Harmful Algae 76 (2018) 1–10

Contents lists available at ScienceDirect

Harmful Algae
journal homepage: www.elsevier.com/locate/hal

Cyanobacteria breakthrough: Effects of Limnothrix redekei

contamination in an artificial bank filtration on a regional water supply
Adam K. Rosea,* , Larelle Fabbroa , Susan Kinnearb
a
School of Medical and Applied Sciences, Central Queensland University, North Rockhampton, QLD, Australia
b
Research Division, Central Queensland University, North Rockhampton, QLD, Australia

A R T I C L E I N F O A B S T R A C T

Article history:
Received 9 October 2017 Mitigation of cyanobacterial or “blue-green algal” blooms is a challenging task for water managers across
Received in revised form 18 March 2018 Australia. In the present study, a regional drinking water source (located in Central Queensland) was
Accepted 30 April 2018 studied to identify the potential risks posed by cyanobacteria. Data were collected from the drinking
water source (a lagoon) as well as the drinking water supply infrastructure, at monthly intervals between
Keywords: September 2012 and December 2014. In March 2013 there was an extreme rainfall event where
Drinking water floodwaters infiltrated the water supply without passing through bank filtration. The floodwaters also
Queensland compromised the bank filtration via erosion. The pump well and bank filtration system were
Cyanobacteria
subsequently upgraded/maintained in May 2013. Results showed that following the extreme event and
Limnothrix
infrastructure upgrade, two distinct Limnothrix redekei blooms microscopically identified, were detected
Flood
Chlorination in the drinking water supply chain. Further investigations indicated that the species was also present in
Bank filtration the pump well infrastructure, a dark environment, growing on the surface of the newly installed pump
well cement pipe. After observing the occurrence and habitat niche of this species during the present
study, a suggestion was made to minimise cyanobacterial contamination and proliferation within the
water supply chain infrastructure. The preliminary proposal is to use clean sand on the sub-surface layer
of the bank filtration, complemented with biologically active sand as a surface cap. Furthermore, the
culturing techniques reported in this study can potentially be used to optimize assessment for Limnothrix
redekei populations surrounding water extraction points.
© 2018 Elsevier B.V. All rights reserved.

1. Introduction safe water supplies (NHMRC, 2011). When harvesting water from
lakes and rivers, one of the most popular choices for the initial
Maintaining safe and secure water supplies has become barrier is bank filtration. Bank filtration is used widely around the
increasingly challenging throughout Australia as well as globally. world as an economical and reliable technique to remove physical,
The deterioration of water supplies and water quality is frequently biological and chemical contaminants (Huelshoff et al., 2009a).
associated with activities such as population growth, increasing Whilst bank filtration was originally adopted in Europe (Schmidt
urbanisation, and the expansion of agriculture. These pressures et al., 2003); it is now a preferred option for an initial barrier in a
have escalated land and water resource demand and resulted in multi-barrier system and is being implemented across both
over-extraction and poorer water quality. Climate change is also an developed and developing countries (Thakur et al., 2013; Dash
additional stressor on the security of water supplies, due to shifts et al., 2008; Sharma and Amy, 2009; Huelshoff et al., 2009b).
in the duration, frequency and intensity of rainfall patterns, leading Traditional bank filtration, defined as water passing through
to unpredictable wet seasons (IPCC, 2014). This combination of alluvial sediments before entering the drinking water treatment
pressures on water supplies has brought effective water treatment plants, is demonstrably effective as a passive method of improving
and management techniques into sharp focus, as never before. water quality (Tufenkji et al., 2002). A number of studies have
Implementing a multi-barrier approach to minimise contam- validated that bank filtration alone is effective in achieving
inants is recognised as the most efficacious practice in securing guidelines for drinking water standards, through its positive
effects on a range of physical, chemical and biological variables
(Table 1). In particular, bank filtration is a very effective method for
removing cyanobacterial cells and the associated toxins (see
* Corresponding author. Table 1 for references).
E-mail address: a.rose@cqu.edu.au (A.K. Rose).

https://doi.org/10.1016/j.hal.2018.04.010
1568-9883/© 2018 Elsevier B.V. All rights reserved.
2 A.K. Rose et al. / Harmful Algae 76 (2018) 1–10

Table 1
Selected studies/experiments with reported improvements in physical, chemical and biological variables after utilising bank filtration.

Selected Studies/Experiments

Physical Improvements
Turbidity (NTU) Weiss et al. (2005), Dash et al. (2008), Dash et al. (2010), Thakur and Ojha (2010) and Freitas et al. (2012)
Electrical Conductivity (mS/ Freitas et al. (2012)
cm)
Hardness Freitas et al. (2012)
pH Freitas et al. (2012)
Temperature ( C) Gross-Wittke et al. (2010)

Reduction or Removal of Chemicals

Metals Lorenzen et al. (2010) and Freitas et al. (2012)
Nutrients Dash et al. (2008), Lorenzen et al. (2010) and Freitas et al. (2012)
Pharmaceuticals Heberer et al. (2004) and Stauder et al. (2012)
Pesticides Hoffmann and Gunkel (2011a) and Stauder et al. (2012)
Chemical Oxygen Demand Dash et al. (2008)
(COD)
Dissolved Organic Carbon Grünheid et al. (2005), Hoffmann and Gunkel (2011b) and Harvey et al. (2015)
(DOC)
Particulate Organic Carbon Gunkel et al. (2009)
(POC)
Cyanotoxins Chorus and Bartram (1999), Miller et al. (2001), Dillon et al. (2002), Chorus et al. (2003) and Grützmacher et al. (2009)
Natural Organic Matter Metge et al. (2010) and Hoffmann and Gunkel (2011a)
(NOM)
Biological Oxygen Demand Freitas et al. (2012)
(BOD)

Reduction or Removal of Organisms

Cryptosporidium sp. Ray et al. (2003), Weiss et al. (2005) and Metge et al. (2010)
Giardia sp. Ray et al. (2003) and Weiss et al. (2005)
Escherichia coli (E. coli) Wang (2002), Shamrukh and Abdel-Wahab (2008), Dash et al. (2010) and Thakur et al. (2013)
Coliforms Wang (2002), Weiss et al. (2005), Dash et al. (2008), Dash et al. (2010) and Freitas et al. (2012)
Viruses Harvey et al. (2015)
Cyanobacteria Chorus and Bartram (1999), Miller et al. (2001), Dillon et al. (2002), Chorus et al. (2003), Babica et al. (2005), Freitas et al. (2012), Romero
et al. (2014) and Harvey et al. (2015)

Over 40 toxin-producing cyanobacterial genera have been reported in dialysis patients (Jochimsen et al., 1998; Carmichael
identified to date, globally (Sivonen and Jones, 1999; Bernard et al., et al., 2001). There is also a growing body of literature particularly
2011), with a further unknown number of “potentially” toxic that based on animal models, that suggests that cyanotoxin
species. Worldwide, about 60% of cyanobacterial samples investi- exposure may be linked with a range of both acute and chronic
gated have contained toxins (WHO, 2003), with Australian samples effects, following a range of exposure types (Carmichael, 2001).
averaging between 42% and 84% in terms of testing positive for Given the present state of knowledge, it is prudent for water
toxicity (Sivonen and Jones, 1999). The incidence of cyanobacterial managers to regard any bloom formations as potentially harmful
blooms in freshwater supplies appears to be increasing in many and, therefore, corrective action should be taken (WHO, 2003).
parts of the world (Sanseverino et al., 2016; Wells et al., 2015) and In previous studies, it has been reported that bank filtration is
as testing methods become more sophisticated and with increased effective in removing cells of a number of cyanobacterial species, as
emphasis on research in this area in other regions of the world, it is well as the intracellular and extracellular toxins. For example,
anticipated that additional toxic species will be detected (WHO, these include Cylindrospermopsis raciborskii (Romero et al., 2014),
2003). Of recent interest has been Limnothrix redekei, which has Microcystis sp. (Babica et al., 2005; Grützmacher et al., 2009),
demonstrated toxicity in mice and tadpoles, however, the Aphanizomenon sp. (Freitas et al., 2012), and Limnothrix redekei
molecular structure of the toxins has not been elucidated (Freitas et al., 2012), as well as the toxins cylindrospermopsin and
(Humpage et al., 2012; Bernard et al., 2011; Daniels et al., 2014). microcystin. It appears that toxin removal and/or breakdown
Risk management associated with cyanobacteria in water occurs in the uppermost layers of sediment, where the soil is highly
supplies is challenging, as blooms can form in a number of biologically active (Grützmacher et al., 2009; Chorus and Bartram,
different zones of the water body, including the surface 1999; Chorus et al., 2003; Dillon et al., 2002; Mohamed et al.,
(epilimnion), sub-surface (metalimnion and hypolimnion) and 2006). In some studies, however, it has been reported that there
on benthos (Codd et al., 2005; Gaget et al., 2017). Whilst it has been are cyanobacterial cells capable of penetrating the bank filter,
readily accepted that blooms are most common in warm, still including both unicellular cyanobacteria and fragments of
waters of high nutrient concentrations, insidious populations are filamentous cyanobacteria (Rachman et al., 2014; Zamyadi et al.,
also known to occur in less obvious and visible locations, such as in 2012; Pazouki et al., 2016; Lahti et al., 2001). This is problematic
subsurface waters or in benthic mats (Mohamed et al., 2006). considering cell contamination can cause undesirable effects on
Potentially toxic cyanobacteria have also recently been reported in drinking water quality (e.g., unpleasant taste/smell through
underground and/or bottled water (Gkelis and Vlamis, 2017). geosmin and methylisoborneol) (Suurnäkki et al., 2015; Graham
Limnothrix redekei has also been reported to survive in the dark, if et al., 2010); and it is of particular concern because of the threat of
there is a carbon source (Daniels, 2016) or if it is associated with a toxin contamination in waters intended for human consumption.
heterotrophic bacteria (Meffert, 1989). There have been no Artificial bank filtration is a filtration method whereby the
reported deaths from consumption by healthy individuals of water from a natural river or lake is drawn through a bank that has
water containing cyanotoxins, however, some deaths have been been purposefully positioned, using either local or imported
 A.K. Rose et al. / Harmful Algae 76 (2018) 1–10 3

materials. Artificial bank filtration thus lends itself well to which makes this study site unique, with the sections upstream of
applications where avoidance of cyanobacterial cells and/or toxins the lagoon being in near pristine condition (Rose et al., 2017).
is required. For example, to avoid excessive cyanobacterial cells Three study sites were investigated; the natural lagoon from
entering the drinking water treatment facility, water is often which the drinking water is drawn; the raw water pipe leading
drawn from areas in the water body where cyanobacterial cells are from the lagoon; and the pre-chlorinated section of the pipe just
low and/or modifications to the intake areas are made so as to before the waters enters the treatment plant (Fig. 1).
reduce the number of these cells in the water treatment facility
(Stitz et al., 2013). There is, however, a paucity of information in the
2.1. Site descriptions
literature with regards to the efficiency of artificial bank filtration
structures in addressing a range of water quality problems
2.1.1. Drinking water supply (natural lagoon)
including cyanobacterial contamination, particularly with variable
The lagoon has a surface area of approximately 75,000 m2, and
in-stream conditions. Another area of study where there is very
has an average depth of 1 m–2 m, extending to 3 m deep in some
little scientific information is the impact of bank damage or
areas. The surrounding land is used for low intensity dry land
‘washout’ on filtration capacity following extreme water flow
agricultural grazing primarily with beef cattle, with no develop-
events such as flooding.
ment occurring upstream of this site. During flood events the water
Given the previously described voids in research, combined
level can reach as high as 15 m above the base line water level.
with the knowledge that contamination events in small remote/
The drinking water supply is drawn from the lagoon which
regional supplies often occur during extreme events (NHMRC,
passes through artificial bank filtration materials (Fig. 1). The
2011), the present project focussed on the effectiveness of
artificial bank filtration is comprised of a combination of medium –
compromised artificial bank filtration following extreme water
coarse sand (0.25 mm–1 mm) and capped with rocks, ranging in
flooding events and infrastructure upgrades. The study focussed on
size between 25 and 75 cm, and all materials being sourced from a
the cyanobacterial risk to the water supply of a small rural
local quarry. Due to geographical constraints the bank was
Australian township. The scope of the study included sampling and
artificially constructed as a pre-filter to the drinking water
observations during an extreme weather event, which led to fast
treatment plant, to reduce biological material entering from the
flowing flood waters damaging the bank filtration and associated
lagoon. During floods, a water level increase of 4 m is enough to
infrastructure. This provided a novel opportunity to examine the
completely immerse the pump station and cause some scouring,
efficacy of the bank filtration system, under routine as well as
the rocks were incorporated to help reduce erosion following flood
extreme in-stream conditions.
events. The distance travelled through the bank filtration is
between 5 m and 15 m and the passage time is relatively fast, being
2. Methods
only minutes to pass from the lagoon to the pump well.
This study focussed on a subtropical unregulated catchment on
the Australian eastern seaboard. A small regional town in Central 2.1.2. Raw water pipe
Queensland currently sources its drinking water from the The outlet is 4.5 km distant from the lagoon intake, it is located
permanent lagoon, which is located at the base of the headwaters between the permanent lagoon and drinking water treatment
of a relatively unmodified ephemeral creek. Whilst seasonal plant. The pipe comprises sections of cast iron, with intervals of an
ephemerality is a common stream condition in this region, most asbestos product. The pipe was installed in the 1970s and has been
other catchments have experienced some degree of modification, in continuous operation for about 40 years.

Fig. 1. Schematic of the bank filtration and sampling sites, main illustration depicts the water supply creek cross-section; the inset represents the top view. Each section is
colour coded to compare with Fig. 2. Green spots are areas which viable cultures were collected, red spot no viable cultures produced. Picture not to exact scale (sizing is
approximate for illustrative purposes). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Source: Author’s own.
4 A.K. Rose et al. / Harmful Algae 76 (2018) 1–10

2.1.3. Pre-chlorinated pipe 2.2.3. Cyanobacteria sampling

This sampling point was situated 6 km from the lagoon intake at Algal samples were collected at monthly intervals from
the drinking water treatment facility. The water consists of 90%– September 2012 to December 2014. Three samples of 1 L each
100% raw water from the lagoon and up to 10% bore water, sourced were pooled from each of the lagoon and supply chain sites, from
from a bore close to the drinking water treatment plant. The raw which duplicate 50 mL water subsamples were collected. Care was
water from the lagoon is mixed for a number of reasons, including taken to use techniques which followed the protocols of the
to reduce the volume of water drawn from the lagoon, as well as to Queensland Government Sampling Procedures for Drinking
add minerals to the drinking water. The water is pre-chlorinated Waters (QGOV, 2009) and Hötzel and Croome (1999). Half the
with an average residual of 1.04 mg/L free chlorine and 1.35 mg/L samples were preserved using Lugol’s iodine prepared as described
total chlorine before entering the treatment plant. in Hötzel and Croome (1999), while the remaining half were left
unpreserved for cell identification and culturing.
2.2. Field methods
2.3. Laboratory methods
2.2.1. Physical variables
2.3.1. Enumeration and identification
Physical and chemical water quality variables for the lagoon
Enumeration of algae was undertaken with the preserved
were recorded at monthly intervals between September 2012 and
samples using the Sedgwick rafter counting method of Hötzel and
December 2014. The suite of water quality variables included
Croome (1999). A minimum of 30 fields were counted to give 95%
temperature ( C), conductivity (mS/cm), total dissolved solids (mg/
confidence in the record of species present (McAlice, 1971) and
L), salinity (ppt), pH, oxidation reduction potential (mV), turbidity
where possible, a minimum of 100 cells were counted to provide a
(NTU) and dissolved oxygen (% saturation). A minimum of three
maximum counting error of 20% (Lund et al., 1958). Species
readings were recorded on each occasion and the average value
identifications were made using a number of sources (Anagnos-
was used. A YSI 6600 multiprobe sonde was used to measure water
tidis and Komárek, 1988; Gkelis et al., 2005; Bernard et al., 2011;
quality; this instrument was serviced and calibrated in the
Komárek et al., 2014).
laboratory prior to any field research according to the manufac-
On occasion individual cells were hard to differentiate, due to
turer’s recommendations (YSI, 2009).
inconspicuous cell walls. To overcome this problem the micro-
scopes light source was altered in intensity and using different
2.2.2. Chemical variables phases allowed for the cell wall to be identified. Morphological and
Water chemistry samples from the lagoon and drinking water ecological behaviours in the field and laboratory were compared to
supply network were collected quarterly between December 2012 genetically confirmed Limnothrix redekei specimens from the
and December 2014. These included total and dissolved metals, existing collection at CQUniversity (Fabbro et al., 2010).
nutrients, salts, disinfection by-products and other inorganic
analytes. Water samples were collected and stored on ice according 2.3.2. Applied culture method and photomicrography
to the protocols of the Queensland Government’s ‘Sampling In the laboratory, the unpreserved samples were mixed gently
Procedures for Drinking Waters’ (QGOV, 2009). The samples were and aliquots of 1 mL were placed on a sterile Petri dish, using a
then sent to a NATA approved laboratory for analyses (Australian sterile pipette. The Petri dish was covered and sealed using Para
Laboratory Services Pty Ltd, Gladstone). film and set aside for a minimum of 1 day either on a window sill or

Fig. 2. Co-occurrence of rainfall (mm) and the concentration of Limnothrix redekei cells between September 2012 and December 2014, in the drinking water supply lagoon,
raw water and pre-chlorinated pipes. The large black arrow indicates maintenance/upgrade on the pump well infrastructure and the bank filtration.
 A.K. Rose et al. / Harmful Algae 76 (2018) 1–10 5

in a growth chamber set at 25  C with a 16:8 h light: dark 3.3. Limnothrix redekei
photoperiod. The Petri dishes were then rinsed thoroughly with
reverse osmosis water and inspected using a microscope at 400 Limnothrix redekei cell concentrations were detected in all three
magnification. To assess suitable culturing media, samples with sites of the drinking water supply chain (Fig. 2). Average cell
remaining Limnothrix redekei attached to the Petri dish had either concentrations were similar in the raw and pre-chlorinated
ASM-1 (Gorham et al., 1964), autoclaved lagoon water or distilled sections of the supply chain pipe, whilst concentrations in the
water added and these were then returned to the window sill or lagoon were on average an order of magnitude lower (Table 2).
growth chamber. After a minimum of 1 month, Limnothrix redekei There was no correlation between Limnothrix redekei and water
was successfully subcultured using traditional techniques de- quality variables, suggesting a tolerance to a range of conditions.
scribed in Bernard et al. (2011).
3.4. Water quality variables and nutrients
2.4. Statistical analyses
Nutrient concentrations throughout the study were, on average,
Statistical analyses were completed using PRIMER-E7 software. less than the Queensland Water Quality Guidelines (Table 2)
Analysis of similarities (ANOSIM) was used to compare variation (Department of Environment and Heritage Protection, 2009). The
between sites. To reduce artefacts, any variables having values less findings from an ANOSIM provided data indicating that nutrient
than the limits of reporting (LOR) for the majority of the study were concentrations at the three sites were similar on average
omitted from statistical analyses. throughout the study (Global R value = 0.079, p = 0.093). There
was a gradual increase in nutrient concentrations as water flowed
3. Results through the supply chain, likely due to groundwater inputs. TKN is
nitrogen bound in organic substances, unlike the increase in TP and
3.1. Rainfall TRP, TKN was on average lower in the pre-chlorinated section. The
p value was not significant, however, this may be related to the
Three significant rain events occurred during the study, in small number of replicates. The R value is a scaled measure of the
January 2013, and March and December 2014 (Fig. 2). As a result of separation between groups, and can still be used to interpret
the January 2013 rain event the lagoon was completely flushed and relative values. The water quality variables (hardness, Cl , Ca2+, Na+
artificial bank compromised, whereas the rain events in March and and pH) at each site were similar, with a Global R value of 0.044
December 2014 served only to fill the lagoon with water as a result (p = 0.19) (Table 2). There were gradual improvements in selected
of draining from the upstream sections of the creek. The water water quality variables (hardness, Cl , Ca2+, Na+ and pH) as the
level during the flood in January 2013 exceeded the height of the water flowed through each section of the supply chain.
pump well. During this time, water was, therefore, also able to
enter the pump well without passing through the artificial bank 3.5. Field and laboratory observations
filtration.
3.5.1. Culture
3.2. Infrastructure maintenance It proved readily possible to isolate and culture Limnothrix
redekei from samples collected from both the raw water pipe and
Following the January 2013 flood, the pump well and pre-chlorinated pipe. It was much easier to culture sub-samples
surrounding bank were compromised, with sections of the bank grown from the pre-chlorinated pipe because the chlorine
surrounding the pump well being washed away. In May 2013, the treatment had been effective in eliminating many other species,
pump well and associated piping were replaced and the bank or reducing these to non-detectable numbers. By contrast, it was
reconstructed to surround the pump well. During this process, field not possible to successfully culture the species from the lagoon
observations showed that many rocks with living Limnothrix waters. Cultures were also able to be grown from samples taken
redekei biofilms which had previously been situated on the top from the sands of the artificial bank filtration materials as well as
surface layer of the artificial bank were relocated to the bottom scrape samples collected from mats growing on the rocks. There
layer of the pump well. was no preferred medium for culturing these cells, with viable

Table 2
Descriptive statistics for Limnothrix redekei cell counts and general water quality variables from Baffle Creek between September 2012 and December 2014.

Water Quality Variable Lagoon Raw Pre-Chlorinated

Range Average (SD) Range Average (SD) Range Average (SD)

Limnothrix redekei (cells/mL) 0–1050 126 (246) 0–15,400 2107 (3596) 0–13825 1896 (3644)
NH3 (mg/L) <LOR–0.024 0.010 (0.006) <LOR–0.230 0.069 (0.074) <LOR–0.133 0.025 (0.042)
NO2 (mg/L) <LOR <LOR <LOR <LOR <LOR <LOR
NO3 (mg/L) <LOR–0.487 0.083 (0.153) <LOR–0.443 0.091 (0.143) <LOR–0.709 0.204 (0.266)
TKN (mg/L) 0.200–0.400 0.256 (0.073) 0.200–0.400 0.250 (0.076) 0.050–0.300 0.156 (0.082)
TN (mg/L) 0.200–0.400 0.270 (0.073) 0.200–0.400 0.263 (0.074) 0.050–0.300 0.202 (0.105)
TP (mg/L) <LOR–0.060 0.021 (0.017) 0.010–0.050 0.026 (0.014) <LOR–0.110 0.039 (0.005)
TRP (mg/L) <LOR <LOR <LOR–0.020 0.010 (0.007) <LOR–0.040 0.011 (0.012)
Hardness (mg/L) <LOR–60.0 30.4 (20.2) <LOR–74.0 36.2 (23.4) <LOR–117.0 59.8 (35.2)
Cl (mg/L) 7.0–22.0 11.6 (6.1) 13.0–97.0 50.3 (27.5) 28.0–102.0 57.1 (24.9)
Ca2+ (mg/L) 1.0–9.0 4.1 (2.5) 4.0–13.0 7.3 (3.6) 8.0–27.0 16.0 (6.3)
Na+ (mg/L) 16.0–55.0 34.0 (12.7) 22.0–55.0 36.1 (10.7) 36.0–66.0 47.4 (9.3)
pH (mg/L) 6.4–8.1 6.9 (0.5) 6.4–7.3 7.0 (0.3) 6.6–8.2 7.1 (0.5)

LOR = limit of reporting; SD = standard deviation.

6 A.K. Rose et al. / Harmful Algae 76 (2018) 1–10

cultures developing with all three media types. Whilst growth infrastructure, specifically the newly installed pump well. The
rates were not specifically measured during culturing, cells in the old pump well was excavated and a new pump well installed, and
autoclaved lagoon water were the first to be visibly detected, during this time period Limnothrix redekei and many other species
possibly due to the Limnothrix redekei cultures being more rapidly of algae and bacteria were observed attached to the rocks and sand
acclimated to the lagoon water (e.g. nutrient, conductivity, pH). surrounding the pump well. After the new pump well was
installed, the biologically active sand and rocks were placed around
3.5.2. Morphological differences of Limnothrix redekei the pump well. This is the most probable time for large numbers of
The presence of central polar gas-vacuoles in Limnothrix redekei Limnothrix redekei to pass into the water supply, considering it is
samples was dependant on the place these were sampled. possible for Limnothrix redekei to survive and even thrive in
Filaments from the surface of the lagoon and culture displayed conditions where sunlight is minimal, and the motile nature of the
very conspicuous gas-vacuoles, whilst filaments from the supply species (Rücker et al., 1997; Nicklisch et al., 2007).
chain or scrape samples collected from biofilms lacked central The large variation in cell numbers between sites may have
polar gas-vacuoles (Fig. 3). resulted from sections of the Limnothrix redekei biofilm sloughing
Limnothrix redekei cultures when disturbed, in this case shaken off the pump well in small portions. It is also conceivable that
vigorously manually for 5 min, firstly distribute throughout the Limnothrix was growing in the pipes, however it was not possible to
culture vessel and after 10 min pull together to form a colony like accurately determine if this occurred. The only field evidence for
structure (Fig. 4). The specimen photographed in Fig. 5 illustrates this possibility is that samples taken from a section of the replaced
what the colony from Fig. 4 looks like in the field. pipe section did result in a viable culture, however the point of
origin of the Limnothrix redekei population may have been caused
4. Discussion from isolated cells in the water column, rather than from a
population that was attached or growing at this site.
4.1. Potential penetrative pathways
4.2. Proliferation through ecological adaptions
There are three potential pathways that may have allowed for
Limnothrix redekei to breakthrough or penetrate the artificial bank Limnothrix redekei was initially described as a cyanobacterial
filtration materials. First, during the extreme event of January species common in turbulent waters (Dokulil and Teubner, 2000).
2013, the depth of the flood waters increased above the supply From laboratory observations during the present study and others
chain infrastructure, thus allowing waters to bypass the bank (Bernard et al., 2011), however, it is revealed that Limnothrix redekei
filtration and enter the drinking water supply chain directly may be present in the benthos or the plankton. Planktonic material
through the pump well opening. The second potential break- has the potential to disperse and form new benthic mats. Evidence
through may have occurred during the same extreme event, for this was observed when disturbed cultures of Limnothrix
however, instead of entering through the pump well opening it is redekei featured the cells assemble together to form a ball-like
possible these cells entered by passing through the artificial bank structure (Figs. 4 and 5), a phenomenon that has also been
filtration materials. The hydraulic pressure and damage sustained described in the field by Bernard et al. (2011). When cultured in the
by the flood waters compromised the bank and may have made the laboratory, Limnothrix redekei preferentially grows on all surfaces
passage of Limnothrix redekei through the bank possible. It is clear (Bernard et al., 2011). When all surfaces are covered it is, however,
that Limnothrix redekei was present in the water supply following proposed that vacuolation increases and cells float to the top
the major flood event, albeit cell concentrations were not in great where these combine (Fig. 5).
numbers (Fig. 2). Clumping together following disturbance (Figs. 4 and 5) or
The third potential pathway for cyanobacteria passage into the when expelled by the benthic mat has ecological benefits through
drinking water supply may have occurred during the subsequent increasing the probability that another viable colony will develop.
maintenance and upgrade of the drinking water supply Furthermore, Limnothrix redekei ecological clumping behaviours

Fig. 3. Limnothrix redekei with gas-vacuoles. Sample A from the lagoon, B and C from culture surface. Limnothrix redekei without gas-vacuoles. Sample E was from the pre-
chlorinated pipe, D and F were from the raw water pipe.
 A.K. Rose et al. / Harmful Algae 76 (2018) 1–10 7

Fig. 4. Limnothrix redekei clumping behaviour, A (0 min), B (2 min), C, (4 min), D (6 min), E (8 min) and F (10 min). Minutes following disturbance.

may protect some cells from chlorination, as it might be expected (1982) and Foy et al. (1976), who noted that photosynthetic
that only cells present on the exterior of the clumping ball would pigment reduction is an adaptation to greater amounts of sunlight.
be exposed to chlorine. Subsequently, filaments that are located By comparison, visual inspection (but not detailed pigment
internally may avoid the chlorination treatment, due to the lack of analyses) suggested that the same cells that were detected at
direct contact; and thus may survive to potentially contaminate sites where there was less sunlight (e.g. Pump Well), had greater
surfaces in the drinking water treatment facilities. If the organism pigmentation, again consistent with the findings of Foy and Gibson
passes through into the drinking water treatment plant infra- (1982). The ecological reason for the increase and decrease in
structure, light may be extremely variable (e.g. open aeration photosynthetically active pigments is considered a practical
tanks) and Limnothrix redekei would have to adapt to changing light adaptation to a range of light conditions, giving Limnothrix redekei
regimes. a competitive advantage as compared with other cyanobacteria
The Limnothrix redekei detected on rocks exposed to a greater and algae that do not have the ability to modify photosynthetic
amounts of sunlight appeared to have had photosynthetic pigment content. Limnothrix redekei may possess the ability to
pigments that were dull and bleached with a yellow-brown survive dark conditions, with studies showing that if it has a carbon
colouring. This is similar to that described by Foy and Gibson source (Daniels, 2016) or it is associated with heterotrophic
bacteria (Meffert, 1989) it will survive.

4.3. Identification using microscopy aided by a simple culturing

technique

Limnothrix redekei was first classed in the Order Oscillatoriales,

however, more recently taxonomic revision places it within Order
Synechococcales and Family Pseudanabaenaceae (Komárek et al.,
2014). Important features originally associated with the morpho-
logical identification of Limnothrix redekei were the presence of
parietal thylakoids, sheath absent, regular filaments and gas-
vacuoles (Meffert, 1989; Gkelis et al., 2005; Komárek et al., 2014). If
the gas-vacuoles are absent, however, the task of identification
becomes much more difficult (Bernard et al., 2011). The quality of
the laboratory microscope is also very important for identification:
a microscope with phase contrast is preferred, due to the gas
vacuoles being more easily identified. During the present study, it
Fig. 5. Limnothrix redekei floating ball structure in a subcultured field specimen. was observed that the central polar gas-vacuoles were only
8 A.K. Rose et al. / Harmful Algae 76 (2018) 1–10

sometimes present (e.g. Fig. 3). The gas-vacuoles were only present the reality that some laboratories rely on dated microscope
in samples collected from the surface of the lagoon waters and in infrastructure which lacks the capacity to identify specific
the surface of laboratory cultures; whereas gas-vacuoles were not microbial features, clearly limit the opportunities for effectively
detected in filaments found in the pipes (both chlorinated and identifying whether Limnothrix redekei are present in samples. A
unchlorinated) or on the benthic mats present on the artificial bank combination of these three factors may help explain why
filtration. Fortunately, the culturing of the population provided Limnothrix redekei has not been as readily identified in Australian
some assistance in the morphological identification of Limnothrix water bodies as it has elsewhere. As such, culturing techniques are
redekei. important in aiding an accurate identification.
The culture technique generated in the present study was based Use of the culture method developed in the present study may
on reflecting the Limnothrix redekei ecology observed in the field help simplify the identification process of Limnothrix redekei,
and laboratory. When grown on a window sill or growth cabinet, which in turn can help in informing managers if the cyanobacteri-
filaments present in the sample attached to the Petri dish where um is present. Understanding the ecological behaviours of
these cells remained even after being rinsed systematically. After Limnothrix redekei also provides valuable information that can
all other cells present in the sample were washed away, a be used to monitor for the species as well as facilitate
monospecific culture was produced. The benefit of this ‘low implementation of strategies in avoiding future contamination
technology’ culture method is that it does not require specialised in the drinking water supply infrastructure. Extreme events similar
infrastructure or the skilled expertise required for conducting to those experienced during the present study are predicted to
other techniques. It is also a very cost effective method that can be increase in frequency in the future, and during these events the
conducted with water sourced at the site of the water supply or possibilities of contamination are increased. To minimise risk in
with distilled water. If budget is no constraint, culturing can also be the future, the proposed maintenance/upgrade protocol is
achieved using the ASM-1 media used for many other cyanobac- recommended in conjunction with field monitoring to include
terial species (Gorham et al., 1964). The method can be used to any pump wells.
easily ascertain if there are Limnothrix redekei present near or To develop an enhanced understanding of the toxicity of
around any water supply intake points, thus making risk assess- Limnothrix redekei and its ecological behaviours, additional
ments more straightforward. experimentation is required. Future studies should include the
identification of the toxin, toxicity tests and chlorination experi-
4.4. Proposed protocol for preventing penetration ments to establish the most effective dosing rates. More research is
required to ascertain whether Limnothrix redekei is able to continue
Considering the analytical results and observations in the to grow in other sections of drinking water supply chains including
present study, a change to drinking water supply infrastructure the drinking water treatment plant, storage reservoirs and supply
maintenance protocols is proposed. The change to protocols would chain pipes. The ability of cyanobacterial species to adapt to every
be to use clean sands and rocks to cover the subsurface layers of the ecological niche is very well documented, therefore, it is not
pump well. The older biologically active sands and rocks that are surprising that Limnothrix redekei are able to survive and even
already on site should be used to cap the subsurface layers. This has thrive in manmade structures. This leads to an emerging need to
benefits in that it increases the competition by other commensal identify, prevent and manage Limnothrix redekei in regional water
microbes and thus inhibits or limits the rate of development of supplies, including monitoring that is focused on presence within
populations of Limnothrix redekei, so this species would be less structures of the treatment plants.
likely to dominate the biota. The biologically active layer also has
benefits in cyanotoxin removal (Grützmacher et al., 2009; Chorus
et al., 2003; Miller et al., 2001; Dillon et al., 2002; Chorus and Acknowledgements
Bartram, 1999), reduction of metals (Freitas et al., 2012; Lorenzen
et al., 2010), reduction or elimination of pathogenic bacteria (Weiss Appreciation is given to James Kinder, Amie Anastasi, Neville
et al., 2005; Ray et al., 2003; Thakur et al., 2013; Wang, 2002; Dash Doyle and Christopher O’Neill for their constructive comments.
et al., 2010; Shamrukh and Abdel-Wahab, 2008) and water quality Special thanks go to Kevin Jeffery for assistance with data
variable adjustment (Freitas et al., 2012). Biofilms grown in collection, and Anthony Kodel for his magnificent photographic
conjunction with cyanobacterial populations are more efficient at skills. Finally thank you to Gladstone Regional Council for funding
removing cyanotoxins (Babica et al., 2005), which makes the the research. The Authors declare they have no actual or potential
decision to cap with existing pre-conditioned soil a logical option competing financial interests. [CG]
in water supply chains.
References
5. Conclusion
Anagnostidis, K., Komárek, J., 1988. Modern approach to the classification system of
cyanophytes. 3—oscillatoriales. Arch. Hydrobiol. Suppl. 80, 327–472.
Limnothrix redekei, historically, has been a problematic species Babica, P., Blaha, L., Marsalek, B., 2005. Removal of microcystins by phototrophic
throughout Europe, causing many difficulties for water treatment biofilms: a microcosm study (6 pp). Environ. Sci. Pollut. Res. 12 (6), 369–374.
Bernard, C., Froscio, S., Campbell, R., Monis, P., Humpage, A., Fabbro, L., 2011. Novel
managers (Rücker et al., 1997; Köhler and Hoeg, 2000; Teubner, toxic effects associated with a tropical Limnothrix/Geitlerinema-like cyano-
2000). Recently Limnothrix redekei has been reported throughout bacterium. Environ. Toxicol. 26 (3), 260–270.
Australia, in particular Central Queensland (Bernard et al., 2011; Carmichael, W.W., Azevedo, S., An, J.S., Molica, R., Jochimsen, E.M., Lau, S., Rinehart,
K.L., Shaw, G.R., Eaglesham, G.K., 2001. Human fatalities from cyanobacteria:
Stitz et al., 2013). Due to the observed morphological variability of chemical and biological evidence for cyanotoxins. Environ. Health Perspect. 109
Limnothrix redekei, it is possible that Limnothrix redekei is present in (7), 663.
many other regions of Australia. The lack of central, polar gas Carmichael, W.W., 2001. Health effects of toxin-producing cyanobacteria: “the
CyanoHABs”. Hum. Ecol. Risk Assess.: Int. J. 7 (5), 1393–1407.
vacuoles has, however, made the identification of these microbes Chorus, I., Bartram, J., 1999. Toxic Cyanobacteria in Water: A Guide to Their Public
complicated, which could possibly result in under-reporting or Health Consequences, Monitoring and Management. E & FN Spon, London.
misreporting. Furthermore, morphological differences aside, the Chorus, I., Bartel, H., Grützmacher, G., Böttcher, G., 2003. Removal of algae and
cyanobacterial toxins during slow sand and bank filtration. Conference Wasser
ecological niche in Australia also appears different to that known
Berlin 51–57.
from Europe, with the species exhibiting benthic rather than Codd, G.A., Morrison, L.F., Metcalf, J.S., 2005. Cyanobacterial toxins: risk manage-
planktonic ecological niches. These considerations, combined with ment for health protection. Toxicol. Appl. Pharmacol. 203 (3), 264–272.
 A.K. Rose et al. / Harmful Algae 76 (2018) 1–10 9

Daniels, O., Fabbro, L., Makiela, S., 2014. The effects of the toxic cyanobacterium and death after exposure to microcystins at a hemodialysis center in Brazil. N.
Limnothrix (strain AC0243) on Bufo marinus larvae. Toxins 6 (3), 1021–1035. Engl. J. Med. 338 (13), 873–878.
Daniels, O., 2016. Autecology, Allelopathy and Toxicity of Limnothrix (strain Köhler, J., Hoeg, S., 2000. Phytoplankton selection in a river—lake system during two
AC0243): Multiple-Organism Studies Using Laboratory Cultures (Unpublished decades of changing nutrient supply. The Trophic Spectrum Revisited. Springer,
Doctoral Dissertation), Doctorate of Philosophy. CQUniversity. pp. 13–24.
Dash, R., Mehrotra, I., Kumar, P., Grischek, T., 2008. Lake bank filtration at Nainital, Komárek, J., Kastovsky, J., Mares, J., Johansen, J.R., 2014. Taxonomic classification of
India: water-quality evaluation. Hydrol. J. 16 (6), 1089–1099. cyanoprokaryotes (cyanobacterial genera) 2014, using a polyphasic approach.
Dash, R.R., Prakash, E.B., Kumar, P., Mehrotra, I., Sandhu, C., Grischek, T., 2010. River Preslia 86 (4), 295–335.
bank filtration in Haridwar, India: removal of turbidity, organics and bacteria. Lahti, K., Rapala, J., Kivimäki, A., Kukkonen, J., Niemelä, M., Sivonen, K., 2001.
Hydrol. J. 18 (4), 973–983. Occurrence of microcystins in raw water sources and treated drinking water of
Department of Environment and Heritage Protection, 2009. Queensland Water Finnish waterworks. Water Sci. Technol. 43 (12), 225–228.
Quality Guidelines, Version 3. ISBN 978-0-9806986-0-2. Lorenzen, G., Sprenger, C., Taute, T., Pekdeger, A., Mittal, A., Massmann, G., 2010.
Dillon, P., Miller, M., Fallowfield, H., Hutson, J., 2002. The potential of riverbank Assessment of the potential for bank filtration in a water-stressed megacity
filtration for drinking water supplies in relation to microsystin removal in (Delhi, India). Environ. Earth Sci. 61 (7), 1419–1434.
brackish aquifers. J. Hydrol. 266 (3), 209–221. Lund, J., Kipling, C., Le Cren, E., 1958. The inverted microscope method of estimating
Dokulil, M.T., Teubner, K., 2000. Cyanobacterial dominance in lakes. Hydrobiologia algal numbers and the statistical basis of estimations by counting. Hydro-
438 (1), 1–12. biologia 11 (2), 143–170.
Fabbro, L.D., Bernard, C., Monis, P.T., 2010. Improved morphometric and genetic McAlice, B.J., 1971. Phytoplankton sampling with the Sedgwick-Rafter cell. Limnol.
tools for better identification and management of blue green algae. The Oceanogr. 19–28.
Australian Coal Industry’s Research Program (ACARP), Environmental Reports. Meffert, M.-E., 1989. Planktic unsheated filaments (Cyanophyceae) with polar and
CQUniversity and South Australia Water Corperation. central gas vacuoles II: biology, population dynamics and biotopes of
Foy, R., Gibson, C., 1982. Photosynthetic characteristics of planktonic blue-green Limnothrix redekei (Van Goor) Meffert. Archiv für Hydrobiologie 116 (3),
algae: changes in photosynthetic capacity and pigmentation of Oscillatoria 257–282.
redekei Van Goor under high and low light. Br. Phycol. J. 17 (2), 183–193. Metge, D., Harvey, R., Aiken, G., Anders, R., Lincoln, G., Jasperse, J., 2010. Influence of
Foy, R., Gibson, C., Smith, R., 1976. The influence of daylength, light intensity and organic carbon loading, sediment associated metal oxide content and sediment
temperature on the growth rates of planktonic blue-green algae. Br. Phycol. J. 11 grain size distributions upon Cryptosporidium parvum removal during
(2), 151–163. riverbank filtration operations, Sonoma County, CA. Water Res. 44 (4), 1126–
Freitas, D., Cabral, J., Paiva, A., Molica, R., 2012. Application of bank filtration 1137.
technology for water quality improvement in a warm climate: a case study at Miller, M., Critchley, M., Hutson, J., Fallowfield, H., 2001. The adsorption of
Beberibe River in Brazil. J. Water Supply: Res. Technol.-Aqua 61 (5), 319–330. cyanobacterial hepatotoxins from water onto soil during batch experiments.
Gaget, V., Humpage, A.R., Huang, Q., Monis, P., Brookes, J.D., 2017. Benthic Water Res. 35 (6), 1461–1468.
cyanobacteria: a source of cylindrospermopsin and microcystin in Australian Mohamed, Z.A., El-Sharouny, H.M., Ali, W.S., 2006. Microcystin production in
drinking water reservoirs. Water Res. 124, 454–464. benthic mats of cyanobacteria in the Nile River and irrigation canals, Egypt.
Gkelis, S., Vlamis, A., 2017. Can cyanobacteria infect underground water sources? Toxicon 47 (5), 584–590.
Evidence from a small scale monitoring of natural mineral water drinking NHMRC, 2011. Australian Drinking Water Guidelines Paper 6 National Water Quality
source. Adv. Oceanogr. Limnol. 8 (1) . Management Strategy. National Health and Medical Research Council, National
Gkelis, S., Rajaniemi, P., Vardaka, E., Moustaka-Gouni, M., Lanaras, T., Sivonen, K., Resource Management Ministerial Council. Commonwealth of Australia,
2005. Limnothrix redekei (Van Goor) Meffert (Cyanobacteria) strains from Lake Canberra.
Kastoria, Greece form a separate phylogenetic group. Microb. Ecol. 49 (1), Nicklisch, A., Shatwell, T., Köhler, J., 2007. Analysis and modelling of the interactive
176–182. effects of temperature and light on phytoplankton growth and relevance for the
Gorham, P., McLachlan, J., Hammer, U., Kim, W., 1964. Isolation and culture of toxic spring bloom. J. Plankton Res. 30 (1), 75–91.
strains of Anabaena flos-aquae (Lyngb.) de Breb. Int. Verein Theor. Angew Pazouki, P., Prévost, M., McQuaid, N., Barbeau, B., de Boutray, M.-L., Zamyadi, A.,
Limnol. 15, 796–804. Dorner, S., 2016. Breakthrough of cyanobacteria in bank filtration. Water Res.
Grünheid, S., Amy, G., Jekel, M., 2005. Removal of bulk dissolved organic carbon 102, 170–179.
(DOC) and trace organic compounds by bank filtration and artificial recharge. QGOV, 2009. Sampling Procedures for Drinking Waters. Queensland Health. http://
Water Res. 39 (14), 3219–3228. www.health.qld.gov.au/publichealthact/documents/sample_drinkwater_pr.
Graham, J.L., Loftin, K.A., Meyer, M.T., Ziegler, A.C., 2010. Cyanotoxin mixtures and pdf.
taste-and-odor compounds in cyanobacterial blooms from the Midwestern Rücker, J., Wiedner, C., Zippel, P., 1997. Factors controlling the dominance of
United States. Environ. Sci. Technol. 44 (19), 7361–7368. Planktothrix agardhii and Limnothrix redekei in eutrophic shallow lakes.
Gross-Wittke, A., Gunkel, G., Hoffmann, A., 2010. Temperature effects on bank Hydrobiologia 342, 107–115.
filtration: redox conditions and physical-chemical parameters of pore water at Rachman, R.M., Li, S., Missimer, T.M., 2014. SWRO feed water quality improvement
Lake Tegel, Berlin, Germany. J. Water Clim. Change 1 (1), 55–66. using subsurface intakes in Oman, Spain, Turks and Caicos Islands, and Saudi
Grützmacher, G., Wessel, G., Klitzke, S., Chorus, I., 2009. Microcystin elimination Arabia. Desalination 351, 88–100.
during sediment contact. Environ. Sci. Technol. 44 (2), 657–662. Ray, C., Melin, G., Linsky, R.B., 2003. Riverbank Filtration: Improving Source-Water
Gunkel, G., Beulker, C., Hoffmann, A., Kosmol, J., 2009. Fine particulate organic Quality, vol. 43. Springer Science & Business Media.
matter (FPOM) transport and processing in littoral interstices–use of Romero, L., Mondardo, R., Sens, M., Grischek, T., 2014. Removal of cyanobacteria and
fluorescent markers. Limnol.–Ecol. Manage. Inland Waters 39 (3), 185–199. cyanotoxins during lake bank filtration at Lagoa do Peri, Brazil. Clean Technol.
Hötzel, G., Croome, R., 1999. A Phytoplankton Methods Manual for Australian Environ. Policy 16 (6), 1133–1143.
Freshwaters. Land and Water Resources Research and Development Corpora- Rose, A.K., Fabbro, L., Kinnear, S., 2017. Hydrogeochemistry in a relatively
tion, Canberra. unmodified subtropical catchment: insights regarding the health and aesthetic
Harvey, R.W., Metge, D.W., LeBlanc, D.R., Underwood, J., Aiken, G.R., Butler, K., risks of manganese. J. Hydrol.: Reg. Stud. 13, 152–167.
McCobb, T.D., Jasperse, J., 2015. Importance of the colmation layer in the Sanseverino, I., Conduto, D., Pozzoli, L., Dobricic, S., Lettieri, T., 2016. Algal Bloom and
transport and removal of cyanobacteria, viruses, and dissolved organic carbon Its Economic Impact. EUR 27905 EN; 102788 660478.
during natural lake-bank filtration. J. Environ. Qual. 44 (5), 1413–1423. Schmidt, C.K., Lange, F., Brauch, H., Kühn, W., 2003. Experiences with Riverbank
Heberer, T., Mechlinski, A., Fanck, B., Knappe, A., Massmann, G., Pekdeger, A., Fritz, B., Filtration and Infiltration in Germany. DVGW-Water Technology Center (TZW),
2004. Field studies on the fate and transport of pharmaceutical residues in bank Karlsruhe, Germany, pp. 17.
filtration. Groundwater Monit. Remediat. 24 (2), 70–77. Shamrukh, M., Abdel-Wahab, A., 2008. Riverbank filtration for sustainable water
Hoffmann, A., Gunkel, G., 2011a. Bank filtration in the sandy littoral zone of Lake supply: application to a large-scale facility on the Nile River. Clean Technol.
Tegel (Berlin): structure and dynamics of the biological active filter zone and Environ. Policy 10 (4), 351–358.
clogging processes. Limnol.–Ecol. Manage. Inland Waters 41 (1), 10–19. Sharma, S., Amy, G., 2009. Bank filtration: a sustainable water treatment technology
Hoffmann, A., Gunkel, G., 2011b. Carbon input, production and turnover in the for developing countries. Water, Sanitation and Hygiene: Sustainable Develop-
interstices of a Lake Tegel bank filtration site, Berlin, Germany. Limnol.–Ecol. ment and Multisectoral Approaches. Proceedings of the 34th WEDC Interna-
Manage. Inland Waters 41 (3), 151–159. tional Conference, United Nations Conference Centre, Addis Ababa, Ethiopia,
Huelshoff, I., Greskowiak, J., Gruetzmacher, G., 2009a. Analysis of the Vulnerability 18–22 May 2009, Water, Engineering and Development Centre (WEDC)
of Bank Filtration Systems to Climate Change by Comparing Their Effectiveness Loughborough University of Technology, pp. 662–666.
Under Varying Environmental Conditions. . Sivonen, K., Jones, J., 1999. Cyanobacterial Toxins. Toxic Cyanobacteria in Water, A
Huelshoff, I., Greskowiak, J., Wiese, B., Grutzmacher, G., 2009b. Relevance and Guide to Their Public Health Consequenses, Monitoring and Management.
Opportunities of Bank Filtration to Provide Safe Water for Developing and World Health Organisation.
Newly Industrialised Countries. . Stauder, S., Stevanovic, Z., Richter, C., Milanovic, S., Tucovic, A., Petrovic, B., 2012.
Humpage, A., Falconer, I., Bernard, C., Froscio, S., Fabbro, L., 2012. Toxicity of the Evaluating bank filtration as an alternative to the current water supply from
cyanobacterium Limnothrix AC0243 to male Balb/c mice. Water Res. 46 (5), Deeper Aquifer: a case study from the Pannonian Basin, Serbia. Water Resour.
1576–1583. Manage. 26 (2), 581–594.
IPCC, 2014. Climate Change 2014–Impacts, Adaptation and Vulnerability: Regional Stitz, L., Kinnear, S., Fabbro, L., 2013. A role for aeration and intake depth in
Aspects. Cambridge University Press. managing toxic Cylindrospermopsis: a comparison between off-stream and
Jochimsen, E.M., Carmichael, W.W., An, J., Cardo, D.M., Cookson, S.T., Holmes, C.E., riverine environments in the Fitzroy Basin, Australia. Lakes Reserv.: Res.
Antunes, M.B., de Melo Filho, D.A., Lyra, T.M., Barreto, V.S.T., 1998. Liver failure Manage. 18 (2), 179–196.
10 A.K. Rose et al. / Harmful Algae 76 (2018) 1–10

Suurnäkki, S., Gomez-Saez, G.V., Rantala-Ylinen, A., Jokela, J., Fewer, D.P., Sivonen, K., Wang, J., 2002. Riverbank filtration case study at Louisville, Kentucky. Riverbank
2015. Identification of geosmin and 2-methylisoborneol in cyanobacteria and Filtration. Springer, pp. 117–145.
molecular detection methods for the producers of these compounds. Water Res. Weiss, W.J., Bouwer, E.J., Aboytes, R., LeChevallier, M.W., O’Melia, C.R., Le, B.T.,
68, 56–66. Schwab, K.J., 2005. Riverbank filtration for control of microorganisms: results
Teubner, K., 2000. Synchronised changes of planktonic cyanobacterial and diatom from field monitoring. Water Res. 39 (10), 1990–2001.
assemblages in North German waters reduce seasonality to two principal Wells, M.L., Trainer, V.L., Smayda, T.J., Karlson, B.S., Trick, C.G., Kudela, R.M.,
periods. Ergebnisse Der Limnologie 55, 565–580. Ishikawa, A., Bernard, S., Wulff, A., Anderson, D.M., 2015. Harmful algal blooms
Thakur, A., Ojha, C., 2010. Variation of turbidity during subsurface abstraction of and climate change: learning from the past and present to forecast the future.
river water: a case study. Int. J. Sediment Res. 25 (4), 355–365. Harmful Algae 49, 68–93.
Thakur, A., Ojha, C., Singh, V., Gurjar, B., Sandhu, C., 2013. Removal of pathogens by YSI, 2009. YSI 6-Series Multiparameter Water Quality Sondes User Manual. . www.
river bank filtration at Haridwar, India. Hydrol. Processes 27 (11), 1535–1542. ysi.com/media/pdfs/069300-YSI-6-Series-Manual-RevH.pdf.
Tufenkji, N., Ryan, J.N., Elimelech, M., 2002. Peer Reviewed: the Promise of Bank Zamyadi, A., MacLeod, S.L., Fan, Y., McQuaid, N., Dorner, S., Sauvé, S., Prévost, M.,
Filtration. ACS Publications. 2012. Toxic cyanobacterial breakthrough and accumulation in a drinking
WHO, 2003. Guidelines for Safe Recreational Water Environments. Volume 1: water plant: a monitoring and treatment challenge. Water Res. 46 (5),
Coastal and Fresh Waters. World Health Organization. 1511–1523.