Geology Reference
In-Depth Information
Table 2. Bulk compositions of stalagmite samples ( ppm in calcite)
mm below top
Mg
P
S
Mn
Cu
Zn
Sr
Y
Ba
Pb
Number of analyses
Obi84
0 - 10
903
66
42.4
0.99
2.2
3410
32.4
0.020
174
500
8
11 - 20
976
52
6.7
1.01
2.7
4590
35.3
0.040
196
770
6
.20
1040
67
10.5
0.87
3.4
6420
30.9
0.039
191
1360
7
Obi12
0 - 10
1160
56
30.6
1.02
4.5
4010
32.6
0.015
244
370
8
11 - 20
1280
67
12.8
0.67
3.8
5970
29.2
0.014
234
630
6
.20
1250
93
17.3
1.3
5.4
5690
30.9
0.020
237
510
5
were conducted at relatively fast growth rates using
high ionic strength solutions.
For other solutes, the partition coefficient
concept is not so applicable and it is also more dif-
ficult to obtain water analyses for many species.
Borsato et al. (2007) found that at Ernesto metals
such as Pb and Zn, as well as P, strongly adsorbed
to collection vessels because of being transported
in the form of unstable colloids. Obir water
samples are normally collected monthly, or less
frequently from cumulative collection vessels,
although some direct samples were taken from
drip SH4 from 2002 to early 2004. On realizing
the importance of colloidally transported elements,
in November 2005, we re-analyzed remaining pre-
viously collected samples from drips SH3 and
SH4, acidifying samples to release adsorbed
elements, and carried out ICP-MS analysis. Data
from SH3 (feeding Obi55) and SH4 (feeding
Obi84) are presented in Table 4 and results are gen-
erally consistent between the two. The Fe, Al and Si
probably represent colloidal particles, but organic
colloids could be even more important (Fairchild
& Treble 2009). Pb, Zn and Y (all elements inferred
to be colloidally transported at Ernesto Cave)
have, as expected, high variations in abundance in
contrast with Sr and Ba which are expected to be
very largely present as free ions. There were no
obvious seasonal variations in abundance.
Empirical distribution coefficients, as in
equation (1), between the dripwater data of
Tables 1 and 4 and the composition of the top
10 mm of the stalagmite are shown in Table 5.
The calculated values for Zn and Pb (1) are con-
sistent with experimental and theoretical data sum-
marized by Rimstidt et al. (1998) and Curti (1999)
and with the evidence of strong sorption of these
elements to calcite surfaces (Zachara et al. 1991;
Godelitsas et al. 2003; Chada et al. 2005). The
values of 1 for Cu and Y are quite inconsistent
with data in Rimstidt et al. (1998) and Curti (1999)
where values 1 can be predicted, using rare
earth elements as a model for Y. This discrepancy
can be accounted for if: (1) the partition coefficients
for organic colloidal substances are ,1; and (2) Cu
and Y are more tightly bound to colloids than are
Zn and Pb. The net effect would be that Y and Cu
are
not
so
readily
available
for
incorporation
into calcite.
Stable isotopes
The mean stable isotope composition of the top
2mm (c. 20 years) of growth of sample Obi84 is
26.41 + 0.36 and 27.83 + 0.23‰ for d
13
C and
d
18
O, respectively (n ¼ 200). In the case of d
13
C
there is a direct comparison with the aqueous d
13
C
composition since the fractionation between
CaCO
3
and dissolved inorganic carbon at the pH
values of interest is very small (M ¨hlinghaus et al.
2007). The mean d
13
C value for drip SH4 (which is
very close to those from SH3) is 210.25 + 0.74‰
(n ¼ 19) Table 1. However, the pattern of seasonal
variation of aqueous d
13
C (with a maximum of
28.36‰ representing as one of only two points
higher than 29‰) is less strongly developed than
for the more slowly dripping points SH1 and SH2,
which display increases of around 3 and 8‰ from
the low summer levels of 211‰ (Sp¨tl et al.
2005). Hence the mean difference of 4‰ between
SH4 drip water and the stalagmite is likely to
reflect
continued
degassing
perhaps
with
some
kinetic effects during precipitation.
The d
18
O composition of dripwater feeding SH4
is 210.23 + 0.10‰ (n ¼ 18). Using the exper-
imentally determined fractionation factors of Kim
& O'Neil (1997), such water should precipitate
calcite at equilibrium at a temperature of 5.8 8C
with a composition of 28.5‰. This is 0.7‰ lower
than the observed composition and might indicate
a small kinetic effect, but such a discrepancy is
very commonly observed in natural speleothems
(McDermott et al. 2005) and is not apparent using
the
older
experimental
values
of
Friedman
&
O'Neil (1977).
No detailed treatment of the variation in stable
isotopes with depth is attempted here. However,
we can note that over time the isotopes show rapid
Search WWH ::
Custom Search