Geology Reference
In-Depth Information
et al. 2008). The Sr/Ca ratios varies by around a
factor of 2 in dripwater during the monitoring
period at Gib04a (Fig. 6b) which is comparable
with the range of calcite Sr in Figure 10, making it
unnecessary to invoke kinetic effects to account
for annual Sr variability at this site (cf. Fairchild
& Treble 2009). The data in Figure 10 were obtained
using a 1 micron beam in 10 micron steps. They
are not affected by mixing of different calcite
types caused by sampling parallel to irregular
boundaries, but suffer from a different problem in
that the precise position of the irregular LCC -
DCC boundary cannot precisely be mapped along
the Sr line traverse. However it can be seen that Sr
variations closely follow the pattern of variation in
d
13
C, but Sr values are more constant during
winter deposition of LCC and in summer, where
d
13
C is more uniformly high (i.e. a plateau with
rounded shoulders caused by micromill sampling),
Sr values peak then gradually decline.
Time lines corresponding to mid-April and mid-
November are shown on Figure 10 and it is apparent
that the Sr cycle begins its rise at about the time
of the April transition, but reaches a short-lived
maximum well before the November transition in
July or August. The Sr values then decline through
the autumn and the minimum is reached after the
November transition. Neither the poor 'square
wave' resolution of the Micromill nor the uncer-
tainty of the fabric boundaries with respect to Sr
can influence these phase relations, which must
have a different explanation. One possibility is
that the timing reflects the influence of drop interval
on degassing and PCP onto the straw stalactite
above Gib04a. The discharge log in Figure 5
shows that discharge generally increases for a
month or two in mid-summer to autumn, but
because the logged years do not cover the four
years of speleothem growth shown in Figure 10a
precise comparison is not possible. Nevertheless,
if discharge regularly increases in late summer, as
it seems to, there would be less time between
drops for degassing and therefore less PCP, so Sr/
Ca would fall. We suggest timing of the Sr peak rep-
resents the onset of increased summer flow at
Gib04a; flow is then slowly reduced again resulting
in increased Sr/Ca values before the November CO
2
rise reduces degassing more completely. After the
sharp winter switch in pCO
2
, PCP slows down and
Sr/Ca falls sharply to the winter minimum plateau.
Thus the details of the d
13
C and Sr cycles appear
to be subtly decoupled from each other with respect
to fabric development. This is consistent with the
results of the degassing - calcite precipitation pro-
cess modeled above in which degassing rates are
highly responsive to seasonally variable cave air
pCO
2
whereas calcite precipitation rates (around
10
3
of Sr at a far slower rate. Banner et al. (2007) also
showed that calcite precipitation can temporally
cease under high cave air pCO
2
conditions and the
sharper junction between LCC back to DCC may
represent a temporary cessation of growth, also
marked by steps in Sr concentration in the calcite.
Clearly, even at this resolution and high level of
confidence in correlating fabrics and chemical pro-
files, the fabric - isotope - trace element relations
still remain rather ambiguous. However the jumps
in d
13
C clearly mark the suppression and restoration
of degassing as cave air pCO
2
levels rise and fall
sharply in mid-November and mid-April accom-
panied with synchronous changes in d
18
O.
Because no clear evidence exists of kinetic
enhancement of d
13
C during seasonal cycles
several other processes may control the irregular
annual d
18
O excursions to heavier values seen in
Figure 10, including changes in the composition of
the drip water, and will be discussed further else-
where. The minimum d
18
O values, as discussed in
Mattey et al. (2008), represent the dripwater compo-
sitions recorded when the cave environment is most
conducive to equilibrium precipitation (i.e. lowest
d
13
C and highest cave air pCO
2
) and these values
correlate well with the weighted mean d
18
Oof
winter precipitation over a 54-year period (Mattey
et al. 2008).
Conclusions
Detailed monitoring of three drip sites in NSM
reveals a strongly coherent seasonal pattern of
dripwater compositions despite each site having sig-
nificantly different flow paths and discharge pat-
terns. Calcite saturation is closely linked to regular
seasonal variations in cave air pCO
2
which is
highest between November - April. The seasonal
switch to low pCO
2
in the summer is caused by
chimney ventilation linked to temperature differ-
ences between the exterior and interior of the
cave. Advection of isotopically homogenous
CO
2
-rich ground air derived from deeper levels in
the Rock maintains high cave air pCO
2
levels result-
ing in winter speleothem deposition of columnar
calcite having the lowest d
13
C and d
18
O values.
Flushing by the outside atmosphere lowers cave
air pCO
2
levels in summer leading to higher
degrees of dripwater degassing rates precipitation
a dark compact calcite having elevated d
13
C values.
A coupled CO
2
degassing - calcite precipitation
model links the development of annual cycles in
d
13
C and dripwater evolution to switching pCO
2
driven by seasonal cave ventilation. The model
shows that drip water supersaturation is consistent
with degassing rates 10
4
greater than the rate of
slower, see above) drive the PCP enrichment
calcite
precipitation
and
also
accounts
for
the
Search WWH ::
Custom Search