Factors controlling growth of modern tufa: results of a field experiment - Tufas and Speleothems: Unravelling the Microbial and Physical Controls - page 148

change in sample weight. The difference between
the tablet weight after and before the experiment
was accepted as the mass increment of tufa. The
rate of tufa growth was calculated by dividing
the mass increment by the total surface area of the
tablet and the exposition time.

time. The Ca concentration and TDS are covariant
in all sites; however, there is no evident correlation
between these two parameters and seasonality. At
the H ´ j site they reach their maximum values
in November 2002 and March 2002/2003. Con-
versely, in Karw ´ w the highest concentration of
Ca was noted in October 2003, while at the
Z ´ zriv ´ site in November 2002. The Z ´ zriv ´ site
is characterized by the smaller variation of Ca con-
centration in the course of a year.
The L ´ ˇ ky site experienced the most dramatic
changes in Ca concentration (Fig. 4). These reflect
fluctuation of mixing proportion between water
derived from shallow and deep circulation. For
example, the Ca concentration decreased from
129.9 to 97.2 mg/l within a two-week period in
March 2003. It resulted from significant influx of
surface runoff due to snow melting in the moun-
tains. The Ca concentration was highly elevated in
June - December 2003 reaching up to 201.1 mg/l.
This, in turn, was the effect of limited surface
runoff, which resulted in greater proportion of the
deep-circulation water substantially elevating the
TDS of the surface stream (Table 2).
The SI calc. values also fluctuated in particular
stations over the period of the experiment (Fig. 4).
All but one calculated values of this parameter
were positive, clearly indicating the overall ten-
dency of calcite to precipitate. The lowest value,
20.03, was calculated for water from the Z ´ zriv ´
site (Z2 station) collected in December 2003. The
outstanding positive values, elevated up to 1.22 -
1.27, characterize the water from the L ´ ˇ ky site
(L3 station) in early March and June 2003 respect-
ively. It clearly reflects the domination of deep cir-
culating component in the water supplying the
studied site.

Texture

The textures of the studied tufa were examined. All
the limestone tablets after weighting were cut paral-
lel to their shorter edge and the standard thin sec-
tions were made. The additional sections were
made from the tufa grown on some copper tablets.
In the latter case the deposits were carefully scaled
off and sunken in boxes filled with ARALDITE.
After the resin solidified, thin sections were made.
Small portions of tufa were sampled from the
tablets to be studied under scanning electron micro-
scope (SEM) JEOL 5410, coupled with a micro-
probe (EDS) Voyager 3100 (Noran product). As
these samples were previously heated in 50 8C
their internal structure could be destroyed. To
study the original structure of fresh tufa additional
samples were collected in the close vicinity of the
tablets. To prevent collapse of the organic structure
some samples were lyophilized prior to examin-
ation. The samples were mounted on SEM holders
with silver glue and coated with C or Au.

Water chemistry

Chemistry of water that fed growing tufas differed
markedly between the sites. The L´ˇky site (L3
station) had the highest values of TDS, and Ca
content: 968.45 mg/l and 145.9 mg/l, respectively
(Table 2). The water at this site represents the
Ca - Mg - SO 4 - HCO 3 type. The water at the
Karw´w site belongs to Ca - Mg - HCO 3 type
while in other sites to the Ca - HCO 3 type. Studied
waters were dominated by Ca ion, the mean annual
concentration of which ranges from 84.4 at Karw ´w
(K1 station) to 67.4 in Z´zriv´ (Z1 station). How-
ever, this content is nearly twice lower than that at
the L´ˇky site. At the sites with lower Ca content
concentration of other ions is also lower, which is
reflected in TDS values ranging between 522.2
(Karw ´w site) and 305.9 (Z´zriv´ site).
Mean SI calc. values were calculated for the water
collected in the nearest stations upstream of the
points where the tablets were exposed. Mean value
of this parameter is highest at the L´ˇky site, equal-
ling 0.94 (L3 station). In other places SI calc. varies
from 0.87 (H6 station) to 0.48 (Z2 station).
Some parameters change seasonally (Table 2).
The general characteristic is lowering of the water
temperature between autumn and spring, which
obviously mirrors the lower air temperature in that

Tufa deposited on tablets and its rate
of growth

In the course of the experiment almost all tablets
were covered with tufa. Substantial differences
could be noticed even in the field between the
amounts of tufa growing on different tablets,
between individual points and between individual
sites studied (Figs 5 & 6). The water level in the
studied sites also changed. In the autumn 2003 it
drastically decreased in response to prolonged
drought. The tablets at the H´j upper waterfall and
dam points, and at the Z´zriv´ site became exposed
and tufa growth on them stopped (Figs 3g & 6b).
The rate of tufa growth is calculated from the
tufa mass increments (Figs 7 & 8). The results are
of semiquantitative character only due to the follow-
ing reasons: (i) asymmetrical growth of tufa on
the tablets; (ii) unknown real area of nucleation