Hydrobiologia (2006) 559:23–76 Ó Springer 2006
Eutrophication in Australian rivers, reservoirs and estuaries – a southern
hemisphere perspective on the science and its implications
J. Richard Davis1,3,* & Klaus Koop2
CSIRO Land and Water, ACT 2601, GPO Box 1666, Canberra, Australia
Department of Environment and Conservation, NSW 1232, P.O. Box A290, Sydney South, Australia
Present address: Land & Water Australia, GPO Box 2182, 2601, Canberra, ACT, Australia
(*Author for correspondence: E-mail: email@example.com)
Received 19 December 2005; accepted 18 March 2005
Key words: eutrophication, nutrients, nutrient limitation, phosphorus, nitrogen, fertiliser, diﬀuse sources, rivers,
estuaries, phytoplankton, akinetes, algal blooms, cyanobacteria, blue green algae, stratiﬁcation, sediments, sediment
nutrient ﬂuxes, bioavailability, residence time, biomanipulation, Murray-Darling Basin, Australia
Australian science has made rapid advances in the last decade in understanding eutrophication processes
in inland waters and estuaries. The freshwater research on which these advances are based was triggered
by well-publicised blooms of cyanobacteria during the 1980s and early 1990s, particularly a 1000 km
long bloom on the Darling River. In estuaries the study which greatly enhanced our understanding but
simultaneously served to stimulate further research into estuarine eutrophication, the Port Phillip Bay
Study, was initially designed to address perceived problems of toxicants in the Bay but provided
profound insights into drivers for, and ecosystem responses to, eutrophication. Subsequent estuarine
research has largely been stimulated by management questions arising from AustraliaÕs increasing
coastal development for residential purposes. The research has shown that some of the beliefs extant at
the time of the blooms were incorrect. For example, it is now clear that stratiﬁcation and light pen-
etration, not nutrient availability, are the triggers for blooms in the impounded rivers of southeastern
Australia, although nutrient exhaustion limits the biomass of blooms. Again, nitrogen seems to play as
important a role as phosphorus does in controlling the biomass of these freshwater blooms. The
research has also shown that aspects of eutrophication, such as nutrient transport, are dominated by
diﬀerent processes in diﬀerent parts of Australia. Many of the biophysical processes involved in
eutrophication have now been quantiﬁed suﬃciently for models to be developed of such processes as
sediment-nutrient release, stratiﬁcation, turbidity and algal growth in both freshwater and estuarine
systems. In some cases the models are reliable enough for the knowledge gained in particular water-
bodies to be applied elsewhere. Thus, there is now a ﬁrm scientiﬁc foundation for managers to rely
upon when managing algal blooms. Whilst these ﬁndings have already been presented to managers and
communities throughout Australia, there is still a considerable way to go before they are absorbed into
their modus operandi.
Introduction and coastal waters are added, including the costs to
ﬁsheries, aquaculture and tourism, the full cost of
Algal blooms in Australian freshwaters cost the algal blooms in Australia is much higher. Signiﬁ-
community between AUD180 and 240 million cantly, the report by Atech (2000) showed that
every year (Atech, 2000). When blooms in estuaries these costs are incurred by urban water users,
dryland farmers and irrigators in roughly equal lakes in Victoria), algal problems in large water
proportions. Thus, algal blooms are a problem storages (e.g., Cannon et al., 1970) including
that aﬀects the whole Australian community. serious illness following a cyanobacterial out-
Eutrophication is the process of increasing break in Solomon Dam (Hawkins et al., 1985)
organic enrichment of an ecosystem, where the and seasonal blooms in riverine impoundments in
organic enrichment causes changes to the ecosys- the Murray-Darling Basin (MDB). Algal blooms
tem (Nixon, 1995; see also National Research received widespread public attention with the
Council, 2000). This has usually been characterised 1991/92 bloom that covered a distance of some
by increased supplies of nutrients to that system 1000 km in the Barwon-Darling River system
but, as we shall argue below, in Australian fresh- (Bowling & Baker, 1996) and a major cocco-
water systems solar radiation and stratiﬁcation also lithophorid bloom in Jervis Bay in late 1992
play important roles. The organic enrichment fre- (Blackburn & Cresswell, 1993). The Barwon-
quently takes the form of phytoplankton blooms, Darling bloom caused losses conservatively esti-
both cyanobacterial and eukaryotic blooms, mated at AUD9m (NSW Blue-Green Algal
although in many Australian eutrophic estuaries it Taskforce, 1992).
manifest itself as blooms of ﬁlamentous macroal- The 1991 Barwon-Darling River bloom gal-
gae. As emphasised by the National Research vanised government action. Algal management
Council (2000), it is important to understand strategies were drawn up for Victoria (Victorian
clearly the cause and eﬀect relationship between Blue-Green Algal Project Team, 1995), New South
nutrients and their ecological eﬀects as confusion Wales (NSW Blue-Green Algal Task Force, 1992)
may ‘‘impede mitigation eﬀorts because proposed and the MDB (Anon, 1994b), and Queensland
changes may not bring about desired eﬀects’’. A formed a Water Quality Taskforce which reported
comprehensive knowledge of these cause and eﬀect on the managerial, operational and situational
relationships is vital to eﬀectively target appropri- aspects of algal blooms. These reports all recog-
ate management actions. nised that there was insuﬃcient understanding of
In a comprehensive review of eutrophication in the processes governing the triggering, formation
coastal ecosystems, Cloern (2001) pointed out that and decline of algal blooms under Australian
the traditional eutrophication model based on environmental conditions and called for research
nutrient inputs as the signal leading to enhanced to be carried out to clarify these processes.
phytoplankton production is increasingly being Two coordinated R&D programs (the CSIRO
replaced by emphasis on the study of system-spe- Blue-Green Algal Program and the National
ciﬁc attributes that act to modulate complex direct Eutrophication Management Program) were
and indirect ecosystem responses. In Australia commenced. Further site-speciﬁc projects were
eutrophication research and management atten- initiated to provide answers for management
tion has been focussed on algal blooms, particu- authorities faced with eutrophication problems in
larly toxic cyanobacterial blooms in fresh waters. estuarine and fresh waterbodies. These initiatives
Some work has been done on phytoplankton re- are described in more detail later.
sponses in coastal systems where mainly diatoms This paper assesses current understanding of
and dinoﬂagellates are involved. That bias is re- the eutrophication process in Australian fresh-
ﬂected in this review, although in recent years waters and estuaries following this research,
there is an increasing move, particularly in studies without undertaking a comprehensive interna-
of estuaries, to studying more complex system tional review. Ocean waters are not reviewed.
characteristics and interactions. The state of knowledge about eutrophication in
Blooms of cyanobacteria have been a recog- the early 1990s is used as a benchmark against
nised phenomenon in Australian fresh waters which the improvement in knowledge is mea-
since late in the 19th Century (Francis, 1878). sured. Some assumptions of that time have been
Throughout the 1970s and 80s there were per- conﬁrmed; some have been overthrown; while
sistent cyanobacterial blooms in the Peel-Harvey others remain unresolved. Finally, some of the
estuary in Western Australia (EPA, 1988), occa- broader management implications of this new
sional blooms in other estuaries (e.g., Gippsland understanding are discussed.
The Australian environment material and this is further exacerbated during
episodic ﬂood events when turbidity may increase
Mainland Australia is an old, highly weathered, by several orders of magnitude. Rivers along the
and dry continent. About 30% of the landmass is southeastern seaboard are faster ﬂowing and
covered by desert with an annual rainfall of less shorter than those draining inland and, conse-
than 200 mm; some 90% receives a median annual quently, are generally less turbid during non-ﬂood
rainfall of less than 800 mm. Among the inhabited ﬂows. Coastal rivers along the Queensland coast
continents, Australia has the lowest annual rainfall are highly episodic, being driven by cyclones dur-
and the highest proportional loss of surface water ing the monsoon season. During this period, they
through evaporation and transpiration. Variability are usually extremely turbid, because of erosion in
in rainfall is extreme, with both seasonal variation their catchments.
and ﬂuctuations between high rainfall years and Floodplains are crucial for the ecology of
drought. Most of tropical and subtropical eastern Australian inland rivers, being signiﬁcant sources
Australia is inﬂuenced by el nin˜o and la nin˜a or of carbon and nutrients for the river ecosystem.
ENSO events with a periodicity of approximately Lateral transport of sediment and nutrients to the
7 years, causing periods of high ﬂow in rivers to be ﬂoodplain during over-bank ﬂows, followed by the
followed by drought periods (e.g., Harris, 1996). return transport of organic matter and nutrients to
During droughts ﬂow, even in the largest rivers, the river channel are major drivers of ecological
ceases altogether and the rivers themselves become processes in the river-ﬂoodplain ecosystem. Input
a series of stagnant pools. In the Murray-Darling of carbon from the catchment to the main channel
river system in eastern Australia, for example, is the major energy source to the river ecosystem
river ﬂow, which averages approximately for driving microbial food webs (Robertson et al.,
14 000 Gl yr)1, may vary between 40 000 Gl yr)1 1999). In these clay-rich systems, phosphorus is
during wet years and zero ﬂow during times of sediment-bound and little of it is bioavailable. As a
drought (Newman, 1998). Similar, although less result, internationally accepted limits of phospho-
well-understood, climate cycles also aﬀect the west rus concentration in waterways are not applicable
coast of Australia (Bureau of Meteorology to these systems. Nitrogen is delivered via
Research Centre et al., 1999). decomposition of ﬂoodplain vegetation and from
Australian inland rivers characteristically have animal wastes. There is little information on
very low gradients through most of their course as nitrogen ﬁxation and atmospheric deposition of
a consequence of a long history of erosion pro- nitrogen as input sources to rivers.
cesses that have created great inﬁlled basins with Variability in discharge in Australian arid zone
very low reliefs and predominantly clay-rich river streams is high, while variability in stream dis-
systems. For example, the Murray River in the charge in high rainfall areas is higher than in
MDB, has a slope of only 0.2 mm km)1 for about comparable streams in other parts of the world
the last 500 km of its course (Thoms et al., 1998). (Lake, 1995; Finlayson & McMahon, 1988). The
This has led to the development of extensive structures of river channels and ﬂoodplains, as
ﬂoodplains that are temporary storages for alluvial well as indigenous plants and animals, have
material adjacent to the main river channel, along evolved to accommodate this variable ﬂow.
with numerous meanders and billabongs Probably the most profound anthropogenic
(oxbows). After heavy rain, there is extensive inﬂuence on Australian rivers, however, has been
ﬂooding of these near-river areas. Biodiversity is through ﬂow regulation by a series of dams and
high on these ﬂoodplains because of the diversity reservoirs in the upper catchments of most of the
of terrestrial and aquatic habitats. major rivers and weirs in the lowland regions.
Australian landscapes are typically highly These rivers are managed as rural water supply
weathered with high percentages of clay material. systems for the major irrigation areas that have
Erosion in the upper catchments of most rivers of been developed within the MDB, parts of
eastern Australia, particularly gully erosion and Queensland and the northwest of Western Aus-
bank slumping, provides a high sediment load. tralia. Apart from reducing in-stream ﬂows and
Thus, rivers are turbid with high loads of colloidal disconnecting river channel and ﬂoodplain these
new ﬂow regimes tend to be delivered at the time catchment and ecological processes are driven by
of year that suits the needs of irrigators rather than material and chemical loads entering the estuary
the in-stream environment. It has been only re- from the river (see also National Research Council,
cently that river management has begun to con- 2000). Studies suggest that, even in permanently
sider the importance of river ﬂows to aquatic open estuaries in temperate Australia (such as some
ecosystems and measures are slowly being imple- of the drowned river valley estuaries in southeast-
mented to address this issue. ern Australia) on an annual average only about
AustraliaÕs coastal regions have also been 30% of the material transported to the estuary
aﬀected by European settlement. Australia has a reaches the nearshore ocean (SKM, 1997). This is
coastline of almost 37,000 km with 758 major likely to be diﬀerent in tropical areas where the
estuaries (Digby et al., 1998). Relatively little work greatest proportion of loads are delivered in a rel-
has been done on Australian estuaries, compared atively small number of ﬂoods (e.g., Mitchell et al.,
to northern hemisphere estuaries. Only a handful 1996; Eyre, 1998) and in a small number of estu-
of AustraliaÕs estuaries have been studied inten- aries in Tasmania, where year-round high runoﬀ
sively and those studies have mostly been of leads to short residence times of the water (e.g.,
estuaries near the larger population centres of CSIRO Huon Estuary Study Team, 2000).
Brisbane, Hobart, Melbourne, Perth and Sydney.
There are no comprehensive data sets comparable
Eutrophication in Australia – historical perspective
to those in Chesapeake Bay in the eastern USA
and the Baltic Sea in Europe.
Phytoplankton blooms including toxic cyano-
It is diﬃcult to extrapolate from the Northern
bacterial blooms have occurred in AustraliaÕs
Hemisphere estuarine studies because, apart from
estuaries, lakes and rivers for at least 100 years.
some estuaries in Tasmania, Australian systems
In this section, we brieﬂy review four occasions
are dominated by features such as overall low
when blooms in each of these types of water
freshwater inﬂow, interspersed with episodic
body have caused major management diﬃculties,
events, which deliver the bulk of water. As a ﬁrst
and the eﬀectiveness of the management
step towards developing management strategies
for these systems the National Land and Water
Resources Audit (NLWRA, 2001) developed a
three-part estuarine classiﬁcation system based on
the interplay of geology, hydrology and ecology:
Francis (1878) is often quoted as being the ﬁrst
1. a relatively small number of estuaries with scientiﬁc paper to describe a toxic algal bloom.
permanent connections to the sea, mainly Codd et al. (1994) describe the background to this
large drowned river valleys, bloom in Lake Alexandrina, an estuarine lake at
2. wave dominated barrier estuaries, most with the mouth of the Murray River in South Australia,
trained entrances that are permanently open, (Fig. 1a) and FrancisÕ subsequent investigations.
or with untrained entrances that remain par- The species causing the problem was Nodularia
tially open, and spumigena. The bloom occurred when water levels
3. estuaries, saline coastal lagoons and creeks were low and water temperatures relatively high.
with entrances that are mostly closed. Scums formed and ﬂoated to the shore of the Lake
where cattle, horses and sheep died after drinking.
All estuaries in category 3 as well as the barrier The post mortem results on sheep that died at the
estuaries in group 2 with untrained entrances that lakeshore were identical to the results on sheep
close occasionally are termed Intermittently Closed that had been fed known amounts of fresh scum
and Open Lakes and Lagoons (ICOLLs) (Roy et (Francis, 1878). There was no sign of the scums in
al., 2000). Because of the episodic ﬂow regime of the stomachs of the experimental sheep, implying
many of its rivers, Australia has a disproportion- that the material had been absorbed into the
ately large number of ICOLLs. These estuaries act bloodstream. Thus, Francis concluded that the
as terminal sinks for material exported from the Nodularia was responsible for the deaths. Codd
South Johnston River
NT Herbert River
Swan and Avon River
PERTH N SW
CANBERRA A CT
Wilson Inlet Lake Alexandrina VIC
MELBOURNE Lake Mokoan
Port Phillip Bay
0 200 400 600 800 1000km
T A S
lgo MacIntyre R
Gw y d
LI Chaffey Dam
h lan Hawkesbury R
Adelaide rum Sydney
0 100 200 km
Figure 1. Location of the sites in Australia where the eutrophication research reviewed in this paper has been done (a) and a more
detailed location map of the MDB (b).
et al. (1994) describe how an informal water tion in the Serpentine River feeding the estuary,
quality monitoring program operated through the with toxic cyanobacteria and nuisance dinoﬂa-
observations of police oﬃcers stationed in the gellates occurring in the 4 years following the
lower Murray area and how community education opening of the channel (Rose, 1998). Thus, the
had alerted landowners to the dangers to stock elimination of Nodularia blooms and general
when scums appeared in water storages. There is improvement in estuarine water quality since
no information on the eﬀectiveness of these man- 1994 is almost certainly due to the opening of the
agement actions. Dawesville Channel. Rose (1998) attributes this,
at least partly, to the increased salinity of bottom
The Peel-Harvey estuary waters preventing the germination of Nodularia
The Peel-Harvey estuary, south of Perth (Fig. 1a), The eﬀect of the nutrient reduction program is
had experienced nuisance macro-algal growths less certain. A phosphorus reduction program had
from the 1960s to the early 1990s. The potentially already been instituted in the catchment in 1983.
toxic cyanobacterium Nodularia spumigena was From 1988 this was strengthened by setting strict
noted in one of the rivers feeding the estuary in the load targets and through controls over develop-
early 1970s. The ﬁrst large cyanobacterial bloom, ments in the catchment, increased education of
attributed to excessive nutrient loads from the farmers, controls over point source inputs and
surrounding catchments, occurred in the estuary in development of ameliorative measures such as
1974. The shallowness (average depth 1 m) and spreading bauxite wastes with high P binding
warm water temperatures of the estuary, the re- capacities. A 1993 review of water quality data
stricted ﬂushing with the ocean through the nar- (Bott, 1993; Humphries & Robinson, 1995)
row Mandurah channel and the high nutrient claimed that total P loads to the estuary had
input provided ideal conditions for plant growth. decreased signiﬁcantly since the introduction of
Phosphorus was identiﬁed as the nutrient limiting the phosphorus reduction program. However, a
the growth of the Nodularia. 1994 review (Anon, 1994a) concluded that reduced
A wide range of management responses were river ﬂows prior to 1996 and errors in the original
considered by the Peel-Harvey Study Group nutrient load estimates made it impossible to show
(1988) with seven of the most promising being that there had been a statistically valid reduction
reviewed in detail in a subsequent study (Kinhill in TP loads entering the estuary. Jakowyna (2000)
Engineers, 1988). The latter review showed that a has also concluded that it is not possible because
combination of increased salinities by construction of natural variability to conclude whether the
of a second channel to the ocean and a reduction program has been successful or not.
in phosphorus inputs to the estuary would reduce
the frequency of Nodularia blooms to acceptable Lake Mokoan
levels. The new entrance to the ocean (Dawesville
Channel) was completed in 1994. Lake Mokoan is a shallow water storage
A statutory review of these measures (EPA, (362,000Ml when full) covering the site of a natural
1999) stated that the predicted eﬀects of the swamp in northeastern Victoria (Fig. 1a). When
Dawesville Channel on estuarine water quality the lake was formed in 1970 as a supplementary
had been achieved. Nodularia blooms, which had storage for irrigation purposes, it was covered in
occurred 9 years out of 12 prior to the channel, extensive beds of native ribbon weed (Vallisneria
had not been recorded in the 5 years since the sp.), water clarity was good (NTU about 10) and
Channel was opened. Harvested macroalgal loads nutrient levels were low. The Lake was almost fully
had fallen from 60,000 tonnes in 1979 to drawn down during a drought in 1982/3 and the
5000 tonnes in 1996 and all common water lakebed, consisting of ﬁne clays, was fully exposed
quality parameters (chlorophyll, Secchi depth, for an extended period during which the macro-
etc) had improved. Seagrasses have been more phytes died. The exposed sediments were mixed
abundant since 1994. Although estuarine water into the water column when the lake reﬁlled over
quality had improved, there had been deteriora- the next few years. High turbidity (300NTU in
1993) and nutrient levels were recorded and the coordinating agency across the Basin) mounted
macrophytes were unable to regenerate under these research programs into the causes of the problem
low light conditions. Cyanobacteria (Microcystis while simultaneously undertaking management
aeroginosa) were ﬁrst recorded in bloom concen- actions based on best available information.
trations in 1989 and have recurred every year since. A ‘‘run of river’’ study (Donnelly et al., 1992),
This has restricted the use of the water for both commissioned by MDBC, proposed a model to
irrigation and recreation. explain the occurrence of the bloom. This model
A management program consisting of control- envisaged the build-up of P-enriched, low carbon
ling the water regime, nutrient load reductions from sediments under generally oxic conditions between
the catchment and re-establishing macrophyte beds, major ﬂood events. During extended periods of
was instituted in the early 1990s to restore the Lake low ﬂow, an inﬂux of sulphate rich, saline
to its previous macrophyte dominated state (Loone groundwaters into the river caused the ﬁne sedi-
& Lloyd, 1996). A review of the program in 1998 ments of the river to ﬂocculate and allowed light to
(Water ECOsciences, 1998) found that, although penetrate into the water body. Phytoplankton
turbidity levels had dropped to about 150NTU, the started growing and, as algal detritus was depos-
water was still too turbid to support macrophytes ited on the bottom, sulphate reducing bacteria
and that nutrient and chlorophyll concentrations became active. As oxygen levels decreased in the
had not been reduced signiﬁcantly. The water bottom waters, sediment-bound P was released as
management program was reﬁned and expanded as soluble P into the water column, promoting the
a result of the review (Lloyd, 1998). Additional growth of the phytoplankton leading to bloom
management options were canvassed during the conditions. The authors noted that the nutrients in
review (Sandercock, 1998) and re-establishing the bottom sediments originated from catchment
macrophytes was reaﬃrmed as crucial to improving sources and these too needed to be investigated.
the LakeÕs water quality. There had been little A series of studies were then initiated into the
experience in rehabilitating shallow lakes in Aus- sources of nutrients within the river, the biogeo-
tralia but overseas experience has shown that a chemistry of river sediments, algal ecology and
range of measures have to be instituted simulta- hydrology of the river. The results from these
neously for successful rehabilitation (e.g., Moss individual studies are discussed later. The overall
et al., 1996a; van Dijk & van Donk, 1991). conclusion was that the hypothesis advanced by
The experience of Lake Mokoan has demon- Donnelly et al. (1992) was supported and that the
strated the diﬃculty of reducing nutrients and combination of low ﬂows, saline, sulphate-en-
turbidity suﬃciently and for long enough to turn a riched groundwaters and relatively P-rich clays
high-turbidity, high-nutrient, cyanobacteria-domi- within the river system, derived from the relatively
nated system back into a microphyte-dominated large basaltic areas in the catchment, triggered the
system. Both states appear to be stable and it takes 1991/92 bloom.
considerable energy to move between the two. The
Lake remains eutrophic with turbidity still drop-
ping slowly but with little change in nutrient or State of knowledge in early 1990s
Various task forces set-up by State governments
The Darling River and the MDBC during the early 1990s (NSW Blue-
Green Algal Task Force, 1992; Schonfeldt, 1993;
The 1000 km long bloom on the Darling River in Anon, 1994b; Victorian Blue-Green Algal Project
the summer of 1991/2 caught both national and Team, 1995) together with two scientiﬁc reviews
international attention (Fig. 1a, b). The rapidity of commissioned by the Land and Water Resources
the bloomÕs development and its extent caught Research and Development Corporation (Harris,
river managers by surprise. As there had been no 1994) and the MDBC (Donnelly, 1994) provide a
extensive studies of eutrophication in the river snapshot of the state of knowledge about algal
prior to the bloom, the NSW government and the blooms in southeastern Australia in the early 1990s
Murray-Darling Basin Commission (MDBC – the (summarised in Table 1).
Table 1. Summary of the state of knowledge at the beginning of the 1990s about factors and processes resulting in eutrophication of Australian freshwaters and estuaries
Issue/process Summary of knowledge References
Anthropogenic sources Diﬀuse sources are a major contributor to nutrient loads in MDB; in dry years point sources, Gutteridge Haskins &
mainly STP, are signiﬁcant producers of bioavailable nutrients Davey P/L (1992)
P transport is closely linked with soil movement and erosion management is seen as essential; NSW Blue-Green Algal
few data on loads of diﬀuse source P from diﬀerent sources, particularly at larger scales Task Force 1992; Schonfeldt (1993)
P transport believed to be mainly by surface ﬂows, but some evidence of sub-surface transport
Phosphate fertilisers P fertiliser seen as potential source, but magnitude unknown NSW Blue-Green Algal
Task Force (1992)
Internal nutrient Simple calculation suggest that P stocks in sediments in the MDB may be two orders of Schonfeldt (1993)
sources magnitude greater than water column standing stocks; biogeochemical links well documented
from European and North American studies, but few Australian data
Speculation about importance of nutrient cycling by zooplankton and ﬁsh
In-water physical/chemical processes
Phosphorus P from sewage treatments plants generally bioavailable Simmons & Cheng (1981)
bioavailability P in stormwater refractory except when stormwater runoﬀ occurs immediately after fertiliser Schonfeldt (1993)
application in rural areas, though there was little general agreement
Nutrient limitation P limits phytoplankton biomass in freshwaters, although there were a few alternate views Schonfeldt (1993), Anon (1994b)
(eﬀect of ﬂows)
Nitrogen N accepted as a factor contributing to phytoplankton blooms in freshwater, Forsberg (1975)
but not in a controlling way
N widely accepted as the limiting nutrient in estuaries; knowledge of biogeochemistry of Bauld & Millis (1977); Newell (1990);
estuaries extremely limited; only a few published papers Skyring et al. (1992)
Temperature Eﬀects of changes in temperature on algal (particularly cyanobacterial) succession well studied; Nicholas (1980), NSW
indirect eﬀects also documented Blue-Green Algal Task
Turbidity Turbidity in lakes and rivers favours cyanobacteria because of their ability to Ganf & Oliver (1982),
regulate cell buoyancy Smith (1986), Geddes (1988),
No process understanding of the role of light in controlling phytoplankton biomass and
succession in freshwaters
River ﬂow Correlation between increased river ﬂow and decreased algal numbers observed in Hotzel & Croome (1994), Jones (1993);
a few studies; not suﬃcient information to set ﬂow targets Schonfeldt (1993)
Biological and ecological interactions
Phytoplankton Increasing reports of toxic cyanobacterial blooms in inland and toxic dinoﬂagellate Bowling & Baker (1996),
dynamics blooms in coastal waters Falconer et al. (1983), May &
McBarron & May (1966),
McBarron et al. (1975),
Negri et al. (1995) and
Runnegar et al. (1988)
Contributing factors are buoyancy regulation and vertical migration to Ganf & Oliver (1982),
exploit light and nutrient regimes, plus ability of both groups to form Fay & Van Baalen (1987)
resting stages that can survive adverse conditions in sediments, including
during periods when sediments dry out
In coastal waters, shipsÕ ballast water identiﬁed as a major source Hallegraeﬀ & Bolch (1992),
Hallegraeﬀ & Fraga (1998)
Macrophyte Conﬂicting evidence of competition between macrophytes and van Donk (1991),
competition phytoplankton, with few published studies Malthus et al. (1990)
Food web eﬀects Food web manipulation (biomanipulation) in its infancy in Australia; Matveev et al. (1994a)
some indication that it might be a useful tool
It is clear that, in the early 1990s, the range of aquatic environments, it is the dissolved inorganic
factors controlling eutrophication in both fresh- forms of N and P that are most readily available
waters and estuarine systems was well understood for assimilation by algae (e.g., Gabric & Bell,
in principle. However, much of this understanding 1993). Whereas analyses of the inorganic species,
arose from research conducted in the Northern NH4 and NOx, appear to give reliable estimates of
Hemisphere. While the basic processes leading to bioavailable N, bioavailable P is much more dif-
high algal concentrations are the same throughout ﬁcult to measure because of its high aﬃnity to
the world, the relative importance of these particles. Studies by Oliver et al. (1993) and Oliver
processes and how they are inﬂuenced by climate et al. (2000) have shown that the ﬁltrate from
and the particular physical and chemical charac- 0.45 lm ﬁltered water commonly used to estimate
teristics of diﬀerent landscapes may profoundly ﬁlterable reactive phosphorus (FRP) still contains
inﬂuence their relative importance and thus the some P attached to the ﬁne clay sediments of the
options for managing them. For example, ﬂow and rivers of the MDB and that ﬁltration through
turbidity were known to be correlated with algal 0.2 lm ﬁlters provides better estimates of FRP in
biomass in Australia, although the reason for the this part of Australia (McKelvie et al., 1995).
correlation was not clear. These ﬁltration techniques will tend to under-
estimate bioavailable P, however, because they do
not measure phosphate reversibly bound to parti-
Current state of knowledge cles that can become available to phytoplankton as
the dissolved fraction is taken up in algal growth.
Scientiﬁc initiatives Using iron-oxide coated paper (ﬁlter strips) is a
simple technique which extracts both dissolved
Various coordinated eutrophication R&D pro- and loosely bound orthophosphate from unﬁltered
grams were developed in freshwater and estuarine water samples and thus provides a truer measure
systems in response to the increased concern about of bioavailable P than ﬁltration techniques. Oliver
eutrophication. These included the CSIRO Blue- et al. (1993) have shown good correspondence
Green Algal Program (Davis, 1997); the National between this ‘‘desorbable’’ P (DP) and the P
Eutrophication Management Program (1995– requirements of phytoplankton calculated from
2000); the Port Phillip Bay Study (Harris et al., growth bioassays. Field studies have shown that
1996); the Moreton Bay Study (Dennison & Abal, this DP may make up anything between 0 and
1999); the Huon Estuary Study (CSIRO Huon 80% of the total P in rivers of the MDB, high-
Estuary Study Team, 2000); the Swan Estuary lighting the inaccuracies arising from continued
Study (Hamilton & Turner, 2001); and the Wilson usage of TP as a measure of bioavailable phos-
Inlet Study (Thompson & Twomey, 2000). phorus. Gel probes (Davison & Zhang, 1994) are a
The Darling River Study (Oliver et al., 2000), the relatively new technique with possibilities for wide
Chaﬀey Dam Study (Sherman et al., 2001) and the application, including measurements of bioavail-
Johnstone Study (Hunter & Walton, 1997) have able phosphorus in situ.
been the major site-speciﬁc freshwater manage- To target management eﬀorts eﬃciently, it is
ment studies conducted in recent years (Fig. 1a, b). important to know the bioavailability of phos-
These coordinated studies, together with other phorus from diﬀerent sources. Thus, Gerdes &
individual research projects conducted by Uni- Kunst (1998) have shown that, while 72% of the
versities, Cooperative Research Centres, the total P in eﬄuent from sewage treatment plants
CSIRO and government agencies have greatly (STP) was bioavailable, only 30% of the total P in
expanded our knowledge base on eutrophication. eroded material entering the river Ilmenau in
Germany was bioavailable. They also showed that
Nutrient delivery to waterways this proportion increased to 59% when the soils
from which the material was sourced were ferti-
Bioavailability lised, suggesting that fertiliser introduced sig-
Although the plant nutrients nitrogen and phos- niﬁcant amounts of bioavailable material into the
phorus occur in many diﬀerent chemical forms in runoﬀ. Sherman et al. (2001) also found that FRP
from the Peel River in northwestern NSW made It is diﬃcult to interpret the ecological impor-
up about 30% of the total P load entering Chaﬀey tance of point and diﬀuse source nutrient loads
Dam (Fig. 1). The source of this dissolved P is simply from their relative size (Eyre et al., 1997).
unknown because the tracer techniques used in Loads are delivered from point and diﬀuse sources
these studies consider particulate matter only. The in quite diﬀerent ways: point sources such as STP
authors stressed the importance of tracing FRP deliver a constant discharge all year round to a
rather than total P if good management decisions particular location, while most diﬀuse sources
are to be informed by these studies. operate primarily during episodic storm events
In a study of the Goulburn River in Victoria, over a wide area of the waterway. Eyre et al. (1997)
Oliver & Webster (2001) found that 53, 32 and 18% compared the ecosystem responses of two rivers in
of the TP coming from an irrigation drain, a dryland northern NSW that have diﬀerent mixes of nutri-
sub-catchment and a STP, respectively, were readily ent sources. They found that in the Richmond
available to algae. The majority of this available River, which has only diﬀuse nutrient sources, P
phosphorus occurred in dissolved form. Not and N concentrations in water and sediments did
surprisingly, the particles from the STP were largely not change signiﬁcantly over time and chlorophyll
organic while those from the other two sources were a concentrations remained low even though
predominantly inorganic. Despite these diﬀerences phosphorus and nitrogen loads had increased by
in particle types, the percentage of the phosphorus factors of 2.5 and 3 respectively over the last 50
associated with particle surfaces but rapidly years. On the other hand, the Brunswick River
exchangeable was, on average, similar for each received about 10% of its nutrient load from
source (23, 25 and 29% respectively). sewage eﬄuent and its estuary was subject to fre-
quent phytoplankton blooms.
Diﬀuse vs. point sources Thus, it is not only the quantity of nutrient
Diﬀuse sources of nutrients usually dominate the loading to rivers that is important, but also the
input loads to rivers and estuaries. For example, in timing, location and nature of the loading. In dry
the Chesapeake Bay system in the US, Boynton periods, residence times are long, water velocity is
et al. (1995) found that diﬀuse loads contributed low and particles settle out of the water. During
approximately 60% of the total nitrogen and total these periods there is likely to be greater light
phosphorus loads to the Bay, whereas point source penetration into the surface waters, and hence
inputs contributed 28 and 35% of the TN and TP nutrients from point sources will often lead to the
load respectively. Atmospheric deposition made development of algal blooms. Kerr (1994) found
up the remainder. Diﬀuse sources are even more that blue green algal blooms in the Hawkesbury-
dominant in Australia. In one of AustraliaÕs most Nepean River near Sydney, Australia, were
urbanised catchment, the Hawkesbury-Nepean strongly associated with low ﬂow and point source
river system near Sydney, Davis et al. (1998a) discharges from STP. In a period of low river ﬂow
estimated that diﬀuse sources still contributed in 1994, which led up to the development of a large
approximately 70–80% of the TN and TP loads summer algal bloom, he calculated that 75% of the
with point sources contributing the remainder. In phosphorus and 84% of the nitrogen delivered to
less populated areas, the diﬀuse source contribu- the river was from STPs. Similarly, Eyre et al.
tion can be as high as 96–98% of the total load (1997) found that the majority of the low ﬂow
(McKee et al. 2000a). nutrient load in the Brunswick River estuary was
Meyer et al. (2001) have summarised the small from a STP and that phytoplankton blooms were
amount of information available on atmospheric common during this period when the upper estu-
wet and dry deposition of N in Australia. They ary was poorly ﬂushed.
report annual mean deposition rates ranging from
0.3 kg ha)1 in oceanic areas to approximately Nutrient limitation
10 kg ha)1 in urban centres, values which are of the Following Howarth (1988) a nutrient is deﬁned to
same order as the average annual Australia-wide be limiting if it causes an increase in the rate of
emission of about 3 kg ha)1. These values are rel- net primary production when added to a system.
atively small compared with catchment loads of N. This may lead to changes in the structure of the
community. It is thus the eﬀect on primary Phosphorus limitation has been observed in
production or growth rate that is important and Australian estuaries. For example, nutrient limi-
not the eﬀect on the ﬁnal biomass achieved. The tation occurred from late spring to autumn in the
latter is aﬀected by a range of factors unre- Swan River estuary (Thompson & Hosja, 1996)
lated to limitation such as self-shading in with nitrogen being the nutrient most likely to
phytoplankton. limit algal biomass during this period. However,
It has become clear in the literature of the last these authors observed a low degree of phosphorus
decade that the earlier generalisation that limitation, supporting an earlier study (Jack, 1987)
phosphorus is limiting in freshwater systems (e.g., that had found a correlation between phosphorus
Vollenweider, 1976; Schindler, 1977, 1978; Harris, loading and chlorophyll a concentration.
1994) whereas nitrogen limits primary production In Australia other factors, particularly light,
in marine and estuarine systems (Ryther & Dun- also exert inﬂuence on plant growth. In addition,
stan, 1971; Howarth, 1988) is an oversimpliﬁcation. changes to the light spectrum at depth due to the
Elser et al. (1990) reviewed studies of 77 North presence of light-absorbing substances in the water
American lakes and concluded that ‘‘Little support (gelbstoﬀ), may further reduce phytoplankton
can be found .… for the conventional wisdom that production (Kirk, 1994; Koop, Larkum & Drew,
P is the predominant primary limiting nutrient in unpubl. data). It appears that plant growth (in-
freshwater’’. They found that, apart from serious cluding phytoplankton growth) in many turbid
deﬁciencies in experimental design of a large Australian rivers is primarily light limited but once
number of the studies, there was no statistical the water enters weir pools or dams, where ﬂows
diﬀerence between the responses to P and N are reduced and particulates in the water can settle
enrichments. Positive growth responses to both N out, this limitation is removed and phytoplankton
and P enrichment had been recorded in 86% of the blooms occur.
studies; 47% had shown increased growth when P In blackwater streams and estuaries carrying
alone was added and 40% when N alone was high loads of coloured dissolved organic matter
added. Moss et al. (1996b) have also pointed out (CDOM) light penetration is signiﬁcantly reduced.
that, although it is theoretically true that nitrogen- Spectral composition can also be inﬂuenced. This
ﬁxing cyanobacteria in lakes can take advantage of has been found in TasmaniaÕs Huon estuary,
high levels of dissolved phosphorus and atmo- which carries high loads of humic substances
spheric nitrogen for growth, in practice ‘‘the (CSIRO Huon Estuary Study Team, 2000).
development of blue-green algal nitrogen ﬁxers in
lakes is slow and the phytoplankton as a whole External nutrient sources and transport
may be strongly nitrogen-limited during mid-
summer’’. Point source pollution to coastal rivers and estuaries
Similar results have been found in Australian As has been found elsewhere in the world, sewage
freshwaters (Robertson, 1999). Studies by Grace pollution of coastal rivers, estuaries, and nearshore
et al. (1997) and Oliver & Webster (2001) in the waters in Australia has been implicated in eutro-
Darling River and in the Murrumbidgee and phication. Most of the sewage produced in coastal
Murray Rivers (Oliver, reported in Robertson areas of Australia is discharged to estuaries and
1999) have shown that, when nutrient limitation coastal waters following treatment, although the
was detected, nitrogen was most frequently the extent of treatment varies. Environmental eﬀects
nutrient limiting phytoplankton growth. Limita- from major sewage outfalls to coastal waters,
tion was related to the time of year with nutrients especially through deepwater outfalls, have been
controlling primary production during late spring diﬃcult to detect because of the highly dynamic
and early summer. Both nitrogen and phosphorus nature of the coastal environment (Koop &
can limit phytoplankton growth in deep storages Hutchings, 1996). The much smaller outfalls dis-
such as Lakes Dartmouth and Hume (Matveev & charging directly to the coastline in New South
Matveeva, 1997) and Chaﬀey Dam (Sherman Wales have been shown to have only local eﬀects
et al., 2001). (e.g., Krogh, 2000).
Discharges to coastal rivers and estuaries, atmospheric N onto agricultural lands and directly
however, have contributed to signiﬁcant into water bodies has also risen, so that Smith
eutrophication problems in many areas. Thus, et al. (1999) estimate that Northern Hemisphere
chronic cyanobacterial blooms in the Hawkesbury inputs are approx 40–90 kg ha)1 yr)1 in heavily
River near Sydney (Fig. 1b) have been linked with industrialised areas.
discharges of bioavailable N and P from STPs These excess nutrients can potentially be
upstream (Kerr et al., 1996). A vigorous scientiﬁc transported to waterbodies, particularly in areas
debate about the relative importance of point and where the ﬂow pathway to streams has become
diﬀuse sources of nutrients for eutrophication and saturated with phosphorus (Johnes & Hodgkin-
the discussion about the relative signiﬁcance of N son, 1998). Although there has not been a com-
and P limitation (see above) have made it diﬃcult prehensive study of eutrophication across Europe,
for environmental managers to decide on nutrient there are clear indications from more restricted
reduction strategies, particularly given the high studies of major increases of nutrients in water-
cost implications. ways over the last 30 years. For example, a third
Spatial statistical analysis techniques developed of 200 lakes in the UK have been found to have
in the Port Phillip Bay Study (Harris et al., 1996) had more than 500% increase in nutrient loads
and further developed in the study of Moreton over the last 50 years (Moss et al., 1996a) and a
Bay (Dennison & Abal, 1999) are now available to study of four European countries found that
map the extent of sewage signals around sewage agriculture contributed more than 60% of the TN
outfalls. The latter study developed a novel loads and 40% of the TP loads (Kristensen & Ole
method of tracing sewage plumes using measure- Hanson, 1994). Thus, it is not surprising that
ments of stable isotopes of N in sediments and Carpenter et al. (1998) claim that eutrophication is
naturally occurring aquatic plants and in cultured the most widespread water quality problem in the
algae (Dennison & Abal, 1999). In that way the US and many other Northern Hemisphere nations.
eﬀects of STP located on separate rivers could be Although there had been considerable research
distinguished in Moreton Bay (Fig. 2a and b). into sources and pathways for nitrogen, particu-
Rush (2005) has further developed this technique larly nitrate, from agricultural sources, relatively
using an algal bioindicator. She was clearly able to little attention had been paid to diﬀuse sources of
show a sewage-nitrogen signal in the algae phosphorus. As a generalisation, managers had
throughout the Hawkesbury-Nepean river system. focussed on point sources of phosphorus, partly
because a signiﬁcant fraction of the phosphorus is
Diﬀuse source pollutants – phosphorus discharged from point sources, partly because point
In Europe there has been an intensiﬁcation of sources are easier to identify and manage, and
agriculture, particularly in the post WWII period partly because phosphorus was believed to be
as countries that suﬀered food shortages attempt strongly bound to soils and therefore relatively
to gain food security (Johnes, 1999). Over 40% of immobile (Oenema & Roest, 1997). Fertilisers and
EC lands are now used for agriculture. There have animal manures are believed to have been applied at
been a number of studies that quantify the such rates and for so long that in many areas soil P
resulting nutrient surpluses in agricultural lands has built up to levels well in excess of crop needs and
(Foy & Withers 1995; Vitousek et al., 1997; Car- has the potential to signiﬁcantly increase phos-
penter et al., 1998). Heathwaite (1999) quotes an phorus levels in waterways (Sharpley et al., 1999).
annual phosphorus surplus of 10 kg/ha in the UK This pool of phosphorus is mobilised through both
(Sharpley & Withers 1994) and 26 kg/ha in the surface and subsurface runoﬀ as well as through soil
US, and Iserman (1990) quotes between 55 and erosion. In Australia, the importance of diﬀuse
88 kg/ha for parts of Europe. Vitousek et al. phosphorus sources has been accepted since the
(1997) estimate that the global agricultural N early 1990s (Gutteridge Haskin & Davey, 1992).
fertiliser usage has increased from <10 million The Australian environment and Australian
tonnes to 80 million tonnes over the last 50 years land and waterway management practices diﬀer
and that the spreading of farm animal wastes has considerably from those in the Northern Hemi-
increased even more sharply. Deposition of sphere. Animal densities and crop densities are
3.9 Deception Bay Island
Samsonvale 11. 1 M O R E T O N
Nth 11. 3 B A Y
Bramble Bay 2.2
LEGEND 9.1 1.6
11. 3 Algae 2.5
8.7 Seagrass r Waterloo Bay 2.4
Ri 8.7 5.4 2.5
2. 0 North
9.9 9.7 Stradbroke
ne 2.5 Island
is ba 9. 3
9.8 Lo 3.6
0 10 20km
Figure 2. d15N Sewage signatures in naturally occurring material (a) and sewage plumes based on d15N signals in indicator algae (b) in
Moreton Bay, subtropical eastern Australia. Ambient d15N signatures of sediments and primary producers near sewage outfalls were
characterised by high d15N signatures, while remote sites had low values (a). Note that use of indicator algae allowed discrimination of
sewage signals from sources relatively close to one another (b). (Redrawn from Dennison & Abal (1999)).
relatively low (except in designated irrigation is high and increasing. Between 1983/4 and 1996/7
areas) and, for most dryland agricultural areas, water use increased by 65% (AATSE/IEA, 1999),
fertiliser usage rates are much lower than in Eur- with most of the increase arising from increased
ope and the US. On the other hand, river water use use for irrigated agriculture.
Deception Bay Island
Lake M O R E T O N
B A Y
Pin Bramble Bay
0-3 N (ppt)
ve Waterloo Bay
STP ne Island
0 10 20km
Figure 2. (Continued)
With the exception of speciﬁc rivers near ur- phosphorus and 29% of total nitrogen in dry
ban areas or draining irrigation areas, simple periods and 83% of total phosphorus and 84%
mass budgets readily demonstrate the dominance of total nitrogen in wet years in the MDB.
of diﬀuse inputs in Australia catchments. Compared to dry years, the total loads were 3–4
Gutteridge Haskin & Davey (1992) estimated times greater in average years and 20 times
that diﬀuse sources accounted for 28% of total greater in wet years.
Thus, from the early 1990s, a prime issue for as very sensitive tracers of the geologic origins of
Australian natural resource managers and scien- the sediment material (Fig. 3). Diﬀerent rock types
tists was identifying the sources of diﬀuse nutrients. contain distinctive isotopic ratios of these elements
It was critical to know where the nutrients were and the source rock types can be calculated from
originating and what pathways they were taking the mix of trace elements and isotopes in the river
from sources to waterbodies, so that limited funds sediments. In particular, inorganic fertiliser
could be targeted at the major contributing areas. obtained from phosphatic rock carries with it the
Possible agricultural sources included inorganic isotopic signature of the rock from which it was
fertiliser, animal wastes, natural soil stores or the mined. Thirdly, mineral magnetics can also indi-
breakdown of plant materials. Bek & Robinson cate the origin of sediments. The combination of
(1991) identiﬁed diﬀuse anthropogenic sources diﬀerent magnetic minerals in river sediments
such as fertilisers, and point sources such as STP depends on the parent rock from which the
to be the most likely sources of phosphorus. material was originally eroded.
Phosphorus was assumed to travel via surface Caitcheon et al. (1995) found that approxi-
pathways and nitrogen was assumed to travel mately 90% of the sediments in Chaﬀey reservoir –
primarily in dissolved form via both surface and a reservoir with a catchment of approximately
sub-surface ﬂows. 42,000 ha, a surface area of 542 ha and a capacity
Nutrient losses can be linked to catchment of 62,000 Ml in north-western New South Wales
characteristics. Thus, yields of TP and TN have (Fig. 1) – were derived from basaltic material. The
been related to land use, rainfall, land cover, slope concentration and ratios of Cs137 and Pb210
and soil type in diﬀerent parts of Australia (Young showed that about half the sediments came from
et al., 1996). These empirical results have been surface sources and the other half came from
codiﬁed into a simple annual average nutrient subsurface sources. The analysis showed that the
export model (Davis & Farley, 1997) which has surface sources were very limited in areal extent;
been used to identify likely sources of TN and TP approximately 2 km2 or 1% of the catchment area.
within catchments in New South Wales and parts Thus, this surface erosion could be identiﬁed as
of Queensland and Victoria. While this empirical most probably originating from a small amount of
approach is simple to implement and provides cropping that occurs on basaltic soils in the upper
managers with guidance about likely sources of catchment. The subsurface component is most
nutrients, it is indirect, does not take account of probably derived from streambank erosion as the
local conditions and is generally not sophisticated river channel passes through ﬂoodplains alluvium
enough to account for diﬀerent forms of nutrients derived from upland basaltic soils.
or diﬀerent size ﬂow events. The dO18 deviation (2.1 ppt) of the ortho-
A number of experimental techniques have phosphate was strongly characteristic of basaltic
now been developed to determine the origins of phosphate (0.3 ppt) and showed no signiﬁcant
the sediment-attached phosphorus in the rivers contribution from fertiliser phosphate (23 ppt).
across Australia. Source and transport processes Thus, the tracer results indicated that the soluble
diﬀer in diﬀerent regions of Australia (Davis orthophosphate in the storage came from subsur-
et al., 1998b). face drainage of basalt soils which are naturally
rich in phosphorus rather than from fertilisers.
Murray-Darling Basin Further support for this conclusion came from
Three of these techniques have been used in the Martin & McCulloch (1999) who, in a later study
MDB to determine the sources of sediment-at- using rare earth isotopic techniques, conﬁrmed
tached phosphorus. First, the diﬀerent penetration that most of the reservoir sediments came from
depths of Cs137 – an episodic nuclear bomb fallout actively eroding areas in the Peel River head-
product during the 1950s – and Pb210 – a contin- waters. They concluded that the contribution of
uous fallout product – can be used to distinguish fertiliser phosphorus to the sediment phosphorus
river sediments originating from surﬁcial and was less than 0.2% in this catchment (Fig. 3).
deeper erosion events. Secondly, trace element Martin & McCulloch (1999) also applied their
isotopes (Sr and the rare earths) can also be used techniques further downstream to suspended
basalt metased soil 10
metasedimentary rock C metased soil comp
basaltic soil reservoir sediment
C basaltic soil comp fertilizer
er 10 1
0.1 1 10 100 1000
Nd (ppm) / P (wt%)
Figure 3. Plots of 143Nd/144Nd vs. Nd/P for sediment from Chaﬀey Dam and fertiliser. Bulk mixing relations between average fertiliser
and basaltic soils as well as mixing proportions are shown. Nd/P ratios of rocks, soils and dam sediments are at least a factor of
20 lower than fertiliser, so as little as 1% addition of fertiliser can be readily distinguished. The data suggests there is little fertiliser
incorporation into the soils and sediments. Nd – neodymium; P – phosphorus. (Redrawn from Martin et al. (1999)).
sediments from the Namoi River near its junction Wallbrink et al. (1998) used Cs137 and Pb210
with the Barwon River. At this point, the Namoi tracing techniques to show that about 85% of the
drains a catchment of 43 000 km2. There was sediment in the Murrumbidgee River was derived
very little evidence of sediments derived from the from sub-surface sources, 13–16% came from
basaltic soils of the Chaﬀey catchment at this cultivated land and a small percent (<2%) came
location. The phosphorus concentration was 2.5 from uncultivated land (Fig. 4). The sub-surface
times lower than in the suspended sediments de- sources could be either collapse of streambanks or
rived from soils of the basaltic uplands. Again, gully erosion. This result supports Neil & Fog-
there was strong evidence of sub-soil derived ertyÕs (1991) earlier conclusion that channel sedi-
material (Davis et al., 1998b). There was no ments were the major sources of sediments in the
evidence of fertiliser-derived phosphorus in these ﬁrst order streams of this part of the MDB.
sediments although, because of the admixtures Although Wallbrink et al. (1998) did not trace the
of geologies in this catchment, they could not origins of the phosphorus attached to these sedi-
rule out the possibility of a small fertiliser ments, they showed that the phosphorus concen-
contribution. tration on the suspended sediments in both the
350 surface soil
Pbex concentration (Bq kg)
150 surface soil
Murrumbidgee, & tributaries,
50 suspended sediments
Murrumbidgee, headwater channel/gully sources
0 10 20 30 40
Cs concentration (Bq kg )
Figure 4. Plots of lead vs. caesium concentrations in diﬀerent source material in the Murrumbidgee River catchment and in river
sediments. The close correspondence of the signals from headwater channel and gully sources and the sediment suspended in the river
suggest that most of the river sediment is from sub-surface sources with only a small proportion derived from agricultural and
uncultivated land. (Redrawn from Wallbrink (1996)).
tributaries and the main channel of the Mur- ment. That is, the sediments come from reworking
rumbidgee River (Fig. 1) was consistent with that of the ﬂoodplains rather than from fresh erosion of
on the ﬁne particle fraction (<2 lm) in the the upper catchment. The phosphorus concentra-
catchment soils. From this they concluded that tion in the river sediments is less than that on the
the phosphorus on suspended (and presumably natural soils from which the material is derived
also bottom) sediments could be adequately ac- and has not changed signiﬁcantly over the last
counted for by the native phosphorus content of 280 years. Presumably, the in-water P concentra-
the eroded soils. Although they could not rule out tions have also remained relatively constant
a contribution from fertilisers, there was no need throughout the period of European settlement.
to invoke any anthropogenic source of diﬀuse Thus there was no evidence for any anthropogenic
phosphorus. increases in P concentration in this river.
The Darling-Barwon River drains These studies in three major and one minor
650,000 km2, including the fertile and erodible inland river of the MDB all show no trace of fer-
basaltic soils of the Darling Downs. An analysis tiliser-derived phosphorus although it could not be
(Caitcheon et al., 1999) of the major element conclusively ruled out in two of the studies. There
chemistry of the river sediments showed that the was strong evidence of natural phosphorus enter-
material came originally from sedimentary-gra- ing the river from accelerated erosion of subsoils
nitic rocks and not from the basaltic areas of the and streambanks. Surface erosion can dominate at
catchment. The weathering of the particles showed paddock scale but this source is overwhelmed by
that they had probably been derived from the greater subsoil losses from eroding gullies and
lowland areas rather than from the upper catch- streambanks.
Tropical Queensland soluble organic N and NH4 concentrations were
In 1993, a major desktop study of coastal Queens- relatively constant over time despite large changes
land catchments (Moss et al., 1993) concluded that in ﬂow. The nitrate-N concentrations were clearly
sediment and nutrient loads had increased in coastal driven by base ﬂows although these concentrations
streams, wetlands, estuaries and the GBR lagoon did rise if there was surface ﬂow and interﬂow. In
over the last 100 years. Urban sources were believed common with the previous studies, they found that
to be the most signiﬁcant point source contributors horticultural crops (bananas and sugarcane) con-
of nutrients and agriculture was the major diﬀuse tributed signiﬁcantly greater yields of TP (about
contributor. Point sources were largely insigniﬁcant 7 kg ha)1 yr)1) than pastures and rainforest
in terms of total load, but in many instances could (about 2 kg TP ha)1 yr)1).
have local signiﬁcance. From national soil testing data, Reuter (2001)
The studies that have been undertaken since has shown high levels of phosphorus in the soils
then have been driven by a concern for the integrity underlying the cultivated crops areas of north
of the Great Barrier Reef and its lagoon and so Queensland, compared with other Australian
analyses have concentrated on delivery from estu- States. Concentrations were commonly in the
aries rather than on the upstream processes range 40–150 mg P kg)1 using the Colwell phos-
(e.g., Hunter et al., 1996; Eyre 1993). These studies phorus test. Although the data do not identify the
have typically derived the total and areal loads of origins of the phosphorus, it is extremely likely
sediment, nitrogen and total phosphorus (e.g., that the nutrient comes from fertiliser additions to
Moss et al., 1993; Mitchell et al., 1996) and, the crops.
sometimes, attributed these nutrient loads to land Direct tracing of sediments and attached
use types. nutrients in rivers has only been carried out in
Thus, Bramley & Johnson (1996) found that Queensland in the Herbert catchment and a
nutrient export rates from intensive agriculture smaller study of Berner Creek in the Johnstone
were greater than those from other land uses in the River catchment. While the results of the Herbert
Herbert catchment irrespective of the type of study are unavailable, those from Berner Creek
production being undertaken. Mitchell et al. show the diﬀerent processes that operate in these
(1996) conclude that fertiliser accounts for sig- tropical areas compared to the MDB (Wallbrink
niﬁcant nutrient input to the lower Herbert river, et al., 2001). The majority (70%) of the sediment-
principally from sugarcane areas. Large nutrient bound phosphorus comes from cultivated lands
loads are moved down rivers during tropical (bananas in the steeper uplands and sugar cane on
storms. They estimated that 600 tonnes of nitrogen the lower, ﬂatter areas). Surface erosion is the
and 65 tonnes of phosphorus were exported from major pathway for the transport of sediment to
the lower catchment in a single large storm event rivers from these cropping systems, with only 11%
that was monitored in 1994. In a major study of of the phosphorus originating from sub-soil sour-
the Johnstone River, Hunter et al. (1996) also ces. Overall, 42% of the phosphorus exported
found that large sediment loads (67 000 tonnes), from the catchment originated from fertiliser; the
total nitrogen (343 tonnes) and total phosphorus majority of this fertiliser phosphorus originates
(104 tonnes) were moved during a cyclonic storm from the cultivated lands.
in the Johnstone River in 1994. As part of the Eyre (1993) has carried out one of the few
same study, Walton & Hunter (1997) found that studies in tropical Queensland that attempted to
horticultural crops produced signiﬁcantly greater uncover the processes controlling nutrient trans-
yields of suspended sediments, total phosphorus port within the waterways. In his study of the river
and total nitrogen than did pastures and rain- and estuary of the Moresby River, he concluded
forest. These three contaminants were delivered that particle-attached phosphorus is eroded from
quite diﬀerently. The suspended sediment and canelands during the wet season and deposited
particulate phosphorus concentrations increased into drainage lines and the upper estuary. It then
rapidly with ﬂow, so that the majority of the moves slowly down the estuary during the dry
annual loads were carried during large ﬂow events. season under the inﬂuence of tidal movement. This
However, dissolved phosphorus (PO4 and DOP), result is not necessarily contrary to the conclusions
of Hunter et al. (1996) about the dominance of terised by deep duplex and lateritic soils with high
storm ﬂows, since Eyre analysed estuarine sedi- phosphorus retention abilities. Most of the natural
ments and did not monitor ﬂow events. It is likely vegetation of the catchment has been cleared for
that both results are true, depending on whether agriculture since European settlement in 1829
cyclonic storms occur during the period of interest. (Avon River System Management Committee,
1993). The introduction of seasonal, shallow-roo-
Southwest Western Australia ted crops has led to increased recharge and
Nutrient dynamics have been studied for many increased surface runoﬀ.
years in Western Australia because of eutrophi- The hydrologic changes brought about by these
cation of estuaries in the southwest of the State. changes coupled with increased potential for soil
Deeley et al. (1999) showed that sandy coastal soils erosion and bank degradation and the widespread
with low phosphorus binding capabilities, use of fertilisers have resulted in considerable
replacement of native perennial vegetation with increases in nutrient export from the catchment
annual grasses, the rise in groundwater levels and since European settlement (Viney & Sivapalan,
the regionÕs marked Mediterranean climate all 2001). These authors modelled pre-European dis-
make this area especially susceptible to eutrophi- charges and particulate phosphorus yields and
cation. Fertilisers are the main sources of nutri- concluded that, prior to clearing, discharge was
ents, even at moderate application rates, (Hodgkin only about 20% of current levels and the TP yield
& Lenanton, 1981; Hodgkin & Hamilton, 1993; was just over 6% of the current yield. They esti-
Deeley et al., 1999) although point source inputs mated that ﬂow-weighted average concentrations
are known also to be important in at least one of for the entire catchment have increased by a factor
the catchments of the Peel-Harvey estuary of 3.3 for TP and 1.9 for soluble phosphorus since
(Fig. 1a). European settlement. In one catchment on the
The Swan/Avon River drains a large catchment coastal plain, Ellen Brook, the current concentra-
(121,000 km2), exiting to the Indian Ocean tion of total phosphorus is 5.6 times the estimated
through the city of Perth (Fig. 1a). The river has pre-European concentration. These predicted
experienced regular phytoplankton blooms and, in changes in phosphorus concentrations are similar
early 2000, a major cyanobacterial bloom to the concentration changes detected over the
(Hamilton, 2000). Because of the aesthetic and same period by Gerritse et al. (1998) in sediment
recreational importance of this waterway, there cores taken below the Ellen Brook conﬂuence.
was an intensive catchment, river and estuary Peters & Donohue (2001) analysed nutrient
study (Hamilton & Turner, 2001) to understand concentrations and ﬂuxes in 15 tributaries to the
the causes of these blooms and develop manage- Swan-Canning Estuary for the period 1986–1996
ment options. Phytoplankton are known to be and 1986–1992, respectively. The Avon River
predominantly nitrogen limited during the spring contributed 65% of the nitrogen load but only
and summer months when the major blooms ap- 32% of the phosphorus load. Most of the phos-
pear (Thompson & Hosja, 1996; Thompson et al., phorus load originates from the catchments of the
1997) and so the study included a component to coastal plain through rapid leaching of phospho-
identify the sources of both nitrogen and rus from fertiliser applied to the infertile sandy
phosphorus. soils (Gerritse et al., 1990). This is consistent with
The Swan River catchment includes a large Pionke et al.Õs (1990) conclusion that nutrients are
area of the Darling Plateau (drained by the Avon being added to some horticultural activities con-
River), a series of catchments on the coastal plain, siderably in excess to crop uptake (5· for N and
and the urban area of Perth. Perth and the coastal 10· for P) throughout the Swan coastal region.
catchments experience a Mediterranean climate Most of the added phosphorus is in soluble form
with most rain falling from autumn through to (orthophosphate and organic complexes), whereas
spring and a pronounced dry period from the remainder of the Swan catchment provides
November to May. The soils of the coastal plain primarily particulate phosphorus.
are predominantly sandy with poor nutrient The Ellen Brook catchment (664 km2 or 0.6%
retention capabilities, while the plateau is charac- of the total Swan catchment area) alone contrib-
uted a remarkable 62% of the FRP and 42% of soils with higher Fe content can accumulate
the annual TP ﬂux to the estuary. Although the phosphorus for a number of decades. Average
superphosphate application rates are relatively low phosphorus storage in the surface soils has in-
in this catchment (Gerritse & Adeney, 1992) there creased from about 930 to 3200 kg ha)1 with river
is a high (25–50%) and rapid loss of the applied sediments showing elevated phosphorus concen-
phosphorus to the stream. In recent years, the ﬂow trations within the upper few centimetres (Weaver,
adjusted TP concentrations have declined in Ellen quoted in Davis et al., 1998b).
Brook (Donohue et al., 2001) although still Transport of both particulate-attached phos-
remaining very high (median TP of 0.54 mg l)1 phorus and dissolved phosphorus via surface
from 1987 to 1994 declining to 0.37 mg l)1 by runoﬀ and leaching through the soil proﬁle are
1998). The reasons for this decline are unclear. known to occur, although the contributions of
Ellen Brook is also a major N source along surface erosion, gully erosion and streambank
with some vegetated areas on the Darling plateau. collapse to the particulate fraction are not well
Thus, Ellen Brook has the highest annual average known. However, McKergow et al. (2001) in a
organic N (1.8 mg l)1) and total N (2.1 mg l)1) study of a small catchment on the south coast
concentrations of any of the Swan river tributaries found that streambank erosion contributes little to
(Peters & Donoghue, 2001). The N/P ratio of the the TP load. Weaver & Reed (1998) conclude that
annual loads from this tributary is about 3 (the signiﬁcant areas with sandy soils of low PRI and
bioavailable ratio is even lower), thus biasing histories of fertiliser application are already
downstream waters towards nitrogen ﬁxing cya- showing signs of phosphorus saturation with the
nobacteria. Groundwater nutrient concentrations potential for major releases of phosphorus to
are exceptionally high at 5 mg l)1 TP and rivers if fertiliser applications continue.
12 mg l)1 TN under horticulture on the sands
within the Ellen Brook sub-catchment. Sharma Subsoil movement of phosphorus
et al. (1996) have forecast that the P concentra- The standard transport paradigm states that
tions in Ellen Brook baseﬂows will rise steeply phosphorus is transported in surface runoﬀ, par-
over the next few years as saturated plumes from ticularly attached to soil particles, except in areas
the sandy soils of the catchment reach the Brook. where there are sandy soils with low phosphorus
Nutrient sourcing and transport processes in sorption capacities. However, studies have now
the southern coastal catchments of WA have also shown that macropores have the potential to
been studied for a number of years because of transport phosphorus through heavy, duplex soils
actual or suspected eutrophication of estuaries via sub-surface pathways.
along this coast. The soils of these catchments are Kirkby et al. (1997) report a study of phos-
commonly duplex with siliceous sands overlying phorus transport processes in a duplex soil in
laterite and clay in the higher landscape and sands South Australia. Phosphorus, from fertiliser
and gravels at lower elevations (Davis et al., applications, was transported in dissolved form
1998b). The sandy soils have low phosphorus through macropores into the B horizon and pos-
retention indices (PRI). Much of the area has been sibly into the watertable. The fate of this phos-
converted from native bush to agriculture in the phorus was not known but it was likely to have
last 40 years (Weaver & Reed, 1998). There is a travelled along the A/B or B/C soil horizons to
monotonic increase in phosphorus concentration either enter streams directly or indirectly from
with time since clearing from 3 mg kg)1 (uncle- return ﬂow lower down the landscape. Kirkby
ared) to 29 mg kg)1 (>30 years) which Weaver et al. (1998) extended this work to 10 soil types,
and Reed attribute to the addition of phosphatic ranging from sands to heavy clays, taken as large
fertilisers. However, some of this diﬀerence may (1 m diameter) cores from ﬁeld sites in South
also be attributable to the increased clearing of less Australia. Under laboratory conditions simulating
productive, sandier soils over the period. The natural storm events, they found that phosphorus
sandy soils of this region become saturated with was ﬂushed rapidly through heavier soils when
phosphorus within about 10 years at typical fer- macropores were present, even when the soils had
tiliser application rates (13 kg P ha)1), while the high P-adsorption characteristics. The macropores
prevented the necessary contact time between the Residence time and eﬀective load
applied phosphorus and the soil particles. Depositional environments such as lakes and some
Macropores need occupy only 5% of the soil estuaries act as sinks for nutrients. In coastal rivers
volume to completely alter the phosphorus trans- without major estuaries, there is greater through-
port pathways. The prevalence of macropores in put, with much of the load being delivered in short
the ﬁeld is not known, but macropores were storm runoﬀ events (Cosser, 1989; Cullen, 1991;
present to some extent in all 10 cores from this Harris, 1996; McKee et al., 2000b). Sediments and
region and other studies have shown them to be nutrients are transported downstream in ﬂood-
common soil features. waters through the estuary and out onto the con-
This work was extended into the ﬁeld in paired tinental shelf. Similarly, in estuaries with constant
catchments at two sites (Stevens et al., 1999). One high ﬂows, such as those in Tasmania where water
site (Myponga) had loamy sands over heavy clays residence times are short, the bulk of material
while the other (Mt Bold) had silty clay loams transported by the river may be exported to
over medium heavy clays. Surface, A/B and B/C coastal waters. Although some nutrients are
horizon ﬂows were intercepted after ﬂow dis- deposited in rivers during these storm events, the
tances of between 100 m (Myponga) and 300 m residence time of most of the catchment-derived
(Mt Bold). Phosphatic fertilisers were applied at nutrients may only be a few days or at most a few
both sites at a typical agricultural rate of weeks. Once these nutrients are exported to the
15 kg P ha)1. Over 2 years of monitoring at the lower estuary and inshore coastal waters, pro-
Myponga site, over 95% of the ﬂow and the total cesses of dilution, dispersion, sedimentation and
phosphorus was transported via macropore ﬂow denitriﬁcation will tend to minimise any potential
along the A/B soil horizon. Even at the Mt Bold these nutrients have to contribute to eutrophica-
site with its heavier textured surface soils, 45% of tion, at least in the short-term.
the water over the 2 years moved via subsurface In the sub-tropical Richmond River catchment
ﬂows, although the proportion of TP that moved in northern NSW, for example, McKee et al.
via this pathway is not reported. The macropores (2000b) found that over 97% and 94% respectively
were capable of transporting both dissolved and of the diﬀuse phosphorus and nitrogen loads were
particulate phosphorus. Thus, in the ﬁrst year at exported from the catchment to the ocean during a
the Myponga site between 90 and 100% of the storm. Nutrient exports to the ocean were found to
phosphorus was in the particulate form, whereas be proportional to ﬂushing time; the larger ﬂoods
the following year between 90 and 100% was with rapid ﬂushing times exported larger propor-
dissolved. The reasons for this change in form are tions of the diﬀuse load to the ocean than smaller
not well understood. ﬂoods with longer ﬂushing times. Temperate rivers
While these experiments have shown that for might be expected to export a smaller fraction of
diﬀerent soil types, subsurface processes are the diﬀuse load to the ocean due to their generally
important (and often dominant) transporters of smaller ﬂooding regime (Eyre, 1998). In a pre-
phosphorus in both dissolved and particulate form liminary assessment of nutrient exports to the
over distances of hundreds of meters, it is yet to be ocean from several temperate NSW estuaries,
shown whether signiﬁcant loads of nutrients can SKM (1997) estimated that, on average, approxi-
be transported to streams via this mechanism. The mately 30% of the wet weather diﬀuse load into
management signiﬁcance of these studies is that (1) these systems would be immediately exported to
management activities such as grassy swales and the ocean. Thus, only a fraction of the diﬀuse load
buﬀers that are designed to trap particulate-borne that enters a coastal river system during storm
phosphorus in surface ﬂows may be ineﬀective if events contributes to eutrophication problems in
large proportions of the nutrients are moving the river and estuary.
beneath the surface, and (2) even though there is On the other hand, inland ﬂowing rivers such
no evidence of surface erosion from the tracing as those in the Murray Daring Basin have long
techniques that have been employed in the MDB, residence times for most of the year due to the very
fertiliser phosphorus may be still transported into low slope of their lower reaches and the con-
waterways via macropores. struction of impoundments along their length for
irrigation purposes. Even in times of ﬂood these 10 cm of the sediments in this river reach. The ﬁne
inland rivers have residence times ranging from clays (which contained over half this phosphorus
weeks to months. The implications of these long load) had a 10% organic content and so there
residence times are discussed below. existed the potential for large phosphorus releases
under anoxic conditions, depending on the avail-
Internal nutrient sources ability of the carbon and phosphorus, the redox
conditions within the river and transport mecha-
Sediment-water interactions nisms from the sediments to the water column.
Bottom sediments in shallow aquatic ecosystems, They calculated that even a 1% release of this
as well as acting as sinks for nutrients in settling available phosphorus into the water column would
particulate matter, may also be important sources increase the dissolved phosphate concentration
of regenerated nutrients. To date, freshwater from 10 to 70 lg/l. Sherman et al. (1998) in their
studies have concentrated primarily on the role of study of the same region of the Murrumbidgee
sediments in phosphorus cycling (e.g., Oliver et al., River found no evidence of anoxia over 2 summer–
2000) although, recently, nitrogen recycling has autumn periods, largely because of regular inﬂows
also begun to receive attention. of colder, oxygenated waters from upstream dam
Chambers et al. (1997) estimated the phospho- releases. Thus, even though the sediments
rus load contained in the sediments of a 3.5 km contained a large phosphorus store, there was little
stretch of the Murrumbidgee River, southern NSW opportunity for it to be released into the water
(Fig. 1). They found that none of the phosphorus column while rivers such as the Murrumbidgee have
was charge-bound to the sediments – the weakest regular inﬂows for irrigation purposes.
form of binding – implying that all such phosphorus It has been widely believed that these sediment
had already been released into the water column stores of phosphorus would be released into the
prior to the cores being extracted. They estimated water column under anoxic conditions. However,
that the potential load of weakly bound phosphorus Bourgues et al. (2000), using sediments from the
was approximately 1 tonne per km from the top Yarra River (Fig. 5), showed that phosphorus
Yarra River Sediment (25°C)
Dissolved Oxygen (mg O2/L)
5 FRP 2.5
0 1 2 3 4 5 6 7 8
Figure 5. Results from Yarra River sediment incubation experiments. Note that phosphorus ﬂuxes from these sediments remained low
even after extended periods of anoxia. (Redrawn from Bourgues et al. (2000)).
ﬂuxes did not necessarily increase even when sed- to 150 lmol N m)2 h)1 at 19 sites in the Brisbane
iments remained anoxic for some days. Similar River in southeastern Queensland, and Bourgues
results have been obtained by Bourgues et al. et al. (unpublished) measured rates ranging from
(1998) in sediments from the Hawkesbury River 150 to 380 lmol N m)2 h)1 in the Goulburn and
near Sydney and by Douglas et al. (1997). A Yarra rivers in Victoria. Thus, these ﬁrst direct
nutrient diagenesis model developed by Harper measurements of denitriﬁcation rates in Austra-
(pers comm.) accounts for these eﬀects. The model lian freshwater systems imply that denitriﬁcation
incorporates the adsorption of phosphorus by is not the major loss pathway for nitrogen from
both Fe(OH)3 and non-iron compounds as well as Australian rivers.
NH3 sorption. It predicts a very wide range of PO4 Denitriﬁcation rates similar to those reported
ﬂux responses, depending on the depth of the above were recorded for inorganic nitrogen ﬂuxes
Fe(OH)3 layer, the adsorptive properties of the Fe from sediments in Port Phillip Bay (Nicholson
and non-Fe sediment phases, bioturbation and et al., 1999). Using mass balance calculations and
other mixing processes, and the mineralisation rate direct measurements of N2 generation, it was also
of particulate organic matter. These responses found that inorganic N ﬂuxes were much lower
range from a pulse of P into the water column than expected from Redﬁeld stoichiometry, sug-
within days of anoxia involving only a small gesting that a signiﬁcant proportion of the organic
release of P, to a slow buildup and long attenua- N reaching the sediments was denitriﬁed. Direct
tion of sediment-water P ﬂuxes involving large measurements using N2/Ar ratios conﬁrmed high
loads of P over many months. These diﬀerent denitriﬁcation rates (Harris et al., 1996; Heggie
sediment response types have important implica- et al., 1999) in many sediment facies in the Bay.
tions for management of waterbodies. For exam- These studies by the Port Phillip Bay team were
ple, if waterbodies contain ‘slow responseÕ seminal in establishing the hypothesis that Aus-
sediments there may be little release of phosphorus tralian estuaries have high rates of denitriﬁcation
into the water during short-term anoxic events. and high denitriﬁcation eﬃciencies, thus suggest-
Nitrogen ﬂuxes have been measured from ing that they would be able to process relatively
sediments in a small number of Australian rivers. high inputs of nitrogen. The studies also found,
Studies by Bourgues et al. (2000) in the Goulburn however, that where sediment respiration rates
River, show that sediment-water nutrient were high (such as at the mouth of the Yarra River
ﬂuxes were variable with maximum rates of entering Port Phillip Bay) denitriﬁcation eﬃciency
5 mmol m)2 d)1 of ammonium in summer and was close to zero and most of the organic N
undetectable phosphorus ﬂuxes. NOx ﬂuxes were deposited in the sediments was returned to the
always into the sediment suggesting denitriﬁcation water as ammonium.
was actively occurring, driven by water column A later study in Moreton Bay oﬀ Brisbane,
nitrate (up to –1.7 mmol NOx-N m)2 d)1). southern Queensland, found that muddy sedi-
Bourgues et al. (unpublished) showed low rates of ments were ‘‘sensitively poised’’ in respect of
denitriﬁcation with 50% of samples below detec- denitriﬁcation eﬃciency (Dennison & Abal, 1999).
tion and maximum rates of 20 lmol N m)2 h)1 in When sediment oxygen concentrations were
river sediment samples. These rates are similar to reduced by about 50% of ambient, ammonium
rates measured by Hanington and co-workers ﬂuxes increased by a factor of 5 and nitriﬁcation
(unpubl.) in a tributary of the Hawkesbury River was suppressed. This phenomenon has not previ-
in Sydney. They found that denitriﬁcation could ously been recorded from coastal systems, but it
remove less than 1% of the total NOx load dis- suggests that the muddy sediments in subtropical
charged from a sewerage treatment plant into the Moreton Bay may be particularly sensitive to
river over a distance of 15 km. Nielsen (1992) changes in oxygen concentrations or organic
measured rates between 34 and 51 lmol N m)2 h)1 loading. Estimates of N input and N export to the
in river sites and a single rate of 122 lmol N ocean suggest that the diﬀerence between those
m)2 h)1 immediately downstream of a STP two terms is small and N loss through denitriﬁ-
releasing NOx in its eﬄuent. Udy (2001) found cation correspondingly small. However, denitriﬁ-
similar denitriﬁcation rates ranging from about 5 cation eﬃciency for the Bay calculated from the
relationship between the percentage of nitrogen results were in line with earlier studies of both soils
load removed through denitriﬁcation and log res- (Sparling et al., 1985) and sediments (Fabre, 1988)
idence time (Nixon et al., 1996) suggests that some where the phosphorus release on rewetting was
25% of the N inputs may be denitriﬁed by Mor- attributed to decomposition of killed bacteria.
eton Bay sediments. Thus, there is considerable Qui & McComb (1995) found support for this
uncertainty in the estimates of N loss through conclusion and estimated that air-drying killed
denitriﬁcation from the estuary. about 76% of the microbial biomass. Possible
Sediment ﬂux studies in Wilson Inlet, an inter- mechanisms for phosphorus release on rewetting
mittently closed estuary in southwest Western included (1) autolysis and drying-induced
Australia (Fig. 1a), conﬁrmed the importance of destruction of micro-organisms, (2) stimulation of
denitriﬁcation in regulating N ﬂuxes in Australian decomposition of organic matter during drying
estuaries. Fredericks & Heggie (2000) found N so that phosphate was ready for release after
ﬂuxes of 2–3 mmol m)2 d)1 (as ammonium) in re-wetting, (3) crystalline structural changes lead-
spring and summer, declining to around ing to inactivation of binding sites for phosphate.
1.5 mmol m)2 d)1 in winter. Denitriﬁcation eﬃ- They claimed that the ﬁrst process was dominant
ciencies ranged from around 50% in spring to be- when the P concentrations in the overlying waters
tween 70 and 80% in late summer and autumn. were relatively low, as occurred in their experi-
Contrary to ﬁndings in Port Phillip Bay there was ments. Baldwin (1996) measured the phosphate
no relationship between denitriﬁcation eﬃciency adsorption isotherms for sediments from a single
and sediment carbon load. Clearly these three de- site at a northern NSW reservoir. These sediments
tailed studies of denitriﬁcation in estuarine systems were divided into three groups (a) never dried, (b)
in southern Australia have shown that the model of oxidised but remained moist and (c) desiccated and
high denitriﬁcation eﬃciency suggested by the Port oxidised. The results implied that it was the oxi-
Phillip Bay study does not apply to all estuaries. dation, rather than the desiccation, which changed
the P absorption characteristics of the sediments.
Nutrient release from sediments after drying Conversely, Mitchell & Baldwin (1998) found
Drawdown of reservoirs is used as a deliberate that phosphorus release from dried lake sediments
management technique to control nuisance plant was almost completely suppressed after rewetting,
growth in Europe and, to a lesser extent, in irrespective of whether the rewetted sediments
Australia. More commonly, given AustraliaÕs were held under oxic or anoxic conditions. Phos-
variable climate, wetlands, storage reservoirs and phate concentrations in water overlying control
rivers dry out following periods of low rainfall. sediments, that were never dried out, reached
Major ecological changes have been observed in 2000 lg l)1 within 8 days of the onset of anoxia,
some water bodies after reﬁlling. For example, whereas there was no detectable phosphorus con-
Lake Mokoan shifted from a macrophyte domi- centration after 30 days in waters above rewetted
nated to an algal dominated system after reﬁlling sediments. Most importantly, this study added
following the 1982 drought (Water ECOsciences, strong evidence to earlier work (e.g., Gachter &
1996). A number of processes may be responsible Meyer, 1993) that showed that P release from
for these shifts including changes in nutrient sediments is a biologically mediated process. Thus,
release from sediments after re-wetting. Baldwin the addition of formaldehyde completely sup-
et al. (2002) provide a review of the literature on pressed all P release from anoxic sediments that
this topic. had not been desiccated, the addition of acetate
Qui & McComb (1995) took cores from an enhanced the release of P above that seen in the
urban wetland and subjected them to a prolonged controls and the addition of glucose slightly sup-
(40 day) period of air-drying followed by reﬂoo- pressed P release. Both carbon sources even pro-
ding under both aerobic and anaerobic conditions. moted P release from the desiccated sediments,
They found an enhanced P release from the albeit at a much slower rate than from the control
reﬂooded sediments compared to controls that sediments. They interpret these results as showing
were never dried out. The P release was greater in that bacteria play a major role in the release of
the anaerobic than the aerobic sediments. These phosphorus and that the bacteria were carbon
limited after desiccation. The apparently contrary in New South Wales, had found that the ‘‘phy-
ﬁnding by Qui & McComb (1995) of large P toplankton population of the Darling River at
releases after rewetting of desiccated sediments Burtundy appears to be most strongly inﬂuenced
arises because of the large size of their samples and by the high variability in ﬂow and changes in
the consequent load of phosphorus available from turbidity rather than by the seasonal temperature
the dead bacteria after desiccation (Baldwin, pers. pattern or availability of nutrients’’. Previously,
comm.). Sullivan et al. (1988) had found that phyto-
Given the extent of bacterial mortality follow- plankton abundance was regulated by ﬂow in the
ing desiccation, we would expect that nitrogen Murray River and Jones (1993) had come to a
processes in sediments would also be aﬀected by similar conclusion for the Murrumbidgee River.
desiccation and rewetting. Thus, Qui & McComb Jones et al. (1997) extended this work to the
(1995) found that wetland sediments exhibited rivers on the east coast of Australia and showed
increased N mineralisation, nitriﬁcation and that the major rivers fell into one of three cate-
denitriﬁcation after desiccation and rewetting, and gories. In Type I rivers there was a linear, inverse
Boon et al. (1997) reported reduced methanogen- relationship between cyanobacterial population
esis after rewetting dried sediments. However, and river ﬂow; in Type II rivers populations
Mitchell & Baldwin (1999) found little evidence of developed under prolonged low ﬂows, but not
nitriﬁcation in either the control (always wet) or always; and in Type III rivers signiﬁcant cyano-
desiccated sediments and they concluded that, at bacterial populations did not develop under either
least for their samples from a water storage in high or low ﬂows. The duration of low ﬂows, the
southern NSW, nitriﬁcation could not act as the nutrient status and the turbidity of the water body
source of NO3 for subsequent denitriﬁcation pro- determined the size of the cyanobacterial popula-
cesses. They found that denitrifying bacteria were tion, if one developed. They concluded that ﬂow
nitrate and probably carbon limited. N2 produc- was the primary determinant of the onset of
tion was similar in sediments from the control sites cyanobacterial populations, not only in the Dar-
and the desiccated sites under both unaltered ling and Murrumbidgee Rivers, but also in a wider
conditions and with the addition of nitrate and range of Australian rivers.
carbon sources to overcome input limitations. However, the mechanisms underlying the
Unlike the ﬁndings of Qui & McComb (1995), they cyanobacteria-ﬂow relationship were not clear.
did not observe any release of mineral N when the When the link between stable water and cyano-
desiccated sediments were re-wetted. However, bacterial populations in rivers had been noted
like Boon et al. (1997), they did ﬁnd that methane previously (e.g., Reynolds, 1984), ﬂow had been
production was suppressed when the desiccated thought to play a stimulating or auxiliary role. The
sediments were re-wetted compared with the con- above Australian studies suggested that ﬂow was
trol sediments. Addition of carbon sources stimu- more important than either nutrient availability or
lated methane production more in the control temperature in controlling algal growth in rivers.
sediments than in the dried sediments, as would be Increasingly, the importance of residence times
expected if the bacterial population of the latter (i.e., of ﬂow regimes) on river system behaviour is
had been reduced due to the desiccation. being recognised as a key parameter in eutrophi-
Thus the eﬀect of desiccation on N processes is cation in Australia. As pointed out by Harris
less clear than is its eﬀects on P release. Some N (1996), widely ﬂuctuating residence times in Aus-
processes are suppressed (e.g., methanogenesis), tralian rivers profoundly inﬂuence system biogeo-
others appear to be unaltered (denitriﬁcation) chemistry and thus the development and
while the results for mineralisation and nitriﬁca- sustainability of algal blooms. During times of
tion diﬀer with diﬀerent studies. drought, the residence time of water in rivers is
increased and eutrophication presents a problem.
Stratiﬁcation, turbidity and light This is exacerbated in Australia by extensive river
regulation and water abstraction, which further
Hotzel & Croome (1994), in a study of long-term
¨ reduces ﬂows in rivers in times of low precipitation
monitoring data from a site on the Darling River and runoﬀ.
Webster et al. (1997) and Sherman et al. (1998), not nutrients was limiting cyanobacterial growth.
in a study of Maude weir pool on the Murrum- Turbidity dropped below 100NTU – the critical
bidgee River, ﬁrst explained the mechanisms value for light limitation of algal growth – when
behind this ﬂow-cyanobacterial relationship in salinity increased above 300 lS cm)1. These saline
Australian rivers. Australian inland rivers tend to inﬂows came from groundwater intrusions at
be highly turbid and so heat is readily absorbed in speciﬁc locations along the river, one such plume
the surface waters during periods of low ﬂow, having a conductivity of 25,300 lS cm)1 (Wil-
leading to stratiﬁcation of stationary water in weir liams, 1993). Thus, in this river, these intrusions
pools and other impoundments. Sherman et al. trigger cyanobacterial blooms by penetrating the
(1988) showed that at discharge rates below water column when ﬂows are low and the pressure
1000 Ml d)1 the Maude weir pool was persistently in the surrounding aquifer drops. The increased
stratiﬁed. Once stratiﬁcation set in, the common salinity causes the ﬁne clays in the river to ﬂoc-
diatom (Aulacoseira sp.) sank out of the photic culate, light becomes available to phytoplankton
zone and ceased to grow because of light limita- and, providing physical and biological conditions
tion, while the buoyant Anabaena sp. dominated are suitable, cyanobacterial blooms develop.
the photic zone. Anabaena populations peaked In a subsequent study of the lower Murray
after approximately 14 days if the stratiﬁcation River (Bormans et al., 1997; Burch & Ganf, 1998),
persisted that long (Fig. 6). It appears that the it was found that stratiﬁcation occurred only
population had depleted the available phosphorus. brieﬂy during periods of low wind. Bormans &
When discharge rates increased above 1000 Ml d)1 Webster (1997) used a simple measure of vertical
the weir pool was diurnally stratiﬁed, i.e., total mixing to predict the duration of stratiﬁcation
mixing of the water column occurred daily. Under over a 6-year period, based on the meteorological
these conditions the diatom had a competitive record, and concluded that stratiﬁcation never
advantage. persisted for more than a few days at a time.
This understanding opened up the possibility of Consequently, cyanobacterial blooms should not
management strategies for river impoundments be able to develop to nuisance sizes in the ﬂowing
based on breaking stratiﬁcation through ﬂow parts of the river. In contrast to the Murrumbidgee
manipulation (Webster et al., 2000). Although the River, where ﬂow dominates the control of strati-
only times when cyanobacterial blooms were ﬁcation, wind and ﬂow are both important for
observed was when ﬂow was low and stratiﬁcation controlling stratiﬁcation in the lower Murray
was established, there were two occasions during River.
the period of study when physical conditions were More recently this work has been extended to
suitable but cyanobacterial populations did not combine the interactions between light availability,
increase as predicted in the surface layers. The nutrients, and ﬂow and stratiﬁcation in dams and
authors speculate that low nutrient concentra- weir pools in a study of the tropical Fitzroy River
tions, zooplankton predation or some other in Queensland. This river has highly episodic ﬂows
unknown factor must have limited phytoplankton driven by summer rainfall followed by many
numbers at these times. Thus, stratiﬁcation is a months of low or no ﬂow. Bormans et al. (2004),
necessary but not suﬃcient condition for cyano- studying a section of the river upstream of the
bacterial blooms in these impounded river systems Rockhampton Barrage, found a pattern of
(Sherman et al., 1998). decreasing ﬂow after tropical storms, leading to
In a subsequent study in a weir pool on the increased sedimentation after the ﬂow had ceased
Darling River, Oliver et al. (2000) came to a sim- altogether. This sedimentation sequestered a large
ilar conclusion. Cyanobacterial concentrations proportion of the nutrients, both P and N, from
were found to be inversely related to discharge, the water column. During the low ﬂow period, the
with large blooms only occurring when discharge barrage pool stratiﬁed leading to decreases in
was below 500 Ml d)1, but not on all occasions (a oxygen in the hypolimnion. Rates of oxygen
Type II river in JonesÕs scheme). Light penetration consumption were about 0.5 mg l)1 d)1 (Bormans
was critical to the development of blooms; for at et al., 2004) and all hypolimnetic oxygen was
least half of the 44 months of the study, light and consumed within 5 months. During this time water
(a) 6000 5
Discharge (MLd )
3000 2 (°C)
100 0.01 m
Turbidity (arbitrary units)
90 2.50 m
(c) 50 30000
Abundance (cells mL )
Chlorophyll (mg m )
26-Dec 5-Jan 15-Jan 25-Jan 4-Feb 14-Feb
Figure 6. Transition between diatoms (Aulacoseira spp.) and cyanobacteria (Anabaena spp.) related to ﬂow and turbidity in a turbid-
river weir pool, Maude Weir pool, on the Murrumbidgee River in southeastern Australia. (a) Discharge and diﬀerence in temperature
between top and bottom water (DT); (b) turbidity at 4 depths in the weir pool measured by beam attenuation at 880 nm and
(c) Anabaena and Aulacoseira abundance (5 m integrated sample) and in situ chlorophyll a ﬂuorescence (F2.5). Note the switch from
diatoms to cyanobacteria when ﬂow was reduced and stratiﬁcation persisted for more than about 2 weeks. (composed from Sherman
et al. (1998)).
column nitrate was taken up by the sediments and enough for nutrients to be released from sediments
denitriﬁed. Signiﬁcant growth of phytoplankton into the hypolimnion; in others, the stratiﬁcation
was not recorded before the ratio of euphotic persists long enough for anoxia to set in and for
depth to mixing depth approached unity (Fig. 7). large nutrient loads to be released from sediments.
At that stage the hypolimnion was anoxic, nitrate In some rivers, residence times are long enough for
had been completely consumed by the sediments sedimentation of particles, leading to greater light
and anoxic conditions had led to release of phos- penetration of surface water; in others, the parti-
phorus and ammonia from the sediments. The cles are ﬁne enough to remain in suspension
algal community in surface waters consisted of between ﬂow events.
a mixture of nitrogen ﬁxing and non-ﬁxing It had been widely accepted that cyanobacteria
cyanobacteria. The study showed that nutrient could use their buoyancy and ability to sequester
concentrations in the surface waters were not phosphorus to migrate diurnally between well-lit
suﬃcient to support this phytoplankton growth. surface waters during the day and nutrient-rich
However, nutrients were entrained from the bottom waters at night (Fogg & Walsby, 1971;
hypolimnion through diﬀusive ﬂux and other Ganf & Oliver, 1982). This gave cyanobacteria a
processes such as wind-induced seiching. strong competitive advantage over other algae.
This body of work has ﬁrmly established the In a recent analysis, Bormans et al. (1999)
importance of stratiﬁcation in controlling eutro- could ﬁnd no evidence that vertically migrating
phication in impoundments on Australian rivers cells reach suﬃcient depth in stratiﬁed systems to
although the eﬀects of, and the control over, enable them to take advantage of the nutrient-rich
stratiﬁcation vary as a consequence of the char- bottom waters. Rather, they found that physical
acteristics of the river section. In some rivers, mechanisms, such as periodic deepening of the
stratiﬁcation is controlled by ﬂow; in others, wind epilimnion relative to the hypolimnion, were suf-
is important as well. In some rivers, the stratiﬁ- ﬁcient to explain replenishment of the nutrient-
cation persists long enough to allow cyanobacteria depleted epilimnion to satisfy the primary
to dominate the surface waters but not long producer requirements. The most important
Zeu / Zm ~ 1
Zeu / Zm << 1
Last day well mixed
10 Jul 9 Aug 8 Sep 8 Oct 7 Nov 7 Dec 6 Jan 5 Feb 7 Mar
Figure 7. Relationship between euphotic depth, mixing depth and algal production in a weirpool on the Fitzroy river in subtropical
eastern Australia. Note that signiﬁcant algal production was recorded only once the ratio of euphotic depth to mixing depth reached
unity (Redrawn from Bormans et al. (2004)).
consequence of the overthrow of the migration Nutrients can be transported directly from catch-
hypothesis is that, during a stratiﬁcation event, ment inﬂows to phytoplankton, although the eﬃ-
cyanobacterial growth in surface waters will be ciency of this mechanism depends on a number of
limited by the nutrients that leak across the ther- factors, including the temperature diﬀerence
mocline. There are a number of mechanisms con- between the inﬂow and the storage. In Burrinjuck
trolling this leakage. Reservoir during summer, inﬂows are relatively
Deep reservoirs usually start to stratify in spring warm and are captured by the surface layer of the
as the surface waters warm. As soon as persistent storage with the consequent direct supply of fresh
stratiﬁcation is established, the oxygen concentra- nutrients into the photic zone (Lawrence et al.,
tion declines rapidly in the hypolimnion and 2000). However, the load of nutrients delivered
enhanced release of nutrients from the sediments through this process was calculated to be very
commences. The released phosphorus and nitro- small (less than 0.5% of the annual FRP load). On
gen, especially ammonia, remain in the hypolim- the other hand, the inﬂows studied in Chaﬀey
nion until the storage mixes the following May or Dam mixed with reservoir waters in the shallow
June because of seasonal cooling of the epilimnion inlet zones before dropping to the level where
coupled with suﬃcient wind to provide the energy their density matched that of the water column
for mixing. This brings a large load of nutrients (Sherman et al., 2001). These sediment and nutri-
into the photic zone where they are available to ent enriched waters then spread out into the res-
phytoplankton. Thus, Jones & Poplawski (1998) ervoir at this level (Fig. 9). Larger sediment
found in Barron Pocket Reservoir (61,000 Ml, particles dropped out during this process and the
49 m deep) in southeastern Queensland that both band of nutrient enriched water (100 lgl)1 TP,
TKN and TP concentrations increased roughly mainly FRP, in the event shown) remained at
linearly in the bottom waters during the period of 4–11 m depth for some weeks until it was gradu-
stratiﬁcation until there was up to 1.6 mg l)1 TKN ally mixed into surface waters as the surface layer
concentration diﬀerential and 0.2 mg l)1 TP dif- cooled, fuelling a phytoplankton bloom many
ferential between bottom and top waters at the time weeks after the inﬂow occurred.
of overturn (Fig. 8). There was a clear relationship between hypo-
Other mechanisms, apart from annual over- limnetic phosphorus concentrations and the mag-
turn, can transport nutrients into the photic zone. nitude of the algal response in Chaﬀey Dam in the
2.0 Total Kjeldahl Nitrogen (TKN) 0.25
Total Phosphorus (TP)
TKN (mg L )
Aug Dec Apr Aug Dec Apr Aug Dec Apr Aug Dec
1993 1994 1995 1996
Figure 8. Phosphorus (Total P) and Nitrogen (Total Kjeldahl N) data from Barron Pocket Reservoir in subtropical eastern Australia.
Data are diﬀerences in concentrations between surface and 20 m depth. Increases in concentrations are associated with periods of
stratiﬁcation of the water column in the reservoir. (Redrawn from Jones & Poplawski (1998)).
1 February 97 21 February 97 4 March 97 1 March 97 25 March 97 1 April 97
0.30 0.32 0.34 0.28 0.30 0.32 0.28 0.30 0.32 0.34 0.28 0.30 0.32 0.28 0.30 0.32 0.30 0.32
Conductivity (dS m )
Figure 9. Conductivity proﬁles in Chaﬀey Dam, eastern Australia before and following an inﬂow event in February 1997. The
horizontal line indicates the trajectory of the surface layer maximum depth. (Redrawn from Sherman et al. (2001)).
following year (Fig. 10). Maximum hypolimnetic gen in the hypolimnion at this point and, presum-
P concentration before the breakdown of stratiﬁ- ably, providing a mechanism for nutrients to mix
cation in late autumn explained 88% of the vari- up into the epilimnion. Bormans et al. (2004) also
ation in algal abundance the following year. noted a marked diurnal tilting of the thermocline in
Similar patterns were observed in Prospect Res- the shallow Rockhampton Barrage during pro-
ervoir (Schladow & Hamilton, 1995) and Burrin- longed wind events.
juck Reservoir (Lawrence et al., 2000). In these Diﬀerential cooling was found to be another
deep storages, nutrients that accumulate in the important transport mechanism in Chaﬀey reser-
hypolimnion over one spring and summer largely voir (Sherman et al., 2001). This process occurs
determine the following seasonÕs mean algal
Tilting of the thermocline provides another 30
vertical transport mechanism. Romero & Imberger 2
Mean annual biovolume (mm L )
y = -3.21 + 35.1x r = 0.88
(1999) observed a downward deﬂection of the destratifier operated
thermocline in Warragamba Dam, with warmer
water from the surface being pushed downwards 20
and a deepening of the mixed layer following wind
events (Fig. 11). Subsequently the thermocline was
re-established, potentially transporting nutrients
into the photic zone. Hamilton et al. (1995) mod-
elled how plunging inﬂows into Prospect Reser- 10
voir, downstream of Warragamba Dam,
introduced oxygen into the hypolimnion, thus
reducing the potential for nutrient and metal
release from sediments under anoxia. They also 0
observed the thermocline tilting due to a prolonged 0 0.2 0.4 0.6 0.8 1.0
wind event, followed by a series of daily oscillations Hypolimnetic TP (mg L )
of the thermocline after the wind eased and the Figure 10. Mean annual algal biovolume vs. hypolimnetic total
waterbody relaxed. These oscillations (‘internal Phosphorus concentration in Chaﬀey Dam just prior to water
column turnover at the start of the algal growing season. Open
wavesÕ) were believed to induce a turbulent mixing circles are measurements in years when a destratiﬁer was
zone when they broke on the wall of Warragamba operating (note the lower algal biovolume). (Redrawn from
Dam, accounting for an observed increase in oxy- Sherman et al. (2001)).
Figure 11. Wind speed and temperature proﬁle in Warragamba Dam near Sydney, eastern Australia. Note the increase in temperature
at the 25 m depth following a moderate wind event on 7–8 May and a much more pronounced deepening of the surface mixed layer
after prolonged winds between 9 and 11 May. Here the deepening of the thermocline lasted about 2 days. This may be a mechanism for
pumping nutrients to the surface from deeper water while at the same time oxygenating deeper water. (Redrawn from Romero &
when water cools more quickly along the margins the photic zone during rapid drawdown of a res-
of the reservoir than in its middle, primarily ervoir when there are large water releases from the
during winter but also on a small number of dam. The eﬀects on nutrient release of drying
occasions during spring and summer. The cooler, bottom sediments when these storages are drawn
denser water from the margins of the storage down have been discussed in a previous section.
ﬂows along the bottom of the lake until it meets a This knowledge about physical processes in
layer of equal density and then spreads into the deep and shallow Australian waterbodies has now
water column at that level. Sturman et al. (1999) advanced to the point where detailed hydrody-
have studied this process in a shallow lake in namic and chemical models have been developed.
Western Australia and have developed a simple The Princeton Ocean Model (Blumberg & Mellor,
model of this convective process. They comment 1987) has been modiﬁed to predict the physical
that, even for heavily sheltered waterbodies which behaviour of stratiﬁed river pools (Bormans &
would normally be assumed to be stratiﬁed and Webster, 1998) and the growth of a non-buoyant
anoxic, diﬀerential cooling eﬀectively prevents diatom and a buoyant cyanobacterial species
stratiﬁcation occurring and can lead to the under stratiﬁed conditions (Bormans & Condie,
exchange of nutrients between deeper and surface 1998). The model has been developed speciﬁcally
waters. In fact, Sherman et al. (2001) found that for the conditions observed in the Maude weir
this process, which has long been recognised but pool and so is driven by the stratiﬁcation process.
little studied, was the most important transport It omits both the eﬀects of nutrient limitation on
process during winter in Chaﬀey Reservoir and algal growth and the advection of algal species
was believed to account for the occurrence of downstream because neither was observed to be
winter cyanobacterial blooms. important in the Maude study. Also, the DYR-
The above mechanisms operate regularly during ESM hydrodynamic model of water storages has
the annual cycle. Lawrence et al. (2000) and Sher- been extended to include water quality (Hamilton
man et al. (2001) describe other mechanisms that & Schladow, 1997; Schladow & Hamilton, 1997).
operate spasmodically. These include the entrain- It has been applied successfully to a variety of
ment of highly nutrient enriched pore waters into water storages (Hamilton, 1999).
Stratiﬁcation drives estuarine processes, of phytoplankton in the dam, with algal species
including phytoplankton production in cold tem- composition and succession determined by a
perate Tasmania (CSIRO Huon Estuary Study complicated interplay of ﬂow, temperature regime,
Team, 2000). The Huon estuary is characterised mixing and nutrients. During times of high
by a shallow, fresher layer ﬂowing seaward over a inﬂows, particularly in spring, inputs of silica from
deeper, more saline layer ﬂowing more slowly up- the catchment and turbulent mixing appear to
river. This is similar to the characteristic physical favour diatoms. Later, during summer, inﬂows are
structure of Northern Hemisphere temperate es- reduced and thermal stratiﬁcation becomes estab-
tuaries. Nitrogen budgets suggested that inorganic lished. During that period green algae dominate
N advected in marine bottom waters and driven the phytoplankton. As nutrients are reduced in the
by the two layer physical circulation dominated epilimnion during mid to late summer a number of
inﬂux of bioavailable nitrogen to the estuary. possible scenarios emerge, depending on bioavail-
Based on these physical conditions, the study able nitrogen. When nitrogen is available as
postulated: strong light limitation of phyto- nitrate, motile green algae tend to be dominant;
plankton in winter but less in spring and summer; when ammonia dominates, non-N-ﬁxing cyano-
light limitation of benthic algae all year round; bacteria are favoured; and when bioavailable N is
depletion of surface nutrients by microalgae; and, very low, the phytoplankton is dominated by
favourable conditions for dinoﬂagellates in summer nitrogen ﬁxing cyanobacteria.
and autumn. These favourable conditions included
a stable stratiﬁcation regime in which they could Food chain interactions – biomanipulation
migrate vertically to optimise their position in the Much of our understanding of predation on algae
water for access to light and nutrients, and the comes from attempts to manipulate food chains in
potential to maintain a position which prevented lakes in order to reduce nuisance phytoplankton
them from being ﬂushed out to sea. Dinoﬂagellates (Shapiro et al., 1975; Shapiro & Wright, 1984;
may also be able to scavenge nutrients, particularly Carpenter et al., 1985). ‘‘Top-down’’ control
nitrogen, from the marine bottom waters advected through biomanipulation (Shapiro et al., 1975)
into the estuary and thus avoid being limited by was seen as an alternative to the more traditional
nutrients delivered by land runoﬀ. ‘‘bottom-up’’ approach through reductions in
nutrient inputs. In a review, McQueen (1998)
Ecological processes in freshwaters suggests that, although biomanipulation can ini-
tially produce quite spectacular results, longer-
Algal succession in water storages term investigations frequently show that these
In their study of Chaﬀey Dam, Sherman et al. eﬀects are not sustained and the resulting algal
(2001) recorded 58 taxa of phytoplankton in the biomass is as high or higher than before the
11 years of sampling between 1987 and 1997. They manipulation began (see also DeMelo et al., 1992).
found no clear algal successional sequence, This implies that biomanipulation is not a one-oﬀ
although they noted patterns of occurrence. Thus, activity and that monitoring and further inter-
many of the major taxa showed marked increases vention will be required.
in biovolume during spring and early summer. During the early 1990s there was a vigorous
Ceratium biomass was usually low during spring debate in Australia about whether biomanipula-
and summer, but their biomass increased during tion would work in local waters (Boon et al., 1994;
autumn together with a number of other major Matveev et al., 1994b). There are now a number of
taxa (Microcystis, Anabaena and Aulacoseria). studies from Australia that show the potential for
Cyanobacteria attained their highest biomass zooplankton grazers to control phytoplankton. In
during mid to late summer and autumn. particular, it has been shown that Daphnia cari-
Lawrence et al. (2000) have reviewed water nata, a large cladoceran, can be present in high
quality and algal data from the period 1976–1998 densities in Australian systems (Mitchell & Williams,
from Burrinjuck Reservoir, a water storage with a 1982) and that it eﬀectively grazes phytoplankton
capacity of 1026 Gl and a surface area of about (e.g., Ganf & Shiel, 1985; Merrick & Ganf, 1988;
6000 ha. There was a general successional pattern King & Shiel, 1993; Kobayashi, 1993; Kobayashi
et al., 1996, 1998; Matveev & Matveeva, 1997). In The triggers of akinete formation and germination
addition, D. carinata is able to feed on cyano- are very unclear. Paerl (1988) suggested that it was
bacteria (Kobayashi, 1993) while being resistant to caused by nutrient levels, and Van Dock and Hart
certain cyanobacterial toxins (Matveev et al., (1996) showed that P-limitation triggered diﬀer-
1994b). Kobayashi et al. (1998) found that the entiation in a cultured strain of Anabaena circi-
weight speciﬁc clearance rate of D. carinata was nalis. However, Baker (1999) could ﬁnd no
between two and 65 times higher (0.27–3.83 l mg- correlation between environmental variables,
dw)1 d)1) than that of the resident zooplankton including nutrient concentrations, and akinete
community in a reservoir in Sydney. This suggests abundance in two species of Anabaena in the lower
that D. carinata is eﬀective at reducing phyto- Murray River. Blackburn & Thompson (2000)
plankton where planktivorous ﬁsh are absent or concluded that akinetes form and germinate lar-
occur in low numbers and where phytoplankton gely in the same conditions that are suitable for
production is suﬃciently high to sustain large optimum vegetative growth and that phosphorus
D. carinata populations (see also Burns, 1998). They limitation is not a primary cause of akinete for-
stress that, although D. carinata can eﬃciently graze mation. They found, however, that diﬀerent
down phytoplankton communities in lakes, it is not strains of A. circinalis held under diﬀerent labo-
clear that they are suitable for removing nuisance ratory conditions react in diﬀerent ways to P lim-
algae, the targets of such management initiatives. itation. For example, in one strain they showed a
Burns et al. (1989) and Forsyth et al. (1992) have correlation between akinete formation and P lim-
shown that the ﬁlamentous Anabaena reduce the itation in batch culture, but not in a chemostat.
grazing rate of D. carinata. Boon et al. (1994) These anomalies make it diﬃcult to extrapolate
showed that, during a bloom of two Anabaena laboratory data to ﬁeld conditions.
species in the Darling River, only a small number of Blackburn and Thompson (2000) also investi-
D. carinata had ingested these cyanobacteria. gated A. circinalis dormancy requirements. They
The ﬁrst full-scale trial of the eﬀectiveness of found that not all strains formed akinetes and that
top-down control of phytoplankton in Australia mixing diﬀerent strains did not enhance akinete
was commenced in 1998 in two reservoirs in formation, i.e., there was no evidence of chemical
southeastern Queensland. According to the tro- cues between diﬀerent strains. No endogenous
phic cascade model (Carpenter & Kitchell, 1993; dormancy requirements were found and akinetes
Carpenter et al., 1985) by reducing the biomass of germinated readily under the right environmental
planktivorous ﬁsh, pressure on zooplankton conditions (Jameson & Blackburn, pers. comm.).
grazers is reduced and, in turn, their grazing on However, the cell growth rate of germlings was
algae will increase. Australian bass were intro- greater than that of vegetative cells. This may be
duced into one of the reservoirs to act as predators important in development of Anabaena blooms.
of the planktivorous ﬁsh. Data available to date
indicate that some changes may have occurred in Ecological processes in estuaries
the zooplankton and phytoplankton communities
of the reservoir after addition of Australian bass Stratiﬁcation and anoxia
(Macquaria novaemaculata) compared to the con- A characteristic of many Northern Hemisphere
trol reservoir which was not stocked. However, no estuaries is that they are stratiﬁed vertically, lead-
clear pattern has emerged and it will be some years ing to hypoxic or anoxic bottom waters (Diaz &
before it is clear whether the introduction of the Rosenberg, 1995; National Research Council,
bass has led to the diﬀerences in zooplankton and 2000). In that regard, estuaries in Tasmania, such
phytoplankton communities between the treat- as the Derwent and Huon estuaries near Hobart,
ment and control reservoirs. are more similar to those in the Northern
Hemisphere than to those in mainland Australia.
Algal reproduction Consequently, the drivers of ecological processes
Cyanobacteria can form resting stages (akinetes) are expected to be similar in these estuaries to those
which can survive for many years in sediments and in the northern hemisphere. A recent detailed study
germinate when conditions become favourable. of the Huon estuary has conﬁrmed this (CSIRO
Huon Estuary Study Team, 2000). The study found marked eﬀects arise from ﬂuctuations in the way
that the estuary was a highly stratiﬁed salt wedge organic matter is processed and in the diﬀerent
estuary with strong riverine input and, conse- nutrient cycling pathways between open and
quently, relatively short water residence times of closed states. A 3-year study in Wilson Inlet in
hours to days in the surface layer and days to southwest Western Australia (Fig. 1) found that
1 week for the whole water column. ecological function could be described in four
Most estuaries in mainland Australia are very distinct phases (Fig. 12):
diﬀerent from those in Tasmania, with longer
residence times and irregular river inﬂows as the 1. Dry with closed entrance (mid-summer [Jan] to
major drivers of ecological functions. Although the start of the winter rains [late May]): River-
there are a number of large estuaries formed by ine input to the estuary is low, the entrance
drowned river valleys (e.g., the Hawkesbury near is closed. Wind mixing generates a uniform,
Sydney, NSW), most tend to be small and shallow vertically mixed water body. Phytoplankton
with intermittent contact with the ocean. These production is low (<500 mg C m)2 d)1),
types of estuaries are thought to be well mixed probably due to low nutrient levels of both N
most of the time with few examples of extensive and P. Chlorophyll a levels do not exceed
hypoxia in bottom waters. Detailed studies in 2 lg l)1.
Wilson Inlet, southwest Western Australia 2. Wet with closed entrance (winter [Jun–Aug]):
(Tompson & Twomey, 2000), have shown that Winter rains increase freshwater ﬂows to the
periodic anoxia does occur even in shallow, see- estuary, supplying an estimated 20–30% of the
mingly well mixed systems characteristic of most annual input of both N and P. The water body
of the coast of Australia. This has now been is well mixed. Inorganic N levels in the water
recorded in a number of other Australian estuaries increase to 2.0–3.6 lmol l)1, but chlorophyll
(NSW EPA, unpubl. data). The anoxic layer in levels and primary production remain low.
these systems tends to be very thin, patchy and 3. Wet with open entrance (spring [Aug–Sept]):
ephemeral and is probably driven by saltwater When the water level reaches 1 m above mean
incursions from the ocean that establish short sea level the estuary entrance is artiﬁcially
periods of stratiﬁcation. Biochemical oxygen de- opened, followed by outﬂow of fresher estu-
mand by sediments can then rapidly draw down arine water and a much smaller counter ﬂow
the oxygen in the thin, saline bottom layer, leading of saline oceanic water into the estuary setting
to hypoxic or anoxic conditions. This has sig- up a stratiﬁed water column. The saline water
niﬁcant implications for sediment nutrient cycling, forms a thin lens of dense bottom water,
and may decrease the denitriﬁcation potential of within which oxygen is rapidly drawn down.
these systems for periods of time. This leads to a hypoxia-driven sediment
nutrient release of both NH3 and PO4 (Fred-
Eﬀects of estuarine entrance dynamics ericks & Heggie, 2000). The annual peak in chl
Many of AustraliaÕs estuaries fall into the category a occurs some 24 days (23 ± 2.9 – remark-
of ICOLLs, i.e., they have mouths that are closed ably consistent from year to year) after strat-
for periods of time, occasionally for many years iﬁed conditions have been established.
(Roy et al., 2000). This is a feature that Australian 4. Dry with open entrance (spring [Sept] to
estuaries share with estuaries in other Southern summer [Jan]): Precipitation and runoﬀ
Hemisphere countries (e.g., Koop et al., 1983). decrease with easing of winter rains and the
Isolation from the ocean imposes signiﬁcant con- channel to the ocean shoals and eventually
straints on the estuarine fauna and may, in special closes by accretion of marine sands. Nutrient
cases, wipe out all of the benthos when the con- concentrations in the estuary return to low
nection is restored, as in the case of the Bot River levels and chl a and phytoplankton produc-
estuary in southwestern South Africa (Koop et al., tion is low because of nutrient limitation.
1983). It also depopulates ﬁsh species that are
dependent on migration to the sea for breeding. In This study clearly demonstrates that opening
terms of eutrophication, however, the most the mouth of the estuary after a period of isolation
1995 1996 1997 1998 1999
Ammonium (µgL )
Jan July Jan July Jan July Jan July Jan July
1995 1996 1997 1998 1999
Figure 12. Salinity and ammonium data during the period 1995–1999 in Wilson Inlet, southwestern Western Australia. Note high
salinities in bottom waters of the estuary immediately after the mouth opened to the sea, leading to hypoxic conditions in bottom
waters (not shown), which resulted in increased ﬂuxes of ammonium from sediments. (Redrawn from Thompson & Twomey
triggers a predictable set of events. Thus, the lag Western Australia (Fig. 1). Most of the nutrients
between opening of the mouth and peak phyto- were taken up by epiphytes, highlighting the
plankton biomass in the estuary was found to be a importance of the macrophytes as substratum for
remarkably consistent 50 ± 14.8 days. Further these physiologically active plants. The study also
studies of similar systems will need to establish showed that Ruppia is able to take up nutrients
whether the pattern detected in Wilson Inlet can be through its leaves. Thus, the macrophyte beds
generalised across a range of ICOLL systems. represent signiﬁcant competitors for phytoplank-
ton and healthy macrophytes decrease the likeli-
Macrophytes hood of phytoplankton blooms in these systems.
There have been relatively few studies of the eﬀects Australia has the greatest diversity of seag-
of nutrients, nitrogen and phosphorus on macro- rasses in the world, with 30 species recorded, and
phytes in Australian estuaries, although a study by the greatest area of temperate seagrass beds
Dudley and Walker (2000) has highlighted the (Larkum et al., 1989). The total seagrass area in
importance of Ruppia megacarpa in maintaining Australia is around 51,000 km2 (Australian State
low nutrient concentrations in Wilson Inlet in of the Environment Committee, 2001), although
there have been signiﬁcant declines in seagrass toplankton growth, but a more complex range of
meadows in Australia in recent years. Thus, responses, including uptake by benthos, vascular
Kirkman (1997) estimated that about 450 km2 of plants, microbial processes and changes in food
seagrass had been lost in the preceding decade web structures.
through anthropogenic stress, including nutrient In Australia, management questions are also
enrichment (see also Abal & Dennison, 1996). moving estuarine scientists from studying indi-
vidual phenomena towards an understanding of
Importance of catchment loads system function. The most common question to-
Most Australian estuaries are sinks for the day is how estuarine ecosystems will respond to
majority of the material exported from the catch- increased development pressures on the coastal
ment and so catchment exports profoundly inﬂu- zone. Since the development of an ecosystem
ence estuarine processes (Harris, 1996, 2001; process model for Port Phillip Bay (Murray &
Scanes et al., 1997). There is a range of models Parslow, 1999), there have been a number of at-
available for estimating catchment exports, but the tempts to model a range of ecological processes
majority require extensive data for calibration. involved in estuarine eutrophication in response
The paucity of data exclude the widespread use of to changes in nutrient loading (Gippsland Coastal
these catchment models. The studies in Port Phillip Board, 2001; Coade, 2002). Initially, these models
Bay (Harris, 1996) highlighted the importance of have predicted the eﬀects of nitrogen as a single
estimating catchment loads because of predictions stressor.
that, above a certain threshold load, denitriﬁcation
potential would be drastically reduced and the Bay
would suﬀer serious eutrophication problems. Knowledge gaps
Equally, Dennison & Abal (1999) state that muddy
sediments in some parts of Moreton Bay are likely Although knowledge about eutrophication has
to be sensitive to increases of organic loadings advanced considerably in both freshwater and
resulting in drastic reductions of denitriﬁcation estuarine systems over the last decade in Australia
eﬃciency. there are still some important research topics that
Scanes et al. (1997) have collated estimates of need to be pursued. The following section is
catchment exports for a range of estuaries in New divided into two parts: topics where basic research
South Wales and attempted to relate these to is required to understand processes, and topics
perceived levels of estuarine degradation. Signals where the ﬁndings of the last decade need to be
at either end of the spectrum, i.e., near-pristine generalised and tested before they can be turned
systems and highly degraded systems were very into practical tools for managers. Given the recent
well correlated with the estimates of catchment advances, the majority of the research eﬀort
loads, but the majority of systems with interme- should be directed towards the latter topics.
diate levels of disturbance had much less clear
relationships. In attempting to relate catchment New knowledge generation
loads to estuarine nutrient concentrations, the
relationship became even less clear. Nitrogen cycle
Cloern (2001) has pointed out that estuarine The research reported above makes it clear that
systems appear not to react to nutrient loadings nitrogen can be as important as phosphorus in
in the same way that freshwater lakes do driving the development of algal blooms in fresh-
(Vollenweider, 1976). In estuaries, there is only a waters at particular times of year. Consequently,
poor correlation between loads and phytoplank- there has been an increase in the research into the
ton biomass. In an earlier review of a large aquatic N cycle in freshwaters in recent years,
number of estuarine studies, Borum (1996) had mainly concentrating on 1-D processes in the
demonstrated only a low correlation between sediments and between the sediments and the water
annual loads of N and phytoplankton production. column. However, there is only a limited under-
Cloern (2001) argues that coastal eutrophication standing of N sources and transport processes at
does not exhibit a single response, such as phy- present, as well as nitrogen loss mechanisms.
Ground–surface water interactions toxins. Topics that need investigation include:
Groundwaters can be a major source of dissolved triggers for toxin release from cells; cheap reliable,
contaminants during times of low ﬂow, when algal robust ﬁeld-tests for presence of toxins; and novel
blooms are prevalent. Apart from some empirical methods for degrading toxins.
evidence, there is very little understanding of the
role that groundwaters may play in triggering and Phytoplankton ecology
sustaining phytoplankton blooms in shallow water While knowledge relating to physical and chemical
bodies. Thus, there is evidence that the 1991/2 controls over eutrophication has advanced greatly
Darling River bloom was driven by groundwater in the last decade, there is still only a limited
intrusions rich in both sulphate and salts; moni- understanding of the ecological interactions
toring data also show that there are high nitrate involving phytoplankton and cyanobacteria. More
levels (>10 mg/l) in groundwaters adjacent to riv- sophisticated biomanipulation methods and
ers, particularly in the southern part of the MDB bloom predictions could eventually emerge from
(Bolger & Stevens; 1999); and the research improvements to our understanding in this area.
reported above shows that phosphorus can Topics that could be pursued in this area include:
potentially enter streams via sub-surface ﬂows. preferential feeding by zooplankton; viral, bacte-
Primary research is needed into these transport rial and fungal infections of cyanobacteria; and
pathways and the inﬂuence of nutrients, sulphate habitat preferences of algal predators.
and salts on the formation and maintenance of
algal blooms. Akinetes and phytoplankton cysts
Although recent research has cleared up some of
Biogeochemical processes the confusion about the formation and germina-
Ecological processes in waterways are largely dri- tion of akinetes in Anabaena circinalis, there is still
ven by delivery of organic matter to the water and very little understanding of the role that these
microbially mediated transformations in situ, par- seeds play in the life-cycle of Anabaena, or other
ticularly in sediments. In spite of the advances cyanobacterial species. The ﬁnding that cells
described above, our understanding of microbial grown from akinetes multiply more rapidly than
processes in both fresh and estuarine waters in vegetatively reproduced cells illustrates how this
Australia is still rudimentary. For example, nutri- knowledge could be important for the manage-
ent transformation processes in sediments ment of blooms.
(including nitriﬁcation–denitriﬁcation couplings)
have been studied in only a handful of systems. Generalising our current understanding
More information is required on how sediment
nutrient ﬂuxes inﬂuence primary production in the In order to develop sound eutrophication man-
water. In another example, there is good empirical agement strategies, it is important to be able to
evidence that inorganic phosphorus ﬂuxes from move from detailed studies at speciﬁc locations to
sediments are mediated by microbial action, but generalisations that are applicable to typical sys-
there is very little understanding of the processes tems. A framework showing how this generalisa-
involved. tion typically proceeds is shown in Figure 13.
Algal biotoxins Linkages between catchment and in-water processes
Approximately 50% of sampled algal blooms in Catchment nutrient loads are the principal drivers
the MDB waters are toxic (Baker & Humpage, of ecological processes in receiving waters. The
1994). However, the major R&D programs funded loads to the receiving waters generally increase
in the last decade did not address the issue of with increased levels of disturbance. Although
toxins. A workshop held in 1998 (CRC Water highly variable across Australia, most of the pro-
Quality and Treatment, 1999), concluded that cesses that deliver nutrients to waterways are now
research into this topic was urgently needed well known. Frequently, however, suﬃcient data
because of the potential human, animal health, are not available to drive standard catchment
and environmental eﬀects from cyanobacterial models. In addition, many of the processes that we
Figure 13. Process for generalising research results from speciﬁc studies and developing management tools.
now know are important in delivering nutrients to ventions. While there is a small number of site
waterways – gully erosion, macropore pathways, speciﬁc ecological-physical models (Harris et al.,
streambank collapse, ﬂoodplain deposition and 1996; Bormans & Webster 1999; Hamilton &
entrainment – are not included in the models Turner, 2001) we need to gain more experience in
developed in the northern hemisphere. Recently, applying these models to diverse waterbodies
Young & Prosser (2001) have developed models of before suﬃcient conﬁdence can be gained to apply
nutrient loading in Australian rivers (NLWRA, them as routine management tools.
2001) that are suited for data-poor situations and
that incorporate some of these nutrient transport Biomanipulation
processes. These models need to be expanded to Preliminary results from the biomanipulation
include dynamic processes and with ﬁner spatial project in deepwater storages in SE Queensland
resolution to make them more relevant for suggest that the technique may work in Australian
regional managers. conditions. Assuming that the technique continues
to show promise, it should be trialled in other deep
Linkages between in-stream ecology and storages where experience can be built up under
hydrodynamics diﬀerent operating conditions, diﬀerent climate
Hydrodynamics is clearly the principal driving regimes and in using other native ﬁsh species.
force in many waterbodies and models of these Other food chain manipulations, apart from top-
physical processes are well advanced, while models down predation, should also be trialled.
of freshwater sediment-water column interactions
are at an earlier stage of development. Both these Modelling tools for managers
types of models need to be linked with in-stream There is a demand for user-friendly modelling
ecological models in order to assess the ecological tools that can be used directly by natural resource
eﬀects arising from various management inter- managers without recourse to the primary scientists.
A recent promising way of achieving this has been salinity, eutrophication and pesticides, and the
to provide managers with a simple interface to a recent National Action Plan for Salinity and Water
look-up table that has been developed from the Quality also requires quantiﬁable targets. The
detailed scientiﬁc modelling of a large number of above research shows that the setting of targets for
‘cannedÕ scenarios (Gippsland Coastal Board, eutrophication must involve the main factors
2001; NLWRA, 2001; Coade, 2002). In this way causing nuisance algal growths, not just phospho-
the scientiﬁc models and their outputs are vetted rus and nitrogen targets as in the past. Thus, in the
by scientiﬁc experts, while managers do not need MDB, targets for nuisance algae would have to be
to acquire the specialist expertise required to run set taking light (turbidity), stratiﬁcation (ﬂow) and
these models when wanting to explore options for nutrients into account. There is suﬃcient infor-
managing particular aquatic ecosystems. mation available for this to be done in the MDB
and possibly southwest Western Australia and
Implications for management some coastal Queensland rivers.
Monitoring is a critical part of ensuring that
The outputs from the eutrophication research these targets are met. The design of the monitoring
conducted over the last decade in Australia have program, like the establishment of targets them-
provided a range of management actions that are selves, must be based on a thorough knowledge of
likely to be eﬀective in the Southern Hemisphere. the underlying processes to ensure that the critical
Many of the options are not new, but the new parameters are monitored at the appropriate
understanding will help managers make better intervals and locations, and that the data collected
choices among the range of options. can be interpreted correctly.
Appropriate modelling – solid foundation
Diﬀuse sources dominate nutrient input to most for management
Australian rivers. Techniques to control these
sources are relatively mature in Australia, given The site-speciﬁc insights that have been gained in
the advances in knowledge about nutrient sources the last 10 years need to be generalised to a range
and pathways and the formation of community of diﬀerent waterbodies. The development and
groups to tackle this form of pollution. application of computer models, based on these
For waterbodies where the volume of annual insights, is a step in that direction. However, these
inﬂows is much less than the volume of the water- models often require input data that simply are not
body, management of catchment nutrient sources available without prohibitive investment. Con-
may not immediately lead to a reduction in algal struction of simple, conceptual models and deci-
production and biomass because of the dominance sion trees (albeit based on a modern understanding
of internal nutrient sources. In these larger water- of the processes involved) provide an alternative.
bodies, internal nutrient sources can be managed While they do not have the predictive capability of
once the particular mechanisms that govern nutri- quantitative computer models, they can be applied
ent release and transport from the sediments in that relatively easily to broad classes of waterbodies
waterbody are identiﬁed. The research reported using readily available data. The study of Moreton
above provides the basis for this identiﬁcation. Bay in southeastern Queensland illustrates
the power of this approach. Simple diagrams of
Target setting and monitoring the processes that control the eutrophication of the
Bay, based on an understanding of the processes
There is increasing emphasis these days on estab- occurring (Dennison & Abal, 1999) have generated
lishing region-speciﬁc targets that reﬂect desirable sustained investment in a wastewater strategy
and achievable standards for water quality within integrated with an ongoing research and
speciﬁc regions. Thus, the MDBC has endorsed an monitoring eﬀort. Furthermore, these diagram-
Integrated Catchment Management Strategy that matic models can be used widely in managing si-
includes quantiﬁable targets for issues such as milar coastal ecosystems.
In another example, Oliver et al. (2000) have these ecosystems, phytoplankton biomass may be
developed binary decision trees for a weirpool on able to be controlled by controlling the light cli-
the Darling River and for a lake (Fig. 14) mate, either through maintaining turbidity or
integrating the main factors leading to the risk of deepening the surface layer in which buoyant cells
blooms or the likely composition of blooms. are circulating. While it is feasible to deliberately
Neither of these trees, by themselves, allows induce turbidity (through strategies such as
management interventions to be assessed but they upstream releases of turbid water) precisely when
do provide a structured approach to considering cells are starting to grow rapidly, the practicality
interventions. of this technique has yet to be assessed. Increases
in turbidity will aﬀect all primary producers and so
the techniques may have undesirable eﬀects on
Controlling catchment sources of N and P macrophytes and benthic microalgae.
Deepening the surface layer can also control
Subsurface sediments have been identiﬁed as the the light climate. This technique has been used in
predominant sources of particulate phosphorus in non-Australian waterbodies and an initial trial is
the MDB (except in areas of intensive agriculture) being undertaken in Adelaide of this technique
and presumably in other heavily eroded, poorly (Burch et al., 2000). Thus far the trial has dem-
fertilised parts of the country. Gully erosion and onstrated that the surface layer can be deepened
slumping of streambanks need priority attention in but algal cell numbers have been too low to show
these areas. Healthy riparian vegetation may whether this will control phytoplankton numbers.
reduce the inﬂux of P to rivers more by stabilising Sherman et al. (2000) have shown that this tech-
streambanks than by reducing overland transport nique may not work in all situations. The cost
of nutrients. eﬀectiveness of the technique also needs to be
Surface erosion processes dominate in crop demonstrated.
areas of tropical Queensland and southwest Wes-
tern Australia. The sources of phosphorus diﬀer in
these two areas; excessive fertiliser applications are Flow management
most likely responsible in Queensland, whereas
low adsorptive sandy soils transmit even moderate We now know that stratiﬁcation in deep storages,
fertiliser applications to steams through surface shallow weir pools, and in many shallow estuaries
and shallow sub-surface pathways in Western is an important driver for eutrophication because
Australia. Management programs, focussing on of its profound eﬀect on mobilising internal sour-
reductions in fertiliser application, are already ces of nutrients from the sediments. River ﬂows
underway in these regions to limit nutrient losses can be used to break the stratiﬁcation in weir pools
to waterways. (Webster et al., 2000) and so remove the advantage
Knowledge about catchment sources, trans- that buoyant nuisance species of algae possess
formations and transport of nitrogen is less (Oliver & Ganf, 2000) as well as preventing anoxic
advanced than that of phosphorus. Nevertheless, conditions becoming established in bottom waters.
it appears that the riparian zone plays an This technique has been trialled with success in
important role in promoting nitrogen transfor- some inland rivers of the MDB.
mation processes, including denitriﬁcation, and so In coastal rivers, algal blooms sometimes
establishment and management of vegetated occur upstream of the salt wedge, particularly
riparian areas is a critical part of diﬀuse nitrogen during low ﬂows (Kerr et al., 1996). Water
management. releases from upstream water storages can break
the salt wedge and eﬀectively ‘‘ﬂush’’ the system.
Light limitation reduces algal growth Trials are presently being conducted in the
Hawkesbury-Nepean River near Sydney to gauge
It is now clear that primary productivity is limited the eﬀectiveness of this technique and to assess
by light availability for signiﬁcant periods of the whether it has adverse eﬀects on other parts of
year in many Australian aquatic ecosystems. In the ecosystem.
Figure 14. A decision support tree identifying environmental conditions leading to particular types of cyanobacterial blooms along
with indicator threshold values for the interactions. Key: Bclad biomass of cladocerans, Bphyto biomass of phytoplankton, ZEU euphotic
depth, ZMIX depth of mixing, u* shear velocity, V ﬂoating or sinking velocity of phytoplankton, tw time that the wind blows, L lake
fetch, cs current speed. (Redrawn from Oliver & Ganf (2000)).
Managing internal loads The vertical position of nutrients that are
injected into a storage can be simply estimated
Algal biomass is usually limited by nutrient from knowledge of the temperature and salinity of
availability in the euphotic zone once there is the inﬂows compared with the temperature and
suﬃcient light. There are many pathways for salinity proﬁle of the storage. While not in a
nutrients to reach the euphotic zone. In many deep position to control these inﬂows, the manager can
storages, the annual autumnal mixing of bottom anticipate their consequences. Thus, injections into
waters into the euphotic zone carries nutrients the euphotic zone (especially surface injections) are
released from sediments during extended periods likely to lead to rapid algal growth if algae are
of anoxia the preceding spring and summer. In nutrient limited. If the nutrients are injected below
these cases, management can focus on preventing the euphotic zone, then they will be sequestered
hypolimnetic anoxia or suppressing sediment until the surface layer deepens enough to bring
release of nutrients. Aeration of bottom waters them into the euphotic zone. Thus, they can fuel
may work in some cases, although energy costs can blooms some months after the inﬂows. If the
be high. Breaking stratiﬁcation through devices inﬂows are dense enough, they can ﬂow along the
such as bubble plumes can also be eﬀective if the bottom, re-oxygenating the bottom waters and
energy costs are not excessive. However, if turned shutting down sediment nutrient release for a
on too late after anoxia has set in or after deep period. With well-targeted monitoring and good
intrusions of nutrient rich fresh inﬂows have scientiﬁc advice a water storage manager can be
occurred, these devices can be counterproductive prepared for these events.
and promote algal blooms by mixing large loads of
nutrients from the hypolimnion into the euphotic Nitrogen vs. phosphorus
zone (Sherman et al., 2001). They need to be
operated with a good understanding of the The current consensus in Australia is that both
waterbody. Alternatively, anoxia can be left to nitrogen and phosphorus, rather than just one
occur but the sediments can be capped with both supposedly limiting nutrient, need to be considered
passive and chemically active barriers (Azcue when developing management strategies to reduce
et al., 1998), including modiﬁed clays that bind nutrient inputs to waters. This is important when
phosphorus even under strong anoxia (Douglas considering options for managing point sources,
et al., 1999). such as STP. The Moreton Bay Catchment Water
Nutrients can be transferred throughout the Management Strategy for southeastern Queens-
year by seiches that can set-up internal waves that land committed to drastic reductions of nitrogen
break on shores and mix bottom waters to the in sewage eﬄuent after the importance of N in
surface. In some circumstances, planting wind- eutrophication had been demonstrated (Dennison
breaks on strategic shores may reduce the extent of & Abal, 1999).
seiching, but this is likely to be successful only if The origins, transformations and movement of
detailed hydrodynamic modelling supports the nitrogen through the landscape is less well known
strategy. than are the sources and movement of phos-
For water storages that are periodically heavily phorus. There is a growing consensus (Hart &
drawn down, Sherman et al. (2001) suggest that Grace, 2001) that nitrogen loads need to be con-
sediment nutrient releases can be inhibited by trolled either at source or by creating favourable
using these periods to kill the bacteria in the sed- conditions for denitriﬁcation in receiving water
iments that drive the nutrient release mechanisms. bodies. Diﬀerences in hydrology mean that river
However, they note that this technique is not sediments are able to denitrify a relatively small
suitable if downstream users are dependent on proportion of the N load, whereas estuaries may
adequate water supplies. Lawrence et al. (2000) remove more than 50% of the N load and fre-
also point out that rapid drawdown from epilim- quently up to 70–80%. Thus maintaining or cre-
netic waters will quickly reduce the epilimnion ating favourable conditions for denitriﬁcation in
thereby promoting blooms by bringing nutrient estuaries will have a relatively greater eﬀect on N
enriched waters into the euphotic zone. removal than it will in rivers.
Although there remain concerns about the relationship between ﬂow and cyanobacteria
potential for promoting nitrogen ﬁxing cyano- abundance observed in rivers of the MDB the
bacterial species if algae become nitrogen limited, early 1990s.
there is evidence that nitrogen ﬁxing species, be- It is also clear that the characteristics of the
cause of their special physiological requirements, Australian environment do inﬂuence the occur-
do not always take advantage of nitrogen deﬁcient rence of algal blooms. Relatively low settlement
conditions. densities mean that diﬀuse nutrient sources pre-
dominate over much of the country; episodic
rainfall patterns, particularly in tropical regions,
Integration with other natural resource mean that the water in even large storages can be
management initiatives replaced within days with major eﬀects on nutrient
dynamics and possibly ecology; the impoundment
Eutrophication is only one of a range of environ- of most inland rivers of South-East Australia for
mental issues facing natural resource managers in irrigation means that signiﬁcant sections of these
Australia. Salinisation of Australian dryland rivers stratify during the summer creating ideal
areas, for example, is currently a high priority. physical conditions for cyanobacteria; and the
Revegetating catchments and installing engineer- high turbidity of many rivers and some estuaries
ing works to control dryland salinity will inevita- from sediments makes the phytoplankton light
bly modify runoﬀ and the delivery of sediments limited for signiﬁcant periods of the year.
and major and minor ions to rivers and estuaries. In spite of these advances in our understand-
In another example, ﬂow management for river ing, there are still important knowledge gaps.
health reasons should be integrated with ﬂow Prime amongst these are: a detailed understand-
regulation to control algal growth in weir pools. ing of the nitrogen cycle from sources through
Thus, management strategies developed for deal- transformations to sinks; the factors governing
ing with other environmental issues are likely to the succession of one algal species by another as
have signiﬁcant eﬀects on eutrophication. There in-water conditions change; the factors governing
will need to be a high degree of coordination nutrient release from sediments, including bacte-
between diﬀerent initiatives to ensure eﬃcient use rial mediation of phosphorus release; the role of
of resources and to obtain the best environmental akinetes in seeding blooms; and the reasons why
outcomes. This will pose a signiﬁcant challenge to certain strains of cyanobacteria release more
natural resource managers, the community and the toxins than others.
political process. The most important task now facing research-
ers is to generalise from the studies that have led to
our current understanding of eutrophication pro-
Conclusions cesses, so that the results can be extended to most
fresh and estuarine waterbodies across Australia.
It is clear that eutrophication in Australian con- Models are a powerful means of generalising. The
ditions is now considerably better understood than models that have been developed so far of physical
it was in the early 1990s. Much of the speculation and chemical processes within waterbodies need to
about the dominant processes operating in fresh- be tested and reﬁned in diverse environments
water bodies at that time has now been replaced by before they can be applied across Australia. On the
a process-based understanding founded on solid other hand, catchment nutrient sourcing and
experimental evidence. These advances have been transport models have yet to include the current
greatest in the identiﬁcation of the diﬀerent sour- knowledge about diﬀerent species of nutrients, the
ces of phosphorus in diﬀerent catchments across role played by streambank erosion and ﬂoodplain
Australia and in the understanding of the physical processes in some parts of Australia, and the role
processes operating in freshwater systems. It is of subsurface pathways for nutrient transport.
now clear, for example, that stratiﬁcation of turbid Waterbody managers will need to have a wide
weirpools, and the consequent advantage that array of tools at their disposal. We now possess
confers on buoyant cyanobacteria, lies behind the suﬃcient understanding to make more eﬃcient use
of long-standing techniques for managing water- enhanced by grazing on components of the microbial web.
bodies, such as destratiﬁcation and biomanipula- Mitteilungen. Internationale Vereinigung fur Theoretische
tion, and to develop and trial novel management und Angewandte Limnologie 24: 879–883.
Atech, 2000. Cost of algal blooms. LWRRDC Occasional Pa-
solutions. Control of stratiﬁcation through ﬂow per 26/99. Land and Water Resources Research and Devel-
management, control of light accession (or, more opment Corporation, Canberra, ACT, Australia, 42 pp.
speciﬁcally, the ratio between euphotic and mixed Australian State of the Environment Committee, 2001. Coasts
depth), the use of special nutrient binding materi- and Oceans, Australia State of the Environment Report 2001
als, and the deliberate drying of reservoir sedi- (Theme Report). CSIRO Publishing, Canberra, Australia,
ments to inhibit nutrient release are all examples of Avon River System Management Committee, 1993. Avon River
such novel techniques. System Management Strategy. Waterways Commission,
There is an increasing demand from both Perth, Western Australia.
communities and governments for improvements Azcue, J. M., A. Zeman & U. Forstner, 1998. International
in the quality of AustraliaÕs fresh and estuarine review of application of subaqueous capping techniques for
remediation of contaminated sediments. Proceedings of the
water resources. The research of the last decade 3rd International Congress for Environmental Geotechnics,
has put Australian managers in a much stronger Lisbon, Sept, 7–11, 1998.
position to meet this demand for eutrophication Baker, P. D., 1999. Role of akinetes in the development of
management. However, the existence of knowl- cyanobacterial populations in the lower Murray River,
edge, even when it is widely accepted within the Australia. Marine and Freshwater Research 50: 265–279.
Baker, P. & A. R. Humpage, 1994. Toxicity associated with
scientiﬁc community, is no guarantee that it will be commonly occurring cyanobacteria in surface waters of the
utilised. Considerable eﬀort is required to turn new Murray-Darling Basin, Australia. Australian Journal of
knowledge into a new paradigm, from which Marine and Freshwater Research 45: 773–786.
practical changes, such as changes in management Baldwin, D. S, 1996. The eﬀects of exposure to air and sub-
sequent drying on the phosphate sorption characteristics of
structures and new ﬁeld programs, can emerge.
sediments from a eutrophic reservoir. Limnology and
While there has already been good progress in Oceanography 41: 1725–1732.
transmitting this understanding to managers Baldwin, D. S., A. M. Mitchell & J. M. Olley, 2002. Pollutant-
through workshops, presentations and publica- sediment interactions: Sorption, reactivity and transport of
tions, there is still much more to be done to turn it phosphorus. In Haygarth, P. & S. Jarvis (eds), Agriculture,
into practical management outcomes. Hydrology and Water Quality. CAB International Press,
New York, 265–279.
Bauld, J. & N. F. Millis, 1977. Role of bacteria in nitrogen
cycling in the Werribee zone. Environmental Studies Series
References No. 166. Ministry for Conservation, Melbourne, Victoria,
AATSE/IEA, 1999. Water and the Australian Economy. Aus- Bek, P. & G. Robinson, 1991. Sweet Water or Bitter Legacy.
tralian Academy of Technological Sciences and Engineering, State of the Rivers – Water Quality in New South Wales.
and the Institution of Engineers Australia. Canberra, Department of Water Resources, Sydney, Australia.
Australia. Blackburn, S. I. & G. Cresswell, 1993. A coccolithophorid
Abal, E. G. & W. C. Dennison, 1996. Seagrass depth range and bloom in Jervis Bay, Australia. Australian Journal of Marine
water quality in Southern Moreton Bay, Queensland, Aus- and Freshwater Research 44: 253–260.
tralia. Marine and Freshwater Research 47: 763–771. Blackburn, S. I. & P. Thompson, 2000. The interaction of
Ajani, P., G. Hallegraeﬀ & T. Pritchard, 2001. Historic over- physics, biology and nutrient regimes on the initiation and
view of algal blooms in marine and estuarine waters of New development of algal blooms. National Eutrophication
South Wales, Australia. Proceedings of the Linnean Society Program ﬁnal report of project CSF1. Land and Water
of New South Wales 123: 1–22. Resources Research and Development Corporation, Can-
Anon, 1994a. Securing the future: Achievements of the Peel berra, ACT, Australia.
Inlet and Harvey Estuary management strategy 1989–1994. Blumberg, A. F. & G. L. Mellor, 1987. A description of a three-
Waterways Commission, Department of Transport, dimensional coastal ocean circulation model. In
Department of Agriculture, Perth, Western Australia. Nihoul, J. C. J. & B. M. Jamart (eds), Three-Dimensional
Anon, 1994b. The Algal Management Strategy for the Murray- Models of Marine and Estuarine Dynamics. Elsevier Ocea-
Darling Basin. Murray Darling Basin Ministerial Council. nography Series 45: 55–88.
Murray-Darling Basin Commission, Canberra, ACT, Bolger, P. & M. Stevens, 1999. Contamination of Australian
Australia. groundwater systems with nitrate. LWRRDC Occasional
Arndt, H. & B. Nixdorf, 1991. Spring clear-water phase in a Paper 03/99. Land and Water Resources Research and
eutrophic lake: control by herbivorous zooplankton Development Corporation, Canberra, Australia.
Boon, P. I., S. E. Bunn, J. D. Green & R. J. Shiel, 1994. Bowling, L. C. & P. D. Baker, 1996. Major cyanobacterial
Consumption of cyanobacteria by freshwater zooplankton: bloom in the Barwon-Darling River, Australia, in 1991, and
implications for the success of ‘‘top-down’’ control of underlying limnological conditions. Marine and Freshwater
cyanobacterial blooms in Australia. Australian Journal of Research 47: 643–657.
Marine and Freshwater Research 45: 875–887. Boynton, W. R., J. H. Garber, R. Summers & W. M. Kemp,
Boon, P. I., A. Mitchell & K. Lee, 1997. Eﬀects of wetting 1995. Inputs, transformations, and transport of nitrogen and
and drying on methane emissions from ephemeral ﬂood- phosphorus in Chesapeake Bay and selected tributaries.
plain wetlands in south-eastern Australia. Hydrobiologia Estuaries 18: 285–314.
357: 73–87. Bramley, R. G. V. & A. K. L. Johnson, 1996. Land use impacts
Bormans, M. & S. A. Condie, 1998. Modelling the distribution on nutrient loading in the Herbert River. In Hunter, H. M.,
of Anabaena and Melosira in a stratiﬁed river weir pool. A. G. Eyles & G. E. Rayment (eds), Downstream Eﬀects of
Hydrobiologia 364: 3–13. Land Use. Dept. of Natural Resources, Queensland,
Bormans, M. & I. T. Webster, 1997. A mixing criterion for turbid Australia, 93–96.
rivers. Environmental Modelling and Software 12: 329–333. Burch, M. & G. Ganf, 1998. Critical ﬂow and population de-
Bormans, M. & I. T. Webster, 1998. Dynamics of temperature velopment of the cyanobacteria Anabaena and Microcyctis in
stratiﬁcation in lowland rivers. Journal of Hydraulic Engi- the Murray-Darling river system. Final report – project
neering 124: 1059–1063. ACW1. Land and Water Resources Research and Devel-
Bormans, M. & I. T. Webster, 1999. Modelling the spatial and opment Corporation, Canberra, Australia.
temporal variability of diatoms in the River Murray. Journal Burch, M. D., J. D. Brookes, P. Tarrant & P. Dellaverde, 2000.
of Plankton Research 21: 581–598. Mixing it up with Blue-Green Algae. Water 27: 21–22.
Bormans, M., P. Ford, L. Duivenvoorden & L. Fabbro, 2000. Bureau of Meteorology Research Centre, CSIRO Atmospheric
Management strategies for control of cyanobacterial blooms Research, CSIRO Land and Water, CSIRO Mathematical and
in the Fitzroy River Barrage. Final report of Project CEM7. Information Sciences, 1999. Towards understanding climate
Land and Water Resources Research and Development variability in south western Australia: Research reports on the
Corporation, Canberra, Act, Australia. First Phase of the Indian Ocean Climate Initiative. Published
Bormans, M., H. Maier, M. Burch & P. Baker, 1997. Tem- by: Indian Ocean Climate Initiative Panel Address: Executive
perature stratiﬁcation in the lower River Murray, Australia: Oﬃcer IOCIP, C/- Water, Rivers Commission Hyatt Centre, 3
implications for cyanobacterial bloom development. Marine Plain St. East Perth, WA 6004 Australia.
and Freshwater Research 48: 647–654. Burns, C. W., 1998. Planktonic interactions with an austral
Bormans, M., B. S. Sherman & I. T. Webster, 1999. Is buoy- bias: Implications for biomanipulation. Lakes & Reservoirs:
ancy regulation in cyanobacteria an adaptation to exploit Research and Management 3: 95–104.
separation of light and nutrients? Marine and Freshwater Burns, C. W., D. J. Forsyth, J. F. Haney, M. R. James,
Research 50: 897–906. W. Lampert & R. D. Pridmore, 1989. Coexistence and
Bormans, M., P. W. Ford, L. Fabbro & G. Hancock, 2004. exclusion of zooplankton by Anabaena minutissima var
Onset and persistence of cyanobacterial blooms in a large attenuata in Lake Rotongaio, New Zealand. Archiv fur ¨
impounded tropical river, Australia. Marine and Freshwater Hydrobiologie Beiheft Ergebnisse der Limnologie 32: 63–82.
Research 55: 1–15. Caitcheon, G., T. Donnelly, P. J. Wallbrink & A. S. Murray,
Borum J., 1996. Shallow waters and land/sea boundaries. In 1995. Nutrient and sediment sources in Chaﬀey Reservoir
Eutrophication in Coastal Marine Ecosystems, Coastal and catchment. Australian Journal of Soil and Water Conser-
Estuarine Studies. Amer. Geophys. Union, Vol 52, vation 8: 41–49.
Washington D.C. 179–203. Caitcheon G. G., J. M. Olley, G. Hancock & P. J. Wallbrink,
Bott, G. M., 1993. Environmental and land use factors aﬀecting 1999. Exposing the phosphorus myth: Looking for evidence
phosphorus losses from estuary catchments of South Wes- of anthropogenic phosphorus in inland basins. International
tern Australia. Proceedings, First Int. Conf. on Diﬀuse Conf on Diﬀuse Pollution IAWQ. CSIRO Land and Water,
(Nonpoint) Pollution – Sources, Prevention, Impact, Perth, Australia, 287–295.
Abatement. IAWQ, Chicago, USA Sept. 1993. Cannon, D., J. C. Dean & I. C. Smalls, 1970. Growth and
Bourgues, S., M. Carpenter, G. Coade, P. Scanes & K. Koop, control of two species of Anabaena (Cyanophyta) in a water
1998. Oxygen and nutrient ﬂuxes quantiﬁed under experi- supply reservoir. Bulletin of the Australian Society for
mental conditions on sediment from deep and shallow sites Limnology 3: 15.
in Berowra Creek. Report to Hornsby Shire Council. Water Carpenter, S. R., N. F. Caraco, D. L. Correll, R. W. Howarth,
Science Section, NSW Environment Protection Authority, A. N. Sharpley & V. H. Smith, 1998. Nonpoint pollution of
Sydney, Australia, 35 pp. surface waters with phosphorus and nitrogen. Ecological
Bourgues, S., B. Hart, P. Ford, M. Harper, M. Grace, 2000. Applications 8: 559–568.
Nutrient release from river sediments: Phase II – Validation Carpenter, S. R. & J. F. Kitchell, 1993. The Trophic Cascade in
and application of sediment-release model. Technical Re- Lakes. Cambridge University Press, Cambridge, UK.
port. Final report to the National Eutrophication Manage- Carpenter, S. R., J. F. Kitchell & R. Hodgson, 1985. Cascading
ment Program. Land and Water Resources Research and trophic interactions and lake productivity. Bioscience 35:
Development Corporation, Canberra, Act, Australia. 634–639.
Chambers, L., J. Olley, H. Crockford & A. Murray, 1997. Conference on Diﬀuse Pollution, CSIRO Land and Water,
Conditions aﬀecting the availability and release of phos- Perth, WA, May 1999, 176–190.
phorus from sediments in the Maude Weir pool on the DeMelo, R., R. D. France & J. McQueen, 1992. Biomanipula-
Murrumbidgee River, New South Wales. In Davis, J. R. tion: Hit or myth? Limnology and Oceanography 37: 192–
(ed.), Managing Algal Blooms: Outcomes from CSIROÕs 207.
Multi-Divisional Blue-Green Algal Program. CSIRO Land Dennison, W. C. & E. G. Abal, 1999. Moreton Bay Study: A
and Water, Canberra, Australia, 41–50. Scientiﬁc Basis for the Healthy Waterways Campaign. South
Cloern, J. A., 2001. Our evolving conceptual model of the East Queensland Regional Water Quality Management
coastal eutrophication problem. Marine Ecology Progress Strategy, Brisbane, Queensland, Australia, 245 pp.
Series 210: 223–253. Diaz, R. & R. Rosenberg, 1995. Marine benthic hypoxia: a
Coade, G., 2002. Lake Macquarie Integrated Catchment Model review of its ecological eﬀects and the behavioural responses
– Eutrophication Potential in Lake Macquarie. Final report of benthic macrofauna. Oceanography and Marine Biology
to Lake Macquarie Taskforce. NSW Environment Protec- Annual Review 33: 245–303.
tion Authority, Sydney, NSW, Australia. Digby, M. J., P. Saenger, M. B. Whelan, D. McConchie,
Codd, G. A., D. A. Steﬀensen, M. D. Burch & P. D. Baker, B. Eyre, N. Holmes & D. Bucher, 1998. A physical classiﬁ-
1994. Toxic blooms of cyanobacteria in Lake Alexandrina, cation of Australian estuaries. Report to Urban Water
South Australia – learning from history. Australian Journal Research Association of Australia. Centre for Coastal
of Marine and Freshwater Research 45: 731–736. Management, Lismore, Australia.
Cosser, P. R., 1989. Nutrient concentration – ﬂow relationships Donnelly, T. H. (ed.), 1994. Review and scoping study of
and loads in the South Pine River, South Eastern Queens- catchment phosphorus sources. Consultancy Report
land. 1. Phosphorus loads. Australian Journal of Marine and 94/26, CSIRO Division of Water Resources, Canberra,
Freshwater Research 40: 613–30. Australia.
CRC Water Quality and Treatment, 1999. Priorities for Algal Donnelly, T. H., J. M. Olley, A. S. Murray & R. J. Wasson,
Research in Australia. Occasional Paper No. 2. CRC Water 1992. Algal blooms in the Darling River: ‘‘Run of River’’
Quality and Treatment, Adelaide, Australia. study Wentworth to Collarenabri, December 1991. Consul-
CSIRO Huon Estuary Study Team, 2000. Huon Estuary tancy Report 92/13, CSIRO Division of Water Resources,
Study – Environmental research for integrated catchment Canberra, Australia.
management and aquaculture. Final report to Fisheries Donnelly, T. H., P. W. Ford, D. McGregor & D. Allen, 1999.
Research and Development Corporation. Project 96/284. Anthropogenic changes to a billabong in New South Wales.
CSIRO Division of Marine Research, Marine Laboratories, 1. Lagoon evolution and phosphorus dynamics. Marine and
Hobart, Australia. (www.marine.csiro.au/Press Releases- Freshwater Research 50: 689–698.
folder/00releases/huon2306/summary.html). Donohue, R., W. A. Davidson, N. E. Peters, S. Nelson &
Cullen, P., 1991. Regional Catchment Management and B. Jakowyna, 2001. Trends in total phosphorus and total
Receiving Water Quality, the Monkey Creek Project. Land nitrogen concentrations of tributaries to the Swan-Canning
and Water Resources Research and Development Corpora- estuary, 1987 to 1998. Hydrological Processes 15: 2411–2434.
tion report. Canberra, ACT, Australia. Douglas, G., D. Hamilton, R. Gerritse, J. Adeney & D. Coad,
Davis, J. R., 1997. Managing Algal Blooms: Outcomes from 1997. Sediment geochemistry, nutrient ﬂuxes and water
CSIROÕs Multi-Divisional Blue-Green Algal Program. quality in the Swan River estuary, Western Australia. In
CSIRO Land and Water, Canberra, Australia, 113 pp. Davis, J. R. (ed.), Managing Algal Blooms. CSIRO Land
Davis, J. R. & T. F. N. Farley, 1997. CMSS: policy analysis and Water, Canberra, ACT.
software for catchment managers. Environmental Modelling Douglas, G. B., J. A. Adeny & M. Robb, 1999. A novel tech-
and Software 12: 197–210. nique for reducing bio-available phosphorus in water and
Davis, J. R, T. F. N. Farley, W. J. Young & S. M. Cuddy, sediments. In Barber, C., B. Humphries & J. Dixon (eds),
1998a. The experience of using a decision support system for Int. Conf. On Diﬀuse Pollution. CSIRO Land and Water,
nutrient management in Australia. Water Science and Perth, Australia, 517–523.
Technology 37: 209–216. Dudley, B. J. & D. I. Walker, 2000. Plant/water nutrient
Davis, J. R., A. Hamblin, E. OÕLoughlin, N. Austin, R. Banens, dynamics in a southwestern Australian estuary. Biologia
P. Cornish, J. Cox, P. Hairsine, M. McCulloch, P. Moody, Marina Mediterranea 7: 43–46.
J. Olley, B. Prove, I. Smalls & D. Weaver, 1998b. Phos- Elser, J., E. R. Marzolf & C. R. Goldman, 1990. Phosphorus
phorus in the landscape. Occasional Paper 16/98. Land and and nitrogen limitation of phytoplankton growth in the
Water Resources Research and Development Corporation, freshwaters of North America: A review and critique of
Canberra, ACT, Australia, 28 pp. experimental enrichments. Canadian Journal of Fisheries
Davison, W. & H. Zhang, 1994. In situ speciation measure- and Aquatic Sciences 47: 1468–1477.
ments of trace components in natural waters using thin ﬁlm EPA, 1988. Peel Inlet and Harvey Estuary Management
gels. Nature 367: 546–548. Strategy. Environmental Review and Management Pro-
Deeley, D. M., E. I. Paling & R. Humphries, 1999. Why south gramme Stage 2. Report and Recommendations of the
western Australian estuaries are highly susceptible to Environmental Protection Authority. Environmental Pro-
nutrient enrichment. Proceedings of the International tection Authority Bulletin 363. Perth, WA, Australia.
EPA, 1999. Review of the Environment Protection (Peel Inlet – algae by blue-green algae in the plankton of a stratiﬁed lake.
Harvey Estuary) Policy 1992. Environmental Protection Journal of Ecology 70: 829–844.
Authority, Perth, WA, Australia. Ganf, G. & R. J. Shiel, 1985. Feeding behaviour and limb
Eyre, B., 1993. Nutrients in the sediments of a tropical north- morphology of two cladocerans with small intersetular dis-
eastern Australian estuary, catchment and nearshore coastal tances. Australian Journal of Marine and Freshwater Re-
zone. Australian Journal of Marine and Freshwater search 36: 69–86.
Research 44: 845–866. Geddes, M. C., 1988. The role of turbidity in the limnology of
Eyre, B., 1998. Transport, retention and transformation of Lake Alexandria, River Murray, South Australia; Compar-
material in Australian estuaries. Estuaries 21: 540–551. isons between clear and turbid phases. Australian Journal of
Eyre, B., 2000. Regional evaluation of nutrient transformation Marine and Freshwater Research 39: 201–209.
and phytoplankton growth in nine river-dominated sub- Gerdes, P. & S. Kunst, 1998. Bioavailability of phosphorus as a
tropical east Australian estuaries. Marine Ecology Progress tool for eﬃcient P reduction schemes. Water Science and
Series 205: 61–83. Technology 37: 241–247.
Eyre, B., L. McKee, A. Ferguson & J. Gay, 1997. Ecosystem Gerritse, R. G. & J. A. Adeney, 1992. Nutrient exports from
response to nutrient loading – Two northern NSW estuarine various land uses on the Darling Plateau in Western Australia:
examples. In Koop, K. (ed.), Proceedings of the Coastal eﬀects on stream water quality. Report No. 92/41. CSIRO,
Nutrients Workshop, 30–31 October 1997. Australian Water Division of Water Resources, Perth, Western Australia.
and Wastewater Association, Artarmon, Sydney, Australia, Gerritse, R. G., C. Barber & J. A. Adeney, 1990. The impact of
100–105. residential urban areas on groundwater quality: Swan
Fabre, A., 1988. Experimental studies on some factors inﬂu- Coastal Plain of Western Australia. Water Resources Series
encing phosphorus solubilization in connection with the 3. CSIRO, Perth, Western Australia.
drawdown of a reservoir. Hydrobiologia 159: 153–158. Gerritse, R. G., P. J. Wallbrink & A. S. Murray, 1998. Accu-
Falconer, I. R., A. M. Beresford & M. T. C. Runnegar, 1983. mulation of phosphorus and heavy metals in the
Evidence of liver damage by toxin from a bloom of the blue- Swan-Canning Estuary, Western Australia. Estuarine and
green alga, Microcystis aeruginosa. Medical Journal of Coastal Shelf Science 47: 165–179.
Australia 1: 511–514. Gippsland Coastal Board, 2001. www.vcc.vic.gov.au/gcboard/.
Fay P. & C. Van Baalen, 1987. The Cyanobacteria. Elsevier, Grace, M. R., B. T. Hart, R. L. Oliver & C. M. Rees, 1997.
Amsterdam. Algal blooms in the Darling River – Are phosphates to
Finlayson, B. & T. A. McMahon, 1988. Australia v. the World; blame? Chemistry Australia, August 1997, 2–5.
A comparative analysis of streamﬂow characteristics. In Gutteridge Haskin & Davey P/L, 1992. Investigation of
Warner, R. F. (ed.), Fluvial Geomorphology of Australia. Nutrient Pollution in the Murray-Darling River System.
Academic Press, Sydney, Australia, 17–40. Report to Murray Darling Ministerial Council, Murray
Fogg, G. E. & A. E. Walsby, 1971. Buoyancy regulation and Darling Basin Commission Report No 311/1048/0504,
the growth of planktonic blue-green algae. Mitteilungen der Canberra, Australia.
internationalen Vereinigung fur Limnologie 19: 182–188.
¨ Hallegraeﬀ, G. & C. J. Bolch, 1992. Transport of diatom and
Forsberg, C. 1975. Nitrogen as a growth factor in freshwater. In dinoﬂagellate resting spores in shipsÕ ballast water: implica-
IAWPR Conference on Nitrogen as a Water Pollutant, tions for plankton biogeography and aquaculture. Journal of
Copenhagen, Denmark. Plankton Research 14: 1067–1084.
Forsyth, D. J., J. F. Haney & M. R. James, 1992. Direct Hallegraeﬀ, G. M. & S. Fraga, 1998. Bloom dynamics of the
observation of toxic eﬀects of cyanobacterial extracellular toxic dinoﬂagellate Gymnodinium catenatum, with emphasis
products on Daphnia. Hydrobiologia 228: 151–155. on Tasmanian and Spanish coastal waters. In Anderson, D.
Foy, R. H. & P. J. A. Withers, 1995. The contribution of M., A. D. Cembella & G. M. Hallegraeﬀ (eds), Physiolo-
agricultural phosphorus to eutrophication. Proceedings No. gical Ecology of Harmful Algal Blooms. Springer Verlag,
365. The Fertiliser Society, Greenhill House, Thorpewood, 59–80.
Peterborough, UK. Hamilton D. P., 1999. Numerical modelling and reservoir
Francis, G., 1878. Poisonous Australian lake. Nature 18: 11–12. management: applications of the DYRESM model. In
Fredericks, D. J. & D. T. Heggie, 2000. Are sediments a sig- Tundisi, J. G. & M. Straskraba (eds), Theoretical Reservoir
niﬁcant source of nutrients in Wilson Inlet. Final Report Ecology and its Applications. International Institute of
Project AGS2 to Land and Water Resources Research and Ecology. Brazilian Academy of Sciences and Backhuys
Development Corporation, Canberra, ACT, Australia, Publishers, 153–173.
128 pp. Hamilton, D. P., 2000. Record summer rainfall induced ﬁrst
Gabric, A. J. & P. R. F. Bell, 1993. Review of the eﬀect of recorded major cyanobacterial bloom in the Swan River.
nonpoint nutrient loading on coastal ecosystems. Australian The Environmental Engineer 1: 25.
Journal of Marine and Freshwater Research 44: 261–83. Hamilton, D. P., G. Schladow & I. H. Fisher, 1995. Controlling
Gachter, R. & J. S. Meyer, 1993. The role of micro-organisms the indirect eﬀects of ﬂow diversions on water quality in an
in mobilisation and ﬁxation of phosphorus in sediments. Australian reservoir. Environment International 21: 583–590.
Hydrobiologia 253: 103–121. Hamilton, D. P. & S. G. Schladow, 1997. Prediction of water
Ganf, G. & R. L. Oliver, 1982. Vertical separation of light and quality in lakes and reservoirs Part. I – model description.
available nutrients as a factor causing replacement of green Ecological Modelling 96: 91–110.
Hamilton, D. P. & J. V. Turner, 2001. Integrating research and Western Australian attempt at integrated catchment man-
management for an urban estuarine system: The Swan- agement. Water Science and Technology 32: 255–264.
Canning estuary, Western Australia. Hydrological Processes Hunter, H. M. & R. S. Walton, 1997. From land to river to reef
15: 2383–2385. lagoon: land use impacts on water quality in the Johnstone
Harris, G. P., 1994. Nutrient loadings and algal blooms in River catchment. Queensland Department of Natural
Australian waters – a discussion paper. LWRRDC Occa- Resources, Brisbane, Queensland, Australia.
sional Paper No. 12/94. Land and Water Resources Re- Hunter, H. M., R. S. Walton & D. J. Russell, 1996. Con-
search and Development Corporation, Canberra. temporary water quality in the Johnstone River catchment.
Harris, G. P., 1996. Catchments and aquatic ecosystems: In Hunter, H. M., A. G. Eyles & G. E. Rayment (eds),
Nutrient ratios, ﬂow regulation and ecosystem impacts in Downstream Eﬀects of Land Use. Queensland Department
rivers like the Hawkesbury-Nepean. Cooperative Research of Natural Resources, Queensland, 339–345.
Centre for Freshwater Ecology, Canberra, ACT, Australia, Iserman, K., 1990. Share of agriculture in nitrogen and phos-
57 pp. phorus emissions into the surface waters of Western Europe
Harris, G. P., 2001. Biogeochemistry of nitrogen and phos- against the background of their eutrophication. Fertilizer
phorus in Australian catchments, rivers and estuaries: eﬀects Research 26: 253–269.
of land use and ﬂow regulation and comparisons with global Jack, P. 1987. Nutrient monitoring in the Swan River estuary.
patterns. Marine and Freshwater Research 52: 139–149. In John, J. (ed.), Curtin University Environmental Studies
Harris, G. P., G. Batley, D. Fox, D. Hall, P. Jernakoﬀ, R. Group Report No.1. Curtin University, Perth, 45–64.
Molloy, A. Murray, B. Newell, J. Parslow, G. Skyring & S. Jakowyna, B. N., 2000. Nutrients in tributary inﬂows to the
Walker, 1996. Port Phillip Bay Environmental Study Final Peel-Harvey estuarine system, WA. Status and trends. Water
Report. CSIRO, Canberra, Australia, 239 pp. Resources Technical Series WRT 23, Water and Rivers
Hart, B. T. & M. R. Grace , 2001. Sources, Transformations, Commission, Perth, WA.
Eﬀects, Management of Nitrogen in Freshwater Ecosystems. Johnes, P. J., 1999. Diﬀuse pollution: sources, control strategies
Report on Nitrogen Workshop, September 2000. Land and and policy implications in the European community. In
Water Australia, Canberra, Australia. Barber, C., B. Humphreys & J. Dixon (eds), Proceedings,
Hawkins P. R., M. T. C. Runnegar, A. R. B. Jackson & I. R. International Conference on Diﬀuse Pollution. CSIRO
Falconer, 1985. Severe epatotoxicity caused by the tropical Land, Water, Wembley, W.A., Australia, 50–64.
cyanobacterium (blue green alga) Cylindrospermopsis raci- Johnes, P. J. & R. A. Hodgkinson, 1998. Phosphorus loss from
borskii (Woloszynska) seenaya and subba raju isolated from a catchments: pathways and implications for management.
domestic water supply reservoir. Appl. Environ. Microbiol. Soil Use and Management 14: 175–185.
50: 1292–1295. Jones, G. J., 1993. Weirpool conditions stimulating cyanobac-
Heathwaite, A. L., 1999. Delineating critical source areas of terial blooms in the Murrumbidgee River. In Roberts, J. &
nutrient export in agricultural catchments. In Barber, C., R. Oliver (eds) The Murrumbidgee: Past and Present.
B. Humphries & J. Dixon (eds), International Conf. on CSIRO, Griﬃth, Australia, 70–82.
Diﬀuse Pollution. CSIRO Land and Water, Perth, Australia: Jones, G., Davis, M., Orr, P. & Fairweather, P., 1997. Setting
342–351. target river ﬂows for the prevention of cyanobacterial
Heggie, D. T., G. W. Skyring, J. Orchardo, A. R. Longmore, blooms. Final Report to the Land and Water Resources
G. J. Nicholson & W. M. Berelson, 1999. Denitriﬁcation and Research and Development Corporation – Project CWN10.
denitriﬁcation eﬃciencies in sediments of Port Phillip Bay: CSIRO Land, Water, Canberra, ACT, Australia.
direct determinations of biogenic N2 and N-metabolite ﬂuxes Jones, G. J. & W. Poplawski, 1998. Understanding and man-
with implications for water quality. Marine and Freshwater agement of cyanobacterial blooms in sub-tropical reservoirs
Research 50: 589–596. of Queensland, Australia. Water Science and Technology 37:
Hodgkin, E. P. & B. H. Hamilton, 1993. Fertilizers and 161–168.
eutrophication in southwestern Australia – setting the scene. Kerr, R., 1994. Water Quality – Hawkesbury-Nepean River
Fertilizer Research 36: 95–103. System, June 1990–June 1993. NSW Environment Protec-
Hodgkin, E. P. & R. C. Lenanton, 1981. Estuaries and coastal tion Authority. EPA 94/103. Chatswood, Sydney, Australia.
lagoons of South Western Australia. In Neilson, B. J. & L. Kerr, R., M. Root & A. Church, 1996. Hawkesbury-Nepean
E. Cronin (eds), Estuaries and Nutrients. Humana Press, River: Eutrophication upstream of Wisemans Ferry, the
Clifton, New Jersey, 307–321. impact of discharges from South and Eastern creeks. July
Hotzel, H. & R. Croome, 1994. Long-term phytoplankton
¨ 1990–June 1995. NSW EPA report 96/35. Sydney, NSW,
monitoring of the Darling River at Burtundy, New South Australia.
Wales: Incidence and signiﬁcance of cyanobacterial blooms. King, C. R. & R. J. Shiel, 1993. Functional response to Daphnia
Australian Journal of Marine and Freshwater Research 45: carinata King when feeding on on the ﬁlamentous diatom
747–759. Melosira granulata. Australian Journal of Marine and
Howarth, R. W., 1988. Nutrient limitation of net primary Freshwater Research 44: 761–768.
production in marine ecosystems. Annual Review of Ecology Kinhill Engineers, 1988. Peel Inlet and Harvey Estuary Man-
and Systematics 19: 89–110. agement Strategy. Environmental Review and Management
Humphries, R. & S. Robinson, 1995. Assessment of the success Programme Stage 2. Prepared for Department of Marine
of the Peel-Harvey estuary system management strategy – a and Harbours, and Department of Agriculture, Perth WA.
Kirk, J. T. O., 1994. Light and Photosynthesis in Aquatic Loone, J., & L. Lloyd, 1996. Lake Mokoan Restoration Pro-
Ecosystems. Cambridge University Press, Cambridge, UK, ject. Management and Implementation Plan. Water ECO-
509 pp. sciences report to Goulburn-Murray Water, report 81/96,
Kirkby, C. A., L. J. Smythe, J. W. Cox & D. J. Chittleborough, Armidale, Victoria.
1997. Phosphorus movement down a toposequence from a Malthus, T. J., E. P. H. Best & A. G. Dekker, 1990. Assessment
landscape with texture contrast soils. Australian Journal of of the importance of emergent and ﬂoating leaved macro-
Soil Research 35: 399–417. phytes to trophic status in the Loosdrecht Lakes (The
Kirkby, C. A., J. W. Cox & D. J. Chittleborough, 1998. Final Netherlands). Hydrobiologia 191: 257–263.
Report Project UAD7 – Movement of Phosphorus through Martin, C. E. & M. T. McCulloch, 1999. Nd-Sr isotopic and
Soils. Land and Water Resources Research and Develop- trace element geochemistry of river sediments and soils in a
ment Corporation, Canberra, Australia. fertilised catchment, NSW, Australia. Geochimica et
Kirkman, H., 1997. Seagrasses of Australia. Technical paper, Cosmochimica Acta 63: 285–303.
State of the Environment. Department of the Environment Matveev, V., 2000. Unpublished Milestone 3 report project
and Heritage, Canberra, Australia. CLW2: whole lake biomanipulation for the reduction of
Kobayashi, T., 1993. Filtering rates of Daphnia carinata King nuisance micro-algae. Land and Water Australia, Canberra,
(Crustacea: Cladocera) on the blue-green algae Microcystis Australia.
aeruginosa Kutz and Anabaena cylindrica (Lemm.). Austra-
¨ Matveev, V. & L. Matveeva, 1997. Grazer control and nutrient
lian Journal of Ecology 18: 231–234. limitation of phytoplankton biomass in two Australian res-
Kobayashi, T., P. Gibbs, P. I. Dixon & R. J. Shiel, 1996. ervoirs. Freshwater Biology 38: 49–65.
Grazing by a river zooplankton community: importance of Matveev, V., L. Matveeva & G. J. Jones, 1994a. Phytoplankton
microzooplankton. Marine and Freshwater Research 47: stimulation by mosquitoﬁsh in the presence of large Daphnia.
1025–1036. Verhandlungen der internationalen theoretischen und
Kobayashi, T., A. Church, S. Hardiman & L. Gallagher, 1998. angewandten Limnologie 25: 2193–2197.
Grazing by a resident macrozooplankton community and Matveev, V., L. Matveeva & G. J. Jones, 1994b. Study of the
non-resident Daphnia carinata King: A preliminary in situ ability of Daphnia carinata King to control phytoplankton
incubation study. Lakes & Reservoirs: Research and Man- and resist cyanobacterial toxicity: implications for bioma-
agement 3: 193–203. nipulation in Australia. Australian Journal of Marine and
Koop, K. & P. Hutchings, 1996. Disposal of sewage to the Freshwater Research 45: 889–904.
ocean – a sustainable solution?. Marine Pollution Bulletin May, V. & E. J. McBarron, 1973. Occurrence of the blue-green
33: 121–123. alga, Anabaena circinalis Rabenh., in New South Wales and
Koop, K., R. Bally & C. D. McQuaid, 1983. The ecology of toxicity to mice and honey bees. Journal of the Australian
South African estuaries. Part XII, The Bot River, a closed Institute of Agricultural Science 39: 264–266.
estuary in the southwestern Cape. South African Journal of McBarron, E. J. & V. May, 1966. Poisoning of sheep in New
Zoology 18: 1–10. South Wales by the blue-green alga Anacystis cyanea (Kuetz)
Kristensen, P. & H. Ole Hanson, 1994. European Rivers and Dr. and Dail. Australian Veterinary Journal 42: 449–453.
Lakes: Assessment of their Environmental State. Environ- McBarron, E. J., R. I. Walker, I. Gardner & K. H. Walker,
ment Monographs 1, European Environment Agency, Co- 1975. Toxicity to livestock of the blue-green alga Anabaena
penhagen, Denmark. circinalis. Australian Veterinary Journal 51: 587–588.
Krogh, M., 2000. Ocean outfalls in NSW. Unpublished report, McKee, L. J, B. D. Eyre & S. Hossain, 2000a. Intra- and
Water Science Section, NSW Environment Protection Au- interannual export of nitrogen and phosphorus in the
thority, Sydney, Australia, 127 pp. subtropical Richmond River catchment, Australia. Hydro-
Lake, P. S., 1995. Of ﬂoods and droughts: river and stream logical Processes 14: 1787–1808.
ecosystems in Australia. In Cushing, C. E., K. W. Kum- McKee, L. J., B. D. Eyre & S. Hossain, 2000b. Transport and
mings & G. W. Minshall (eds), River and Stream Ecosys- retention of nitrogen and phosphorus in the sub-tropical
tems. Elsevier, Amsterdam, 659–694. Richmond River estuary, Australia – A budget approach.
Larkum, A. W. D., A. J. McComb & S. A. Shepherd, 1989. Biogeochemistry 50: 241–278.
Biology of Seagrasses. Elsevier Press, Oxford. McKelvie, I. D., D. M. W. Peat & P. J. Worsfold, 1995.
Lawrence I., M. Bormans, R. Oliver, G. Ransom, B. Sherman, Techniques for the quantiﬁcation and speciation of phos-
P. Ford & B. Wasson, 2000. Physical and nutrient factors phorus in natural waters. Analytical Procedure 32: 437–445.
controlling algal succession and biomass in Burrinjuck McKergow, L., D. Weaver, I. Prosser, R. Grayson & A. Reed,
Reservoir. Technical Report – Project UOC12. University of 2001. Before and after riparian management: sediment and
Canberra, Canberra, Australia. nutrient exports from a small agricultural catchment, Wes-
Lewin, K., 1998. Lake Mokoan Restoration Program: Lake tern Australia. Paper to Third Australian Stream Manage-
Mokoan Revegetation trial. Technical Report 5/98. Water ment Conference, August 2001, Brisbane, Australia.
ECOsciences, Armidale, Victoria. McQueen, D. J., 1998. Freshwater food web manipulation: a
Lloyd, L., 1998. Lake Mokoan Restoration Program: Water powerful tool for water quality improvement, but mainte-
regime management guidelines for Lake Mokoan. Technical nance is required. Lakes & Reservoirs: Research and Man-
Report 9/98. Water ECOsciences, Armidale, Victoria. agement 3: 83–94.
McQueen, D. J., J. R. Post & E. L. Mills, 1986. Trophic rela- Newman, R., 1998. Hydrology of the lower lakes and Coorong.
tionships in freshwater pelagic ecosystems. Canadian Jour- In Jensen, A., M. Good, P. Tucker & M. Long (eds), River
nal of Fisheries and Aquatic Sciences 43: 1571–1581. Murray Barrages Environmental Flows. Dept. of Environ-
Merrick, J. & G. G. Ganf, 1988. Eﬀects of zooplankton grazing ment and Natural Resources, Adelaide, South Australia:
on phytoplankton communities in Mt Bold Reservoir, South 14–21.
Australia, using enclosures. Australian Journal of Marine Nicholas, D., 1980. Mineral nutrient requirements and utilisa-
and Freshwater Research 39: 503–523. tion by algal ﬂora of freshwater lakes. Australian Water
Meyer, C. P., R. W. Gillett & I. E. Galbally, 2001. The Resources Council Research Report No. 74/72. Canberra,
atmospheric nitrogen cycle over Australia. In Hart, B. T. & Australia.
M. R. Grace (eds), Nitrogen Workshop 2000. Land and Nicholson, G. J., A. R. Longmore & W. M. Berelson, 1999.
Water Australia, Canberra, Australia, 65–73. Nutrient ﬂuxes measured by two types of benthic chamber.
Mitchell, A. M. & D. S. Baldwin, 1998. Eﬀects of desiccation/ Marine and Freshwater Research 50: 567–572.
oxidation on the potential for bacterially mediated P release Nielsen, L. P., 1992. Denitriﬁcation in sediment determined
from sediments. Limnology and Oceanography 43: 481–487. from nitrogen isotope pairing. FEMS Microbial Ecology 86:
Mitchell, A. M. & D. S. Baldwin, 1999. The eﬀects of sediment 357–362.
desiccation on the potential for nitriﬁcation, denitriﬁcation Nixon, S. W., 1995. Coastal marine eutrophication: a
and methanogenesis in an Australian reservoir. Hydrobio- deﬁnition, social causes and future concerns. Ophelia 41:
logia 392: 3–11. 199–219.
Mitchell, A. W., H. M. Reghenzani, H. M. Hunter & Nixon, S. W., J. W. Ammerman, L. P. Atkinson, V. M.
R. G. V. Bramley, 1996. Water quality and nutrient ﬂuxes Berounsky, G. Billen, W. C. Boicourt, W. R. Boynton, T. M.
from river systems draining to the Great Barrier Reef Marine Church, D. M. DiToro, R. Elmgren, J. H. Garber, A. E.
Park. In Hunter, H. M., A. G. Eyles & G. E. Rayment (eds), Giblin, R. A. Jahnke, N. J. P. Owens, M. E. Q. Pilson & S. P.
Downstream Eﬀects of Land Use. Dept. of Natural Seitzinger, 1996. The fate of nitrogen and phosphorus at the
Resources, Queensland, Australia, 23–33. land-sea margin of the North Atlantic Ocean. Biogeochem-
Mitchell, B. D. & W. D. Williams, 1982. Population dynamics istry 35: 141–180.
and production of Daphnia carinata (King) and Simocepha- NLWRA, 2001. www.nlwra.gov.au.
lus exspinosus (Koch) in waste stabilization ponds. NSW Blue-Green Algal Task Force, 1992. Blue-Green Algae.
Australian Journal of Marine and Freshwater Research 29: Final Report of the Task Force. Dept. of Water Resources,
565–576. Parramatta, NSW, Australia.
Moss, A. J., G. E. Rayment, N. Reilly, E. K. Best, 1993. A Oenema O., C. W. J. Roest, 1997. Nitrogen and phosphorus
preliminary assessment of sediment and nutrient exports losses from agriculture into surface waters. In Roijackers, R.,
from Queensland coastal catchments. Environment Techni- R. H. Aalderink & G. Blom (eds), Eutrophication Research,
cal Report No. 5. Department of Environment and Heritage, State-of-the-Art, Department of Water Quality Management
Brisbane, Queensland. and Aquatic Ecology, Wageningen Agricultural University,
Moss, B., J. Stansﬁeld, K. Irvine, M. Perron & G. Phillips, Netherlands, 13–25.
1996a. Progressive restoration of a shallow lake: a 12-year Oliver, R. L., B. T. Hart, G. B. Douglas & R. Beckett, 1993.
experiment in isolation, sediment removal and biomanipa- Phosphorus speciation in the Murray and Darling Rivers.
tion. Journal of Applied Ecology 33: 71–86. Chemistry Australia, 60: 392–397.
Moss, B., J. Madgwick & G. Phillips, 1996b. A Guide to the Oliver, R. L., B. T. Hart, J. Olley, M. Grace, C. Rees &
Restoration of Nutrient-Enriched Shallow Lakes. Broads G. Caitcheon, 2000. The Darling River: Algal growth and
Authority and Environment Agency UK, 180 pp. the cycling and sources of nutrients. Report to the Murray-
Murray, A. G. & J. S. Parslow, 1999. Modelling of nutrient im- Darling Basin Commission (Project M386). CRC for
pacts in Port Phillip Bay – a semi-enclosed marine Australian Freshwater Ecology and CSIRO Land and Water, Canberra,
ecosystem. Marine and Freshwater Research 50: 597–611. ACT, Australia, 200 pp.
National Research Council, 2000. Clean Coastal Waters: Under- Oliver, R. L. & G. G. Ganf, 2000. Freshwater blooms. In
standing and Reducing the Eﬀects of Nutrient Pollution. Na- Whitton, B. A. & M. Potts (eds), The Ecology of Cyano-
tional Academy Press, Washington, DC, USA, 405 pp. bacteria. Kluwer Academic Publishers, 149–194.
Negri, A. P., G. J. Jones & M. Hindmarsh, 1995. Sheep mor- Oliver, R. & I. Webster, 2001. Unpublished Final Report
tality associated with paralytic shellﬁsh poisons from the Project MDR17: Algal availability of phosphorus discharged
cyanobacterium Anabaena circinalis. Toxicon 33: 1321–1329. from diﬀerent catchment sources. Land and Water Austra-
Neil, D. & P. Fogarty, 1991. Land use and sediment yield on lia, Canberra, ACT, Australia.
the Southern Tablelands of NSW. Australian Journal of Soil Paerl, H. W., 1988. Nuisance phytoplankton blooms in coastal,
and Water Conservation 4: 33–39. estuarine and inland waters. Limnology and Oceanography
Newell, B. S., 1990. An appraisal of the eﬀects of the Werribee 33: 823–847.
Treatment Complex discharge on the ecology of Port Phillip Peel-Harvey Study Group, 1988. Peel Inlet and Harvey Estuary
Bay. MMBW Environmental Services Series No. 90/007. management study: Environment review and management
Melbourne and Metropolitan Board of Works, Melbourne, program, Stage 1. Dept. Conservation and Land Manage-
Victoria, Australia. ment, Perth, WA.
Peters, N. E. & R. Donohue, 2001. Nutrient transport to the Aiding Nutrient Management in Coastal Rivers and Estu-
Swan-Canning estuary, Western Australia. Hydrological aries. 30–31 October 1997. Australian Water and Wastewa-
Processes 15: 2555–2577. ter Association Inc., Artarmon, Sydney, Australia, 89–99.
Pionke, H. B., M. L. Sharma & J. K. Hosking, 1990. Eﬀect of Schindler, D. W., 1977. Evolution of phosphorus limitation in
irrigated horticultural cropping on groundwater quality: lakes. Science 195: 260–262.
Swan Coastal Plain, Western Australia. CSIRO Water Re- Schindler, D. W., 1978. Factors regulating phytoplankton
sources Series No. 2, CSIRO Division of Water Resources, production and standing crop in the worldÕs freshwaters.
Perth, WA. Limnology and Oceanography 23: 478–486.
Qui, S. & A. J. McComb, 1995. Planktonic and microbial Schladow, S. G. & D. P. Hamilton, 1995. Eﬀect of major ﬂow
contributions to phosphorus release from fresh and air dried diversion on sediment nutrient release in a stratiﬁed reser-
sediments. Freshwater Research 46: 1039–1045. voir. Marine and Freshwater Research 46: 189–195.
Reuter, D., 2001. Assessments of soil fertility in AustraliaÕs Schladow, S. G. & D. P. Hamilton, 1997. Prediction of water
agricultural lands. Interim report to National Land and Water quality in lakes and reservoirs. Part II – model calibration,
Resources Audit, project ‘‘Nutrient Balance in regional sensitivity analysis and application. Ecological Modelling 96:
farming Systems and Soil Nutrient Status’’, May 2001. 111–123.
Reynolds, C. S., 1984. The Ecology of Freshwater Phyto- Schonfeldt, C., 1993. Algal Management Strategy: Technical
plankton. Cambridge University Press, Cambridge, UK. Advisory Group. Murray Darling Basin Commission,
Reynolds, C. S., 1989. Relationships among the biological Canberra, Australia.
properties, distribution and regulation of production by Shapiro, J., 1990. Biomanipulation: the next phase – making it
planktonic cyanobacteria. Toxicity Assessment: An Inter- stable. Hydrobiologia 200/201: 13–27.
national Journal 4: 229–235. Shapiro, J., V. Lamarra & M. Lynch, 1975. Biomanipulation: An
Robertson, A. I., 1999. Limiting nutrient workshop 1997. ecosystem approach to lake restoration. In Brezonik, P. L. &
LWRRDC Occasional Paper 7/99. Land Water Resources J. L. Fox (eds), Proceedings of a Symposium on Water Quality
Research and Development Corporation, Canberra, ACT, Management through Biological Control. University of
Australia, 17 pp. Florida Press, Gainesville, Florida, 85–96.
Robertson, A. I., S. E. Bunn, P. I. Boon & K. F. Walker, 1999. Shapiro, J. V. & D. I. Wright, 1984. Lake restoration by
Sources, sinks and transformations of organic carbon in biomanipulation: Round Lake, Minnesota, the ﬁrst two
Australian ﬂoodplain rivers. Marine and Freshwater years. Freshwater Biology 14: 371–383.
Research 50: 813–829. Sharma, M. L., P. G, Kin, D. E. Herne, & J. D., Byrne, 1996.
Romero, J. R. & J. Imberger, 1999. Seasonal horizontal gra- Land use eﬀects on water quality in the Ellen Brook
dients of dissolved oxygen in a temperate Austral reservoir. catchment: experimental/modelling. Annual Report 96/1,
In Tundisi, J. G. & M. Straskraba (eds), Theoretical July 1994–June 1995, CSIRO, Perth, Western Australia.
Reservoir Ecology and its Applications. International In- Sharpley, A. N. & P. J. A. Withers, 1994. The environmentally-
stitute for Ecology, Brazilian Academy of Sciences and sound management of agricultural phosphorus. Fertilizer
Backhuys Publishers, 211–226. Research 39: 133–146.
Rose, T., 1998. Dawesville Channel: has the Peel-Harvey been Sharpley, A. N., T. Daniel, T. Sims, J. Lemunyon, R. Stevens &
ﬁxed? A review four years later. Institute of Municipal R. Parry, 1999. Agricultural phosphorus and eutrophication.
Engineers, Mandurah, WA. US Department of Agriculture, Agriculture Research Ser-
Roy, P. S., R. J. Williams, P. J. Gibbs, A. R. Jones, B. Coates, vice, University Park, PA, ARS-149, 42 pp.
L. Yassini, J. P. Hudson, P. Scanes, R. J. West & S. Nichol, Sherman, B., J. Whittington & R. Oliver, 2000. The impact of
2000. Structure and function of south-east Australian estu- artiﬁcial destratiﬁcation on water quality in Chaﬀey Reser-
aries: A framework for addressing management issues. . voir. Archiv fur Hydrobiologie. Special Issue Advances in
Estuarine and Coastal Shelf Science 53: 351–384. Limnology 55: 15–29.
Runnegar, M. T. C., A. R. B. Jackson & I. R. Falconer, 1988. Sherman, B. S., P. F. Ford, P. Hatton, J. Whittington, D. Green,
Toxicity to mice and sheep of the cyanobacterium (blue- D. Baldwin, R. L. Oliver, R. Shiel, J. van Berkel, R. Beckett,
green algae) Anabaena circinalis. Toxicon 26: 599–602. L. Grey & W. Maher, 2001. The Chaﬀey Dam Story – Final
Rush, J., 2005. Stable isotope of nitrogen as a tracer of sewage report to CRC Freshwater Ecology, Canberra, Australia.
pollution in the Hawkesbury-Nepean River. Ph.D. thesis, Sherman, B. S., I. T. Webster, G. J. Jones & R. L. Oliver, 1998.
University of Sydney, NSW 2006, Australia, 247 pp. Transitions between Aulacoseira and Anabaena dominance
Ryther, J. H. & W. M. Dunstan, 1971. Nitrogen, phosphorus, in a turbid river weir pool. Limnology and Oceanography
and eutrophication in the coastal marine environment. 43: 1902–1915.
Science 171: 375–380. Simmons, B. L. & D. M. H. Cheng, 1981. Phosphorus assimi-
Sandercock, C., 1998. Lake Mokoan Restoration Program: lation in the Nepean River, New South Wales. In Procs
review of lake restoration approaches of relevance to large AWWA 9th Federal Convention, Perth, 1981.
shallow lakes. Technical Report 4/98. Water ECOsciences, Sims, J. T., R. Simard & B. C. Joern, 1998. Phosphorus loss in
Armidale, Victoria. agricultural drainage: historical perspective and current
Scanes, P., G. Coade, D. Large & T. Roach, 1997. Developing research. Journal of Environmental Quality 27: 277–293.
criteria for acceptable loads of nutrients from catchments. In SKM, 1997. Preliminary assessment of nutrient discharges to
Koop, K. (ed.), Coastal Nutrients Workshop. Research estuaries. Report prepared for the NSW Environment
Protection Authority. Sinclair Knight Merz, St. Leonards, Van Dock, W. & B. T. Hart, 1996. Akinete diﬀerentiation in
Sydney, Australia. Anabaena circinalis (Cyanophyta). Journal of Phycology 32:
Skyring, G. W., A. R. Longmore, A. W. Chiﬃngs & 557–565.
C. J. Crossland, 1992. Nutrients. In Hall, D. N. (ed.), Port Van Donk, E., 1991. Perspectives for submerged macrophytes
Phillip Bay Environmental Study: Status Review. Technical in shallow lake restoration projects in the Netherlands.
Report No. 9. CSIRO Port Phillip Bay Environmental Hydrobiological Bulletin 24: 125–131.
Study, Melbourne, Victoria, Australia. Victorian Blue-Green Alga Project Team, 1995. Nutrient
Smith, V. H., 1986. Light and nutrient eﬀects on the relative Management Strategy for Victorian Inland Waters.
biomass of blue-green algae in lake phytoplankton. Department of Conservation and Natural Resources, Mel-
Canadian Journal of Fisheries and Aquatic Sciences 43: bourne, Australia.
148–153. Viney, N. R. & M. Sivapalan, 2001. Modelling catchment
Smith, V. H., G. D. Tilman & J. C. Nekola, 1999. Eutrophi- processes in the Swan-Avon River Basin. Hydrological
cation: impacts of excess nutrient inputs on freshwater, Processes 15: 2671–2685.
marine and terrestrial ecosystems. Environmental Pollution Vitousek, P. M., J. Aber, R. W. Howarth, G. E. Likens,
100: 179–196. P. A. Matson, D. W. Schindler, W. H. Schlesinger & G.
Sparling, G. P., K. N. Whale & A. J. Ramsay, 1985. Quanti- D. Tilman, 1997. Human alteration of the global nitrogen
fying the contribution from the soil microbial biomass to the cycle: causes and consequences. Ecological Applications 7:
extractable P level of fresh and air-dried soils. Australian 737–750.
Journal of Soil Research 23: 613–621. Vollenweider, R. A., 1976. Advances in deﬁning critical loading
State of Victoria, 1995. Nutrient management strategy for levels for phosphorus in lake eutrophication. Memorie
Victorian inland waters. State Government of Victoria, dellÕIstituto Italiano di Idrobiologia 33: 53–83.
Melbourne, Australia. Wallbrink, P. J., C. E. Martin & C. J. Wilson, 2001. Con-
Stevens, D. P., J. W. Cox & D. J. Chittleborough, 1999. tribution of sediment and sediment bound P from diﬀerent
Pathways of phosphorus, nitrogen and carbon movement landuses to two contrasting streams in temperate and tro-
over and through texturally diﬀerentiated soils, South pical Australia. Proc. 3rd Australian Stream Management
Australia. Australian Journal of Soil Research 37: 679–693. Conference, Brisbane, Australia. Pages 617–623.
Sturman, J., C. E. Oldham & G. N. Ivey, 1999. Steady con- Wallbrink, P. J., A. S. Murray, J. M. Olley & L. J. Olive, 1998.
vective exchange ﬂows down slopes. Aquatic Sciences 61: Determining sources and transit times of suspended sedi-
260–278. ment within the Murrumbidgee tributaries, NSW, Australia,
Sullivan, C., J. Saunders & D. Welsh, 1988. Phytoplankton of using fallout 137Cs and 210Pb. Water Resources Research 34:
the River Murray, 1980–1985. Murray Darling Basin Com- 879–887.
mission. Canberra, ACT, Australia, 61 pp. Wallbrink, P. J., J. M. Olley, A. S. Murray & L. J. Olive, 1996.
Thompson, P. A. & W. Hosja, 1996. Nutrient limitation of The contribution of channel banks and gully walls to total
phytoplankton in the upper Swan River estuary, Western phosphorus loads in the Murrumbidgee River. In Ruther-
Australia. Marine and Freshwater Research 47: 659–667. ford, I. & M. Walker (eds), National Conference on Stream
Thompson, P. A. & Twomey, L., 2000. The phytoplankton Management, Merrijig. Cooperative Research Centre for
ecology of Wilson Inlet. National Eutrophication Catchment Hydrology, Adelaide, South Australia, 1–6.
Management Program Final Report UTA8. Land and Walton, R. S. & H. M. Hunter, 1997. Water quality modelling
Water Research and Development Corporation, Canberra, with HSPF in a tropical catchment. 24th Hydrology and
Australia. Water Resources Symposium, Auckland, N.Z., 24–27 Nov.
Thompson P., J. Adeney & R. Gerritse, 1997. Phytoplankton in 1997. Institution of Engineers, Canberra, ACT, Australia,
the Swan River, Western Australia: Research results 1993 to 469–474.
1996 and management implications. In Davis, J. R. (ed.), WATER ECOsciences, 1996. Lake Mokoan Restoration Pro-
Managing Algal Blooms: Outcomes from CSIROs Multi- ject, draft restoration strategy. Report 64/96, Water
Divisional Blue-Green Algal Program. CSIRO Land and ECOsciences, Armidale, Victoria, Australia.
Water, Canberra, Australia, 1–14. WATER ECOsciences, 1998. Lake Mokoan Restoration Pro-
Thoms, M., P. Suter, J. Roberts, J. Koehn, G. Jones, T. Hill- gram. Technical reports 1–10/98. Water ECOsciences,
man & A. Close, 1998. Report of the River Murray Scientiﬁc Armidale, Victoria, Australia.
Panel on Environmental Flows. River Murray – Dartmouth Weaver, D. M. & A. E. G. Reed, 1998. Patterns of nutrient
to Wellington and the Lower Darling River. Murray-Darling status and fertiliser practice on soils of the south coast of
Basin Commission, Canberra, 168 pp. Western Australia. Agriculture, Ecosystems and Environ-
Udy, J. W., 2001. Diﬀerent methods for measuring rates of ment 67: 37–53.
nitrogen transformation: strengths and weaknesses. In Webster, I., G. Jones, R. Oliver, M. Bormans & B. Sherman,
Hart, B. T. & M. R. Grace (eds), Nitrogen Workshop 2000, 1997. Control strategies for cyanobacterial blooms in weir
Land and Water Australia, Canberra, Australia, 86–90. pools. In Davis, J. R. (ed.), Managing Algal Blooms: Out-
Van Dijk, G. M. & E. Van Donk, 1991. Perspectives for sub- comes from CSIROÕs Multi-Divisional Blue-Green Algal
merged macrophytes in shallow lake restoration projects in Program. CSIRO Land and Water, Canberra, ACT,
The Netherlands. Hydrobiologia 24: 125–131. Australia, 31–40.
Webster, I. T., B. S. Sherman, M. Bormans & G. Jones, 2000. limitation. Australian Journal of Plant Physiology 22:
Management strategies for cyanobacterial blooms in an 331–340.
impounded lowland river. Regulated Rivers Research and Young, W. & I. Prosser, 2001. Modelling long term average
Management 16: 513–525. nutrient loads in large-scale river networks. Paper to Third
Williams, R. M., 1993. Saline groundwater inﬂows to the Australian Stream Management Conference, August 2001,
Darling River from Mungindi to Wentworth: Run of river Brisbane, Australia.
study 1991/2. NSW Dept. of Water Resources Technical Young, W. J., F. M. Marston & J. R. Davis, 1996. Nutrient
Report TS92.027, Sydney, Australia. exports and land use in Australian catchments. Journal of
Wood, M. D. & R. L. Oliver, 1995. Fluorescence transients Environmental Management 47: 165–183.
in response to nutrient enrichment of nitrogen- and Zann, L. P., 1995. The state of the marine environment report
phosphorus-limited Microcystis aeroginosa cultures and for Australia. Technical summary. Dept. of the Environ-
natural phytoplankton populations: a measure of nutrient ment, Sport and Territories, Canberra, Australia.