, B., Elmore, R.D., Cogoini, M., and Ferry, S., 1998, Widespread chemical remagnetization: Orogenic fluids or burial diagenesis of clays?: Geology, v. 26, p. 603–606. Katz, B., Elmore, R.D., Cogoini, M., Engel, M.H., and Ferry, S., 2000, Associations between burial diagenesis of smectite, chemical remagnetization, and magnetite authigenesis in the Vocontian trough, SE France: Journal of Geophysical Research, v. 105, p. 851–868. Kemp, R.A., 1995, Role of micromorphology in paleopedological research: Quaternary International, v. 51/52, p. 133–141. Kemp, R.A., Derbyshire, E., Xingmin, M., Fahu, C., and Baotian, P., 1995, Pedosedimentary reconstruction of a thick loess-paleosol sequence near Lanzhou in north-central China: Quaternary Research, v. 43, p. 30–45. Kluth, C.F., and Coney, P.J., 1981, Plate tectonics of the ancestral Rocky Mountains: Geology, v. 9, p. 10–15. Kukla, G., and An, Z., 1989, Loess stratigraphy in central China: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 72, p. 203–225. Kukla, G., Heller, F., Liu, X.M., Xu, T.C., Liu, T.S., and An, Z.S., 1988, Pleistocene climates in China dated by magnetic susceptibility: Geology, v. 16, p. 811–814. Leonard, A.B., 1951, Stratigraphic zonation of the Peoria Loess in Kansas: Journal of Geology, v. 59, p. 323–332. Liu, X.M., Shaw, J., Liu, T., Heller, F., and Yuan, B., 1992, Magnetic mineralogy of Chinese loess and its significance: Geophysical Journal International, v. 108, p. 301–308. Machette, M.N., 1985, Calcic soils of the southwestern United States, in Weide, D.L., ed., Soils and Quaternary geomorphology of the southwestern United States: Geological Society of America Special Paper 203, p. 1–21. Mack, G.H., 1997, Paleosols for sedimentologists (2nd edition): Geological Society of America Short Course Notes, 114 p. Mack, G.H., and James, W.C., 1994, Paleoclimate and the global distribution of paleosols: Journal of Geology, v. 102, p. 360–366. Mack, G.H., James, W.C., and Monger, H.C., 1993, Classification of paleosols: Geological Society of America Bulletin, v. 105, p. 129–136. Mack, G.H., Cole, D.R., and Trevino, L., 2000, The distribution and discrimination of shallow, authigenic carbonate in the Pliocene–Pleistocene Palomas basin, southern Rio Grande rift: Geological Society of America Bulletin, v. 112, p. 643–656. Maher, B.A., 1986, Characterisation of soils by mineral magnetic measurements: Physics of the Earth and Planetary Interiors, v. 42, p. 76–79. Maher, B.A., 1998, Magnetic properties of modern soils and Quaternary loessic plaeosols: Paleoclimate implications: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 137, p. 25–54. Maher, B., and Taylor, R.M., 1988, Formation of ultrafine-grained magnetite in soils: Nature, v. 336, p. 368–371. Maher, B., and Thompson, R., 1991, Mineral magnetic record of the Chinese loess and paleosols: Geology, v. 19, p. 3–6. Maher, B.A., Thompson, R., and Zhou, L.P., 1994, Spatial and temporal reconstructions of changes in the Asian palaeomonsoon: A new mineral magnetic approach: Earth and Planetary Science Letters, v. 125, p. 461–471. Malicse, A., and Mazzullo, J., 1996, Early diagenesis and paleosol features of ancient desert sediments: Examples from the Permian basin, in Crossey, L.J., Loucks, R., and Totten, M.W., eds., Siliciclastic diagenesis and fluid flow: Concepts and applications: SEPM (Society for Sedimentary Geology) Special Publication 55, p. 151–161.


685

TRAMP et al. Mallory, W.W., 1972, Regional synthesis of the Pennsylvanian System, in Mallory, W.E., ed., Geologic atlas of the Rocky Mountain region: Denver, Colorado, Rocky Mountain Association of Geologists, p. 111–127. Mallory, W.W., 1960, Outline of Pennsylvanian stratigraphy of Colorado, in Guide to geology of Colorado: Denver, Rocky Mountain Association Geologists, p. 23–33. Mason, J.A., and Jacobs, P.M., 1998, Chemical and particlesize evidence for addition of fine dust to soils of the Midwestern United States: Geology, v. 26, p. 1135–1138. Matlack, K.S., Houseknecht, D.W., and Applin, K.R., 1989, Emplacement of clay into sand by infiltration: Journal of Sedimentary Petrology, v. 59, p. 77–87. Maynard, J.B., 1992, Chemistry of modern soils as a guide to interpreting Precambrian paleosols: Journal of Geology, v. 100, p. 279–289. McFadden, L.D., Wells, S.G., and Jercinovich, M.J., 1987, Influences of eolian and pedogenic processes on the origin and evolution of desert pavement: Geology, v. 15, p. 504–508. Miller, K.B., McCahon, T.J., and West, R.R., 1996, Lower Permian (Wolfcampian) paleosol-bearing cycles of the U.S. Midcontinent: Evidence of climatic cyclicity: Journal of Sedimentary Research, v. 66, p. 71–84. Nuccio, V.F., Johnson, S.Y., and Schenk, C.J., 1989, Paleogeothermal gradient and timing of oil generation in the Belden Formation, Eagle basin, northwestern Colorado: The Mountain Geologist, v. 26, p. 31–41. Pimental, N.L., Wright, V.P., and Azevedo, T.M., 1996, Distinguishing early groundwater alteration effects from pedogenesis in ancient alluvial basins: Examples from the Palaeogene of southern Portugal: Sedimentary Geology, v. 105, p. 1–10. Rabenhorst, M.C., Wilding, L.P., and Girdner, C.L., 1984, Airborne dusts in the Edwards Plateau region of Texas: Soil Science Society of America Journal, v. 48, p. 621–627. Rankey, E.C., 1997, Relations between relative changes in sea level and climatic shifts: Pennsylvanian–Permian mixed carbonate-siliciclastic strata, western United States: Geological Society of America Bulletin, v. 109, p. 1089–1100. Reheis, M.C., 1990, Influence of climate and eolian dust on the major-element chemistry and clay mineralogy of soils in the northern Bighorn basin, USA: Catena, v. 17, p. 219–248. Reheis, M.C., and Kihl, R., 1995, Dust deposition in southern Nevada and California, 1984–1989: Relations to climate, source area, and source lithology: Journal of Geophysical Research, v. 100, p. 8893–8919. Reheis, M., Goodmacher, J.C., Harden, J.W., McFadden, L.D., Rockwell, T.R., Shroba, R.R., Sowers, J.M., and Taylor, E.M., 1995, Quaternary soils and dust deposition in southern Nevada and California: Geological Society of America Bulletin, v. 107, p. 1003–1022. Retallack, G.J., 1981, Fossil soils: Indicators of ancient terrestrial environments, in Niklas, K.J., ed., Paleobota-

686

ny, paleoecology and evolution: New York, Praeger Publishers, p. 55–102. Retallack, G.J., 1988, Field recognition of paleosols, in Reinhardt, J., and Sigleo, W.R., eds., Paleosols and weathering through geologic time: Principles and applications: Geological Society of America Special Paper 216, p. 1–20. Retallack, G.J., 1990, Soils of the past—An introduction to paleopedology: London, Unwin-Hyman, 520 p. Retallack, G.J., 1991, Untangling the effects of burial alteration and ancient soil formation: Annual Review of Earth and Planetary Sciences, v. 19, p. 183–206. Retallack, G.J., 1993, A coulour guide to paleosols: Chichester, UK, John Wiley and Sons, 175 p. Retallack, G.J., 1994, The environmental factor approach to the interpretation of paleosols, in Luxmoore, R.J., ed., Factors of soil formation: A fiftieth anniversary retrospective: Soil Science Society of America Special Publication 33, p. 31–64. Reynolds, R., Belnap, J., Reheis, M., Lamothe, P., and Luiszer, F., 2001, Aeolian dust in Colorado Plateau soils: Nutrient inputs and recent change in source: Proceedings of the National Academy of Sciences, v. 98, p. 7123–7127. Robb, A.J., III, 1992, Rain-impact microtopography (RIM): An experimental analogue for fossil examples from the Maroon Formation, Colorado: Journal of Sedimentary Petrology, v. 62, p. 530–535. Singer, M.J., Verosub, K.L., Fine, P., and TenPas, J., 1996, A conceptual model for the enhancement of magnetic susceptibility in soils: Quaternary International, v. 34–36, p. 243–248. Soreghan, G.S., Elmore, R.D., Katz, B., Cogoini, M., and Banerjee, S., 1997, Pedogenically enhanced magnetic susceptibility variations preserved in Paleozoic loessite: Geology, v. 25, p. 1003–1006. Soreghan, G.S., Elmore, R.D., and Lewchuk, M.T., 2002a, Sedimentologic-magnetic record of western Pangean climate in upper Paleozoic loessite (lower Cutler beds, Utah): Geological Society of America Bulletin, v. 114, p. 1019–1035. Soreghan, M.J., Soreghan, G.S., and Hamilton, M.A., 2002b, Paleowinds inferred from detrital-zircon geochronology of upper Paleozoic loessite, western equatorial Pangea: Geology, v. 30, p. 695–698. Tandon, S.K., and Narayan, D., 1981, Calcrete conglomerate, case-hardened conglomerate, and cornstone— Comparative account of pedogenic and non-pedogenic carbonates from the continental Siwalik Group, Punjab, India: Sedimentology, v. 28, p. 353–367. Thorp, J., Johnson, W.M., and Reed, E.C., 1951, Some post-Pliocene buried soils of central United States: Journal of Soil Science, v. 2, p. 1–19. Tweto, O., and Lovering, T.S., 1977, Geology of the Minturn 159 quadrangle, Eagle and Summit Counties, Colorado: U.S. Geological Survey Professional Paper 956, 96 p. Verosub, K.L., Fine, P., Singer, M.J., and TenPas, J., 1993,

Pedogenesis and paleoclimate: Interpretation of the magnetic susceptibility record of Chinese loesspaleosol sequences: Geology, v. 21, p. 1011–1014. Vine, H., 1987, Wind-blown materials and West African soils: An explanation of the ‘‘ferallitic soil over loose sandy sediments’’ profile, in Frostnik, L., and Reid, I., eds., Desert sediments: Ancient and modern: Geological Society [London] Special Publication 35, p. 171–183. Walker, T.R., 1967, Formation of red beds in modern and ancient deserts: Geological Society of America Bulletin, v. 78, p. 353–368. Walker, T.R., Waugh, B., and Grone, A.J., 1978, Diagenesis in first-cycle desert alluvium of Cenozoic age, southwestern United States and northwestern Mexico: Geological Society of America Bulletin, v. 89, p. 19–32. Wang, Y., Evans, M.E., Rutter, N., and Ding, Z., 1990, Magnetic susceptibility of Chinese loess and its bearing on paleoclimate, Geophysical Research Letters, v. 17, p. 2449–2451. Wilson, M.D., 1992, Inherited grain-rimming clays in sandstones from eolian and shelf environments: Their origin and control on reservoir properties, in Houseknecht, D.W., and Pittman, E.D., eds., Origin, diagenesis and petrophysics of clay minerals in sandstones: SEPM (Society for Sedimentary Geology) Special Publication 47, p. 209–225. Woods, S., Elmore, R.D., and Engel, M., 2000, The occurrence of pervasive chemical remanent magnetizations in sedimentary basins: Implications for dating burial diagenetic events: Journal of Geochemical Exploration, v. 67–70, p. 381–386. Woods, S., Elmore, R.D., and Engel, M., 2002, Paleomagnetic dating of the smectite to illite conversion: Testing the hypothesis in Jurassic sedimentary rocks, Skye, Scotland: BiSolid Earth and Planets, v. 107, no. 5, 12 p. Wright, V.P., 1991, Calcretes: An introduction, in Wright, V.P., and Tucker, M.E., eds., Calcretes: International Association of Sedimentologists, Reprinted Series, v. 2, p. 1–22. Wright, V.P., 1992, Paleosol recognition: A guide to early diagenesis in terrestrial settings, in Wolf, K.H., and Chilingarian, G.V., eds., Diagenesis III: Amsterdam, Elsevier, p. 591–619. Wright, V.P., and Vanstone, S.D., 2001, Onset of Palaeozoic glacio-eustasy and the evolving climates of low latitude areas: A synthesis of current understanding: Geological Society [London] Journal, v. 158, p. 579–582. Yaalon, D.H., and Ganor, E., 1973, The influence of dust on soils during the Quaternary: Soil Science, v. 116, p. 146–155. Zheng, H., Oldfield, F., Wintle, A.G., Robinson, S.G., and Wang, J.T., 1990, Partly pedogenic origin of magnetic variations in Chinese loess: Nature, v. 346, p. 737–739. MANUSCRIPT RECEIVED BY THE SOCIETY 17 MARCH 2003 REVISED MANUSCRIPT RECEIVED 19 AUGUST 2003 MANUSCRIPT ACCEPTED 13 SEPTEMBER 2003 Printed in the USA
