Geology Reference
In-Depth Information
were also systematically sampled for detailed water
analysis by means of a Perkin-Elmer Optima
5300DV (Perkins-Elmer, Waltham, MA, USA)
inductively coupled plasma optical emission spec-
trometer (ICP-OES) for dissolved metal ions and
titration with 0.02 M hydrochloric acid using the
'BDH 4.5' indicator for bicarbonate. The water
used for experiments was taken from a chalk
spring in Welton (East Yorkshire, UK). This water
was stored at 5 8C, sterilized by UV treatment and
placed into the flume reservoirs within 1 week of
collection. Water was not replenished during
periods of equilibrium experimentation, and a
small loss (,10%) volume was incurred via evapor-
ation over this period. Detailed experiments con-
cerning the impact of photoperiod and temperature
on these systems revealed deviation of sterile
Flume 4 from physicochemical expectation. It is
therefore likely that some unknown chemical or
biological agent infected this Flume during the
experiment, meaning that the results of Flume 3
should be taken as a more reliable representation
of sterile systems.
Table 1. Temperature and photoperiod conditions
used during treatments contributing to Experiment
2. Temperature was controlled via an in-line chiller
unit, and lighting via automated over-head
hydroponic lamps
Experiment
name
Photoperiod
length (hours)
Water
temperature (8C)
'mid-summer'
18
14
'cold summer'
18
8
'warm winter'
6
14
'mid-winter'
6
8
thereby likely exaggerating oscillations driven by
microbe metabolic processes. Nevertheless, the
carbonate-biofilm-water chemical system these
experiments represent is the same as at field sites,
so this issue is likely only to affect the magnitudes
of effects described.
The mesocosm facilities at the University of Hull
allow independent control on ambient temperature
(via in-line chillers) and photoperiod length (via
lamps) that cannot be achieved in field studies. Fur-
thermore, comparison of the response of colonized
Flumes (1 and 2) to sterilized Flumes (3 and 4)
allows distinction of purely biological from
physico-chemical effects. To exploit this, during
Experiment 2 exposed the systems used in Exper-
iment 1 to a series of 'treatments', the conditions
for which are summarized in Table 1. These exper-
iments were designed to reveal the independent and
combined influences of temperature and photo-
period length on colonized and non-colonized
systems. Each treatment ran for 3 weeks, with the
first week of data disregarded for analysis. Treat-
ments were separated by a 2-week period during
which the flumes were maintained under 'mid-
summer' conditions.
Experiment design
Two experiments are reported in this manuscript;
Experiment 1 was designed to investigate diurnal
cyclicity in a system in approximate equilibrium,
and Experiment 2 was designed to investigate the
impact of changes
in photoperiod and water
temperature.
In Experiment 1 all four flumes were run under
'mid-summer conditions', with water temperature
set by the in-line chillers at 14 8C and the day
length at 18 hours, and flume water was allowed
to evolve freely over a period of three months.
The initial evolution of water chemistry is domi-
nated by carbonate precipitation, and this is
described in detail elsewhere (Rogerson et al.
2008). Subsequent to attainment of approximate
equilibrium in terms of Ca
2þ
Results
Characterisation of the diurnal cycle
in colonized and sterilized systems
(aq)
and HCO
3(aq
2
, the
diurnal cycle of pH and conductivity was recorded
for a further 2 months in order to establish
whether diurnal changes were cyclic and what
chemical and biological factors control this cycli-
city. As this data represents near-equilibrium
conditions, it reflects levels of calcite saturation
(SI
calcite
¼ lnf[Ca]
(aq)
[CaCO
22
]
(aq)
/k
sp
g) slightly
below that usually associated with tufa precipitation
(SI . 2.5; Pentecost 1992). Maintaining perfectly
constant hydrochemistry under a condition of
rapid calcite precipitation is extremely challenging
in a recirculating flume, and inevitable small varia-
bility in alkalinity may have overprinted the photo-
synthetic variability this study is concerned with.
Consequently, the flumes systems are relatively
weakly buffered in comparison to many field sites,
Figure 1a-d show the evolution of pH in Flumes
1-4 respectively during Experiment 1. The impact
of photoperiod cycling in Flumes 1 (Fig. 1a) and 2
(Fig. 1b) is clear, with the 6-hour period of dark con-
sistently characterized by lower pH than the 18-hour
period of light. The amplitude of day/night vari-
ations in both flumes increases with time, in close
association with the decrease in dissolved calcium
(Fig. 1e). Note that the decrease in dissolved
calcium is primarily related to precipitation of
calcite, and that evolution of bulk water chemistry
over this experimental period is fully described else-
where (Rogerson et al. 2008). Consequently the low
Search WWH ::
Custom Search