SSSAJ Grow Your Career with SSSA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (18)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Muhs, D. R.
Right arrow Articles by McGeehin, J. P.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Muhs, D. R.
Right arrow Articles by McGeehin, J. P.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Muhs, D. R.
Right arrow Articles by McGeehin, J. P.
Soil Science Society of America Journal 65:1761-1777 (2001)
© 2001 Soil Science Society of America

DIVISION S-5 - PEDOLOGY

Impact of Climate and Parent Material on Chemical Weathering in Loess-derived Soils of the Mississippi River Valley

D. R. Muhs*,a, E. A. Bettis, IIIb, J. Beena and J. P. McGeehinc

a U.S. Geological Survey, MS 980, Box 25046, Federal Center, Denver, CO 80225
b Dep. of Geoscience, 121 Trowbridge Hall, The Univ. of Iowa, Iowa City, IA 52242
c U.S. Geological Survey, MS 955, National Center, Reston, VA 20192

* Corresponding author (dmuhs{at}usgs.gov)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Peoria Loess-derived soils on uplands east of the Mississippi River valley were studied from Louisiana to Iowa, along a south-to-north gradient of decreasing precipitation and temperature. Major element analyses of deep loess in Mississippi and Illinois show that the composition of the parent material is similar in the northern and southern parts of the valley. We hypothesized that in the warmer, wetter parts of the transect, mineral weathering should be greater than in the cooler, drier parts of the transect. Profile average values of CaO/TiO2, MgO/TiO2, K2O/TiO2 and Na2O/TiO2, Sr/Zr, Ba/Zr, and Rb/Zr represent proxies for depletion of loess minerals such as calcite, dolomite, hornblende, mica, and plagioclase. All ratios show increases from south to north, supporting the hypothesis of greater chemical weathering in the southern part of the valley. An unexpected result is that profile average values of Al2O3/TiO2 and Fe2O3/TiO2 (proxies for the relative abundance of clay minerals) show increases from south to north. This finding, while contrary to the evidence of greater chemical weathering in the southern part of the transect, is consistent with an earlier study which showed higher clay contents in Bt horizons of loess-derived soils in the northern part of the transect. We hypothesize that soils in the northern part of the valley received fine-grained loess from sources to the west of the Mississippi River valley either late in the last glacial period, during the Holocene or both. In contrast, soils in the southern part of the valley were unaffected by such additions.

Abbreviations: AGCM, atmosphere general circulation models


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
ONE OF THE MOST pressing issues in society today is the possibility of human-induced climate change. Addition of greenhouse gases to the atmosphere may result in global warming, based on atmospheric general circulation models (AGCMs) (Hansen et al., 1988). However, because AGCMs produce simplified forecasts of future climates, it is imperative that they undergo validation tests. An effective method for validation is to compare AGCM models of past climates with geologic data on past climates, particularly warm periods. A commonly studied, recent warm-climate interval is the last, or Sangamon, interglacial period that had its peak warming ~125000 yr ago (Muhs and Szabo, 1994). The pedologic record potentially contains paleoclimatic information about the last interglacial period in the form of the Sangamon paleosol or its stratigraphic equivalent (e.g., the Eemian paleosol in Europe). There have been attempts to infer a longer, warmer, or wetter last interglacial period from studies of this soil or its equivalent in the midcontinent of North America (Ruhe, 1969; Hall and Anderson, 2000), Alaska (Muhs et al., 2001), and China (Maher et al., 1994). Critical to paleoclimatic interpretations, using paleosols, are reliable climofunctions for modern soils, which give an understanding of soil formation as a function of climate, all other factors being equal.

The importance of climate as a soil-forming factor has been appreciated by pedologists for more than 100 yr, starting with Dokuchaev (1883) and emphasized by Jenny (1941)(p. 281) 60 yr ago. Since that time, there have been many soil climosequence studies. Much of the work of the past two decades has been summarized recently by Birkeland (1999)(p. 430). One of the most commonly used geologic settings for soil climosequence studies is loess that covers a large region and spans gradients of temperature and precipitation. Such studies have been conducted in loess-dominated landscapes in China (Maher et al., 1994), Argentina (Alvarez and Lavado, 1998), and New Zealand (Webb et al., 1986). In the USA, soil climosequence studies have been made across the east-west precipitation gradient in the prairie-dominated loess belt (Jenny and Leonard, 1934; Wells and Riecken, 1969; Ruhe, 1984a) and across the north-south precipitation and temperature gradients in the forested loess uplands along the Mississippi River Valley (Torrent and Nettleton, 1979; Ruhe, 1984b,c). It is fortunate that many soil climosequences have been conducted in loess, as this is also the parent material for the Sangamon soil in much of the midcontinent of North America (Frye et al., 1968; Ruhe, 1969; Pye and Johnson, 1988; Rutledge et al., 1996; Markewich et al., 1998).

Several studies along the Mississippi River Valley suggest that significant chemical weathering of primary minerals has taken place in soils developed in last-glacial (Peoria) loess. Krinitzsky and Turnbull (1967), Snowden and Priddy (1968), Pye and Johnson (1988), and Markewich et al. (1998) all reported loss of both carbonate minerals and feldspars in soils derived from Peoria Loess in Mississippi and Arkansas. Beavers et al. (1963), Jones and Beavers (1966), Fehrenbacher et al. (1965), and Jones et al. (1967) reported alteration of silt-sized minerals and loss of soluble elements (Ca and K) relative to an insoluble element (Zr) in loess-derived soils in Illinois. These workers attributed the spatial variability in CaO/ZrO2 and K2O/ZrO2 values in Illinois to distance from the loess source, with higher (less weathered) values close to the source and lower values distant from the source. Although this interpretation is certainly reasonable, it is difficult to decouple the parent material factor (loess-thinning) from the climate factor, because many of the transects studied in Illinois also span a precipitation gradient. For example, the lower CaO/ZrO2 and K2O/ZrO2 values in southern Illinois that Jones and Beavers (1966) attribute to thinner loess also occur in a region of significantly higher precipitation than in northern Illinois, where CaO/ZrO2 and K2O/ZrO2 values are higher.

By far the most comprehensive study of loess-derived soils in the Mississippi River Valley is a transect from southern Mississippi to southern Minnesota studied by Ruhe (1984b)(c). Ruhe attempted to decouple loess sedimentation from climate variables by limiting his sample sites to areas immediately adjacent to the Mississippi River Valley, where presumably there would be a uniform parent material. He reported that soils in the northern part of the transect are thinner, less deeply leached, and have higher clay contents and base status than those in the southern part. Ruhe (1984c) attributed the differences in leaching depth and base status to climate; he felt the difference in clay content was because of local differences in loess sedimentation (Ruhe, 1984b). In contrast to other workers (Beavers et al., 1963; Fehrenbacher et al., 1965; Jones and Beavers, 1966; Jones et al., 1967; Krinitzsky and Turnbull, 1967; Snowden and Priddy, 1968; Pye and Johnson, 1988; Markewich et al., 1998), Ruhe thought that Mississippi River Valley loess-derived soils had experienced little chemical weathering and thought that the impact of climate was limited to the southward trend of decreasing base status. Nevertheless, evidence for chemical weathering of primary minerals reported by other workers (cited above), using either mineralogical or major element abundance data, suggests that a climate influence is detectable. Ruhe's (1984c) interpretations of moderate chemical weathering using only base status is debateable, because it is an indirect and partial indicator of primary mineral weathering. Furthermore, Ruhe's pedons were cultivated; base status could have been altered by fertilizer additions and soil amendments (e.g., liming).

We report results of a new soils study along the Mississippi River Valley, similar to Ruhe's (1984b)(c) transect. Major element chemistry of bulk soil samples is used to assess degree of chemical weathering. New stratigraphic data and geochemical analyses of deep loess provide the baselines for comparison of soils with a similar parent material. Our major study goal is to assess which chemical parameters are most useful in inferring past climates from buried soils, particularly the Sangamon soil.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
We sampled soils from northern Louisiana to northern Iowa on well-drained but uneroded upland localities where deep Peoria Loess exists (Fig. 1) . With the exception of pedons 3, 4, 5, 6, and 11, the soils were uncultivated. In the southern part of the transect, soils were described and sampled from hand-dug pits or deep roadcuts; in the northern part of the transect, soils were described and sampled from cores taken with a hydraulic drilling rig. Abbreviated soil descriptions and complete chemical data are given for all pedons in Table 1.



View larger version (65K):
[in this window]
[in a new window]
  		Fig. 1. Map showing the distribution of loess in the central USA, location of pedons sampled, and last-glacial paleowinds (Muhs and Bettis, 2000). Peoria Loess distribution slightly modified from Thorp and Smith (1952); Bignell Loess distribution compiled by the authors using data from Caspall (1972), Martin (1993), Johnson (1993), Kuzila (1995), Pye et al. (1995), and Muhs et al. (1999).

 

View this table:
[in this window]
[in a new window]
  		Table 1. Morphology and major and trace element concentrations in Mississippi River valley loess-soil transect.{dagger}

 
Pedons sampled in Louisiana, Mississippi, Tennessee, and Kentucky are in the Memphis series (fine-silty, mixed, active, thermic Typic Hapludalfs), those in southern Illinois are in the Alford series (fine-silty, mixed, superactive, mesic Ultic Hapludalfs), and those in northern Illinois and Iowa are in the Fayette series (fine-silty, mixed, superactive, mesic Typic Hapludalfs). All soils are Typic (Memphis and Fayette) or Ultic (Alford) Hapludalfs developed primarily under what was oak (Quercus)-hickory (Carya) deciduous forest in presettlement time (Kuchler, 1964). Past vegetation and climate differed from that of the present, but based on pollen studies, most of the region has probably been in mesic deciduous forest at least since about 9000 14C yr BP, with perhaps some mid-Holocene prairie occupation in parts of the northern Mississippi River Valley (Webb et al., 1993; Baker et al., 1996). In the southern portion of the transect, northern Louisiana, mean annual precipitation is ~1550 mm and mean January and July temperatures are 10°C and 28°C, respectively (based on 1961–1990 means for Baton Rouge, LA; all climatic data cited herein are from National Climatic Data Center, Asheville, NC, unpublished data, 1991). In northern Iowa, mean annual precipitation is as low as ~780 mm and mean January and July temperatures are -8°C and 23°C, respectively (based on 1961–1990 means for Prairie du Chien, WI, immediately east of our northernmost pedon). The southernmost part of the valley rarely has freezing temperatures, whereas the northern valley has 5 mo with mean daily minima below freezing (0°C). Precipitation has a winter maximum in the southern part of the valley, up to southern Illinois. From central to northern Illinois and Iowa, there is a very distinct summer maximum for precipitation. Ruhe (1984b)(c) pointed out that effective moisture is greater in the southern valley not only because of greater overall precipitation, but also because maximum precipitation occurs during cooler months, when evapotranspiration is lower.

After description and sampling, total-soil splits from each horizon were pulverized in a shatterbox; major element chemistry was determined by wavelength-dispersive x-ray fluorescence and trace element chemistry was determined by energy-dispersive x-ray fluorescence. Abundances of soluble elements were normalized to relatively insoluble Ti and Zr to develop weathering ratios (Birkeland, 1999, p. 430). Profile average weathering ratios were determined by weighting all horizons in a pedon by horizon thicknesses and summing, similar to the approach used by Ruhe (1984b)( c) for other parameters.

Major element composition of deep, unaltered loess was determined for samples obtained by drilling at two localities near Moline and Morrison (4 and 5 on Fig. 1) in northern Illinois, one locality (8 on Fig. 1) in southwestern Illnois, and published geochemical data from three localities at Natchez and Vicksburg, Mississippi (near pedons 21 and 22 on Fig. 1) reported by Pye and Johnson (1988). Pedon localities 5 and 4 correspond to the Rapids City B and Morrison sections originally described by Frye et al. (1968); loess geochemical data for these sections are in Muhs and Bettis (2000). Pedon 8 is at the Greenbay Hollow loess section described by Hajic (1990) and Grimley et al. (1998); geochemical data are reported here for the first time. Organic matter, charcoal, and conifer needle macrofossils from the Rapids City B loess were processed for accelerator mass spectometric radiocarbon dating at the USGS graphite extraction line laboratory in Reston, VA. After preparation at the USGS, abundances of 14C in the Rapids City B samples were determined at the Lawrence Livermore National Laboratory.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Loess Stratigraphy and Geochemistry
Loess stratigraphy and new radiocarbon ages agree well with previous studies of loess in the Mississippi River valley (Fig. 2 and 3) . At the Rapids City B loess section in northern Illinois (Fig. 2), radiocarbon ages of the Farmdale soil, which underlies Peoria Loess, indicate that last-glacial loess deposition began sometime after ~23000 14C yr BP and spruce (Picea) needles farther up in the section indicates Peoria Loess deposition was in progress by ~21000 14C yr BP. Radiocarbon ages around clay-rich marker beds from a section ~37 km southwest of Greenbay Hollow reported by Grimley et al. (1998) indicate that Peoria Loess deposition was still in progress ~18000 14C yr BP (Fig. 3). The youngest Peoria Loess deposition in Illinois may have occurred sometime after ~13000 14C yr BP and could be as young as ~10000 14C yr BP (Frye et al., 1968; Grimley et al., 1998; Wang et al., 2000). These ages are in reasonable agreement with 14C and thermoluminescence ages of Peoria Loess reported from Tennessee, Arkansas, and Mississippi (Snowden and Priddy, 1968; Pye and Johnson, 1988; Oches et al., 1996; Rutledge et al., 1996; Rodbell et al., 1997; Markewich et al., 1998).



View larger version (31K):
[in this window]
[in a new window]
  		Fig. 2. Stratigraphy of Rapids City B, Illinois loess section (locality 5 on Fig. 1), new accelerator mass spectrometric (AMS) radiocarbon ages, major element ratios (data from Muhs and Bettis, 2000, except for TiO2, reported for the first time here), and ranges of major element ratios in deep, unleached Peoria Loess from the southern Mississippi River valley (data from Pye and Johnson, 1988).

 


View larger version (35K):
[in this window]
[in a new window]
  		Fig. 3. Stratigraphy of Greenbay Hollow, Illinois loess section (locality 8 on Fig. 1), major element ratios (data from this study) and ranges of major element ratios in deep, unleached Peoria Loess from the southern Mississippi River valley (data from Pye and Johnson, 1988). Radiocarbon ages shown are from correlative flood deposits from a locality in adjacent Missouri and are from Grimley et al. (1998).

 
Comparison of deep, unweathered loess geochemistry in the Illinois sections with data reported by Pye and Johnson (1988) from localities in Mississippi shows that loess composition does not differ significantly in the northern and southern portions of the valley (Fig. 2, 3, and 4) . Plots of individual samples of deep, unweathered loess show that concentrations of CaO, MgO, K2O, Na2O, Al2O3,and Fe2O3 in northern Mississippi Valley loesses (from Illinois) are similar to those in southern Mississippi Valley loesses, from sections at Vicksburg and Natchez (Fig. 4). Furthermore, both northern and southern Mississippi Valley loesses have distinctly different concentrations of CaO, MgO, Al2O3, and Fe2O3 than loesses found either to the west (Iowa) or the east (Indiana). Mean values and ranges for CaO/TiO2, MgO/TiO2, K2O/TiO2, Na2O/TiO2, Al2O3/TiO2, and Fe2O3/TiO2 are not significantly different among the northern Illinois (Rapids City B and Morrison) sections, the central Illinois (Greenbay Hollow) section, and the southern Mississippi (Natchez and Vicksburg) sections. We conclude, therefore, that loess composition (soil parent material) was initially similar throughout the valley and, based on the radiocarbon chronologies above, pedogenesis began at about the same time, perhaps around 13000 to 10000 14C yr BP.



View larger version (19K):
[in this window]
[in a new window]
  		Fig. 4. Plots of concentrations of CaO, MgO, K2O, Na2O, Al2O3, and Fe2O3 in deep, unweathered loess from the northern Mississippi Valley, the southern Mississipi Valley, western Iowa, and southwestern Indiana. Northern Mississippi Valley loess samples are Peoria loess from the Morrison, Rapids City B, and Greenbay Hollow sections in Illinois (this study); southern Mississippi Valley loess samples are unleached Peoria loess from the Vicksburg and Natchez sections, Mississippi reported by Pye and Johnson (1988). Shown for comparison are ranges of concentrations for the same elements from the Loveland, Iowa loess section (Muhs and Bettis, 2000), and the Mount Vernon, Indiana loess section (section described by Hayward and Lowell, 1993; geochemical data are from the present authors).

 
Soil Morphology and Chemistry
Ruhe (1984b)(c) reported that soils along the Mississippi River valley are thicker, redder, and lower in clay in the southern portion of the valley compared with the northern portion of the valley. Our observations are in agreement with some of the trends reported by Ruhe. Ruhe (1984b) reported that soil B horizon colors become redder, from 10YR to 7.5YR to 5YR as one moves from the north to south in the valley. Most northern valley pedons we studied have 10YR hues and most southern valley soils have 7.5YR hues (Table 1). This is in agreement with Ruhe's observations and loess soil studies by Lindbo et al. (1997) and Rhoton et al. (1998). The Bt horizons of most soils in the southern part of the valley are silt loams whereas many of the Bt horizons in soils of the northern part of the valley are silty clay loams. Geochemical evidence for higher clay content in northern valley soils, which we discuss below, is in agreement with Ruhe's observations. Ruhe (1984b)(c) reported that sola in the southern valley are thicker than those in the northern valley. We tested this conclusion by detailed measurements of solum thickness. In our studies, we considered the solum depth to terminate where there was no evidence of subangular blocky, columnar, or prismatic structure (Table 1). This depth usually occurs well below where the deepest clay films occur. On the basis of this definition of solum thickness, we observed no systematic geographic relation for overall solum thickness. Soil Bt horizons, defined by those depths where there are well-defined clay films, are actually thicker in the northern valley (Fig. 5) .



View larger version (25K):
[in this window]
[in a new window]
  		Fig. 5. Solum and Bt horizon thicknesses shown as a function of mean annual precipitation (1961–1990 means) in the transect.

 
In soil climosequence studies, it is generally assumed that in regions with higher rainfall, there should be greater depletion of primary minerals (Birkeland, 1999, p. 430). We therefore compared soluble elements that proxy for primary minerals (Ca and Mg for calcite, dolomite, and other Ca- and Mg-bearing minerals such as hornblende, Na for plagioclase and hornblende, and K for mica) with Ti and Zr. Profile average CaO/TiO2 shows an inverse relation to mean annual precipitation (Fig. 6) . The coefficient of determination shows that the precipitation gradient explains about 75% of the variation in this weathering ratio. The same relation is obtained if Zr is used as the stable element instead of Ti. Profile average MgO values show similar trends and a similar degree of explanation, whether we use Ti or Zr as the stable element (Fig. 6).



View larger version (26K):
[in this window]
[in a new window]
  		Fig. 6. Profile average values of CaO/TiO2, MgO/TiO2, CaO/ZrO2, and MgO/ZrO2 as a function of mean annual precipitation in the transect.

 
Plagioclase is the main Na carrier, and profile average Na2O/TiO2 values show a logarithmic decline with increasing precipitation (Fig. 7) . Profile average Na2O values, normalized to ZrO2, also show a decline with increasing precipitation, although the relation is linear. Na2O/ZrO2 values have a high coefficient of determination that explains nearly 90% of the variability. The two primary minerals that are the main carriers of K are potassium feldspar and mica (biotite, muscovite, and illite). Profile average K2O/TiO2 and K2O/ZrO2 both show linear declines with increasing precipitation, although the coefficients of determination are not as high as for the other elements (Fig. 7). The lower coefficient of determination for K vs. precipitation could be because of the fact that K is found in both weatherable (mica) and resistant (potassium feldspar) minerals.



View larger version (24K):
[in this window]
[in a new window]
  		Fig. 7. Profile average values of K2O/TiO2, Na2O/TiO2, K2O/ZrO2, and Na2O/ZrO2 as a function of mean annual precipitation in the transect.

 
Concentrations of soluble trace elements, relative to stable Zr, are consistent with the major element weathering ratios (Fig. 8) . Strontium substitutes for Ca in Ca-bearing minerals and shows a decline in abundance relative to Zr with mean annual precipitation. Both Ba and Rb substitute for K in potassium feldspar and mica and both elements show significant decreases in abundance with mean annual precipitation, although neither has the degree of explanation shown for K.



View larger version (21K):
[in this window]
[in a new window]
  		Fig. 8. Profile average values of Sr/Zr, Ba/Zr, and Rb/Zr as a function of mean annual precipitation in the transect.

 
Ruhe (1984b) showed that profile average clay contents of loess-derived soils in the Mississippi River valley decrease with increasing precipitation. He attributed this counterintuitive finding to the effects of local sediment supply rather than any pedogenic process. Because total (as opposed to dithionite or oxalate-extractable) Al2O3 and Fe2O3 are highly correlated with clay content (Markewich et al., 1998; Mason and Jacobs, 1998; Muhs and Bettis, 2000), these values, normalized to TiO2, tend to be highest in soil Bt horizons (Fig. 2 and 3; Table 1). Profile average Al2O3/TiO2 and Fe2O3/TiO2 values are thus indirect measures of profile average clay content. In our transect, profile average Al2O3/TiO2 and Fe2O3/TiO2 values are negatively correlated with precipitation, in agreement with Ruhe's finding of higher clay contents in the northern part of the valley (Fig. 9) .



View larger version (28K):
[in this window]
[in a new window]
  		Fig. 9. Profile average values of Al2O3/TiO2, Fe2O3/TiO2, Al2O3/ZrO2, and Fe2O3/ZrO2 as a function of mean annual precipitation in the transect. Range of Al2O3/TiO2 and Fe2O3/TiO2 values in eastern Nebraska loess are from three localities (Plattsmouth, Lincoln, and Elba) reported by Muhs and Bettis (2000).

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
CONCLUSIONS
 REFERENCES
 
Profile average values of CaO/TiO2, MgO/TiO2, K2O/TiO2, Na2O/TiO2, Sr/Zr, Ba/Zr, and Rb/Zr in loess-derived soils of the Mississippi River Valley all show decreases as a function of precipitation. The trends for the major elements are unchanged if Zr is used as a stable index element rather than Ti. Because these soils all lack carbonates, the systematic decreases in CaO/TiO2 and Sr/Zr are probably because of greater depletion of noncarbonate, Ca-bearing minerals, such as hornblende or Ca-plagioclase in the southern part of the valley. Hornblende and biotite weathering might explain the similar trend for MgO/TiO2 values.

Both feldspars and micas are common primary minerals in Mississippi Valley loess (Snowden and Priddy, 1968; Pye and Johnson, 1988; Markewich et al., 1998). Previous studies suggest that plagioclase feldspar has undergone depletion in modern soils in at least the southern part of the valley (Krinitzsky and Turnbull, 1967; Pye and Johnson, 1988; and Markewich et al., 1998). The trend of decreasing Na2O/TiO2 with increasing precipitation supports these observations, although some of the Na depletion may also come from alteration of hornblende. Both potassium feldspar and mica are carriers of K, Ba, and Rb. Theoretical concepts of mineral stability (such as the Goldich stability series, discussed in Birkeland, 1999, p. 430) combined with empirical studies of rock and loess weathering (Colman, 1982, p. 51 and Markewich et al., 1998) suggest that potassium feldspar is a fairly stable mineral in midlatitude soils. Therefore, we interpret the declines in K2O/TiO2, Ba/Zr, and Rb/Zr as a function of increasing precipitation to be primarily the result of mica weathering. Mica is found in both the silt and clay fractions of Mississippi River Valley loess (Snowden and Priddy, 1968).

Many soil climosequence studies, including those conducted in loess, have shown an increase in clay content with increasing precipitation (Birkeland, 1999, p. 430). The usual interpretation of this trend is that greater precipitation produces greater chemical weathering of primary, silt-sized minerals and alteration to clay minerals. Because Ca, Mg, Na, and K all show depletions relative to either stable Ti or Zr in the more humid southern part of the Mississippi Valley, we would have expected that this part of the valley should also show higher clay contents, the usual products of chemical weathering. However, both Ruhe's (1984b) and our results show the opposite trend. Although local sedimentation variability (e.g., loess with higher clay contents at certain points along the transect) could explain this trend, a more random pattern of geographic variability might be expected. Furthermore, deep loess profiles in Illinois show Al2O3 concentrations and Al2O3/TiO2 values similar to those in Mississippi (Fig. 2, 3, and 4). This is supported by Ruhe's (1984b) data, which show that clay contents in the C horizons of loess-derived soils have no systematic relation to latitude. We conclude from these observations that soils in the northern part of the Mississippi River Valley may have received secondary additions of clay-rich sediment, a conclusion also reached by Mason and Jacobs (1998), based on mass balance calculations of chemical data from modern, loess-derived soils.

As pointed out by Mason and Jacobs (1998), there could be at least two explanations for the clay enrichment in northern Mississippi Valley soils. One possibility is that during the final stages of Peoria Loess sedimentation, fine-grained loess derived from a distant source (as well as coarse-grained loess from the local source) was deposited in the upper Mississippi Valley. Recent particle-size and geochemical data suggest that clay and fine silt-enriched dust from a source west of the Missouri River (most likely Nebraska) was deposited in western Iowa during the final stages of Peoria Loess sedimentation (Muhs and Bettis, 2000). It is possible that even finer-grained components of loess from this distant source, or from western Iowa, traveled as far east as western Illinois and added clay minerals enriched in Al and Fe to coarser-grained loess derived from the Mississippi River. Radiocarbon ages show that Peoria Loess deposition in Nebraska did not cease until ~10500 14C yr BP (Maat and Johnson, 1996; Muhs et al., 1999), similar to or younger than the final stages of loess deposition in Illinois (Frye et al., 1968; Grimley et al., 1998). This hypothesis could be tested by Pb-isotope compositions of coarse silt, fine silt, and clay fractions of loess in northern Illinois, following the methods outlined by Aleinikoff et al. (1998)(1999).

Another explanation for the high clay mineral content of the northern valley soils is that there have been secondary, fine-grained dust additions to soil surfaces during the Holocene. There is stratigraphic and geochronologic evidence for Holocene loess deposition (referred to as the Bignell Loess; Fig. 1) in the Great Plains region of Nebraska, Kansas, and Colorado (Pye et al., 1995; Maat and Johnson, 1996; Muhs et al., 1999). Even where the Bignell Loess is not identified as a discrete stratigraphic unit in the field, both silt and clay mineralogical differences show that it comprises the upper part of modern soils in some areas of Nebraska (Kuzila, 1995). Because the Great Plains region lies upwind of the northern Mississippi River Valley (Fig. 1), fine-grained components of Bignell Loess could have traveled eastward to Illinois and Iowa. The sediment record from Elk Lake, MN contains Al-rich clays and fine silts that are interpreted to be distant-source eolian sediments that date from a mid-Holocene warm and dry period (Dean, 1997).

It is difficult to determine whether fine-grained eolian additions to soils of the northern Mississippi River Valley occurred during the last glacial period or the Holocene, and it is possible that deposition occurred during both periods. Use of immobile element ratios or Pb-isotopic compositions of potassium feldspars could identify a secondary source that is distinct from the local source, but such data say little about the timing of secondary, fine-grained dust additions. The addition of such material, however, most likely came from a source west of the Mississippi River. The distribution of both Pleistocene and Holocene loess to the west of the Mississippi River Valley and dominantly westerly modern winds and paleowinds (Fig. 1) could explain why soils in Illinois were affected by fine-grained dust additions, but soils south of Illinois were not.

Despite the possible freshening effect that fine-grained late-glacial or Holocene additions of loess may have had, northern valley soils nevertheless show evidence of chemical weathering. Ratios of Na2O/TiO2, a proxy for degree of plagioclase depletion, are shown as a function of depth for four pedons in the northern valley in Fig. 10 ; also plotted is the range of Na2O/TiO2 for unweathered loess in eastern Nebraska (same localities as those in Fig. 9). These plots show that in the upper horizons of all pedons, Na2O/TiO2 values are lower than the range of values for eastern Nebraska loess. Thus, even in the northern valley, where weathering rates may be slowest, the rate of plagioclase depletion has kept ahead of the rate of younger eolian accretion.



View larger version (36K):
[in this window]
[in a new window]
  		Fig. 10. Plots of Na2O/TiO2 values as a function of depth for four pedons in the northern part of the transect. Shown for comparison is the range of Na2O/TiO2 values for eastern Nebraska loess (three localities are Plattsmouth, Lincoln, and Elba).

 

   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
REFERENCES
 
Profile average weathering ratios of mobile elements (Ca, Mg, Na, K, Sr, Ba, and Rb) to immobile elements (Ti and Zr) show that loess-derived soils of the Mississippi River Valley are more chemically weathered in the humid, southern part of the valley. Depletion of the primary minerals calcite, dolomite, hornblende, plagioclase, and mica are the likely causes of these chemical trends. Greater alteration of primary minerals in the southern part of the valley may be a function of higher rainfall and warmer temperatures, which promote greater chemical weathering.

If primary mineral weathering is greater in the humid southern part of the valley, clay mineral content is expected to be higher there than in the drier, northern part of the valley. Both Al and Fe are highly correlated with clay mineral content. Our studies indicate higher profile average Al and Fe relative to Ti or Zr in the northern part of the valley. These results, while unexpected, are consistent with similar chemical and particle-size data reported by other workers. The trend observed in this study suggests that the northern valley soils may have received fine-grained secondary additions of clay-sized particles from a distant source west of the Mississippi River, based on modern and paleowind directions. Testing of the distant-source hypothesis could be accomplished with Pb-isotopic analyses of different size fractions of the loess and soils (Aleinikoff et al., 1998, 1999).

Our results differ from those of Ruhe (1984b)(c), who concluded that climate had little impact on modern loess-derived soils of the Mississippi River Valley and that soils in this region have experienced little chemical weathering. Primary mineral alteration is shown by a number of elemental ratios, and all show a high degree of correlation with mean annual precipitation. However, a trend of higher clay content with northern valley soils suggests that the assumption of a uniform parent material for all soils along the valley may not be valid. Thus, extreme care must be used in interpreting chemical data of paleosols for paleoclimatic interpretations. Nevertheless, the results indicate that major and trace element weathering ratios, when utilized with caution, could be useful paleoclimatic indicators in the interpretation of loess-derived paleosols, such as the Farmdale and Sangamon soils.


   ACKNOWLEDGMENTS
 
This study was supported by the Earth Surface Dynamics Program of the U.S. Geological Survey and by the Iowa Department of Natural Resources, Geological Survey Bureau. We thank the landowners at Morrison (Pauline Meakins) and Rapids City (Dan Van Besien) for access to their property, and Ed Hajic and Deborah Quade for help with drilling. Richard Pullman of the NRCS kindly provided descriptive data and samples for Pedon 11. Helaine Markewich, Milan Pavich, and three anonymous reviewers read an earlier version of the paper and made helpful comments for its improvement.

Received for publication November 10, 2000.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 




This article has been cited by other articles:


Home page
 Soil Sci.Home page
C. A. Stiles and K. A. Stensvold
Loess Contribution to Soils Forming on Dolostone in the Driftless Area of Wisconsin
Soil Sci. Soc. Am. J., May 1, 2008; 72(3): 650 - 659.
[Abstract] [Full Text] [PDF]

Home page
 Clay MineralsHome page
M. EGLI, A. MIRABELLA, G. SARTORI, D. GIACCAI, R. ZANELLI, and M. PLOTZE
Effect of slope aspect on transformation of clay minerals in Alpine soils
Clay Minerals, September 1, 2007; 42(3): 373 - 398.
[Abstract] [Full Text] [PDF]

Home page
 J. Environ. Qual.Home page
R. Burt, M. A. Wilson, M. D. Mays, and C. W. Lee
Major and Trace Elements of Selected Pedons in the USA
J. Environ. Qual., November 1, 2003; 32(6): 2109 - 2121.
[Abstract] [Full Text] [PDF]

Home page
 Soil Sci.Home page
M. Aide and C. Smith-Aide
Assessing Soil Genesis by Rare-Earth Elemental Analysis
Soil Sci. Soc. Am. J., September 1, 2003; 67(5): 1470 - 1476.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (18)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Muhs, D. R.
Right arrow Articles by McGeehin, J. P.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Muhs, D. R.
Right arrow Articles by McGeehin, J. P.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Muhs, D. R.
Right arrow Articles by McGeehin, J. P.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Vadose Zone Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome