Seasonal record from recent fluvial tufa deposits (Monasterio de Piedra, NE Spain): sedimentological and stable isotope data - Tufas and Speleothems: Unravelling the Microbial and Physical Controls - page 136

calculated using the average measured water d 18 O
values for each period, by means of the O'Brien
et al. (2006) formula:

considered. Neither were there regular variations
through space among the samples of the three
tablets in fast flow conditions.
Although the rhythmic pattern is not the rule for
d 13 C, some recent tufa systems have registered it.
In the River Piedra, a few intervals recorded it
(e.g. tablet 3). In the case studied by Chafetz et al.
(1991) the d 13 C seasonal pattern presented higher
values in the summer sediment than in winter sedi-
ment, which was attributed to the increase in 12 C
loss by degassing in summer. In contrast, the seaso-
nal pattern of the example studied by Matsuoka
et al. (2001) had low summer values, which were
explained by local seasonal phenomena that were
able

T (8C) ¼ 15:310 4:478(d 18 O calcitePDB
d 18 O waterSMOW ) þ 0:14[0:227
1:0412(d 18 O calcitePDB
d 18 O waterSMOW )]

In the case studied by these authors, the temperature
range of tufa formation was similar to that of the
Parque del Monasterio de Piedra. In the River
Piedra, the calculated temperatures, which corre-
spond to a theoretical equilibrium precipitation,
show a high degree of agreement with the measured
temperatures, displaying the same seasonal trend,
although the calculated temperatures present a
narrower range than the measured ones (Fig. 12).
Similar differences have been reported by Lojen
et al. (2004), although a larger disagreement
between calculated and measured temperatures
was found; this was linked to warmer temperatures
with wider ranges.
In the River Piedra, the measured water tempera-
ture range was 12.0 8C in the studied period. For the
measured water d 18 O composition, which remained
almost constant through the studied period, this
temperature range corresponds to a variation in
d 18 O of the carbonate sediment of 2.9‰ PDB,
assuming an approximate change of 0.24‰ per 8C
for the equilibrium precipitation (Andrews 2006).
The average measured d 18 O range in the sedi-
ment from tablets is 1.6‰, which roughly corre-
sponds to a temperature range of 6.7 8C. These
data are closer to the range of calculated water temp-
eratures (5.4 8C) than to the measured ones (12 8C).
To some extent, this difference can be explained
by the fact that the analysed sediment corresponded
to a period of several months, while the measured
water temperatures represented punctual conditions
within the periods. In spite of that difference, the
calculated water temperatures are quite similar to
the actual ones. The cool period of 2002 - 2003
and the warm period of 2002 yield calculated
water temperatures too high and too low, respect-
ively, for the kind of periods; this is related to the
above mentioned anomaly in d 18 O detected in
tablets 2 and 5. As discussed above, high sedimen-
tation rates in this cool period are linked to the
warmer-than-normal mean temperature of Novem-
ber and December of 2002 (about 4 8C higher com-
pared to the same months in 2001), so that most
sediment of this cool interval probably was depos-
ited under warmer conditions than the average of
the

to compensate the degassing effect. In

the d 13 C

general, behaviour is more complex

because it can suffer the influence of several

factors, both biotic and abiotic.
The rather irregular variation of d 13 C compo-
sition through time and space for most of the
studied intervals in such a short stretch of the
River Piedra can be attributed to changing influence
of the several variables affecting tufa formation on a
local scale. Changes in d 13 C of the stream water that
affect the entire fluvial system do not explain such
irregular variation and must be disregarded. This
is the case of seasonal changes in temperature, and
of variations in water d 13 C caused by input of soil-
derived CO 2 , introduction of atmospheric CO 2 into
soil in arid conditions and variations in aquifer
d 13 C composition or in aquifer contribution to the
stream. Only processes that operate on a local
scale can be invoked: for example, 13 C enrichment
caused by a local increase of mechanical CO 2 degas-
sing or local changes of biofacies (type of flora and
its physiological activity). However, the expected
slight increase in d 13 C composition due to photosyn-
thesis is not always traceable (Shiraisi et al. 2008).
In particular, in fast flow environments, the effect
of this process commonly becomes masked by the
strong influence of the mechanical degassing,
which increases the 13 C content in water.
To sum up, the generally irregular d 13 C pattern
through space and time might indicate that the
several variables that affect the d 13 C composition,
such as the fractionation processes mentioned
above, had changing contributions through the
studied warm and cool periods. For instance, it
cannot be excluded that the sediment sampled
at each interval in a particular subenvironment
could present some slight differences in terms

of biofacies.

Temperature calculations

The theoretical water temperature at which the
sediment of warm and cool periods was formed,
in a theoretical equilibrium precipitation, was period. However, the warmer temperature