Page 6 - GSA Today December 2022

D (cm .s ) = diffusion constant for CO consumption by silicate weathering can be 500-m.y. increments to calculate standard
–1
2
CO
2
2
in air (=0.162); α (fraction) = ratio of diffu- calculated from stoichiometry of Equations deviations as the height of the open box
sion constant for CO in soil divided by dif- 1–4 and carbon consumption by apatite (Fig. 3).
2
fusion constant for CO in air (=0.1, range weathering from stoichiometry of Equation 5.
2
0.08–0.12); L (cm) = original depth to water STEPWISE BIOTIC ENHANCEMENT
table (after decompacted using Equation 6). DATABASE, ERROR CALCULATIONS, OF WEATHERING
The duration of soil formation in years AND ALTERNATIVES The results of mass transfer calculations of
(A in k.y.) can be calculated from carbonate Detailed accounts of each of the paleosols paleosols ranging back in age to 3700 Ma
nodule diameter (D in cm: r = 0.57, s.e. = 1.8, used in the compilation for these calcula- show three orders of magnitude increases in
2
p = <0.001) for calcareous soils (Retallack, tions have all been published elsewhere: nutrient depletion of both phosphorus and
2005), or thickness of profile (T in cm: r = citations and component data, including alkali and alkaline earths, but on different
2
0.79, s.e. = 140, p = 0.01) for non-calcareous error estimates for individual profiles, are time schedules (Figs. 3A–3B). Most of the
unconformity paleosols (Markewich et al., listed in the supplemental material . Criteria range of alkali and alkaline earth depletion
1
1990): for quality of data outlined by Rye and was achieved by the Great Oxidation Event
Holland (1998) were used to select paleosols (GOE) of 2.45 Ga, but phosphorus depletion
.

A D 0 34 , (12) for the compilation. Full petrographic and rose markedly at both the GOE and the
3 92
.
geochemical data, as well as bulk density Neoproterozoic Oxidation Event (NOE) of 0.8
A 4 915. . . (13) determinations, were essential for all hori- Ga. These changes may reflect increased rates
T
343 4
zons (Equations 7 and 8). Also needed was of nutrient procurement due to increased
Mean annual precipitation (P in mm) can evidence of at least moderate development, biological productivity at those times.
be obtained by the CIA-K proxy, effectively such as argillic, calcic, or gypsic horizons Alkali and alkaline earth depletion rose
a chemical index of alteration without diage- (Retallack, 2013, 2018, 2022b). To be steadily from 3.5 to 2.4 Ga under acid-sulfate
netically problematic K (I as mole fraction: included, paleosols had to have chemical weathering by anaerobic bacterial soil micro-
r = 0.72, s.e. = 182, p = <0.0001; Sheldon et weathering demonstrated by tau analysis biomes (Retallack, 2018; Retallack et al.,
2
al., 2002), or compaction-corrected depth to (Brimhall et al., 1992). Weakly developed, 2016), now restricted to waterlogged soils
calcic horizon (D in cm: r = 0.52, s.e. = 147, gleyed, and inadequately documented paleo- and playa lakes (Benison and Bowen, 2015).
2
p = <0.0001; Retallack, 2005): sols were not included. The paleosol data- Alluvial paleosols from 3.5 to 3.0 Ga contain
base includes profiles on bedrock unconfor- desert roses of sulfate minerals, such as bar-
P 221 1 e 0 0197 I , (14) mities (Rye and Holland, 1998), as well as ite and gypsum, as evidence for weathering
.
.
within sedimentary sequences (Retallack, by strong sulfuric acid rather than weak car-
P D 0 0132 . (15) 2013, 2018, 2022b). Virtually all suitable bonic acid (Retallack, 2018; Retallack et al.,
24
.
64
137
.
5
.
2
Precambrian paleosols are included in 2016). The microbiome of desert rose paleo-
The normalized value of μmol F.cm . the database, along with most suitable sols dated to 3.0 Ga is permineralized with
2
mm .a , where F is the sum of the four Phanerozoic paleosols for which data was silica, and its microfossils, analyzed for cell-
–1
–1
alkaline and alkaline earth bases, or μmol available. Errors for the calculations were specific carbon-isotopic-composition, reveal
G.cm .mm .a , where G is the sum of based on standard errors of transfer func- an anaerobic community of purple sulfur
–1
2
–1
phosphorus depletions, become proxies for tions (Equations 12–15) and Gaussian error bacteria, actinobacteria, and methanogens
global CO consumption if multiplied by propagation from partial derivatives of (Retallack et al., 2016).
2
modal mean annual precipitation, which is transfer equations summed in quadrature as Other paleosols in the data set formed in
764 mm in the modern world, with a stan- outlined by Retallack et al. (2021). humid climates on bedrock (supplemental
dard error of 704 mm (Beck et al., 2005). Some of the transfer functions used are material [see footnote 1]) and were thick,
This modal mean annual precipitation may compromised by other variables: Equations clayey profiles, with little evidence of solu-
have changed in deep time, but the current 14 and 15 for paleoprecipitation include ble salts (Rye and Holland, 1998). These do
understanding of paleoprecipitation from components of temperature (Sheldon et al., not stand out as anomalies in Figure 3 com-
paleosols shows mainly arid to subhumid 2002) and paleoproductivity, respectively pared with paleosols with soluble salts
estimates (Retallack, 2013, 2018; Retallack (Breecker and Retallack, 2014), which con- (Retallack, 2022c) because they were nor-
et al., 2016), comparable with today (Beck tribute to cited standard errors. Warmth and malized for mean annual precipitation
et al., 2005). Estimates of exposed land area high precipitation can also compromise age (Equations 12–13) and duration of forma-
in deep time are from published areas of estimates of paleosols using nodule size tion (Equations 14–15). CO consumption
2
continental crust and hypsometric curves (Retallack, 2005) and depth of weathering rates of Paleoproterozoic and Archean
(Cawood and Hawkesworth, 2019). These (Markewich et al., 1990), again within stan- paleosols are too low (Fig. 4) to explain
changing land areas were proportionally dard error of the data used for the transfer paleotemperatures under a faint young sun
scaled to a modern land area of 148,429,000 function. Although individual paleosol (Kasting, 2010). Likely sulfur bacteria and
km , and carbon consumption to modern depletion rate standard deviations were methanogens in paleosols support the idea
2
global silicate weathering (Ciais et al., small, the variance of estimated depletion that other greenhouses gases, such as meth-
2013) of 0.3 PgC.a (Pg = 10 g). Carbon rates is large, so rates were pooled by ane, ethane, and SO , formed a greenhouse
–1
15
2
1 Supplemental Material. Table S1. Base and phosphorus depletion and paleoenvironments of 97 well-studied paleosols. Go to https://doi.org/10.1130/GSAT.S.20126417
to access the supplemental material; contact editing@geosociety.org with any questions.
6 GSA TODAY | December 2022