The study was carried out at the Syngenta Jealott’s Hill Research Station. For a diagram
of the study site showing microcosm layout see Figure 2.1
Each individual microcosm (Figure 2.2) consisted of a rectangular fibreglass tank
1.0 m in width, 4.5 m long) divided into 3 discrete sections of different depth and
length. Microcosm depth and length of the ‘shallow,’ ‘medium,’ and ‘deep,’ sections
were approximately 0.3 m x 2 m, 0.5 m x 1.5 m and 0.9 m x 1 m respectively. A total
of 24 microcosms were available. The microcosms were sunk into the ground with the top
edge at ground level, in order to buffer and stabilise the temperature in the microcosms.
Turf was laid in between the microcosms. The microcosms were uniquely labelled M01
– M24 as shown in Figure 2.1.
Sediment was added to a depth of approximately 10 cm to each section and water
was added to give depths of approximately 10 cm (shallow section), 30 cm
(medium section) and 70 cm (deep section). The length of these sections was 2 m,
1.5 m and 1 m, respectively, although the 2 m shallow section also incorporated
a mud bank, which sloped from the edge to approximately half the length of the
shallow section (see Figure 2.2). The approximate water volume of the shallow,
medium and deep sections was 117, 450 and 700 litres respectively, giving an overall
microcosm capacity of approximately 1267 litres.
Mixing and addition of sediment and water was carried out prior to sampling.
Sediment was transferred by wheelbarrow (~80L capacity) from source pond 5
See Figure 2.1) and distributed between the microcosms to give a depth of
approximately 10 cm in each section. The sediment was levelled with a spade and
initially sufficient water (taken from source ponds 4 and 1 (see Figure 2.1) was
added to cover the sediment to protect the microcosms from frost damage. The
pondsbwere later topped up with water, from the same source, to the required depth.
Mud banks were incorporated into the shallow sections of the microcosms, to
form a gradual transition of depth. Two wheelbarrow loads(approximately 50L per
load) of sediment was added to the shallow region of each microcosm, and then
smoothed over with a spade to form a slope. A double layer of mesh (approximately
1 cm2 mesh size, purchased from “World of Water Aquatic Centre”, Shinfield,
Reading, Berkshire) was placed on top of the slope to help hold it in place, followed
by another one and a half wheelbarrow loads of sediment to finish forming the slope.
When completed, the bank extended to approximately 1 m into the shallow region
from the top of the microcosm edge (see Figure 2.2). Each microcosm was prepared in
a similar manner to promote uniformity of experimental units.
Samples of water and sediment were sent to Natural Research Management Ltd (NRM)
for analysis of parameters including selected nutrients, metals and organic compounds.
These data confirmed that the sediment and water tested were suitable for use in
the study.
Much of the biota was added along with the naturalised sediment and water
(described in Section 2.1.3.1). Further enhancement of the phytoplankton or periphyton
communities was not considered necessary as the initial addition via the sediment
and water and rate of natural colonisation was considered to provide a sufficient
introduction. However, due to the short-term nature of the study, abundances
of macrophytes, macroinvertebrates and zooplankton were enhanced via planting
and addition. This was carried out prior to any pre-treatment sampling. Additions
were made to each microcosm in a similar manner in order to promote uniformity of
experimental units. After commencing biological sampling, no further additions of biota
were made to the microcosms.
After completion of the microcosm set-up phase (the addition of sediment, water
and biota), the systems were monitored for water level and macrophyte cover using
methods described below.
2.1.4.1 Water level assessments
The water levels in the microcosms were monitored at least fortnightly throughout
the study period. A piece of tape was placed on the microcosm lip, 1 m down from the
bank end of the shallow section of each microcosm to mark the positioning for water
depth measurements. Water depth, to the nearest centimetre, was measured at this
position in the microcosms using a ruler placed in front of the tape marker.
Water level adjustments were made when fluctuations were +/- 5cm from the
desired level, or when considered necessary. Microcosm water additions were
made to the deep section using water pumped from source pond 4 (see figure
2.1), filtered through a 100 µm mesh filter. Where water needed to be removed,
the water was pumped out via a 100 µm mesh filter and poured down the drain.
The microcosms were set up to include a natural shallow zone providing a good cover
of marginal vegetation, in addition to the deeper areas supporting lower numbers of
macrophytes. The different sections aimed to provide a range of habitats and enhance
the diversity of flora and fauna in the systems. Microcosms were prepared as similarly
as possible.
A list of species present in the microcosms was prepared and the percentage cover
(when viewed from above) of submerged/floating/emergent plants, open water and
bare-ground was also visually assessed on these days. Assessments were made in
5% increments with any group present at <5% being scored as <5%.
All samples were taken in a fully randomised manner. The only exception to this
was the Hydrolab (used for physic-chemical determinations) where post-application
measurements were done in ascending treatment order to prevent cross-contamination.
Watersamples were taken for various biological and chemical measurements. On each
sampling occasion, the deep section of each microcosm was sampled using a
mechanical depth integrating, trap door water column sampler (as illustrated in
Figure 2.3). The sampler was lowered vertically through the water column, with
the entrance flap open, until it was approximately 10 cm above the sediment.
The sampler was then sealed, by pulling the lever-handle, and the entire water column
then lifted vertically from the microcosm. If a large amount of filamentous algae was
present, the sample was discarded and another sample taken. The sampler contents were
then poured into a bucket. This was repeated until approximately 12 litres were collected
in the bucket.
The combined water sample was mixed vigorously with a wooden stirrer and sub-samples
were taken for the following determinations :
1. Alkalinity/Hardness and Chlorophyll a analysis. See Sections 2.4.2.2 and 2.4.3
2. Phytoplankton. See Section 2.4.4.
3. Zooplankton. See Section 2.4.5.
Excess water was then returned carefully to the deep section of the sampled microcosm.
Measurements of water temperature, dissolved oxygen (DO), pH, turbidity and
conductivity were carried out in situ on all of the study microcosms, using a
“Hydrolab” H20 or YSI6820 multiparameter instrument. The probe was placed at
approximately mid-depth in the deep section of the microcosm and a record of each
parameter was taken, once the readings had stabilised.
The alkalinity and total hardness of water from each microcosm were measured.
A 500 mL labelled bottle was immersed in the combined water column sample (see Section 2.4.1),
and the bottle was filled and then capped. The sample was taken to the laboratory for
processing and analysis, as soon as possible after sampling.
In the laboratory, 100 mL of the sample was removed and alkalinity was determined
by titration with a standard hydrochloric acid (HCl) solution in the presence
of bromocresol green/methyl red indicator. A further 50 mL was removed and
total hardness was determined by titration against ethylenediaminetetra-acetic
acid (EDTA) disodium salt in the presence of Erichrome Black indicator.
The calculations used were as follows: -
Alkalinity (as mg L-1 CaCO3) = Volume of HCl titre (mL) x Molarity of HCl x 500.5
Total hardness (as mg L-1 CaCO3) = Volume of EDTA disodium titre (mL) x 1000
Volume of water sample
Water column sampling for the analysis of chlorophyll-a was carried out. A 500 mL bottle
was immersed in the combined water column sample (see Section 2.4.1), and the bottle was
filled and then capped. The sample was taken to the laboratory to be processed as soon as
possible for alkalinity, hardness and chlorophyll-a determinations (see Section 2.4.2 for
details of alkalinity and hardness measurements). Chlorophyll-a content was determined by
fluorometry.
Chlorophyll a concentration in the microcosm water samples was determined using
a bbe Algae Analyser Cuvette Fluorometer. The measurements were performed as follows:
Background adjustment – a subsample(> 25 mL) of the microcosm water sample was
filtered under vacuum through a 0.45 mm filter membrane. The filtered sample
was transferred to a glass cuvette, placed in the fluorometer for analysis in order to
provide a background reading.
Measurement of the test sample – a subsample (> 25 mL) of well-mixed, unfiltered
microcosm water was poured into a cuvette via a coarse mesh filter (approximately 0.5 mm)
to eliminate large particles of e.g. filamentous algae. The sample was analysed by
fluorometery, taking into account the background adjustment described above, to give total
chlorophyll-a in mg/L.
Phytoplankton samples were taken using the water column sampler, as described in
Section2.4.1. A 250 mL bottle was immersed in the combined water column sample
(see Section 2.4.1) filled and then capped. The sample was taken to the laboratory to
be processed and preserved as soon as possible by adding approximately 6 mL of
Lugol’s-iodine solution to the sample bottle.
Samples were transferred to Institut fuer Gewaesserschutz or Technische Universität
München for analysis. A summary of the methods used to identify and count the
phytoplankton samples are as follows.
Samples were prepared for identification and enumeration according to the following method.
The phytoplankton sample was shaken thoroughly and the tubular chamber was then filled with
the sample water. When the settling volumes deviated from 50 ml, the desired volume was
pipetted into the tubular chamber, and the remaining volume was made up with water. For
sedimentation the samples were allowed to stand for at least 24 hours. The supernatant water
was removed by inserting a glass disc between the tube and the base chamber (containing the
sample to be counted) and excess water was taken up with cellulose.
The samples were evaluated using an inverted microscope at a magnification of 400x.
A strip of the counting chamber (width 100 µm) along its diameter (2.6 cm) was counted
with the aid of a counting grid. In cases where the individual count of a species (or
other taxonomic level being evaluated) consistently exceeded 15 in an area measuring
100 x 100 µm, the whole strip was not counted - instead 10 quadrants measuring 100 x
100 µm were counted for this particular species/category. Care was taken to ensure
that the 10 quadrants were uniformly distributed over the whole counting strip. Where
this was carried out it was recorded on the data sheet. The count from these 10 quadrants
was multiplied by 26 to account for the total area of the strip. Following identification
and counting, the number of individuals per mL of each taxon was determined.
Zooplankton samples were taken using the water column sampler, as described in
Section 2.4.1. A 12 L subsample of the microcosm water was removed from the composite
water column sample using a plastic measuring jug, and poured through a 100 mm mesh
size conical plankton net with detachable end sieve. The inside of the plankton net
was washed into the end sieve with tap water from a low-pressure hose. The end sieve
was then detached from the net and the contents washed into a 125 mL sample bottle with
tap water from a washbottle. The amount of water used was kept to a minimum, so that
the bottle was only approximately one quarter to one third full. The sample was taken
to the laboratory to be processed and preserved as soon as possible by topping up the
volume with industrial methylated spirit (IMS) to give approximately a 70% aqueous IMS
solution. Two drops of 0.4% rose bengal stain were then added.
Samples were transferred to Institut fuer Gewaesserschutz or Technische Universität
München for analysis. A summary of the methods used to identify and enumerate the
zooplankton samples are as follows.
The zooplankton sample was filtered through a gauze of 63 µm mesh size. Organisms
remaining on the gauze were rinsed off with water into a counting chamber along with
any organisms remaining in the sample bottle. The samples were evaluated under a
stereomicroscope with transmitted light illumination. The following groups were identified
and counted in the zooplankton samples: Rotatoria, Cladocera, Copepoda, Insecta
(Chaoboridae) and Ostracoda. All adult Crustacea (Cladocera, Copepoda and Ostracoda) were
identified to species level if this was possible without extensive .preparation of the
organisms. No size classifications or egg counts were made, with the exception of the
Copepoda for which counts of nauplii and copepodites were included. Following identification
and enumeration, the number of individuals per litre of each taxon was determined taking
into consideration the sampling volume, the counted area and the total area of the counting
chamber.
Macroinvertebrate sampling was carried out using 3 techniques:
1. Enhanced Surface Area Substrate samplers (subsequently referred to as ‘ESAS’) –
artificial colonising substrates to sample benthic and epiphytic dwellers.
2. Sweep-netting(‘NETS’) – to sample swimming and epiphytic organisms.
3. Emergence Traps(‘ET’) – to sample emergent adult insects.
During the study conducted in 2000 only techniques 1 and 2 were implemented.
Due to the small size of the test system and the resultant potential for destructive
sampling of the populations, the organisms collected by ESAS and NETS were identified
and enumerated live. The organisms were then returned to the originating microcosm.
Each ESAS comprised of a plastic colonising area (constructed from 14 polypropylene
pall rings); weights (sealed plastic bottles containing sand); retaining mesh
(1 mm) at the base (to avoid loss of macroinvertebrates from the sampler when
raised through the water column), and a floating marker attached by a cord (see
Figure 2.4).
Two sets of ESAS (labelled A and B), were employed during the sampling period.
Each set consisted of two samplers in the shallow section and one in each of
the medium and deep sections. The samplers were labelled in order to identify
the microcosm number, position (shallow, medium or deep) and set (A or B) in
each microcosm. They were placed in the microcosms so that the mesh was in
contact with the sediment surface, away from each other and from plants where
possible. Each set of samplers was left to colonise for two weeks before
sampling, and was returned to the microcosms for recolonisation on completion
of sample processing. ESAS sets A and B were alternated to ensure the two-week
colonisation was maintained.
On each sampling occasion, the ESAS positioned in the medium and deep sections of
the microcosms were removed and processed as a combined sample (MD). When these
samples had been processed, the two samplers in the shallow (S) section were removed.
The sampling procedure is detailed below:
1. The ESAS were carefully removed from the microcosm by pulling on the attached
cord. They were placed in a bucket labelled with microcosm number and section (‘S’ or ‘MD’)
and containing a small amount of water from the originating microcosm.
2. The organisms were then gently washed from the samplers and bucket using
a low pressure hose into a large tray. Any excess plant material or debris was removed
from the tray using forceps, making sure no organisms were removed from the sample.
The samplers were also checked to ensure that all organisms were removed; however, where
an organism could not be dislodged it was identified and recorded on the sample datasheet.
The sample was concentrated by passing it through a 250-425 mm mesh sieve. It was then
transferred from the sieve to a counting tray, labelled with the microcosm number and
section (‘S’ or ‘MD’), by inverting the sieve and washing with water from a low-pressure hose.
3. Macroinvertebrates were identified and counted live, by transferring individuals
carefully using either a plastic pipette (the end of the plastic pipette had been cut
to prevent damaging the organisms) or forceps into a separate tray. Individuals
left in the originating tray were counted systematically from top to bottom, left to right.
Where necessary a magnifying lens or microscope was used to aid identification. Organisms
were identified to species level when possible or genus/family level where organisms were
too small, immature or identification of live specimens to a lower level was not practically
feasible.
4. Following identification and enumeration, and after all ESAS had been sampled,
organisms were returned to their microcosm and section depth. The samplers were also
returned to their original positions in the microcosms after completion of
sample processing.
5. Counting results were captured on hard copy data sheets
and were then checked and transferred from these sheets into an electronic database.
A single macroinvertebrate sample was taken from the shallow (S) section and a
combined sample from the medium and deep (MD) section of each microcosm, using
a 1 mm mesh sweep net supported by a 15 cm x 15 cm frame. For the shallow
sample, the water was swept for a 30 second period incorporating as much of the
section as possible, tapping the sediment and plants. The medium and deep
combined sampling was split into two 15-second periods, one for each section.
Netting in the medium and deep sections was carried out in continuous top to
bottom sweeps, incorporating as much of the water depth and width as possible.
The netting procedure was timed using a stopwatch.
The samples were transferred to sample trays (labelled with microcosm number and
section) containing a little water from the appropriate microcosm. Each sample
was transferred by inverting the net over the sample tray, and the water in the
tray used to gently rinse the net of all organisms.
The organisms were then gently washed from the trays using a low pressure hose into
a large tray. Any excess plant material or debris was removed from the tray
using forceps, making sure no organisms were removed from the sample. The
sample was concentrated by passing it through a 425mm mesh sieve. Following this
clean-up process, the sample was transferred from the sieve to a counting tray, labelled
with the microcosm number and section (‘S’ or ‘MD’), by inverting the sieve and washing
with water from a low-pressure hose.
Macroinvertebrates were identified and counted live as described for the ESAS sampling
(see above) and data recorded in the same way. Following sample processing, and after
completion of all net sampling, organisms were returned to their originating microcosms
and depth section.
Emergent adult insects were sampled using emergence traps placed over the shallow
section of each microcosm (see Figure 2.5 for a diagram showing the trap design).
Each trap consisted of three panels of white mesh (104 x 26 mesh per inch), and a front
panel of clear, non-toxic PVC plastic with one sleeve opening for access to the interior.
The bottom of the trap was open with four corner loops, which were attached to a plastic
frame (145 x 60 cm) made from PVC tubing (diameter: 27mm) using cable ties.
‘A’ – main part of trap ‘B’ – plastic frame design
The traps were positioned over the shallow section of a microcosm starting from the
1 m tape mark to 60 cm towards the shallow/medium section divide (the emergence trap
frame being 60cm wide). Pegs were placed in the ground next to the microcosm to attach
the frame securely, and to prevent the trap blowing away. When not in use the traps were
placed away from the study microcosms.
The traps were placed over the microcosms to collect emergent insects for a minimum
of one night every two weeks during the study period. The frequency of sampling depended
on weather conditions and sample size. Traps were positioned late afternoon for
collection the following morning. In the morning the emergence trap of a microcosm would
be carefully lifted and emergent insects collected using a battery-powered pooter to
carefully suck up the adult insects. The fly sample was then carefully transferred into
a labelled bag partially filled with air to avoid sample damage. Large organisms, such
as Anisoptera and Zygoptera, were identified live at the study site and then allowed to fly
away. These taxa counts were recorded on the activity record for that day.
All the sample bags were taken to the laboratory for transfer to glass vials. Transfer was
aided by placing the samples in the freezer for approximately 30 minutes to temporarily stun
the flies. After this time the bags were removed from the freezer and the fly sample transferred,
by pooter, from the plastic bag into a pre-labelled vial containing 70% IMS.
Identification of these samples was carried out to family with further identification
of selected samples to produce a species list.