RICE UNIVERSITY
Broadening Theories of Soil Genesis: Insights from Tanzania and Simple Models
by
Mark Gabriel Little
A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE
Doctor of Philosophy
APPROVED, THESIS COMMITTEE:
Cin-Ty ee, Chair, Assistant Professor Earth Science
Andreas Liittge, Associ e Professor Earth Science and Chemistry
~~d/Maso~Olh~ Pl"()fussor Civil and Environmental Engineering
Jo Anderson, W. Maurice Ewing Pr: essor in Oceanography Earth Science
Carrie Masiello, Assistant Professor Earth Science
HOUSTON, TEXAS MAY2007
ABSTRACT
Broadening Theories of Soil Genesis:
Insights from Tanzania and Simple Models
by
Mark Gabriel Little
Three basic assumptions of soil formation are challenged herein: the degree of chemical weathering decreases with depth; increased physical weathering due to high topographical gradients causes an in?rease in chemical weathering; and the mineral soil derives from the transformation of in situ parent material. The first part presents an investigation into the degree and nature of chemical weathering during soil formation on a volcanic substrate on Mt. Kilimanjaro in northern Tanzania. The degree of weathering was found to increase with depth in the soil profile. Observations show that the upper and lower layers of the weathering profile have undergone different weathering histories.
The presence of a buried paleosol or enhanced weathering due to lateral subsurface water flow may explain the observations, the latter having novel implications for the transport of dissolved cations to the ocean. The second part presents a model to test the link between chemical weathering associated with soil fmmation and erosion associated with mass wasting. The predicted ratios suspended/dissolved ratios, however, are all higher than observed in rivers, the discrepancy worsening with increasing topographic relief. This discrepancy arises from the fact that in regions of high relief, mass wasting are so high that soil mantles do not reside on hillslopes long enough to allow for significant chemical weathering. The discrepancy between the model and observations can be explained by: over-estimate of predicted suspended load; absence of chemical weathering of deltaic/alluvial sediments from the model; or chemical weathering associated with groundwater weathering. The third part presents data from a sequential extraction on a basaltic soil from Mt. Meru in Northern Tanzania. The behavior of relatively immobile elements is consistent with soil formation being accompanied by mass loss due to chemical weathering. However, superimposed on this mass loss appears to be enrichment of most elements measured. These data suggest that the surface of the Meru soil columns may have experienced "re-fertilization" by the deposition of volcanic ash. ACKNOWLEDGEMENTS
In the fall of2003, I arrived in Houston to work with Cin-ty A Lee, the most inquisitive, challenging and creative advisor I could ask for. Cin-ty has shaped my concept of good science and the societal role of the scientist. He has also provided vital research funds and much of my stipend (through departmental funds and a Packard
Fellowship). The bulk of my stipend has been generously provided by a Ford Foundation
Fellowship and the Office of the President of Rice University. I came to Rice with the support of many former professors including Richard O'Connell, Daniel Schrag, John
Shaw, Joseph Kalt and Otto Solbrig at Harvard and my senior thesis advisor, Maria Zuber, from M.I.T.
Conversations with my thesis committee, Andreas Liittge, Mason Tomson, John
Anderson and Carrie Masiello, have been central to my progress and research. Others have also been wonderful resources: Arnaud Agranier, Philip Bedient, Jim Blackburn,
Joe Cibor, Paul Harcombe, Zhengxue Li, Neal Lane, Julie Morgan, Glen Snyder and
Carter Sturm. Musa Naroro, Felix John and Nuru and A.S.S. Mbwana, George Sayulla and Rama Ngatoluwa were all invaluable during fieldwork in Tanzania. I am also grateful to the James A. Baker III Institute for Public Policy for providing me a way to engage the Houston community. And I am indebted to my students who inspired me while suffering through my disorganization.
Of course, none of this would be possible without my parents, Shade Keys Little and Bernadette Gray-Little, my Grandma Little and all of my antecedents. v
TABLE OF CONTENTS
1 TITLE PAGE
11 ABSTRACT lV ACKNOWLEDGEMENTS v TABLE OF CONTENTS
1 INTRODUCTION
5 PART I: On the formation of an inverted weathering profile on
Mount Kilimanjaro, Tanzania: buried paleosol or groundwater
weathering?
52 PART II: Physical and chemical weathering in mountainous
regions: insights from a model linking chemical weathering to soil
formation, creep and mass wasting
96 PART III: Element distribution in soils from Mount Meru,
Tanzania: a combined bulk and leach study
157 APPENDIX A: Investigating the distribution of labile metals in a
vertisol in southeastern Texas
180 APPENDIX B: Bulk composition tables from Mweka, Olmani and
Monduli
200 NOTES
202 BIBLIOGRAPHY 1
INTRODUCTION 2
Soils form a thin chemical boundary layer between the oxidizing, wet atmosphere
6 8 and the solid earth. Compared to the radius of the earth (~ 10 · m) or the thickness of the
4 5 2 2 continents (~ 10 · m), soils are relatively thin (~ 10 to 1o- m). As such, soils account for
small fraction of the total mass of the planet. However, soils are located at the boundary
between the solid and fluid earth and are therefore exceedingly important for a number of processes including the global C02 cycle, sediment supply to the oceans and numerous
other human activities including agriculture.
The capacity of C02 to retain solar energy in the atmosphere as heat was
suggested as early as 1896 by Svente Arhhenius. The recognition of C02 as an important
greenhouse gas has become common knowledge as the combustion of fossil fuels has
driven an unprecedented rise in atmospheric C02 (UNFCCC, 1998) (Somerville et al.,
2007) (Petit et al., 1999). The relationship between C02 and climate on long time scales
has been strongly asserted by the correlation of C02 measurements and temperature
proxies in the Vostok ice core (Petit et al., 1999). C02 is removed from the atmosphere
when silicate rocks are chemically weathered and on long time scales weathering of
silicate rocks is the main sink for C02 (Berner, 1995; Berner et al., 1983a). As the by
product of this process, soils retain the reaction products of chemical weathering
reactions and are a record of these processes.
Chemical weathering reactions liberate some fraction of every element contained
in the protolith. These dissolved elements may be transported to the oceans via rivers.
The time averaged flux of dissolved material can be deduced from the composition of the
soil mantle when compared to samples of unweathered protolith. Physical weathering,
e.g. soil creep and mass wasting, liberates un-dissolved solid material that may also be 3 transported to the ocean via rivers as suspended load. Both the dissolved and suspended fluxes are of interest to the sedimentology community and they are important for marine
nutrient supply. Areas of high topographic relief are of particular interest because the
major rivers of the world have their source in mountainous regions, mass wasting is
correlated with topographic gradients and tectonic uplift has been related to weathering
rates (Gaillardet et al., 1999; Millot et al., 2002; Raymo et al., 1988).
Soils have direct impact on humans because they provide nutrients and a growth
substrate for agriculture. Thus understanding the fate of soil nutrients and soil formation
rates are important for agricultural planning in natural systems. Soils are also host to a
variety of inorganic elements, e.g. Ph (Othman et al., 1997) that may be pernicious or
toxic to humans. Therefore the mobility of elements of health concern is important to
human health. In East Africa, soil erosion from overgrazing is a growing problem as
population demands stress the agricultural and livestock sectors (Little et al., 2001;
Scoones and Toulmin, 1999; Solomon et al., 2000). As the population and agricultural
use of volcanic soils in Northern Tanzania increases, information about the nutrient
availability as well as the distribution of heavy metals will be come more important
(Anderson, 1982; Mlingano et al., 2006).
The work presented here addresses aspects of each of these soil-related concerns.
The first part presents an investigation into the degree and nature of chemical weathering
during soil formation on a volcanic substrate on Mt. Kilimanjaro in northern Tanzania.
The degree of weathering was found to increase with depth in the soil profile.
Observations show that the upper and lower layers of the weathering profile have
undergone different weathering histories. The presence of a buried paleosol or enhanced 4 weathering due to lateral subsurface water flow may explain the observations, the latter having novel implications for the transport of dissolved cations to the ocean. The second part presents a model to test the link between chemical weathering associated with soil
formation and erosion associated with mass wasting. The predicted ratios
suspended/dissolved ratios, however, are all higher than observed in rivers, the
discrepancy worsening with increasing topographic relief. This discrepancy arises from
the fact that in regions of high relief, mass wasting are so high that soil mantles do not
reside on hillslopes long enough to allow for significant chemical weathering. The
discrepancy between the model and observations can be explained by: over-estimate of
predicted suspended load; absence of chemical weathering of deltaic/alluvial sediments
from the model; or chemical weathering associated with groundwater weathering. The
third part presents data from a sequential extraction on a basaltic soil from Mt. Meru in
Northern Tanzania. The behavior of relatively immobile elements is consistent with soil
formation being accompanied by mass loss due to chemical weathering. However,
superimposed on this mass loss appears to be enrichment of most elements measured.
These data suggest that the surface of the Meru soil columns may have experienced "re
fertilization" by the deposition of volcanic ash.
Each part of this work challenges basic theory of soil genesis: the degree of
chemical weathering decreases with depth; increased physical weathering due to high
topographical gradients causes an increase in chemical weathering; and the mineral soil
derives from the transformation of in situ parent material. When taken in sum, the data
suggests that we broaden our concept of how soils form. 5
PART 1:
On the formation of an inverted weathering profile on Mount Kilimanjaro,
Tanzania: buried paleosol or groundwater weathering?
Published in Chemical Geology, volume 235, pages 205-
221 and authored by Mark Gabriel Little and Cin-ty A Lee. 6
On the formation of an inverted weathering profile on Mount Kilimanjaro,
Tanzania: buried paleosol or groundwater weathering?
Synopsis
This is an investigation into the degree and nature of chemical weathering during soil formation on a volcanic (phonolite) substrate on the southern slopes of Mt. Kilimanjaro in northern Tanzania. The high field strength elements Nb and Ta were used to estimate enrichments and depletions relative to the bedrock. The degree of weathering was found to increase with depth in the soil profile. At depths greater than 200 em, Si, Na, K, Ca, and Mg have been depleted by nearly 100 % while AI has been enriched, resulting in a highly aluminous soil residue (40-50 wt% Ah03). At depths shallower than 200 em, the soil is also depleted in Si, Na, K, Ca and Mg though not to the extents seen at depths greater than 200 em. The lower degrees of weathering in the upper 200 em are also evidenced by the fact that the layer above 200 em is characterized by slight positive Eu anomalies relative to other rare earth elements whereas the deeper layer exhibits no Eu anomalies. The rare earth element systematics are consistent with preferential weathering of the glassy matrix in the upper 200 em, leaving behind plagioclase phenocrysts, which are enriched in Eu. In the deeper layer, weathering appears to be so extensive that both
Eu-rich plagioclase phenocrysts and Eu-poor glass/ash have largely weathered away.
These observations collectively show that the upper and lower layers of the weathering profile have undergone different weathering histories. Four scenarios may explain the apparent inverted weathering profile: re-precipitation followed by erosion, Aeolian deposition, a buried paleosol, and enhanced weathering due to lateral subsurface water 7 flow. The first hypothesis fails to explain the massive losses of Si, Na, K, Ca, and Mg
below 200 em. The Aeolian deposition hypothesis is also untenable because it
contradicts the trace element and REE's behaviors. The latter two hypotheses are both
reasonable; however, the buried paleosol model is inconsistent with some physical and
geochemical observations and the subsurface flow model requires the influence of
hydraulic conditions not tested in this study. It is concluded here that either the buried
paleosol model or the subsurface flow model can explain the formation of the Machame
soils with the latter having novel implications for the transport of dissolved cations to the
ocean. 8
1. Introduction
Quantifying the rate of continental weathering is of fundamental importance for understanding atmospheric C02 drawdown (Berner, 1995; Berner et al., 1983a) and the magnitude of the fluxes of various elements into the marine environment, many of which have paleo-oceanographic significance (Gaillardet et al., 1999; Kump et al., 2000; Sak et al., 2003). In a quasi-steady-state system, the continental weathering flux can, in theory, be estimated anywhere along the pathway between the source and the sink, the latter represented by marine sediments and the former represented by chemical weathering of the continents. This means that one approach in measuring the continental weathering flux is to measure the flux of solutes into the marine environment at the mouths of major river systems (Dessert et al., 2001; Gaillardet et al., 1999). This approach is convenient because the watershed areas of large river systems effectively average over large regions of continental crust, and as such, are believed to be fairly representative of average continental weathering rates. Thus, if the goal is to quantify global weathering rates and fluxes into the marine environment on large scales, the riverine perspective should be sufficient. However, if the goal is to better understand the mechanisms of continental weathering, the averaging effect of rivers is no longer a benefit. To understand the mechanisms of continental weathering, attention must then be focused toward understanding the source terms in addition to the riverine fluxes.
In this context, the most obvious source term to turn to is the chemical weathering flux associated with soil formation on either stable geomorphic surfaces (non-eroding) or steady-state landforms, where erosion and conversion of bedrock to soil roughly balance.
However, other source terms may also exist. For example, subsurface water flow may 9 result in chemical weathering of the deep subsurface, but its significance in the global
context is unclear largely because of the greater inaccessibility of the deep subsurface
compared to soil profiles. Nevertheless, recent studies seem to hint that subsurface water
flow may contribute to the chemical weathering flux of continents (Basu et al., 2001 ). If
such subsurface water is significant and bypasses the river mouth by directly entering the
oceans, the solute flux measured at river mouths may underestimate the true continental
weathering flux to oceans for some elements.
A young, but unusually mature weathering profile was investigated. The soil is
situated on an intermediate volcanic substrate on the southern slopes of Mt. Kilimanjaro
in northeastern Tanzania, a region having a tropical equatorial climate. This weathering
profile was fortuitously exposed by a road cut, affording access to the deep, and
otherwise inaccessible, portions ofthe weathering profiles in this region. Importantly, in
this weathering profile, the degree of weathering appears to be greatest at depth, which is
the inverse of what is seen in typical soil profiles where weathering is top-down. This
inverted profile may be the result of a buried paleosol or enhanced chemical weathering
at depth due to focused subsurface lateral water flow. The latter interpretation, if correct,
would underscore the importance of subsurface water weathering at least locally. 10
2. Study Area and description of weathering profile
Samples were collected from a road cut-exposed weathering profile (03°11.275' S,
37°14.157' E, elevation 1639 m) in the Machame region of northern Tanzania on the
southern slope of Mt. Kilimanjaro (Figure 1). The andosols in this region were formed
on phonolitic lava and ash flows. The weathering profile examined in this study is
situated on a relatively undissected surface, which suggests relatively little erosion. Most
erosion is confined to isolated ravines bordering large swathes of non-eroding
geomorphic surfaces. The studied weathering profile is situated on the edge of one of these uneroded geomorphic surfaces and has been conveniently exposed as a
consequence of a deep ravine and a road cut. Lava flows on the southern slope of
Kilimanjaro have been dated between roughly 0.17 to 0.51 million years (Dawson, 1992;
Evemden and Curtis, 1965). The modem, average annual rainfall in the Machame region
is 1929 mm/yr (Anderson, 1982).
The relatively unweathered proto lith of the Machame soils is a bluish-gray
lithified welded ash of phonolitic composition. Proto lith samples were obtained from
blocks of bedrock exposed near the base of the road cut. The groundmass of the protolith
is largely aphanitic (~90% of thin section area) and is dominated by small plagioclase
lathes and tiny dark green to black grains of magnetite-ulvospinel. Occasional
phenocrysts are olivines with some of those having alteration rims of iddingsite. The
thickness of the sampled weathering profile is ~5 m and consists of three main horizons
(from bottom to top): a dark gray, semi-consolidated saprolite composed of gibbsite
gravel and weathered clasts; a friable, grayish brown, aluminum rich gravel with large
voids; and a dark brown, organic rich, sandy loam layer. A very shallow Oi horizon, less 11 than 5 em thick, consisting primarily of forest litter, was also present, but not sampled. 8
soil samples were collected at 30-60 em intervals down to a maximum depth of ~370 em. 12
3. Methods
Machame soil and proto lith samples were dried for 24 hours at 105°C. Samples were then ground by hand with a ceramic mortar and pestle. Six grams of each sample powder and USGS rock powder standards BHV01, BHV02, and BIR1 were sent to
Washington State University at Pullman's GeoAnalytical Lab for XRF determination of
Si, Ti, AI, Fe, Mn, Mg, Ca, Na, K, and P concentrations and for determination of Loss on
Ignition (LOI), defined as the fraction of mass lost after heating each sample for 16 hours at 900 °C (Cornelius, 2004; Johnson et al., 1999).
Sample powders, a blank, and BHV01, BHV02, and BIR1 USGS rock standards were prepared for inductively coupled plasma mass spectrometry (ICP-MS) analysis at
Rice University using a series of acid dissolutions. Approximately 80 mg of the
Machame soil and rock powders, BHV02, and BIR1 were weighed; 2 aliquots of approximately 40 and 80 mg ofBHV01 were weighed for use as an external standard.
Powders were placed in Teflon beakers with approximately 0.5 mL each of concentrated
Seastar HN03 and HCl04. These solutions were sealed and ultrasonicated for ~20 min.
Beakers were then heated overnight at 115 °C. The following day, beakers were removed from the oven, unsealed, and placed on heating pads at ~ 17 5 °C until the solutions had dried down. This process of adding acids, overnight heating, and drying down was repeated 2 additional times; once again with the addition of 0.5 mL each of concentrated
HN03 and HCI04 and finally with the addition of 1.0 mL ofHCl04 only.
Following the third and final dry down, ~ 1 mL of 2% HN03 was added to the samples. Beakers were placed on a hot plate at ~ 100 °C for about 2 hours.
Approximately 280 J.tL of a 440.81 ppb Indium solution (to serve as an internal standard) 13 was precisely weighed into a 125 mL capacity, low density polyethylene bottle. The contents of the Teflon beakers were carefully washed into the polyethylene bottles using dilute HN03. A small drop of 1M HCl was added to keep the dissolved Fe in solution.
The contents of the bottles were then diluted up to 100 mL ( ~ 1 ppb Indium in the final solution) with dilute HN03 and weighed again.
Machame samples, blank, BHV01, BHV02, and BIR1 were all run on a Finnigan
Element II, single collector ICP-MS at Rice University. The ICP-MS was run in low and medium resolution modes (m/~m=300, 4000), the latter allowing for the analysis of Na,
Mg, AI, P, K, Ca, Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Zr, and Nb, for which isobaric molecular interferences can occasionally be a problem in low resolution. The In values were used to correct for instrumental drift. The procedural blank was used to correct for blanks introduced during the dissolution process. The external precision is <3%. 14
4. Results
The concentration data from the XRF analysis were consistent with the ICP-MS data for the major elements. For internal consistency, only ICP-MS data is reported in
Table 1 (with the exception of Si which is lost as volatile SiF4 during the dissolution process). The partly unweathered protolith (i.e., bedrock) has 52 wt.% Si02, 7.2% K20,
4.4% Na20, and 26% Ah03 with an Mg# (Mg/(Mg+Fe) x 100) of21. The intermediate
Si02 contents and high total alkali contents classifies this protolith as a phonolitic tephrite.
This protolith appears to have unusually high Ah03 and low Mg# compared to magmatic rocks having similar Si02 contents (Ah03 contents typically range between 15-20 wt. %).
The low Mg# of this rock is likely due to extensive crystallization of mafic minerals (e.g., olivine) and possible loss of Mg associated with small degrees of dissolution that may have already occurred in the proto lith sample. The slightly high Ah03 content of the protolith also suggests that the protolith may have already experiences some weathering.
The major element concentrations of the soil samples are very different from the relatively unweathered bedrock protolith and suggest extensive weathering (Fig. 2). The
Si02 concentrations in the bedrock range from 3 to 4 wt. % below 200 em and from 17 to
22 % above 200 em, indicating extensive Si dissolution in all parts of the soil column. Si loss appears to be greatest in the deepest section of the soil profile. The elements, Na, K,
Ca, and Mg, all of which are relatively mobile like Si, have also been lost from the soil to varying degrees, and like Si, appears to be most depleted from the lower layer. The zonations in Si, Na, K, Ca, and Mg contents suggest that the lower layer is more developed. This is consistent with Zr, Hf, Nb, and Ta concentrations, which increase with depth by more than a factor of two. As discussed in Section 5.2, these elements are 15 relatively immobile, hence their concentrations should increase with progressive weathering dissolution. Two other elements often used as immobile element tracers, AI and Ti, may have been mobilized as the LOT-corrected AI and Ti contents are higher than the protolith throughout the soil column (Fig. 2).
The rare earth elements (REEs) also display some consistent patterns (Fig. 3).
Except for Ce, REE concentrations are slightly depleted (relative to the protolith) throughout the soil column, being most depleted at depths greater than 200 em. The
depletion ofREEs appears also to be accompanied by subtle fractionations in the relative
abundances ofthe REEs. A positive Ce anomaly (relative to the other REEs) has
developed throughout the soil column. In some cases, the Ce concentrations are enriched
over the proto lith and in some cases depleted, but there is no apparent systematic
correlation with depth unlike the gross behaviors of the other REEs. Above 200 em, the
protolith-normalized abundance patterns are convex upwards and show subtle positive Eu
anomali~:s (relative to the other REEs). The convex REE profiles suggest at face value
that the light REEs and heavy REEs may have been more efficiently depleted than the
middle REEs. At depths greater than 200 em, the protolith-normalized abundance
patterns are remarkably flat (i.e. no Eu anomaly) except for the positive Ce anomaly.
Finally, one suite of elements shows clear surface enrichments. The major
element oxides P20 5 and Ti02 have been enriched from ~0.4 to 1% and ~ 1 to 4%,
respectively, from the protolith to surface layer. Co, Pb, Cu, Cr, Ni, and V show much
larger surface enrichments: at the surface these elements may be 2 orders of magnitude
greater than concentrations in the protolith and at depth (Fig. 2). U and Th show 16 moderate enrichments throughout the soil column though the enrichments below 200 em are greatest (Fig. 2). 17
5. Estimating mass changes during soil formation
The elements Zr, Hf, Nb, and Ta are generally considered immobile
(conservative), hence their weight concentrations should increase with increasing degree of mass dissolution associated with weathering. In most soil profiles, it is the shallowest parts that have been undergoing weathering for the longest times, thus the concentrations of immobile elements should be elevated at shallow levels, gradually decreasing with depth down to the bedrock. However, the Machame soils are enriched in Zr, Hf, Nb, and
Ta at depth (Fig. 2). In this section, the amount of dissolution experienced in the soil column is quantified. In section 6, possible mechanisms for the formation of an apparent inverted soil profile are discussed.
5.1 Calculating chemical weathering and element mobility
The Chemical Depletion Factor (CDF) and the Chemical Index of Alteration (CIA) are two proxies for chemical weathering (Nesbitt and Young, 1984; Riebe et al., 2003).
The CDF is an estimate of mass changes during soil formation based on immobile elements where,
(1)
The CDF is an approximation of the fraction of net mass removed during so that values approach zero for the least weathered soil and values approach 1 for highly weathered soils. The CIA is an estimate of the amount of chemical weathering based on the ratio of
Al, which is relatively refractory, to the sum of Al and more mobile major elements Ca and Na where, 18
(2)
As a rock is weathered and the Ca and Na are preferentially dissolved, the CIA increases from initial values of around 0.6 for upper continental crustal rocks to a maximum of 1.0.
These two proxies provide a generic method of comparison to other soils and within the profile.
In order to determine the mobility of individual elements, the relative mass losses
and additions are estimated by considering a mass balance approach, that is,
(3)
where Mo represents the total mass of the original parent material, Msoil is the mass
remaining in the soil residue, M11 is the net mass change during weathering by such
processes as chemical weathering, aeolian deposition, or authigenic precipitation from
solute-laden water derived from other parts ofthe weathering profile or precipitation
(Brimhall and Dietrich, 1987b). The lowercase mo represents the mass of one element in
the original parent material, msoil is that element in the soil, and m11 is the net mass change
of that element during weathering. The effects of adding water or organics are explicitly
ignored; therefore, all concentrations referred to herein are LOI-corrected. Before
estimating the relative amount of each element lost, the relationship between the mass of
the original proto lith, the mass of the soil produced, and the concentrations of immobile
elements in the protolith and the soil must be identified. The mass ratio of original
protolith to current soil can be estimated by,
--=--Msoil (4) Mo C~ml 19
where C~ is the weight concentration of an immobile element i in the original bedrock protolith and c;ou is the concentration of the same immobile element in the soil. It
follows that the ratio of the mass change of a mobile element, j, to mass of that element in
the protolith can be expressed by,
c~ *Ci cJ mj -c-i- soil- 0 Ll. = ----"so=il ____ (5) mi 0
This equation allows for the estimation of the relative change in mass of a particular
element lost or gained.
5.2 Identification of an appropriate immobile element tracer
The first step in estimating mass losses and additions is to correct measured
elemental concentrations in the dry soil and bedrock for volatiles, such as organic carbon
and water. The correction factor is determined from the loss on ignition (LOI in Table 1),
which is a measure of how much organic carbon, water and other volatiles present in the
soil but not likely to be derived from the protolith itself. The second step is to identify an
appropriate immobile element tracer. In reality, no elements exhibit true immobility;
however, there are a few that have been proven to be relatively immobile. Ti and Al are
often treated as immobile elements during various weathering processes (April et al.,
1986; Colman, 1982; Cramer and Nesbitt, 1983; Gouveia et al., 1993; Middelburg et al.,
1988; Sak et al., 2003; Taylor and Blum, 1995; Teutsch et al., 1999). Their popularity is
largely due to the fact that measurements ofTi and Al are routine. However, elements
such as Zr, Nb and Ta have also been used (Bain et al., 1993; Brimhall et al., 1991;
Brimhall et al., 1988; Kirkwood and Nesbitt, 1991; Kurtz et al., 2000). What all ofthese 20 elements have in common is that they are relatively high field strength elements (ionic charge/ionic radius) and hence have limited solubility in water. The question is which one of these elements is likely to be the least mobile. AI is actually the most mobile of these elements, consistent with its lower field strength compared to Zr, Ti, Nb and Ta
(Vitousek et al., 1997). Nevertheless, Ti has also been shown to be mobile (Brimhall et al., 1988). Under conditions ofTi and AI mobility, it has been suggested that Nb and Ta may remain relatively immobile (Kurtz et al., 2000).
The behaviors ofNb, Ta, Zr, Hf, AI, and Ti are fairly consistent in the Machame soils. An additional element, Hf, is considered because of its similar geochemical behavior to Zr. Table 3 presents the proto lith to soil concentration ratios, ~~ , ofNb, csoil
Ta, Zr, Hf, Ti, and Al. The values calculated from Nb, Ta, Zr, Hf, and AI suggest a spectrum of modest net mass change(+/- 20 %) above 150 em depth and large losses(>
T:1 40 %) between below 150 em (Ti is a distinct outlier because the C~; values suggest 70- Csoil
75% mass removal above 150 em). The geochemical behavior of these elements can be shown by plotting Ta versus Nb, Hfversus Zr, A}z03 versus Nb, and Ti02 versus Nb in
Figure 4. Also plotted is the protolith reference line corresponding to the Ta/Nb, Hf/Zr,
A}z03/Nb, and Ti/Nb ratio of the proto lith sample. Perfectly immobile elements should plot on the dotted line because while the absolute concentrations of two immobile elements will change due to dilution or concentration effects, their relative abundances should not. It can be seen from Figs. Sa and b that Nb/Ta and Zr/Hf ratios are remarkably constant as evidenced by the fact that the data points fall close to the protolith reference line. This suggests that Nb, Ta, Zr, and Hf are behaving as immobile elements. In 21 contrast, the Al/Nb ratios of the most weathered samples deviate from the protolith reference line (Fig. 4c ). This suggests that AI may have been added since Nb has now been demonstrated to be immobile. Ti /Nb ratios also deviate from the protolith reference line for samples above 200 em, indicating that Ti has been added (Fig. 4d).
Collectively, these observations suggest that Nb, Ta, Zr, and Hf are relatively immobile whereas AI and Ti are not. However, the apparent immobility of Zr and Hf may be partly coincidental. The LOI-corrected Zr and Hf concentrations appear to be slightly depleted
(by about 10%) relative to the proto lith, which suggests that Zr and Hf may have actually suffered small losses (Fig. 5). These observations are consistent with other observations that have shown that Zr (and by inference Hf) may be slightly mobile under intense weathering conditions. For example, during bauxite formation on volcanic rocks in
Southern Brazil, Zr has been shown to be more mobile than Fe and Nb using iso volumetric mass balance methods (Melfi et al., 1996). In all ensuing discussions, Nb and
Ta are identified as relatively immobile element tracers.
5.3 Estimating mass fluxes
Nb and Ta concentrations are used in conjunction with Eq. 3 to estimate the total amount of original proto lith mass remaining in the soil residuum. If there has been significant mass loss, the mass concentrations ofNb and Ta (corrected for LOI) should be greater than that of the protolith. It can be seen from Figure 4 that the LOI-corrected Nb and Ta concentrations are slightly higher (10 %) than that of the protolith in the upper
200 em of the weathering profile and significantly higher (200 %) in the lower 200 em.
These features indicate that there has been mass loss throughout the weathering profile, with most of the loss occurring in the lower half of the profile. What is peculiar is that 22 the LOI-corrected Si02 contents in the upper 200 em appear to be substantially less than that of the proto lith, implying that much greater amounts of dissolution may have
occurred than implied from Nb and Ta concentrations in the upper 200 em. There are three possible explanations for this. The first is that there may have been mass addition
in the upper 200 em, which dilutes the concentration of an immobile element. The
second is that the presumably unweathered proto lith, as pointed out in the Results section,
has probably already undergone some amount of dissolution, which cannot be quantified
if a truly pristine proto lith is unavailable. The third is that the soil protolith is not
homogenous with depth. The first two scenarios imply that the estimates of mass
dissolution using Nb and Ta as immobile element tracers are probably minimum
estimates. All of these scenarios will be discussed in greater detail in section 6. For the
remainder of section 5, an homogenous protolith is assumed.
In Table 1 and in Figure 2, the CIA and the CDF are reported for each soil sample.
The CIA factors are very close to unity throughout the column indicating a high level of
weathering. Even the proto lith sample is 0.80, typical of a weathered bedrock sample.
However, there is a discemable difference between the samples above 183 em, which
have a CIA of0.97, and those below which have a CIA of 1.00. The CIA of the soil
sample at 183 em is 0.98, closer to the more shallow samples than the deeper ones;
however, it may also represent a transition from less to more weathered material. The
CDF, an estimate of net mass loss, indicates that the original protolith underwent a net
~ 10 % loss of mass above 183 em and a net loss of ~50 % below 183 em. The soil
sample at 183 em again shows an intermediate net loss of ~25%. The CIA and CDF both
indicate that weathering throughout the soil column increases with depth. 23
The mass balance calculations can be extended to estimating the mass dissolution
j flux of each element. Figure 5 is a plot of m11 for the major elements, Si02, CaO, MgO, ml 0
Ah03, Na20, P20s, MnO, FeO, K20, and Ti02 and the trace elements Zr, Nb, Eu, Ph, Cu,
U, and Th calculated from Eq. 5. These values generally represent time-averaged losses during pedogenesis that contribute to the dissolved load that eventually makes it to rivers.
Given their weight percentages in the bulk rock, it is clear from Fig. 5 that the time- averaged dissolution flux should be dominated by Si, K, and Na. Interestingly, there appears to be a net addition ofFe in the upper 200 em and a net addition of AI below 200 em, suggesting that Fe and AI and have either been redistributed in the soil column or that excess Fe and AI has been introduced from outside of the soil system. Ti, Fe, and P exhibit somewhat unusual behaviors in that they are all added above 200 em but relatively immobile below. AI is the only major element significantly added below 200 em. The last panel in Fig. 5 shows a sum of all of the major and minor element losses throughout the soil column as well as the Chemical Depletion Factor. While there is a discrepancy between the two estimates of total mass loss, they are well correlated and the absolute deviation between the estimated decreases with depth. 24
6. Possible explanations for the inverted weathering profile
j The CIA, Table 1 and Fig. 2, and the relative mass losses, mfl , in Fig. 5 show m; 0 that there have been huge mass losses throughout the soil column with the more depleted
layers occurring below 200 em. There are at least four hypotheses that may explain some part or this entire phenomenon. Before exploring these hypotheses, the possibility that
lateral movement of material through colluvial processes and landslides has led to the
development of these geochemically stratified zones is ruled out. Below, 200 em, the soil
is extremely weathered, but possesses a rigid, highly porous structure which would most
likely be destroyed with significant lateral movement of solid material. The horizons
above 200 em appeared to be laterally homogenous and without large clasts both of
which suggest that any recent contribution to the soil from colluvial processes is limited
and insignificant.
Surface enrichments of Pb, Cu, and P must also be mentioned. These enrichments
remain unexplained by the four natural, non-anthropic hypotheses presented below.
There is a wealth of research documenting the surface enrichment of Pb, Cu and other
metals through anthropogenic deposition and bio-cycling (Alloway, 1995; Blaser et al.,
2000; Erel et al., 1997; Halamic et al., 2003; Hamelin et al., 1989; Han and Banin, 1995a;
Hansmann and Koppel, 2000; Keller and Domergue, 1996; Lee et al., 1998; Olson and
Skogerboe, 1975; Othman et al., 1997; Teutsch et al., 2001; Wilcke et al., 2001; Wilkens
and Loch, 1998). In agricultural settings, fertilization and manure application can lead to
persistent P enrichments (Carpenter et al., 1998). While atmospheric dust cannot account
for Pb, P, and Cu, exclusively, other anthropic depositions such as fertilizer, manure, and 25 particulates from exhaust and smoke could account for surface enrichments of these
elements.
Each of the four hypotheses will be discussed and explored noting their
plausibility in the context of the data.
6.1 Re-precipitation followed by erosion
One way to achieve higher concentrations of certain major elements at the surface
is to invoke recent erosion (Fig. 7a). This scenario could work as follows. The first step
is the development of a deep, mature soil profile from the top down. Soluble elements
such as Ca, Sr, Na, and Mg would be leached throughout the soil column with the upper
parts experiencing greater depletion. Fe and other particle reactive elements, such as Mn
and Mg, leached from the upper part of the soil column, will migrate downward and re
precipitate onto clay surfaces and/or form amorphous Fe and Mn hydroxides, generating
an enrichment front just beneath the depleted upper layer. If the upper horizons above
the enriched layer are subsequently removed through rapid physical erosion (i.e., faster
than the timescales of leaching and downward migration and re-precipitation of particle
reactive elements), the enriched layer could be exposed at the surface, giving the artifact
of an inverted soil profile. In other words, this hypothesis would require the upper 200
em of the soil profile to represent a paleo-enrichment front.
Re-precipitation may be able to explain the Fe added above 200 em in the
Machame profile (Fig. 5) and may also explain Ti which has a concentration profile (Fig.
2) and relative mass loss profile (Fig. 5) very similar to the respective profiles ofFe.
Like Fe, Ti has been shown to be mobile in highly weathered tropical environments 26
(Cornu et al., 1999). Cornu et al., (1999) showed significant Ti losses through the formation of anatase precipitates as well as general mobility due to plant-soil cycling.
Some re-precipitation processes may also explain the slight Th enrichments observed in the upper 200 em (Fig. 5) as Th may be mobilized in association with dissolved organic
matter or colloids in some marine environments (Guo et al., 1997). Th was found to be
mobilized through dissolution/re-precipitation processes or association with organic
colloids in a tropical watershed in Cameroon (Braun et al., 2005). The presence of
colloids, organic matter, and water provide the conditions necessary for the mobilization
ofTh. Therefore, the enrichment of Fe, Ti, and Thin a zone ofre-precipitation may be
feasible.
Despite these successes, the model simply fails to explain the majority of the
major elements. Ca, Mg, Na, K, and Si are not particle reactive; hence, their higher
abundances in the upper 200 em cannot be explained by this same process. This process
is thus considered untenable in explaining the inverted soil profile.
6.2 Aeolian dust deposition
Atmospheric dust deposition is another hypothetical mechanism that might be
invoked to explain high elemental concentrations above 200 em relative to concentrations
below 200 em (Fig. 7b ). Dust deposition has been suggested to explain the development
of some bauxite formations which, like the Machame soils, have high Ah03 contents
(Brimhall et al., 1988). Marine aerosol deposition has also been shown to be critical to
soil formation and maintaining nutrient budgets on the Hawaiian Islands (Chadwick et al.,
1999). At more than 250 km from the Indian Ocean, it is doubtful that marine aerosols 27 have a significant impact on soils in Mt. Meru; thus, the discussion is restricted to
continentally derived dust. For the Machame region, the dust deposition hypothesis
requires that dust is deposited at the surface of the soil column. The dust must then be
mixed down into the soil column by burrowing animals, roots, clay shrink/swell, and/or
other biological and physical processes (Brimhall et al., 1991 ). The dust may originate
locally or be blown in from great distances (Chadwick et al., 1999; Grousset and Biscaye,
2005). As a result, enrichments in atmospherically deposited elements are superimposed
onto a mature soil profile. A closer look at what elements are enriched and the nature of
their enrichments can help to assess the plausibility of this scenario in explaining the
apparent inverted soil profile.
Firstly, the profiles of the major elements do not seem to be consistent with
surface deposition. Ca, Mg, Si, Pb, P, Fe, and Ti are enriched at the surface but abruptly
decay in concentration below 122 em. If these surface enrichments are associated with
external (Aeolian) inputs, then the downward depth distribution of these elements would
be expected to decay monotonically with depth due to downward transport by
bioturbation (Brimhall et al., 1991). The step function stop in concentration of the above
mentioned elements is inconsistent with an external surface input. The only element to
show a monotonic decay with depth is Cu.
The applicability of the Aeolian hypothesis can be further assessed by examining
the magnitude and composition of the elemental enrichments at the surface. If the
surface enrichments above ~200 em are caused by dust inputs, the composition of the soil
below 200 em should be the baseline composition of the profile. A minimum percentage
of the soil column that originated as dust by summing all elemental mass enrichments 28
above the baseline concentrations can then be calculated. At least 38% of the total soil
dry mass down to 200 em must be dust-derived in order to account for the concentration profiles of the major elements. The chemical composition of the dust can also be
calculated resulting in the following major element composition: 66% Si02, 23% FeO,
7% Ti02, 2% K20, 1% MgO, 1% CaO, 1% Na20, and 0% Ah03. Dust of this
composition most likely represents a mixture of quartz, iron oxides (magnetite, spinel),
and Ti-bearing oxides (rutile and ilmenite), implying an origin from felsic continental
crust. Such material would be expected to be enriched in the high field strength elements
due to the presence ofTi-bearing oxides (which could be used to explain the high Ti in the upper layer). If this material were a significant component of the upper layer of the
Machame soil profile, the high field strength element systematics would be controlled to
a large extent by the dust. This dust, however, would be characterized by low Nb/Ta
ratios. This is because the source of this dust, upper continental crust, typically has
Nb/Ta ratios between 10-12 (Rudnick and Fountain, 1995), much lower than most
intraplate volcanic rocks, such as the Machame proto lith. In fact, the hypothetical dust
should have even lower Nb/Ta ratios because it is known that rutile, the dominant
repository for Nb and Ta in silicate rocks, prefers Ta over Nb and hence typically has a
low Nb/Ta ratio compared to the host rock (Klemme et al., 2004). The fact that the
Nb/Ta ratios of the Machame soils are high (~ 20, a typical value for mantle-derived rocks)
and constant throughout the soil column implies that dust of the required composition did
not fall in amounts sufficient to generate the enrichments seen in the upper layer.
Furthermore, from Fig. 3c it is clear that the soil/chrondrite REE ratios for the shallow
Machame soils do not exhibit the behavior expected with a significant contribution of 29
continental crust because there is no negative Eu anomaly as well as high soil/chondrite ratios (Kurtz et al., 2001). Thus, dust deposition cannot adequately reconcile the major and trace element data and is therefore an unreasonable hypothesis to explain the inverted profile.
6.3 Buried Paleosol
The two chemically distinct soil layers may represent two soils of different ages
formed from two different lava flows (Fig. 7c). In this scenario, a soil, now represented
as the Machame soil below 200 em, formed on a lava flow-chemically identical to the
material sampled as the Machame protolith. After this soil develops and is extensively
weathered, it is covered by a subsequent lava flow. The overlying lava flow then
undergoes weathering itself and forms a second soil horizon, now represented as the soil
above 200 em. Assuming similar weathering rates for the development of the relic soil
and the modem soil and given the significantly more depleted horizons below 200 em,
this model requires that the relic soil was directly exposed to the atmosphere much longer
than the modem soil has been.
This model could, in theory, explain why the upper layer is less weathered than
the deeper layer. It can also explain the behavior ofEu relative to the other REE's. The
Machame protolith is rich in plagioclase phenocrysts. These plagioclase phenocrysts
have preferentially partitioned Eu into their mineral lattices over other REEs, a process
which left the complementary glassy matrix poor in Eu. Because glass lacks crystalline
structure, the glassy matrix is more susceptible to chemical weathering than the
plagioclase phenocrysts. According to this two lava hypothesis, the relic soil below 30
200cm was weathered more intensely, destroying nearly all primary minerals, including plagioclase. This would explain the positive Eu anomaly above 200 em and lack of an
anomaly below 200 em (Fig. 2).
Some characteristics of the soil, however, are inconsistent with a buried paleosol
model. First, the soil profile lacks physical evidence of a bake zone in the field. If the
soil below 200 em did represent an older relic soil, a bake zone would be expected to
form between the soils. This bake zone would have probably been highly resistant to
weathering and should have been preserved. Second, the depth behavior of chemical
weathering, manifest in terms of CIA and CDF, is inconsistent with a buried paleosol. If
the change in composition at ~200 em indeed represents a transition from a younger, less
weathered soil and to an older paleosol, a decrease in the degree of weathering as this
boundary is approached from above should be observed. Conversely, as the transition is
approached from below, an increase in the degree of weathering should be observed.
The constant weathering degrees in the upper and lower layers may be inconsistent with
this hypothesis. However, the buried paleosol model is consistent with the bulk of the
geochemical data including Al and is, therefore, a viable mechanism for soil formation in
the Machame region.
6.4 Lateral subsurface water flow
A fourth hypothesis involves enhanced weathering due to lateral subsurface water
flow, (Fig. 7d). Meteoric waters enter the subsurface at higher elevations. This
subsurface water then flows from its recharge region down the flanks of the mountain,
the flow being driven by the hydraulic head imposed by mountain topography. This 31 model yields the following predictions. Progressive reaction of a package of groundwater along its flow path should lead to 1) consumption of oxygen due to
oxidative weathering (i.e., 0.5 02 + 2 FeO = Fe203) and 2) consumption of C02 due to 2 acidic breakdown of silicate (e.g., C02 + H20 + CaSi03 = HC03- + Ca + + HC03-). As
such, weathering by subsurface water flows would be large only near recharge point but
small further along the flow path. However, in volcanically active regions, there is likely
a constant background flux of volcanic gasses permeating through the volcano (Aiuppa et
al., 2000; Cruz et al., 1999a). Such gases will be reducing but rich in C02, and as a
consequence, flux of such gasses into the subsurface water could increase acidity but not
lead to an increase in 0 2. Thus, in volcanically active regions, subsurface groundwater
should be acidic and reducing
If this model is applicable to the study site, the implication is that the entire soil
profile has formed from the same proto lith as a result of stratification in weathering
processes. The soil above 200 em was formed from typical top-down weathering while
the soil below 200 em was created contemporaneously by subsurface water flow, the two
weathering regimes being distinguished by their different weathering rates (below 200
em is faster) and possibly by different redox conditions. The former is suggested by the
step function distribution of Ca, Mg, Na, and Si (Fig. 2) and the shallow Eu anomaly (Fig.
3a). The latter is suggested by the contrasting behaviors ofU and Thin the two layers
(Fig. 6). Th appears to be enriched in the upper 200 em (i.e., high Th!Nb) and depleted
below 200 em, indicating that Th is relatively immobile in the top layer but mobile in the
bottom layer. U appears to be enriched relative to Nb at all depths, but the enrichment in
U is greatest in the bottom layer. The U/Nb systematics indicates that U has been added 32 at all depths, particularly in the bottom layer. These features suggest that the bottom
layer may have been somewhat reducing and the top layer more oxidizing. This redox
stratification would allow forTh mobility and preferential immobilization ofU in the more reducing bottom layer and Th immobility and variable behavior ofU in the more
oxygenated upper layer (Langmuir, 1997). Focused subsurface water flowing through the bottom layer would be consistent with these observations. In fact, the net additions of
U throughout the soil column, particularly in the bottom layer, are probably not solely
sustainable by leaching ofU from the surface followed by re-precipitation at depth. In
this model, net additions ofU are due to immobilization of dissolved U in subsurface
water being laterally advected through the system. The upper 200 em corresponds to the
unsaturated vadose zone, which would be relatively oxygenated. Therefore, the
subsurface flow model is consistent with the major and trace geochemical data, except AI,
and with the physical data and thus remains a viable hypothesis to explain the formation
of the Machame soils. 33
7. Implications of lateral subsurface water flow
Based on the discussion in Section 6, the data appear to be best explained by the buried paleosol scenario or the subsurface water flow model-with the caveat that Fe and
Ti behavior may be the result of mobility due to re-precipitation or association with
colloids. While neither hypothesis can be ruled out, the subsurface water flow model
deserves particular attention because it can explain the unusual U/Th systematics. If this
bias is correct, there are some novel implications of lateral subsurface water flow. This
case study would suggest that groundwater weathering could increase local weathering
rates by at least a factor of two (Table 3). This means that if only typical top-down soil
formation processes are considered, weathering rates could be under-estimated. The
importance of subsurface water weathering may not be confined to this case study. Chen
et al. (1988) concluded that bauxites in the Tatun volcanic region of northern Taiwan
were also formed by enhanced weathering associated with groundwater flow (Chen et al.,
1988). In south Asia, Basu et al. (2001) showed that groundwater may be as important as
surface water from the Ganges-Brahmaputra river system in delivering dissolved Sr to the
oceans (Basu et al., 2001). Finally, Manga (1998) has shown that there is extensive
groundwater flow in the Oregon Cascades arc (Manga, 1998). What all of these regions
have in common is a large and steady recharge zone of meteoric water and a strong
topographic gradient to drive groundwater flow. These regions are also subject to
volcanic degassing or metamorphic de-carbonation reactions.
If weathering by subsurface water flow turns out to be significant in volcanic or
tectonically active regions, this phenomenon is not only important for soil formation and
ocean trace element budgets, but possibly for the global carbon cycle. The weathering of 34 silicate rocks is a slow, but permanent means of sequestering carbon (Berner et al.,
1983a). The data suggests that weathering associated with subsurface water flow is 30% more effective than top down soil formation processes at mobilizing Ca. For Mg, the subsurface water is 90 % more effective at this site than top down processes. It is possible that the enhanced weathering by subsurface water flow in this study site is facilitated by the addition of volcanic C02 slowly seeping out of the mountainside.
Assuming that Ca and Mg are the primary cations responsible for silicate COz
sequestration, this study suggests that estimates of the COz drawdown capacity of silicate
weathering, at least in volcanically active regions, may need to account for weathering by
subsurface water flow. If so, weathering by subsurface water flow may be a means of
modulating the amount of COz emitted from volcanic environments. 35
8. Conclusions
The following conclusions were made after investigating an inverted weathering profile formed on a volcanic substrate on the southern slopes of Mount Kilimanjaro,
Tanzania, a region having a tropical equatorial climate:
1) The degree of chemical weathering increases with depth into the soil profile.
2) The depletion of mobile major elements, Si, Na, K, Ca, and Mg, and the relative
immobility of AI have created soils of 30-40 wt% Ah03 at depths greater than
200 em.
3) The refractory elements Nb, Ta, Zr, and Hf are shown to be immobile throughout
the soil column while Ti and AI are not. Ti and Fe are enriched above 200 em
and immobile below while AI appears to be slightly depleted at shallow depths
but enriched at greater depths.
4) All geochemical and physical evidence suggests the presence of two distinct
weathering regimes separated at 200 em.
5) Two models can explain the bulk of the chemical and physical observations:
either a buried paleosol or intense weathering associated with lateral subsurface
water flow.
6) If subsurface water flow is responsible for the formation of the Machame soils,
inputs of trace elements and Ca and Mg into the ocean budget are locally
significant, and, if regionally representative, they should have implications for the
global C02 cycle. 36
Table 1. LOI-corrected major element oxide concentrations
Wt% Scm 30 91 122 183 244 305 366 Protolith LOI 33.25 31.84 29.94 31.38 32.96 28.34 27.67 28.63 7.95 Na20 0.38 0.35 0.38 0.35 0.27 0.05 0.11 0.12 4.42 MgO 0.76 0.67 0.73 0.66 0.67 0.30 0.33 0.26 1.11 Ah03 30.17 30.08 31.16 30.07 41.20 65.03 67.09 79.59 25.90 Si02 * 30.51 30.31 30.37 30.32** 24.45 5.38** 3.53 4.44 56.78 P20s 1.09 1.06 1.09 1.00 0.80 0.52 0.63 0.72 0.36 K20 0.78 0.72 0.74 0.73 0.55 0.09 0.13 0.09 7.17 CaO 0.59 0.51 0.60 0.50 0.37 0.03 0.03 0.04 1.89 Ti02 4.32 3.99 4.27 3.83 3.62 2.18 2.78 2.46 1.31 MnO 0.18 0.17 0.18 0.16 0.19 0.24 0.32 0.34 0.25 FeO 17.49 16.97 17.96 16.79 16.22 12.11 15.20 14.46 7.29 TOTAL 86.26 84.82 87.47 84.40 88.35 85.93 90.15 102.51 106.49
All values determined using medium resolution ICP-MS unless indicated *Values determined by XRF **Values interpolated due to problems in XRF 37
Table 2. Minor and trace element concentrations by depth
~~m Scm 30 91 122 183 244 305 366 Protolith Li 15.9 16.1 17.9 16.8 25.0 45.6 63.2 56.2 54.1 Be 2.00 1.94 1.99 2.00 1.95 1.49 1.17 1.36 13.0 Sc ~ 12.7 13.0 13.7 12.6 15.4 11.3 13.8 15.2 8.62 v* 173 170 178 167 141 9.51 6.44 4.45 2.98 Cr* 138 127 136 129 103 5.71 4.50 3.66 3.47 Co ~ 11.5 10.7 11.5 9.98 9.13 5.32 5.11 9.32 4.34 Ni * 22.9 22.3 23.3 22.2 19.2 33.2 10.4 2.20 0.52 cut 307 99.3 38.6 31.4 18.7 10.4 11.5 11.0 9.43 Zn * 186 137 116 110 111 164 207 197 288 Ga 42.3 41.8 45.5 45.4 43.7 61.1 66.7 71.9 41.9 Rb 23.79 22.1 23.0 23.7 15.6 4.03 5.77 4.40 211 Sr 128 119 136 127 75.4 10.5 5.19 3.51 128 y 18.7 17.5 19.1 19.4 13.2 8.23 8.66 8.63 64.87 Zr* 1420 1520 1520 1390 1900 2850 3570 3480 2200 Nbt 398 410 446 399 480 678 855 796 517 Cs 2.11 1.98 2.11 2.12 1.41 0.130 0.153 0.128 2.14 Ba 274 282 296 282 185 183 60.1 36.2 705 La 51.7 49.3 54.5 57.5 42.6 90.3 86.9 69.5 228 Ce 240 219 243 241 293 345 332 425 394 Pr 11.0 10.5 11.6 11.6 8.22 9.33 9.02 7.69 37.7 Nd 41.6 37.4 40.0 41.4 29.7 22.3 20.9 18.6 108 Sm 7.96 7.50 8.06 7.86 5.93 2.84 2.74 2.64 17.6 Eu 2.23 2.12 2.27 2.27 1.58 0.54 0.49 0.49 3.69 Tb 0.927 0.882 0.951 0.960 0.682 0.397 0.385 0.380 2.19 Gd 6.24 5.87 6.29 6.39 4.45 2.24 2.13 2.05 14.3 Dy 4.80 4.57 4.90 4.91 3.55 2.15 2.13 2.09 11.8 Ho 0.807 0.765 0.817 0.823 0.602 0.379 0.395 0.382 2.12 Er 2.20 2.09 2.21 2.24 1.71 1.15 1.23 1.21 6.28 Tm 0.323 0.311 0.330 0.328 0.269 0.187 0.203 0.195 0.981 Yb 2.16 2.08 2.16 2.16 1.92 1.27 1.36 1.34 6.59 Lu 0.277 0.266 0.277 0.277 0.249 0.162 0.170 0.171 0.943 Hf 24.1 24.0 24.8 24.6 30.4 49.7 58.1 50.7 34.0 Ta 21.8 22.5 24.7 23.1 25.5 40.1 50.1 41.4 26.1 Tl 0.218 0.217 0.226 0.229 0.156 0.0376 0.0283 0.0367 0.114 Pb 52.4 50.7 52.3 52.3 44.7 21.3 24.5 19.6 ll5.0 Th 52.4 51.8 55.7 53.9 56.9 74.8 87.6 77.5 56.4 u 9.50 9.43 10.0 9.76 10.8 23.9 21.7 22.4 9.09
All values determined using low resolution ICP-MS unless indicated t Values determined using medium resolution ICP-MS 38
Table 3. Protolith to soil ratios, Co I Csoib forTi, Al, Zr, Hf, Nb, and Ta
Co I Csoil Scm 30 91 122 183 244 305 366 Protolith Ti 0.30 0.33 0.31 0.34 0.36 0.60 0.47 0.53 AI 0.86 0.86 0.83 0.86 0.63 0.40 0.39 0.33 Zr 1.12 1.07 1.10 1.18 0.84 0.60 0.48 0.49 Hf 1.02 1.05 1.04 1.03 0.81 0.53 0.46 0.52 Nb 0.94 0.93 0.88 0.96 0.78 0.59 0.47 0.50 Ta 0.87 0.86 0.80 0.84 0.75 0.51 0.41 0.49 CIAt 0.97 0.97 0.97 0.97 0.98 1.00 1.00 1.00 0.80 CDFtt 0.10 0.11 0.16 0.10 0.24 0.45 0.56 0.50
All data corrected for LO I t Chemical Index of Alteration, CIA= Al20 3 (Nesbitt and Young, 1984) A/20 3 + CaO + Na20 tt Chemical Depletion Factor, averaged Nb and Ta, CDF = 1- Xrock (Riebe et al., 2003) X soil 39
Figure Captions
Figure 1. Road cut in the Machame region of Tanzania on the southern slopes ofMt.
Kilimanjaro. The gray circles indicate the approximate depths at which samples were taken.
Figure 2. The loss on ignition (LOI), LOI-corrected element concentrations (Si02, CaO,
MgO, Ah03, Na20, K20, FeO, Ti02, Zr, Hf, Ta, Nb, Yb, Ph, Cu, U, and Th), Chemical
Depletion Factor (CDF), and Chemical Index of Alteration (CIA) are plotted by depth.
Also plotted as a dotted line or marked symbol are the LOI, element concentrations, and
CIA of the proto lith. The Chemical Index of Alteration is defined as
CIA= Al20 3 (Nesbitt and Young, 1984). The Chemical Depletion Factor Al20 3 + CaO + Na20
is averaged for Nb and Ta, where CDF =1- X rock (Riebe et al., 2003). X soil
Figure 3. Three plots ofREE concentration ratios (La, Ce, Pr, Nd, Sm, Eu, Tb, Gd, Dy,
Ho, Er, Tm, Yb, and Lu). Soil/protolith REE concentrations are plotted for plotted
samples above 200 em ( o ), and samples below 200 em ( •) in ppm in (A). Also plotted
are soil/upper continental crust (B) and soil/chondrite (C) concentration ratios for the
REE's. In (B) and (C), ratios are also plotted for the protolith sample as(*).
Figure 4. Highly refractive elements that are commonly used for mass balance
calculations are plotted against each other. Concentrations ofTa v. Nb (A), Hfv. Zr (B),
Ah03 v. Nb (C), and Ti02 v. Nb (D) plotted for bedrock (*), samples above 200 em ( o ), 40 and samples below 200 em ( •) in ppm. The dotted line from the origin through the bedrock data is the evolution path available to ideal, immobile elements. Samples that plot upslope from the bedrock datum have a net mass loss; samples that plot down-slope from the bedrock datum have a net mass gain. The slopes of each solid origin-protolith
line and the squares of the correlation coefficients (r2) between each line and the soil
sample data are reported. The dotted lines represent a linear best fit for the data above
200 em in (C) and below 200 em in (D) for comparison to the other data.
j Figure 5. The relative mass change, mA , is plotted for Si02, CaO, MgO, Ab03, Na20, m6
K20, P20s, MnO, FeO, Ti02, Zr, Nb, Eu, Pb, Cu, U, and Th from Eq. 5. Also plotted are
two estimated of the total mass loss during pedogenesis. The solid line is the sum of all
1 1 1 "L..,.mA- -"L....-t- c~ * c sot!- co mass differences calculated using Eq. 5, Csott , and the dashed line
is the Chemical Depletion Factor (CDF). Discrepancies between these two lines
represent measured mass losses unaccounted for by the immobile element ratios.
Figure 6. U/Nb and Th/Nb ratios plotted for bedrock ( +), samples above 200 em ( o ), and
samples below 200 em ( • ).
Figure 7. Schematic models of four hypotheses to explain the formation of the Machame
soil profile. Hypotheses A describes re-precipitation of particle reactive elements
followed by latter-day surface erosion. Hypothesis B requires the addition of many
elements by Aeolian dust deposition. Hypothesis C proposes that the soil was formed 41 from two different lava layers. Hypothesis D calls for enhanced weathering due to lateral subsurface water flow. An inset for the subsurface flow model that shows a potential recharge mechanism is included 42
FIGURE 1 43
FIGURE2
LOI (%) Si02 (%) CaO(%) MgO (%l) 20 40 60 !LO 0.5 Hl 1.5 2.0 0.0 0.5 1.0· 1.5 ·-~~·~~~-~~4 .----~~----
100
•
Na20 (%) K20 (%) 4 6 a
.. I
&
P205(%) FeO (%) 0.5 1.0 5 10 15 20
I 1ooj 44
FIGURE 2, continued
Ti02 (%) Zr (ppm) Nb (ppm) Eu (ppm) [) 2 4 II 0 2000 4000 500 1000 1500 0 2 4 8 0 .. .. • ,. • • ' ...... • • ,...... ~·· ...... ~ If • ... • ..
Yb (ppm) Pb (ppm) Cu (ppm) U (ppm) 0 2 4 8 8 0 50 100 0 200 400 800 0 10 20 30 110 0 • • ... 100 + + .. •• ...• .. --~ ...... =200 it .. .. 0 • 300 • ~· .. • g 400 .Jt... Th (ppm) CDF (Nb,Ta) CIA 0 50 100 150 0.0 0.2 CIA 0.6 0.75 Cl85 0.95 1.06 0 •
100 ...... • .ol • """' .[200 .. .. • .ol .. 0i 300 ... .. • ...... rD. 400 45
FIGURE3
REE Soii/Protolith Ratios 1.6 A o Above 200 em 1.2 • Below 200 em u0 )K Protolith ·a 0.8 -II) u 0.4
0 La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
REE Soii/UCC Ratios 10
8 0 0 ::::> (.) 6 ·a -II) 4 (.)
2
0 La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
REE Soii/Chondrite Ratios 10000 c 1000 :~!!! -c... c .r::0 100 u() -2 10 u-
1 La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu 46
FIGURE4
80 100 ...... --- 0 Soil < 200 :em • Soil> 200 em /./ X Prokllilti 80 so - linear (Prota11h) I Linear{~! >21!~~ • / 60 / ! jV/ ! 40
,./ 20
/ A B,.,,,:_j 0 ~------~ 1500 1j}00 1500 0 2.000 4000 6000 Nb (ppm) Zr (ppm)
so~------e-----~
• 4 60 0 ;Q 0 -·("') 0 40 £j <(
2:0 c 0 0-F-----.- 0 500 1000 1500 () 500 1000 1500 Nb (ppm) Nb {ppm) 47
FIGURES
Si02 CaO MgO -1.0 -0.8 ..().6 -0..4 ~.2 0.0 -1.0 -0.8 -1:1.8 .Q.ol -0.2 CHJ -1.0 -0.8 -0.6 -OA .Q.2 0.0 ····-···--··---··-·-·-· .---·~· 0 ~----~y-- • • 1
. I. r..;_..~.t-~. I i\1203 Na20 K20 0.0 0.2 0.
t
P205 .o.,;. O.G• ~::J.·5 '1.0 t.5 2.0 0
7lI ; 100 ... E .. g :6200 c.. c:,l!ll· •
3!JC; 4 • •oc· c. 48
FIGURE 5, continued
Ti02 Zr Nb -I 0 2 -04 -0.2 G.O 0.2 -0.4 -G.2 O.G 0.2 0 • • ...... 100 •• • • I .. .. :[ 200 • " Q) ... 0 • .. 300 ... • • ...... 400
Eu Pb Cu -1.0 -0.8 -0.8 -0.4 -0.2 0.0 -1 0 1 2. 3 4 0 111 20 30 4() Or • ;. • 100 • • • • • • ._,~ ,G200 • • • c. Q) 1 0 ... t liOO ' ... • ...... • • 400
u Th Total Mass Loss 0.0 02 0.4 0.6 0.8 1.0 -0.2 0.0 0.2 -1.11 -0.8 -0.6 -G.4 -0.2 0.0 0
• • I
100 ' • ... • ' ...... • I I ' K ... :[200 • Q) 0 " • /CDF 300 •.. .. I • ... 49
FIGURE6
U/Nb VS. Th/Nb 0.04 ~ ' +U 4t' +U -Th . +Th l 0.03
• 1 ..Q o 6' oo z j :s 0.02 ------*--' ------
)K Proto~th 0.01 · o Above 200 em • Below 200 em -U +Th 0.00 ' 0.05 0.10 0.15 Th/Nb 50
FIGURE7
Re ... precipitation followed erosi,on (A) (adapted from Nesbitt et al., 1980 and Brim haM et aL, 1991)
-Modern Surface
1, Soluble Elements -200 em Mobilized (Alkali/Alkaline Metals, etc.)
Bedrock
Aeolian dust deposition (B) (adapted from Brimhall et aL. 1991)
. ' 2. Aeolian Dust , Deposition "'
Elements 51
FIGURE 7, continued
Two lava layers (C)
-Modern Surface
-200 em
Bedrock
Groundwater weathering (D) 52
PART II:
Physical and chemical weathering in mountainous regions: insights from a model
linking chemical weathering to soil formation, creep and mass wasting
Submitted to American Journal of Science in the winter of
2006 and authored by Mark Gabriel Little and Cin-ty A Lee. 53
Physical and chemical weathering in mountainous regions: insights from a model
linking chemical weathering to soil formation, creep and mass wasting
Synopsis
The magnitude and nature of chemical weathering is fundamental to understanding C02 drawdown and the marine budget of elements of paleo-oceanographic significance. The relationship between chemical weathering and erosion is unclear because chemical weathering, occurring during soil formation on hillslopes, is a quasi-continuous process but erosion can be episodic, especially in rapidly uplifting regions or regions of high relief where mass wasting undoubtedly dominates physical erosion. Here, the link between chemical weathering associated with soil formation and erosion associated with mass wasting is investigated. Landform evolution in regions ofhigh topographic relief is assumed from the outset to be controlled by a cyclic process in which a soil mantle gradually forms on a hillslope, but after reaching a critical thickness such that the basal shear stress exceeds the Coulomb-failure criterion, mass wasting occurs. Averaged over
>10 kyrs, the statistical periodicity (0.5-10 kyr for steep slopes) of mass wasting is controlled by the rate of soil formation and the slope angle. These results are combined with simple, empirically constrained dissolution models to predict maximum and minimum bounds on the ratio of suspended/dissolved load in rivers. The predicted ratios, however, are all higher than observed in rivers, the discrepancy worsening with increasing topographic relief. This discrepancy arises from the fact that in regions of high relief, physical erosion rates (mass wasting) are so high that soil mantles do not reside on hillslopes long enough to allow for significant chemical weathering. This 54 conclusion is at odds with the view that tectonic uplift, erosion and chemical weathering are intimately linked. The discrepancy between the model and observations can be explained in several ways: 1) river water measurements are typically made in alluvial/deltaic regions where sedimentation is occurring, hence the suspended load predicted in the model is over-estimated, 2) chemical weathering of deltaic/alluvial sediments was not accounted for by the model, and 3) chemical weathering associated with groundwater weathering was not included. 55
1. Introduction
Quantifying the magnitude and nature of chemical weathering of the continental crust is of fundamental importance for understanding drawdown rates of atmospheric
C02 (Berner et al., 1983b; Raymo and Ruddiman, 1992; Walker et al., 1981) and the global marine budget of various elements, such as those that serve as nutrients for biological productivity or various paleo-oceanographic tracers (McCauley and DePaolo,
1997; Peucker-Ehrenbrink and Ravizza, 2000). Chemical weathering of the continental
crust can occur on land or in rivers. In situ processes on land, such as soil formation,
contribute dissolved material to rivers via surface runoff and ground water. Material physically removed from continents through erosive processes, such as downhill creep
and mass wasting, can contribute to the oceanic budget of dissolved material if it is
chemically weathered in rivers as suspended load or during temporary storage as bedload.
The sum of in situ (during soil formation) and river dissolution (during suspension in
rivers or as temporary storage in river bedload) processes is the source term for dissolved
material at the mouths of the rivers of the world and which enters the oceans (fig. 1A).
What are the relative contributions of erosion-induced chemical weathering and in
situ chemical weathering? Intuitively, in situ chemical weathering might be expected to
dominate in geomorphically stable regions, such as continental shields, but what about
tectonically active regions and regions of high topographic relief? In very tectonically
active regions, the dominant form of physical erosion or denudation is mass wasting (for
example, landslides or debris flows) rather than downhill creep (Sadler and Morton,
1989), the latter which dominates regions of subdued topography. Examples where
mass-wasting clearly controls hillslope erosion are seen in mountainous regions 56
extending from the less vegetated slopes at very northerly or southerly latitudes to highly vegetated slopes in more equatorial regions. An example of the latter from the southern
Sierra Madre cloud forests in Chiapas, Mexico is shown in figure 1B. Thus, mass
wasting probably plays an important role in the physical and chemical weathering of
continents
If regions undergoing rapid tectonic uplift are characterized by rapid denudation
rates, intuition again suggests that the total reactive surface area of rocks and sediments
should be rapidly increased, in tum leading to enhanced chemical weathering rates. In
other words, physical and chemical weathering should be correlated. The increase in
marine 87 Sr/86Sr around 40 Ma ago has been interpreted to reflect an increase in the rates
of silicate weathering (and atmospheric C02 composition) due to uplift oftbe Himalayas
(Raymo and Ruddiman, 1992). Indeed, a log-log linear relationship between suspended
load, attributed to physical weathering and suspended load, attributed to chemical
weathering, has been suggested by geochemical studies of major rivers (Gaillardet et al.,
1999). The implication is that increased rates of physical denudation cause a
corresponding increase of dissolved load in rivers.
More recently, however, it has been shown that the solute content of streams
draining rapidly uplifting regions are dominated by dissolution of easily soluble calcite
veins, which make up less than 1% of exposed metamorphic rocks, rather than
dissolution of silicates (Jacobson et al., 2002). Unlike silicate dissolution, calcite
dissolution (followed by precipitation of carbonate in the oceans) does not result in a net
drawdown of atmospheric C02. These observations suggest that the silicate weathering
directly associated with uplift and erosion may be smaller than previously thought, 57 inconsistent with the log-log linear relationship between physical and chemical weathering mentioned above.
Here, a simple model is developed, which attempts to link chemical weathering, soil formation and mass wasting in regions of high topographic relief. The function of this model is to define a relationship between riverine solute flux (chemical weathering) and suspended (un-dissolved) flux. We focus on the specific case of regions ofhigh topographic relief where we believe that mass wasting is the dominant process of physical weathering. The ratio of solute to suspended load predicted by this model is lower than that actually observed in the world's major rivers. The implications of this discrepancy are discussed in detail. 58
2. Bedrock to soH co111version
Constitutive Equations
Soil creep.-
Downhill creep, the continuous transport of soil down topographic gradients has been modeled as a diffusive process (Dietrich et al., 2003). Only the soil mantle participates in creep, that is, the bedrock is strong enough that it does not undergo downhill creep until after it is converted to soil. Downhill creep in the soil mantle is the net result of discrete, smalllengthscale processes acting in concert to agitate soil particles.
These processes include bioturbation and other processes such as the contraction and expansion of the soil due to freezing, thawing and absorption of water, where the volume
3 1 1 flux q0 of soil per unit elevation contour length (m m- yr- ) is given by
(1) where dy/dx is the local slope (assumed to be negative),y is the local elevation, xis the
2 1 distance from the crest of the hill and k is the diffusivity of the soil (m y{ ), which represents a measure of how fast soil particles can move as the net result of the different processes that agitate soil particles (fig. 1C).
Equation (1) describes downhill material transport as a diffusive process. Values
2 1 reported fork fall in the range of 49±37 cm y{ (Dietrich et al., 2003; Reneau, 1988).
Dietrich et al. (2003) assert that this linear relationship is valid primarily for slopes less than 20 % ( ~ 11 °). For steeper slopes, a non-linear relationship with slop~: has been suggested (Roering et al., 2001) due to small-scale mass wasting, which is a more efficient transport mechanism than creep. Because mass wasting will be modeled explicitly in the next section, the downhill transport of soil on steep slopes is broken 59 down into two components. The first is downhill creep .. 1ssumed to follow a linear diffusive law. The second is mass wasting, assumed to occur only when the soil mantle reaches a critical thickness such that the basal shear stress at the soil-bedrock interface exceeds the strength of the soil. Avcraged over the recurrence rates of mass wasting episodes, the combination of mass wasting and linear soil creep will result in an overall non-linear transport behavior.
Soil production.-
The rate at which bedrock is converted to soil is also controlled by a combination of physical and chemilcal processes. As in downhill creep, the physical processes of soil formation are the sum of many discrete processes, such as the development of cracks and the break up of rocks by freeze-thaw, release of overburden pressures, burrowing of organisms and the infiltration ofplant roots (Berner, 1991; Gabet and Reichman, 2003).
These discrete physical processes in turn increase the reactive surface area allowing for chemical weathering of the bedrock to occur. Collectively, all of these processes result in the net conversion of bedrock to soil. First, the assumption is made that on large enough length scales, the efficiency of soil formation decreases on average with depth. More precisely, the rate of bedrock to soil conversion decreases exponentially with the thickness of the soil mantle as demonstrated by cosmogenic nuclide studies of steady state soil mantles on eonvex hillslopes (Heimsath et al., 2002). This is a reasonable relationship because it is likely that the density of roots and animal burrows decreases with depth. In addition, the effects of freeze/thaw, clay expansion, etc. in creating or widening cracks in the bedrock are likely to be suppressed beneath thick soil mantles 60 because the amount of water infiltrating to the bedrock decreases with increasing thickness of the soil mantle. The exponential decrease in the soil production takes on the following relationship (Heimsath et al., 1997),
dH -Ha -=se (2) dt 0
1 1 where H (m) is the soil thickness, dH/dt (m yf ) is the rate of soil production, &0 (m yf )
is the maximum soil production rate (that occurs when soil thickness His zero) and a (m·
1 ) is the inverse of the characteristic lengthscale of soil production. Undoubtedly, these
parameters will vary according to rock type, climate, vegetation density, etc. However,
recent field studies show that reasonable maximum production rates &0 range from 52 to
2078 m/My with characteristic lengthscales (Va) of~ 0.5 m (Dietrich et al., 2003).
Conservation ofMass
These constitutive equations can now be incorporated into a mass conservation
equation. This results in an expression for the change in mass per unit time and surface
area ofhillslope:
(3a)
3 where Psis the dry density of the soil (kg/m ), dH!dt is the rate at which the soil mantle
thickens, cfy!dx2 is the curvature of the hillslope and all other symbols as previously
defined. Equation (3a) can of course be further simplified to
(3b) 61
The first term on the right represents the soil production term. The second term on the right represents the material transport term due to soil creep. For a convex hillslope cfy!dx2 is negative and creep tends to thin the soil mantle. For a concave hillslope cfy!dx2 is positive and creep tends to thicken the soil.
If the goal is to investigate long-term evolution oflandscape form (for example, elevation changes), equation (3b) would now be re-expressed in terms of the elevation variable y. However, because the eventual goal is to investigate the relative contributions to the solute flux by soil formation and mass wasting, only timescales long enough to average out several cycles of soil formation and mass wasting are considered. The entire life of an eroding mountain is not of concern. For this reason, the landscape form is assumed to be constant on these time scales. This means that landscape form, or y(x) can be treated as roughly constant (over a few cycles), from which it follows that the transport term is also constant.
Equation (3b) is a nonlinear differential equation. Thus, before solving equation
(3b) explicitly, some simple end member cases are considered to illustrate the equation's meaning. For example, on the convex portion of a hill (including the hill divide), the soil mantle will reach a steady-state thickness due to the competing effects of soil production and the thinning effects of downhill creep on convex hillslopes. The steady-state thickness when dH/dt = 0 is given by
2 . 1 ( k d yJ H(oo) = --ln --·---2 (4) a & 0 dx which means that the steady-state soil mantle thickness can be determined from the local curvature of the hillslope and the constants of soil production D and soil diffusivity k.
However, for concave hillslopes (cfy!dx2>0), no steady-state exists because downhill 62
creep results in the progressive thickening of the soil mantle (note that the quantity in parentheses becomes negative and hence undefined). Thus, the transition between
steady-state and non-steady-state conditions at long times must occur at the inflection
2 point of a hillslope landform, that is where the curvature, cfy!dx , is zero. If curvature is
zero, soil mantle thickness grows logarithmically with time:
(5)
This equation describes the thickening of soil on a perfectly horizontal surface
(dyldx=cFy/dx2=0) or at the inflection point of a hill (where downhill creep in and out of
a unit volume are equal). This equation also approximates long, steep slopes, where
curvature is confined primarily to the crests of ridges or hills (this approximation will be
used in the next section when steep hillslopes are considered).
The special case in which the soil production term is zero is considered below. In
this case, the solution to equation (3b) is as follows
(6)
where H0 is the initial soil mantle thickness. In other words, in the absence of soil
production, a soil mantle on a convex hillslope (that is, cfy!dx2 < 0) will eventually thin
linearly to zero (on a concave hillslope, it will thicken due to accumulation). The more
general solution to equation (3b) is given by
H(t) ln[l-(enht (1; +beaH•)-e )] =! 0 0 (7) b =kd2y dx2 63
Application
Figure 2 shows how soil mantle thickness H varies as a function of time for various parameters (note that soil diffusivity, k, has been kept constant at 0.005 m2/s for all models). In figure 2A, it can be seen that hillslopes exhibiting high negative curvature
(convex hills) reach a steady-state Hand as the curvature decreases (or straightens), soil mantle growth converges to that for a perfectly horizontal surface. In figures 2B and C, a hillslope with extremely low curvature (that is, cty/d:l-~ 0) is assumed and varied the
1 maximum soil production rate So (50-2000 m yr" ) and the characteristic lengthscale of soil production lla(.2-0.5 m) over the observed range of these parameters in the soil production studies summarized in (Dietrich et al., 2003) (note that the values of a cited in
Dietrich et al. are off by a factor of 100). The parameter space considered in these figures covers much of that seen in nature because the soil production studies adopted in this study were done over a range of different bedrock protolith and climates (Dietrich et al., 2003). As expected, it can be seen in figure 2B that soil mantle thickness increases more rapidly if the maximum soil production rate is larger. Thus, bedrock lithologies, which are more easily weathered, will yield higher soil production rates and therefore faster growth of soil mantle thicknesses. Figure 2C shows how soil mantle thickness varies with V? which can be interpreted as an inverse measure of the characteristic lengthscale to which chemical and physical weathering processes (root infiltration, animal burrowing, water seepage, freeze/thaw, etc.) can extend beneath the soil surface.
Not surprisingly, soil production is faster when the characteristic lengthscale (1/a) is
larger. 64
3. Mass Wasting
Constitutive Equations
Equation (7) can now be used to estimate how long it takes for a soil mantle to reach unstable conditions, that is, critical time, tc. What follows can be easily modified from classic textbooks, but the details are presented here for completeness. In the model, the soil mantle will slide down the bedrock-soil interface if the basal shear stress Tb at the
bedrock-soil interface exceeds the well-known Coulomb failure criterion:
(9)
where Tc is the critical yield stress (Pa), TN is the stress acting normal to the bedrock-soil
interface, S is the internal strength of the bedrock-soil interface, rjJ is the angle of friction
and P0 is the pore pressure (Po = pwgH cos B , where Pw is the density of water and His
the thickness of the soil mantle taken perpendicular to the bedrock-soil interface). See
figure 1D for a diagram of most of these parameters.
Assuming that the hillslope is very long compared to the soil mantle thickness
(this is reasonable considering steep hillslopes), the local force balance is approximated
by an infinite slope, that is, downhill pressure gradients can be ignored (that is, an infinite
slope). The basal shear stress Tb and normal stress TN are then given by
Tb = PwsgH sinO (10) TN= PwsgH cosB
where g is the gravitational acceleration, Pws is the wet soil density and B is the slope
angle of the bedrock-soil interface. Substituting equation (1 0) into equation (9) and
rearranging, mass wasting occurs when soil mantle thickness (for a given slope angle)
exceeds a critical value He that is 65
s H ~He=---.------(11) g(Pws smB- (Pws- Pw)cosBtantjJ)
Landslides will be most prone on the steepest parts of a hill, which must coincide (for hill profiles that can be parameterized by a continuous function) with the inflection of the hillslope profile (that is, cty/dx2 =0). AsH is a function of time ( eq 7), equation (11) in theory provides a straightforward means of determining the periodicity tc of landslides for
a given point on a hill slope. If H does not exceed He, such as might be the case if the
slope angle is too small, then mass wasting will never occur and H will eventually reach
its steady-state value.
Application
As shown above, He is a function of S, tjJ and B. Figure 3A plots He as a function
of() for reasonable range of Sand t/J. In figure 3B, those ranges are superimposed on a
plot of soil growth. In these calculations, the soil is saturated with water when it fails.
This is a reasonable assumption because the soil becomes unstable in its saturated state
well before it becomes unstable in its unsaturated state. It can also be seen in figure 3B
that slope instability occurs well before the soil mantle thicknesses approach their steady-
state values and depends almost solely on slope but not on local curvature for slopes of
30° or more. The independence from curvature is due to the fact that, initially, soil
2 mantle thicknesses on different parts of a hill (that is, different tfy!dx ) grow roughly at
the same rate. It is only after long times when the effects of creep and curvature start to
play a larger role, but slope instability is reached before the soil mantle growth curves for
different curvatures begin to diverge significantly. Thus, for a given slope and lithology,
there is a fairly well-defined critical time, tc, for slope failure. Thus, in the model, the 66 soil mantle thickens over thousands of years (eq 5) until a critical thickness is reached at which time the slope fails, thereby renewing the bedrock surface and re-initiating the cycle. High slopes yield shorter periodicities, whereas low slopes yield longer periodicities. On short timescales, random factors such as earthquakes are likely to
enhance soil instability. However, over long timescales, these random factors should
average out so that the model here describes the average periodicity of mass wasting for a
given landform. 67
4. Implications for chemical weathering inputs to the marine environment
The total chemical weathering input into the marine environment from regions of high relief can be represented by the sum of dissolved material from two sources. One source is the soil protolith, which chemically weathers during soil formation. Another
source of dissolved material is the soil removed from the mountain during mass wasting
events, some of which dissolves as bed load, which is stored as river bedload on average
for timescales equal to the time between consecutive mass wasting events.
Chemical Weathering during Soil Formation
First, the process that transforms the original protolith into soil should be
quantified. The mass ratio of original proto lith to current soil for a given package of soil
can be estimated from ratios of immobile or perfectly conservative elements (for example,
a high field strength element such as Zr, Ti, or Nb) in the soil to the proto lith (Brimhall
and Dietrich, 1987a) such that
-j
M 0 Csoil --~-- (12) Msoil C~
where M 0 is the mass of the original protolith, Msoil is the mass of the soil, c:on is the
depth-averaged concentration of an immobile element, i and C~ is the concentration of i
in the original bedrock protolith.
In order to bracket chemical weathering rates during soil formation, maximum
and minimum bounding cases are considered. More realistic, but still empirical,
weathering rate laws should fall somewhere within the maximum and minimum
bounding cases. The fractional rate of chemical weathering is first assumed to be 68 constant, that is, as the soil/proto lith interface propagates downwards into the bedrock, the chemical weathering rate (or mass loss) from the soil is assumed to be constant at any given depth and hence with time. This is clearly a maximum bound on the time- or depth-integrated weathering rates because bioturbation, carbonic acid and pore waters usually decrease with depth into the soil column reducing the weathering rate. An expression that represents a constant fractional rate of loss of mass during soil formation follows:
(13)
where fJ is the constant weathering rate and M::::X is the ratio of mass chemically lost Mo from the soil during soil formation to the mass of the original protolith. The goal is to estimate the chemical weathering flux associated with soil formation averaged over several mass wasting cycles of periodicity tc, expressed as Mt:!/: (kg/m2/y). Integrating equation (13) with respect to time yields the mass loss as a function of time M::::X = f3t. Mo
Integrating again up to time tc and dividing by tc gives the average mass lost during one
landslide cycle _.!._ fJtcMo. Dividing this quantity yet again by tc yields the average 2 chemical weatheringjlux over several soil formation/landslide cycles:
}jMAX =M * f3 (14) loss o 2
A minimum rate of chemical weathering loss during soil formation is also considered by assuming an exponential decrease in the fractional weathering rate with age of soil assuming the following relationship: 69
(15)
1 where r (y{ ) is the inverse of the characteristic timescale of chemical weathering formation and A is the initial, maximum rate of the change of the ratio. Integration of
equation ( 15) with respect to time yields the mass loss as a function of time.
MMIN ~=l-e-rt (16) Mo
Following the same procedures for the derivation of equation (14), equation (16) is twice
integrated and twice divide by fc. The integration and first division by tc yield the average
mass loss over one land slide cycle. The final division by tc yields the average chemical
weathering
-.MIN [ 1 e--;t, -1] M loss = M 0 * - + 2 fc rfc
(17)
Equations (14) and (17) are maximum and minimum bounds on the chemical weathering
flux during pedogenesis.
Chemical Weathering during Transport as Bedload
Chemical weathering associated with the material removed from the slope during
mass wasting events is not considered. This solid material is temporarily present in the
river as suspended or bedload. The transport time of the suspended load is taster than the
timescale of any significant chemical weathering; hence, dissolution during suspended
transport is ignored such that bedload weathering is the only source of dissolved load in
the river. However, for simplicity, all wasted material is assumed to be transported to the 70 river as bedload. The average residence time of river bedload does not exceed the recurrence intervals of mass wasting if the production of sediments (mass wasting) and river output of sediments is roughly in steady-state (averaged over >kyr timescales).
Thus, for the model, the average time that solid material from land slides remains in the river is the maximum probable time interval ofland sliding, tc.
For simplicity, maximum and minimum rates of chemical weathering are again considered. A maximum rate of chemical weathering in the river is constant with the amount of solid material added. A minimum rate will be assumed to decrease exponentially with the age of the bedload material itself. With progressive weathering, the soil becomes more dominated by refractory phases and clays that that are closer to equilibrium with the aqueous environment than their original bedrock proto1ith.
The proposed maximum and minimum bounds on the rate of bedload dissolution are mathematically identical to those proposed for the rate of soil formation in Section
4.1. In order to adapt equations (14) and (17) for bedload dissolution, the mass ofrock
available for soil formation, M 0 , is replaced with the mass of bedload which is the mass
of soil at the time of land sliding, M 0 - M 10,, , such that
MMAX (M MMAX) * /3' river = o - lo!i'S l
(18)
-MIN. ( MIN ) [ 1 e-r'tc -1 Mr1ver = Mo -Mi Modeling Dissolved and Suspended Load in the River: Applying the Model In order to estimate the flux of dissolved load to the rivers, [J, y and Mo must be estimated and equations (14), (17), (18) and (19) must be solved. For simplicity, fJ = [J' andy = y'. fJ andy are estimated graphically using data from three soils, two from Tanzania and one from Hawaii-all formed from a basaltic protolith (fig. 3C). The Machame slopes of Mt. Kilimanjaro in Tanzania have a depth averaged soil/protolith Nb and Ta ratio of C:oit ~ 1.4 (Little and Lee, 2006). At this site, 30% of the original C'0 protolith mass was lost during soil formation, that is Mt::: = 0.3, which occurred over a Mo maximum of 0.17 to 0.51 million years (Dawson, 1992; Evemden and Curtis, 1965). At Mt. Meru, the other Tanzanian site, M{:!N = 0.2, at a maximum age of 0.1 to 0.2 million Mo years (Bagdasaryan et al., 1973; Dawson, 1992). At Kohala in Hawaii, M{;!N = 0.58- 0.84 based on Zr ratios from six soils which formed over a maximum of Mo 0.13 to 0.21 million years (Teutsch et al., 1999). fJ = 1.4 * 10-6 yr-1 andy= 1.4 * 10-6 yr- 1 (fig. 3C). Values for M0 can be roughly estimated as a function of He, the soil thickness at the time of failure and a soil-to-protolith conversion such that M = C :au * p' . *H (21) o ci SOli c 0 Where p 'soil, is the dry density of the soil. In figure 3D, M/':::, M/!N, M;!~ and M:::: are all plotted against time. Average parameters for the calculation of critical soil 72 thickness, He, were used in this calculation (a= 2 m-I, k= 0.005 m2/s, d2y/dx2 = -0.00001 and eo=0.00005). As expected, the flux from weathering in the river is smaller than the loss during pedogenesis. The maximum weathering rate produces more dissolved flux than the minimum rate; however, the difference is trivial. An increase in f3 or y represents a faster weathering rate and will contract the curves in the time axis and will increase the predicted annual weathering flux. A decreases in f3 or y represents a slower weathering rate and will extend the curves in the time axis and will decrease the predicted annual weathering flux. Note that the curvature in the linear (maximum) weathering flux is derived from the logarithmic growth of the soil mantle (fig. 2). Figure 4 plots the annual chemical weathering flux associated with soil formation against that associated with wasted mass in rivers as a function of the slope-dependant periodicity of landsliding, fc, for a range of bedrock to soil conversion rates 80 (0.00005 m 1 1 yr- < 80 < 0.002 m yr- ). Critical time (tc), or landslide frequency, is indicated on shaded regions. The darkest regions in figure 4 represent steep slopes and high landslide frequency. In these regions, the solute flux from soil formation is positively correlated with that from bedload dissolution. This correlation occurs because steeper slopes give rise to more frequent landslides which, in turn, transport solid material to the rivers for dissolution and which expose more unweathered material for soil formation. However, on more gentle slopes ( < 20 °), the chemical weathering flux averaged over several landslide cycles must decrease because the rate of weathering is controlled by the critical soil thickness, He, which thickens at an ever decreasing rate (fig. 2). The dependence of the dissolved flux on the soil formation rate explains why the soil loss term (abscissa) eventually decreases after an initial increase (fig. 4). Beyond these similarities, there are 73 three main differences between the two dissolution flux plots for the maximum and minimum rates (Figures 4a and b, respectively). First, the maximum rate produces higher flux from river dissolution and soil formation than the minimum case. Second, for the first 50,000 years, maximum rate input terms follow a one-to-one relationship, but bedload dissolution outpaces the minimum soil flux by a factor of more than 8:1 for the minimum rate case. The large mass loss during maximum soil formation leaves a relatively smaller amount of material in the soil column for mass wasting and thereafter bedload dissolution. Lastly, the long-term sense of the two plots is counterclockwise for the maximum rate and clockwise for the minimum rate. Despite these differences, the ratio of chemical weathering associated with soil formation to wasted mass fall within the linear leg of figure 4 because for the relatively steep slopes (> 20 °) of interest, failure occurs within 20 ky (fig. 3B). Figure 5 plots the total chemical weathering flux (from both soil formation and bedload dissolution) versus the total average suspended load associated with mass wasting, estimated as the total mass of soil at the time of a landslide. M suspended = M soil - M river (22) As in previous figures, the ratios are plotted as a function oflandslide periodicity, fc. The 1 shaded region represents a range of bedrock to soil conversion rates Go (0.00005 m y{ < 1 Go< 0.002 m y{ ). Also shown is the global array of solute flux versus suspended flux as determined from a global survey of river waters, (chemical weathering)=0.39(physical 0 66 weathering) · (Millot et al., 2002); note that the solute flux used in constructing the global array data does not include Si, 0, nor any anions and as such under-estimates the total solute mass flux. 74 This model does not reproduce the global array. First, instead of a positive correlation, the model suggests a negative correlation between suspended and dissolved load. Second, in regions of high relief, where slopes are >20°, the predicted dissolved to suspended load ratio falls below the global array. Plotting below the global array indicates that the predicted dissolved load is too low or the suspended load is too high. One way of reconciling the discrepancy between the model and observations is to increase the chemical weathering rates fJ and y during soil formation by 10 times. However, the empirical calibrations of fJ and y are based on soils formed on basaltic substrates under wet, tropical conditions; basalt weathering rates are probably already on the high end of the weathering rate spectrum. Thus, it is not likely that chemical weathering rates in the soil and bedload have been underestimated. In any case, the fact that the model predicts a negative correlation between solute and suspended load indicates that enhanced physical weathering does not immediately give rise to enhanced chemical weathering. This negative correlation and the solute deficit can be rationalized as follows. In regions of high topographic relief, denudation is so fast that the residence time of soils on hillslopes is small; hence there is not enough time for extensive chemical weathering. Only when slope angles are low do denudation rates also becomes low and soil mantle residence times become longer. In the extreme case of no relief, denudation drops to zero and the rivers should then have a very high solute/suspended load ratio. 75 5. Resolving the discrepancy between model predictions and observations Based on the above discussions, this inability to match the global river array in terms of its magnitude and sense (fig. 5) implies that all sources of solute into rivers have not been accounted for or that the suspended load flux has been overestimated. Below are three possible scenarios and discuss how these scenarios can be tested. Measured Suspended Load in Rivers Underestimate the True Suspended Load A fundamental assumption in the model is that when mass wasting is averaged over long enough timescales or when the watershed area is significantly larger than the affected area for individual landslides, the resultant suspended load is simply the time averaged mass wasting flux. The model also assumed that the average residence time of bedload material in the watershed region of rivers is equal to the average residence time of soil mantles on the hillslope. However, on short timescales, this is probably not the case. Flushing ofbedload material may in fact be very episodic. If bedload flushing is associated with very large floods that happen only once in several decades or more, actual point measurements of suspended load in rivers are unlikely to detect these sudden flushes of bedload material. Finally, the measurements of suspended load and dissolved load in the global array database of (Millot et al., 2002) are dominantly measured at the mouths of rivers or in the floodplains, not in the headwaters or watershed areas. This means that most of the riverine measurements are made in depositional areas, i.e. the river delta, rather than erosional areas. Sediment deposition will decrease the suspended load. 76 These hypotheses can be tested by determining the sediment burial rate of rivers in their deltaic or alluvial plains. Although beyond the scope of this paper, this quantity allows for the estimation of the fraction of the total un-dissolved flux (physical erosion) represented by actual suspended load measurements and the fraction lost from suspension as deltaic sediments. These discussions suggest it would be worth incorporating a sedimentation component in the model to make it more applicable on a larger scale. Chemical Weathering in the Alluvial Fan or Floodplain ofRivers is not Accounted for In the alluvial or deltaic region of rivers, the residence time of sediments in one given location is likely to far exceed the residence time of soil mantle and river bedload in the actively eroding watershed regions. Alluvial/deltaic sediments may thus have more time to undergo chemical weathering, thereby contributing more solute flux to the rivers without a large contribution of suspended load (because active deposition is occurring). However, deltaic chemical weathering cannot occur indefinitely. More rapid erosion in the watershed regions will cause more rapid sedimentation in the deltaic regions. Rapid sedimentation isolates earlier deposited sediments from a fresh new supply of water, slowing down chemical weathering considerably. In other words, more rapid sedimentation causes a decrease in the contribution of chemical weathering from the sediments. Knowing the exact trade-offs will require further sophistication of the model. Solute Flux Associated with Groundwater Weathering is not Accounted 77 Another way to reconcile the discrepancy between the model predictions and the global array of Millot et al. is to invoke the role of groundwater weathering. Groundwater is often high in total dissolved solids and carries little suspended load. From a global perspective, the role of groundwater weathering has received less attention. For this reason follows a discussion on the importance of groundwater. High solute concentration alone is not enough to make groundwater flow the "missing" input. It is the total solute flow associated with groupdwater that is of interest. In order to balance the discrepancy, groundwater must deliver a dissolved flux that is ten times greater than the predicted dissolved load. To assess this hypothesis, the groundwater solute concentration and the total flow rate of groundwater are required of a given watershed. Groundwater concentrations of Ca, Mg, Na and K (Table 2 and fig. 6A) from Mt. Etna, the Bengal Basin, Furnas volcano in the Azores, the Guiana Shield, the Rhine Valley, peninsular Malaysian karst and Netherlands peat are compared (Aiuppa et al., 2000; Crowther, 1989; Cruz et al., 1999b; Dowling et al., 2003; Durand et al., 2005; Negrel and Lachassagne, 2000; Schot and Wassen, 1993) to the concentrations in the Amazon, Ganges, Brahmaputra, Purari and Limpopo rivers (Gaillardet et al., 1999). Based on this admittedly limited compilation, groundwater appears, on average, to have higher concentrations of dissolved elements than rivers. In Table 2, these values have been converted to watershed area-normalized groundwater and riverine fluxes where appropriate data were available (this flux is the product of concentration and flow rate divided by the catchment area). Figure 6B shows that groundwater from Mt. Etna and the Bengal Basin deliver a dissolved flux ten times greater than that of the rivers. Thus, 78 in some regions, groundwater solute fluxes may be high enough to provide the dissolved input missing from this model. The possible importance of groundwater weathering discussed above can be further explored by examining whether groundwater flows in mountainous regions in general are large. This can be assessed by examining whether groundwater flow is large enough to affect heat removal by advection. In this context, groundwater has been shown to explain the anomalously low apparent heat fluxes in the tectonically active Oregon Cascades (Ingebritsen et al., 1989; Manga, 1998) and possibly even in the Sierra Nevada (Wang et al., 2005), an extinct Mesozoic arc in California. To illustrate this concept, two heat flux transects across the Sierra Nevada and one across the Peninsular ranges are examined (fig. 7), the latter representing the southern extension of the extinct Mesozoic arc batholiths (heat flow data taken from (Clark, 1957; Henyey, 1968; Henyey and Lee, 1976; Henyey and Wasserburg, 1971; Lachenbruch and Sass, 1980; Lachenbruch et al., 1976a; Lachenbruch et al., 1976b; Mase et al., 1982; Mase et al., 1980; Pollack et al., 1991; Roy et al., 1968; Sass et al., 1982; Sass et al., 1971; Wang and Munroe, 1982). In the Sierra Nevada, heat fluxes fall1argely between 30-50 m W/m2 except in the immediate vicinity of Long Valley and the Owens Valley, where there is ongoing volcanism associated with very recent Basin and Range tectonism. In the Peninsular Ranges, the 2 heat fluxes range between 30-60 mW/m • These heat fluxes are more typical of stable, Archean cratons (>2.5 Gy in age) (Nyblade, 1999). Tectonically young areas, such as the late Mesozoic Sierran and Peninsular Ranges batholiths, should be characterized by surface heat fluxes greater than 60 mW/m2 assuming that all mantle heat is transported through the lithosphere and to the surface by conduction. 79 That the apparent surface heat fluxes are substantially lower than that expected for their young tectonothermal ages suggests that an advective component may be involved in the transport of heat. Advective heat transport via groundwater flow could explain this discrepancy as suggested by Manga et al. for the Cascades (Ingebritsen et al., 1989; Manga, 1998). However, groundwater flow obviously should only play a role ifthere is enough replenishment of groundwater in the mountains. In this context, the heat flux has been depressed more in the Sierras than in the Peninsular Ranges. It is tempting to speculate that this difference may be associated with the greater precipitation rates in the western flank of the Sierras (80 to 150 cm/yr) than in the Peninsular Ranges (40 to 80 cm/yr) (USDA-NRCS, 1999). These observations, if generalizable, suggest that so long as there is sufficient precipitation, groundwater flows could be just as significant as riverine flow or precipitation runoff. The presence of deep saprolitic bedrock in tropical climates attests to the importance of groundwater or subsurface weathering. More extensive studies of groundwater weathering should be encouraged. 80 6. Conclusions A model linking the continuous process of soil formation and chemical weathering to the stochastic process of mass wasting on hillslopes in regions of high topographic relief was presented. This model links the frequency of mass wasting to the rate of soil formation, which depends on hillslope curvature and bedrock-to-soil conversion rate. A periodic cycle is predicted in which a soil mantle gradually forms on a sloping bedrock surface. After the soil mantle reaches a critical thickness, the entire soil mantle fails catastrophically at the bedrock-soil interface (mass wasting). The soil formation and mass wasting models were combined with a simple model for chemical weathering, allowing one to estimate the total dissolution associated with soil formation during stable times and of wasted mass during temporary storage as river bedload or as suspended river load. These observations suggest that greater topographic relief, while leading to enhanced physical erosion, does not lead to an increase in chemical weathering. On the contrary, a negative correlation between physical erosion and chemical weathering for a given set of parameters is predicted. In retrospect, this makes intuitive sense because in regions of higher topographic relief, the residence time of a soil mantle is shorter, leaving little time for chemical weathering. The models also predict a solute to suspended load ratio much lower than actually measured in the world's major rivers. Three scenarios are suggested to resolve this discrepancy: 1) measured suspended loads under-estimate the total un-dissolved load due to sedimentation in alluvial or deltaic plains, 2) chemical weathering in the alluvial/deltaic regions of rivers is not accounted for in the models and/or 3) chemical weathering associated with groundwater flow is not accounted for. While the model did not account for these additional factors, this model 81 provides the framework for future models coupling mass wasting, soil formation and sedimentation. 82 Table 1. Major Element Concentrations in Groundwater and Rivers Groundwater mg/L Ca Mg Na K concentration Mt. Etna (Aiuppa et al., 2000) 28 58 105 16 Bengal Basin (Dowling et al., 2003) 110 52 349 8.5 Furnas volcano, Azores (Cruz et al., 1999) 29 8 142 80 (Negrel and Guayana Shield 15 3.8 8.0 0.94 Lachassagne, 2000) Rhine Valley (Tricca et al., 1999) 99 8.3 50 6.3 Peninsular Malaysian (Crowther, 1989) 51.3 4.8 Karst (Schott and Wassen, Netherlands Peat 2525 102 207 59 1993) River concentration mg/L Ca Mg Na K Amazon (Gaillardet et al., 1999) 5.4 0.90 1.8 0.82 Ganges " 23 6.5 9.6 2.6 Brahmaputra " 14 2.1 3.8 3.9 Purari " 21 3.2 2.6 1.0 Limpopo " 19 12 21 4.6 Groundwater flux tonslkm2/yr Ca Mg Na K Mt. Etna (Gaillardet et al., 1999) 17 36 63 10 Bengal Basin (Dowling et al., 2003) 66 31 209 5.1 Furnas (Cruz et al., 1999) 2.9 0.8 14 8 River flux tonslkm2/yr Ca Mg Na K Amazon (Gaillardet et al., 1999) 6 2.0 1.0 0.9 Ganges " 11 4.5 3.0 1.2 Brahmaputra " 12 1.8 3.3 3.4 Purari " 57 8.8 7.2 2.8 Limpopo " 1.1 1.2 0.7 0.3 83 Figure Captions Figure 1. Cartoon depicting model parameters (A). Photos indicating landslides on mountain slopes in Chiapas, Mexico (B). Cartoon showing the main variables used in calculating the rate of soil formation (C) and Cartoon showing the main variables used in calculating the criteria for soil failure (D). Figure 2. Growth of the soil mantle, H, a function of time in thousands of years and 2 three model parameters: hillslope curvature, -cfy!dx ; soil production, e; and characteristic lengthscale of soil formation, a. ln A, the curvature cfy!dx2 is varied while 1 2 1 1 keeping all other parameters constant ( £ 0 =0. 001 m yr- , k=O. 005 m yr- , a=2 m- ). In B, maximum soil growth rate £0 is varied while curvature is assumed to approach zero 2 1 1 (k=0.005 m yr- , a=2 m- ). InC, the characteristic lengthscale of soil production 1/ais 1 2 1 varied while curvature is assumed to approach zero, £0 =0. 001 m yr- and k=O. 005 m yr- . Figure 3. Modeling critical soil thickness, critical soil time, dissolution rate constants and dissolved flux. Critical soil thickness, He, depends on slope, B and various strength parameters (A). The internal angle of friction, if>, is a function of soil type and the strength of the bedrock soil interface, S, is the function of a variety of factors. The inset lists angles of internal friction, if>, for various materials (Holtz and Kovacs, 1981 ). In (B), 1 2 1 soil thickness is plotted in black for a range of curvatures (0 m- :::; cfy!dx :::=: 0.01 m- ). In (B), also ploted are the range of He from (A) for slopes of20°, 30° and 40° as shaded rectangles producing a range of critical times, fe, for each slope. In (C), maximum, fJ and minimum, y, rate constants are estimated. The Machame and Meru soils are from 84 northern Tanzania (Little and Lee, 2006) and the Kohala soils are from Hawaii (Teutsch et al., 1999), all from basalt protolith. Using the /3 andy estimates, the maximum and minimum flux from chemical losses during soil formation and dissolution of river bedload are calculated (D). Maximum estimates are in black; minimum are in gray. Chemical loss during soil formation is in solid lines; bedload dissolution is in dotted lines. 3 P 'soil= J . 6 tonnes m - and --.C :oil = 1 .4 c0 Figure 4. Ratio of dissolved fluxes in the model for maximum (A) and minimum (B) weathering rates. The dissolved flux derived from dissolution of the bedload and from soil formation are plotted along the abscissa and ordinate axes respectively. The ratio of these fluxes are represented by shaded polygons whose extent is defined by the average landslide periodicity, tc, as indicated and the initial rate of soil formation, eo, which spans 0. 00005 m yr-1~ &0 ~ 0. 002 m yr-1 in these figures. Figure 5. Dissolved flux versus suspended load. Dissolved flux represents the sum of the input of dissolved mass to the river derived from soil formation and dissolution of bedload. Suspended load is an estimate of bedload after dissolution in the river. (A) Maximum (linear) chemical weathering rate and (B) minimum (negative exponential) chemical weathering rate. Open polygons represent the predicted ratio of dissolved to suspended load for the indicated landslide periodicity over a range of initial soil 1 1 formation rates (0.00005 m yr- ~ &0 ~ 0.002 m yr- ). Diagonal solid line with diamond symbols represents the global river array (Millot et al., 2002). The gray shaded regions 85 represent the range of predicted tc for slopes of 20°, 30° and 40° over a range of soil strength characteristics (see Fig. 5a). Figure 6. Concentration (A) and flux (B) ofCa, Mg, Na and Kin selected groundwater and rivers. Flux was calculated as the product of the elemental concentrations and total water flow divided by catchment area (Aiuppa et al., 2000; Crowther, 1989; Cruz et al., 1999b; Dowling et al., 2003; Durand et al., 2005; Gaillardet et al., 1999; Negrel and Lachassagne, 2000; Schot and Wassen, 1993). Figure 7. Plots of heat flow and elevation across the Sierra Nevadas, (A) and (B) and Peninsular Range, (C). The location of the transects are shown on the Predicted heat flow for quaternary volcanic regions is ~ 100 m W m-2 and is shown as a dark gray swath across all plots. 86 FIGURE 1 A 87 FIGURE 1, continued d2 J. = Curvature c 'tN =Normal ght D 88 FIGURE2 3 Hillslope A Curvature om-' -E 0.001 m- 1 -2 0.003 m·' ::I:... 0.005 m·' .c..... O.ol m·' a. cu 0.02m·' "a -d2yldx2 ·--0 "' 0 0 20 40 60 80 Time (ky) 3 Soil Production B 0.002m/y 0.001 m/y 2 0.0005 m/y -E 0.0002 m/y -::I: 0.0001 m/y O.OOOOSm/y E 0 0 20 40 60 80 Time (ky) 3 Soil Formation c Lengthscale Inverse 2.om·' 2- 2.5m·1 3.0m·' -E 3.sm·' 4.0m·1 -::I: 4.Sm-' 1 a 0 --~------~ 0 20 40 60 80 Time(ky) 89 FIGURE3 10 Material Type 2.0 B 1.6 ..-..1.2 E :Co.s 0.4 0.0 0 4 8 12 16 Time (ky) 90 FIGURE 3, continued 0.5 c 0.4 0.3 MACHA ME ..c:E~I 0 ~ 0.2 0.1 0 0 100 200 300 400 500 Tc (ky) _..._ 2.0 ,- >.I D N E 1.6 C/) c: 0 ~1.2 X ::::l lJ.. -o 0.8 Q) > Chern loss 0 ~ 0.4 ---===== River loss 0 0.0 0 20 40 60 80 100 Tc(ky) 91 FIGURE4 8 7 A ,.-- 6 '->- 1.4My N E 5 ~ U'J 4 s:: 0 ~ 3 ~ .2 2 .~ 1 0 0 1 2 3 4 5 Mriver (tons km-2 yr1) 5 8 4 ,... -L.. >.. N 3 E' ~ U'J c:: 2 0 -(f) -en 0 1 ~ N T""v 0 0 0.5 1 1.5 2 2.5 Mnver (tons km-2 yr1) 92 FIGURES X ::::::5 u., al > 0 (/) (/) Cl 0.1 +------,.L-~'-----~ 1 10 100 1000 10000 Suspended Flux (tons km-2 y-1} 100~------~ T"" -I >. <"}! E ~ fJ) 10 c -.9 10 100 1000 10000 Suspended Flux (tons km-2 y-1) 93 FIGURE6 Concentration 10000 • Groundwater 1000 A. rrv1t Etna II Bengal Basin <>Furnas, .A.zores -..._....I 100 Ll 4 Guiana Sheild 0) • l!il Rhine Valley • Malays ian Karst ...__.E 10 •* 6 Ne1hertands F)eat 1 azon Ganges Brahmaputra Purari limpopo Flux ,..-...... 100 ~ N I E 10 ~ m c _.0 - 1 94 FIGURE7 100 4000 A -~ 80 E 3000 $ m CD E 60 m< - cr ~ -::J u:: 40 (ij -3 Q) • • J: • • 1000 - 20 • • • 0 ·0 -122 -121.6 -121.2 -120.8 -120.4 Longitude 100 . 4000 B ~ -E 80 3000 m ~ ~ g 60 • m 3: • 2000 6' 0 -::J u:: 40 • • 3 co •··.: .... - -Q) '••' J: .' . 1000 - 20 • .:···~r·._. • 6 0 0 -116.8 -116.6 -116.4 -116.2 -116 Longitude 95 FIGURE 7, continued 96 PART III: Element distribution in soils from Mount Meru, Tanzania: a combined bulk and leach study formation, creep and mass wasting Submitted to Applied Geochemistry in the spring of 2006 and authored by Mark Gabriel Little and Cin-ty A Lee. 97 Element distribution in soils from Mount Meru, Tanzania: a combined bulk and leach study Synopsis A modified BCR (Bureau Commun Reference) sequential extraction was performed on a basaltic soil (phono-tephrite) from Mt. Meru in Northern Tanzania in order to determine the relative contribution of water soluble, carbonate and exchangeable, oxide and organic fractions to the bulk composition of the soil. Elemental compositions were determined by ICP-MS and corrected for loss on ignition. Relatively immobile elements, such as Zr, Hf and Al, are enriched by 10 to 30 % compared to the unweathered protolith, consistent with soil formation being accompanied by mass loss due to chemical weathering. However, superimposed on this mass loss appears to be enrichment of elements such as Fe, Ca and Mg, especially towards the surface. In some cases, the bulk concentrations of these elements at the surface exceed that of the proto lith. These data suggest that the surface of the Meru soil columns may have experienced "re-fertilization" by the deposition of volcanic ash. From the carbonate and exchangeable leach, there is evidence of clay rich horizons which may sequester as much as 5 % of the bulk K. The concentration of calcium carbonate appears to decrease with depth, but the largest incorporation of Sr and Ba into carbonates occurs below 114 em. Fe and Mn oxides scavenge more than 10 to 20% of total Ti, V, Co, Cu, Zr and Pb below 114 em. The organic fraction sequestered significant fractions of total Al, Cu, REE's and Pb throughout the soil column. 98 1. Introduction The objective of this study was to assess the mobility of major, minor and trace elements during chemical weathering of a basaltic proto lith through a case study of a basaltic (phono-tephrite) soil profile on the southern slopes of Mount Meru in northern Tanzania. Of particular interest are the extent of element mobilization in the bulk soil and the absorption of elements by authigenic phases. In order to analyze the authigenic phases, a series of chemical leaches intended to separate the water soluble, carbonate and exchangeable, oxide and organic soil fractions was performed. The bulk chemistry data allows for the quantification of net chemical losses from the soil during weathering as well as net mass additions by atmospheric dust deposition. The leaching experiments provide insights into the mechanisms by which various elements are redistributed within the soil column during soil formation, which in tum may have implications for transport of anthropogenically introduced heavy metals. 99 2. Materials and methods 2.1 Sample description Samples were collected from soils formed on a massive basalt flow northern Tanzania on the southern slope ofMt. Meru (03°19.604' S, 36°52.602' E, elevation 1363 m) where annual rainfall is> 1000 mm/yr (Mlingano et al., 2006). 22 soil samples were collected to a maximum depth of 177 em using a 40 mm diameter split tube sampler (Dormer Engineering). Meru basalts have been dated between 0.1 to 0.2 million years (Bagdasaryan et al., 1973; Dawson, 1992). The volcanic soils in this region were formed from a highly alkalic, augite-, plagioclase- and nepheline-bearing basaltic protolith (phono-tephrite ), which was exposed approximately 100 meters from the sampling site in a fresh, ~5 m deep ravine. The unweathered protolith consists of ~20% plagioclase phenocrysts (0.5-0.1 mm in diameter), set within an aphanitic groundmass made up of small phenocrysts of augite, plagioclase, occasional nepheline and Fe-Ti oxides. 2.2 Bulk soil analysis Meru soil and proto lith samples were dried for 24 hours at 105°C. The water content was calculated as the mass of water lost during drying divided by mass of the dry soil. A portion of the dry soil samples were heated to 550 °C for 4 hours, the% mass lost reported as loss on ignition (LOI). Soil porosity was calculated assuming a constant dry 3 soil density of 2.5 g/cm . Soil pH was estimated as the pH of a 1:2 mixture of dry soil and milipore water. Aliquots of the dried samples were ground by hand with a ceramic mortar and pestle. Sample powders, a blank and BHV01, BHV02 and BIR1 USGS rock standards 100 were prepared for inductively coupled plasma mass spectrometry (ICP-MS) analysis using a series of acid dissolutions at Rice University. Approximately 80 mg of the Machame soil and rock powders, BHV02 and BIR1 were weighed; 2 aliquots of approximately 40 and 80 mg ofBHV01 were weighed for use as an external standard. Powders were placed in Teflon beakers with approximately 0.5 mL each of concentrated Seastar HN03 and HCl04. These solutions were sealed and ultrasonicated for ~20 min. Beakers were then heated overnight at 115 °C. The following day, beakers were removed from the oven, unsealed and placed on heating pads at~ 175 °C until the solutions had dried down. This process of adding acids, overnight heating and drying down was repeated 1 additional time. Following the second and final dry down, ~1 mL of2% HN03 and~ 0.1 mL of concentrated HCl were added to the samples. Beakers were ultrasonicated and then placed on a hot plate at ~ 100 °C for about 4 hours. After samples had dissolved completely, the beakers were removed from the hotplate. The contents of the Teflon beakers were diluted to 100 mL in polyethylene bottles using 2 wt. % HN03. Solutions were spiked with a pure Indium solution to yield a concentration of~ 1 ppb In, which was used to monitor instrumental drift. Meru samples, blank, BHV01, BHV02 and BIR1 were all run on a Finnigan Element II, single collector inductively coupled plasma mass spectrometry (ICP-MS) at Rice University. The ICP-MS was run in low and medium mass resolution modes (m/8m=300, 4000), the latter allowing for the analysis of Na, Mg, AI, P, K, Ca, Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Zr and Nb, for which isobaric molecular interferences can occasionally be a problem in low mass resolution. 101 2. 3 BCR (modified) leaching technique A variety of sequential leaching techniques were investigated for this study (Alloway, 1995; Asami et al., 1995; Cabral and Lefebvre, 1998; Gibson and Farmer, 1986; Han and Banin, 1995b; Marin et al., 1997; Martinet al., 1987; Palumbo et al., 2000; Tessier et al., 1979). After two trial runs were made using the Tessier et al. (1979) technique, a modified BCR technique as reported by Marin et al. (1997) and Alloway (1995) was chosen the best choice for the small sample sizes (Fig. 1). Duplicate leaches were performed for 6 of the 21 soil samples to assess reproducibility. The objective of the first leach step was to extract the water-soluble fraction. Approximately 60 mg of each sample were accurately weighed and placed into a small plastic vial with a screw cap. Approximately 1.5 mL ofMillipore water was added to the sample vial. The vial and its contents were then shaken three times for ~30 minutes each over the course of 24 hours. After 24 hours had passed, the vial was shaken again for ~ 10 minutes and centrifuged for 20 minutes. The supernatant was finally pipetted out and stored in a separate vessel (the centrifugation procedure was repeated twice). The supernatant was combined and stored as the water-soluble fraction. After the water soluble fraction had been removed, the next step involved an acetic acid attack in order to extract elements adsorbed onto the surfaces of clays or sequestered in carbonates. Approximately 1.5 mL of an acetic acid solution (25% CH3COOH buffered to pH 5 with CH3COONa) was added to the un-dissolved residues from the previous leaching step. The vial containing the resulting suspension was shaken three times for ~30 minutes each over the following 24 hours. The vial was then shaken 102 again for ~ 10 minutes, followed by 20 minutes of centrifuging. The resulting supernatant was pipetted out and stored as the carbonate fraction. The final extraction was used to remove metals associated with Fe-Mn oxyhydroxides (oxide fraction). Approximately 1.5 mL of 1M hydroxyl amine hydrochloride solution (1M NH20H-HCl in a 25% acetic acid solution) was added to vial containing the un-dissolved residues from the acetic acid leaching step. The vial containing this suspension was shaken three times at ~30 minute durations over the following 24 hours. The sample vial was then centrifuged for 20 minutes and the supernatant was pipetted out for storage as the oxide fraction. The final leaching step was designed to extract the organic-bound metals (organic fraction). Un-dissolved residues from the previous step were washed into Teflon bottles with ~3 mL ofO.Ol M HN03. 2 mL ofH202 was added and the bottles were placed on hot plates and heated at 80°C until dry. Another 2 mL of H20 2 was added, followed again by heating at 80°C until dry. Approximately 3 mL ofO.Ol M HN03 was then added to the bottles and shaken. The resulting solution was centrifuged and the supernatant was pipetted out and stored as the organic fraction (to extract all supernatant, another 1.5 mL of H20 was added and centrifuge procedure was repeated) For ICP-MS analysis, the water soluble, carbonate and exchangeable, oxide and organic fractions were diluted up to 15 mL with 2% HN03. An In spike was added to each fraction to yield an In concentration of 1 ppb (internal standard for drift correction). The leach samples, procedural blanks and reference standards (BHV01, BHV02 and BIRl) were run on a Finnigan Element II, single collector ICP-MS at Rice University following techniques discussed above. 103 3. Results 3.1 Physical characteristics The water content, loss on ignition (LOI) and porosity profiles share the same overall depth-dependant behavior. The water content of the soil is greater than 40 wt.% throughout the soil column. From 2 to 26 em, water content nearly doubles from 48 to 87 %. From 26 to 61 em, high water contents of around 80% are maintained. Between 78 and 123 em, the water content is ~67 %. The water content increases from 46 to 55% below 132 em. The protolith LOI is low (1 %). Between 2 and 18 em, LOI of the soil decreases from a profile-wide maximum of29 to 17 wt. %. LOI increases abruptly to 26 % at 26 em and decreases steadily below to a minimum of 2 % at 177 em. The porosity also decreases between 2 and 18 em from 81 to 7 5 % and increases to a maximum of 82 % at 26 em. Porosity generally decreases below 26 em, the minimum value of 70 % reached at 166 em. In contrast to bulk physical characteristics, the pH of the soil increases between 2 and 18 em from 5.7 to 6.0, the maximum value. The pH decreases to 5.6 at 61 em and remains at ~5.6 to the base of the profile. The Chemical Index of Alteration, CIA= Al20 3 (where each ofthese oxides is presented as Wt. %), of the Al20 3 +CaD+ Na20 proto lith is 0.52 (Nesbitt and Young, 1984). The CIA of all soil samples is greater throughout the soil and increases from ~0.6 at the surface to ~0.75 at depth. The Mg#, molar Mg/(Mg+Fe), is~ 0.35 between 2 and 141 em but falls to protolith values of0.30 between 149 and 177 em. 3.2 Bulk soil composition 104 The bulk soil concentrations ofNa, Mg, Al, P, K, Ca, Ti, Mn and Fe are reported as metal-oxides in Table 1. The bulk soil concentrations ofLi, Be, Sc, Ti, V, Cr, Co, Ni, Cu, Zn, Ga, Rb, Sr, Y, Zr, Nb, Mo, Cd, Cs, Ba, La, Ce, Pr, Nd, Sm, Eu, Tb, Gd, Dy, Ho, Er, Tm, Yb, Lu, Hf, Ta, Tl, Pb, Th and U are reported in Table 2. Concentrations of all major element oxides and Sr, Ba and La are plotted against depth in Figure 2. Na20, K20 and Rb are all depleted near the surface (~50%) but more depleted below 61 em (~80 %). Cs also decreases with depth from near-protolith values above 61 em to 50 % depletions below. Li generally decreases with depth; but Li is enriched over the protolith concentrations by ~30% above 61 em and hovers around the protolith concentration below. Between 2 and 96 em, MgO and CaO are enriched over proto lith values by 100 and 40 %, respectively. Both subsequently decrease below 96 em, approaching a protolith concentration below 150 em. In contrast, the other alkaline metals are not generally enriched, nor do they exhibit an overall trend similar to MgO and CaO. Be concentrations are equivalent to the protolith throughout the profile. Sr is depleted by ~ 20 % at the top and bottom of the profile and by more than 100 % in the middle. Ba is depleted by >20% from the surface down to 123 em. Below 123 em, Ba is enriched. MnO and Fe20 3 are enriched moderately throughout the profile and up to 50% at 2 em and between 52 and 123 em. Ti02 is also moderately enriched down to 52 em; but, the greatest enrichment is below 61 em (>50%). The other period 4 transition metals, Sc, V, Cr, Co, Ni and Zn, are also enriched throughout the profile and behave most like Fe20 3• The lowest concentrations of these metals are found at or near the bottom of the soil profile and the highest concentrations are found at 61 and 114 em where these 105 maxima represent 2 to 10- fold enrichments relative to the protolith. Cu is also enriched throughout the profile by more than 300 %, but Cu does not show any dynamic behavior with depth. Ah03 is consistently enriched relative to the proto lith by~ 10 % in all soil samples. Ga, Y, Zr, Nb, the REE's (La, Ce, Pr, Nd, Sm, Eu, Tb, Gd, Dy, Ho, Er, Tm, Yb and Lu), Hf, Ta, U, Th and Pb are enriched by 10 to 30% over the protolith concentrations. In general, REE's are constant with depth though higher REE concentrations are found in deeper samples. There appears to be no fractionation between the light, medium and heavy REE's; however, there is a negative Ce anomaly in most samples, a negative Eu anomaly in all samples and a positive Gd anomaly in all samples relative to the protolith. P205 is enriched throughout the profile and generally decreases with depth; however, enrichments at the surface and at 61 em are greater than any other major element (>200% enrichment). Soil Tl is enriched throughout the soil profile and, in general, increases with depth. 3.3 Water soluble elements The water soluble concentrations of the major elements are reported as oxides in Table 3. The water soluble concentrations of minor and trace elements are reported in Table 4. Be, Sc, Cr, Ni, Cu and Pb data fall below the limit of detection (Table 4). Concentrations of all major element oxides, Sr, Ba and La are plotted against depth in Figures 2 and 3. No significant volume increase was observed while the samples were immersed in water. 106 Na20, K20, Li, Rb and Cs concentrations are more than 3 orders of magnitude below the bulk soil concentrations and all tend to decrease with depth. The main exception to this pattern is a moderate increase in water soluble K20 below 114 em. MgO, CaO and Sr are approximately 3 orders of magnitude below the bulk soil and also decrease with depth, though MgO concentrations between 22 and 61 em fall slightly higher than this general trend. Water soluble Ba is also 3 orders of magnitude below the bulk soil; however, Ba does not decrease with depth and tends to fluctuate by 100 % throughout the profile. Ti02, Fe203, V, Y, Nb, REE's and Ta all tend to increase moderately with depth, especially below 96 em. However, the water-soluble yield for these elements is quite low, 3-4 orders of magnitude below the bulk soil. Generally, deeper samples do have overall higher REE concentrations than the more shallow samples and there is a negative Ce anomaly in all samples. Ga also increases with depth but is only 3 orders of magnitude below the bulk soil. Ah03, Zr, Hf, Tl, Th and U are all fairly constant with depth and are all depleted by more than 3 orders of magnitude relative to the bulk soil. In contrast, MnO, P20 5, Co and Zn all decrease with depth in the water leach are ~ 3 orders of magnitude below the bulk soil. 3. 4 Carbonate and exchangeable fraction The concentrations of the major elements in the carbonate and exchangeable leach are reported as oxides in Table 5 (Na20 is not reported because Na is in the buffered acetic acid wash used in this step). The concentrations of minor and trace elements in the 107 carbonate and exchangeable leach are reported in Table 6. Rb concentrations fall below the limit of detection (Table 6). Concentrations of all major elements, Sc, Zn, Sr, Ba and La are plotted against depth in Figures 2 and 3. The alkali metals, K20, Li and Cs, are ~3 orders of magnitude below the bulk soil concentrations. K20 and Cs display a strong surface depletion down to 14 ern juxtaposed on a C-shaped profile with minima between 78 and 114 em and a significant increase below 132 ern. Li is fairly constant with depth except for some high outliers between 132 and 149 ern. The alkaline metals, MgO, CaO, Be and Sr, are 2 to 3 orders of magnitude lower than the bulk soil concentrations. MgO and CaO both display a shallow, 18 ern deep, surface depletion juxtaposed on a profile that generally decreases with depth. Sr is very similar to MgO and CaO, but shows an increase in concentration below 132 ern not seen in the major elements. Be and Ba are both fairly constant between 2 and 114 em, but increase significantly below 114 ern. Ba in the carbonate and exchangeable leach represents up to 30% of the bulk soil Ba. Sc, Cu, Y, Ga and REE's generally increase with depth in the carbonate and exchangeable leach and are 2 to 3 orders of magnitude below the bulk soil concentrations. The Cu data does show a high outlier at 61 ern. Sc increases are quite significant such that Sc concentrations in the carbonate and exchangeable fraction at the base of the profile approach 10% of the bulk soil concentrations. As for the REE's, there is a clear negative Ce anomaly in all samples, a strong positive Eu anomaly in the shallow samples and a slight negative Eu anomaly in the deep samples. Cr concentrations are reduced by 108 more than 3 orders of magnitude relative to the bulk soil; however, there is no clear trend in the data. P20s, Ti02, MnO, V, Co, Ni, Zn, Nb and Pb all tend to decrease with depth in the carbonate and exchangeable leach and are 2 to 3 orders of magnitude below bulk concentrations. However, P20s, Ti02, MnO, V, Zn and Pb show an 18 em surface depletion juxtaposed on the depth-dependant decrease and there are high outliers in the Ni data between 22 and 30 em. Fe20 3, Tl and U increase between 2 and 52 em and remain fairly constant below. Concentrations ofFe203 in the carbonate and exchangeable leach are ~ 3 orders of magnitude below the bulk soil; however,> 20% of bulk Tl and U is released in the carbonate and exchangeable leach. Ah03, Zr, Hf and Th increase between 2 and 52 em and decrease below 132 em. The yield of these elements in the carbonate and exchangeable leach is 2 to 3 orders of magnitude below the bulk soil concentrations. 3. 5 Oxide fraction The concentrations of the major elements in the oxide leach are reported as oxides in Table 7. The concentrations of minor and trace elements in the oxide leach are reported in Table 8. Cs, Tl and Ni concentrations fall below the limit of detection (Table 8). Concentrations ofthe major element oxides, Zn, Sr, Ba and La are plotted against depth in Figures 2 and 3. Approximately 10 % of the bulk soil K20 and Rb are released in the oxide leach and the both tend to decrease from the surface down to 61 em, below which the 109 concentration is low and constant. Li falls 2 orders of magnitude below bulk soil levels and decreases down to 61 em; however, Li increases below 114 em. Some alkaline metals, MgO, CaO and Sr, are lower in the oxide fraction by ~3 orders and tend to behave like the alkali metals: decrease from the surface down to 61 em, below which the concentration is low and constant. Ba in the oxide leach is also ~ 3 orders of magnitude below the bulk soil and generally decreases with depth; however Ba concentrations fluctuate significantly. Over half of the total Be is in the oxide leach. Moreover, Be decreases between 2 and 14 em, increases down to 87 em and remains relatively constant below that depth. The concentrations ofMnO, Sc, Co, Zr, Hf and Th are constant between 2 and 114 em and increase linearly below. 1 to 10% of the bulk soil MnO, Co and Zr is released in the oxide leaches while the yield of Sc, Zr, Hf and Th is less than 0.1 % of the bulk. To a lesser extent, Ti02, Fe20 3, V, Cu, Nb, Ph U share the same behavior, but the increase below 114 em is less significant. Less than 1% ofbulk TiOz, between 1 and 10 % ofbulk Fe20 3 and Nb and as much as 10 to 30% of bulk soil V, Cu, Pb and U are released in the oxide leach. Y and the REE's tend to increase continually with depth in the oxide fraction which constitutes ~ 1 % of the bulk soil concentrations. REE tend to be more concentrated in the deeper samples. There is also a significant negative Ce anomaly in all samples and a negative Eu anomaly in some samples. Ah03, P20 5 and Zn all decrease with depth. Ah03 in the oxide leach represents > 10% of the bulk soil concentration. Ah03 is fairly constant between 2 and 114 and decreases linearly below. P20 5 decreases between 2 and 78 em and below 114 em, but 110 <10% ofthe bulk soil P20s is released in the oxide leach. ~10% ofthe bulk soil Zn is released in the oxide leach where an 18 em deep surface enrichment is juxtaposed on a bimodal distribution of relatively high concentrations above 61 em and relatively low below. The decameter-scale variation within Ga and Ta concentrations are greater than any observable overall trend. Approximately 10% ofbulk Ga is in the oxide leach; however, oxide leach Ta is <1 %of the bulk. 3. 6 Organic fraction The concentrations of the major elements in the organic leach are reported as oxides in Table 9. The concentrations of minor and trace elements in the organic leach are reported in Table 10. V and Cr concentrations fall below the limit of detection (Table 10). Concentrations of the major element oxides, Cu, La, Ph and U are plotted against depth in Figures 2 and 3. The alkali metals, K20, Li, Rb and Cs all decrease with depth in the organic leach with removes 1 to 10 % of these elements from the soil. K20, Rb and Cs decrease significantly between 2 and 61 em and remain constant below that depth while Li data tends to decrease throughout the profile, though the data is more scattered. Cs data from 167 em is anomalously high. Most of the alkaline metals, MgO, CaO, Sr and Ba, also decrease with depth. MgO and Ba in the organic leach are more than 2 orders of magnitude lower than the bulk soil concentrations. 3-4% ofbulk CaO and nearly 10% ofSr is released in the organic leach. MgO data below 158 em and CaO, Sr and Ba data between 96 and 123 em 111 fall above the overall tendency to decrease with depth. Be is fairly constant with depth and more than 10% of total Be is in the organic leach; however, the data is scattered. Most of the other elements analyzed also generally decreased with depth: Ah03, P20s, Ti02, MnO, Fe203, Co, Ni, Cu, Zn and Ga. ~10% ofthe bulk soil Ah03, P20s, Ni, Zn and Ga and nearly 50 % of bulk Cu are released in the organic leach. Organic Ti02 falls 3 orders of magnitude below the bulk soil and only~ 1 %of bulk MnO, Fe20 3 and Co were released. Some data points fell above this decreasing trend. Ah03, P20s, Ti02, Cu and Ga have high outliers between 87 and 114 em; CaO, MnO and Fe203 have high outliers between 52 and 61 and 87 and 114 em; and Zn data is high between 52 and 61 em. Zr, Nb,Y, REE's, Hf, Ta and Th are all fairly immobile, though these elements do exhibit a slight decrease with depth. Most of these elements also have high outliers from 52 to 61 and/or 87 to 114 em. And ~ 1 % of most of these elements are released in the organic leach though ~30% of the bulk REE's are released. Samples below 149 em have a negative Ce anomaly and all other samples have clear positive Ce anomalies. All samples show a clear a negative Eu anomaly in the organic leach. Only Sc and Pb increase with depth in the organic leach. Sc falls 2 orders of magnitude below the bulk soil; however, more than 10% ofthe Pb in the soil was released in the organic leach. 112 4. Discussion 4.1 Surface-enrichments in the bulk soil The elemental concentrations of bulk soil profiles are characterized by enrichments at the surface relative to deeper sections of the soil profile. In many cases, elemental concentrations are even higher than that of the proto lith. There are two ways to achieve these concentration enrichments: 1) selective removal of soluble elements during chemical weathering concentrates immobile or refractory elements retained in the soil residue, or 2) external inputs, such as atmospheric deposition, re-enrich the soil column. Dissolution alone cannot explain the observed enrichments of various elements relative to the protolith. The maximum concentration enrichments allowed by mass dissolution are estimated below. AI, Ti, Zr, Nb, Hf and Ta are the least mobile elements in the bedrock and soil, hence, changes in their concentrations in the bulk soil should reveal any bulk mass losses or gains (Brimhall et al., 1991; Brimhall et al., 1988; Chadwick et al., 1990; Kurtz et al., 2000; Little and Lee, 2006; Melfi et al., 1996; Sak et al., 2003; Sastri and Sastry, 1982; Taylor and Blum, 1995; Teutsch et al., 1999). If any two of these elements were perfectly immobile, their ratio would be constant throughout the soil and bedrock proto lith. Figure 4a shows that all soil and protolith samples have a similar Zr/Hf ratio, i.e. they plot on the protolith-origin line, suggesting that Zr and Hf are relatively immobile in the Meru soils (element-element ratios of AI, Ti, Nb and Ta also plot close to the protolith-origin line). The net(%) mass loss/gain, f!Mass, can be estimated from the soil/proto lith ratios of any of these immobile elements, i.e. AMass = 100 * [ C~r~tolith 1] , csoi/ 113 where C~rotolith is the protolith/soil ratio of any immobile element. Figure 4b plots the c~oil 11Mass calculated from Al, Ti, Zr, Nb, Hf and Ta. Zr, Hf, Al and Ta all predict approximately 10 to 30 % mass losses throughout the profile. Ti and Nb behave slightly differently: Ti predicts mass losses >40% between 61 and 141 em and Nb predicts mass gains between 2 and 123 em. Given the consistent behavior of Zr, Hf, Aland Ta and the well-behaved Zr/Hf ratios, Zr is used to calculate mass losses throughout the soil column. Figure 4c plots the concentration ratios of soil to protolith averaged over the top 114 em. The dashed line, which passes through the Zr soil/protolith ratio, represents the maximum enrichment that can be achieved by a perfectly conservative (immobile) element during chemical weathering. The soil/proto lith ratios of many elements, most notably the REEs, fall on this line. However, Mg, P, Ca, Ti, Mn and Fe ratios fall above the Zr line. The enrichments in these elements thus cannot be solely explained by the concentrating effects of mass loss on conservative elements during soil formation. It can be seen in Figure 4c that most element enrichments are confined to the upper 114 em, which is the reverse of that expected for typical soil profiles where the shallowest samples tend to be the most weathered and hence most depleted in soluble elements. The decrease in soil/proto lith ratio for many soluble elements is denoted by the endpoints of the gray triangles representing the average soil/proto lith ratio below 150 em. The Mg# of the bulk soil also follows a reverse depth zonation. It is high at depths shallower than 114 em but similar to the protolith deeper than 150 em. Collectively, these observations suggest that the soil shallower than 114 em was formed from or contains a significant component of materials, which are higher in Mg, P, Ca, Ti, Mn and Fe relative to the weathered soil residue at depth. 114 There are a number of ways to explain these surface enrichments. They could 1) be artifacts of a layered or heterogeneous proto lith or 2) derive from surface-deposited material external to the present-day soil column (Capo and Chadwick, 1999; Chadwick et al., 1999). Another possibility is that this soil profile represents an inverted weathering column, wherein the lower part is more deeply weathered due to enhanced weathering associated with subsurface water flow (Little and Lee, 2006). The subsurface water flow scenario was proposed by Little and Lee (2006) for a lateritized soil column on the slopes ofMount Kilimanjaro,just east ofMt. Meru. However, in that study, there was a substantial increase in immobile elements (Zr, Hf, Nb, Ta) with depth, corresponding to enhanced mass loss at depth. The lack of any significant variation in immobile elements with depth in the Meru site suggests that there .are probably not strong depth zonations in the degree of chemical weathering in the present study, so the subsurface water flow scenario is not applicable to the Meru site. Thus, only the first two hypotheses are likely to pertain to the Meru site. While a heterogeneous protolith cannot be ruled out, a number of observations seem to require some addition of a foreign soil component, such as dust. Na and K increase towards the surface yet Na20/K20 ratios are constant throughout the soil and protolith despite major changes in bulk soil composition (Figure 4d). Given the high solubility ofNa and K, it seems highly unlikely that Na and K can be elevated at shallow depths and that the Na!K ratio can remain constant in soil residues. This dilemma can be resolved if the shallow soil had been "re-fertilized" by less weathered materials with higher N a and K and similar Na!K ratios as the protolith itself. One possibility could be the addition of dust in the form of juvenile volcanic ash. Such a scenario would also be consistent with the 115 enrichments in Ni and Mg towards the surface. These elements might be expected to be leached out during chemical weathering, but they could be re-enriched if fresh volcanic ash (e.g., unweathered) had been introduced. In the following sections, the processes which may control the behavior and distribution of elements within the soil column are discussed. 4. 2 Carbonate and exchangeable fractions The clay fraction probably contains kaolinite and illite because plagioclase (~30% of thin section cross area) and high concentrations of alkali metals characterize the protolith. However, kaolinite has a low cation exchange capacity, which is further reduced by the low pH of the soils(< 6.0), and thus it does not contribute significantly to the exchangeable fraction. The high exchange capacity swelling clays (e.g. montmorillonite) are not abundant because no significant volume change occurred during the water leach. Therefore, illite is probably the primary absorbent in the clay fraction. Illite, like all 2:1 clays, has a very strong affinity for monovalent K+ which is incorporated between basal oxygen on the tetrahedral sheets. Above 123 em, this exchangeable K20 fraction mirrors the bulk soil (Fig. 2), indicating the presence of a shallow illite-bearing horizon between 6 and 30 em. However, the exchangeable K20 concentration spikes below 123 em so that> 7% of the bulk K20 is in the exchangeable fraction at these depths. This spike may indicate the presence of a secondary, deeper illite-bearing layer. The dissolution of carbonates probably controls the yield of CaO and MgO in the carbonate and exchangeable leach because they should not be significantly absorbed by 116 illite or kaolinite. Ca and Mg are the most abundant carbonate-forming cations in the soils and they are both surface enriched down to 61 em, thus the total amount of carbonate may decrease below 61 em; however, increases in Sr and changes in Sr/CaO ratios below 114 em indicate that the composition of the carbonate changes significantly (Fig. 3). Sr/CaO ratios can be used to track the sources of Ca in natural systems because Ca and Sr are often incorporated into the same minerals (Blum et al., 2002). The bulk soil Ca and Sr are not correlated; however, in the carbonate and exchangeable leach, Sr/Ca ratios are correlated (Fig. 6a). The carbonate fraction of soil samples above 123 em and the protolith share the same Sr/CaO ratio; however, the ratio changes drastically below 114 em suggesting a change in the Sr/CaO ratio of the carbonate minerals being formed at these depths. This change in Sr/CaO is unique to the carbonate leach: the Sr/CaO ratio of the water soluble fraction (removed before the carbonate leach) and the oxide fraction (removed after the carbonate leach) do not deviate from the protolith ratio (Fig. 6b). Like Sr and Ca, Ba also occurs as a divalent cation that forms weakly soluble carbonates in nature (Sulkowski and Himer, 2006; Velde, 1992). Ba/CaO behaves like Sr/CaO in the carbonate and exchangeable leach (Fig. 6c ). However, below 114 em, the Ba/CaO ratio of the soil increases dramatically such that >30% of the bulk Ba is in carbonate at these depths (Figs. 5a and 6c ). Carbonate and exchangeable Sc increases exponentially with depth, such that the greatest change in concentration occurs below 114 em while the REE' s, represented by La, increase linearly with depth (Fig. 3). U increases more abruptly such that below 78 em, >35% ofbulk U in the bulk soil is in the carbonate (Fig. 5a). Neither Sc, the REE's 117 nor U correlate with CaO (Fig. 6d), but below 114 em, high concentrations of carbonate and exchangeable K, Sr, Ba, Sc, REE's and U are available. 4. 3 Scavenging ofmetals by metal hydroxides More than 10% of bulk AI, Fe and Mn are released in the oxide leach indicating the presence of metal hydroxides and oxyhydroxides common to highly weathered tropical soils (Alloway, 1995). Al-oxides do not effectively scavenge other elements; however Figure 7 shows a positive correlation between Ti, V, Cu, Co and Pb and Fe or Mn, consistent with reported Fe and Mn oxide scavenging behavior (Alloway, 1995; Keller and Vedy, 1992; Teutsch et al., 1999). This positive, linear correlation implies that increases in Fe or Mn oxides will result in the absorption and hence retention of labile Ti, Cu, V, Pb, Co or Zr. Although Ti, Zr and Fe are highly correlated, the very low ratios ofTi02/Fe203 and Zr/Fe203 in the oxide leaches (Fig. 7a) must be due to the low availability of labile Ti and Zr during soil formation and this is consistent with the relatively immobile behavior ofTi and Zr. The high ratios ofCu, Co, V and Pb relative to Fe and Mn in the oxide leaches indicates higher availability of these elements for absorption. These elements are mobilized during chemical weathering, but some fraction is re-immobilized by absorption onto Fe-Mn-oxyhydroxides. A feature of interest is where the regression line of a given metal versus Fe or Mn intercepts. Most of the elements (Pb, Co, Ti, Zr) plotted against Fe or Mn (Fig. 7) have regressions that go through the origin or just slightly above the origin. These elements are probably quantitatively adsorbed by Fe-Mn oxyhydroxides. In contrast, the Cu and V versus Fe20 3 regressions intercept they-axis (Cu or V) at negative values, which implies 118 that although these elements are scavenged by Fe-Mn oxyhydroxides, they are probably not as efficiently scavenged as Pb, Co, Ti and Zr. 4. 4 Metals in the organic fraction A large fraction of the bulk Alz03, Cu, La (REE's), Pb and U was released in the organic leach (Fig. 5c). The presence of Al-organic complexes above 123 em could explain why > 10 % of the bulk Alz03 is released in the organic leach (Fig. 5c ). Such complexes were found to be a significant pool of AI in soils with a similar pH(~ 5.5) formed from volcanic ash in France (Prevosto et al., 2004) .. Bulk soil enrichments of Cu and Pb are often associated with organic matter and most of the bulk enrichment of Cu above 123 em and ofPb below 26 em can be accounted for in the organic fraction (Alloway, 1995; Gibson and Farmer, 1986; Keller and Vedy, 1992; Palumbo et al., 2000; Rahman et al., 1996; Teutsch et al., 1999). REEs are relatively immobile a yield of>50 % of the bulk soil REEs in the organic leach was unexpected (Fig. 5c ); however, the capacity of organic coatings to absorb REEs in seawater has been demonstrated, thus it is reasonable that a significant portion of REEs are incorporated into organic matter (Byrne and Kim, 1990). The organic fraction duplicates for all elements analyzed are less consistent than the oxide, carbonate and exchangeable, water soluble and bulk fractions (Figs. 2 and 3). The inconsistent solubilization of the organic-bound soil fraction, the oxidation of sulfides by H20 2, or the solubilization of some clay minerals may all contribute to the scatter (Keller and Domergue, 1996; Marin et al., 1997; Sulkowski and Rimer, 2006; Tessier et al., 1979). 119 5. Conclusions A detailed leach study on a volcanic soil from Mt Meru in northern Tanzania was presented. The bulk behavior of the relatively immobile elements, Zr, Hf and Al, predicts a 10 to 30% mass loss throughout the soil column. Superimposed on this mass loss are enrichments ofMg, P, Ca, Ti, Mn and Fe which most likely derive from a layered protolith or from the surface deposition of volcanic ash. The behavior of the major and minor elements in the carbonate and extractable leach indicate the presence of a shallow, K-bearing clay layer above 30 em; Ca carbonate above 61 em; and below 114 em both Sr and Barich carbonates and a secondary K-bearing clay layer. Al, Fe and Mn oxides were also present throughout the soil column, though Fe and Mn oxides were most effective at scavenging Cu, Pb, Co and Cu below 114 em. Metal-organic complexes sequestered significant fraction of Al, Cu, REE's and Pb throughout the soil column. 120 Table 1. Bulk physical properties, major element concentrations (LOI corrected) and extent of chemical weathering % 2cm 10 18 26 43 52 61 78 87 96 w.c. 48 65 52 87 71 83 79 63 67 67 LOI 29 25 17 26 23 24 23 21 22 18 Porosity 81 79 75 82 78 81 81 78 77 77 pH 5.7 5.9 6 5.8 5.7 5.7 5.6 5.6 5.6 5.6 Na20 3.72 5.8 5.66 3.35 3.26 2.54 1.84 1.44 1.26 1.22 MgO 3.38 3.59 2.99 2.89 3.02 3.64 3.97 4.07 3.93 3.52 Ah03 22.36 23.98 22.22 22.14 21.48 24.37 23.22 20.28 21.44 21.4 P20s 1.36 1.03 0.97 1.01 1.06 1.14 1.5 1.23 1.08 1.1 K20 2.4 3.35 3.26 2.26 2.1 1.77 1.21 0.94 0.93 0.97 CaO 10.69 10.87 9.1 8.81 8.77 9.77 9.95 9.39 9.04 7.9 Ti02 2.76 2.77 2.53 2.06 2.47 2.58 3.45 3.39 3.2 2.95 MnO 0.31 0.29 0.25 0.3 0.29 0.34 0.36 0.31 0.3 0.29 Fe203 12.13 12.47 10.55 10.31 11.17 13.25 14.78 13.71 13.82 13.22 Si02 :t 41 36 42 47 46 41 40 45 45 47 CIA 0.61 0.59 0.6 0.65 0.64 0.66 0.66 0.65 0.68 0.7 MW# 0.36 0.36 0.36 0.36 0.35 0.35 0.35 0.37 0.36 0.35 % 114cm 123 141 149 158 167 177 Prot LOD w.c. 70 67 49 50 55 54 55 LOI 16 16 13 12 7 9 2 1 Porosity 79 76 72 72 73 70 71 pH 5.5 5.6 5.6 5.5 5.5 5.6 5.6 Na20 1.23 1.25 1.54 1.77 1.54 1.9 1.54 10.45 0.00015 MgO 4.15 3.93 3.12 2.85 2.46 2.34 2.13 1.72 0.00014 Ah03 20.52 21.06 21.86 25.1 22.45 23.8 22.21 19.12 0.001 P20s 1.02 0.96 0.89 0.78 0.62 0.71 0.56 0.41 0.00011 K20 1.01 1.05 1.2 1.43 1.34 1.5 1.42 5.63 0.0053 CaO 8.82 8.51 7.44 6.96 6.17 6.02 5.64 7 0.002 Ti02 3.27 3.28 3.42 2.81 2.94 3.15 2.94 2.09 0.00014 MnO 0.3 0.3 0.28 0.26 0.28 0.3 0.28 0.23 7.4E-06 Fe203 14.5 13.77 11.75 12.42 10.81 10.92 9.9 8.11 0.00059 Si02 t 45 46 49 46 51 49 53 45.24 CIA 0.67 0.68 0.71 0.74 0.74 0.75 0.76 0.52 Mg# 0.36 0.36 0.34 0.31 0.31 0.30 0.30 0.30 Water content (W.C.), loss on ignition (LOI), porosity and element oxides are reported as a percent. All soil concentrations determined using medium resolution ICP-MS and protolith (Prot) concentrations using XRF. The Chemical Index of Alteration (CIA) is 121 defined as A/203 and the molar Mg number (Mg#) is defined as A/20 3 + CaO + Na20 molar Mg Mg+Fe t Si02 soil concentrations were estimated as the difference between the sums of the major element oxides for each soil sample. 122 Table 2. Bulk minor and trace element concentrations by depth l!l!m 2cm 10 18 26 43 52 61 78 87 96 Li 20.9 23.5 23.0 22.4 21.7 24.4 19.9 14.4 15.6 15.7 Be 6.40 6.47 6.36 6.95 7.11 7.70 7.72 6.62 6.89 6.88 Set 9.94 11.1 8.41 7.74 8.10 11.1 12.1 12.8 12.4 11.1 Vt 237 250 206 204 210 264 292 280 279 232 Crt 23.3 24.9 21.6 16.9 17.8 25.2 28.5 22.8 12.8 11.7 Cot 28.2 29.3 26.1 24.4 25.4 32.3 36.2 35.9 35.3 33.8 Nit 10.8 12.1 11.1 11.5 8.85 11.7 13.0 12.3 12.2 12.3 Cut 55.7 46.3 44.6 82.5 48.0 54.3 49.5 49.7 55.2 65.1 Znt 361 351 269 287 326 352 368 281 239 236 Ga 32.4 31.9 30.2 31.1 31.8 36.1 35.7 33.2 34.4 33.7 Rb 71.2 77.5 105 60.9 57.7 54.5 37.8 22.6 22.3 23.3 Sr 1464 1605 1530 1252 1196 1101 951 819 792 768 y 48.8 46.5 43.9 47.0 46.6 55.6 54.3 49.8 52.3 51.1 Zrt 603 629 562 589 574 666 661 633 654 599 Nbt 185 163 179 137 155 149 197 201 181 181 Cs 1.22 1.21 1.27 1.11 1.10 1.57 1.15 0.503 0.436 0.466 Ba 1758 1994 1646 1612 1615 1828 1396 1116 1219 1327 La 195 190 172 188 178 236 208 176 198 196 Ce 310 315 286 291 271 321 298 293 327 319 Pr 28.7 29.7 26.0 27.5 26.3 33.0 30.3 28.9 32.5 31.2 Nd 99.0 97.7 88.5 91.7 89.8 108.4 104.8 104.0 112.6 108.7 Sm 16.3 16.0 14.7 15.0 15.3 17.8 17.9 17.9 19.0 18.4 Eu 5.88 5.98 5.38 5.47 5.62 6.36 6.26 5.98 6.40 6.40 Tb 2.75 2.62 2.52 2.50 2.70 2.91 3.26 3.17 3.08 3.08 Gd 18.2 17.6 16.7 16.9 17.8 21.0 21.0 21.0 23.1 22.0 Dy 13.0 12.5 11.8 12.2 12.7 14.6 15.2 14.9 15.7 15.5 Ho 1.60 1.54 1.43 1.50 1.51 1.75 1.74 1.67 1.78 1.72 Er 4.14 3.98 3.75 3.88 3.92 4.48 4.49 4.29 4.52 4.38 Tm 0.560 0.539 0.507 0.526 0.532 0.597 0.603 0.574 0.606 0.584 Yb 3.58 3.43 3.28 3.37 3.50 3.95 3.97 3.92 4.08 3.95 Lu 0.465 0.443 0.417 0.434 0.444 0.500 0.506 0.490 0.513 0.495 Hf 10.4 9.88 8.93 9.37 8.99 10.6 10.5 10.3 10.4 9.89 Ta 11.1 8.71 8.85 9.24 8.66 9.17 9.80 10.76 7.93 9.09 Tl 0.352 0.327 0.344 0.377 0.381 0.618 0.460 0.390 0.347 0.387 Pb 22.8 20.7 19.2 21.3 22.7 25.9 24.6 22.9 23.5 23.2 Th 22.3 22.3 20.6 21.1 20.5 22.3 21.8 21.6 22.2 21.1 U:j: 5.12 5.17 4.68 4.93 4.91 5.15 4.99 5.15 5.78 5.44 ~Values determined using medium resolution ICP-MS; all other values determined using low resolution ICP-MS 123 Table 2. continued ~~m 114cm 123 141 149 158 167 177 Prot LOD Li 17.0 16.9 19.1 26.0 18.4 13.8 10.9 15.8 0.0090 Be 6.58 6.47 6.10 7.02 6.43 7.12 6.73 6.45 0.0092 Set 12.9 12.4 9.84 9.19 8.43 8.37 7.63 3.63 0.0042 Vt 267 262 231 172 204 198 198 153 0.011 Crt 12.2 8.41 7.98 7.66 6.48 14.79 5.97 2.94 0.87 Cot 36.7 35.6 29.4 26.1 26.5 27.9 26.3 12.4 0.0047 Nit 12.8 11.8 9.92 9.12 7.49 7.88 6.67 1.45 0.033 Cut 61.4 60.2 53.2 56.7 63.9 63.3 57.9 15.4 0.25 Znt 253 218 224 163 161 166 158 142 0.29 Ga 33.8 32.9 31.8 35.0 32.6 34.1 31.0 27.0 0.0036 Rb 22.6 22.9 21.8 26.3 25.5 36.0 25.6 121.8 0.0053 Sr 788 805 1063 1214 1224 1424 1347 2062 0.0056 y 50.8 49.1 42.5 47.1 48.9 54.0 52.3 41.8 0.0012 Zrt 592 628 667 573 648 644 622 523 0.058 Nbt 178 170 233 218 201 245 218 207 0.0095 Cs 0.451 0.419 0.396 0.371 2737 2312 2090 1.22 0.00027 Ba 1438 1522 2266 2861 180 200 187 2049 0.033 La 183 182 149 172 328 341 317 158 0.0013 Ce 308 319 315 336 30.8 34.6 32.9 261 0.0020 Pr 31.4 31.0 27.1 30.1 107.9 116.7 113.7 23.8 0.00042 Nd 112.0 108.6 94.8 104.5 18.2 19.7 19.0 78.5 0.0018 Sm 19.1 18.8 16.3 17.7 7.23 7.35 7.06 12.9 0.0010 Eu 6.71 6.48 6.54 7.27 2.95 3.27 3.20 5.43 0.00036 Tb 3.37 3.02 2.72 2.98 21.5 24.2 23.2 2.33 0.00010 Gd 23.1 22.0 19.3 21.1 15.0 16.5 15.8 14.8 0.00047 Dy 16.1 14.7 13.4 14.8 1.72 1.88 1.80 11.0 0.00067 Ho 1.75 1.71 1.52 1.71 4.36 4.89 4.70 1.36 0.00012 Er 4.45 4.29 3.85 4.30 0.592 0.669 0.640 3.60 0.00037 Tm 0.591 0.577 0.520 0.582 3.95 4.43 4.22 0.495 0.000062 Yb 3.98 3.89 3.56 3.91 0.497 0.562 0.542 3.15 0.00043 Lu 0.499 0.485 0.441 0.486 11.1 11.5 11.1 0.406 0.00055 Hf 10.1 10.5 10.8 10.9 10.18 10.97 10.77 8.59 0.00085 Ta 7.94 7.98 11.61 8.97 0.531 0.672 0.584 8.39 0.00048 Tl 0.445 0.423 0.506 0.489 20.5 21.9 21.1 0.168 0.00048 Pb 22.1 20.7 19.6 21.7 22.1 23.3 21.9 14.7 0.0076 Th 20.2 20.0 20.9 23.5 5.26 5.33 5.01 18.0 0.00039 u~ 4.79 4.71 4.70 5.64 18.4 13.8 10.9 3.53 0.0017 ~Values determined using medium resolution ICP-MS; all other values determined using low resolution ICP-MS 124 Table 3. Water soluble major element concentrations by depth J.!J.!m 2cm 6 10 10 14 14 18 22 26 30 Na20 242.7 308.5 407 477.5 437.4 457.9 367.5 641.4 342.8 334 MgO 214.6 155.5 103 110.6 82.6 88.07 64.32 149.4 162.8 163.5 AhOJ 65.63 53.14 98.63 61.61 75.2 70.41 56.27 50.24 36.55 49.08 P20s 22.22 16.9 13.81 16.76 9.758 12.05 7.237 12.56 12.99 11.17 K20 368.9 372 334.6 387.9 291.6 313 235.8 294.3 236.3 186.8 CaO 1067 899 557.6 627.1 388.4 432.7 267.2 498.4 427.7 339.3 Ti02 1.341 1.169 2.463 1.281 1.738 1.688 1.082 1.373 0.86 1.254 MnO 28.19 24.39 12.53 13.08 8.118 8.101 6.203 14.37 16.84 16.91 Fe203 13.28 11.37 24.94 12.01 14.66 13.7 16.33 12.88 8.316 11.24 J2J2M 43cm 52 61 61 78 87 96 96 114 123 Na20 301.1 243 206.1 248.4 263.9 221.8 286.3 322.3 252.6 232.8 MgO 166.5 193.3 152.8 176.2 49.02 24.75 21.74 25.76 15.46 15.3 Ah03 51.78 67.12 53.08 102.2 54.9 42.18 73.16 50.68 58.26 72.9 P20s 10.68 10.16 6.827 12.49 6.455 38.59 3.755 4.867 4.728 2.756 K20 161.2 111.1 77.92 91.29 39.72 14.57 5.468 10.95 15.71 28.4 CaO 358 252.2 201.1 235.9 131 90.86 57.91 73.03 63.39 79.65 Ti02 1.309 2.028 2.023 3.924 2.424 2.06 3.362 2.248 2.869 3.1 MnO 16.6 24.12 27.16 30.2 16.31 8.664 10.58 11.22 9.723 9.41 Fe203 12.8 16.55 20.36 27.09 14.14 12.75 18.74 12.95 20.35 16.7 J2J2M 132cm 132 141 149 149 158 167 177 177 LOD Na20 185.9 223.9 194.7 166.2 174.8 129.7 142.5 131.2 124.9 11 MgO 6.69 12.43 4.79 3.937 4.133 1.15 2.586 4.158 2.183 5.8 AhOJ 63.03 167 102.8 140 53.27 63.91 109.5 83.73 62.19 7 P20s 1.894 6.884 2.605 3.008 3.385 1.155 3.799 4.01 3.713 4.9 K20 28.68 36.62 41.47 72.05 69.96 136.2 105.4 119.6 112.8 21 CaO 36.45 55.88 34.92 27.79 30.93 17.78 20.89 58.57 21.11 10 Ti02 3.088 8.767 5.07 8.513 2.898 4.996 7.695 6.182 4.748 3.4 MnO 3.668 5.324 3.595 2.741 1.908 1.53 1.475 1.388 1.074 0.36 Fe203 16.42 45.82 27.59 46.18 16.12 36.34 37.21 31.61 23.31 6.8 All values determined using medium resolution ICP-MS 125 Table 4. Water soluble minor and trace element concentrations by depth ppm 2cm 6 10 10 14 14 18 22 26 30 43 Li 0.0448 0.0409 0.0223 0.0272 0.0252 0.0292 0.0268 0.0517 0.0334 0.0408 0.0271 Be 0.0047 0.0088 Set 0.0006 0.0005 0.0004 0.0004 Vt 0.1049 0.087 0.0945 0.0946 0.0954 0.0932 0.0681 0.0753 0.0698 0.0718 0.0612 Crt 0.029 0.0531 Cot 0.0587 0.0618 0.0519 0.0586 0.0487 0.0531 0.0389 0.0519 0.0479 0.04 0.041 Nit 0.019 0.039 0.0109 0.0385 0.0274 0.0399 0.0349 0.0381 0.0341 0.0694 Cut 15.6 1.57 0.152 Znt 3.27 9.18 6.43 1.9 0.804 4.82 16.5 3.23 Ga 0.0113 0.0091 0.0193 0.0137 0.0203 0.0188 0.0193 0.0145 0.0103 0.0123 0.0119 Rb 1.81 1.37 1.12 1.2 1.31 1.36 1.3 1.69 1.51 1.34 1.24 Sr 22.12 19.3 12.2 12.39 9.29 9.25 6.05 11.12 9.54 8.71 8.56 y 0.0312 0.0276 0.0302 0.0254 0.025 0.0232 0.0238 0.0379 0.0315 0.0342 0.0305 Zrt 0.262 0.262 0.231 0.239 0.176 0.194 0.13 0.19 0.214 0.262 0.195 Nbt 0.0546 0.047 0.0686 0.0483 0.0515 0.0569 0.0349 0.0453 0.0342 0.0453 0.0448 Cs 0.0096 0.0097 0.0097 0.0103 0.0095 0.0096 0.0064 0.0103 0.0143 0.015 0.011 Ba 1.84 2.51 1.14 0.804 0.479 1.51 0.226 0.599 4.77 1.27 1.17 La 0.0757 0.0684 0.0872 0.0682 0.0745 0.065 0.0596 0.0941 0.0854 0.0998 0.0818 Ce 0.0488 0.0437 0.0942 0.0632 0.0812 0.0733 0.0704 0.0885 0.0626 0.0738 0.0643 Pr 0.0098 0.009 0.0118 0.0091 0.0103 0.0087 0.0084 0.013 0.0111 0.0127 0.0104 Nd 0.0349 0.0313 0.0415 0.0321 0.0347 0.0303 0.0299 0.0463 0.0398 0.0447 0.0366 Smt 0.0052 0.0045 0.0059 0.0048 0.0051 0.0044 0.0043 0.0069 0.0057 0.0065 0.0053 Eu 0.0021 0.0021 0.0021 0.0016 0.0016 0.0016 0.0014 0.0022 0.003 0.0023 0.0018 Tb 0.0009 0.0008 0.0011 0.0007 0.0009 0.0006 0.0007 0.0011 0.0009 0.0011 0.0009 Gd 0.0062 0.0057 0.0074 0.005 0.0063 0.0046 0.0056 0.0085 0.0068 0.0078 0.0064 Dy 0.0047 0.0043 0.0052 0.0038 0.0046 0.0033 0.0041 0.006 0.005 0.0056 0.0047 Ho 0.0009 0.0007 0.0008 0.0007 0.0008 0.0006 0.0007 0.001 0.0008 0.0009 0.0008 Er 0.0024 0.0023 0.0024 0.0021 0.0022 0.0019 0.002 0.0032 0.0027 0.0029 0.0025 Tm 0.0004 0.0004 0.0004 0.0003 0.0004 0.0003 0.0002 0.0005 0.0004 0.0004 0.0004 Yb 0.0024 0.0022 0.0024 0.0021 0.002 0.0018 0.0019 0.0031 0.0027 0.003 0.0026 Lu 0.0004 0.0003 0.0003 0.0003 0.0003 0.0003 0.0002 0.0004 0.0004 0.0004 0.0003 Hf 0.0033 0.0034 0.0031 0.0028 0.0023 0.0024 0.0017 0.0024 0.0027 0.003 0.0026 Ta 0.0006 0.0006 0.0007 0.0004 0.0005 0.0006 0.0004 0.001 0.0003 0.0005 0.0005 Tl 0.0031 0.0033 0.0019 0.0031 0.0011 0.0018 0.0007 0.0023 0.0037 0.0038 0.0035 Pb 0.0517 0.451 0.0628 0.0076 0.0312 0.0268 Th 0.0033 0.003 0.0047 0.0048 0.0041 0.0043 0.0034 0.0038 0.0034 0.0037 0.0036 U± 0.0101 0.0099 0.0097 0.0089 0.009 0.0079 0.0084 0.0107 0.0124 0.0136 0.0118 :t Values determined using medium resolution ICP-MS; all other values determined using low resolution ICP-MS 126 Table 4. continued ppm 52 em 61 61 78 87 96 96 114 123 Li 0.0267 0.021 0.0285 0.0188 0.0088 0.0134 0.0127 0.0141 0.0132 Be Set 0.0004 0.0005 0.0005 0.0005 0.0005 Vt 0.0892 0.0757 0.0993 0.0742 0.0617 0.0731 0.0743 0.0723 0.0792 Crt 0.0272 0.178 Cot 0.0487 0.0553 0.0656 0.0402 0.0189 0.0387 0.0217 0.019 0.0137 Nit 0.0162 0.0269 0.109 0.0351 0.0952 0.199 0.0123 Cut 0.354 0.215 Znt 0.342 1.21 2.25 1.91 Ga 0.0136 0.0114 0.0174 0.0125 0.0125 0.0164 0.0125 0.0171 0.0162 Rb 1.34 1.22 1.33 0.73 0.453 0.494 0.515 0.477 0.475 Sr 6.8 5.28 5.59 2.13 1.17 1.26 1.26 1.42 1.61 y 0.0442 0.0413 0.0477 0.055 0.0472 0.0543 0.0505 0.0653 0.046 Zrt 0.166 0.142 0.201 0.149 0.0862 0.106 0.106 0.103 0.104 Nbt 0.047 0.0466 0.0886 0.0566 0.0515 0.0732 0.0527 0.067 0.0737 Cs 0.0157 0.0135 0.0152 0.0078 0.0042 0.005 0.0053 0.0043 0.0045 Ba 1.62 1.28 1.48 0.742 0.496 0.679 0.601 0.812 1.15 La 0.142 0.121 0.145 0.167 0.158 0.192 0.18 0.247 0.182 Ce 0.0872 0.0674 0.093 0.113 0.124 0.144 0.121 0.174 0.145 Pr 0.0167 0.0148 0.0182 0.0248 0.0245 0.0299 0.0283 0.0415 0.0306 Nd 0.0597 0.0531 0.0631 0.0897 0.0883 0.109 0.101 0.153 0.115 Smt 0.0086 0.0073 0.0089 0.0125 0.0119 0.0148 0.0135 0.0204 0.0149 Eu 0.0028 0.0026 0.0029 0.0038 0.0034 0.0042 0.0039 0.0057 0.0044 Tb 0.0013 0.0013 0.0013 0.0019 0.0018 0.0023 0.0017 0.003 0.0021 Gd 0.0105 0.0093 0.0093 0.0143 0.0134 0.0178 0.0127 0.0234 0.0169 Dy 0.0071 0.0065 0.0069 0.0096 0.0087 0.0108 0.008 0.014 0.0098 Ho 0.0012 0.0012 0.0013 0.0015 0.0013 0.0015 0.0013 0.0018 0.0013 Er 0.0036 0.0034 0.0038 0.0044 0.0038 0.0044 0.0039 0.0051 0.0037 Tm 0.0005 0.0005 0.0005 0.0006 0.0005 0.0006 0.0006 0.0007 0.0005 Yb 0.0033 0.0031 0.0037 0.004 0.0034 0.0039 0.0036 0.0046 0.0033 Lu 0.0004 0.0004 0.0005 0.0005 0.0005 0.0005 0.0005 0.0006 0.0004 Hf 0.0021 0.0018 0.0025 0.0018 0.001 0.0015 0.0012 0.0014 0.0013 Ta 0.0005 0.0008 0.0013 0.0011 0.0011 0.0015 0.001 0.0015 0.0018 Tl 0.0079 0.0057 0.0095 0.0045 0.002 0.0024 0.0039 0.0026 0.0029 Pb 0.0146 0.0341 0.0085 Th 0.0024 0.002 0.0027 0.0019 0.0014 0.0015 0.0018 0.0017 0.0014 U:j: 0.0133 0.0118 0.0116 0.0123 0.0097 0.0115 0.0103 0.0104 0.0091 ~Values determined using medium resolution ICP-MS; all other values determined using low resolution ICP-MS 127 Table 4. continued ppm 132 em 132 141 149 149 158 167 177 177 LOD Li 0.0212 0.0233 0.0155 0.014 0.0067 0.0024 0.0047 0.004 0.016 Be 0.003 0.0027 0.0064 0.0061 0.029 Set 0.0007 0.0007 0 0.0005 0.0044 Vt 0.079 0.119 0.109 0.14 0.104 0.129 0.2 0.177 0.163 0.022 Crt 0.0198 0.0218 0.1228 0.26 Cot 0.0068 0.0193 0.0068 0.0134 0.0041 0.0633 0.0066 0.0059 0.0061 0.016 Nit 0.0166 0.1306 0.072 46.64 0.0216 0.0416 0.12 Cut 0.19 0.539 0.407 2.2 Znt 1.52 0.706 0.484 0.447 1.56 2 Ga 0.0165 0.0315 0.0233 0.0315 0.016 0.0177 0.0247 0.0197 0.0157 0.0045 Rb 0.316 0.358 0.333 0.365 0.364 0.485 0.395 0.4 0.377 0.016 Sr 1.06 1.23 1.1 1.27 1.14 1.92 1.72 1.94 1.82 0.013 y 0.0464 0.0567 0.0522 0.0618 0.0467 0.0463 0.0605 0.0537 0.0476 0.0032 Zrt 0.109 0.199 0.151 0.233 0.114 0.151 0.189 0.172 0.128 0.049 Nbt 0.0795 0.201 0.125 0.205 0.0729 0.1099 0.163 0.131 0.101 0.014 Cs 0.0026 0.0034 0.0022 0.0022 0.002 0.0023 0.0026 0.0028 0.0023 0.0015 Ba 0.975 1.6 1.43 2.66 1.26 2.32 1.3 1.13 0.79 0.02 La 0.198 0.239 0.214 0.231 0.198 0.174 0.192 0.164 0.145 0.0096 Ce 0.178 0.272 0.25 0.29 0.165 0.159 0.178 0.149 0.12 0.0083 Pr 0.0334 0.0412 0.0371 0.0399 0.0343 0.0284 0.0325 0.0287 0.0249 0.003 Nd 0.127 0.15 0.137 0.145 0.126 0.104 0.12 0.106 0.0932 0.013 Smt 0.0165 0.0198 0.0184 0.02 0.0168 0.0141 0.0169 0.0149 0.0129 0.0031 Eu 0.0046 0.0057 0.0053 0.0062 0.0047 0.0043 0.0047 0.0042 0.0035 0.0009 Tb 0.0024 0.0025 0.0027 0.0029 0.0019 0.002 0.002 0.0018 0.0015 0.0006 Gd 0.0185 0.0187 0.0199 0.0217 0.0155 0.0155 0.0164 0.0145 0.0124 0.0036 Dy 0.0107 0.0111 0.0118 0.0138 0.0089 0.0092 0.01 0.0089 0.0075 0.0025 Ho 0.0014 0.0018 0.0016 0.0019 0.0014 0.0013 0.0016 0.0014 0.0012 0.0004 Er 0.0038 0.0047 0.0045 0.0054 0.0039 0.0038 0.0046 0.0042 0.0037 0.0012 Tm 0.0006 0.0007 0.0006 0.0007 0.0006 0.0005 0.0007 0.0006 0.0006 0.0004 Yb 0.0035 0.0042 0.004 0.0049 0.0036 0.0036 0.0046 0.004 0.0035 0.0013 Lu 0.0005 0.0006 0.0006 0.0007 0.0006 0.0004 0.0007 0.0006 0.0005 0.0004 Hf 0.0015 0.0022 0.002 0.0027 0.0015 0.0019 0.0024 0.0021 0.0016 0.0011 Ta 0.0017 0.0037 0.0024 0.0037 0.0015 0.0023 0.0035 0.003 0.0021 0.0008 Tl 0.0018 0.0032 0.002 0.0016 0.0023 0.0015 0.0021 0.0021 0.002 0.0015 Pb 0.072 0.0317 0.0085 0.0172 0.0245 0.0082 0.034 Th 0.0015 0.0038 0.0032 0.006 0.0028 0.004 0.0032 0.0031 0.0022 0.0031 U:j: 0.0073 0.0082 0.0088 0.0118 0.0086 0.0083 0.0092 0.0074 0.0068 0.0043 ~Values determined using medium resolution ICP-MS; all other values determined using low resolution ICP-MS 128 Table 5. Carbonate and exchangeable fraction major element concentrations by depth ppm 2 em 6 10 10 14 14 18 22 26 30 MgO 479.6 498.4 342.7 277.4 155.5 455.7 713.3 854.9 Ah03 1586 1678 1841 1744 2175 2079 1171 2474 2175 2829 P20 5 55.55 45.18 32.9 32.9 25.67 27.42 13.21 37.9 39.36 37.05 K20 517.5 445.1 703.1 633.6 1021 946.6 785.5 677.4 446.4 357.6 CaO 10340 11360 8175 7062 6209 5649 3041 6373 7587 6545 Ti02 8.705 7.845 8.687 5.552 9.263 8.311 4.909 8.809 8.418 10.57 MnO 325.8 368.5 211.4 195.8 143.9 143.2 77.53 222 345.8 335.4 Fe20 3 65.47 62.88 77.51 75.78 91.88 99.7 70.91 117.5 85.29 117.5 ppm 43 em 52 61 61 78 87 96 96 114 123 MgO 605.2 794.1 526.5 413.9 266.1 139.2 146.7 89.67 95.18 Ah03 2273 3269 2953 3926 3950 3008 3401 3690 3164 3548 P20 5 37.2 44.92 35.03 46.1127.07 17.77 24.31 27.01 22.2 23.55 K20 363 313.8 212.5 233 155.2 115.4 144.9 125.3 150 210.9 CaO 6043 4830 3064 3529 2093 1387 1697 1641 1457 1746 Ti02 10.84 9.461 7.74 7.871 8.811 6.002 8.951 6.582 6.351 7.352 MnO 305.2 419.6 374.6 482.5 311.6 186.1 250.8 271.1 196 215.3 Fe20 3 127.6 122.8 108.9 148 123.9 97.03 112.8 117.3 99.19 113.4 ppm 132 132 141 149 149 158 167 177 177 LOD MgO 72.33 80.09 57.16 44.48 200 Ah03 2677 3660 2624 1870 2048 1174 1137 963 977.8 7.1 P20 5 15.8 23.19 15.38 10.2 11.06 8.342 12.4 10.3 9.955 4 K20 323.8 374.7 573.4 705.4 679.3 995 1741 1746 1634 53 CaO 1320 1586 1382 1124 1191 1174 1725 1694 1703 28 Ti02 5.252 5.779 6.516 4.018 4.673 6.211 3.886 11 MnO 114.8 152.1 128.4 81.65 88.98 68.57 95.13 87.09 88.01 4.7 Fe20 3 91.69 137.4 117.4 91.88 101.3 81.31 121.7 97.75 94.84 23 All values determined using medium resolution ICP-MS 129 Table 6. Carbonate and exchangeable fraction minor and trace element concentrations by depth ~~m 2cm 6 10 10 14 14 18 22 26 Li 0.208 0.188 0.122 0.136 0.139 0.122 0.104 0.233 0.225 Be 0.316 0.314 0.289 0.372 0.313 0.403 0.173 0.336 0.356 Sc~ 0.0045 0.0069 0.0075 0.0076 0.0034 0.0101 0.0101 v~ 0.218 0.203 0.216 0.172 0.21 0.177 0.0994 0.168 0.164 Cr~ 0.135 0.0611 0.053 0.104 0.0506 0.0553 Co~ 0.797 0.811 0.888 0.93 0.904 0.943 0.534 0.817 0.802 Ni~ 0.0936 0.0703 0.0899 0.114 0.105 0.119 0.0561 0.172 0.208 Cu~ 1.24 1.16 0.9313 0.562 0.886 0.511 0.685 1.5 1.62 Zn~ 16.7292 20.1299 14.2239 22.8119 10.4568 15.4634 5.22 12.1247 11.2268 Ga 0.0676 0.0615 0.0943 0.0863 0.12 0.125 0.0809 0.123 0.0937 Rb 4.17 8.77 Sr 313 321 256 216 242 204 131 225 250 y 1.1 0.929 1.07 0.773 1.17 0.881 0.902 1.82 1.49 Zr~ 3.1 3.18 3.09 2.23 2.93 2.28 1.65 4.12 3.85 Nb~ 0.276 0.256 0.289 0.205 0.26 0.213 0.13 0.31 0.256 Cs 0.0739 0.0549 0.11 0.0923 0.185 0.147 0.113 0.118 0.113 Ba 186 218 113 86.7 60.6 45.5 23.1 84.1 169 La 6.21 5.06 5.94 3.41 6.6 4.06 4.89 9.43 8.59 Ce 2.17 1.88 3.58 2.12 4.82 3.04 3.6 6.02 4.25 Pr 0.438 0.365 0.452 0.31 0.512 0.371 0.387 0.762 0.636 Nd 1.29 1.08 1.34 1.02 1.48 1.2 1.12 2.23 1.85 Sm 0.186 0.159 0.193 0.157 0.214 0.176 0.16 0.319 0.268 Eu 0.0768 0.0736 0.0704 0.0611 0.0701 0.0611 0.0489 0.103 0.101 Tb 0.0264 0.0225 0.0274 0.0215 0.0299 0.0245 0.0226 0.0444 0.0375 Gd 0.196 0.164 0.196 0.156 0.213 0.183 0.16 0.322 0.273 Dy 0.138 0.116 0.141 0.113 0.155 0.129 0.117 0.233 0.195 Ho 0.0262 0.0222 0.0266 0.0201 0.0292 0.0229 0.0225 0.0443 0.0372 Er 0.0654 0.0576 0.0694 0.0523 0.075 0.06 0.0584 0.114 0.0962 Tm 0.0077 0.0072 0.0083 0.0064 0.0094 0.0075 0.0071 0.0135 0.0114 Yb 0.0414 0.0361 0.0421 0.0372 0.0477 0.0425 0.036 0.0707 0.0603 Lu 0.006 0.0057 0.0064 0.0052 0.0067 0.006 0.0052 0.0102 0.0091 Hf 0.0604 0.0643 0.059 0.0362 0.0512 0.0345 0.028 0.0691 0.0705 Ta 0.0033 0.0029 0.003 0.0017 0.0023 0.0018 0.0012 0.0038 0.0024 Tl 0.0581 0.0472 0.0491 0.0385 0.048 0.0364 0.0315 0.0654 0.0787 Pb 0.56 0.387 0.433 0.55 0.484 0.44 0.185 0.559 0.728 Th 0.0917 0.193 0.258 0.207 0.322 0.251 0.223 0.436 0.379 U:j: 0.926 0.918 0.841 0.502 0.791 0.477 0.467 1.1 1.24 ~Values determined using medium resolution ICP-MS; all other values determined using low resolution ICP-MS 130 Table 6. continued l!l!m 30 em 43 52 61 61 78 87 96 96 114 Li 0.246 0.191 0.217 0.154 0.176 0.15 0.116 0.166 0.153 0.167 Be 0.447 0.347 0.48 0.44 0.891 0.572 0.427 0.503 0.806 0.478 Sc:j: 0.0138 0.014 0.025 0.0329 0.0328 0.0574 0.0472 0.064 0.052 0.0676 V:j: 0.183 0.179 0.201 0.182 0.214 0.249 0.17 0.201 0.186 0.168 Cr:j: 0.064 0.0638 0.0716 0.302 0.0691 0.0396 0.147 0.0299 Co:j: 0.857 0.833 1.02 0.95 1.35 0.897 0.5 0.521 0.63 0.352 Ni:j: 0.17 0.073 0.0613 0.0676 0.109 0.0891 0.0434 0.054 0.0825 0.0184 Cu:j: 1.44 0.9949 1.23 10.5 1.92 1.81 1.67 2.39 3.22 2.45 Zn:j: 33.1303 26.1076 14.0773 8.08 18.4243 5.32 2.87 3.2 6.03 1.2 Ga 0.113 0.112 0.101 0.0783 0.119 0.122 0.124 0.137 0.229 0.138 Rb 3.42 2.1 Sr 235 213 200 127 144 83.8 59.2 73.7 64.6 63 y 1.51 1.66 2.18 1.93 2.05 2.99 3.05 3.79 3.1 3.94 Zr:j: 4.82 4.18 4.55 4 4.48 5.62 3.58 4.22 3.63 3.68 Nb:j: 0.315 0.318 0.226 0.18 0.208 0.216 0.134 0.184 0.152 0.132 Cs 0.104 0.0922 0.143 0.11 0.117 0.0726 0.0591 0.068 0.0585 0.0663 Ba 231 225 339 231 253 203 186 262 216 264 La 9.3 9.44 14.3 11.7 10.7 19.5 21.8 28.8 17.8 30.4 Ce 4.4 4.46 4.84 3.32 2.91 7.76 9.27 10.7 7.6 12 Pr 0.656 0.703 0.988 0.851 0.86 1.64 1.93 2.46 1.89 2.81 Nd 1.91 2.04 2.88 2.51 2.8 4.85 5.73 7.31 6.26 8.47 Sm 0.274 0.294 0.408 0.365 0.425 0.701 0.799 1.01 0.904 1.17 Eu 0.109 0.114 0.161 0.135 0.167 0.224 0.24 7 0.314 0.295 0.355 Tb 0.0382 0.0412 0.0571 0.0507 0.0596 0.0907 0.1 0.125 0.11 0.139 Gd 0.277 0.3 0.416 0.372 0.434 0.676 0.755 0.95 0.847 1.08 Dy 0.204 0.218 0.289 0.267 0.306 0.455 0.491 0.607 0.53 0.6641 Ho 0.0375 0.0407 0.053 0.0486 0.0538 0.081 0.0828 0.1015 0.0862 0.108 Er 0.0993 0.105 0.134 0.123 0.136 0.201 0.207 0.246 0.208 0.257 Tm 0.0121 0.0129 0.0155 0.0145 0.0165 0.0235 0.0242 0.0288 0.0254 0.0295 Yb 0.0632 0.0675 0.085 0.0774 0.0909 0.128 0.13 0.155 0.146 0.161 Lu 0.0095 0.01 0.0124 0.0114 0.0136 0.0184 0.0185 0.022 0.0208 0.0224 Hf 0.0863 0.0764 0.09 0.0787 0.0758 0.114 0.0746 0.0904 0.0646 0.0771 Ta 0.003 0.0039 0.0029 0.0026 0.003 0.0044 0.0028 0.0039 0.0029 0.0034 Tl 0.0769 0.0804 0.161 0.107 0.105 0.106 0.0686 0.0836 0.0682 0.095 Pb 1.1 0.886 1.01 0.763 0.925 0.783 0.478 0.625 0.573 0.468 Th 0.477 0.427 0.437 0.421 0.402 0.733 0.591 0.636 0.487 0.585 U:j: 1.46 1.29 1.61 1.41 1.23 2.24 1.96 2.29 1.54 1.96 t Values determined using medium resolution ICP-MS; all other values determined using low resolution ICP-MS 131 Table 6. continued ppm 123 132 132 141 149 149 158 167 177 177 LOD Li 0.243 0.523 0.609 0.744 0.402 0.426 0.176 0.285 0.238 0.262 0.035 Be 0.553 0.5 0.952 0.607 0.525 0.906 0.589 1.54 1.51 1.56 0.082 Sc:j: 0.0896 0.0891 0.1 0.135 0.146 0.121 0.237 0.396 0.395 0.374 0.024 V:j: 0.185 0.133 0.173 0.159 0.105 0.102 0.0963 0.143 0.106 O.ll06 0.037 Cr:j: 0.0519 0.132 0.0367 0.107 0.0728 0.28 Co:j: 0.288 0.112 0.151 0.1066 0.0676 0.0723 0.077 0.0917 0.0912 0.089 0.053 Ni:j: 0.0264 0.0454 0.0974 0.0309 0.0315 0.19 Cu:j: 2.83 1.86 3.1 2.38 1.81 2.5 1.74 5.35 3.79 3.34 0.26 Zn:j: 1.25 1.28 4.01 1.73 1.12 2.62 0.62 3.41 3.68 3.13 2.8 Ga 0.175 0.157 0.375 0.184 0.172 0.38 0.188 0.578 0.566 0.564 0.0081 Rb 2.79 3.94 7.79 7.09 6.91 0.15 Sr 75.7 77 81.8 105 106 106 188 359 355 352 0.43 y 5.25 4.5 4.74 5.56 5.8 5.54 8.13 11.9 11.8 11.5 0.0066 Zr:j: 3.89 2.91 3.43 3.32 2.12 1.85 1.41 1.34 1.04 1.02 0.12 Nb:j: 0.154 0.105 0.115 0.114 0.0692 0.0564 0.0829 0.119 0.0409 0.0563 0.014 Cs 0.0749 0.069 0.0689 0.0878 0.0745 0.0692 0.0869 0.155 0.137 0.135 0.0039 Ba 403 604 627 1017 1068 1022 1224 943 774 763 0.094 La 41.2 36 29.2 43.2 44.5 31.6 57.2 57 55.2 52.2 0.021 Ce 16.1 17.1 14.1 23.6 20.9 14.1 22.9 20.4 18.2 17.2 0.021 Pr 3.83 3.41 3.33 4.14 4.02 3.6 5.21 6.64 6.53 6.24 0.0048 Nd 11.6 10.3 11.1 12.4 12.6 11.8 16.6 22.3 22.1 21.3 0.019 Sm 1.6 1.4 1.59 1.71 1.73 1.73 2.32 3.22 3.21 3.1 0.0058 Eu 0.493 0.468 0.553 0.604 0.612 0.651 0.791 1.04 1.02 1 0.0025 Tb OJ88 0.166 0.186 0.205 0.213 0.209 0.285 0.395 0.392 0.393 0.0009 Gd 1.48 1.28 1.45 1.56 1.6 1.58 2.17 3.04 3.03 3.03 0.0045 Dy 0.8919 0.7922 0.8848 0.9821 1.04 1 1.4 1.94 1.92 1.9 0.004 Ho 0.144 0.126 0.139 0.159 0.173 0.164 0.237 0.329 0.33 0.316 0.0019 Er 0.343 0.304 0.331 0.385 0.422 0.398 0.585 0.829 0.827 0.799 0.0062 Tm 0.0391 0.035 0.0396 0.0455 0.0506 0.0485 0.07 0.104 0.103 0.1 0.001 Yb 0.213 0.195 0.233 0.251 0.274 0.285 0.385 0.604 0.606 0.594 0.0076 Lu 0.0297 0.0278 0.0318 0.0358 0.0401 0.0391 0.0563 0.0885 0.0899 0.0872 0.0007 Hf 0.0828 0.0641 0.0582 0.0752 0.053 0.0369 0.0485 0.0363 0.0295 0.0292 0.0028 Ta 0.004 0.0031 0.0031 0.0034 0.0025 0.002 0.003 0.004 0.0032 0.0032 0.0022 Tl 0.105 0.0916 0.0824 0.1225 0.0886 0.0775 0.0878 0.103 0.112 0.107 0.0014 Pb 0.544 0.33 0.502 0.469 0.294 0.355 0.176 0.238 0.174 0.183 0.17 Th 0.654 0.533 0.543 0.662 0.505 0.396 0.39 0.341 0.302 0.291 0.0082 u; 2.39 2.os 1.12 2.39 2.36 1.61 2.s 2.19 2.09 2 o.oo94 t Values determined using medium resolution ICP-MS; all other values determined using low resolution ICP-MS 132 Table 7. Oxide fraction major element concentrations by depth ppm 2 em 6 10 10 14 14 18 22 26 30 MgO 112.8 166.9 122.8 147.8 175.2 93.5 80.7 86.58 Ah03 24960 26780 26460 26690 24830 29570 24550 27400 27240 32150 P20s 607.5 511.8 479.7 470.4 493.7 551.8 465 466.4 409.1 455.5 K20 594.3 388.6 882.6 995.7 1208 1493 1234 455 186 300 CaO 3479 4055 2944 2931 2619 3073 2749 2547 2519 2461 Ti02 226.4 241.5 235.1 244 232.6 268.6 230.4 267.5 306.9 342.7 MnO 291.1 288.4 153 180.2 102.1 145 142.4 280.9 289.9 234.4 Fe203 5049 5634 4800 5176 4210 5072 4377 5412 5937 6548 ppm 43 em 52 61 61 78 87 96 96 114 123 MgO 138.71 79.2 51.79 42.04 15.63 12.27 8.031 14.85 6.517 Ah03 26520 25850 28510 28880 28350 29450 26070 244402738024930 P20 5 395.8 337.6 368.3 395.2 264.9 295.2 292 275.1 317.9 289.2 K20 596 139 CaO 2419 1635 1547 1302 629.3 544.4 553.7 504.4 502.8 468.2 Ti02 277.4 302.9 344.5 361.9 402.6 410.8 364.5 336.2 397.7 353.7 MnO 221.4 292.2 344.2 388.1 261 321.2 255.5 279 292.5 387.5 Fe203 5614 6126 6853 7015 6558 6626 5947 5496 6446 6015 ppm 132 em 132 141 149 149 158 167 177 177 LOD MgO 8.64 11.4 13.44 7.716 19 Ah03 23850 24770 20840 19300 20100 12490 11870 9507 9742 97 P20 5 266.9 280.8 248.6 226.8 221.7 161.7 171.3 135.8 127.4 12 K20 15.61 41.57 37.43 23.55 21.23 230 CaO 298.5 351.1 295.8 303.4 299.8 312 487 375.2 340.1 72 Ti02 424.8 436.4 404.8 405.6 415.9 348.7 481.8 421.8 419.1 31 MnO 614.9 707.4 522.1 832.7 844.6 798.3 1202 1152 1109 13 Fe20 3 7210 7273 6976 7709 7737 7172 9154 7957 7842 50 All values determined using medium resolution ICP-MS 133 Table 8. Oxide fraction minor and trace element concentrations by depth ~~m 2cm 6 10 10 14 14 18 22 26 30 Li 0.537 0.473 0.518 0.561 0.606 0.717 0.611 0.417 0.35 0.398 Be 2.74 2.79 2.31 2.03 1.86 1.98 1.98 2.65 2.91 3.17 Sc:j: 0.0228 0.0263 0.0196 0.0279 0.0172 0.0302 0.0189 0.0162 0.0145 0.0154 V:j: 5.06 5.39 4.67 4.82 3.97 4.55 4.08 4.78 5.25 5.7 Co:j: 1.86 2.05 1.65 1.91 1.17 1.79 1.42 1.78 1.83 1.79 Ni:j: 0.257 0.358 0.339 0.353 0.237 0.359 0.331 0.869 1.04 0.906 Cu:j: 1.27 2.26 2.17 1.97 2.13 1.72 7.83 3.58 5.26 6.4 Zn:j: 64.5 86.4 49.6 57.7 29.7 36.5 27.9 40.7 49 42.1 Ga 3.01 2.84 3.39 3.44 3.76 4.22 4.01 4.19 3.76 4.25 Rb 4.54 2.3 4.25 4.28 8.04 9.14 10.6 5.95 2.85 2.17 Sr 134 127 119 118 122 139 126 98.9 85.5 82.3 y 0.954 0.783 1.16 1.13 1.71 1.88 2.02 1.49 0.994 1.23 Zr:j: 14.9 14.9 14.2 15.7 13.2 16.9 14.7 14.1 14.6 16.2 Nb:j: 8.63 9.39 8.68 8.51 8.31 9.1 8.48 10.3 12 13.9 Ba 320 336 317 306 243 276 223 277 269 276 La 5.56 4.07 6.58 6.78 10.2 11 10.9 8.41 5.55 7.21 Ce 2.95 2.34 4.87 4.79 8.94 9.63 10.2 5.98 3.23 4.32 Pr 0.418 0.332 0.543 0.537 0.868 0.953 0.997 0.665 0.429 0.558 Nd 1.31 1.06 1.68 1.65 2.65 2.92 3.05 2.06 1.33 1.72 Sm 0.177 0.148 0.227 0.22 0.346 0.373 0.393 0.268 0.184 0.231 Eu 0.122 0.119 0.137 0.124 0.153 0.162 0.166 0.142 0.116 0.129 Tb 0.026 0.0212 0.0338 0.0323 0.051 0.0555 0.0578 0.0398 0.0268 0.0337 Gd 0.19 0.156 0.239 0.233 0.374 0.407 0.42 0.291 0.192 0.249 Dy 0.131 0.109 0.165 0.162 0.25 0.274 0.288 0.204 0.138 0.174 Ho 0.0214 0.018 0.027 0.0266 0.0395 0.044 0.0459 0.0335 0.0229 0.0281 Er 0.0598 0.051 0.0729 0.0729 0.106 0.122 0.125 0.0912 0.0625 0.077 Tm 0.00794 0.00703 0.0099 0.0099 0.0143 0.0164 0.0168 0.0123 0.00839 0.0109 Yb 0.051 0.0432 0.0596 0.0623 0.0861 0.0979 0.0966 0.0735 0.0511 0.0647 Lu 0.00681 0.00579 0.00832 0.00845 0.0113 0.0128 0.0132 0.0098 0.0073 0.00888 Hf 0.215 0.214 0.196 0.248 0.176 0.257 0.171 0.14 0.133 0.147 Ta 0.0218 0.0227 0.0145 0.00698 0.0103 0.0099 0.0106 0.0137 0.0203 0.0245 Pb 6.03 4.99 4.51 4.51 4.46 5.76 6.05 7.37 Th 0.206 0.272 0.264 0.57 0.234 0.681 0.268 0.124 0.0897 0.0684 U:j: 0.427 0.408 0.42 0.408 0.38 0.425 0.428 0.495 0.561 0.576 :1: Values determined using medium resolution ICP-MS; all other values determined using low resolution ICP-MS 134 Table 8. continued J!J!ID 30 em 43 52 61 61 78 87 96 96 114 Li 0.398 0.504 0.392 0.252 0.208 0.158 0.135 0.151 0.125 0.162 Be 3.17 2.95 3.43 3.95 3.4 3.84 4.08 3.59 2.97 3.9 Sci 0.0154 0.0235 0.0214 0.0307 0.0494 0.0384 0.0472 0.0425 0.0615 0.0485 Vi 5.7 5.46 8.18 10.8 10.9 13.1 13.5 11.9 11 13 Coi 1.79 1.83 2.34 2.91 3.48 2.3 2.53 2.04 2.35 2.29 Nii 0.906 0.286 0.412 0.289 0.263 0.289 0.235 0.148 0.21 Cui 6.4 4.42 7.25 6 3.92 6.11 10.1 8.72 10.1 14.3 Zni 42.1 47.1 52.8 51.5 46.1 20.8 20.9 20 22.7 15.8 Ga 4.25 3.65 3.05 3.08 3.14 3.45 4.07 3.54 3.2 3.93 Rb 2.17 3.07 2.82 2.04 1.86 0.975 0.936 0.931 0.819 0.958 Sr 82.3 93.1 59.7 46 43.1 17.2 15.2 14.4 13.7 13.2 y 1.23 1.34 1.21 1.61 1.67 2.96 4.67 4.52 4.13 5.67 Zri 16.2 19.4 16.3 19.5 23.7 21.3 24.7 20.3 21.6 20.8 Nbi 13.9 10.6 10.9 11.7 11.8 14.8 16.1 14.3 12.9 16.4 Ba 276 303 332 256 246 157 177 191 165 225 La 7.21 8.27 7.47 8.99 9.5 16.2 27 28.3 27.2 34.3 Ce 4.32 4.56 2.96 3.33 3.46 8.31 14.5 13.1 12 17.4 Pr 0.558 0.594 0.517 0.652 0.687 1.5 2.7 2.73 2.51 3.65 Nd 1.72 1.79 1.61 2.04 2.12 4.57 8.23 8.34 7.51 11.3 Sm 0.231 0.236 0.221 0.284 0.291 0.62 1.07 1.08 0.958 1.46 Eu 0.129 0.136 0.137 0.141 0.133 0.212 0.342 0.339 0.302 0.455 Tb 0.0337 0.0349 0.0316 0.0425 0.0447 0.0881 0.152 0.147 0.137 0.202 Gd 0.249 0.253 0.234 0.304 0.319 0.639 1.1 1.06 1.01 1.51 Dy 0.174 0.178 0.162 0.218 0.225 0.438 0.728 0.712 0.663 0.963 Ho 0.0281 0.0296 0.0274 0.0371 0.0379 0.0714 0.113 0.11 0.0981 0.138 Er 0.077 0.0816 0.073 0.101 0.103 0.193 0.296 0.286 0.256 . 0.35 Tm 0.0109 0.0114 0.0099 0.0139 0.0139 0.0255 0.0395 0.0368 0.0334 0.0459 Yb 0.0647 0.0678 0.0596 0.0819 0.0838 0.154 0.231 0.222 0.201 0.27 Lu 0.00888 0.00935 0.00816 0.0112 0.0116 0.0208 0.0316 0.0299 0.0268 0.0357 Hf 0.147 0.221 0.147 0.163 0.246 0.152 0.171 0.146 0.187 0.128 Ta 0.0245 0.0348 0.0247 0.0287 0.0183 0.0273 0.0256 0.0255 0.0169 0.0227 Pb 7.37 6.42 6.07 6.78 6.47 6.83 6.37 6.57 Th 0.0684 0.195 0.0916 0.0933 0.214 0.073 0.0925 0.0658 0.186 0.0732 Ui 0.576 0.544 0 0.631 0.626 0.978 1.07 0.839 0.794 0.887 *Values determined using medium resolution ICP-MS; all other values determined using low resolution ICP-MS 135 Table 8. continued l!l!m 123 em 132 132 141 149 149 158 167 177 177 LOD Li 0.266 0.645 0.603 0.819 0.867 0.903 0.498 0.553 0.519 0.461 0.056 Be 3.87 3.76 3.37 3.45 3.95 3.56 3.44 3.4 2.96 3.03 0.44 Sc~ 0.0576 0.101 0.126 0.133 0.225 0.268 0.348 0.528 0.446 0.46 0.059 v~ 13.3 15.8 15.8 14.9 17 17.1 17.2 23.9 21.4 21.2 0.18 Co~ 2.91 4.35 5.66 3.74 6.31 6.17 5.9 8.86 8.48 7.91 0.25 Ni~ 1.4 Cu~ 11.4 13.9 13.8 12.5 17.9 15.9 12.8 20.9 16.3 15.5 1.5 Zn~ 14.8 22.8 25.4 27.8 26.4 29.5 17.1 32.1 22.8 24.2 6.9 Ga 3.95 4.73 4.67 4.52 5.01 4.89 4.24 4.34 3.87 3.71 0.11 Rb 0.823 0.69 0.682 0.754 0.987 1.03 1.26 1.52 1.26 1.14 0.078 Sr 11.5 8.3 8.86 9.08 10.1 9.99 11.7 18.6 15.4 14.5 0.29 y 6.41 7.43 7.12 7.23 9.17 9.52 10.3 13.7 12.5 12.5 0.015 Zr~ 24.4 33.7 39.1 46.1 62.1 67.9 67.1 90.3 77.3 77.2 0.25 Nb~ 16.3 20 18.3 20.5 21.7 19.1 18.6 21 15.5 14.6 0.16 Ba 219 196 191 177 189 204 162 121 97 0.43 La 40.4 41.1 41.7 36 43 47 47.2 59.8 54.6 52.1 0.025 Ce 21.1 36.3 34.6 39.5 52.5 54.2 57.5 75.3 75.6 64.8 0.031 Pr 4.29 4.89 4.89 4.79 5.9 5.97 6.25 8.06 7.39 7.27 0.008 Nd 13.2 15.3 15.2 15 18.7 19.6 21.1 27 24.6 23.7 0.04 Sm 1.69 2.02 1.97 1.99 2.58 2.67 2.99 3.79 3.47 3.41 0.018 Eu 0.516 0.593 0.576 0.589 0.748 0.771 0.843 1.048 0.955 0.936 0.0063 Tb 0.227 0.276 0.271 0.27 0.363 0.381 0.415 0.542 0.498 0.481 0.0021 Gd 1.71 1.99 1.98 1.98 2.52 2.66 2.9 3.85 3.5 3.42 0.013 Dy 1.1 1.33 1.31 1.33 1.84 1.87 2.09 2.7 2.48 2.39 0.0088 Ho 0.158 0.192 0.189 0.202 0.283 0.29 0.326 0.425 0.379 0.373 0.0025 Er 0.408 0.5 0.495 0.528 0.756 0.766 0.877 1.16 1.04 1.03 0.0067 Tm 0.0534 0.0657 0.0659 0.0719 0.106 0.109 0.125 0.169 0.151 0.148 0.0021 Yb 0.312 0.401 0.393 0.438 0.646 0.659 0.766 1.03 0.929 0.902 0.048 Lu 0.0416 0.0523 0.051 0.0571 0.0847 0.0862 0.102 0.142 0.128 0.124 0.0045 Hf 0.162 0.224 0.297 0.365 0.568 0.653 0.74 1.07 0.919 0.939 0.015 Ta 0.0287 0.0238 0.0132 0.0206 0.0272 0.00778 0.0351 0.0169 0.0139 0.00719 0.0098 Ph 6.78 7.45 7.14 8.1 7.16 0.27 Th 0.0912 0.198 0.39 0.356 0.837 1.16 1.13 1.5 1.18 1.22 0.0089 u~ 0.893 0.945 0.932 0.917 1.18 1.24 1.12 1.27 1.03 1.05 0.015 +Values determined using medium resolution ICP-MS; all other values determined using low resolution ICP-MS 136 Table 9. Organic fraction major element concentrations by depth ppm 2 em 2 6 10 14 18 22 26 26 30 MgO 149.5 121.8 107.5 130.8 130.7 135.4 109.3 98.92 85.46 79.33 AhOJ 45480 32160 40780 35100 34000 26660 30640 31320 28220 34660 P20 5 1143 818.6 992 750.2 592.5 806.3 690.9 654 268.7 795.6 K20 672.9 442.3 470 808.2 823 863.6 430.7 338.9 324.2 325.7 CaO 5410 2843 4464 4617 4272 3895 3594 3797 2703 3905 Ti02 63.39 46.6 74.76 93.81 148.9 147.5 69.41 37.24 15.96 87.56 MnO 135.6 111.7 121.4 85.59 68.9 63.11 88.38 127.7 106.9 104.2 Fe20 3 1344 1145 1191 984.2 964.1 712.6 744 930.2 498.9 790.1 ppm 43 em 43 52 52 61 78 78 87 96 114 114 MgO 90.25 94.13 102.5 81.59 72.81 58.24 45.25 60.29 65.01 49.95 19.08 Ah03 33740 33530 39630 22990 26210 19560 13680 42310 44580 32170 4448 P20s 592.3 619.1 967.7 287.3 622.8 364.7 376.3 2014 2286 1196 114.9 K20 369.3 389.6 207.6 146 60.13 20.85 20.39 21.21 19.01 13.64 CaO 4178 3310 5265 2818 3344 2605 2811 2892 3740 3926 3292 Ti02 91.46 68.05 136.5 36.66 110.8 69.98 50.38 232 271.4 352.2 29.09 MnO 103 104.5 157.8 111.1 139.2 81.89 64.27 103.9 122.4 110.1 30.93 Fe203 787.4 573.8 807 628.3 1379 552.7 336 924.7 1086 846.1 249.2 ppm 123 em 123 132 141 149 158 158 167 167 177 LOD MgO 34.36 29.82 29.13 20.88 38.23 33.22 23.04 116.35 47.71 55.67 6.9 AhOJ 14820 11020 13750 9040 18990 13040 8273 17860 10740 9783 63 P20s 406.6 220.5 223.7 21.34 11 K20 7.531 9.037 40.78 19.52 31.53 10.15 67.46 43.65 37.97 23 CaO 2569 3566 1539 1794 1817 1759 1301 2282 1337 1624 34 Ti02 79.28 39.94 52.91 16.35 24.57 39.18 40.8 13.14 45.8 24.42 10 MnO 84.11 56.59 55.25 37.45 43.77 46.45 34.77 50.76 38.99 37.8 0.41 Fe203 345.6 294.6 264.6 201.5 313.3 324.9 308 285.7 339.1 248.8 14 All values determined using medium resolution ICP-MS 137 Table 10. Organic fraction minor and trace element concentrations by depth J!J!ID 2cm 2 6 10 14 18 22 26 26 30 43 Li 1.83 1.73 1.35 1.26 0.981 1.06 1.87 2.19 2.58 1.91 1.7 Be 1.23 0.869 1.09 0.947 1.06 0.737 1.02 1.16 1.08 1.12 1.31 Sc~ 0.0963 0.0867 0.0741 0.0615 0.0528 0.0696 0.0795 0.0488 0.0712 0.0787 v~ 0.469 0.164 0.564 0.491 1.41 0.449 0.275 0.0951 0 0.261 0.256 Cr~ 0.336 0.149 0.354 0.168 0.126 2.97 0.0948 0.131 0.107 0.139 0.139 Co~ 2.54 1.91 2.03 2.09 1.33 1.55 1.53 1.61 1.46 1.56 1.69 Ni~ 1.64 1.12 1.39 1.31 1.42 1.01 1.76 2.06 1.96 1.56 0.694 Cu~ 39.4 25.5 35 33.5 31.2 23.7 46.9 53 49 41.3 32.7 Zn~ 23.4 13.5 21.8 13.4 7.07 6.95 7.72 10.2 9.48 10.6 11.3 Ga 2.41 2.52 2.48 2.17 1.99 2.05 2.68 2.81 2.42 2.64 2.7 Rb 6.81 4.74 3.2 5.69 9.86 11.9 4.73 3.08 2.83 2.8 3.1 Sr 115 68.9 99.3 119 119 115 91.6 88.5 66.4 85.3 89.1 y 20 16.4 19.3 16.2 14.1 12.8 20.7 21.4 20.6 18.4 20.7 Zr~ 4.76 9.5 4.65 4.11 5.42 7.48 4.64 3.25 1.61 7.61 6.83 Nb~ 0.649 0.818 0.796 1.12 2.19 2.68 1.2 0.33 0.147 1.74 1.65 Cs 0.146 0.0539 0.0931 0.15 0.209 0.143 0.124 0.104 0.0764 0.108 0.117 Ba 173 160 182 298 278 169 224 197 166 220 220 La 81.9 71.9 87.6 74.2 58.7 59.8 92.4 95.8 87.5 86.9 94.7 Ce 69.3 78.7 78.2 77.8 56.1 81.4 102 99.3 84.4 86 87.2 Pr 9.9 8.5 11.2 9.7 8.2 8.9 12.4 12.3 11.1 11 11.9 Nd 29.8 27.2 34.5 29.8 26.6 28.2 39.1 38.4 34.1 33.9 36.8 Sm 4.34 4.08 5.03 4.37 4.09 4.19 5.53 5.49 5.06 4.97 5.31 Eu 1.36 1.25 1.56 1.45 1.37 1.28 1.7 1.69 1.54 1.55 1.66 Tb 0.936 0.858 0.994 0.921 0.823 0.85 1.1 1.11 1.07 0.999 1.04 Gd 6.01 5.63 6.94 5.94 5.37 5.63 7.67 7.64 7.01 6.72 7.27 Dy 4.61 4.13 5.07 4.4 3.97 4.04 5.55 5.53 5.13 4.96 5.38 Ho 0.554 0.474 0.575 0.495 0.451 0.418 0.613 0.633 0.587 0.563 0.61 Er 1.47 1.26 1.53 1.3 1.17 1.09 1.64 1.7 1.6 1.48 1.62 Tm 0.201 0.176 0.212 0.178 0.164 0.149 0.225 0.238 0.223 0.209 0.225 Yb 1.18 1.03 1.26 1.06 0.97 0.9 1.31 1.39 1.31 1.22 1.32 Lu 0.145 0.127 0.156 0.13 0.118 0.109 0.163 0.174 0.162 0.151 0.164 Hf 0.106 0.162 0.104 0.107 0.109 0.168 0.15 0.0955 0.067 0.191 0.173 Ta 0.0401 0.0132 0.0113 0.0392 0.0445 0.1441 0.0877 0.0174 0.0084 0.0998 0.0895 Tl 0.0637 0.0428 0.0469 0.0534 0.0447 0.0393 0.0527 0.0713 0.0591 0.0867 0.0753 Pb 1.15 1.91 1.45 1.31 1.75 1.06 1.12 2.35 1.46 1.92 2.33 Th 0.319 0.621 0.472 0.565 0.679 0.712 0.34 0.448 0.241 0.506 0.582 u~ 1.13 1.02 1.42 1.08 0.995 0.807 1 1.34 1.23 1.16 1.24 t Values determined using medium resolution ICP-MS; all other values determined using low resolution ICP-MS 138 Table 10. continued l!l!m 43 52 52 61 78 78 87 96 114 114 123 Li 1.77 2.16 2.34 1.32 0.888 0.94 0.876 0.88 0.694 0.595 0.398 Be 1.22 1.53 1.08 1.19 0.752 0.571 1.24 1.32 1.27 0.18 0.612 Sc* 0.0809 0.102 0.123 0.148 0.129 0.0936 0.25 0.272 0.181 0.0419 0.127 v* 0.0527 0.27 0.174 0.359 0.143 0.198 0.38 0.411 1.17 0.0806 0.123 cr; 0.0988 15 0.118 0.142 Co* 1.82 2.13 1.8 2.08 1.39 1.31 1.79 2.11 1.71 0.368 0.918 Ni* 0.888 0.708 0.425 0.544 0.243 0.0953 0.361 0.316 0.113 Cu* 30.5 26.3 22.2 20 19.5 16.7 24.5 27.2 23.4 7.73 10.5 Zn* 12.6 22.4 13.9 12.2 5.77 4.89 7.04 6.86 2.6 1.69 Ga 2.31 2.62 2.72 2.51 2.56 2.11 4.02 4.48 2.68 1.25 2.12 Rb 3 2.57 2.28 1.65 0.83 0.742 0.836 0.861 0.76 0.524 0.61 Sr 73.5 94.6 56 59.1 42.1 43.3 45.5 58.3 56.7 47.8 38 y 20.8 29.9 26.6 27.2 18.7 17.1 24.1 26.6 24.5 6.56 16.3 Zr* 6.87 9.67 3.1 10 6.05 3.35 26.9 27.7 22.7 0.594 7.05 Nb* 1.45 2.45 0.484 1.54 1.18 0.774 5.18 6.15 5.28 0.161 1.32 Cs 0.105 0.211 0.128 0.116 0.0508 0.0407 0.0565 0.0614 0.0592 0.0238 0.0331 Ba 230 201 201 155 90.2 64.3 135 164 204 49.3 125 La 92.9 143 117 114 68.2 59.8 98.4 110 87.9 27.4 55.5 Ce 92.6 113 83.4 80.5 86.2 78.1 135 158 80.5 45.5 76.8 Pr 11.3 16.7 12.9 13.6 11.2 9.3864 16.6 19.4 15.5 5.04 10.6 Nd 33.9 53 41.6 44 35.4 31.3 54.5 63.8 53.7 18.4 36.6 Sm 5.01 7.45 6.01 6.44 5.7 5.08 8.13 9.43 8.21 3.02 5.76 Eu 1.57 2.21 1.82 1.95 1.67 1.45 2.4 2.78 2.48 0.824 1.69 Tb 1.05 1.45 1.19 1.26 1.08 1.04 1.45 1.71 1.47 0.588 1.06 Gd 6.86 10.4 8.33 8.99 7.49 6.81 10.5 12.1 10.6 4.07 7.53 Dy 4.99 7.05 6.09 6.29 5.44 4.86 6.98 7.95 6.98 2.72 5.28 Ho 0.588 0.841 0.731 0.788 0.614 0.536 0.823 0.908 0.836 0.236 0.564 Er 1.57 2.21 1.92 2.09 1.6 1.4 2.14 2.33 2.16 0.567 1.46 Tm 0.218 0.301 0.267 0.287 0.225 0.189 0.299 0.323 0.298 0.0727 0.201 Yb 1.28 1.8 1.59 1.71 1.34 1.16 1.8 1.97 1.81 0.502 1.25 Lu 0.158 0.221 0.197 0.212 0.167 0.14 0.223 0.242 0.224 0.0567 0.154 Hf 0.186 0.258 0.116 0.248 0.192 0.12 0.683 0.693 0.451 0.0325 0.183 Ta 0.108 0.146 0.0304 0.106 0.138 0.091 0.708 0.766 0.458 0.0113 0.131 Tl 0.077 0.0739 0.0797 0.0636 0.0412 0.0522 0.0453 0.0529 0.0613 0.044 0.0513 Pb 1.32 1.41 3.46 3.89 4.26 3.34 4.02 3.29 2.28 2.8 2.49 Th 0.392 0.53 0.341 0.74 0.659 0.472 1.98 2.18 1.94 0.393 0.779 U:j: 1.09 0.937 1.21 1.07 1.11 0.857 1.51 1.31 0.869 0.336 0.656 t Values determined using medium resolution ICP-MS; all other values determined using low resolution ICP-MS 139 Table 10. continued ~~m 123 132 141 149 158 158 167 167 177 LOD Li 0.4 0.955 0.894 1.35 1.07 1.24 0.208 0.545 0.503 0.2 Be 0.444 0.767 0.57 1.44 1.24 0.91 1.34 1.11 1.08 0.012 Sc:j: 0.0932 0.151 0.112 0.158 0.334 0.317 0.268 0.358 0.318 0.011 V:j: 0.124 0.0331 0.0134 0.055 Cr:j: 8.81 0.119 0.26 Co:j: 0.635 0.383 0.236 0.273 0.293 0.1 0.345 0.296 0.337 0.022 Ni:j: 0.13 Cu:j: 8.88 3.51 1.69 3.94 2.63 0.808 1.24 1.24 2.25 0.27 Zn:j: 0.636 0.325 0.831 1.23 1.28 0.985 1.5 Ga 1.94 2.18 1.36 1.97 1.77 1.53 1.8 1.42 1.5 0.008 Rb 0.562 0.422 0.526 0.458 0.403 0.334 2.73 2.21 0.499 0.01 Sr 51.8 23.6 31.2 29.1 25.7 21.1 42.3 25.4 26.7 0.13 y 12.5 14 7.53 12 11.7 9.04 12.8 9.06 10.6 0.026 Zr:j: 2.29 4.79 0.475 0.959 0.195 0.235 0.277 0.503 0.166 0.15 Nb:j: 0.534 0.773 0.229 0.178 0.0467 0.077 0.0624 0.104 0.0182 0.037 Cs 0.0241 0.0226 0.0298 0.0312 0.0282 0.0149 0.2328 0.2877 0.0359 0.002 Ba 78.6 96 58.3 46.3 26 24.9 18.6 11.9 11.4 0.38 La 45.3 46.8 27.8 42.9 39.1 33.2 47.1 33.4 36.5 0.066 Ce 76.5 103 70.1 118 93.6 84.2 110 77.1 83.2 0.029 Pr 8.73 9.88 5.62 9.61 8.49 6.95 9.87 6.74 7.92 0.013 Nd 30.6 33 19.5 30.9 27.8 22.8 33.4 22.7 26.4 0.049 Sm 5 5.48 3.28 5.26 4.73 4.05 5.41 3.85 4.41 0.015 Eu 1.39 1.6 0.92 1.47 1.29 1.1 1.46 1.03 1.19 0.005 Tb 0.972 1.01 0.63 0.936 0.866 0.767 0.975 0.729 0.838 0.005 Gd 6.39 7.06 4.06 6.36 5.92 4.93 6.73 4.74 5.63 0.002 Dy 4.51 5.08 2.96 4.76 4.32 3.65 4.86 3.46 4.02 0.01 Ho 0.438 0.545 0.3 0.509 0.471 0.39 0.506 0.377 0.424 0.002 Er 1.09 1.4 0.767 1.31 1.24 1.02 1.33 0.99 1.13 0.01 Tm 0.145 0.199 0.108 0.192 0.181 0.151 0.195 0.146 0.167 0.001 Yb 0.94 1.24 0.681 1.18 1.11 0.932 1.23 0.912 1.04 0.005 Lu 0.112 0.152 0.0826 0.145 0.138 0.115 0.153 0.115 0.131 9E-04 Hf 0.0901 0.16 0.037 0.063 0.0427 0.0362 0.046 0.0388 0.037 0.003 Ta 0.0599 0.105 0.0094 0.0146 0.0058 0.0053 0.0067 0.0048 0.0042 0.007 Tl 0.0606 0.0563 0.0568 0.0742 0.0584 0.063 0.117 0.0708 0.0707 0.003 Pb 2.37 3.49 2.62 2.2 2.54 2.73 2.32 2.42 2.05 0.022 Th 0.392 0.7 0.226 0.225 0.207 0.26 0.124 0.206 0.159 0.004 U:j: 0.535 0.885 0.571 1.28 1.42 1.24 1.22 1.38 1.21 0.007 ~Values determined using medium resolution ICP-MS; all other values determined using low resolution ICP-MS 140 Figure Captions Figure 1. Schematic diagram of the leaching procedure. After samples are dried, they are run through a series of washes intended to attack the water soluble fraction (water), the carbonate fraction (weak acetic acid), the oxide fraction (hydroxylamine hydrochloride) and the organic fraction (hydrogen peroxide). Figure 2. All Concentrations in ppm unless indicated on figure. Water content (WC% ), loss on ignition (LOI%), pH, chemical index of alteration (CIA) and molar Mg#=___!!L_ Mg+Fe (Mg#) are plotted by depth in the first row. Bulk soil concentrations (B%) and the water soluble (W), carbonate and exchangeable (C), oxide (X) and organic (0) soil fractions of the bulk and water soluble fraction ofNa20 is shown because Na was introduced in the buffered actic acid solution used for the carbonate and exchangeable leach. The bulk soil composition of Si02 was estimated as the difference between the sum of all the major element oxide concentrations and unity. The dashed lines represent the bedrock concentrations. Figure 3. All Concentrations in ppm. Bulk soil concentrations (B) and the water soluble (W), carbonate and exchangeable (C), oxide (X) and organic (0) soil fractions of Sc, Co, Cu, Rb, Sr, Zr, Ba, La, Pb and U are plotted by depth. All REE's generally behave like La. The dashed lines represent the bedrock concentrations. 141 Figure 4. All concentrations in ppm unless indicated on figure. Hf/Zr ratios throughout the bulk soil(+) and bedrock (•) are consistent (A). The estimate of the %ilMass of the soil is calculated from soil/bedrock ratios of Al, Ti, Zr, Nb, Hf and Ta (B). Soil/Bedrock ratios averaged over the upper 114 em are plotted for all elements (C). The dotted line runs through a soil/bedrock ratio of 1 representing immobile elements assuming no mass loss. The dashed line runs through the soil/bedrock ratio ofZr, 1.18, representing immobile elements relative to Zr. The termini of the gray triangles represent soil/bedrock ratios averaged over samples below 150 em. Na20/ K20 ratios throughout the bulk soil (+)and bedrock (•) are also consistent (D). Figure 5. Percentage of bulk soil K, Ca, Sr, Ba and U in the carbonate leach (A). Percentage of Al, Mn, Fe, Co, Cu, Zn and Pb in the oxide leach (B). Percentage of Al, Ca, Cu, La, Pb and U in the organic leach (C). All percentages were averaged over the depth ranges indicated on the plot. Figure 6. All concentrations in ppm. Sr/CaO in the shallow (2-114 em) soil (o) and bedrock (dashed line) are consistent; however deep (123-177 em) samples (•) differ (A). Both water soluble ( o) and oxide (x) Sr/CaO ratios are consistent with the bedrock throughout the soil column (B). Ba/CaO in the shallow soil and bedrock are consistent; however deep samples differ (C). U does not correlate well with any element, though U/CaO ratios do tend to higher below 123 em. 142 Figure 7. All concentrations in ppm. Ti02, V, Cu and Pb linearly correlate with Fe203 in the oxide fraction (A). Co and Zr correlate with MnO (B). The bulk bedrock ratio of these elements is plotted as a dashed line. 143 FIGURE 1 Leach;ng Method Ultrapure water• • • ,.. Water solub1e Carbonate & Acetic acid pH 5 • • ·,.. e·xchcmgeable Hydroxylamine hydrochloride • • ot[81'• • ,._ Oxides & 25% acetic acid ma~:..-nl:l &Hydrogen 0.01 M nttnc lle.ro~ ac1d •• f.·.······•.•..• ·.·.·.···.' •· .. -.•...... ;.•~.··· • ..··.· .. · .... Organic fraction 144 FIGURE2 WCo/o LOI% pH CIA Mg# 40 60 eo 100 0 10 20 30 5 5.5 6 0.5 0.6 0.7 0.8 0.2 0.3 0.4 ,I 0 ~ .' • • •• • ~ •••• -.. ~ • • 30 ., .. , < • • • • • • • • -E 60 •• • • • • •• .2,. • • • • • 90 • • • • •• •• • • • • • =at 120 • • •• •• • • c 150 • •• •• • • • • • •• •• • •• 180 • • • • • Na20 Si02* 0 3 6 9%0 400 800 35 40 45 50 55% 0 .. .. ,,, ' • •.v • • 30 • • I • • • • • • -E 110 •• • • s • • • 90 • -• • • • =at 120 • ••- •• c • • 150 • ••- 180 :Bo/o •• MgO 0 2 4 6%0 100 200 300 0 300 600 900 0 100 200 0 100 200 0· ' •• . •• ' ' •• I,.--· • • , •• " • • •• ., 30' • . .. • • • • < r"' • • • • • -E 60 • • • • •• u • •• •• .. • • • • • -.t::. 90 •• • • •• -•• c. -Q;) 120 • .. • • • • c • • •• •• • 150 •• • •• • • • 0 • • • • 180 B /o • -• Al203 15 20 25 30%0 100 200 0 2000 4000 0 1 2 3 4% 0 1 2 3 45% G' '". FIGURE 2, continued P205 0 0.5 1 1.5 2% 0 10 20 3040 0 20 40. 60 100 300 500 700 0 2000 ...... •.. . . ' 0 • • .. !I" •• .,. • •• ,. 30 • ., J •• .t • • • • • -Eao • • •• • • •• • • • • ..,2, • • • • - • i:i 90 • • • • • 0. • • • • 11)120 • • •- • • • Q • • • • .. • ••• ••• •• •• 150 • ..• •• • • 180 ·:aolo • •• •• K20 0123456%0 200 400 0 1000 2000 0 1000 2000 0 500 1000 O• ... \ I l ----.; ~ • • ! • _., • • ••• ,,'·~ ...... -E •• • • • • • • ,B. eo • • .. - .E;GO • •- •• • c. • : • • • • • • ~120 • • •• • • 1!0 • •• • • • 8°/o ·c • • •• 180 • • • • CaO 5 10 15% 0 400 800 1200 0 6000 12000 0 2500 5000 0 4000 0 ' • ••• • • • • ,.. 30 • ..... -··•• :.t· • •- • • • • Eao • •• • • • • • & • • •- • - • =90 • • ..• •• • • • ih2o • • • • • • c • •• •• •• • • • 150 •• • • :e• • •• • • •• ••• • 160. • • • • 146 FIGURE 2, continued Ti02 0 1 2 3 4%0 5 10 0 3 6 9 12 200 300 400 500 0 200 400 0 ! ..• , L• • ••.,. '). ·~ • • tf • ...... , • • • • '·• • • • • • • • • •- •• • • • -• •• ••• •• • •• • - • • • • • • • • • • • • • • • • ••• • • • • • ••• 8°/o-:.• • ·c • ••• 180 • • • • "'- MnO 0 0.2 0.4%0 10 20 30 40 0 250 500 0 500 10001500 0 40 80 120160 0· .. • • ••• , .... 30 • • • •• ••I ., . • • • • • • •• •• • -5 60 • • • • • • • •- • • .c: 90 - - •• •• • •• • • 0. • • • - • • • ~120 • • •• • • • • 150, •• ••- •• •• • •• •••• 180 BYo • • • Fe20 5 10 15%0 20 40 50 100 1504000 7000 10000 0 500 10001500 0 '! . ~ ... ., •• \,'• • .. •• -: "·• •.. • ' • •• ":.• • • • • .• .• • ' • • ••• • • • • • • • • • • • • • • •• ••• • •• • • • •• •• • -• •• •• • • • • • • • • •• •• 1!!0 • • • •• • • ~ w: X • • 11!0 ·c· • 147 FIGURE3 Sc 0 5 10 15 0 0.001 0 0. 1 0.20.30.4 0 0.2 0.4 0.6 0 0.2 0.4 0! •• ••• 30 •"' • I I ( -E • • •• • • 0 60 •• • .. • • - • • • -• ••- 90 • • • c. • • • • =1.1) • • • • • • Cl 120 • • • • ••• 150 • • • • -• • • w - • ··x_. o·. 180 - • " Co 0 20 40 0 0.05 0.1 0 0 .. 5 1 1.5 0 3 6 9 0 1 2 3 0· ' .. , . , . •• .·:~~, • .. .-f' .. 30 • . • • • • -E 60 • • ' • • • • 0 • •• • • -• • • • •- • -.r::. 90• •• •• • ••• • • • c. • • • -• • • ~- 120 • • • • •• • ••• • •• • •• 150· • •• •• • • •• • • • •• 180 • • • - • Cu 0 50 100 0 10 20 0 3 6 9 12 0 6 12 18 24 0 20 40 60 0· .•. .,_ • •• • 'i .... 30 • • J •... • .~ • • • ·' -6 60 •• • •• • ••• •• • - -.r::. • • • '5. 90 • • • ••• • -•• 120 • • -• • • ~ • • ••• • • • • 150 • • • • •• •• (I •- X • •.. 180 • - - Rb 0 50 100 150 0 1 2 0 5 10 0 4 8 12 0 4 8 12 0 ... ' •• . •• •••• • • • • • • "' • • • • 30 • ..• •• JA • .. , . • • • • -g 60· • • • • •• • • • - -.r::. 90· • • • - • • '5. • • • • • 120: • • • • ~ • •• • • • 150 •• • • •• • •• •• •• • 180 • • • • - 148 FIGURE 3, continued Sr 500 1500 2500 0 10 20 30 0 200 400 0 50 100 150 0 50 100 150 0· •• •• .:• n.• '(.· ...., 30 • • ., • 1r ... •I • • • .. • •• 5 60 •• •• • • •• " ..• • • - " • -..c:. 90 • • •• • •• 0. • •• • • ... ~ 120 " •" • •" • •• 150 • • • • •• :.a • • •• ••• 180 • "• • • • Zr 500 600 700 0 0.1 0.2 0.3 0 2 4 6 0 50 100 0 10 20 30 0• • "'" ,_.0 •• _.,, • • ... ·~·.,. 30 • • •• • 't • • • '·• • • -E 60 •• •• • • •• • • • ..£. • • • -• • • ..c:. 90· • • •• • • •• •• 0. • • • • • ~ 120: • • • • • • • • • • ••• • • 150· • - • ·a: w· • - ~ • • 180 • • • • • • Ba 1000 2000 30000 2 4 0 1000 0 200 400 0 100 200 300 0· >It"" j, ,, .• -~ ' ' ' ,. • • Jl! ••• ·~ ,~JA •• 30 • • • • ~ c :.: • • • • • -5 60 • • • •• • • • • • • - • • • • -..c:. 90 • • • • • 0. • • • - 120 • • -• -• • • ~ • ••• • • •• •• • 150 • • •• • • •• p • • C· .·)( •• 180 - • • • • 149 FIGURE 3, continued La 100 200 300 0 0.1 0.2 0.3 0 20 40 60 0 20 40 60 0 50 100 150 0• • J.' l ') ., !.'It'. 30 •• ~ c ,. • • • • • -E 60 • • ••• •• •• •• • 0 • • • • .. 90 • • • • • -J:::. • - • • • • -0..120 • • • • • • • • • 8 • •• • •• • •• • -• 150 • ··a w c· = • • • • 180 • - - -• Pb - 10 20 30 0 0.2 0.4 0.6 0 0.5 1 1.5 0 5 10 0 2 4 6 0 •• •• ' ... • ..... • ...... t, '. ~·• 30 • • • •• • • -.: .. • • • • • -60 •• • ••• •• • •• E • • • • • • & 90 •• • •• •• • • =0..120 • • • • • FIGURE4 Bulk Hf-Zr 14 12 10 <>~ Bedrock~ 4 2 A 0 0 200 400 600 BOO Zr ..6._Mass (o/o) -40 -20 0 20 40 0 30 60 90 120 150 8 180 151 FIGURE 4, continued Soil/Bedrock Ratios c 0 1 +,-, .... ,--,--·-T·"-T""""T""T-,·TTr-"""""··-" TC ...... T .. ,,.T.. ""T "T"TTTT""" ..... , •• ,., •• ".,,.• "T" "T.. T""C"T"Tl 0 0.1 10 1000 100000 Bedrock 8 Bedrock • '(<.6 a:s z #. 4 2 D 0 ~------~ 0 2 4 6 0/o K20 152 FIGURES ~ Q) 40 ...... ,..c(1) c::a>co co 30 oOl ..at: co..c:t... co 20 (_)(.) o>< 0' Q) 10 50 20 10 2-18 em 26-61 em 78-114 em 123-177 em 30 Q) .0 "C Q. ")(20 0 ~ 0 10 2-18 em 26-61 em 78-114 em 123-177 em 153 FIGURE6 Carbonate Sr-CaO 400 ID 2~114cm 1 II Ill 123-177 cml 300 ._ C/)200 II 100 A 0 0 3000 6000 9000 12000 CaO Water & Oxide Sr-CaO 20 X ._ 15 en 10 5 0 8 0 0 400 CaO aoo 1200 154 FIGURE 6, continued Carbonate Ba-CaO 0 2-114 em 111 123-177 em 1200 Ill ..• 800 ea • 1:0 Ill 400 • D oo ClO ~ D 0 D or9 o D 0 c 0 3000 6000 9000 12000 CaO Carbonate U-CaO 4 o 2-114 em II 123-177 em 3 :;) 2 ') 0 ~ D 0 D Cl D 0 1 0 0 0 0 D Cl 0 D 0 .~~------~------~ 0 2000 4000 6000 8000 10000 12000 CaO 155 FIGUIU: 7 Scavenged by Fe Oxides (A) 500 ; X 20 I I ~~ 400 ~~ .. · 15 3oo xi:$ 0 > t= ~ 10 200 100 5 0 --~~--+------+-, --····--··~--·~··· 0 3000 6000 9000 00 3000 6000 9000 Fe20s Fe20a I X)( / ' 16 X / 8 X f F Xj'E" ' 12 X' 6 X ~ :::f / .Q (.) 8 " a..4 / " I " *I ~ / ~X 4 2 I' I / 0 0 0 3000 6000 9000 0 3000 6000 9000 Fe20a Fe203 156 FIGURE 7, continued Scavenged by Mn Oxides (B) 8 X. 6 0 u X 4 X X r·' 2 #/ o------o 500 1000 MnO ;@ X ~' 80 J' ~·· ~: 60 '- N 40 X X 20 0'------0 500 1000 MnO 157 APPENDIX A: Investigating the distribution of labile metals in a vertisol in southeastern Texas 158 Investigating the distribution of labile metals in a vertisol in southeastern Texas Synopsis Concern for human and environmental health inspires a better understanding of the persistence of heavy metals in soils. This is an investigation into the behavior of a suite of elements in a vertisol in southeastern Texas. A dilute 2% HN03 leach was performed on 6 samples down to 122 em depth. Leachable concentrations were determined using an ICP-MS. Chemical weathering is evidenced by surface depletions of elements considered to be mobile relative to the bulk soil such as Ca and Na. However, surface enrichments ofNb, Zr, Zn, Fe, Mn, K, Cd, Co, and Pb cannot be explained by the same process. The concentration ofNb, Zr, Fe, and Mn may increase as a result of the loss of the mobile major elements. Cd, Co and Pb enrichments are most likely to be due to surface deposition. There is also evidence of organic matter in the A/0 horizon and Fe/Mn oxyhydroxides throughout the soil column, either of which may provide an adsorption surface for these metals. Deposition is concluded to be recent as these elements were readily leached from the samples and thus should not persist in a highly weathered surface layer. 159 1. Introduction The purpose of this study is to investigate the nature of labile metals in the soil column. Such information is crucial for contamination and ecological studies because labile metals are available to plants, groundwater, river water, and eventually humans. Quantification and analysis of metals of environmental and ecological importance, including Pb, Cd, Cr, Cu, and Zn, have been studied extensively in soils (Andresen et al., 1980; Blaser et al., 2000; Campos et al., 2003; Mortvedt, 1987; Nicola et al., 2003). Investigations of trace metals have also ranged to analyses of potential inputs such as aerosols, fertilizers, and manures (McBride and Spiers, 2001; Salam et al., 2003). This paper specifically addresses the origin and fate of labile elements, focusing in particular on trace-elements. To these ends, a dilute acid leach was performed on a soil column as a proxy for geologic and/or anthropogenic processes that transport labile trace elements. A sample site in rural, southeastern Texas was chosen in order to complement work done by Stiles et al. (2003) who investigated the effects of weathering on bulk major element systematics. Here, the nature and origin of labile trace elements in these soils is investigated. 160 2. Sample Descriptions Samples were collected at 29°36.105' N, 95°48.683' W,just north of the town of Rosenburg, at 30m elevation in Fort Bend County, Texas on 19 October 2003. This location lies approximately 10 m from Farm Road 723 on former agricultural property that has lain fallow for at least three years. The soils are smectitic vertisols from the Pledger soil series (Service, 2004). This series is derived from the Brazos River and represents a 4000 y terrace (Stiles et al., 2003). Mean annual temperatures range from 20.1 to 23.6 °C and mean annual precipitation is~ 1000 mm (Stiles et al., 2003). Six soil samples were collected at ~30 em intervals to a maximum depth of~ 1.2 m using a 40 mm diameter split tube sampler (Dormer Engineering). The soil samples were cored at 0-5 (A/0 horizon), 28-33,66-71, 104-109, and 119-124 em depth (each is identified by their average depth, e.g., 2.5, 30, 69, 84, 107, and 122 em, respectively). (See Fig. 1 for a schematic stratigraphic depiction of the soil column.) The 2.5 em sample is reddish black (2.5YR 2.5 /2, in situ) silty clay with~ 1 mm diameter roots present. It consists of ~50% fine grained humic material and ~50% a mixture of quartz and feldspar grains all less than 0.005 mm diameter. Slightly larger feldspar grains, <0.01 mm diameter, distinguished the 30 em sample from the 2.5 em sample. In the field at 40 em depth, a transition from a reddish black, humic horizon to a clayey, dark red horizon was observed (2.5YR 3/6, in situ). The 69 em sample was a firm, dark red low plasticity silty clay with quartz and feldspar grains <0.01 mm. The plant content appeared to be no greater than 1% at this depth. The 84 em sample was in the second observed transition zone from dark red clay to the red silty clay (2.5YR 4/8, in situ). This sample was dry to the touch and well compacted. There was no visible macroscale plant 161 content. The 107 em sample was hard, red silty clay. The 122 em sample was a very hard, red silty clay with more silt content than the 107 em sample. These three deepest samples were all dominated by fine quartz grains of <0.05 mm diameter, larger feldspars of up to 0.1 mm diameter, and unidentified black accessory minerals of <0.05 mm diameter. The physical characteristics of the soil columns do not suggest any recent, significant overturn processes; therefore, transport and reactive processes may account for elemental distribution in the soil column. 162 3. Methods Physical Analyses Each sample was cut to the same length from the cored soil column. Each soil sample was weighed while wet (Mwet) and after drying for 72 hours at 1oooc (Mwooc). This allowed for the calculation of soil moisture (fsoil moisture) (Holtz and Kovacs, 1981 ): fsoil moisture=MwetiMI00°C (1) Dry density (pdry) was also calculated from Mwooc and the volume of the cored sample (Vtotai) according to equation 2. Pdry= M10oocfVtotal (2) Approximately 1 g of each dried sample was weighed (M100oc) and then heated at 400°C for 20 minutes in a muffle furnace (M4oooc). The estimation of the volatile fraction (fvolatile) follows from (3) Chemical Analyses The leaching procedure is as follows. ~- 1 g of each 100 °C-dried so ill sample was mixed with ~4 rnL of2% HN03. The acid soil mixture was agitated for 10 minutes and centrifuged for 3 minutes. The supernatant was then removed and the leaching procedure repeated 2 more times with fresh HN03, giving a total of three leaches. The supernatants of the three leaches were combined, micro-centrifuged and analyzed directly by solution ICP-MS. All concentrations are reported relative to the soil mass in ppm. A natural rock standard, USGS basalt standard BIRl, was used as an external standard. To correct for instrumental drift, samples, standard, and blank were spiked with Indium as an internal 163 standard. The ICP-MS was run in low and medium resolution modes (m/&n=300, 4000), the latter allowing for the analysis of Na, P, K, Ca, Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Zr, and Nb, for which isobaric molecular interferences are a problem in LR. The leachable concentration of an element, X, is defined as: X = g element leached I g of dry soil (4) Lead Isotope Analyses In order to assess whether there is an external input to the labile trace metal budget in the soils, the Pb isotopic composition of leachates were analyzed. Separate aliquots of the leachates were directly analyzed for 204Pb, 206Pb, 207Pb, and 208Pb. Because of isobaric interferences from 204Hg and errors introduced from dead-time uncertainties on the multiplier, only the 207PbP06Pb and 208PbP06Pb ratios are reported. The reported ratios represent the average of 100 measurements and the in-run uncertainties are reported as 2 times the standard error (e.g., 2*STD/"n). Mass bias was corrected online using the measured 205TlP03 Tl and assuming an exponential fractionation law (the aliquots were pre-spiked with normal Tl). All isotope measurements were corrected for procedural blank even though the blank correction was negligible (the concentration of the procedural blank was determined by external standard as discussed above). Measurements of United States Geological Survey rock standard BHV02 are within error of accepted values. 164 4. Results Physical Properties All physical characteristics of the samples are reported in Table 1 and Fig. 2. The soil moisture of the samples decreases with depth (Fig. 2A). There are two main factors that control soil moisture: porosity and degree of saturation. An increase in the level of compaction and an increase the silt content are consistent with a decrease in porosity. The 2.5 em sample is a clear outlier; but the relatively low soil moisture is probably due to exposure to air currents and sunlight. The volatile fraction of the soil also decreases with depth (Fig. 2A). A major constituent of volatile mass is expected to be organic carbon. The majority of the water present in the soils was removed during the 100°C drying process; therefore, the fvolatile trend is consistent with the observed decrease in organic matter content with depth. The dry density, pdry, increases with depth (Fig. 28). Dry density is a function of density (p) and porosity () where Pdry = p*(l-<1>) (5) Therefore, the increase of dry density is also consistent with the observed decrease in porosity. Geochemistry The leachable concentrations of trace metals are shown in Table 2 in terms of ~g per g of dry soil (ppm). The elements can be grouped according to their relative concentration profiles. Some elements, usually those in lower concentrations, show no significant surface depletion or enrichment, e.g. the rare earth elements (REE), whose leachable concentrations in the soil range from 0.01 ppm to 10 ppm. A smaller number 165 of elements are clearly depleted at the surface of the soil profile, e.g. Ca, Sr, and Na. A third group of elements are distinctly enriched at the surface and depleted at depth, e.g. Nb, Zr, Zn, Fe, Mn, K, Ph, Cd, Co, Be, Ga, and Cu. These element groups are discussed below. Rare Earth Elements In Fig. 3A, the leachable concentrations of the REE are normalized to average upper continental crust (UCC) values (Rudnick and Fountain, 1995). The first general observation is that the REEs are depleted by a factor of0.2 to 0.7 below UCC. The second observation is a variable but positive Eu anomaly for all samples and a negative Ce anomaly in the 30, 69, and 84 em samples. Those samples with the negative Ce anomalies are generally accompanied by the largest positive Eu anomalies. Fig. 3B plots the REEs against depth. In order to plot all REEs on the same diagram, they are normalized to their 122 em datum. It can be seen that all REEs, except for Ce, are enriched slightly in the upper ~ 100 em relative to the two deepest samples. Importantly, all REEs, except for Ce, show a roughly constant profile with depth from 2.5 to 69 em; Yb, Er, and Lu are constant down to 107 em. The REEs, with the exception of Ce and Eu, exhibit remarkable uniformity with depth. For this reason, in order to directly compare depth profiles of various leached trace metals, all remaining elements are normalized to one of the average REEs, La. This normalization will also minimize any inconsistencies in the leaching technique (Brimhall et al., 1991). Surface-Depleted Elements 166 Fig. 4A shows all elements with a clear depletion profile. There appears to be a strong depletion inCa, Sr, Na, and Pat depths equal to and shallower than 69 em. Ca and Sr exhibit the strongest depletions while Na shows a smaller depleted zone that can only be resolved to 30 em. Pis also depleted, but does not follow the simple shape of the Ca, Sr, and Na profiles. There appears to be a bimodal distribution of Ca/Sr ratios (Fig. 4B). Below 84 em, Ca!Sr>800; above 69 em, Ca/Sr Surface-Enriched Elements Those elements that show distinct enrichments at the soil surface are plotted in FigureS. Figures SA and SB show those elements whose enrichments appear to be confined within the 2.S em datum. Nb, Zr, Zn, and Fe are enriched by more than a factor of 3 relative to the deepest sample (Fig. SA). Mn, K, Ph, Cd, and Co are enriched at the surface by of a factor of 2 relative to the deepest sample (Fig. SB). Figure SC presents those elements whose surface enrichment penetrates deeper than the 2.S em datum. These are Be, Ga, and Cu, which are enriched at the surface by a factor of 3-7. These elements show a more gradual decrease in concentration from 2.S down to 84 em. Lead Isotopes Table 3 and Fig. 6 show the 207PbP06Pb and 208PbP06Pb ratios of leached Ph as a function of depth in the soil column. It can be seen that the 207PbP06Pb ratio in the 2.S em sample is significantly higher than all deeper samples (the 208PbP06Pb ratio shows no significant behavior with depth). 167 5. Discussion The depletions inCa, Sr, Na, and Pin the upper 70 em of the soil column are likely due to the effects of long-term weathering. Ca and Sr have similar chemical properties: both have high solubility and both are incorporated into the same minerals. In particular, source Ca!Sr ratios may vary considerably (Blum et al., 2002; Kataba-Pendias and Pendias, 2001). The upper 69 em, which is characterized by low Ca!Sr (<500) is consistent with atmospheric values (Blum et al., 2002), precipitation, and throughfall (Probst et al., 2000). Below 84 em the Ca!Sr ratio is >800, suggesting mixing with a silicate source containing high Ca!Sr. P increases with depth in the soil column suggesting the presence of apatite. Apatite has a Ca!Sr ratio >2000 and therefore may be responsible for the increase in Ca!Sr ratio below 84 em. In this context, the distinct surface enrichments in Nb, Zr, Zn, Fe, Mn, K, Ph, Cd, Co, Be, Ga, and Cu are problematic. Many of these metals are readily soluble, so a depleted profile similar to Ca, Sr, and Na is expected due to the time-integrated effects of weathering. Their surface enrichments thus suggest that there must be a recent input of these elements superimposed on a preexisting weathering profile, or that there is a steady state enrichment of these elements towards the surface. One possibility is that these elements may be enriched and biocycled in the surface soil. However, enrichments ofPb, which is not a nutrient, strongly suggests contamination (Johnson and Petras, 1998). This is further supported by a 207PbP06Pb ratio at 2.5 em that is significantly different from all deeper samples and hence requires that there is an external source of Ph at the surface (Wilcke et al., 2001 ). Therefore, the surface enriched metals are most likely derived from 168 anthropogenic atmospheric deposition (including deposition of dust, soot, rain, and/or other particles in the atmosphere). If these surface enrichments are recent, it follows that there must be a phase that temporarily sequesters them at the surface. One possibility is organic matter. Indeed the reddish black color of the shallowest sample (2,.5 em) is evidence of enrichment in organic matter. Another possibility is Fe-Mn oxyhydroxides. The red color of the soil is indicative of oxidized Fe. High concentrations of leacheable REEs may reflect REEs scavenged from meteoric waters by Fe-Mn oxyhydroxides in the soil column. The negative Ce anomalies for 30, 69, and 84 em samples and the accompanying positive Eu anomalies also suggest that redox reactions and Fe-Mn oxyhydroxides were involved in the history of these soils and/or the pore waters. Similar negative Ce anomalies have been reported and similarly interpreted in oxidizing groundwater (Dia et al., 2000). 169 6. Conclusions The surface-depleted Ca, Sr, and Na profiles indicate extensive weathering (Fig. 4); yet, superimposed on this weathering profile is a surface enrichment ofNb, Zr, Zn, Fe, Mn, K, Pb, Cd, and Co. These metals must be temporarily held in an authigenic phase in the surface of the soil column. The observations suggest that Fe-Mn oxyhydroxides and/or organic matter can sequester environmentally toxic metals and prevent them from entering the ground water system. However, this sequestration mechanism is temporary as changes in soil pH or physical erosion can remobilize these metals. It is probable that external surface deposition was recent because these elements were readily leached from the samples and thus should not persist in a highly weathered surface layer. The ultimate source for these metals may be direct wet or dry deposition from automobil~::s, exhaust, polluted rainwater, or even fertilizers. 170 Table 1. Physical characteristics of the soil column Depth (em) fvolatile fsoil moisture 2.5 0.031 0.13 1.2 30 0.034 0.18 0.82 69 0.024 0.16 1.4 84 0.029 0.15 1.4 107 0.021 0.083 1.5 122 0.013 0.070 1.5 171 Table 2. Element concentrations and Select Element Ratios m La Ce Pr Nd Sm Eu Tb Gd D Ho 2.5 em 9.11 21.4 2.89 11.8 2.72 0.596 0.385 2.43 1.80 0.305 30em 9.43 14.1 2.94 12.1 2.82 0.625 0.389 2.50 1.79 0.302 69em 8.93 11.0 2.94 12.3 2.70 0.665 0.399 2.57 1.83 0.314 84em 10.2 19.3 3.11 12.4 2.34 0.535 0.304 2.47 1.39 0.233 107em 8.17 16.6 2.53 10.6 1.87 0.420 0.262 1.69 1.23 0.217 122em 5.81 12.0 1.81 7.56 1.41 0.314 0.200 1.28 0.92 0.161 BLANK t 1.5 2.2 0.3 0.8 0.2 1.4 0.03 0.3 0.1 0.02 m Er Yb Lu Sm Eu Hf Ta Tl 2.5 em 0.603 0.460 0.0705 2.72 0.596 0.0299 0.00245 0.0111 30em 0.587 0.445 0.0670 2.82 0.625 0.0199 0.00248 0.00963 69em 0.577 0.492 0.0684 2.70 0.665 0.0157 0.00288 0.00618 84em 0.551 0.481 0.0669 2.34 0.535 0.0183 0.00264 0.00699 107 em 0.588 0.476 0.0671 1.87 0.420 0.0147 0.00232 0.00835 122em 0.444 0.359 0.0510 1.41 0.314 0.0103 0.00174 0.00411 BLANK t 0.1 0.1 0.0 0.2 1.4 0.1 0.3 0.2 m Pb Th u Na* P* K* Ca* Se* Ti* V* 2.5em 10.7 0.088 0.467 22.5 104 737 3760 0.0692 1.50 7.43 30em 3.84 0.073 0.255 23.9 34.7 447 5010 0.0908 1.35 3.78 69em 4.03 0.154 0.155 41.3 154 428 13400 0.182 1.66 2.90 84em 4.94 0.154 0.276 50.5 99.0 411 54200 0.140 1.57 3.37 107em 3.60 0.152 0.275 47.3 136 323 65400 0.147 2.51 2.33 122em 2.50 0.105 0.226 37.6 106 222 46800 0.0910 1.36 1.96 ppt ppt ppt ppb ppb ppb ppb ppt ppt ppt BLANK 15.3 0.2 0.2 156.7 2.0 210.4 40.4 2.3 111.6 5.1 m Cr* Mn* Fe* Co* Ni* Cu* Zn* Zr* Nb* Ca*/Sr 2.5em 1.46 361 1131 3.04 5.34 5.20 7.07 0.544 0.00469 278 30em 1.00 166 399 1.70 3.10 2.99 1.67 0.243 0.00173 244 69em 0.851 152 318 1.39 1.33 1.72 1.07 0.126 0.00190 457 84cm 0.902 215 260 1.70 2.41 1.43 0.789 0.195 0.00188 842 107em 0.768 139 221 0.739 0.612 0.908 0.784 0.148 0.00262 1070 122em 0.354 117 121 0.654 0.314 0.481 0.843 0.0818 0.00092 896 ppb ppt ppb ppt ppt ppb ppb ppt ppt BLANK 0.7 97.1 2.9 4.0 61.2 1.1 1.7 3.3 0.3 *MR-ICP MS, all other samples LR-ICP MS 172 Table 3. Lead isotope ratios lll7Pb I llllipb 2SE lll8pb I llllipb 2SE 2.5 em 0.8283 0.0037 2.054 0.0092 30cm 0.8177 0.0053 2.073 0.012 69cm 0.8184 0.0044 2.027 0.011 84cm 0.8198 0.0022 2.0252 0.005 107cm 0.8192 0.0026 2.0529 0.0056 122cm 0.8151 0.0021 2.0328 0.0051 173 Figure captions Figure 1. Schematic stratigraphic section. Relative humic, clay, silt, and sand content plotted by depth in the soil column. The change in maximum grain size, color, and level of compaction with depth are also reported. Figure 2. {A) Fraction of volatile matter, fvolatile (.6.), and soil moisture, fsoilmoisture (+)in the soil column. (B) Dry density, Pctry ( • ), in the soil column. Figure 3. (A) The leached REE concentrations in the soil divided by Upper Continental Crust (UCC) values (Rudnick and Fountain, 1995). (B) REEs normalized to their 122 em datum. Figure 4. (A) Leached concentrations ofCa (+), Sr (•), Na (.6.), and P (o) normalized to La and to the 122 em datum. (B) Ca/Sr ratio ( .6. ). Figure 5. (A) Leached concentrations ofNb (•), Zr (•), Zn (.6.), and Fe(+) normalized to La and to the 122 em datum. (B) Leached concentrations ofMn (•), K (.6.), Pb (•), Cd (o), and Co (x) normalized to La and to the 122 em datum. (C) Leached concentrations of Be ( • ), Ga ( .6. ), and Cu ( •) normalized to La and to the 122 em datum. '<':t" ...... r- loose I <------firm------> I <---hard---> I <---very hard--> <--reddish black--> I <-----dark red-----> I <------red------> ~~"~E ~E 0 v "CC ~~silt E E E E E E E E u u u u u u u u 0 0 0 0 0 0 0 0 N ~ \,() co 0 N ~ """'" """'" """'" ~ ~ 8 -~ 175 FIGURE2 0.5 1 1.5 2 0 20 .-..... E 40 • .._..0 ..c:...., 60 5} 80 • Cl • 100 • 120 • B 140 176 FIGURE3 REE Normalized to Upper Continental Crust (UCC) 0.8- 0.7 -+-2.5 0.6 -o-3o ...... ,_69 0 o.s ~84 u ---107 ~OA ·s ...... ,_122 (/) 0.3 0.2 0.1 -i OA ~ ~ ~ ~ ~ ~ ~ Th ~ ~ & ~ ~ REE Normalized to 122 em 0 1 2 3 0 20 -<>-La -....ce 40 ...... Pr -E -*"""Nd &60 ...... sm ..£:: -Eu -arso -o-Dy 0 -tr-Ho 100 -+-Yb -a-Lu 120 -o-Er -Gel 140 B 177 FIGURE4 Normalized Concentrations 0 0.2 0.4 0.6 0.8 1.2 0 20 - 40 ~60 ... .c aQ) ao 0 +Ca 100 •Sr 120 &Na oP A 140 Ca/Sr 0 500 1000 1500 0 20 40 -E ..8.- 60 .J:: c. 80 -Q) Cl 100 120 B 140 178 FIGURES Normalized Concentrations 0 2 4 6 8 0 20 -40 5 -.c: 60 0.. G) c 80 100 120 A 0 1 2 3 4 0 20 -40 •Mn 5 .t.K -.c: 60 a.. ePb G) oCd Ceo :::r:co 100 120 B 0 1 2 3 4 5 6 7 8 0 20 -40 E •Be -8. •Ga :5 60 a. •cu G) c 80 100 120 c 179 FIGURE6 207pb f206pb 0.81 0.82 0.83 0.84 0 20 40 -E 0 60 0 £i- ~ 80 1--(>--1 0 100 t---0-----1 120 ~ A 140 20apb 1 206pb 2.00 2.02 2.04 2.06 2.08 2 .. 10 0 20 • 40 -E 0 60 II -.s:::: 15.(J) 80 ~ 0 100 I • I 120 B 140 180 APPENDIXB: Bulk composition tables from Mweka, Olmani and Monduli 181 Table 1. Mweka major elements (MR-ICP-MS) protolith XRF 2cm 6 Na20 5.62 13.5 12.9 9.99 9.60 0.139 0.248 0.297 0.318 MgO 1.89 3.10 2.33 2.73 2.65 0.805 1.19 1.14 1.51 AhOJ 18.1 33.3 32.6 27.3 28.0 52.8 51.2 50.6 55.8 Si02 53.9 51.0 53.8 44.2 49.8 21.7 24.0 19.9 22.5 P20s 0.90 0.426 0.907 0.623 0.470 1.13 1.26 1.34 1.17 K20 3.78 8.76 9.64 7.23 6.35 0.868 0.956 0.940 1.37 CaO 4.60 5.52 5.86 5.95 5.19 0.289 0.384 0.331 0.586 Sc 6.05 7.45 6.75 10.2 8.2 21.0 21.2 16.5 18.1 Ti02 1.93 1.93 1.93 1.93 1.93 3.83 3.74 2.59 2.52 v 21.5 12.0 11.6 14.8 16.8 176 182 238 151 Cr 7.0 2.5 1.6 0.5 0.3 137.6 117.9 101.3 137.4 MnO 0.242 0.438 0.343 0.323 0.284 0.219 0.251 0.272 0.162 Fe203 10.0 13.9 12.5 11.0 12.2 25.6 23.9 30.0 20.9 Co 11.5 11.4 12.1 13.6 17.2 27.0 31.9 17.0 Ni 1.9 0.3 0.3 0.4 0.6 68.0 74.5 101.8 62.1 Cu 25.0 28.8 31.6 29.1 48.1 175.3 217.2 237.3 174.0 Zn 164 156 150 172 170 168 188 192 174 llcm 15 19 24 Na20 0.316 0.270 0.180 0.691 0.253 0.311 0.251 0.308 MgO 0.833 0.792 0.956 0.927 0.931 0.802 0.910 0.981 Ah03 55.5 52.3 58.6 63.0 61.3 62.8 62.1 59.6 Si02 21.5 29.9 18.5 17.0 19.3 19.1 16.3 19.5 P20s 0.922 1.00 0.784 0.790 1.03 1.06 0.689 0.793 K20 1.10 0.859 0.743 1.63 0.858 0.846 0.578 1.07 CaO 0.154 0.303 0.157 0.211 0.148 0.112 0.114 0.198 Sc 16.0 16.8 14.1 21.3 13.8 14.3 14.5 14.1 Ti02 2.72 3.02 2.42 2.47 1.85 2.37 2.06 2.59 v 141 146 162 143 123 138 140 135 Cr 98.1 116.0 147.0 122.3 171.8 124.5 127.0 115.8 MnO 0.247 0.130 0.184 0.169 0.171 0.100 0.143 0.197 Fe203 23.8 18.1 24.8 20.2 21.3 19.7 25.8 23.5 Co 19.2 14.6 26.1 16.2 17.4 13.3 19.4 17.3 Ni 58.8 54.1 75.7 54.1 47.6 59.2 54.8 44.9 Cu 172.1 126.5 214.2 233.0 108.2 176.8 133.2 102.9 Zn 152 113 95.2 92.2 98.4 155 103 162 182 Table 1. continued 28cm 45 55 64 Na20 0.198 0.624 0.464 0.437 0.349 0.141 0.126 0.150 0.145 MgO 1.04 0.906 1.20 1.24 1.22 0.69 0.70 0.84 0.86 Ah03 61.1 53.9 53.5 58.5 60.5 49.9 59.9 60.9 56.1 Si02 16.2 23.3 19.1 18.5 17.3 26.1 16.8 16.7 21.6 P20s 0.697 0.741 0.707 0.718 0.629 0.780 0.779 0.785 0.956 K20 0.740 0.999 1.00 1.09 0.720 0.407 0.553 0.623 0.417 CaO 0.0860 0.329 0.561 0.400 0.695 0.0645 0.129 0.120 0.0949 Sc 10.7 12.1 13.2 14.7 18.9 15.0 18.6 20.4 21.2 Ti02 1.88 2.13 2.58 2.89 2.91 1.59 2.51 2.48 2.81 v 136 172 150 166 162 129 159 164 179 Cr 149.5 134.5 139.6 122.8 140.4 165.9 158.0 182.6 150.2 MnO 0.180 0.295 0.170 0.168 0.190 0.158 0.133 0.134 0.136 Fe203 26.8 25.9 30.3 25.1 24.3 29.7 27.6 26.2 26.1 Co 18.2 24.6 20.0 24.6 22.5 19.6 16.3 17.8 18.4 Ni 56.0 50.1 53.9 59.6 52.2 55.1 56.9 58.1 62.7 Cu 84.6 89.1 82.0 82.8 79.5 89.8 84.5 77.6 90.3 Zn 84.8 104 101 102 110 89.8 108 109 125 83 em 93 102 Na20 0.164 0.248 0.0786 0.0466 0.0588 0.0252 0.0446 MgO 1.09 0.99 1.00 1.65 1.25 1.39 1.16 Ah03 65.7 66.5 68.8 63.8 55.5 68.1 74.3 Si02 13.2 14.6 14.6 13.0 23.0 15.2 10.6 P20s 0.498 0.407 0.602 0.607 0.762 0.457 0.575 K20 0.394 0.386 0.336 0.327 0.273 0.389 0.294 CaO 0.143 0.0838 0.188 0.139 0.483 0.113 0.130 Sc 22.5 23.7 29.4 23.4 21.3 18.3 18.5 Ti02 1.97 2.40 2.90 2.53 2.45 1.87 2.31 v 144 161 179 220 178 160 212 Cr 182.1 143.2 154.9 138.3 116.3 80.7 81.1 MnO 0.240 0.226 0.184 0.331 0.247 0.233 0.671 Fe203 25.3 22.6 19.6 24.4 23.3 18.4 16.0 Co 21.9 19.5 20.7 38.4 35.5 30.7 57.7 Ni 63.1 69.7 72.7 103.1 109.6 108.9 142.1 Cu 55.2 56.4 58.4 62.3 69.1 61.1 116.9 Zn 62.9 78.3 87.7 110 137 121 342 183 Table 1. continued 122cm 131 140 Na20 0.0348 0.0513 0.0729 0.0459 0.0675 0.0450 MgO 1.51 2.64 1.27 1.33 1.02 1.15 Ah03 62.4 50.9 56.4 56.5 53.3 58.2 Si02 11.4 14.6 15.7 14.8 20.9 15.6 P20s 0.419 0.510 0.534 0.727 0.706 0.493 K20 0.420 0.498 0.507 0.692 0.649 0.832 CaO 0.0854 0.065 0.227 0.247 0.106 0.031 Sc 27.1 25.9 26.1 24.7 26.5 23.5 Ti02 2.86 3.90 2.29 3.05 2.26 2.40 v 267 257 166 196 186 189 Cr 399.1 123.1 90.4 141.9 112.0 204.5 MnO 0.392 0.372 0.355 0.630 0.421 0.296 Fe203 27.2 33.8 29.6 28.3 27.1 26.7 Co 36.9 60.2 45.7 47.7 42.6 39.9 Ni 188.3 163.2 260.0 212.6 179.5 216.5 Cu 78.9 71.5 91.2 73.6 85.7 100.2 Zn 128 182 222 224 236 229 145cm 148 Na20 0.0530 0.109 0.064 0.857 1.45 MgO 0.903 0.98 0.87 0.456 0.796 Ah03 56.9 56.4 50.3 65.5 57.4 Si02 18.3 18.9 16.7 19.8 25.7 P20s 0.627 0.598 0.700 0.883 0.765 K20 0.609 0.729 0.617 0.875 1.36 CaO 0.021 0.049 0.083 0.475 0.784 Sc 15.8 16.1 19.2 3.33 3.50 Ti02 1.04 1.18 1.44 0.766 0.597 v 146 109 132 40.8 21.8 Cr 113.8 99.5 113.8 19.7 6.3 MnO 0.284 0.373 0.338 0.324 0.592 Fe203 27.7 26.7 36.1 14.7 14.1 Co 29.9 30.3 34.3 12.1 21.2 Ni 150.6 167.7 136.9 8.3 3.7 Cu 72.6 76.5 64.9 26.2 13.5 Zn 136 147 168 400 88.7 184 Table 2. Mweka minor elements (LR-ICP-MS) proto lith 2cm 6 11 Be 14.8 10.6 12.2 12.8 5.54 5.95 5.35 5.21 5.19 4.79 B 17.8 17.4 16.7 14.5 13.6 15.6 16.2 14.3 11.8 10.9 Rb 230 281 144 448 101 93.8 97.8 95.4 71.7 76.6 Sr 965 844 668 915 132 92.2 70.0 59.3 50.3 67.0 y 76.4 63.8 54.8 65.2 22.9 26.9 25.4 24.7 30.1 27.8 Zr 1029 955 944 1009 1642 1727 1260 1097 1355 1438 Nb 279 282 278 262 587 643 434 386 494 463 Cs 1.67 1.61 1.18 3.59 8.47 8.21 9.69 7.29 6.79 5.70 Ba 1410 1455 1063 1308 365 359 301 238 294 292 La 155 128 123 132 46.1 56.4 49.9 40.4 51.3 Ce 264 197 180 209 191 307 157 159 181 206 Pr 32.8 23.1 24.7 26.0 10.3 14.7 9.88 11.1 11.8 11.4 Nd 127 88.8 98.1 101 42.3 55.7 41.5 48.4 47.9 47.0 Sm 27.6 20.9 20.8 22.0 11.2 13.2 11.4 13.3 12.5 11.3 Eu 8.89 7.22 6.24 6.66 3.32 4.02 4.23 5.03 4.07 3.86 Tb 9.86 7.32 6.65 6.64 3.10 3.55 5.19 5.95 4.54 4.40 Gd 77.0 58.7 50.2 52.1 23.4 27.3 43.0 46.7 34.8 34.4 Dy 45.4 33.8 31.9 30.6 14.5 16.4 24.3 24.4 21.2 21.0 Ho 3.01 2.53 2.47 2.51 1.21 1.38 1.30 1.45 1.42 1.42 Yb 9.28 7.33 7.18 7.73 4.02 4.12 5.46 5.50 4.90 4.96 Lu 1.00 0.801 0.860 0.892 0.407 0.488 0.527 0.494 0.499 0.486 Hf 19.8 19.3 20.0 22.0 27.2 28.9 24.7 20.2 22.9 22.7 Ta 12.9 12.8 13.3 13.4 22.9 25.2 16.8 17.2 20.6 17.5 Pb 5.70 6.63 5.20 4.86 154 142 172 146 98.6 100 Th 15.1 15.4 14.9 19.0 41.5 44.4 33.4 33.9 40.3 36.2 u 3.03 3.27 3.11 4.53 8.03 9.64 6.53 6.78 9.77 7.05 185 Table 2. continued 15cm 19 24 28 Be 4.46 5.15 3.21 4.26 4.60 4.18 3.93 4.28 4.19 B 12.7 11.2 8.49 9.25 16.0 18.6 18.3 20.3 22.9 Rb 61.2 71.8 61.5 69.4 67.2 64.1 67.1 77.8 90.4 Sr 55.9 111 63.6 187 42.5 56.3 47.9 49.7 58.8 y 22.7 26.5 17.7 18.9 21.9 20.5 22.2 17.7 27.4 Zr 1039 1241 916 1001 1003 1183 1087 1014 1054 Nb 366 399 297 343 391 441 416 378 391 Cs 6.41 7.53 5.43 5.98 6.49 6.99 6.34 7.62 7.84 Ba 265 294 193 348 233 262 252 218 220 La 41.9 50.3 35.6 32.7 41.7 37.0 28.3 40.7 Ce 129 172 117 136 150 133 130 105 104 Pr 8.93 10.9 8.05 7.46 8.36 8.05 8.17 7.10 9.71 Nd 38.6 41.6 32.4 31.4 33.7 32.9 31.3 30.9 42.2 Sm 10.3 11.1 8.18 8.48 8.81 8.48 8.07 8.32 10.2 Eu 3.92 4.40 3.05 3.44 3.39 2.70 2.90 3.44 3.80 Tb 4.84 4.53 3.62 3.12 3.73 3.28 3.09 3.72 4.98 Gd 40.8 37.9 31.4 24.8 28.3 25.9 24.6 30.0 37.7 Dy 22.1 21.0 17.8 13.9 14.7 14.3 14.1 15.8 20.0 Ho 1.21 1.29 0.95 0.86 0.97 0.99 1.15 1.00 1.27 Yb 4.85 5.45 3.32 3.32 3.74 3.52 3.88 3.87 4.59 Lu 0.468 0.521 0.425 0.299 0.368 0.346 0.392 0.370 0.441 Hf 20.7 22.5 18.8 13.3 18.6 19.6 20.3 20.3 22.5 Ta 14.4 15.8 12.0 10.6 13.7 15.4 14.9 14.9 15.4 Pb 99.0 116.9 120 77.5 104 133 114 151 149 Th 29.2 30.3 24.1 21.2 23.9 30.7 29.8 26.8 28.3 u 5.63 6.34 3.91 5.73 4.95 5.79 5.82 5.14 6.20 186 Table 2. continued 45cm 55 64 83 Be 4.65 4.56 2.88 4.00 3.61 3.51 4.46 3.94 2.75 2.61 B 18.3 20.2 20.8 20.3 14.8 16.8 18.0 16.0 7.88 6.67 Rb 65.4 95.0 68.1 77.0 50.8 47.6 68.5 62.7 32.2 35.0 Sr 42.7 47.1 48.0 27.9 49.0 66.5 65.2 43.6 20.0 19.9 y 19.2 24.1 12.8 12.5 15.6 22.8 12.0 11.3 9.45 8.49 Zr 1135 1416 989 1184 1568 1343 1031 992 861 885 Nb 425 547 253 292 389 347 235 188 220 236 Cs 7.17 8.33 7.73 9.24 5.54 4.83 4.83 4.82 3.12 2.76 Ba 194 249 159 191 177 215 190 196 134 129 La 31.5 42.9 25.7 23.8 24.6 27.4 19.8 21.7 17.6 16.6 Ce 100 152 106 119 151 130 128 126 103 126 Pr 7.57 9.78 7.46 5.25 8.51 7.02 6.15 6.11 4.92 4.28 Nd 32.7 40.0 30.5 21.0 32.1 29.3 28.1 27.5 21.6 18.2 Sm 8.11 10.2 6.81 5.62 9.59 7.55 10.5 9.76 7.70 6.88 Eu 2.73 3.32 3.00 2.53 3.02 2.47 3.73 3.64 2.82 2.50 Tb 3.47 4.13 4.01 2.90 2.95 3.08 3.30 3.66 2.32 2.11 Gd 31.7 32.9 33.6 24.8 23.9 19.3 26.7 28.3 18.9 Dy 16.4 17.8 16.2 13.9 13.4 10.6 14.4 15.4 11.4 9.48 Ho 0.99 1.23 0.81 0.86 0.96 0.91 0.92 1.01 0.827 0.737 Yb 3.78 4.15 3.23 3.28 4.33 3.32 6.09 5.60 4.30 3.88 Lu 0.342 0.398 0.277 0.297 0.377 0.343 0.555 0.522 0.381 0.366 Hf 19.0 24.4 20.0 23.6 25.8 23.1 24.7 23.9 17.4 17.9 Ta 14.6 19.7 12.4 13.0 15.5 13.5 11.3 9.89 7.95 7.86 Pb 145 166 155 240 138 101 109 112 71.5 71.3 Th 28.6 36.3 24.4 29.4 33.8 32.1 17.7 18.0 15.5 15.6 u 5.94 7.78 4.66 6.49 9.60 8.02 5.28 5.75 4.21 4.62 187 Table 2. continued 93cm 102 122 131 Be 2.82 2.60 2.25 1.80 1.92 2.82 3.40 2.52 2.88 B 18.0 9.25 6.77 4.06 6.06 13.7 12.2 10.1 8.42 Rb 35.6 41.1 28.8 21.0 34.2 53.7 54.9 68.8 74.6 Sr 30.5 144.7 28.7 20.3 28.1 52.9 63.5 69.7 59.6 y 7.86 6.75 6.32 2.89 3.43 5.35 6.30 7.43 5.09 Zr 503 462 401 321 372 543 746 629 603 Nb 158 167 144 141 156 197 308 246 293 Cs 3.13 3.25 3.03 1.97 2.56 5.52 5.98 5.40 6.31 Ba 289 348 110 95 105 226 321 175 185 La 11.4 30.5 18.7 6.3 9.3 10.5 21.4 15.2 14.7 Ce 88 111 92 85 98 105 172 124 183 Pr 3.36 6.47 4.96 1.86 2.45 2.69 5.92 4.45 4.12 Nd 13.8 29.4 21.1 7.8 9.6 10.9 26.4 31.4 16.5 Sm 5.18 8.20 6.87 2.45 2.91 3.35 6.48 6.77 4.29 Eu 3.73 4.48 2.76 1.10 1.25 2.45 3.68 3.66 2.26 Tb 2.12 4.69 3.37 0.87 1.01 1.72 2.98 3.27 1.77 Gd 16.0 40.9 26.8 6.3 7.9 12.7 24.6 24.5 15.0 Dy 8.87 20.3 13.8 3.61 4.69 6.74 13.6 13.6 8.22 Ho 0.522 0.535 0.563 0.239 0.276 0.387 0.498 0.589 0.392 Yb 3.63 3.14 3.10 1.26 1.38 2.06 2.59 3.08 1.76 Lu 0.323 0.298 0.270 0.126 0.135 0.185 0.254 0.294 0.170 Hf 10.6 9.44 8.26 5.86 6.32 12.5 13.5 14.1 13.0 Ta 6.49 7.43 5.60 4.44 4.77 9.11 10.6 10.8 10.4 Pb 66.1 82.0 61.1 44.5 44.2 122 141 156 121 Th 7.5 10.4 8.9 4.8 5.09 10.8 14.4 12.1 13.1 u 3.63 3.48 3.26 2.78 2.65 2.33 3.14 2.84 2.69 188 Table 2. continued 140cm 145 148 Be 3.27 2.87 2.85 1.96 25.0 18.7 25.4 B 15.2 16.2 11.3 9.38 28.6 16.8 24.1 Rb 82.3 88.3 45.4 46.1 30.4 24.1 38.2 Sr 47.8 46.2 48.4 17.9 153 118 277 y 5.49 5.14 17.7 4.49 8.53 6.84 8.16 Zr 782 826 681 509 618 589 668 Nb 330 300 211 211 175 211 214 Cs 6.94 7.68 6.22 5.76 0.183 0.299 0.34 Ba 212 174 269 144 150 167 298 La 10.1 12.7 10.3 8.44 29.5 36.4 Ce 186 182 112 109 142 130 275 Pr 3.21 3.17 2.61 1.97 5.05 7.07 8.85 Nd 13.7 13.3 12.4 9.8 16.0 22.9 25.1 Sm 3.93 4.02 3.93 2.51 5.68 6.67 7.40 Eu 2.43 2.18 2.85 1.70 4.92 3.73 6.02 Tb 1.72 1.97 2.21 1.34 3.93 3.41 3.84 Gd 13.8 15.0 15.8 10.5 31.8 27.7 29.4 Dy 7.86 8.36 9.85 6.37 17.2 17.5 16.9 Ho 0.473 0.504 0.749 0.394 1.10 1.07 1.19 Yb 2.48 2.40 3.67 2.30 10.7 10.0 11.5 Lu 0.233 0.225 0.348 0.191 0.759 0.848 1.01 Hf 18.5 14.2 14.5 13.6 19.4 23.1 25.9 Ta 13.2 11.6 10.7 9.29 15.2 15.5 16.2 Pb 148 151 108 136 34.8 34.8 30.9 Th 15.8 18.9 13.3 12.8 11.2 10.7 10.9 u 2.83 3.95 2.90 2.61 2.52 2.78 2.97 189 Table 3. Olmani major elements (MR-ICP-MS) protolith XRF Scm 91 Na20 3.37 6.58 7.07 6.05 0.913 0.828 1.38 1.09 MgO 17.8 9.8 14.0 17.0 1.19 0.893 3.39 4.53 Ah03 8.26 11.8 14.6 13.0 34.7 34.8 39.2 32.8 Si02 41.4 31.7 39.0 37.8 52.1 48.5 43.8 47.9 P20s 0.610 0.615 0.824 0.726 0.957 2.423 0.546 0.698 K20 1.79 3.38 4.30 3.55 1.44 1.44 1.62 1.21 CaO 11.6 18.2 18.6 16.1 2.59 3.82 2.40 5.00 Sc 29.2 17.2 19.4 19.7 13.9 12.5 9.13 12.0 Ti02 2.53 2.53 2.53 2.53 2.45 3.03 2.33 2.36 v 230 214 225 205 165 189 144 151 Cr 825 368 386 491 85 357 155 126 MnO 0.209 0.264 0.340 0.348 0.384 0.381 0.451 0.410 Fe203 13.8 16.7 20.6 18.6 13.1 14.0 14.3 14.0 Co 69.8 93.7 97.8 39.1 39.0 45.0 47.1 Ni 630 143 335 417 112 85 200 183 Cu 139 325 427 338 100 108 118 119 Zn 96.9 111 152 121 144 192 157 176 107cm 122 183 Na20 0.935 0.783 0.727 0.697 1.18 0.79 0.645 0.421 0.704 MgO 8.37 3.22 4.31 32.6 22.0 23.0 20.3 30.9 17.6 AhOJ 35.6 30.6 35.7 12.8 16.7 14.7 14.1 9.9 13.7 Si02 29.6 44.5 34.2 24.4 27.3 26.7 24.1 25.7 29.8 P20s 0.294 0.439 0.339 0.237 0.253 0.243 0.431 0.371 0.469 K20 1.26 1.05 0.968 0.328 0.376 0.164 0.147 0.175 0.094 CaO 2.49 1.77 2.94 4.11 5.95 6.91 15.3 9.22 13.4 Sc 10.3 7.25 12.9 14.0 18.3 19.7 22.3 13.6 20.8 Ti02 1.47 1.25 1.69 1.41 2.34 2.24 1.55 1.25 2.33 v 131 100 130 123 162 178 163 120 189 Cr 817 398 193 1074 615 1318 1231 585 878 MnO 0.488 0.362 0.438 0.326 0.326 0.314 0.377 0.298 0.329 Fe203 18.5 14.3 17.6 23.2 23.9 25.5 23.5 21.9 21.9 Co 81.8 58.1 67.8 163 138 139 119 162 102 Ni 419 286 221 1685 1233 1068 663 1629 611 Cu 124 105 145 126 170 142 336 264 308 Zn 168 171 206 103 89 99 109 91.2 93.5 190 Table 4. Olmani minor elements (LR-ICP-MS) proto lith Scm 91 107 Li 14.2 17.0 34.5 40.8 31.7 27.4 19.7 31.7 Be 3.02 3.56 7.89 10.40 7.96 7.87 4.62 7.14 B 13.3 12.6 5.78 7.30 5.24 5.55 8.01 8.93 Rb 100 133 55.7 68.4 44.5 40.0 27.9 40.7 Sr 1172 1275 1052 1205 1238 1023 810 871 y 20.7 24.0 36.6 56.9 38.5 41.1 22.0 28.1 Zr 203 236 480 744 510 515 259 344 Nb 96.3 125 242 327 219 251 113 165 Cs 0.871 1.10 2.40 2.20 1.79 1.70 1.33 2.15 Ba 775 1007 1601 1749 1547 1328 1022 1130 La 65.3 82.6 137 205 127 157 59.9 88.6 Ce 110 232 356 226 287 93.3 132 Pr 13.3 16.6 22.6 34.1 22.9 27.1 9.73 13.9 Nd 58.0 69.3 82.4 121.7 84.2 96.7 38.5 51.0 Sm 12.6 14.1 14.2 20.5 15.7 17.5 8.75 10.8 Eu 8.66 9.28 8.69 10.0 10.7 9.02 10.5 10.7 Tb 4.55 4.93 4.10 6.09 5.11 5.29 4.86 5.62 Gd 32.0 33.3 29.1 43.0 36.5 36.3 35.0 39.8 Dy 23.1 24.6 21.4 33.7 27.3 27.6 22.4 25.8 Ho 1.00 1.13 1.42 2.29 1.63 1.75 0.902 1.15 Er 2.34 2.69 3.72 5.79 4.20 4.46 2.28 2.95 Yb 4.00 4.22 4.54 7.42 5.85 5.43 4.18 5.10 Lu 0.400 0.449 0.561 0.868 0.660 0.648 0.418 0.513 Hf 6.11 5.93 9.66 15.3 10.7 10.9 5.24 6.89 Ta 5.54 6.65 8.65 12.5 9.77 12.6 4.81 5.48 Pb 14.9 15.3 38.0 74.0 52.1 70.5 46.2 43.7 Th 8.17 9.89 19.8 32.3 19.5 21.5 9.60 12.8 u 1.49 1.67 4.53 7.31 4.26 4.98 1.78 2.79 191 Table 4. continued 122 em 183 Li 8.8 12.3 7.4 5.4 15.0 Be 2.50 3.08 1.91 1.72 3.15 B 6.28 7.91 7.38 4.89 13.7 Rb 8.6 13.4 2.2 1.5 3.3 Sr 322 366 346 333 602 y 12.8 16.2 10.7 8.9 19.0 Zr 131 167 97.3 83.2 165 Nb 70.2 75.4 54.0 48.0 97.2 Cs 0.195 0.334 0.0656 0.0491 0.098 Ba 286 330 172 133 282 La 34.8 49.5 32.6 28.7 59.2 Ce 44.8 57.4 51.4 95.5 Pr 6.87 8.71 6.07 5.48 10.9 Nd 28.5 35.4 24.9 21.4 43.6 Sm 6.95 8.52 5.88 4.81 9.55 Eu 4.82 5.49 3.52 2.66 5.08 Tb 3.84 4.72 3.31 2.71 4.83 Gd 27.8 35.2 24.7 19.8 35.8 Dy 17.8 21.6 15.1 12.1 22.5 Ho 0.621 0.771 0.508 0.393 0.821 Er 1.52 1.85 1.18 0.96 1.88 Yb 2.94 3.31 2.29 1.73 3.30 Lu 0.295 0.328 0.218 0.161 0.322 Hf 3.34 4.28 2.60 1.99 3.68 Ta 3.32 3.75 2.80 2.26 4.67 Pb 10.0 13.8 6.64 5.74 9.90 Th 4.75 5.97 3.98 3.35 7.26 u 1.2 1.4 0.578 0.520 1.06 192 Table 5. Monduli major elements (LR-ICP-MS) protolith XRF 3cm 8 13 18 Na20 2.82 5.54 5.59 1.73 1.63 2.19 2.84 6.19 1.32 1.68 1.97 MgO 15.7 19.9 22.1 2.18 3.72 2.02 1.94 2.36 1.35 2.27 1.81 Ah03 10.1 19.4 22.7 34.6 24.6 42.4 44.9 34.4 46.3 45.5 45.1 Si02 44.5 42.7 46.8 46.3 53.1 34.7 36.7 48.8 43.5 38.7 45.2 P20s 0.648 0.496 0.540 0.657 0.684 0.276 0.256 0.627 0.280 0.263 0.243 K20 1.34 2.54 2.57 2.82 2.98 4.65 3.86 3.80 2.61 3.23 2.53 CaO 10.5 19.2 19.7 4.96 6.51 3.68 3.93 5.74 3.21 3.59 2.82 Sc 23.4 25.3 27.5 8.91 12.8 4.25 6.12 6.75 10.7 6.78 5.96 Ti02 2.47 2.47 2.47 2.69 2.53 1.25 1.52 1.95 1.97 1.53 1.74 v 236 252 235 202 215 136 155 156 183 132 130 Cr 898 1116 857 144 104 40.4 36.0 19.1 34.6 47.7 27.4 MnO 0.2 0.264 0.250 0.508 0.487 0.585 0.456 0.266 0.505 0.266 0.285 Fe203 13 19.2 18.7 23.1 23.4 27.6 22.9 14.8 18.3 22.3 17.6 Co 100 106 59.1 77.3 79.6 74.4 32.5 94.3 36.7 32.7 Ni 580 616 730 90.9 73.1 59.8 55.0 30.3 64.7 63.9 40.8 Cu 115 275 273 106 108 88.4 103 52.1 89.2 91.5 74.1 Zn 102 97.8 85.7 161 181 124 113 178 168 138 119 23cm 28 33 46 Na20 2.00 1.24 1.10 0.88 1.33 2.74 1.92 2.00 2.26 MgO 4.45 4.29 1.66 2.80 1.73 1.95 1.59 1.75 1.16 AhOJ 37.4 34.8 34.2 21.4 30.5 42.0 36.2 33.8 33.6 Si02 45.1 50.8 36.9 27.3 27.7 40.2 48.0 50.8 53.5 P20s 0.258 0.285 0.226 0.187 0.239 0.197 0.395 0.462 0.385 K20 2.21 1.60 2.37 1.25 3.36 4.92 3.51 3.44 3.72 CaO 7.38 8.17 2.55 4.68 2.98 2.86 3.43 3.53 3.18 Sc 10.1 11.4 5.21 6.18 4.57 5.40 8.38 7.29 7.62 Ti02 2.30 2.01 1.11 0.875 0.974 1.50 2.26 2.60 2.06 v 174 139 302 635 234 94 145 124 118 Cr 35.1 102 24.2 26.8 22.3 27.0 78.2 37.4 23.0 MnO 0.278 0.311 0.257 0.327 12.668 0.277 0.749 0.184 0.168 Fe203 17.9 15.9 31.0 50.9 28.8 15.0 13.4 12.7 11.2 Co 45.5 34.6 31.2 30.7 382 24.7 38.7 20.3 24.3 Ni 42.8 35.6 30.6 23.3 63.5 27.9 50.5 32.9 34.1 Cu 63.1 71.0 61.9 102 71.7 63.1 69.4 54.6 61.5 Zn 129 144 104 126 124 133 167 146 155 193 Table 5. continued 56 em 66 79 84 Na20 1.74 1.43 1.59 1.82 1.70 1.44 1.66 1.35 1.38 MgO 1.77 1.66 1.70 2.86 2.11 2.09 2.01 1.91 1.75 Ah03 39.1 33.7 42.1 45.6 47.8 49.0 50.3 50.7 49.2 Si02 47.1 47.5 44.2 35.4 36.7 37.7 38.5 42.9 43.5 P20s 0.168 0.135 0.376 0.554 0.137 0.089 0.195 0.0669 0.114 K20 3.09 2.55 3.01 2.69 2.67 2.80 2.87 1.69 2.05 CaO 3.08 9.79 3.88 5.63 3.27 4.38 4.16 1.70 2.45 Sc 7.89 6.49 7.16 6.99 6.11 8.02 7.49 5.95 7.23 Ti02 2.05 1.96 2.16 1.74 2.00 1.72 1.74 1.15 1.35 v 136 104 126 98.3 153 77.5 74.1 67.0 70.3 Cr 22.0 18.5 22.7 41.4 19.1 60.4 23.4 22.4 24.4 MnO 0.330 0.434 0.267 0.282 0.698 0.156 0.0986 0.0450 0.123 Fe203 13.1 11.5 12.4 15.2 14.9 12.6 10.5 10.7 10.2 Co 28.5 33.0 27.2 22.1 36.4 9.85 7.39 4.73 11.5 Ni 41.4 27.6 29.3 32.0 32.5 74.9 29.8 35.5 31.8 Cu 67.4 43.9 57.1 58.7 62.1 58.6 49.1 45.0 44.0 Zn 155 131 169 126 136 156 160 138 188 89cm 94 104 132 Na20 1.30 1.66 1.39 1.35 1.79 1.48 1.66 1.37 1.55 1.41 MgO 1.89 2.01 1.70 1.66 2.25 1.65 1.53 1.71 1.43 1.49 Ah03 53.4 53.4 43.1 45.7 42.5 46.3 41.6 42.8 43.3 43.0 Si02 37.8 36.1 47.8 46.2 45.8 38.0 40.4 40.5 42.0 41.7 P20s 0.0640 0.0626 0.0814 0.0926 0.0840 0.115 0.162 0.133 0.0730 0.123 K20 2.02 2.01 1.93 2.08 2.05 2.35 3.42 1.87 2.01 2.05 CaO 1.65 5.63 2.34 1.95 3.04 2.38 2.57 2.33 2.04 2.23 Sc 4.03 5.93 6.26 6.81 7.76 7.44 6.24 7.58 5.83 5.91 Ti02 0.85 1.63 1.41 1.51 1.56 2.14 2.00 1.89 1.60 1.57 v 67.5 88.5 107 103 117 176 116 113 109 106 Cr 25.1 17.4 21.0 24.2 22.7 34.3 25.3 54.9 23.3 23.5 MnO 0.990 0.237 0.314 0.429 0.172 0.185 0.202 0.137 0.436 0.258 Fe203 12.4 9.36 11.6 11.0 12.4 13.5 11.8 12.6 10.9 11.4 Co 45.2 15.6 18.2 16.0 12.8 19.7 19.7 14.9 29.3 16.3 Ni 36.9 36.2 39.0 34.1 31.9 37.2 33.7 35.8 35.1 32.2 Cu 47.4 53.5 45.8 44.6 43.7 49.9 45.5 50.3 37.2 40.8 Zn 124 143 174 161 137 109 112 105 82 107 194 Table 5. continued 1" 7' 11' Na20 1.05 1.27 2.83 3.26 0.527 0.646 MgO 0.87 1.01 0.700 0.513 1.15 0.976 Ah03 36.0 36.3 47.5 40.1 42.9 38.9 Si02 37.3 39.9 30.2 37.0 28.3 33.8 P20s 0.726 0.648 0.169 0.161 0.504 0.519 K20 3.12 4.14 4.26 7.11 2.93 3.22 CaO 1.89 1.93 1.08 0.961 0.772 0.712 Sc 11.9 11.1 6.23 5.07 10.4 9.67 Ti02 5.10 4.42 2.31 2.12 4.61 4.04 v 295 242 181 148 315 263 Cr 53.0 65.6 49.2 45.1 45.3 48.5 MnO 0.376 0.355 0.353 0.304 1.41 1.01 Fe203 19.6 15.7 17.2 14.3 23.9 22.5 Co 33.0 35.5 32.3 27.9 97.0 70.8 Ni 57.1 57.5 37.4 28.6 65.1 59.1 Cu 67.6 68.2 48.6 33.2 72.9 61.9 Zn 148 141 107 81.2 138 143 195 Table 6. Monduli minor elements (LR-ICP-MS) Proto lith 3cm 8 13 Li 11.0 7.96 6.12 16.5 21.4 38.2 38.8 33.4 25.2 27.1 Be 2.72 2.42 2.42 3.95 5.24 6.82 6.32 6.28 6.22 5.81 B 70.9 48.9 38.3 26.9 30.8 84.4 66.2 51.1 27.7 24.8 Rb 70.2 63.3 52.3 105 113 168 203 157 169 128 Sr 1547 1695 2200 1464 1215 1461 2673 1194 1111 1141 y 27.8 27.3 29.1 33.1 50.9 27.0 28.0 26.0 37.2 31.5 Zr 240 237 260 252 482 243 265 268 385 361 Nb 69.2 72.6 81.4 96.7 197 101 93.2 89.2 156 142 Cs 0.617 0.513 0.436 1.15 1.61 2.37 2.53 2.20 2.80 1.51 Ba 1037 1095 1283 1476 1940 1322 2485 1328 1443 1562 La 47.3 49.7 51.3 59.8 92.6 35.4 51.6 48.3 98.2 68.3 Ce 94.6 90.1 93.8 78.8 142 36.8 55.3 49.8 229 174 Pr 11.4 11.4 12.1 10.0 17.7 8.81 9.09 8.08 20.7 11.7 Nd 57.5 57.7 60.6 49.6 91.4 47.0 42.4 39.9 101 55.7 Sm 13.0 12.9 13.0 11.7 19.2 12.0 10.7 10.4 20.0 11.5 Eu 5.72 5.51 5.61 8.00 7.99 18.6 21.3 12.3 7.94 5.50 Tb 3.19 2.76 2.67 4.01 3.90 8.19 7.33 6.56 4.92 3.06 Gd 27.1 23.6 22.0 36.0 30.9 67.6 64.5 58.3 43.1 27.6 Dy 14.3 13.0 12.0 18.7 18.9 31.5 30.9 27.5 24.5 15.4 Ho 1.19 1.12 1.11 1.24 2.01 1.19 1.20 1.20 1.71 1.28 Er 3.02 2.83 2.78 3.31 5.19 3.28 3.25 3.38 4.83 3.49 Yb 3.71 3.28 3.02 5.21 6.00 7.74 6.64 6.70 6.51 4.53 Lu 0.399 0.373 0.355 0.540 0.720 0.761 0.606 0.608 0.772 0.536 Hf 7.66 7.15 6.95 5.76 11.3 7.57 6.28 6.40 12.5 9.03 Ta 5.14 4.91 5.14 4.40 10.3 5.29 4.79 4.81 8.55 6.58 Ph 9.71 8.24 7.09 32.6 53.8 58.3 56.1 41.1 52.6 29.5 Th 7.02 6.70 7.07 7.76 15.2 6.04 7.33 7.53 14.3 12.8 u 1.69 1.62 1.54 1.84 4.01 1.12 1.39 1.43 2.86 2.34 196 Table 6. continued 18cm 23 28 33 46 Li 37.7 32.9 29.9 19.1 26.1 27.2 37.6 39.8 57.9 43.0 Be 6.40 6.15 6.34 4.69 7.22 5.74 5.09 6.84 8.01 11.9 9.81 B 33.5 31.1 36.0 20.8 30.0 22.0 16.1 18.7 23.1 18.2 20.1 Rb 156 138 141 120 182 110 107 136 156 228 183 Sr 1286 941 1470 1684 2850 817 595 1008 979 1154 1376 y 31.5 30.9 33.3 42.5 43.3 19.9 17.6 23.9 27.9 51.4 58.9 Zr 307 313 326 423 430 195 198 258 276 643 580 Nb 129 152 139 170 201 65.7 79.6 104 103 281 253 Cs 1.90 2.09 1.71 1.31 2.19 1.96 1.69 2.32 2.63 3.88 2.92 Ba 1344 1751 1353 1879 3687 1311 738 1334 1425 1882 2117 La 48.7 58.6 51.2 62.6 76.6 53.4 38.4 50.0 63.9 140 153 Ce 70.6 88.7 64.2 62.9 139 70.8 45.2 51.5 86.1 189 227 Pr 10.0 12.1 9.33 14.3 15.4 8.30 6.93 8.54 10.3 23.4 25.4 Nd 47.9 58.4 47.0 74.6 73.2 40.2 32.9 40.4 48.1 102 118 Sm 11.9 12.6 11.2 15.8 15.2 9.63 7.61 9.33 10.9 18.2 23.0 Eu 10.1 10.7 7.71 7.82 10.6 11.8 6.65 12.1 11.7 8.00 10.0 Tb 6.05 6.08 4.47 4.79 4.18 6.60 4.60 6.41 7.44 5.80 7.23 Gd 48.1 36.8 38.7 33.9 51.5 36.2 51.2 59.0 47.2 56.2 Dy 23.3 25.9 20.0 22.1 18.4 27.2 20.2 26.6 30.6 29.7 32.7 Ho 1.16 1.50 1.27 1.44 1.52 1.07 0.854 1.10 1.35 2.31 2.74 Er 3.18 4.03 3.32 4.05 4.04 2.61 2.12 2.91 3.32 6.02 6.97 Yb 6.16 6.29 5.23 5.86 5.52 4.79 3.98 5.76 5.89 7.83 8.44 Lu 0.615 0.660 0.528 0.603 0.620 0.428 0.389 0.547 0.536 0.872 0.920 Hf 7.62 8.78 6.70 10.3 9.84 5.31 5.12 7.05 6.93 17.3 14.0 Ta 5.46 7.02 5.10 10.5 7.91 3.48 3.56 4.78 4.79 13.0 10.9 Pb 37.1 46.2 39.6 55.5 42.1 48.0 35.9 52.9 49.9 61.8 82.4 Th 9.17 8.71 10.0 13.9 13.5 7.58 8.09 9.10 10.8 30.0 26.8 u 1.80 1.69 2.11 3.20 2.71 1.60 1.36 1.59 1.82 6.18 5.55 197 Table 6. continued 56 em 66 79 84 Li 57.2 52.0 83.1 75.6 72.3 122.2 101.9 64.0 55.9 63.7 Be 11.4 10.6 14.0 13.0 12.3 18.3 18.1 10.5 10.5 11.8 B 17.4 13.4 34.5 23.5 24.9 29.1 23.5 12.9 11.4 12.7 Rb 164 170 254 204 208 372 257 142 166 171 Sr 1207 1304 1590 1126 1328 1570 1566 760 862 986 y 52.7 43.1 53.7 41.5 47.1 50.3 64.3 35.8 41.4 43.9 Zr 570 592 569 509 515 618 628 415 414 479 Nb 228 254 199 204 211 210 244 156 167 180 Cs 3.49 3.18 5.01 4.14 3.99 7.61 5.37 3.00 3.00 3.43 Ba 1758 2363 2364 1645 2036 2236 2174 884 1160 1464 La 148 132 112 97.0 103 115 138 80.2 102 92.8 Ce 171 223 107 74.2 95.9 84.6 107 52.6 29.4 66.2 Pr 23.0 23.2 17.7 15.9 16.7 16.7 19.1 12.2 15.1 15.1 Nd 96.5 95.9 81.3 71.8 71.8 74.2 84.0 50.5 62.0 64.4 Sm 17.2 16.7 17.6 15.3 14.9 15.5 18.0 10.5 13.0 13.2 Eu 8.40 8.03 16.1 11.4 12.3 13.8 12.8 6.66 7.51 7.93 Tb 6.37 5.30 10.48 8.49 8.58 9.91 10.0 6.12 6.30 5.85 Gd 51.6 42.0 82.7 67.4 67.9 73.9 76.1 50.1 50.8 45.5 Dy 29.7 24.5 45.2 36.0 37.0 38.7 42.7 28.1 28.2 26.0 Ho 2.18 2.09 2.22 1.92 2.07 2.27 2.66 1.56 1.80 1.76 Er 5.51 5.17 5.37 5.00 5.15 6.13 6.59 3.96 4.65 4.48 Yb 7.15 6.71 9.20 7.90 8.09 10.02 10.54 6.56 6.97 6.54 Lu 0.822 0.778 0.859 0.760 0.783 0.989 1.08 0.630 0.722 0.720 Hf 14.5 14.8 12.5 11.4 11.2 15.7 13.5 9.31 10.0 11.4 Ta 9.69 10.5 8.14 8.37 7.55 8.08 7.48 5.09 6.26 6.47 Pb 62.8 63.5 82.5 80.1 67.4 114.4 74.7 49.4 53.2 59.4 Th 26.5 27.9 20.4 23.2 19.1 24.6 25.1 16.9 20.3 23.2 u 5.18 5.18 3.93 4.51 3.85 3.85 4.75 2.95 3.30 4.44 198 Table 6. continued 89cm 94 104 132 Li 83.8 75.7 49.8 66.4 69.7 51.4 67.7 64.3 49.0 Be 16.4 14.9 8.77 12.5 14.2 11.2 12.2 11.8 9.9 B 16.3 15.6 9.22 12.4 13.7 13.4 10.8 11.3 9.7 Rb 172 171 109 182 158 176 183 173 146 Sr 1354 890 642 1000 1092 1526 1176 1156 1005 y 38.5 37.8 37.6 57.4 60.7 57.0 53.6 45.3 39.5 Zr 417 420 406 508 612 653 700 519 535 Nb 138 157 168 220 214 251 293 197 192 Cs 4.68 4.29 2.46 3.36 3.93 3.34 3.57 3.30 4.50 Ba 2754 1388 789 1266 1364 1962 1955 3216 1668 La 118 80.1 60.9 128 127 144 155 105 101 Ce 125 65.9 50.6 118 134 166 131 161 145 Pr 14.2 13.4 9.88 18.7 23.6 23.8 25.7 19.3 17.4 Nd 60.9 60.3 46.3 76.5 109 110 113 87.1 77.2 Sm 14.2 13.4 10.1 16.5 20.3 21.5 21.2 17.6 15.4 Eu 21.5 10.5 5.52 9.60 8.63 8.66 8.19 10.6 6.21 Tb 10.7 7.74 4.98 8.02 7.10 5.78 4.98 5.93 3.95 Gd 83.1 60.3 44.7 59.1 57.9 45.8 39.9 50.9 33.1 Dy 41.8 34.0 25.1 31.9 34.6 25.2 22.7 27.4 18.5 Ho 2.17 1.82 1.50 2.56 2.38 2.27 2.15 1.97 1.65 Er 6.04 4.60 3.68 6.43 6.03 6.05 5.64 5.08 4.52 Yb 10.08 7.98 5.91 8.12 9.17 8.56 7.39 6.93 6.09 Lu 1.023 0.773 0.559 0.871 1.068 1.002 0.880 0.740 0.703 Hf 11.5 10.1 7.8 12.1 16.8 15.1 17.0 11.9 12.3 Ta 4.92 5.58 4.64 6.93 9.01 9.81 10.8 8.00 7.38 Pb 134.3 74.2 55.5 64.7 72.9 50.1 49.7 47.8 64.8 Th 17.8 17.3 17.3 18.2 31.5 26.1 32.0 20.3 23.2 u 3.03 3.06 3.45 3.65 5.61 6.50 6.44 4.28 4.82 199 Table 6. continued 1" 7' 11' Li 53.9 54.6 56.5 49.1 72.1 68.9 Be 12.8 12.7 10.8 11.4 19.0 18.0 B 10.6 10.5 10.3 9.55 11.7 9.56 Rb 257 248 276 229 560 448 Sr 1081 1122 834 711 990 745 y 65.8 62.5 27.7 28.8 43.9 41.4 Zr 1365 1245 816 830 1148 1239 Nb 581 564 307 301 517 554 Cs 4.96 4.83 6.34 5.06 11.3 11.0 Ba 2371 2279 1548 1494 2510 2387 La 269 216 90.1 93.3 450 388 Ce 460 528 209 277 320 365 Pr 43.0 38.1 16.2 17.9 51.6 46.2 Nd 181 156 75.4 77.2 209 188 Sm 30.8 27.3 14.4 14.2 34.5 30.4 Eu 9.04 8.23 6.00 5.70 11.6 9.55 Tb 4.28 3.69 3.81 3.43 8.62 5.96 Gd 34.1 29.2 31.0 28.5 71.4 48.4 Dy 20.8 18.6 16.2 15.8 36.4 26.9 Ho 2.61 2.60 1.25 1.36 2.48 2.29 Er 6.82 6.80 3.33 3.61 6.50 6.12 Yb 7.90 7.23 5.03 5.01 9.05 8.34 Lu 1.029 0.993 0.571 0.593 1.05 1.04 Hf 29.8 30.0 19.7 19.3 26.5 28.7 Ta 22.4 23.0 12.9 11.9 18.1 19.1 Pb 67.0 71.6 72.5 77.9 157 121 Th 51.0 51.8 29.4 31.4 39.8 42.2 u 10.3 10.6 6.34 7.45 11.5 11.3 200 BIBLIOGRAPHY 201 Aiuppa, A. et al., 2000. Mobility and fluxes of major, minor and trace metals during basalt weathering and groundwater transport at Mt. Etna volcano (Sicily). Geochimica et Cosmochimica Acta, 64(11): 1827-1841. Alloway, B.J., 1995. Heavy Metals in Soils. Heavy Metals in Soils. Blackie Academic and Professional, Suffolk, England, 368 pp. Anderson, G.D., 1982. A survey of the soils and land use potential of the southern and eastern slopes ofKilimanjaro, Tanzania. 1, Institute of Resource Assessment, University of DarEs Salaam and International Development Program, Clark Univeristy, DarEs Salaam. Andresen, A.M., Johnson, A.H. and Siccama, T.G., 1980. Levels oflead, copper, and zinc in the forest floor in the northeast United States. Journal of Environmental Quality, 9: 293-296. April, R., Newton, R. and Coles, L.T., 1986. Chemical weathering in the Adirondack watersheds: Past and present-day rates. Geological Society of America Bulletin, 97: 1232-1238. Asami, T ., Kubota, M. and Orikasa, K., 1995. Distribution of different fractions of cadmium, zinc, lead and copper in unpolluted and polluted soils. Water, Air, & Soil Pollution, 83: 187-194. Bagdasaryan, G.P., Gerasimovskiy, V.I., Polyakov, A.I., Gukasyan, R.K. and Vernadskiy, V.I., 1973. Age of volcanic rocks in the rift zones of East Africa. Geochemistry International, 10: 66-71. Bain, D.C., Mellor, A., Robertson-Rintoul, M.S.E. and Buckland, S.T., 1993. Variations in weathering processes and rates with time in a chronosequence of soils from Glen Feshie, Scotland. Geoderma, 57: 275-293. Basu, A.R., Jacobsen, S.B., Poreda, R.J., Dowling, C.B. and Aggarwal, P.K., 2001. Large Groundwater Strontium Flux to the Oceans from the Bengal Basin and the Marine Strontium Isotope Record. Science, 293: 1470-1473. Berner, R.A., 1991. Weathering, plants, and the long-term carbon cycle. Geochimica et Cosmochimica Acta, 58(8): 3225-3231. Berner, R.A., 1995. Chemical weathering and its effect on atmospheric C02 and climate. In: A.F. White and S.L. Brantley (Editors), Chemical Weathering Rates of Silicate Minerals. Reviews in Mineralogy. Mineralogical Society of America, pp. 565-583. Berner, R.A., Lasaga, A.C. and Garrels, R.M., 1983a. The carbonate-silicate geochemical cycle and its effects on atmospheric carbon dioxide and climate. American Journal of Science, 283: 641-683. 202 Berner, R.A., Lsaga, A.C. and Garrels, R.M., 1983b. The carbonate silicate geochemical cycle and its effect on atmospheric carbon dioxide over the past 100 million years. Am. J. Sci., 283: 641-683. Blaser, P., Zimmermanna, S., Lustera, J. and Shotyk, W., 2000. Critical examination of trace element enrichments and depletions in soils: As, Cr, Cu, Ni, Pb, and Zn in Swiss forest soils. The Science ofthe Total Environment, 249(1-3): 257-280. Blum, J.D. et al., 2002. Mycorrhizal weathering of apatite as an important calcium source in base-poor forest ecosystems. Nature, 417: 729-731. Braun, J.-J. et al., 2005. Present weathering rates in a humid tropical watershed: Nsimi, South Cameroon. Geochimica et Cosmochimica Acta, 69(2): 357-387. Brimhall, G.H. et al., 1991. Deformational Mass Transport and Invasive Processes in Soil Evolution. Science, 255: 695-702. Brimhall, G.H. and Dietrich, W.E., 1987a. Constitutive mass balance relations between chemical composition, volume, density, porosity, and strain in metasomatic hydrochemical systems: results on weathering and pedogenesis. Geochim. Cosmochim. Acta, 51: 567-587. Brimhall, G.H. and Dietrich, W.E., 1987b. Constitutive mass balance relations between chemical composition, volume, density, porosity, and strain in metasomatic hydrochemical systems; results on weathering and pedogenesis. Geochimica et Cosmochimica Acta, 51(3): 567-587. Brimhall, G.H. et al., 1988. Metal enrichment in bauxites by deposition of chemically mature aeolian dust. Nature, 333: 819-824. Byrne, R.H. and Kim, K.-H., 1990. Rare earth element scavenging in seawater. Geochimica et Cosmochimica Acta, 54: 2645-2656. Cabral, A.R. and Lefebvre, G., 1998. Use of sequential extraction in the study ofheavy metal retention by silty soils. Water, Air, & Soil Pollution, 102. Campos, M.L., Pierangeli, M.A.P., L.R.G.Guiherme, Marques, J.J. and Curi, N., 2003. Baseline concentrations of heavy metals in Brazilian Latosols. Communications in Soil Science and Plant Analysis, 34(3, 4): 547-557. Capo, R.C. and Chadwick, O.A., 1999. Sources of strontium and calcium in desert soil and calcrete. Earth and Planetary Science Letters, 170: 61-72. Carpenter, S. et al., 1998. Nonpoint Pollution of Surface Waters with Phosphorus and Nitrogen. Ecological Applications, 8(3): 559-568. 203 Chadwick, O.A., Brimhall, G.H. and Hendricks, D.M., 1990. From a black to a gray box - a mass balance interpretation of pedogenesis. Geomorphology, 3: 369-390. Chadwick, O.A., Derry, L.A., Vitousek, P.M., Huebert, B.J. and Hedin, L.O., 1999. Changing sources of nutrients during four million years of ecosystem development. Nature, 397: 491-497. Chen, C.-H., Liu, K.-K. and Shieh, Y.-N., 1988. Geochemical and isotopic studies of bauxitization in the Tatun volcanic area, northern Taiwan. Chemical Geology, 68: 41-56. Clark, S.P.J., 1957. Heat flow at Grass Valley, California. Amer. Geophys. Union Trans., 38(23). Colman, S.M., 1982. Chemical weathering ofbasalts and andesites: Evidence from weathering rinds. USGS Professional Paper, 1246. Cornelius, D.J., 2004. Washington State University at Pullman's GeoAnalytical Lab. personal communication. Cornu, S. et al., 1999. Evidence of titanium mobility in soil profiles, Manaus, central Amazonia. Geoderma, 91(3): 281-295. Cramer, J.J. and Nesbitt, H.W., 1983. Mass-balance relations and trace-element mobility during continental weathering of various igneous rocks. Sci. Geol. Mem., 73: 63- 73. Crowther, J., 1989. Groundwater chemistry and cation budgets of tropical karst outcrops, peninsular Malaysia, I. calcium and magnesium. Journal of Hydrology, 107: 169- 192. Cruz, J.V., Coutinho, R.M., Carvalho, M.R., Oskarsson, N. and Gislason, S.R., 1999a. Chemistry of waters from Furnas volcano, Sao Miguel, Azores: fluxes of volcanic carbon dioxide and leached material. Journal ofVolcanology and Geothermal Research, 92: 151-167. Cruz, J.V., Coutinho, R.M., Carvalho, M.R., Oskarsson, N. and Gislason, S.R., 1999b. Chemistry of waters from Furnas volcano, Sao Miguel, Azores: fluxes of volcanic carbon dioxide and leached material. Journal ofVolcanology and Geothermal Research, 92: 151-167. Dawson, J.B., 1992. Neogene tectonics and volcanicity in the Noth Tanzania sector of the Gregory Rift Valley: contrasts with the Kenya sector. Tectonophysics, 204: 81-92. 204 Dessert, C. et al., 2001. Erosion of Deccan Traps determined by river geochemistry: impact on the global climate and the 87Sr/86Sr ratio of seawater. Earth and Planetary Science Letters, 188: 459-474. Dia, A. et al., 2000. The distribution of rare earth elements in groundwaters: Assessing the role of source-rock composition, redox changes and colloidal particles. Geochimica et Cosmochimica Acta, 64(24): 4131-4151. Dietrich, W.E. et al., 2003. Geomorphic transport laws for predicting landscape form and dynamics. Geophys. Monograph, 135: 1-30. Dowling, C.B., Poreda, R.J. and Basu, A.R., 2003. The groundwater geochemistry ofthe Bengal Basin: Weathering, chemsorption, and trace metal flux to the ocean. Geochimica et Cosmochimica Acta, 67(12): 2117-2136. Durand, S., Chabaux, F., Rihs, S., Duringer, P. and Elsass, P., 2005. U isotope ratios as tracers of groundwater inputs into surface waters: Example of the Upper Rhine hydrosystem. Chemical Geology, 220: 1-19. Erel, Y ., Veron, A. and Halicz, L., 1997. Tracing the transport of anthropogenic lead in the atmosphere and in soils using isotopic ratios. Geochimica et Cosmochimica Acta, 61(21): 4495-4505. Evemden, J.F. and Curtis, G.H., 1965. The potassium-argon dating of Late Cenozoic rocks in Africa and Italy. Current Anthropology, 6: 343-384. Gabet, E.J. and Reichman, O.J., 2003. The effects of bioturbation on soil processes and sediment transport. Annual Review of Earth and Planetary Sciences, 31: 249-273. Gaillardet, J., Dupre, B., Louvat, P. and Allegre, C.J., 1999. Global silicate weathering and C02 consumption rates deduced from the chemistry of large rivers. Chemical Geology, 159: 3-30. Gibson, M.J. and Farmer, J.G., 1986. Multi-step Sequential Chemical Extraction of Heavy Metals from Urban Soils. Environmental Pollution Series B, 11: 117-135. Gouveia, M.A. et al., 1993. Behavior ofREE and other trace and major elements during weathering of granite rocks, Evora, Portugal. Chemical Geology, 107: 113-126. Grousset, F.E. and Biscaye, P.E., 2005. Tracing dust sources and transport patterns using Sr, Nd and Pb isotopes. Chemical Geology, 222: 149-167. Guo, L., Santschi, P.H. and Baskaran, M., 1997. Interaction of thorium isotopes with colloidal organic matter in oceanic environments. Colloids and Surfaces, A, 120: 255-271. 205 Halamic, J., Galovic, L. and Sparica, M., 2003. Heavy Metal (As, Cd, Cu, Hg, Ph and Zn) Distribution in Topsoil Developed on Alluvial Sediments of the Drava and Sava Rivers in NW Croatia. Geologia Croatica, 56(2): 215-232. Hamelin, B., Grousset, F.E., Biscaye, B.E. and Zindler, A., 1989. Pb isotopes in trade wind aerosols at Barbados: the influence of European Emissions over the North Atlantic. Journal of Geophysical Research, 94: 1624 3-1625 0. Han, F. and Banin, A., 1995a. Long-term transformation and redistribution of potentially toxic heavy metals in arid-zone soils incubated: I. under saturated conditions. Water, Air, & Soil Pollution. Han, F. and Banin, A., 1995b. Selective sequential dissolution techniques of potentially toxic heavy metals in arid-zone soils: the carbonate dissolution step. Communications in Soil Science and Plant Analysis, 26: 553-576. Hansmann, W. and Koppel, V., 2000. Lead-isotopes as tracers of pollutants in soils. Chemical Geology, 171: 123-144. Heimsath, A.M., Chappell, J., Spooner, N.A. and Questiaux, D.G., 2002. Creeping soil. Geology, 30: 111-114. Heimsath, A.M., Dietrich, W.E., Nishiizumi, K. and Finkel, R.C., 1997. The soil production function and landscape equilibrium. Nature, 388: 358-361. Henyey, T.L., 1968. Heat flow near major strike-slip faults in central and southern California, California Institute of Technology, Pasadena, CA. Henyey, T.L. and Lee, T.C., 1976. Heat flow in Lake Tahoe, California-Nevada, and the Sierra Nevada-basin and range transition. Geological Society of America Bulletin, 87(8): 1179-1187. Henyey, T.L. and Wasserburg, G.J., 1971. Heat flow near major strike-slip faults in California. Journal of Geophysical Research, 76(32): 7924-7946. Holtz, R.D. and Kovacs, W.D., 1981. An Introduction of Geotechnical Engineering. Prentice-Hall, Inc., Upper Saddle River, NJ, 733 pp. Ingebritsen, S.E., Sherrod, D.R. and Mariner, R.H., 1989. Heat Flow and Hydrothermal Circulation in the Cascade Range, North-Central Oregon. Science, 243(4897): 1458-1462. Jacobson, A.D., Blum, J.D. and Walter, L.M., 2002. Reconciling the elemental and Sr isotope composition of Himalayan weathering fluxes: insights from the carbonate geochemistry of stream waters. Geochim. Cosmochim. Acta, 66: 3417-3429. 206 Johnson, C.E. and Petras, R.J., 1998. Distribution of zinc and lead fractions within a forest spodosol. Soil Society of America Journal, 62(3): 782-789. Johnson, D.M., Hooper, P.R. and Conrey, R.M., 1999. XRF Analysis ofRocks and Minerals for Major and Trace Elements on a Single Low Dilution Li-tetraborate Fused Bead. Advances in X-ray Analysis, 41: 843-867. Kataba-Pendias, A. and Pendias, H., 2001. Trace Elements in Soils and Plants. CRC Press, Boca Raton, 331 pp. Keller, C. and Domergue, F.-L., 1996. Soluble and particulate transfers of Cu, Cd, Al, Fe and some other major elements in gravitational waters of a Podzol. Geoderma, 71: 263-274. Keller, C. and Vedy, J.-C., 1992. Distribution of copper and cadmium fractions in two forest soils. Journal of Environmental Quality, 23: 987-999. Kirkwood, D.E. and Nesbitt, H.W., 1991. Formation and evolution of soils from an acidified watershed: Plastic Lake, Ontario, Canada. Geochimica et Cosmochimica Acta, 55: 1295-1308. Klemme, S., Prowatke, S., Hametner, K. and Gunther, D., 2004. Partitioning of trace elements between rutile and silicate melts: Implications for subduction zones. Geochimica et Cosmochimica Acta, 69(9): 2361-2371. Kump, L.R., Brantley, S.L. and Arthur, M.A., 2000. Chemical weathering, atmospheric C02, and climate. Annual Review of Earth and Planetary Sciences, 28: 611-667. Kurtz, A. C., Derry, L.A. and Chadwick, O.A., 2001. Accretion of Asian dust to Hawaiin soils: Isotopic, elemental, and mineral mass balances. Geochimica et CosmochimicaActa, 65(12): 1971-1983. Kurtz, A.C., Derry, L.A., Chadwick, O.A. and Alfano, M.J., 2000. Refractory element mobility in volcanic soils. Gelogy, 28(8): 683-686. Lachenbruch, A.H. and Sass, J.H., 1980. Heat flow and energetics of the San Andreas fault system. Proceedings of Conference IX Magnitude of deviatoric stresses in the earth's crust and upper mantle. U.S. Geological Survey Open-File Report, 80- 625: 47-146. Lachenbruch, A.H., Sass, J.H., Munroe, R.J. and Moses, T.H.J., 1976a. Geothermal setting and simple heat conduction models for the Long Valley caldera. Journal of Geophysical Research, 81: 769-784. 207 Lachenbruch, A.H., Sorey, J.H., Munroe, R.J. and Moses, T.H.J., 1976b. The near surface hydrothermal regime ofLong Valley California. Journal of Geophysical Research, 81(5): 763-768. Langmuir, D., 1997. Aqueous Environmental Geochemistry, 1. Prentice Hall, Upper Saddle Rive, NJ, 589 pp. Lee, S.-Z., Chang, L., Yang, H.-H., Chen, C.-M. and Liu, M.-C., 1998. Adsorption characteristics oflead onto soils. Journal of Hazardous Materials, 63: 37-49. Little, M.G. and Lee, C.-T.A., 2006. On the formation of an inverted weathering profile on Mount Kilimanjaro, Tanzania: buried paleosol or groundwater weathering? Chemical Geology, 235: 205-221. Little, P.D., Smith, K., Cellarius, B.A., Coppock, D.L. and Barrett, C., 2001. Avoiding Disaster: Diversification and Risk Management among East African Herders. Development and Change, 32(3): 401-433. Manga, M., 1998. Advective heat transport by low-temperature discharge in the Oregon Cascades. Geology, 26: 799-802. Marin, B., Valladon, M., Polve, M. and Monaco, A., 1997. Reproducibility testing of a sequential extraction scheme for the determinaiton of trace metal speciation in a marine reference sediment by inductively coupled plasma-mass spectrrometry. Analytica Chimica Acta, 342: 91-112. Martin, J.M., Nirel, P. and Thomas, A.J., 1987. Sequential extraction techniques: Promises and problems. Marine Chemistry, 22: 313-341. Mase, C.W., Lachenbruch, A.H. and Munroe, R.J., 1982. Preliminary heat-flow investigation of the California Cascades. U.S. Geological Survey Open-File Report, 82-150: 240. Mase, C.W., Sass, J.H. and Lachenbruch, A.H., 1980. Near-surface hydrothermal regime ofthe lassen "known geothermal resource area," California. U.S. Geological Survey Open-File Report, 80-1230: 31. McBride, M.B. and Spiers, G., 2001. Trace element content of selected fertilizers and dairy manures as determined by ICP-MS. Communications in Soil Science and Plant Analysis, 32(1-2): 139-156. McCauley, S.E. and DePaolo, D.J., 1997. The marine 87Sr/86Sr and delta 180 records, Himalayan alkalinity fluxes, and Cenozoic climate models. In: W.F. Ruddiman (Editor), Tectonic uplift and climate change. Plenum Press, New York, pp. 427- 467. 208 Melfi, A.J., Subies, F., Nahon, D. and Formoso, M.L.L., 1996. Zirconium mobility in bauxites of Southern Brazil. Journal of South American Earth Sciences, 9(3/4): 161-170. Middelburg, J.J., Weijden, C.H.V.D. and Woittiez, J.R.W., 1988. Chemical processes affecting the mobility of major, minor and trace elements during weathering of granitic rocks. Chemical Geology, 68: 253-273. Millot, R., Gaillardet, J., Dupre, B. and Allegre, C.J., 2002. The global control of silicate weathering rates and the coupling with physical erosion: new insights from rivers ofthe Canadian shield. Earth and Planetary Science Letters, 196: 83--98. Mlingano, Agricultural, Research and Institute, 2006. Rainfed agriculture crop suitability for Tanzania, Mlingano Agricultural Research Institute, Tanga, Tanzania. Mortvedt, J.J., 1987. Cadmium levels in soils and plants from some long-term soil fertility experiments in the United States of America. Journal of Environmental Quality, 16: 137-142. Negrel, P. and Lachassagne, P., 2000. Geochemistry of the Maroni River (French Guiana) during the low water stage: implications for water-rock interaction and groundwater characteristics. Journal of Hydrology, 237: 212-233. Nesbitt, H.W. and Young, G.M., 1984. Prediction of some weathering trends of plutonic and volcanic rocks based on thermodynamic and kinetic considerations. Geochimica et Cosmochimica Acta, 48: 1523-1534. Nicola, F.D., Maisto, G. and Alfani, A., 2003. Assessment of nutritional status and trace element contamination of hold oak woodlands through analyses ofleaves and surrounding soils. The Science ofthe Total Environment, 311(1-3): 191-203. Nyblade, A., 1999. Heat flow and the structure ofthe Precambrian lithosphere. Lithos, 48: 81-91. Olson, K.W. and Skogerboe, R.K., 1975. Identification of soil lead compounds from automotive sources. Environ. Sci. Technol., 9(3): 227-230. Othman, I., Al-Oudat, M. and Al-Masri, M.S., 1997. Lead levels in roadside soils and vegetation of Damascus City. The Science of the Total Environment, 207: 43-48. Palumbo, B. et al., 2000. Influence of inheritance and pedogenesis on heavy metal distribution in soils of Sicily, Italy. Geoderma, 95: 247-266. Petit, J.R. et al., 1999. Climate and atmospheric history of the past 420,000 years from the Vostok ice core, Antarctica. Nature, 399: 429-436. 209 Peucker-Ehrenbrink, B. and Ravizza, G., 2000. The marine osmium isotope record. Terra Nova, 12: 205-219. Pollack, H.N., Hurter, S.J. and Johnson, J.R., 1991. A new global heat flow compilation, Departmnet of Geological Sciences, The university of Michigan, Ann Arbor, Michigan. Prevosto, B., Dambrine, E., Moares, C. and Curt, T., 2004. Effects of colvanic ash chemistry on the soils and vegetation of naturally regenerated woodlands in the Massif Central, France. Catena, 56: 239-261. Probst, A., Gh'mari, A.E., Aubert, D., Fritz, B. and McNutt, R., 2000. Strontium as a tracer of weathering processes in a silicate catchment polluted by acid atmospheric inputs, Strengbach, Franch. Chemical Geology, 170: 203-219. Rahman, S., Takaki, H., Tarnai, M. and Nagamoto, Y., 1996. Distribution of zinc, manganese, copper, cobalt and nickel in andosol profiles. Soil Science and Plant Nutrition, 42(4): 881-891. Raymo, M.E. and Ruddiman, W.F., 1992. Tectonic forcing oflate Cenozoic climate. Nature, 359: 117-122. Raymo, M.E., Ruddiman, W.F. and Froelich, P.N., 1988. Influence oflate Cenozoic mountain building on ocean geochemical cycles. Geology, 16: 649-653. Reneau, S.L., 1988. Depositional and erosional history of hollows; application to landslide location and frequency, long-term erosion rates, and the effects of climate change. PhD Thesis, University of California, Berkeley, California, 327 pp. Riebe, C.S., Kirchner, J.W. and Finkel, R.C., 2003. Long-term rates of chemical weathering and physical erosion from cosmogenic nuclides and geochemical mass balance. Geochimica et Cosmochimica Acta, 67(22): 4411-4427. Roering, J.J., Kirchner, J.W. and Dietrich, W.E., 2001. Hillslope evolution by nonlinear, slope-dependent transport: steady-state morphology and equilibrium adjustment timescales. J. Geophys. Res., 106: 16499-16513. Roy, R.F., Decker, E.R., Blackwell, D.D. and Birch, F., 1968. Heat flow in the United States. Journal of Geophysical Research, 73: 5207-5221. Rudnick, R.L. and Fountain, D.M., 1995. Nature and Composition of the Continental Crust: A Lower Crust Perspective. Reviews of Geophysics, 33(3): 267-309. 210 Sadler, P.M. and Morton, D.M. (Editors), 1989. Landslides in a semi-arid environment with emphasis on the inland valleys of southern California, 2. Inland Geological Society, 385 pp. Sak, P.B., Fisher, D.M., Gardner, T.W., Murphy, K. and Brantley, S.L., 2003. Rates of weathering rind formation on Costa Rican basalt. Geochimica et Cosmochimica Acta, 68(7): 1453-1472. Salam, A., Bauer, H., Kassin, K., Ullah, S.M. and Puxbaum, H., 2003. Aerosol chemical characteristics of a mega-city in Southeast Asia (Dhaka-Bangladesh). Atmospheric Environment, 37: 2517-2528. Sass, J.H., Galanis, S.P.J. and Munroe, R.J., 1982. Measurement ofheat flow by a downhole probe technique in the San Joaquin Valley, California. U.S. Geological Survey Open-File Report, 82-819: 27. Sass, J.H., Lachenbruch, A.H., Munroe, R.J., Greene, G.W. and Moses, T.H.J., 1971. Heat Flow in the western United States. Journal of Geophysical Research, 76: 6376-6413. Sastri, G.G.K. and Sastry, C.S., 1982. Chemical Characteristics and Evolution of the Laterite Profile in Hazaridadar Bauxite Plateau, Madhya Predesh, India. Economic Geology, 77: 154-161. Schot, P.P. and Wassen, M.J., 1993. Calcium concentration in wetland groundwater in relation to water sources and soil conditions in the recharge area. Journal of Hydrology, 141: 197-217. Scoones, I. and Toulmin, C., 1999. Soil nutrient budgets and balances: What use for policy? Service, N.R.C., 2004. Official Soil Series Descriptions. United States Department of Agriculture. Solomon, D., Lehmann, J. and Zech, W., 2000. Land use effects on soil organic matter properties of chromic luvisols in semi-arid northern Tanzania: carbon, nitrogen, lignin and carbohydrates. Agriculture, Ecosystems and Environment, 78: 203-213. Somerville, R. et al., 2007. Historical Overview of Climate Change. In: S. Solomon et al. (Editors), Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. Stiles, C.A., Mora, C.I., Driese, S.G. and Robinson, A.C., 2003. Distinguishing climate and time in the soil record: Mass-balance trends in Vertisols from the Texas coastal prairie. Geology, 31(4): 331-334. 211 Sulkowski, M. and Himer, A.V., 2006. Element fractionation by sequential extraction in a soil with high carbonate content. Applied Geochemistry, 21: 16-28. Taylor, A. and Blum, J.D., 1995. Relation between soil age and silicate weathering rates determined from the chemical evolution of a glacial chronosequence. Geology, 23: 979-982. Tessier, A., Campbell, P.G.C. and Bisson, M., 1979. Sequential extraction prodecure for the speciation of particulate trace metals. Analytical Chemistry, 51: 844-851. Teutsch, N., Erel, Y., Halicz, L. and Banin, A., 2001. Distribution of natural and anthropogenic lead in Mediterranean soils. Geochimica et Cosmochimica Acta, 65(17): 2853-2864. Teutsch, N., Erel, Y., Halicz, L. and Chadwick, O.A., 1999. The influence of rainfall on metal concentration and behavior in the soil. Geochimica et Cosmochimica Acta, 63(21 ): 3499-3511. UNFCCC, 1998. Kyoto Protocol to the United Nations Framework Convention on Climate Change, UN, Kyoto. USDA-NRCS, 1999. California Annual Precipitation. USDA-NRCS National Cartography & Geospacial Center, Fort Worth, TX. V elde, B., 1992. Introduction to clay minerals. Chapman & Hall, London. Vitousek, P.M. et al., 1997. Soil and ecosystem development across the Hawaiian Islands. GSA Today, 7: 1-8. Walker, J.C.G., Hays, P.B. and Kasting, J.F., 1981. A negative feedback mechanism for the long-term stabilization of Earth's surface-temperature. J. Geophys. Res. Oceans Atmos:., 86: 9776-9782. Wang, C., Brumm, M. and Manga, M., 2005. Is Heat flow in Sierra Nevada really as low as suggested?, AGU Fall Meeting. Eos Trans. AGU, San Francisco, CA. Wang, J. and Munroe,, R.J., 1982. Heat flow and sub-surface temperatures in the Great Valley, Califomia. U.S. Geological Survey Open-File Report, 82-844: 120. Wilcke, W., Krauss, M., Kobza, J. and Zech, W., 2001. Quantification of anthropogenic lead in Slovak forest and arable soils along a deposition gradient with stable isotope ratios. Journal ofPlantNutrition and Soil Science, 164: 303-307. 212 Wilkens, B.J. and Loch, J.P.G., 1998. Accumuiation of Cadmium and Zinc from Diffuse Immission on Acid Sandy Soils, as a Function of Soil Composition. Water, Air, & Soil Pollution, 96: 1-16.