Hydrobiologia (2017) 787:269–290 DOI 10.1007/s10750-016-2970-9
PRIMARY RESEARCH PAPER
Palaeoecological evidence for sustained change in a shallow Murray River (Australia) floodplain lake: regime shift or press response? Giri Kattel
. Peter Gell . Atun Zawadzki . Linda Barry
Received: 6 March 2016 / Revised: 23 August 2016 / Accepted: 24 August 2016 / Published online: 2 September 2016 Ó Springer International Publishing Switzerland 2016
Abstract Paleolimnological techniques can reveal long-term perturbations and associated stable state transitions of lake ecosystems. However, such transitions are difficult to predict since changes to lake ecosystems can be abrupt or gradual. This study examined whether there were past transitions in the ecological regime of Kings Billabong, a shallow River Murray wetland in southeast Australia. A 94-cm-long core, covering c. 90 years of age, was analysed at 1 cm resolution for subfossil cladocerans, diatoms and other Handling editor: Jasmine Saros G. Kattel P. Gell Water Research Network, Federation University Australia, Mt Helen, VIC 3350, Australia G. Kattel P. Gell School of Applied & Biomedical Sciences, Federation University Australia, Mt Helen, VIC 3350, Australia G. Kattel Hydrology and Water Resources Group, Department of Infrastructure Engineering, The University of Melbourne, Parkville, VIC 3010, Australia G. Kattel (&) State Key Laboratory of Lake Science and Environment, Nanjing Institute of Geography and Limnology, Chinese Academy of Sciences, Nanjing 210008, China e-mail:
[email protected] A. Zawadzki L. Barry ANSTO Institute for Environmental Research, Lucas Heights, NSW 2232, Australia
proxies. Prior to river regulation (c. 1930), the littoral to planktonic ratios of cladocerans and diatoms, and bulk sediment d13C values were high, while the period from c. 1930 to c. 1970 experienced considerable changes to the wetland ecosystem. The abrupt nature of changes of planktonic cladocerans and diatoms, particularly after the onset of river regulation (1930s), was triggered by inundation, high rates of sedimentation and shifts in bulk sediment d15N values. However, the transition of a once littoral-dominated community, to one favouring an increasingly turbid, planktondominated trophic condition following river regulation was relatively slow and lasted for decades. The progression to a new regime was likely delayed by the partial recovery of submerged plant communities and related internal dynamics. Keywords River Murray Floodplain wetland Bulk stable isotopes of carbon and nitrogen Subfossil cladoceran Subfossil diatom, regime shift
Introduction Following the arrival of European settlers in the early 1900s, forest clearance and grazing of domestic animals in the catchments of the River Murray system resulted in accelerated soil erosion and rapid sedimentation (Thoms et al., 1999; Gell et al., 2009). The construction of weirs and dams along the River Murray has modified its natural hydrological patterns,
123
270
particularly the variability, duration and frequency of flows (Thoms et al., 2000), altering the transport of organic carbon from the riparian zone to the river and, subsequently, to wetlands (Bunn & Arthington, 2000). For example, permanent wetlands shifted to intermittent wetlands after flood control and flow reductions, whereas previously seasonal wetlands became permanent with water levels held to that of structures (weirs) in the main river and associated river channels (Kattel, 2012). While these hydrological alterations remain, sustaining pressures on ecosystems, the stressors on the condition of these wetlands are further exacerbated by recent prolonged drought, with declines in March– May rainfall as high as 61% across southeast Australia (Murphy & Timbal, 2008; Mills et al., 2013a, b). This combination of catchment and climate change poses a considerable challenge for programs intended to rehabilitate these wetland systems (Gell & Reid, 2014). Paleolimnological studies undertaken across the upper reaches of the River Murray have revealed the loss of submerged aquatic macrophyte communities in many billabongs following the arrival of European settlers (Reid et al., 2007). In these billabongs, subfossil cladoceran assemblages reflected a rapid transition from littoral chydorid species to assemblages dominated by planktonic bosminids indicating a rapid decline of submerged macrophyte stands (Ogden, 2000). This evidence is supported by rapid transitions from epiphytic to planktonic diatom communities (Reid et al., 2002, 2007). The loss of a macrophyte-dominated regime in upstream River Murray billabongs was inferred to result from landscape clearance, followed by the abstraction of river water for irrigation and the increased use of fertilisers for agriculture (Reid et al., 2007). In recent decades, it has been revealed that macrophyte-dominated shallow lakes and billabongs in the lower reaches of the river have also been rapidly transformed with evidence of fringe-emergent macrophyte (Typha sp.)-dominated wetlands (Gell, 2012) associated with turbid water states (Grundell et al., 2012). Such transformation from submerged aquatic vegetation to emergent macrophyte dominance also appears to have occurred despite waterlogging. The maintenance of stable weir pool levels has reduced germination and recruitment of submerged vegetation and increased other emergent taxa in the lower Murray River system (Blanch et al., 1999, 2000). A synthesis of paleolimnological studies
123
Hydrobiologia (2017) 787:269–290
on wetlands from across the southern Murray–Darling Basin today reveals most of those regulated wetlands, including the permanently inundated systems, to be impacted by increased sediment flux and almost half of those wetlands by elevated nutrient levels (Grundell et al., 2012; Gell & Reid, 2014, 2016). In ‘clear water’ wetlands, submerged macrophytes play an important role in maintaining high water clarity through allelopathic effects, the stabilisation of sediments and through tight nutrient cycling (Blindow et al., 2000). This clarity acts to provide light energy for plant growth through photosynthesis reinforcing the trophic structure of the system. A sudden loss of water transparency, directly as a result of the increased inputs of fine sediment, or indirectly through nutrient enrichment, which leads to the increased growth of phytoplankton, may further impact the light regime of the water body (Kattel et al., 2016). Such a phenomenon may compromise the capacity of plants to assimilate nutrients and so may destabilise the trophic structure (Jeppesen et al., 2001). When perturbations (e.g. nutrient loads) exceed a critical threshold level in the clear-water state, an abrupt shift may occur, leading to unprecedented eutrophication (Carpenter et al., 1999; Wang et al., 2012). Consequently, the feedbacks associated with the loss of plants, comprising increased wind-driven sediment resuspension followed by a reduced light environment, can stabilise a phytoplankton state (Scheffer & Carpenter, 2003). Once a perturbation occurs, however, the response may be slow (Walker & Salt, 2006) as the feedbacks that establish a new state gradually take hold. It has been argued that, despite a highly dynamic ecosystem compared to other terrestrial and marine environments (Hughes et al., 2013), the feedbacks in shallow temperate lakes can be mediated by slowly changing variables such as climate change and also lake ontogenetic processes such as accumulation of nutrients in the lake bed over a longer time scale (RandsaluWendrup et al., 2016). However, at a time of gradual change, a tendency to an abrupt change exists, and this can be brought upon by rapidly changing variables such as the sudden release of water column nutrients derived from flash floods, forest clearing and cultivation in the catchments (Carpenter & Lathrop, 2008). As a result of these linear and non-linear interactions, the system becomes increasingly complex. Today, the investigation of these mechanisms in wetlands has been identified as a research priority for palaeoecology
Hydrobiologia (2017) 787:269–290
(Seddon et al., 2014), and, in a system that shows evidence of change from early European settlement, palaeoecological approaches may represent the sole means of understanding the nature of widespread shift to phytoplankton dominance in Murray River floodplain wetlands. Over the past century, the human-induced rapid geomorphological variations have profoundly modified the landscapes of wetlands of the Murray River system (Gell et al., 2009). The inevitable human impacts, which include land conversion, water extraction, infrastructure development, pollution and the introduction of invasive species in the Murray River landscapes, have brought sustained pressures to wetland ecosystems in the region (Davis et al., 2010). These persistent pressures by humans have also displaced food resources for consumers in the region’s waterways (Bunn & Arthington, 2000). While it is clear now that many of the lower Murray River wetlands have changed with the persistent pressure response, or ‘press response’, it remains uncertain however, whether the persistence of planktonic communities is a direct response to continued high loads of sediments and nutrients, or whether the internal dynamics of these systems is acting to resist recovery to macrophyte-rich wetlands (Gell & Reid, 2016). Recently, the Murray Darling Basin has been the subject of a large number of local and catchment-wide ecosystem assessments (e.g. Norris et al., 2007) leading to the release of the Murray–Darling Basin Plan (MDBA, 2013). In concert, a number of measures have been implemented to rehabilitate the waterways of the River Murray including its floodplain wetlands. Provision of environmental flows and the restoration of landscapes are prime objectives to rehabilitate degraded wetland systems and to promote ecological resilience. However, the ecological condition, relative to a baseline, of many shallow wetlands and billabongs is unknown due to the limited availability of historical data. The identification of benchmark conditions, and variations in the alternative stable states of ecosystems, is essential to guide wetland restoration programs (Bennion et al., 2011). High resolution multiproxy paleolimnological analyses in shallow floodplain wetlands can provide evidence for the nature of change over time, and so inform the management actions that may be most effective. By reconstructing the abundance of key biota over time, palaeolimnology offers an important platform
271
for understanding long-term changes in biodiversity and ecosystem function (Jeppesen et al., 2001; Sayer et al., 2010). However, hitherto multi-proxy paleolimnological records have rarely been used to test the occurrence of regime shifts in shallow wetlands, complete with the shift in stabilising feedbacks that entrench them in a new condition. In this paper, we use multi-proxy paleolimnological approaches to reveal a sustained shift in the condition of the Kings Billabong ecosystem, a lowland River Murray floodplain wetland. The development of a 210Pb-age model for sediment accumulation, analysis of subfossil assemblages of cladocerans and diatoms, and of bulk sediment stable isotopes of organic carbon and nitrogen, is directed towards understanding the nature of change in the Billabong and the dynamics that appear to have entrenched it in a new state.
Materials and methods Study area Kings Billabong is a shallow (2.3 m deep) ox-bow lake or ‘cut-off meander loop’ of the River Murray, near Mildura (northwest Victoria). It lies in a reserve managed by Parks Victoria (Fig. 1). Part of Kings Billabong, which covers c. 2,200 ha, was first gazetted in 1979 as a Wildlife Reserve under the 1958 Land Act (Victoria) following the recommendations of the Land Conservation Council (LCC) for the Mallee Study Area. Historically, Kings Billabong was a significant site of the Nyeri Nyeri aboriginal community providing a rich source of goods and services (Parks Victoria, 2008). In its natural condition, the Billabong was likely intermittently filled at times of high flows, and was exploited as a resource by the aboriginal community (Lloyd, 2012). However, following the intensification of European settlement in the Victorian Mallee region from 1891 to 1923, the landscape of the reserve changed substantially (www.murrayriver. com.au). Kings Billabong was first used as a water storage basin in 1896, and three pumping stations were operational during this time. Between the 1890s and 1960s, Kings Billabong was drained following summer irrigation seasons for the maintenance of the Central Pump suction lines (Lloyd, 2012). After the 1970s, high water levels were maintained throughout the off-peak irrigation season (Lloyd, 2012). Despite
123
272
Hydrobiologia (2017) 787:269–290
Fig. 1 Location map of the sediment core collected from the deepest point (2.3 m) of Kings Billabong, Murray River system northwest Victoria (near Mildura). In proximity to Kings Billabong, regulation impacts commenced with the construction of Locks 11 and 15 at Mildura and Euston between 1927 and
1937. The link between the weir pool of Lock 11 and Kings Billabong leads to the permanent inundation of much of the site. The shaded areas are the areas of water, and green areas are catchment vegetation of the lake and river
the dedication of Kings Billabong as a Wildlife Reserve, intensive land use led to considerable soil compaction and bank erosion in its vicinity (MCMA, 2006). The natural flow of the River Murray has been significantly modified by the construction and operation of a series of locks, weirs and storages (Gippel & Blackham, 2002). The first of these was commissioned in 1922 and most were in place by 1936. The changes brought by this infrastructure have affected the hydrology, and, in particular, the variability, duration and
frequency of flows in the river (Gippel & Blackham, 2002). In proximity to Kings Billabong, regulation impacts commenced with the construction of Locks 11 and 15 at Mildura and Euston between 1927 and 1937 (Mackay & Eastburn, 1990). The construction of these weirs altered downstream river flows and naturally occurring flood pulses into adjacent wetlands, including into Kings Billabong (Gippel & Blackham, 2002). The weir pool of Lock 11 (Mildura) backs up as far as Kings Billabong, which leads to the permanent inundation of much of the site.
123
Hydrobiologia (2017) 787:269–290
Permanent waterlogging in Kings Billabong led to the dieback of river red gum (RRG; Eucalyptus camaldulensis Dehnh.) forests, and the establishment of fringing cumbungi (cat-tail, Typha sp.) vegetation (Parks Victoria, 2008). However, prior to river regulation, the lake may have been dominated by submerged vegetation as revealed by the preservation of charophyte macrofossils (e.g. Nitella sp.) (Kattel unpublished data). A recent survey by the Murray Darling Freshwater Research Centre (MDFRC—Mildura) suggests that Kings Billabong also contains a range of native and introduced fish species including the Murray cod [Maccullochella peelii peelii (Mitchell, 1838)], golden perch [Macquaria ambigua (Richardson, 1845)], Australian smelt [Retropinna semoni (Weber, 1895)], bony herring [Nematalosa erebi (Gu¨nther, 1868)], flathead gudgeon [Philypnodon grandiceps (Kreft, 1864)], flyspecked hardy-head (Craterocephalus stercusmuscarum fulvus Ivantsoff, Crowley & Allen, 1987), and one non-native species, mosquito fish (Gambusia holbrooki Girard, 1859) (Ellis unpublished data). Three groups of native zooplankton recorded include cladocerans (Daphnia sp.; Bosmina meridionalis Sars, 1904), copepods and shrimps (Macrobrachium sp.) (Kattel unpublished data). The loss of RRG forests in Kings Billabong continued until the 1950s as timber was used for steam-operated pumps and paddle boats on the river (Parks Victoria, 2008). The impact of infrastructure development, widespread deforestation, and waterlogging, followed by intensive irrigation of saline dune systems for horticulture, have all led to secondary salinization around the southern section of the billabong. This is mostly evident in the nearby Psyche Bend Lagoon (Fig. 1), which has received saline ground water discharge and irrigation drainage water resulting in the death of gum forests, understory vegetation and wetland macrophytes (Gell et al., 2002). Experimental design The reconstruction of the changes in Kings Billabong relied on three major paleolimnological approaches applied to a 94-cm core: (i) 210Pb-age modelling was carried out to identify the timing of changes in sedimentation rates and chemical and biological indicators, (ii) analysis of multi-proxy, physical, biological and chemical indicators, including grain size, subfossil cladocerans, diatoms and stable isotope
273
ratios of organic carbon and nitrogen in bulk sediment samples, and (iii) the application of ordination techniques to the biological data to reveal the degree to which the wetland’s ecological state had changed during the last century or more. Sediment coring and 210Pb-age modelling In September 2011, eight short sediment cores were collected along a margin-to-centre transect in Kings Billabong, including a 94-cm-long, 80-mm-diameter piston sediment core taken from the deepest point (2.3 m) near the lake centre, for detailed analysis. The chronology for this core was established using the 210 Pb dating approaches as described by Appleby (2001), employing the Constant Initial Concentration (CIC) (Brugam, 1978) and Constant Rate of Supply (CRS) (Appleby & Oldfield, 1978) models. In order to establish a 210Pb-age model, a total of 9 sub-samples, between 0 and 60 cm depth, was analysed for 210Pb activities. Sediment samples were processed following the methods outlined in Harrison et al. (2003). The processed samples were analysed by alpha spectrometry. The 210Pb dating was conducted at the Australian Nuclear Science and Technology Organisation (ANSTO), Lucas Heights, Australia. Measurement of physical, biological and chemical assemblages Dry weight (DW) and loss on ignition (LOI) analysis About 2 g of wet sediment was placed in crucibles and heated overnight in an oven at 105°C. Crucibles with oven-dried sediment were kept in a desiccator to prevent re-absorption of moisture. The dry weight (DW) percentage of sediment was calculated following the methods described by Meyers & Teranes (2001). The crucibles with dried sediments were then placed in a furnace (Carbolyte) at 550°C for 2 h in order to estimate the percentage loss on ignition (LOI). The percentage DW was used for quantifying the number of subfossil cladoceran individuals per gram of dry sediment (e.g. Kattel et al., 2008). Grain size analysis Variation in the physical nature of the sediment, such as changes in grain size, can affect the adsorption of
123
274
unsupported 210Pb to sediments. An increase in unsupported 210Pb activity can occur with an increase in the specific surface area of sediments (He & Walling, 1996). The grain size distribution in sediments was determined to ascertain if there were factors influencing the unsupported 210Pb activity other than radioactive decay. A total of 25 sub-samples (every 2–3 cm intervals down to 60 cm depth) was analysed for grain size analyses at ANSTO. Samples were treated with hydrogen peroxide and sodium hexametaphosphate to oxidise organic matter and to break up clay aggregates. The treated samples were analysed on a Malvern Mastersizer 2000 laser diffraction spectrophotometer to determine grain size distribution in the range of 0.01–2000 lm. Analyses of subfossil cladocerans and diatoms The 94-cm core was subsampled at high resolution (1 cm interval) for the analysis of subfossil cladoceran and diatom assemblages. For cladoceran analyses, approximately 3–4 g of wet sediment was treated with 100 mL of 10% KOH solution, and heated at 60°C on a hotplate for at least 45 min. The sub-sample mixture was sieved through a 38-lm mesh with running tap water. A few drops of safranin were added after washing to stain the remains. Slides were then prepared by pipetting 0.05 mL of each subsample onto microscope slides. Two hundred or more cladoceran remains (carapaces, head shields, postabdomen, ephippia and post-abdominal claws) were counted at 4009 magnification. Badly fragmented, but identifiable, remains were counted. The dry weight percentage of each sediment sample was measured to calculate the counted portion of remains present per gram of dry sediment (Kattel et al., 2008). Cladoceran taxa were identified following Shiel & Dickson (1995) and Szeroczynska & Sarmaja-Korjonen (2007). For diatoms, samples were digested in 10% HCL and 10% H2O2 (Battarbee, 1986). The diatom concentration was determined by mixing a known quantity of divinylbenzene (DVB) marker microspheres (size range 5–10 lm). The DVB is resistant to the organic solvents used in diatom mountants (Battarbee et al., 2001). The stock suspension was made with a concentration of 5 9 106 spheres per ml and the concentration was then precisely determined using an electronic panicle counter (Battarbee et al., 2001). The
123
Hydrobiologia (2017) 787:269–290
microspheres were added at the final stage of preparation to avoid loss during the washing processes. A drop of ammonia and mercuric chloride was used to reduce clumping and avoid the growth of bacteria (Battarbee et al., 2001). Aliquots of 400 and 800 ll of treated sample were dried on coverslips and mounted on microscope slides using Naphrax. Approximately 200 valves, a sum considered adequate to characterise a diatom community (Bate & Newall, 1998), were counted per slide. Broken valves were counted if the central area was present. Counting was undertaken on a Nikon Eclipse E600 microscope under differential interference contrast (DIC), using x1000 magnification (oil immersion lens). Diatom taxa were identified following Krammer & Lange-Bertalot (1986, 1988, 1991a, b) and Sonneman et al. (2000). The counts for each diatom taxon were expressed as a percentage of the total valves counted.
Stable isotope ratios of carbon (d13C) and nitrogen (d15N) analyses of bulk sediment A total of 45 bulk sediment sub-samples (1 cm3 taken at 2 cm intervals) from the dated sediment core was analysed for stable isotope ratios of organic carbon (d13C) and nitrogen (d15N) in the ANSTO laboratories. Each sample was divided into two fractions (Fernandes & Krull, 2008). One fraction of each sample was mixed with 1 mL HCl and placed in a water bath at 60°C for 2 h to remove carbonates. These sub-sample fractions were then rinsed five times with distilled, de-ionised water, followed by drying in an oven at 60°C. Samples were analysed on an Elementar VarioMICRO Elemental Analyser and a Continuous Flow Isotope Ratio Mass Spectrometer (GV Instruments IsoPrime). All d13C analyses were performed on the acid-treated fraction and results were normalised to reference standard IAEA C8 (consensus value: d13CV-PDB = -18.31%) (Le Clercq et al., 1998). All d15N analyses were performed on the untreated fraction, and a two-point normalisation was performed on the data using international reference standards IAEA N-2 (consensus value: d15NAIR = ?20.3% (Bohlke & Coplen, 1995) and USGS-25 (consensus value: d15NAIR = -30.4%) (Bohlke & Coplen, 1995). A number of check standards were included in all analyses.
Hydrobiologia (2017) 787:269–290
Ordination In order to examine how species are associated, and are aligned with sample depths, indirect ordination techniques were applied to the subfossil datasets of cladocerans and diatoms. Detrended correspondence analysis (DCA) was performed to assess the gradient length, or the standard deviation (SD) units, of the respective assemblages. DCA removes the arch effect and gradient compression in correspondence analysis by ‘detrending’ and ‘rescaling’ the first axis (Hill & Gauch, 1980). During rescaling, the positions of the sub-samples along the ordination axis are shifted to make the beta diversity constant (Gauch, 1982; ter Braak & Smilauer, 2002; Leps & Sˇmilauer, 2003). Hill’s N2 diversity index was calculated for both subfossil cladoceran and diatoms samples, since this index provides the effective number of abundant taxa in a sample (Hill, 1973). Only cladoceran species present in at least two of the sub-samples, with a relative proportion of more than 1% in at least one sample, were included for ordination analysis (Korhola, 1999). The first two DCA axes accounted for 31.3% of the cumulative variation of the species data. With a gradient length of 2.3 SD, a principal components analysis (PCA), a linear ordination technique, was used to assess variations amongst samples. The PCA Axis 1 and 2 sample scores, and Hill’s N2 diversity index, were calculated for subfossil cladocerans. The sample scores (eigenvectors), and eigenvalues were used to measure the strength of the PCA (ter Braak & Verdonschot, 1995). DCA was performed on the diatom data using all recorded species. The gradient length of the first two axes was 3.71 ([2.5 SD), thus a correspondence analysis (CA) was chosen to specify taxa co-occurring closest, or furthest apart, and to measure the variation shown by the first two axes of the CA ordination (Rees et al., 2000). The DCA Axis 1 and CA Axis 1 and 2 sample scores, as well as Hill’s N2 diversity index, were calculated for subfossil diatoms.
275
distance dissimilarity coefficient of samples, and stratigraphically constrains this for quantitative zonation. TILIAGRAPH was used for producing stratigraphic diagrams, incorporating the dendrograms from CONISS (Grimm, 1987). The PCA sample scores for Axis 1 and Axis 2, and Hill’s N2 diversity index, obtained following the indirect ordination, were used to compare with the stratigraphic assemblage changes and the littoral:planktonic ratios of subfossil cladocerans. The DCA Axis 1 and 2 scores, and Hill’s N2 diversity index, obtained after indirect ordination, were also compared with the stratigraphic changes in diatom assemblages and the littoral:planktonic diatom ratios.
Results Age modelling The unsupported 210Pb activities in the sediment core analysed from the Kings Billabong were relatively low, therefore the calculated chronology is treated with caution. The unsupported 210Pb activity at the top of the core was only 22 Bq/kg, and decreased to about 3 Bq/ kg at 57 cm depth (Table 1; Fig. 2a). Both CIC and CRS 210Pb dating models were used to develop the core chronology (Table 1). These outcomes from these models were in relatively close agreement (see Fig. 2b). The CRS model suggests that, at 51 cm the sediment was 42 ± 6 years old, which is equivalent to a time of deposition at c. 1969 ± 6 AD. The mass accumulation rate (MAR), based on the CRS model, increased from 0.5 to 0.8 g/cm2/year between 1976 and 2009, while the CIC model returned a constant MAR of 0.75 g/cm2/ year (Table 1). The MAR measurements, and the monotonically deviated 210Pb profiles of the core (Fig. 2a), suggest that CRS is a more reliable model to determine the age of the core (Appleby & Oldfield, 1978). Dating below 60 cm was estimated by extrapolating the MAR of 0.6 g/cm2/year.
Cluster analysis of subfossil cladocerans and diatoms
Changes in physical, chemical and biological assemblages
Constrained incremental sum of squares (CONISS) analysis was performed in TILIA on the subfossil cladoceran and diatom data. Dendrograms were produced to determine the main changes or shifts in biological assemblages. CONISS provides the chord-
Physical assemblages: grain sizes More than 80% of the grains extracted between 3 and 61 cm core depth ranged in size from 2 to 63 lm (Fig. 3). The lowest and highest percentages of grain
123
276
Hydrobiologia (2017) 787:269–290
Table 1 Kings Billabong core
210
Pb activities, CIC and CRS calculated ages and mass accumulation rates
Depth (cm)
Total 210 Pb (Bq/Kg
Supported 210 Pb (Bq/Kg)
0–1
60 ± 5
38 ± 4a
22 ± 6
1±1
1±1
0.7 ± 0.2
5–6
56 ± 3
38 ± 4
18 ± 5
2±1
2±2
0.8 ± 0.2
10–11
49 ± 3
32 ± 3
18 ± 4
5±1
5±2
0.8 ± 0.1
20–21
47 ± 2
30 ± 4
25–26
45 ± 2
30 ± 3
Unsupported 210 Pb (Bq/Kg)
a
Calculated CIC ages (years)
Calculated CRS ages (years)
CRS model mass accumulation rates (g/cm2/year)
17 ± 4
12 ± 2
13 ± 4
0.6 ± 0.1
15 ± 4
16 ± 3
17 ± 4
0.6 ± 0.1
30–31
48 ± 2
33 ± 2
16 ± 3
20 ± 4
22 ± 5
0.5 ± 0.1
42–43
45 ± 2
35 ± 3
10 ± 3
29 ± 5
35 ± 6
0.5 ± 0.1
42 ± 6
0.7 ± 0.4
50–51
46 ± 2
40 ± 3
6±4
35 ± 6
57–58
43 ± 2
40 ± 3a
3±4
40 ± 7
a
Activities were estimated from the nearest sample depth
a
Depth (cm)
0
10
20
30
Sediment age (years)
b
Unsupported 210 Pb Activitiy (Bq/kg) 40
0
0
0
10
10
20
20
30
30
40
40
50
50
60
60
70
70
10
20
30
2006
40
CIC
50
60
CRS
1990
Fig. 2 a King Billabong core unsupported plotted against depth
1970
210
Pb activities plotted against depth, b King Billabong core CIC and CRS model ages
size with less than 2 lm (clay) were measured from 40 to 41 cm and 52 to 53 cm, respectively, while the lowest and highest percentages of grain sizes with more than 63 lm (sand) were at 60 to 61 cm and 20 to 21 cm, respectively (Fig. 3). In general, however, there was only minor variation in grain size throughout the core. Biological assemblages: subfossil cladocerans and zonation
less than 1%, are presented in Fig. 4. The most commonly recorded cladoceran taxa include Bosmina meridionalis; Chydorus sphaericus O.F. Mu¨ller, 1785; Biapertura setigera (Brehm, 1931); Dunhevedia crassa King, 1853; Biapertura affinis (Leydig, 1860) and Alona guttata Sars, 1862. Using the results of CONISS, four zones were identified: KB-CLAD I, KB-CLAD II, KB-CLAD III and KB-CLAD IV (Fig. 4). KB-CLAD I (94–70 cm * pre-1930)
More than 40 species of cladoceran were recorded from the Kings Billabong sediments. These assemblages, other than the rare species recorded at frequencies of
123
This zone was characterised by the maximum abundance of littoral cladocerans as revealed by
µm
µm
277
c. 1900
c. 1930 KB-CLAD I
0
PC
e pl m sa
35-36
20 N
e pl A
e or sc
m sa
A
32-33
2
e or sc
s xi
1
27-28
x de In
A
xi
s
2
25-26
0
A PC .p nc co
g er
42-43
300 400
m ra
y dr of
40-41
500
se
37-38
lC ta To
io at R la
40
c ni to nk
52-53
100
do la
47-48 50-51
200
ns ra ce
45-46
tt Li
20
:P al or
55-56
20
40
60
80
100 0
20
100
20 0
i ph le ta To
0
ia pp
200
300
57-58 60-61
40 20 60 40 20 20 80 60 20
40
ld ie sh ad he is d str e i o s is a tif tir s ar a cu qu sa en r ec isi usu eri ul er a is in la as o b ng tig n id s gu isa ffin a ru n g rr cr ha ra se (u rcu b a g lo sp an exc a a a k a lo ia ad ra r ra ce us us us r ct r ed qu re lla rtu rtu rtu rtu er tu pt o or ev or d or a e e e e a e h d d on o n lo n ia p ia p ia p a p ia p am hy hy hy un Al C A Bi B B Al C C D C B B
Percentage Cladocera
20 20 40 20 60 20 60 95
90
85
80
75
70
65
60
55
50
45
40
35
30
25
20
15
10
5
0
s Bo
in m
20
a
m
io id er
40
lis na
Depth (cm)
In this zone, the abundance of littoral cladocerans began to decline as revealed by lower littoral:planktonic ratios (Fig. 4). The abundance of D. crassa, dominant in the previous zone, declined rapidly (Fig. 4). Meantime, small Alona species, such as Alona guttata, and the planktonic taxon B. meridionalis, began to increase (Fig. 4). Cladoceran ephippia were recorded throughout this zone. Although the N2
s gi de r in i oi ue i ih e a on i t al e r bo s rch am a vid u tta a ru ni pe a a a c a d a g h n on on on o ap cr lo Al Al A A Al D
40
KB-CLAD II (71–51 cm * 1930–1970)
) III
littoral:planktonic ratios (Fig. 4). The littoral assemblage was mainly dominated by Dunhevedia crassa. Other littoral chydorid taxa, including Alona guttata; Chydorus sphaericus and Graptoleberis testudinaria (Fischer, 1851), were also abundant in this zone. The abundance of the littoral taxon, Camptocercus rectirostris Schoedler, 1862, was relatively low in this zone. Neither the PCA Axis 1 scores, nor the N2 Diversity Index, showed any obvious changes within this zone. The abundance of cladoceran ephippia was as high as 150 individuals per gram of dry sediment (Fig. 4). The abundance of planktonic taxa, i.e. Bosmina meridionalis, increased briefly at 87 cm, but remained low throughout this zone (Fig. 4).
) V d el h i ra ria ds ce na ra ea do di ( h la ce tu s C s do li ic te la ra is st to n lC er au nk ra b o e ia la itt ol ig l p ll pt yd t a ta ra Le To G To
Fig. 3 Grain size measurements of the sediment core collected from Kings Billabong
60
80
100
Percentage
Fig. 4 Stratigraphic records of subfossil cladocerans and concentration of their ephippia in Kings Billabong based on CONISS cluster analysis
2
22-23
t en m di
Depth (cm)
4
6
20-21
Total sum of squares
17-18
c. 1970
c. 1990
CONISS
12-13
8 10 12 14 16 18 20
8-9 10-11
KB-CLAD II
5-6
KB-CLAD III
3-4
c. 2000
0-1
KB-CLAD IV
Age (AD)
>6 3
263
<2
µm
Hydrobiologia (2017) 787:269–290
123
278
index showed a minor increase in diversity, the PCA Axis 1 and 2 values did not change greatly (Fig. 4). KB-CLAD III (51–30 cm *1970–1990) The main feature of this zone was a significant increase in the abundance of total planktonic cladocerans and a decline in the abundance of total littoral cladocerans (Fig. 4). Amongst the planktonic taxa, B. meridionalis dominated this zone. However, there was a brief spike in the abundance of the littoral cladoceran D. crassa at 44 cm, coincident with a decline in B. meridionalis. Two relatively abundant littoral taxa, Alona guttata and Alona quadrangularis (O. F. Mu¨ller, 1785), and other less abundant littoral taxon, Camptocercus rectirostris, were also recorded within this zone. More Daphnia post-abdominal claws were recorded in this zone than elsewhere (Fig. 4). KB-CLAD IV (30–0 cm *1990–2000) In this zone, the littoral:planktonic ratios were low and the zone was dominated by B. meridionalis. Four littoral species, including A. guttata, Biapertura longinqua Smirnov, 1971, A. quadrangularis and Chydorus sphaericus were common in this zone. The littoral taxon, Camptocercus rectirostris, was also recorded in low abundance in this zone. However, the dominant littoral taxon, D. crassa declined significantly (Fig. 4). At 15 cm, as many as 300 ephippia were recorded per gram of dry sediment. The PCA Axis 1 scores and the N2 diversity index values remained constant throughout this zone (Fig. 4). Biological assemblages: subfossil diatoms and zonation A total of 90 species of diatoms was recorded from the Kings Billabong sediments. The diatom flora shows distinct changes through the core. The stratigraphic diatom data were divided into five assemblage zones on the basis of a dendrogram derived using CONISS (Grimm, 1987).
Hydrobiologia (2017) 787:269–290
types in this zone (Fig. 5). This dominance is shown by the high littoral:planktonic diatom ratios (Fig. 5). The epiphyte, C. placentula was dominant in the middle part of the zone, and showed a gradual decrease towards the top (Fig. 5). Only two planktonic taxa, Aulacoseira subborealis (G. Nygaard) L. Denys, K. Muylaert & K. Krammer 2003 and A. granulate (Ehrenberg) Simonsen 1979, were recorded in high numbers in this zone (Fig. 5). At 72 cm, a facultative planktonic taxon, Pseudostaurosira brevistriata (Grunow) Williams and Round 1987, was recorded for the first time (Fig. 5). The DCA Axis 1 showed a positive relationship with epiphytic (C. placentula) and benthic (E. adnata) diatoms as well as littoral:planktonic ratios of diatoms (Fig. 5). KB-DIAT II (70–51 cm *1930–1970) The benthic and epiphytic taxa that dominated the previous zone began to decline with a corresponding increase in open water or plankton-dominated taxa such as small species within the Fragilariaceae (Fig. 5). The significant decline in the littoral:planktonic ratios of diatoms coincided with the loss of benthic–epiphytic taxa in this zone (Fig. 5). Both epiphytic species, Cocconeis placentula and Gomphonema truncatum Ehrenberg 1832, began to decline from this zone. Benthic taxa, such as Achnanthidium minutissimum (Ku¨tz.) Czarnecki 1994, only showed small peaks, while the planktonic taxa Aulacoseira granulata (Ehrenberg) Simonsen 1979; A. granulata var angustissima (O. F. Mu¨ller) Simonsen 1979 and A. ambigua (Grunow) Simonsen 1979 became dominant. The planktonic species Cyclotella meneghiniana F.T. Ku¨tzing 1844 is also present in this zone (Fig. 5). Other taxa, known from turbid and disturbed water regimes and believed to be facultative planktonic, such as Pseudostaurosira brevistriata; Staurosira construens var construens and Staurosira construens var venter (Ehrenberg) Hamilton 1992, gradually increased in this zone (Fig. 5). The DCA Axis 1 showed a positive relationship with epiphytic (C. placentula) and benthic (E. adnata) diatoms as well as littoral:planktonic ratios of diatoms (Fig. 5).
KB-DIAT I (94–70 cm *pre 1930) KB-DIAT III (51–35 cm *1970–1980) Benthic and epiphytic taxa, including Epithemia adnata (Ku¨tzing) Bre´bisson 1838 and Cocconeis placentula Ehrenberg 1838, were the predominant
123
In this zone, some planktonic species, including A. subborealis and A. granulata var angustissima,
c. 1900s
la l ta To
20
l: P
ra t to
80
Li
Valves g
-1
200 100
c
ni
to nk
s
60
om at di
60
80 100
l ta To
20
40
ic on kt an pl
80 60 40 20 40 20 40 20
co
20
ira se
40
lis ea or bb su
an
Au
la
r va ta la
la
20
co
ira se
40
gr
u an ta
Au
la
95
90
85
80
75
70
65
60
55
50
45
40
35
30
25
20
15
10
5
0
Au
la
20
co
i se
ra
Au
la
20
co
i se
ra
40
gr
u an a gu bi am
Percentage Diatoms
20
20
40
20
20
20
40
20
20
20
40
a o e um at rc rv tri m m im va va na is a la ris lu tu tis s la um s s ia ev al a el la tu el ul i ica s c a nu ta ii in br ul en en a rv er ph bia t a lica ch cu a n ar m ng nt ru un m i ru na gh ira t t r i a e a ul a bi ta e ba ul s o g t o lo x tin pa nc ce t n e c s et a p t ul bor dna ib up bis hun dia ns ns pi re pen a s pe pt o a um ph s sip d ob la n en uro g o o s o r p m i m i m y a p e c c m i u s r l r m ia ra bre lla ral v ho a c s s i a a a d diu es a a is el a s e on e m a a c ne thid d i r r h r a la ost a e a p a o l i i i i t e i o o l i i r o l t l l h lla n it s s s n h h th o t o e ne em te em t ia ph sig c u an ph n a n ro ro ro p a a p ire lio l l sc sc o lo ud co p h eno m b plo it h it h no om y ro avi hn au au au itz itz an yc e ho ll r yb ta om h t y i oc N N Pl Ep Eu G G N C Ps Am C C D Ep St St Ac R S e Su T r T o G Ac St C
Benthic Epiphyt ic F acult at ive Plankt onic
er nt s en
Plankt onic
t ru ns
Depth (cm)
a im iss st gu
Fig. 5 Diatom stratigraphy in Kings Billabong based on CONISS. Species [3% are on display. The diatoms are grouped as their habitat preference
40
KB-DIAT I
30 20 10
n tio t io t ra Ra en nc co om at di
300
D
CA
20
2 1 x is is Ax Ax nde I CA 2 N D
Total sum of squares
c. 1930
c. 1970
KB-DIAT II
KB-DIAT III
c. 1980
279
CONISS
KB-DIAT V
KB-DIAT IV
c. 1990
c. 2000
Age (AD)
Hydrobiologia (2017) 787:269–290
declined, while other facultative planktonic taxa, such as Pseudostaurosira brevistriata (Grunow) Williams and Round 1987, increased (Fig. 5). The abundance of the small facultative planktonic species, Staurosira construens var construens (Ehrenberg) Hamilton 1992 and S. construens var venter, fluctuated in this zone (Fig. 5). There was no obvious increase in benthic taxa except for a minor increase in the abundance of Epithemia adnata and Planothidium delicatulum (Ku¨tzing) Round and Bukhtiyarova 1996 (Fig. 5). The DCA Axis 1 scores showed some positive relationships between facultative planktonic (P. brevistriata), and epiphytic and benthic (C. placentula, E. adnata) diatoms at the beginning, but this began to decline after c. 1970 (Fig. 5). KB-DIAT IV (35–17 cm *1980–1996) There was no change in the abundance of benthic taxa, including Planothidium delicatulum (Ku¨tzing) Round and Bukhtiyarova 1996, within this zone (Fig. 5). Previously dominant planktonic taxa, such as A. granulata; A. granulata var angustissima and A. subborealis, also became less abundant. The planktonic taxon, A. ambigua, and the facultative planktonic taxa Pseudostaurosira brevistriata; S. construens var construens and S. construens var venter, were highly abundant in this zone (Fig. 5). The DCA Axis 1 values declined at the time the littoral:planktonic ratios decreased, with reciprocal increase in the planktonic assemblage of diatoms (Fig. 5). KB-DIAT V (17–0 cm *1996–2011) The abundance of most benthic and epiphytic species was very low in this zone (Fig. 5). Most planktonic and facultative planktonic species were common, while two facultative planktonic taxa, S. construens var venter and S. construens var construens, were dominant (Fig. 5). Amongst the planktonic taxa, A. granulata and A. ambigua were relatively abundant, while A. granulata var angustissima disappeared from the record (Fig. 5). The DCA axis 1 scores declined along with the littoral:planktonic diatom ratio (Fig. 5). Chemical assemblages: loss on ignition (LOI) and bulk sediment d13C and nitrogen d15N The amount of organic matter present in the sediment, as revealed by percentage LOI, increased gradually
123
280
Hydrobiologia (2017) 787:269–290
Depth (cm)
Fig. 6 Stratigraphic records of LOI and stable isotopes of carbon (d13C) and nitrogen (d15N) of bulk sediment subsamples collected from Kings Billabong, northwest Victoria, Australia
c. 2010 AD
2-3 4-5 6-7 8-9 10-11 12-13 14-15 16-17 18-19 20-21 22-23 24-25 26-27 28-29 30-31 32-33 34-35 36-37 38-39 40-41 42-43 44-45 46-47 48-49 50-51 52-53 54-55 56-57 58-59 60-61 62-63 64-65 66-67 68-69 70-71 72-73 74-75 76-77 78-79 80-81 82-83 84-85
c. 2006AD
c. 1990AD Flood c. 1975 AD
c. 1970 AD
River regulation –Mildura weir construction c. 1930 AD Use of steam operated pump 12
24
36
48 -26 -25 -24 -23 -22 1
from the base of the core to 50 cm. It began to decline from 47 cm being lowest at 38 cm depth (Fig. 6). The stratigraphic records of the stable isotopes of organic carbon (d13C) and nitrogen (d15N) in bulk sediment, as well as the organic content, as revealed by percentage loss on ignition (LOI), are shown in Fig. 6 (Carpenter & Brock, 2006; Scheffer & Jeppesen, 2007). The d13C values began to decline from 75 cm, where it was as high as -22.3%. From 60 cm, the d13C value declined to -25.1 %, and continued to decrease down to 32 cm, becoming as low as -25.6 % (Fig. 6). The d15N values began to increase from 75 cm and, by 60 cm, they were 3.5 % (Fig. 6). The d15N values were relatively constant above 60 cm, becoming highest at 21 cm (Fig. 6). Ordination-species associated with core samples PCA ordination of subfossil cladocerans The PCA of cladoceran species data shows that Axis 1 and 2 explained 59.5 and 14.9% of the total variation, respectively. The inter-species correlations in the PCA show that Axis 1 is associated with taxa such as D. crassa; Biapertura longinqua, B. affinis, B. meridionalis and A. quadrangularis while axis 2 is associated
123
2
3
c. 1900s AD
4
with dominant A. guttata and other less abundant species such as Camptocercus rectirostris (Fig. 7). Taxa such as D. crassa and Graptoleberis testudinaria are skewed towards the basal sediment samples, while Bosmina and Biapertura genera are associated mostly with upper samples. Alona guttata and C. rectirostris, on the other hand, are associated with samples towards the middle of the core (Fig. 7). CA ordination of subfossil diatoms The eigenvalues or the correlation coefficients for the sample and species scores for CA Axis 1 and 2 were 0.29 and 0.14, respectively, while the percentage variation explained by Axes 1 and 2 were only 27% and 13%, respectively. The diatom assemblages in the positive field of axis 1 are closely associated with the basal core samples which are represented by littoralbenthic and epiphytic species such as Epithemia adnata and Cocconeis placentula (Fig. 8). Many planktonic taxa, including A. granulata var angustissima, were associated with CA Axis 2 (Fig. 8). The samples around the middle of the core supported mostly planktonic (A. granulata, C. meneghiniana) and benthic species (Nitzschia amphibia Grunow 1862), Gyrosigma spenceri (W. Smith) Griffith &
Hydrobiologia (2017) 787:269–290
281 41
1.0
Fig. 7 PCA ordination of the subfossil assemblage cladocerans in 94 sediment core samples collected from Kings Billabong. Samples are presented in numbers. Only dominant species are shown with full scientific names
19
59 15
45
36
42 2
9 34 82
74
Axis 2
Dunhevedia crassa
76 79 73 78 80 70 Pseudochydorus sp 75 77
Saycia saycia
35 38 40 43
31
25
44 Camptocercus rectirostris Chydorus barrosi
1 32 24 5
Alona quadrangularis Bosmina meridionalis
Chydorus sphaericus Alonella 18 excisa 4 Ceriodaphnia Biapertura karua 6 Daphnia Biapertura headshield (unidentified) Leydigia australis Alona sp 14 sp Biapertura affinis Chydorus globusus AloXI Monospilus diporus Rhyaus AcraloDadaya 84 56 Alonella sp 33 28 81 Biapertura longinqua setigera 86 Biapertura 8 Graptoleberis testudinaria 67 46 29 49 Alona 83cambouei 93 Biapertura rigidicaudis Anchiostropus minor Alodav Raklab Acroperus elongatus Alona costata 3 37 Pleuroxus denticulatus Arccheopleuroxus Alonella exigua 11spIllsp Alona archeri 68 lattissima 61 27 69Kurzia Alona rectuangula Oxyurella tenuicaudis 47 9192
87 90
85 89 65 AloIX 88 66
63
55 72 30 39 5494 Pleuroxus aduncus 13 50 48 51 52 64 71 62 58
16
23
22
60 53 17 12 21
7 20
26
-1.0
57 10 Alona guttata
-1.5
Henfrey 1856; Eunotia serpentina Ehrenberg 1854, and also facultative planktonic species (Staurosira construens var construens, Pseudostaurosira brevistriata (Fig. 8).
Discussion Physical, biological and chemical assemblage change and alteration of ecological state Sedimentation rates and grain sizes The CRS and CIC age-depth models for Kings Billabong showed relatively close agreement. The low surface activities, typical in Australia, preclude the production of robust chronologies, however, rates of *1 cm/year can be justified as the 210Pb profiles in the CRS model suggested accumulation rates as high as 0.8 g/cm2/year (Fig. 2). The preferred chronology, based on a sedimentation rate as high as *1 cm/year, for this record, would place *1930 AD and *1970 AD with key points of change at 70 cm and at 51 cm, respectively. This rate is not dissimilar to those determined from other neighbouring wetlands connected to the main river (Fluin et al., 2010; Grundell
Axis 1
1.0
et al., 2012). Despite the fact that weirs can increase sediment trapping, the increasing MAR in Kings Billabong, since 1970 (Tab. 1), is a characteristic feature of the modern lower River Murray floodplain wetlands, which can have MAR as high as 2 g cm-2 year-1 (Gell et al., 2009). The amplification of sediment fluxes in Kings Billabong is likely to have been caused by ongoing connection with the River Murray (Gell et al., 2009). The import of fine sediments from the river channel to Kings Billabong following regulation increased the rate of accumulation of allochthonous sediments and organic matter (Table 1) . The analysis of diatom assemblage changes across several wetlands in the lower Murray River floodplain (Gell & Reid, 2014) allows for an assessment of regional scale changes. Several records (e.g. Fluin et al., 2010; Grundell et al., 2012) report the rise in Aulacoseira granulata following regulation in the 1930s. The rise in these river plankters has been attributed to the hydrological changes associated with river regulation (Grundell et al., 2012). Similarly, Aulacoseira subborealis increases substantially at around 1950–1960 in several records (Fluin et al., 2010; Grundell et al., 2012). The existence of common patterns of change in records of river plankton, across
123
282
Hydrobiologia (2017) 787:269–290
Fig. 8 CA ordination of the subfossil assemblage diatoms in 94 sediment core samples collected from Kings Billabong. Subsamples are presented in numbers. Only dominant species are shown with full scientific names
several hundred river kilometres, suggests a common cause, most likely the river as a source. This evidence suggests that a regional biostratigraphy can have the potential to provide greater confidence in the chronologies produced where radiometric activities are low, such as in Kings Billabong. Although the sedimentation rates were high following river regulation, the grain size records did not show an obvious change. The permanent inundation in Kings Billabong after river regulation (1930s) may have reduced the frequency of turbulence in the lake leading to the relatively small variation in grain sizes, which ranged from 1 to 61 lm. In many floodplain wetlands, the river can be the principal source of sediments and so can be responsible for changes in particle sizes. The lack of variation in grain size in the Kings Billabong record suggests that sediments associated with past flooding events, such as those in 1956 and 1973–1975, are not well recorded by grain size.
123
Changes in subfossil cladocerans and diatoms The subfossil assemblages of cladocerans and diatoms in Kings Billabong over the past century have provided strong evidence of interplay between positive and negative feedback mechanisms that may drive possible alternative stable state transitions of regulated shallow wetlands of the lower River Murray system in southeast Australia. Two periods in the past: since regulation in the 1930s, and from the 1970s to the last drought period (2001–2009), the cladoceran and diatom communities of Kings Billabong have undergone considerable change. Unlike in the 1970s, the littoral diatom and cladoceran communities were replaced abruptly by planktonic ones at the time of river regulation (1930s) (Figs. 4, 5). Such abrupt changes in the ecosystem indicate the sudden hydrological alterations of the River Murray with increased disruption of connectivity (Kattel, 2012). However,
Hydrobiologia (2017) 787:269–290
the changes documented for the 1970s are gradual, and could be associated with various factors including the combined climatic and human-induced hydrological factors. The nutrient mass balances in arid wetlands are directly influenced by changes in loading and hydraulic retention time and also indirectly by alterations in trophic structure, including changes and recovery of macrophyte coverage over the longer time ¨ zen et al., 2010). Rising temperatures, period (O together with prolonged drought, further amplify the change of trophic structures and dissolved oxygen ¨ zen et al., 2010). concentrations in the water column (O Despite the high littoral:planktonic ratios before regulation, the PCA Axis 1 and N2 index of cladocerans did not reveal a close relationship, suggesting that the subfossil cladoceran assemblage change did not come with an alteration in diversity. Although the N2 index for diatoms did not change greatly, the close correlation between the DCA Axis 1 and littoralbenthic diatom ratios until 70 cm depth suggests higher biological diversity prior to river regulation and a simplification of the assemblages since. In some instances (e.g. in the 1990s), the timing of the shift to a planktonic flora and fauna coincided with an increase in the abundance of cladoceran resting eggs indicating increased ecological stress for zooplankton in Kings Billabong (Fig. 4). A range of physical and chemical factors including variation in cold climates, dehydration and altered concentration salts and pH in wetlands, and ecological factor such as increased fish predation pressure, has been described to drive cladoceran populations to undergo a resting phase by producing a large number of fertile eggs (Sarmaja-Korjonen, 2003). Following river regulation, in particular, stress may have affected the cladoceran populations as a result of prolonged inundation, sustained sedimentation and nutrient enrichment followed by eutrophication. Lloyd (2012) reported that, throughout this period, the water level of Kings Billabong was maintained permanently. Although subfossil records of fish were not available for this study, the constant water level in Kings Billabong may have resulted in stress amongst the zooplankton populations. The higher preservation of Daphnia evident after the 1970s also indicates increased trophic interactions in Kings Billabong during this period (Fig. 4). For example, the increase in the depth gradient can lead to niche overlapping of space for vertical migration of
283
planktivorous fish and Daphnia and result in higher predation pressure by fish (DeMott & Kerfoot, 1982). Also, the gradual loss of littoral submerged macrophytes, due to inundation, can lead to further loss of habitat for chydorids across the horizontal gradient limiting the refuge of horizontally migrating cladocerans (see Jeppesen et al., 2001). Both the PCA and CA first axes show distinct variations in conditions of habitat (littoral-shoreline and deep-open water), which influenced the community dynamics of both cladocerans and diatoms prior to, and after, river regulation (Figs. 7, 8). These conditions appear to be, at one end of the gradient represented by the axes, an ecosystem with submerged vegetation under a clear water regime as indicated by the predominance of littoral cladocerans and diatoms. For instance, epiphytic species of cladocerans and diatoms, such as Dunhevedia crassa and C. placentula and E. adnata, are predominantly associated with samples across the basal sections of the sediment core, or in the negative field of the PCA and CA first axes (Figs. 7, 8). Their occurrence indicates relatively high density of submerged aquatic macrophytes in Kings Billabong with higher water clarity, before regulation. The cladoceran and diatom species associated mainly with PCA and CA axis 2 comprise small littoral species of cladocerans, Alona guttata, and the diatom species, Aulacoseira granulata var angustissima. These taxa have also been reported as preferring turbid and nutrient-rich water bodies in other lakes (Hofmann, 1996; Gell et al., 2005). Similar assemblages of flora and fauna were reported by Blindow et al. (2000) in European shallow lakes, so in the ordination, PCA and CA axis 2 represent a transition assemblage that tend to prefer relative eutrophic, and turbid water conditions. The assemblage that dominates the upper sediment layers is also associated with PCA and CA axis 1 but opposed to the basal assemblages (Figs. 7, 8). The planktonic Bosmina meridionalis, which dominates the cladoceran assemblage, has been reported to occur in a wide range of habitats in Australia (Shiel & Dickson, 1995). The small littoral chydorid, Alona sp. is also an inhabitant of relatively eutrophic waters, and it is often commonly found together with Eubosmina sp. in European lakes (Hofmann, 1996). Bosmina, Chydorus sp. and small Alona in Northern Hemispheric lakes have been found to be within a mixed community, and also to feed opportunistically as a result of changes in the
123
284
wetland environment and available habitat types (DeMott & Kerfoot, 1982). In Kings Billabong, the increase in diatoms, such as Pseudostaurosira brevistriata and Staurosira construens var venter, in association with this mixed cladoceran community, provides evidence of a further change towards phytoplankton dominance, and a turbid water regime (Grundell et al., 2012). Changes in stable isotopes of carbon and nitrogen in bulk sediment organic records The stable isotope signatures in sediment can be continuously altered over time due to shifts in the source of organic matter, both from terrestrial and aquatic sources, as well as the occurrence of postdepositional diagenetic processes (e.g. Meyers & Ishiwatri, 1993). The evidence of changing d13C and d15N signatures in the bulk sediment organic matter in Kings Billabong reveals a sustained change in the trophic status of the ecosystem. Below 75 cm (*1930 AD), all samples had low organic content as revealed by low percentage LOI, d13C values of *-22% and d15N values of *3.0%. These are consistent with other studies nearby, with most organic carbon derived from terrestrial or macrophyte sources and little evidence for N enrichment. For example, Herczeg et al. (2001) reported that, prior to the 1940s; the d13C values in the bulk sediment of Lake Alexandrina (downstream in South Australia) were associated with aquatic macrophytes and terrestrially derived inputs of organic carbon. Our results also suggest low levels of influence of phytoplankton in the system before regulation. In Kings Billabong, a transition was revealed in these chemical indicators from 75 to 60 cm whereupon higher percentage of LOI, lower d13C (*-25%) and higher d15N (3.5%) values were sustained to the surface of the core. It has been reported that phytoplankton generally has a 13Cdepleted signature (e.g. Krull et al., 2009) relative to macrophytes and so this change may be reliable, and is likely to be consistent, with the changes in the diatom flora (Fig. 5). However, Brenner et al. (1999) cautioned the attribution of the response of bulk sedimentderived stable isotope of organic carbon signatures to macrophyte vs planktonic algal-dominated system switch as those organic carbons are also influenced by terrestrial origins. The enrichment of 15N in Kings Billabong likely reflects human impact in the form of
123
Hydrobiologia (2017) 787:269–290
human and animal wastes and the inflow of irrigation return waters rich in fertiliser. For example, in Lake Alexandrina, South Australia, the bulk sediment core samples were found to have 15N-enriched inorganic nitrogen (derived from mineralised animal waste and sewage) as a result of increased agricultural and human activities in the catchment following the 1940s (Herczeg et al., 2001). The 15N enrichment in Kings Billabong, following the 1930s, may also reflect a reduction in the assimilation of nutrients due to the reduced cover of macrophytes followed by phytoplankton growth and other non-planktonic contributions of organic nitrogen to the sedimentary pool. The d15N values in the organic matter of bulk sediment are not a highly reliable indicator of trophic state, as individual lakes may display varying d15N signatures in their sedimentary dissolved inorganic nitrogen (DIN) pools, regardless of trophic state (Gu et al., 1996). However, to some degree, both the d13C and d15N signatures of organic matter in the bulk sediment are useful for qualitative examination of organic carbon energy flow and changes in the trophic state of a lake (Gu et al., 1996). So, on balance, the isotopic evidence supports assemblage evidence for a shift from a macrophyte system to plankton-dominated system, which was triggered by inundation and higher water levels, as well as increased nutrient loads. The timing and nature of ecosystem change in Kings Billabong Kings Billabong was commissioned as an irrigation water storage facility soon after the construction of the nearby river pumps in the early 1890s. This is likely to be the first substantial disturbance to the local hydrology of the site (Rutherfurd, 1990; Lloyd, 2012). A marked impact of humans on land and water was noted since the ‘‘soldier settler’’ projects rewarded the returning soldiers from World War I with the provision of a block of irrigated land for them to work to minimise potential unemployment in the cities (Haisman, 2004). This program would have increased land clearance and disturbance; however, these efforts to colonise the land were not as successful near Kings Billabong, as they were elsewhere in the region (Lloyd & Rees, 1994). A more significant and widespread transition in floodplain wetland hydrology, however, is likely to have come with the phase of weir building from 1917 to 1936 AD. The development in weir pool
Hydrobiologia (2017) 787:269–290
infrastructure along the river modified the natural hydrological patterns, particularly resulting in the reduction of the frequency, duration and magnitude of all but the largest floods (Thoms et al., 2000). Critically, the water levels of many wetlands above the locks were linked to the weir pools eliminating variability and drying (Thoms et al., 2000). As a result of these changes, it is not surprising that Kings Billabong, and other wetlands, experienced substantial ecosystem change. Given the inherent shortcomings associated with radiometric dating on sites with naturally low activities and variable sediment sources (e.g. Gell et al., 2005), it is challenging to attribute early changes in the sediment record to particular events. It is possible that the changes in biological assemblage towards the base of the core may be linked with the first impounding of the billabong or the increased local land development after World War I. However, the increased abundance of key planktonic species of diatoms observed in some River Murray wetlands during the same period indicates that the cause is likely to be a regional, rather than a site specific. For example, at several sites, the increase in the abundance of planktonic (e.g. Aulacoseira sp.) or facultative planktonic taxa is attributed to river regulation (Gell et al., 2007; Grundell et al., 2012). In Kings Billabong, A. subborealis increases between 70 and 75 cm are coincident with the inferred point of weir development after the 1930s. Unusually, it is also abundant before this point, at *90 cm. As it is a river plankter, it is possible that this is attributable to the direct input of river water into Kings Billabong from the 1890s—an historic management action unique to this site. The shift in the biological records between 70 and 75 cm may be principally related to the widespread effects of river regulation associated with the commissioning of a network of weirs between 1922 and 1936, including Lock 11 erected in Mildura in 1927 (Haisman, 2004). The record of cladoceran and diatom assemblages, and the d13C and d15N values in bulk sediments, in Kings Billabong indicates that the density of submerged aquatic plants may have begun to decline rapidly soon after river regulation. Gradual changes in water clarity, and increasing turbidity and eutrophication, may have played a role in the transition in ecological state as shown by increasing phytoplankton development in association with decreasing light penetration. Due to a higher water level ([2 m),
285
submerged macrophyte populations began to decline owing to a reduced light regime. Such a change is consistent with regime shift models (e.g. Scheffer, 1997), which also invoke a shift in feedback effects such as the destabilisation of sediments with reduced densities of submerged vegetation followed by alteration in the recycling of nutrients to phytoplankton. In the Lower River Murray system all wetlands, including those permanently inundated, would respond to nutrient loads and macrophyte dynamics in a similar fashion (Gell & Reid, 2014). Further, flow regulation, permanent inundation and the lack of wetting and drying, would put phytoplankton at an advantage over other autotrophs (Blanch et al., 1999, 2000; Gell et al., 2002). As a result, under this model, once the feedbacks associated with the new state were established, conditions that limit underwater light penetration and maintain elevated nutrient concentrations (e.g. Blindow et al., 1993) became entrenched. Reid et al. (2007) demonstrated the vulnerability of permanently wet and deeper (3–5 m depth) billabongs to the loss of macrophytes through increased turbidity and diminished light regime so the maintenance of constant water levels with turbid conditions may have favoured planktonic forms over littoral forms across a wide range of sites. Further, more shallow sites, and those going through wet and dry cycles, are shown to be more resistant to these impacts suggesting light regime is a principal driver of change in vulnerable, ‘transitioning’ wetlands (Gell & Reid, 2016). Limitations in the radiometric dating down the core preclude a clear date for the initiation of this change; however, it appears likely that the state switch occurred soon after the onset of river regulation. The pressures of permanent inundation by sediment and nutrient-rich waters may have increased the turbidity and reduced the resilience of this wetland, and prevented it returning to a regime of high water clarity and biological diversity (e.g. Reid et al., 2011). However, the evidence of changes in primary producers and first-order zooplankton consumers in Kings Billabong suggests that the transition may not have been abrupt, but gradual. Sometimes, it is difficult to predict the nature of ecological transitions (abrupt vs. gradual) in highly dynamic shallow wetland systems that are characterised by species preferring different habitat structures (Scheffer & Jeppesen, 2007). The time shallow lakes may take to transition following the crossing of
123
286
a threshold may be short and unnoticed (Scheffer et al., 2001). However, gradual transitions have also been witnessed under regime shift models (Walker & Salt, 2006). In such a slow transition, the regime shift from the one state to the other is protracted, and the succession within the state is also slow, but the dynamic behaviour continues to stay close to the equilibrium due to stochastic variation or the influence of more complex attractors around the threshold (Hughes et al., 2013). While the upstream records show abrupt changes during the earlier part of the European settlement (e.g. Reid, 2008), those from Sinclair Flat (Grundell et al., 2012) and also at Kings Billabong to some extent, albeit briefly, during the 1930s regulation phase, occur slowly over an extended period. This rate of regime change tempts the question as to whether these changes are attributable to internal trophic dynamics, which might be expected to adjust or reduce the stressors around the point of equilibrium relatively quickly, or to pressures from the externally derived forces of a larger system. The decline of benthic cladocerans, Dunhevedia crassa, Biapertura affinis and diatoms, Epithemia adnata is reported as a typical occurrence in many shallow wetlands across the Murray River in recent time (Davids et al., 1987; Gell et al., 2007; Grundell et al., 2012). Notwithstanding the high values of Aulacoseira subborealis early in this record, the transition in diatom assemblages in general at Kings Billabong is remarkably similar to those documented nearby at Lake Cullulleraine (Fluin et al., 2010) and Sinclairs Flat (Grundell et al., 2012) located about 80, and over 300 river kilometres, respectively, downstream. In both Sinclair’s Flat and Kings Billabong, benthic taxa (Epithemia sp.; Eunotia serpentina) yield to river plankton (Aulacoseira sp.; especially A. subborealis) and eventually to tychoplanktonic forms (Pseudostaurosira brevistriata, Staurosira sp., Staurosirella sp.). In particular, Fluin et al. (2010) show high post-regulation values for A. subborealis across four sites across the region. The coincident shift in diatom assemblages across widespread wetlands in the Lower Murray River system indicates a regional scale change in the state of the source waters, and the associated wetland ecosystems. The diatom assemblages are known to receive river plankton upon connectivity and to revert to a typical benthic flora when isolated (Gell et al., 2002). Also higher
123
Hydrobiologia (2017) 787:269–290
proportions of plankton and fine sediments are recorded near inlets to wetlands further suggesting the river as a source of these floras (Grundell et al., 2012). So, while the changes are consistent with regime shift models, it remains possible that the assemblage shifts here are responding to ongoing regional scale pressures. Capon et al. (2015) reviewed evidence for regime shifts and concluded that few studies demonstrated the feedbacks responsible for non-linear change, or a lack of recovery after the removal of a stressor. The scale of the changes within the Murray River basin makes it unlikely that the pressures which brought about such widespread changes will be removed. Documenting the existence of, and alteration in, the internal feedback mechanisms that may have driven these changes requires more sophisticated, molecular level assessments, for example, stable isotopes and DNA, of the sedimentary record over time.
Conclusions This study provides multi-proxy evidence for the response of Kings Billabong, one of many shallow floodplain wetlands of the River Murray system in southeast Australia, to anthropogenic disturbance. Both subfossil cladoceran and diatom assemblages, along with the bulk sediment organic carbon and nitrogen stable isotopic (d13C and d15N) values in the sediment core spanning the past c. 100 years, show a gradual change in the wetland ecosystem. The multi-proxy results suggest that submerged aquatic macrophytes have been replaced by a phytoplankton-dominated system. The transition occurred following the arrival of European settlers and is most likely attributable to hydrological changes associated with river regulation. Changing land use may have exacerbated the effect of hydrological change through increased flux of nutrients and fine sediments. In concert, these changes have driven the loss of important species of littoral cladocerans and diatoms, perhaps together with a decline in terrestrial and macrophyte-derived carbon inputs, in Kings Billabong in recent times. Kings Billabong now supports a depleted bulk sediment carbon and less productive system with reduced macrophyte density than it did previously as shown by low d13C values.
Hydrobiologia (2017) 787:269–290
This multi-proxy paleolimnological study has identified the timing and nature of the changes in Kings Billabong during the past century. While these changes may not be abrupt, it remains that they may reflect a regime shift whereby feedbacks that reinforced a macrophyte-dominated system weakened on account of human disturbance and were replaced by those that entrench phytoplankton dominance, such as reduced light regime and the release of nutrients and sediments from the benthos. While paleolimnological approaches remain the principal means by which long-term changes can be explored, establishing the existence of regime shifts is challenging, particularly in large systems that have been substantially impacted by human development. The advancement of palaeoecological approaches in the future can fill in knowledge gaps of baseline conditions and early warning scenarios of regime shifts. This will be especially helpful in wetland ecosystems which are under increased human influences in southeast Australia, and it can inform managers of regional water resources of the true condition of their wetlands. Acknowledgments The AINSE Grant # AINSEGRA11087 to GK supported this research. The laboratory assistance was provided by the Australian Nuclear Science and Technology Organisation (ANSTO), and the field work was supported by the Collaborative Research Network (CRN), and the School of Applied and Biomedical Sciences, Federation University Australia. Rob Milne from the FedUni Centre for eResearch and Digital Innovation assisted to design Fig. 1 using GIS. Iain Ellis from the MDFRC provided further information about the fish species present in Kings Billabong. A part of this research was also written whilst GK was working as a CAS-PIFI Professorial Fellow at the Nanjing Institute of Geography and Limnology Chinese Academy of Sciences (NIGLAS).
References Appleby, P. G., 2001. Chronostratigraphic techniques in recent sediments. In Last, W. M. & J. P. Smol (eds), Tracking Environmental Change Using Lake Sediments, Volume 1: Basin Analysis, Coring and Chronological Techniques. Kluwer Academic Publishers, Dordrecht: 171–203. Appleby, P. G. & F. Oldfield, 1978. The calculation of lead-210 dates assuming a constant rate of supply of unsupported 210-Pb to the sediment. Catena 5: 1–8. Bate, N. & P. Newall, 1998. Techniques in the use of diatoms in water quality assessment: how many valves? In John, J. (ed.), Proceedings of the 15th International Diatom Symposium. Curtin University of Technology, Perth: 153–160. Battarbee, R. W., 1986. Diatoms analysis. In Berglund, B. E. (ed.), Handbook of Paleoecology and Paleohydrology. Wiley, London: 527–570.
287 Battarbee, R. W., V. J. Jones, R. J. Flower, N. G. Cameron, H. Bennion, L. Carvalho & S. Juggins, 2001. Diatoms. In Smol, J. P., H. J. B. Birks & W. M. Last (eds), Tracking Environmental Change Using Lake Sediments. Volume 3: Terrestrial, Algal, and Siliceous Indicators, Vol. 3. Kluwer Academic Publishers, Drodrecht: 155–202. Bennion, H., R. W. Battarbee, C. D. Sayer, G. L. Simpson & T. A. Davidson, 2011. Defining reference conditions and restoration targets for lake ecosystems using palaeolimnology: a synthesis. Journal of Paleolimnology 45: 533–544. Blanch, S. J., G. G. Ganf & K. W. Walker, 1999. Tolerance of riverine plants to flooding and exposure indicated by water regime. Regulated Rivers: Research & Management 15: 43–62. Blanch, S. J., K. W. Walker & G. G. Ganf, 2000. Water regimes and littoral plants in four weir pools of the River Murray, Australia. Regulated Rivers: Research & Management 16: 445–456. Blindow, I., G. Andersson, A. Hargeby & S. Johansson, 1993. Long-term pattern of alternative stable states in 2 shallow eutrophic lakes. Freshwater Biology 30: 159–167. Blindow, I., A. Hargeby, B. M. A. Wagner & G. Anderson, 2000. How important is the crustacean plankton for the maintenance of water clarity in shallow lakes with abundant submerged vegetation? Freshwater Biology 44: 185–197. Bohlke, J. K. & T. B. Coplen, 1995. Interlaboratory comparison of reference materials for nitrogen isotope ratio measurements, taken from an IAEA Technical report, Reference and intercomparison materials for stable isotopes of light elements, IAEA-TECHDOC-825. Brenner, M., T. J. Whitmore, J. H. Curtis, D. A. Hodell & C. L. Schelske, 1999. Stable isotope (d13C and d15N) signatures of sedimented organic matter as indicators of historic lake trophic state. Journal of Paleolimnology 22: 205–221. Brugam, R. B., 1978. Pollen indicators of land use change in Southern Connecticut. Quaternary Research 9: 349–362. Bunn, S. E. & A. H. Arthington, 2000. Basic principles and ecological consequences of altered flow regimes for aquatic biodiversity. Environmental Management 30: 492–507. Capon, S. J., A. J. J. Lynch, N. Bond, B. Chessman, J. Davis, N. Davidson, M. Finlayson, P. Gell, D. Hohnberg, C. Humphrey, R. Kingsford, D. Nielsen, K. Ward, J. Thomson & R. MacNally, 2015. Regime shifts, thresholds and multiple stable states in freshwater ecosystems; a critical appraisal of the evidence. Science of the Total Environment 534: 122–130. Carpenter, S. R. & W. A. Brock, 2006. Rising variance: a leading indicator of ecological transition. Ecology Letters 9: 311–318. Carpenter, S. R. & R. C. Lathrop, 2008. Probabilistic estimate of a threshold for eutrophication. Ecosystems 11: 601–613. Carpenter, S. R., D. Ludwig & W. A. Brock, 1999. Management of eutrophication for lakes subject to potentially irreversible change. Ecological Applications 9: 751–777. Davids, C., M. Stolp & C. J. De Groot, 1987. The cladocerans of the littoral zone of Lake Maarsseveen I. Hydrobiological Bulletin 21: 71–79.
123
288 Davis, J., L. Sim & J. Chambers, 2010. Multiple stressors and regime shifts in shallow aquatic ecosystems in antipodean landscapes. Freshwater Biology 55(Suppl. 1): 5–18. DeMott, W. R. & W. C. Kerfoot, 1982. Competition among cladocerans: nature of the interaction between Bosmina and Daphnia. Ecology 63: 1949–1966. Fernandes, M. & E. Krull, 2008. How does acid treatment to remove carbonates affect the isotopic and elemental composition of soils and sediments? Environmental Chemistry 5: 33–39. Fluin, J., J. Tibby & P. Gell, 2010. Testing the efficacy of electrical conductivity (EC) reconstructions from the lower Murray River (SE Australia): a comparison between measured and inferred EC. Journal of Paleolimnology 43: 309–322. Gauch Jr., H. G., 1982. Multivariate Analysis and Community Structure. Cambridge University Press, Cambridge. Gell, P. A., 2012. Palaeoecology as a means of auditing wetland condition. In Haberle, S. G. & B. David (eds), Peopled Landscapes: Archaeological and Biogeographic Approaches to Landscapes. Terra Australis, Vol. 34. Australian National Uuniversity Press, Canberra: 445–457. Gell, P., S. Bulpin, P. Wallbrink, S. Bickford & G. Hancock, 2005. Tareena Billabong – A paleolimnological history of an everchanging wetland, Chowilla Floodplain, lower Murray-Darling Basin. Marine and Freshwater Research 56: 441–456. Gell, P., J. Fluin, J. Tibby, G. Hancock, J. Harrison, A. Zawadzki, D. Haynes, S. Khanum, F. Little & B. Walsh, 2009. Anthropogenic acceleration of sediment accretion in lowland floodplain wetlands, Murray–Darling Basin, Australia. Geomorphology 108: 122–126. Gell, P. & M. Reid, 2014. Assessing change in floodplain wetland condition in the Murray Darling Basin. The Anthropocene 8: 39–45. Gell, P. & M. Reid, 2016. Muddied waters: the case for mitigating sediment and nutrient flux to optimise restoration response. Frontiers in Ecology and Evolution. doi:10.3389/ fevo.2016.00016. Gell, P. A., I. R. Sluiter & J. Fluin, 2002. Seasonal and interannual variation in diatom assemblages in Murray River connected wetlands in northwest Victoria, Australia. Marine and Freshwater Research 53: 981–992. Gell, P., J. Tibby, F. Little, D. Baldwin & G. Hancock, 2007. The impact of regulation and salinisation on floodplain lakes: the lower River Murray, Australia. Hydrobiologia 591: 135–146. Gippel, C. J. & D. Blackham, 2002. Review of environmental impacts of flow regulation and other water resource developments in the River Murray and Lower Darling River System: includes glossary of terms: final report to Murray Darling Basin Commission. Murray Darling Basin Commission. Grimm, E. C., 1987. CONISS: a FORTRAN 77 program for stratigraphically constrained cluster analysis by the method of incremental sum of squares. Computers Geosciences 13: 13–35. Grundell, R., P. Gell, A. Zawadzki & K. Mills, 2012. Interaction between a river and its wetland: evidence from spatial variability in diatom and radioisotope records. Journal of Paleolimnology 47: 205–219.
123
Hydrobiologia (2017) 787:269–290 Gu, B., C. L. Schelske & M. Brenner, 1996. Relationships between sediment and plankton isotope ratios (d13C and d15N) and primary productivity in Florida lakes. Canadian Journal of Fisheries and Aquatic Sciences 53: 875–883. Haisman, B., 2004. Murray–Darling River Basin Case Study Australia. Background Paper. The World Bank, Washington, DC: 81. Harrison, J., H. Heijnis & G. Caprarelli, 2003. Historical pollution variability from abandoned mine sites, Greater Blue Mountains World Heritage Area, New South Wales, Australia. Environmental Geology 43: 680–687. He, Q. & D. E. Walling, 1996. Use of fallout Pb-210 measurements to investigate longer-term rates and patterns of overbank sediment deposition on the floodplains of lowland rivers. Earth Surface Processes and Landforms 21: 141–154. Herczeg, A. L., A. K. Smith & J. C. Dighton, 2001. A 120 year record of changes in nitrogen and carbon cycling in Lake Alexandrina, South Australia: C:N, d15N and d13C in sediments. Applied Geochemistry 16: 73–84. Hill, M. O., 1973. Diversity and evenness: a unifying notation and its consequences. Ecology 54: 427–432. Hill, M. O. & H. G. Gauch Jr., 1980. Detrended correspondence analysis: an improved ordination technique. Vegetatio 42: 47–58. Hofmann, W., 1996. Empirical relationships between cladoceran fauna and trophic state in thirteen northern German lakes: analysis of surficial sediments. Hydrobiologia 318: 195–201. Hughes, T. P., C. Linares, C. V. Dakos, L. A. van de Leemput & E. H. van Nes, 2013. Living dangerously on borrowed time during slow, unrecognized regime shifts. Trends in Ecology & Evolution 28: 149–155. Jeppesen, E., P. Leavitt, L. De Meester & J. P. Jensen, 2001. Functional ecology and palaeolimnology: using cladoceran remains to reconstruct anthropogenic impact. Trends in Ecology and Evolution 16: 191–198. Kattel, G. R., 2012. Can we improve management practice of floodplain lakes using cladoceran zooplankton? River Research and Applications 28: 1113–1120. Kattel, G. R., R. W. Battarbee, A. W. Mackay & H. J. B. Birks, 2008. Recent ecological change in a remote Scottish mountain loch: an evaluation of a Cladocera-based temperature transfer-function. Palaeogeography, Palaeoclimatology, Palaeoecology 259: 51–76. Kattel, G. R., X. Dong & X. Yang, 2016. A century-scale, human-induced ecohydrological evolution of wetlands of two large river basins in Australia (Murray) and China (Yangtze). Hydrology and Earth System Sciences 20: 2151–2168. Korhola, A., 1999. Distribution patterns of Cladocera in subarctic Fennoscandian lakes and their potential in environmental reconstruction. Ecography 22: 357–373. Krammer, K. & H. Lange-Bertalot, 1991a. Susswasserflora von Mitteleuropa. Bacillariophyceae Teil iii: Centrales, Fragilariaceae, Eunotiaceae. Gustav Fischer Verlag, Stuttgart: 596. Krammer, K. & H. Lange-Bertalot, 1991b. Susswasserflora von Mitteleuropa. Bacillariophyceae Teil iv: Achnanthaceae. Gustav Fischer Verlag, Stuttgart: 437.
Hydrobiologia (2017) 787:269–290 Krammer, K. & H. Lange-Bertalot, 1986. Susswasserflora von Mitteleuropa. Bacillariophyceae, Teil i: Naviculaceae. Gustav Fischer Verlag, Stuttgart: 876. Krammer, K. & H. Lange-Bertalot, 1988. Susswasserflora von Mitteleuropa. Bacillariophyceae Teil ii: Bacillariaceae, Epithemiaceae, Surirellaceae. Gustav Fischer Verlag, Stuttgart: 576. Krull, E., D. Haynes, S. Lamontagne, P. Gell, D. McKirdy, G. Hancock, J. McGowan & R. Smernik, 2009. Changes in the chemistry of sedimentary organic matter within the Coorong over space and time. Biogeochemistry 92: 9–25. Le Clercq, M., J. van der Plicht & M. Gr} oning, 1998. New 14C reference materials with activities of 15 and 50 pM. In: Mook WG, van der Plicht J (eds) Proceedings of the 16th International 14C Conference, Radiocarbon, 40: 295–297. Leps, J. & P. Sˇmilauer, 2003. Multivariate Analysis using CANOCO. Cambridge University Press, Cambridge. Lloyd, L. N., 2012. Kings Billabong Operating Plan. Report to the Mallee CMA. Lloyd Environmental, Syndal, Victoria. Final Draft 22 March 2012. Lloyd, C. & J. Rees, 1994. The Last Shilling: A History of Repatriation in Australia. Melbourne University Press, Carlton. Mackay, N. & D. Eastburn, 1990. The Murray. Murray-Darling Basin Commission, Canberra. MCMA, 2006. Mallee River Health Strategy. Mallee Catchment Management Authority, Mildura. MDBA, 2013. MDBA Basin Plan. Murray Darling Basin Authority, Canberra. Meyers, P. A. & R. Ishiwatri, 1993. Lacustrine organic geochemistry-an overview of indicators of organic matter sources and diagenesis in lake sediments. Organic Geochemistry 20: 867–900. Meyers, P. A. & J. L. Teranes, 2001. Sediment organic matter. In Last, W. M. & J. P. Smol (eds), Tracking Environmental Change Using Lake Sediments, Vol. 2., Physical and Geochemical Methods Kluwer Academic Publishers, Dordrecht: 139–269. Mills, K., P. Gell, P. Hesse, R. Jones, P. Kershaw, R. Drysdale & J. McDonald, 2013a. Paleoclimate studies and natural-resource management in the Murray–Darling Basin I: past, present and future climates. Australian Journal of Earth Sciences 60: 547–560. Mills, K., P. Gell, J. Gergis, P. Baker, M. Finlayson, P. Hesse, R. Jones, P. Kershaw, S. Pearson, P. Treble, C. Barr, M. Brookhouse, R. Drysdale, J. McDonald, S. Haberle, M. Reid, M. Thoms & J. Tibby, 2013b. Paleoclimate studies and natural-resource management in the Murray-Darling Basin II. Unravelling human impacts and climate variability. Australian Journal of Earth Sciences 60: 561–571. Murphy, B. F. & B. Timbal, 2008. A review of recent climate variability and climate change in southeastern Australia. International Journal of Climatology 28: 859–879. Norris, R. H., F. Dyer, P. Hairsine, M. Kennard, S. Linke, L. Merrin, A. Read, W. Robinson, C. Ryan, S. Wilkinson & D. Williams, 2007. Australian Water Resources 2005. A Baseline Assessment of Water Resources for the National Water Initiative. Level 2 Assessment. River and Health Theme, Canberra. Ogden, R. W., 2000. Modern and historical variation in aquatic macrophyte cover of billabongs associated with catchment
289 development. Regulated Rivers: Research & Management 16: 497–512. ¨ zen, A., B. Karapmar, I. Kucuk, E. Jeppesen & Meryem O Beklioglu, 2010. Drought-induced changes in nutrient concentrations and retention in two shallow Mediterranean lakes subjected to different degrees of management. Hydrobiologia 646: 61–72. Parks Victoria, 2008. The Management Plan for Kings Billabong Wildlife Reserve. Parks Victoria, Melbourne. Randsalu-Wendrup, L., D. J. Conley, J. Carstensen & S. C. Fritz, 2016. Paleolimnological records of regime shifts in lakes in response to climate change and anthropogenic activities. Journal of Paleolimnology. doi:10.1007/s10933-016-9884-4. Rees, P. M., A. Ziegler & P. J. Valdes, 2000. Jurassic phytogeography and climates: new data and model comparisons. In Huber, B. T., K. G. Macleod & S. L. Wing (eds), Warm Climates in Earth History. Cambridge University Press, Cambridge: 297–318. Reid, M. A., 2008. Evidence for catastrophic shifts in the trophic structure of floodplain lakes associated with soil erosion. In Schmidt, J., T. Cochrane, C. Phillip, S. Elliot, T. Davies & L. Basher (eds), Sediment Dynamics in Changing Environments. IAHS Press, Wallingford: 584–590. Reid, M. A., R. Ogden & M. C. Thoms, 2011. The influence of flood frequency, geomorphic setting and grazing on plant communities and plant biomass on a large dryland floodplain. Journal of Arid Environments 75: 815–826. Reid, M. A., C. D. Sayer, P. Kershaw & H. Heijnis, 2007. Paleolimnological evidence for submerged plant loss in a floodplain lake associated with accelerated catchment soil erosion (Murray River, Australia). Journal of Paleolimnology 38: 191–208. Reid, M. A., J. Fluin, R. W. Ogden, J. Tibby & A. P. Kershaw, 2002. Long-term perspectives on human impacts on floodplain-river ecosystems, Murray–Darling Basin, Australia. Verhandlung Internationale Vereinigung Limnologie 28: 710–716. Rutherfurd, I., 1990. Ancient river, young nation. In Mackay, N. & D. Eastburn (eds), The Murray. Murray-Darling Basin Commission, ACT, Canberra: 17–36. Sarmaja-Korjonen, K., 2003. Chydorid ephippia as indicators of environmental change biostratigraphical evidence from two lakes in southern Finland. The Holocene 13: 691–700. Sayer, C. D., A. Burges, K. Kari, T. A. Davidson, S. Peglar, H. Yang & N. Rose, 2010. Long-term dynamics of submerged macrophytes and algae in a small and shallow, eutrophic lake: implications for the stability of macrophyte-dominance. Freshwater Biology 55: 487–499. Scheffer, M., 1997. Ecology of Shallow Lakes, 1st ed. Chapman & Hall, London. Scheffer, M. & S. R. Carpenter, 2003. Catastrophic regime shifts in ecosystems: linking theory to observation. Trends in Ecology & Evolution 18: 648–656. Scheffer, M., S. R. Carpenter, J. A. Foley, C. Folke & B. Walker, 2001. Catastrophic shifts in ecosystems. Nature 413: 591–596. Scheffer, M. & E. Jeppesen, 2007. Regime shifts in shallow lakes. Ecosystems 10: 1–3. Seddon, A. W. R., A. W. Mackay, A. G. Baker, H. J. B. Birks, E. Bremen, C. E. Buck, E. C. Ellis, C. A. Froyd, J. L. Gill, L. Gillson, E. A. Johnson, V. J. Jones, S. Juggins,
123
290 M. Macias-Fauria, K. Mills, J. L. Morris, D. Nogue´sBravo, S. W. Punyasena, T. P. Roland, A. J. Tanentzap, K. J. Willis, M. Aberhan, E. N. van Asperen, W. E. N. Austin, R. W. Battarbee, S. Bhagwat, C. L. Balanger, K. D. Bennett, H. H. Birks, C. B. Ramsey, S. J. Brooks, M. de Bruyn, P. G. Butler, F. M. Chambers, S. J. Clarke, A. L. Davies, J. A. Dearing, T. H. G. Ezard, A. Feurdean, R. J. Flower, P. Gell, S. Hausmann, E. J. Hogan, M. J. Hopkins, E. S. Jeffers, A. A. Korhola, R. Marchant, T. Kiefer, M. Lamentowicz, I. Larocque-Tobler, L. Lo´pez-Merino, L. Hsiang Liow, S. McGowan, J. H. Miller, E. Montoya, O. Morton, S. Nogue´s, C. Onoufriou, L. P. Boush, F. Rodriguez-Sanchez, N. L. Rose, C. D. Sayer, H. E. Shaw, R. Payne, G. Simpson, K. Sohar, N. J. Whitehouse, J. W. Williams & A. Witkowski, 2014. Looking forward through the past. Identification of fifty priority research questions in palaeoecology. Journal of Ecology 102: 256–267. Shiel, R. & J. A. Dickson, 1995. Cladocera Recorded from Australia. The Royal Society of South Australia, Adelaide. Sonneman, J., A. Sincock, J. Fluin, M. Reid, P. Newall, J. Tibby & P. A. Gell, 2000. An Illustrated Guide to Common Stream Diatom Species from Temperate Australia. The Murray–Darling Freshwater Research Centre, Identification Guide No. 33, Albury. Szeroczyn´ska, K. & K. Sarmaja-Korjonen, 2007. Atlas of Subfossil Cladocera from Central and Northern Europe. Friends of the Lower Vistula Society, S´wiecie: 84.
123
Hydrobiologia (2017) 787:269–290 ter Braak, C. J. F. & P. F. M. Verdonschot, 1995. Canonical correspondence analysis and related multivariate methods in aquatic ecology. Aquatic Sciences 57: 255–289. ter Braak, C. J. F. & P. Smilauer, 2002. CANOCO Reference Manual and CanoDraw for Window’s User’s guide: Software for Canonical Community Ordination (Version 4.5). Microcomputer Power (Ithaca, NY, USA), 500 pp. Thoms, M., R. Ogden & M. Reid, 1999. Establishing the condition of lowland floodplain rivers: a palaeo-ecological approach. Freshwater Biology 41: 407–423. Thoms, M. C., P. Suter, J. Roberts, J. Koehn, G. Jones, T. Hillman & A. Close, 2000. Report of the River Murray Scientific Panel on Environmental Flows: River Murray– Dartmouth to Wellington and the Lower Darling River, River Murray Scientific Panel on Environmental Flows. Murray Darling Basin Commission, Canberra. Walker, B. & D. Salt, 2006. Resilience Thinking: Sustaining Ecosystems in a Changing World. Island Press, Washington, DC. Wang, R., J. A. Dearing, P. G. Langdon, E. Zhang, X. Yang, V. Dakos & M. Scheffer, 2012. Flickering gives early warning signals of a critical transition to a eutrophic lake state. Nature 492: 419–422.