Palaeogeography, Palaeoclimatology, Palaeoecology 246 (2007) 471 – 487
www.elsevier.com/locate/palaeo
A multi-trace element coral record of land-use changes in the
Burdekin River catchment, NE Australia
Stephen E. Lewis a,⁎, Graham A. Shields a,1 , Balz S. Kamber b , Janice M. Lough c
b
a
School of Earth Sciences, James Cook University, Townsville QLD 4811, Australia
Department of Earth Sciences, Laurentian University, Ramsey Lake Road, Sudbury, Canada
c
Australian Institute of Marine Science, PMB 3, Townsville M.C, QLD 4810, Australia
Received 3 November 2005; received in revised form 18 September 2006; accepted 20 October 2006
Abstract
Previous studies have provided evidence for greater sediment input into the inner Great Barrier Reef (GBR) since the midnineteenth century, presumably caused by an increase in erosion due to intensive animal grazing after European settlement.
Here we report a new trace element (Ba, Y, Mn) study of a Porites coral from Magnetic Island, northeastern Australia that
grew continuously for 175 yr from 1813–1986. Increasing Ba and Y concentrations in the 1850–1890 coral growth bands are
attributed to increased soil erosion caused by the introduction of livestock, mainly cattle, into the adjacent Burdekin River
catchment, the largest contributor of suspended sediment to the GBR lagoon. In our study Y levels continued to rise with
agricultural intensification; however, Ba concentrations reach a plateau after 1890, indicating that Ba/Ca ratios in near-shore
corals may reflect additional factors than sediment input.
Mn concentrations in the coral follow a more complicated pattern, which is clearly related to changes in catchment land-use
and related erosion. Before the initiation of European settlement and the introduction of sheep into the Burdekin River
catchment in 1854, coral Mn concentrations were uniformly low (b 1 ppm), yielding typical values for coral aragonite.
However, after 1854, coral Mn levels rose dramatically, fluctuating in parallel with livestock numbers but with ever smaller
amplitudes before reaching more normal values again by 1900. The coincidence in timing between the initial rise in coral Mn
and the introduction of grazing animals in the Suttor and Belyando sub-catchments pinpoints the trace element source as the
Burdekin River. A decrease in Mn levels since 1865 against a background of generally increasing land clearing, livestock
numbers and sediment flux points to the existence of a transient, surficial Mn reservoir in the southern Burdekin River
catchment, which is tentatively identified here as the pre-European topsoil on the Peak Range basaltic province. A brief return
to moderate Mn levels between 1945–1960 is attributed to the expansion of the cattle industry into the upper Burdekin, Bowen
and Suttor sub-catchments after World War II.
This geochemical record confirms that trace elements in massive corals can be used to trace sediment input from
neighbouring river catchments. In our study, Y concentrations are closely correlated with predicted increases in sediment flux
due to land-use changes and prolonged droughts and so may be more useful than Ba for this purpose. Our Mn results attest to
the rapidity of the ecosystem imbalance caused by the introduction of grazing animals, in particular, sheep into NE Australia.
⁎ Corresponding author. Current address: Australian Centre for Tropical Freshwater Research, James Cook University, Townsville, QLD 4811,
Australia. Fax: +61 7 4725 1501.
E-mail address: [email protected] (S.E. Lewis).
1
Current address: Geologisch-Paläontologisches Institut and Museum Westfälische-Wilhelms Universität Münster, Corrensstr. 24, 48149 Münster,
Germany.
0031-0182/$ - see front matter © 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.palaeo.2006.10.021
472
S.E. Lewis et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 246 (2007) 471–487
Within 30 yr, sheep had efficiently reworked and entirely removed the Mn-rich topsoil, which had conceivably accumulated
during thousands of years of regular burning by Australia's indigenous population.
© 2006 Elsevier B.V. All rights reserved.
Keywords: Coral geochemistry; Burdekin River; Land-use; Great Barrier Reef; European settlement; Coral; Mn; Ba/Ca; Y
1. Introduction
Massive Porites corals of the Great Barrier Reef
(GBR), NE Australia can grow up to several metres in
diameter and live for hundreds of years. Concentrations
of certain trace elements incorporated within the coral's
aragonite lattice reflect their relative proportions with
respect to calcium in ambient seawater modulated only
by minor temperature and vital effects. These elements
can thus be used for tracking environmental changes
over a 100–102 yr timeframe. Secular changes in coralline Ba /Ca (Lea et al., 1989), Mn/Ca (Shen et al., 1992),
Cd/Ca (Shen et al., 1987) and Pb/Ca (Shen and Boyle,
1987) ratios have all been used to investigate environmental parameters such as upwelling, El Niño Southern
Oscillation events (ENSO) and industrial contamination.
Recently, Ba/Ca ratios in corals from NE Australia
have been used to recognise changes in the influx of
fine sediment particles from rivers into the GBR lagoon
(McCulloch et al., 2003; Sinclair and McCulloch, 2004).
An increase in the amplitude and frequency of Ba/Ca
peaks after 1870 was attributed by McCulloch et al. (2003)
to increased soil erosion due to cattle grazing in the
Burdekin River catchment. Barium can potentially be used
to trace sediment transport because it undergoes desorption from suspended particles on entering the more saline,
estuarine environment; the amount of Ba released into the
flood plume is presumed to be proportional to the amount
of fine, suspended material (clay). Other metals such as
yttrium (Y) and the rare earth elements (REE) exhibit
similar estuarine desorption behaviour (Hoyle et al., 1984)
to Ba/Ca and could be applied in the same manner (e.g.
Sinclair, 2005), although this approach has not yet been
fully tested. In addition, Shen and Sanford (1989) proposed that manganese (Mn) concentrations in corals might
also “passively record historical perturbations to the surface ocean environment”.
The Burdekin River (Figs. 1 and 2) watershed area is
the second largest (130,000 km2) of the Great Barrier
Reef catchments and dominates the terrigenous sediment
supply to the central Great Barrier Reef shelf (Belperio,
1983). The river strongly influences salinity in the waters
of inshore coral reefs northwards of the delta (King et al.,
2001). The inshore fringing reefs surrounding Magnetic
Island, from which our coral record derives, are among
the first reefs to be impinged by the Burdekin freshwater
plume during the summer wet season. This locality is
ideal to study changes in farming practices within the
vast Burdekin catchment, an area equal to that of England. Previous studies have postulated that large-scale
cattle grazing, in tandem with settlement, has resulted in
soil erosion and increased sediment flux to the inshore
Great Barrier Reef (Moss et al., 1992; Prosser et al., 2002;
Brodie et al., 2003; Furnas, 2003; McCulloch et al., 2003).
In this paper we present a 175-year trace element (Ba,
Y and Mn) record from a long-lived Porites coral that
spans the period of agricultural expansion and land
clearing in NE Australia. Although coral trace element
records have commonly been interpreted in terms of
land-use changes in adjacent catchments, it can be difficult to distinguish the relative influence of regionally
significant river catchments, such as the Burdekin River,
from that of local river systems. Another complicating
factor involves climatic variations, which exert a considerable influence on runoff and erosion and may mask
the effects of land-use changes. In this study, we attempted to address these issues by compiling a comprehensive history of cattle and sheep statistics as well as other
major land-use changes to more specifically isolate the
impact, if any, of particular settlement events on the
regional, near-coastal seawater chemistry.
2. European settlement and land-use in Queensland
The Burdekin and neighbouring river catchments,
bar their coastal fringe, remained unexplored by Europeans until 1844 (Fig. 1). Only three explorers (Leichhardt, Mitchell and Kennedy) have left detailed
accounts (Leichhardt, 1847; Mitchell, 1848; Beale,
1970) of their expeditions into the land as it presented
itself prior to the arrival of European grazing animals.
While Kennedy's 1848 exploration of the tropical and
mountainous Cape York (Fig. 1) met with great logistical
difficulties (Beale, 1970), Leichhardt's epic 4500 km
journey across the state (Fig. 1) was facilitated by easier
topography but, more importantly, by a terrain largely
free of undergrowth. The frequency with which Leichhardt's journal comments on the ‘openly timbered’,
‘open well-grassed’ forest land (Leichhardt, 1847) is
remarkable and in stark contrast to the scrubby, dense
S.E. Lewis et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 246 (2007) 471–487
473
Fig. 1. Map of Queensland (location within Australia: Inset A). A massive Porites coral was cored in Geoffrey Bay, Magnetic Island (Inset B),
approximately 5 km off the coast near Townsville. The major coastal river catchments east of the Great Dividing Range are outlined, including the
Burdekin (shaded). Detailed descriptions of the pre-European settlement landscapes were documented in the expeditions of Leichhardt (blue line,
1844), Mitchell (green line, 1845) and Kennedy (red line, 1847; 1848). The first sheep run in the Burdekin catchment occurred in 1854 west of the
present day Clermont Township (shown on map). British demand for wool during the American Civil War in the 1860's prompted rapid settlement in
both the Burdekin and Fitzroy River catchments. Suitable sheep and cattle pastures in the Burdekin catchment were sub-divided in 1861. Port Curtis
(1854; Gladstone), Port Denison (1861; Bowen) and Townsville (1865) rapidly emerged to supply the sheep industry and to export wool produced
from these areas. Sheep grazing in the coastal area was an environmental disaster and the major sustainable wool districts are now to the west of the
Great Dividing Range. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
underwood vegetation encountered today (Bowman,
1998). The rapid transformation from an openly treed,
grassy landscape to denser, woodier undergrowth since
European settlement can also be reconstructed from accounts of early sea-based explorers (Flannery, 1994) and
is commonly attributed to the cessation of the Aboriginal
inhabitant's back-burning practice: ‘fire-stick farming’
(Jones, 1969; Fensham et al., 2003). As early as 1848,
the astute observer Thomas Mitchell (Mitchell, 1848)
concluded that fire, grass, kangaroos and human inhabitants formed an inter-dependent existence.
The initial sheep-grazing efforts of the early 1840's in
Queensland focused on the state's extreme SE (Figs. 1
and 2) in the Darling Downs and Logan areas (Knight,
474
S.E. Lewis et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 246 (2007) 471–487
Fig. 2. The Burdekin River catchment and immediate surrounds. The Burdekin River is divided into six sub-catchments: the Belyando, Bowen,
Suttor, Cape, upper Burdekin, and lower Burdekin. The outcropping mafic volcanic rocks within the Burdekin River catchment are shown. These
rocks are typically enriched in Mn and provide a potential source for the elevated Mn concentrations in the Magnetic Island coral. The rest of the
lithology is a mixture of sedimentary, metamorphic and granitoid rocks. The police districts in the vicinity of the Burdekin River catchment from
where cattle and sheep numbers were recorded are highlighted. The statistics from the Rockhampton (until 1863), Emerald, Springsure, Alpha,
Clermont and Belyando districts were included in the Belyando sub-catchment. Statistics for the Suttor and Bowen sub-catchments were combined
due to the sparse districts in this area; the statistics from the Peak Downs, Fort Cooper, Mackay and Bowen districts were used for this location. The
Charters Towers and Cape River district was placed in the Cape River sub-catchment. The Ravenswood, Townsville, Cardwell and Kennedy districts
were incorporated into the upper Burdekin, while the Ayr district was included in the lower Burdekin sub-catchment. Note the Aramac, Barcaldine
and Blackall districts which are all located to the west of the Great Dividing Range. These districts are among the most successful sheep grazing areas
in Queensland. Note: The location of the Belyando police district was taken from the Belyando Crossing. No other reference to the town/district of
Belyando could be found.
1895; Fox, 1919–1923; Campbell, 1936). From these
humble beginnings, squatters rapidly moved northwards
with flocks of sheep sourced from the already established
colonies in the south. New ports were opened along the
coast in 1854 at Port Curtis (Gladstone), in 1861 at Port
Denison (Bowen) and in 1865 at Townsville (Fig. 1) to
facilitate export of wool and supply the rapidly growing
grazing industry with new stock as well as supporting
gold prospectors. The great speed with which new
districts were settled between 1854 and 1865 reflects the
reliance of the newly formed colony on wool exports and
the English textile industry's demand for wool, especially after 1861 when the vital North American raw
cotton supply was halted as a result of the Yankee naval
S.E. Lewis et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 246 (2007) 471–487
blockade during the American Civil War (Fox, 1919–
1923; Menghetti, 1992; Statistics of the Colony/State of
Queensland, 1860–1974).
By 1859, Queensland's official sheep population had
reached 3.2 million and more than doubled to 7.3 million
by 1866 (Statistics of the Colony/State of Queensland,
1860–1974). However, by the end of the American
Civil War in 1865, the wool industry along the coast had
already begun to collapse due to a drop in demand,
various sheep afflictions, drought and, most importantly,
the disappearance of suitable feeding grasses, accompanied by the proliferation of unwelcome coarse
vegetation (Menghetti, 1992). Spear grass (Stipa spp.),
in particular, was notorious for reducing the quality of
sheep fleeces and the health of the animals (Bolton,
1963; Menghetti, 1992). Sheep had been a short-lived,
but disastrous experiment and cattle were to become
475
increasingly important to coastal catchments (Fig. 4)
and the Queensland economy after 1890 (Thorpe,
1996). Other major agricultural industries in the region
have included sugarcane and tobacco (Kerr, 1994);
however, cattle grazing occupies by far the most land in
the large Burdekin River catchment (Furnas, 2003).
3. Land-use changes in the Burdekin catchment
Sheep and cattle numbers reported from police districts within or adjacent to the Burdekin River catchment
were compiled here from the annual statistical records
for the Colony/State of Queensland (where available:
1860–1974). The Burdekin River catchment was
subdivided into five sub-catchments that relate to its
major tributaries, i.e. the Belyando, Bowen and Suttor,
Cape, upper Burdekin, and lower Burdekin sub-
Fig. 3. Sheep numbers for the Burdekin River sub-catchments (A). The statistics indicate that the sheep industry was only prominent in Belyando subcatchment. Major droughts typically coincided with declining sheep numbers, while the increased British demand for wool during the American Civil
War was linked to the initial expansion of the sheep industry in the Burdekin River catchment. Cattle numbers in the Burdekin River sub-catchments
(B). Cattle are the most prominent industry in the Burdekin River catchment and occupy the most land. Cattle numbers increased particularly in the
Bowen, Suttor, Cape, Belyando and upper Burdekin sub-catchments during the 1870's–1880's until significantly declining during the Federation
Drought between 1895 and 1903. The cattle industry significantly grew in the Bowen, Suttor and upper Burdekin sub-catchments after World War II.
476
S.E. Lewis et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 246 (2007) 471–487
Table 1
The coral geochemistry (obtained from ICP-MS) and sheep and cattle data for this study
Calendar years
Mn ppm
Y ppb
Ba/Ca
Ba/Ca⁎
Th ppb
Ca %
Ba ppm
1984–1986
1982–1984
1980–1982
1978–1980
1976–1978
1974–1976
1972–1974
1970–1972
1968–1970
1966–1968
1964–1966
1962–1964
1960–1962
1958–1960
1956–1958
1954–1956
1952–1954
1950–1952
1948–1950
1946–1948
1944–1946
1942–1944
1940–1942
1938–1940
1936–1938
1934–1936
1932–1934
1930–1932
1928–1930
1926–1928
1924–1926
1922–1924
1920–1922
1918–1920
1916–1918
1914–1916
1912–1914
1910–1912
1908–1910
1906–1908
1904–1906
1902–1904
1900–1902
1898–1900
1896–1898
1894–1896
1892–1894
1890–1892
1888–1890
1886–1888
1884–1886
1882–1884
1880–1882
1878–1880
1876–1878
1874–1876
1872–1874
1870–1872
0.52
0.47
0.46
0.47
0.49
0.51
0.47
0.45
0.38
0.44
0.41
0.51
0.62
0.61
0.65
0.55
1.18
1.15
1.45
0.95
0.57
0.54
0.53
0.65
0.56
0.53
0.84
0.67
0.77
0.64
0.45
0.50
0.42
0.45
0.44
0.47
0.47
0.49
0.50
0.41
0.47
0.77
0.66
1.21
1.33
1.53
1.77
1.75
1.68
2.02
3.61
4.58
3.11
2.29
1.16
3.21
2.01
2.87
223.1
241.3
241.0
231.1
238.8
241.6
229.9
206.6
216.2
204.7
197.9
192.9
199.1
238.2
227.0
220.6
225.8
222.0
205.0
243.2
206.4
191.7
239.9
195.5
195.4
182.9
182.7
177.7
217.3
204.0
197.7
174.0
186.1
196.6
195.4
179.7
205.1
202.3
182.9
205.9
183.5
196.0
183.2
214.2
223.4
211.8
194.0
214.3
186.8
181.3
177.8
187.9
188.3
171.0
190.4
186.0
181.2
174.8
6.65
3.72
3.84
3.87
3.38
3.52
3.27
3.00
2.91
2.73
3.05
8.52
2.77
3.65
2.95
3.00
3.04
3.05
2.68
3.42
3.08
4.23
4.23
2.67
4.31
5.75
3.50
4.09
4.59
3.93
4.79
2.84
3.89
5.22
3.73
2.78
3.62
3.44
4.25
3.80
3.43
4.32
3.82
4.16
4.46
5.19
3.36
3.90
3.58
3.85
3.83
4.23
3.32
4.38
3.36
3.23
3.20
3.32
4.18
4.43
5.03
4.51
4.80
5.20
5.26
5.04
4.37
4.60
4.25
4.41
4.40
5.25
5.40
4.91
4.68
4.96
4.58
5.38
4.35
4.11
4.56
4.25
4.12
4.48
4.10
4.74
5.50
4.43
4.24
4.25
4.32
4.71
4.03
4.12
4.54
4.42
4.24
4.49
4.25
4.08
4.23
4.65
4.93
4.53
4.13
4.40
4.09
3.95
4.11
4.53
4.21
4.31
4.80
4.44
3.85
4.39
1.56
1.18
0.80
1.11
1.30
0.69
0.75
0.98
0.72
0.88
0.61
0.97
1.32
3.25
0.91
0.82
0.99
0.98
0.93
1.37
1.05
0.76
1.08
0.81
1.04
0.59
0.69
0.73
0.57
0.99
1.05
0.56
0.72
1.16
0.99
0.77
1.04
0.83
0.79
0.94
1.21
1.16
1.93
0.85
0.88
0.91
0.80
1.26
1.43
1.15
0.84
10.14
1.59
0.91
0.92
1.06
0.88
0.81
39.97
38.55
38.62
37.30
37.36
37.64
37.37
37.67
37.11
37.29
37.27
36.14
37.34
38.78
38.41
35.79
38.51
36.61
37.03
37.45
37.16
35.51
36.61
36.79
36.73
35.92
36.07
35.27
37.26
37.27
36.73
37.00
36.64
36.05
36.60
36.28
36.62
36.18
37.54
35.77
36.34
36.11
37.64
36.78
37.48
36.13
37.28
36.98
37.12
37.13
36.81
37.92
37.20
37.14
37.87
37.95
36.69
36.90
9.10
4.91
5.08
4.94
4.33
4.54
4.19
3.88
3.70
3.48
3.90
10.55
3.55
4.85
3.88
3.68
4.01
3.83
3.39
4.39
3.92
5.15
5.30
3.37
5.43
7.08
4.33
4.94
5.86
5.02
6.02
3.60
4.88
6.45
4.68
3.45
4.55
4.26
5.47
4.65
4.27
5.35
4.92
5.24
5.72
6.42
4.29
4.94
4.55
4.90
4.83
5.49
4.24
5.58
4.36
4.21
4.02
4.20
Sheep (1000's)
Cattle (1000's)
57
69
86
277
314
324
357
485
380
291
259
234
1885
1635
1453
1336
1308
1237
1156
1048
1071
1031
944
594
619
956
983
920
812
834
657
917
805
613
711
639
653
575
596
1101
976
910
761
554
635
805
798
732
726
628
671
630
736
889
1019
854
742
639
784
702
630
500
388
316
766
652
636
1135
1104
886
742
636
895
838
847
855
1495
599
800
1102
1501
1178
867
771
731
823
698
519
409
295
1505
170
S.E. Lewis et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 246 (2007) 471–487
477
Table 1 (continued )
Calendar years
Mn ppm
Y ppb
Ba/Ca
Ba/Ca⁎
Th ppb
Ca %
Ba ppm
Sheep (1000's)
1868–1870
1866–1868
1864–1866
1862–1864
1860–1862
1858–1860
1856–1858
1854–1856
1852–1854
1850–1852
1848–1850
1846–1848
1844–1846
1842–1844
1840–1842
1838–1840
1836–1838
1834–1836
1832–1834
1830–1832
1828–1830
1826–1828
1824–1826
1822–1824
1820–1822
1818–1820
1816–1818
1814–1816
1812–1814
JCp-1
JCp-1 (RSD)
4.23
5.08
9.64
4.20
8.15
8.98
6.46
5.86
0.80
0.79
0.48
0.59
0.60
0.47
0.76
0.47
0.48
0.45
0.53
0.49
0.91
0.52
0.50
0.59
0.51
0.54
0.72
0.67
0.72
0.71
1.48%
161.3
159.7
164.5
163.9
176.1
162.6
145.0
139.4
168.4
141.5
150.2
141.7
143.0
142.5
162.2
149.8
147.9
154.8
154.4
154.4
149.4
155.6
154.8
170.2
137.7
165.4
159.3
168.9
168.4
334.1
1.83%
2.65
2.21
3.13
2.40
2.80
2.74
3.18
2.30
4.27
2.62
1.98
2.22
2.74
2.22
2.33
1.77
2.02
2.64
1.83
1.90
1.80
2.19
2.13
1.94
1.75
1.81
2.53
1.70
1.85
1.93
1.57%
3.81
3.65
3.88
3.86
4.03
3.61
3.61
3.61
3.47
3.41
3.52
3.75
3.50
3.32
3.61
3.69
3.53
3.50
3.41
3.37
3.28
3.35
3.33
3.27
3.79
3.56
3.62
3.81
3.72
0.58
1.18
0.95
1.18
0.84
1.11
0.82
0.75
0.64
0.70
0.90
0.96
2.25
0.95
16.33
0.90
1.41
1.13
1.10
1.47
1.08
0.80
1.11
0.94
0.50
1.39
1.06
1.34
0.83
21.38
3.47%
36.85
37.04
37.45
37.13
36.96
36.40
37.05
36.70
37.58
36.73
36.73
36.28
37.17
36.43
37.40
37.19
35.10
36.87
37.95
36.61
37.11
36.55
36.55
36.27
36.45
37.25
36.09
36.76
36.55
36.56
0.55
3.35
2.81
4.02
3.05
3.55
3.42
4.04
2.90
5.50
3.30
2.49
2.76
3.49
2.78
2.99
2.25
2.43
3.34
2.38
2.38
2.29
2.74
2.66
2.42
2.19
2.31
3.12
2.14
2.32
10.14
1.27%
1541
1783
1296
774
215
304
Cattle (1000's)
149
130
209
83
53
13
⁎McCulloch et al. (2003). Note: Ba/Ca data in 10-6.
catchments (Fig. 2). Police districts were assigned to the
sub-catchments according to geographical proximity
and topography (see Fig. 2), e.g. the Alpha, Belyando,
Springsure, Clermont, Emerald and Rockhampton
(1860–1863) police districts were assigned to the
Belyando sub-catchment while the Bowen and Suttor
River sub-catchment included the districts of Bowen,
Mackay, Fort Cooper and Peak Downs. Statistics from the
Springsure (1864–1893) and Emerald (1945–1973)
districts were used to cover the southern section of the
Belyando sub-catchment only when data from the Alpha
(1895–1939) district were not collated.
Prime grazing land in NE Queensland was first identified by Leichhardt in 1844 and outlined by him in two
letters to Charles and William Archer, two pastoralist
explorer brothers who set out in 1853 in search of
suitable pastures. In 1854 they identified Clermont, part
of which lies within the Burdekin catchment (Belyando
and Suttor sub-catchments), as a suitable location and
returned in 1856–7 with several thousand sheep, by
which time others had already moved in. The first sheep
grazed in the Burdekin River catchment were brought
by Jeremiah Rolfe in 1854 in the vicinity of Mistake
Creek (O'Donnell, 1989), which is within the Belyando
sub-catchment but also borders the Suttor sub-catchment (Fig. 2). Sheep numbers were high in the Belyando
and Suttor sub-catchments from this time and rose dramatically during the 1860's (Fig. 3A). The initial boom
of the sheep industry in the Belyando sub-catchment
peaked in 1872 (approx 1.5 million sheep). Following
this period sheep numbers significantly declined to a
low of ∼430,000 in 1886 before rising again gradually
to a new peak in 1914 (∼1.7 million) and remaining at
relatively consistent numbers until the early 1940's.
This time was when the sheep industry in the Belyando
sub-catchment collapsed (Fig. 3A). With the exception
of the Belyando and Suttor sub-catchments, the sheep
industry was never particularly prominent in the rest of
the Burdekin catchment.
Cattle numbers significantly increased in the Burdekin
catchment after the late 1870's to the 1890's particularly
in the Bowen and Suttor, Belyando, Cape and upper
478
S.E. Lewis et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 246 (2007) 471–487
Burdekin River sub-catchments (Fig. 3B). Cattle numbers
dropped considerably during the late 1890's to early
1900's, but the industry had completely recovered by
1910. In more recent times (post World War II), the cattle
industry became particularly prominent in the upper
Burdekin, Bowen and Suttor sub-catchments (Fig. 3B).
4. Methods
The Porites coral was cored in Geoffrey Bay, Magnetic Island (19.15° S, 146.87° E: inset B, Fig. 1) in
1987 by the Australian Institute of Marine Science (Isdale and Daniel, 1989; Lough et al., 1999). A fossil coral
of mid-Holocene age was also recovered from the adjacent bay (Nelly Bay) during the construction of a marine
harbour in 2001. The outer growth layer of this coral
was U–Th dated providing an age of 5790 yr BP (Lewis,
2006). X-ray prints were used to establish a 130-year
growth record for this fossil specimen.
The coral cores were sliced into 7 mm thick slabs and
cleaned with Milli-Q water and dried before sampling.
Annual density banding identified in X-ray prints and
luminescent lines observed under UV light indicate that
the modern Geoffrey Bay coral began growing in 1813
and contains a continuous record of growth until 1986.
In an effort to remove any possible surface contamination, a 1 mm cleaning cut was performed along the
sampling area (Hendy et al., 2002). In addition, Th concentrations were measured to ensure that there were no
geological detrital components trapped within the coral
(Wyndham et al., 2004). Both corals were sampled at bulk
2-yearly growth increments (biennial resolution) using a
stainless steel drill tip on a moveable stage. The coral
powders were homogenized using an agate mortar and
pestle.
For solution ICP-MS analyses, each sample was prepared as follows: 10 mg of coral aragonite was weighed
into a pre-cleaned 14 mL polystyrene tube and digested
with 10 mL of 2.5% HNO3 (double sub-boiling distilled).
No optically visible residue was found after centrifuging.
From this stock solution, an aliquot was taken into a new
pre-cleaned 14 mL polystyrene tube and diluted by a
factor of 5000 with 2% HNO3 and a nominal internal
standard concentration of 6 ppb. Internal standards used
were In and isotopically enriched 6Li, 61Ni, and 235U (see
Eggins et al., 1997). Samples were analysed by ICP-MS at
the Advanced Centre for Queensland University Isotope
Research Excellence (ACQUIRE) laboratory at the University of Queensland in two consecutive batches (modern coral vs. mid-Holocene coral) of 60-80 samples on a
Thermo Electron X-Series instrument with the Xi interface, preferred over the high-performance interface due to
the high Ca concentrations (Kamber et al., 2003). Residual drift after internal standard normalisation was b 3%
and corrected for by multiple analyses of an external drift
monitor, a mixed coral solution judiciously spiked with
some dolerite solution. A suite of 15 elements were
analysed, of which data for Ca, Mn, Y, Ba and Th are
reported here. Instrument response for Mn, Ba and Y
was calibrated against dilute (5000–20,000×) solutions of
U.S.G.S. dolerite standard W-2. Preferred calibration
values are listed in Kamber et al. (2003). Calcium was
calibrated against G.S.J. coral standard JCp-1 (Okai et al.,
2002; 380,425 ppm), which was run as an unknown for
the other elements. Repeat analyses (n = 6) from repeat
digestions (n = 4) of JCp-1 yielded 1 sigma RSD reproducibilities of: Mn = 1.48%; Y = 1.83%, Ba = 1.27% and
Th = 3.47% (Table 1). The internal precisions (RSD) for
analyses of the Geoffrey Bay coral were at times inferior
for Y and Th, as lower count rates were recorded for these
elements than in JCp-1.
Additional samples from the same Geoffrey Bay
coral slice were taken to confirm that the Mn spikes in
the coral represented lattice-bound Mn. This time the
coral was sampled at homogenised 5-yearly growth increments using the sampling strategies described above.
Each sample was treated with 3 mL of Milli-Q water and
placed in an ultrasonic bath for 15 min then centrifuged
at 3000 RPM for 15 min. 3 mL of 10% hydrogen peroxide (H2O2) was added to each sample, placed in an
ultrasonic bath for 15 min and centrifuged. The samples
were then decanted and rinsed three times using Milli-Q
water. The samples were frozen and then freeze-dried for
at least 24 h or until completely dry. The samples were
analysed by a Varian Vista (in axial view detection mode)
ICP-AES at the Queensland Health Scientific Services
laboratory in Brisbane. The analytical reproducibility for
Mn using this technique was only 44.72%, but both
absolute concentrations and trends were identical within
error in both datasets (Fig. 4A).
5. Results
Atomic Ba/Ca ratios measured in this study remained
around 2 × 10− 6 for growth bands 1814–1870, gently
rising to 4 × 10− 6 (±1) by 1900. The interval 1944–1980
was marked by a fall in Ba/Ca ratios, which were still
significantly elevated compared with pre-1870 levels.
These broad trends are consistent with the laser ablation
ICP-MS data for a coral from Havannah Island (McCulloch et al., 2003), when averaged over the same 2-yearly
intervals as our data (Fig. 4B). There are some notable
offsets between the Ba/Ca records. The Ba/Ca ratios in
our Magnetic Island coral are systematically lower than
S.E. Lewis et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 246 (2007) 471–487
479
Fig. 4. Mn concentrations in the Magnetic Island coral sampled at 2 (red) and 5 (blue) yearly resolutions (A). The H2O2 treated 5-yearly resolution
samples are in excellent agreement with the untreated 2-yearly resolution dataset. This finding suggests that the coral Mn concentration is unaffected
by organics incorporated within the skeleton and is probably substituting for Ca in the aragonite lattice. Coral Ba/Ca ratios from Havannah Island
(green; McCulloch et al., 2003) and Ba/Ca ratios (red) and Y (blue) concentrations from the Magnetic Island coral (B). The coral Ba/Ca record from
Havannah Island (McCulloch et al., 2003), averaged to 2-yearly resolution, yields a significant ( p b 0.00) linear correlation with the Y data (r = 0.86).
The correlations between the Havannah Island coral Ba/Ca record with the Ba/Ca dataset from Magnetic Island (r = 0.47) and cattle numbers in the
Burdekin River catchment (r = 0.49) are lower, but still significant. The coral Y concentrations also yield a significant positive correlation with cattle
numbers in the Burdekin catchment (r = 0.64, p b 0.00, n = 52). (For interpretation of the references to colour in this figure legend, the reader is
referred to the web version of this article.)
McCulloch et al.'s (2003) Havannah Island coral. In
addition, the fall of Ba/Ca ratios in the Magnetic Island
record during the 1940's is not evident in the Havannah
Island coral. Although the laser ablation data, obtained
at much higher spatial resolution, show a more shallow
Ba/Ca increase, both studies reveal a similarly smooth
increase in Ba/Ca from the late 1860's — early 1870's
when data are averaged bi-annually.
Y concentrations in the Magnetic Island coral ranged
between 140 ppb and 250 ppb and displayed a very
similar trend to the Ba/Ca data of McCulloch et al. (2003)
from Havannah Island, even with regard to decadal-scale
fluctuations (Fig. 4B). The correlation between our solution ICP-MS data obtained from Geoffrey Bay and the
corresponding averages of the laser ablation ICP-MS data
from Havannah Island (McCulloch et al., 2003) was
significant (r = 0.86, p b 0.00, n = 87). Coral Y concentrations also rise after 1870 and have continued to increase
throughout the coral record.
Of all the elements analysed, Mn displayed by far the
greatest variation (Fig. 4A) with Mn concentrations in
some parts considerably elevated over those reported in
previous studies of massive coral (Shen et al., 1992;
Abram et al., 2003; Alibert et al., 2003). Manganese
480
S.E. Lewis et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 246 (2007) 471–487
Table 2
Mn concentrations in the mid-Holocene coral
Table 2 (continued )
Years from basal section
Mn ppm
4–6
6–8
8–10
10–12
12–14
14–16
16–18
18–20
20–22
22–24
24–26
26–28
28–30
30–32
32–34
34–36
36–38
38–40
40–42
42–44
44–46
46–48
48–50
50–52
52–54
54–56
56–58
58–60
60–62
62–64
64–66
66–68
68–70
70–72
72–74
74–76
76–78
78–80
80–82
82–84
84–86
86–88
88–90
90–92
92–94
94–96
96–98
98–100
100–102
102–104
104–106
106–108
108–110
110–112
112–114
114–116
116–118
118–120
1.11
1.13
1.06
1.05
0.84
0.97
0.74
1.01
0.84
0.73
1.13
0.81
0.80
0.77
0.89
0.74
1.38
0.68
0.75
0.86
0.84
0.93
0.87
0.95
0.87
0.98
0.91
0.82
0.81
2.14
0.70
0.74
0.97
0.76
0.86
0.64
0.66
0.74
0.65
0.87
0.93
0.78
0.73
0.83
0.69
0.84
0.76
1.04
1.31
0.74
0.98
0.76
0.74
0.66
0.64
0.62
0.58
0.58
Years from basal section
Mn ppm
120–122
122–124
124–126
126–128
128–130
Average
0.68
0.75
0.66
0.96
1.10
0.86 ± 0.23
Data obtained from ICP-OES.
concentrations were constant and low (ca. 0.6 ppm) in
all the growth bands for the years 1813–1854, within
error indistinguishable from the average value recorded
in the mid-Holocene coral (0.86 ± 0.23 ppm; Table 2;
Fig. 6) and identical to Mn levels in the most recent
growth bands. A 10-fold increase in coral Mn concentrations was observed from 1855–1856 (5.86 ppm) with
levels peaking at 9.60 ppm during the 1865–1866 growth
bands. The post 1866 period was marked by a significant
decline in Mn concentrations to 1.16 ppm by 1877–1878,
which was followed by a smaller, but significant rise that
peaked during 1883–1884 (4.58 ppm). The coral Mn
concentrations then fell gradually, returning to baseline
values by 1901–1902 (0.66 ppm). Mn concentrations
remained at baseline levels until 1947–1948 when they
rose to 1.45 ppm in 1949–1950. However, the elevated
coral Mn concentrations quickly returned to baseline
levels by 1955–1956 (∼0.6 ppm) and remained at these
levels for the remainder of the coral record. Mn concentrations showed no significant correlation with the
other trace element parameters measured in this study or
in published studies from the inner GBR (McCulloch
et al., 2003).
6. Discussion
6.1. Ba/Ca ratios
Elevated Ba/Ca ratios in corals from Havannah Island
have previously been linked to increased erosion in the
Burdekin River catchment related to the introduction of
cattle grazing (McCulloch et al., 2003). Havannah Island
is further from the mouth of the Burdekin River than
Magnetic Island and would be less influenced by the
Burdekin River plume under normal circumstances.
However, pre-1870 Havannah Island Ba/Ca ratios are
significantly higher than those of Magnetic Island from
our study, which might indicate that Havannah Island is
additionally influenced by sediment and freshwater input
from the Herbert River, the mouth of which is nearby
(see King et al., 2001) as well as other rivers/creeks in the
local area. Ba/Ca ratios in McCulloch et al. (2003) were
S.E. Lewis et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 246 (2007) 471–487
481
Fig. 5. The coral Y concentrations are significantly correlated with cattle numbers in the Burdekin catchment (A). The rapid increase of cattle numbers
in the Burdekin catchment in the 1870's coincides with a steep rise in coral Y concentrations. Similarly to coral Y concentrations, both Ba/Ca records
respond to the rapid stocking of cattle in the Burdekin catchment (B). The initial rise in coral Mn concentrations coincides with the introduction of
sheep in the Burdekin catchment (C). Because the peak sheep numbers in the Burdekin postdate the peak coral Mn levels, we conclude that there must
have been a transient, labile Mn reservoir in the Burdekin catchment.
482
S.E. Lewis et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 246 (2007) 471–487
shown to relate to regional rainfall with Ba/Ca maxima
correlating significantly with floods in the Burdekin
catchment; however, most of these rainfall events would
also have influenced the Herbert River catchment.
Regardless, although baseline pre-1870 values are not
comparable, both Ba/Ca datasets show a major change
around 1870. In the Havannah Island record, this change
is recorded as an increase in the frequency and amplitude
of Ba/Ca maxima, while in the Magnetic Island record 2year homogenised samples show an abrupt increase in
Ba/Ca ratios from 1870 onwards (Fig. 4B). The Havannah Island and Magnetic Island Ba/Ca records appear to
be offset again from 1944 onward. While the Havannah
Island Ba/Ca values remain high but variable, the Magnetic Island Ba/Ca values shift lower throughout much of
this interval, although both sets of Ba/Ca ratios are significantly elevated with respect to their starting positions
in the early 1800's. Interpretation of Ba/Ca ratios in
corals is complicated by the influence of additional factors other than fine suspended sediment release into the
near-shore seawater. Possible factors include barite trapping following phytoplankton blooms, decaying blooms
of Trichodesmium, physiological perturbations associated with coral spawning and bottom sediment resuspension following dredging (Esslemont et al., 2004; Sinclair,
2005). In this regard, it is important to note that neither
study records the initial rapid stocking of grazing animals in the southern Burdekin River catchment. In addition, neither record is significantly correlated with total
sheep numbers in the Burdekin River catchment.
6.2. Yttrium concentrations
Yttrium also shows promise as an indicator of fine
sediment input into the inner GBR (Sinclair, 2005);
theoretically, it should also be a better indicator than Ba
because of the more conservative behaviour of Y during
estuarine mixing due to the formation of strong aqueous
carbonate complexes in seawater and its insignificant
role in primary productivity, which distinguishes it from
Ba (Luo and Byrne, 2004).
The significant positive correlation (r = 0.86, p b 0.00,
n = 87) between the Geoffrey Bay coral Y concentrations
and the Ba/Ca dataset from Havannah Island (McCulloch et al., 2003) indicates that both elements relate to
regionally important variations in sediment input. While
the records parallel each other (Fig. 4B), Y concentrations in our study show a higher correlation coefficient
with total cattle numbers (r = 0.64; p b 0.00, n = 52; not
shown) in the Burdekin River catchment than both the
Ba/Ca records produced from Havannah (r= 0.49, pb 0.00,
n= 52) and Magnetic Islands ( p N 0.05: insignificant), res-
pectively (Fig. 5A–B). As with the Ba/Ca ratio, coral Y
concentrations do not respond immediately to the initial
settlement in the southern Burdekin catchment and are not
significantly correlated with sheep numbers. The coral Y
record appears to be responding to increased land-use and
sediment runoff in the Burdekin River catchment from
cattle rather than sheep grazing, in which case soil erosion
and sediment runoff have apparently only begun to increase
significantly during the late 1860's and early 1870's as
proposed by McCulloch et al. (2003) and have continued to
rise since then in proportion to fluctuating cattle numbers
(Fig. 5B). Presently, the waters of the upper Burdekin and
Bowen sub-catchments carry significantly more suspended
sediments than the other sub-catchments of the Burdekin
(Bainbridge et al., 2006). The upper Burdekin sub-catchment supplied 1.7 million tonnes of sediment in the
2004/05 wet season, around three times the amount of
sediment issued from the other major sub-catchments
combined (Table 3). Since this sub-catchment contributes
the bulk of sediment exported from the Burdekin River,
the coral record should respond mostly to land use changes in the upper Burdekin. The Bowen sub-catchment
would also significantly contribute to the bulk sediment
concentrations because of the high peak suspended sediment levels (Table 3). In addition, this sub-catchment is
the closest to the mouth of the Burdekin River.
Cattle numbers in the Bowen and Suttor sub-catchments increased mainly during the 1870's–1890's, which
closely coincides with the increasing coral Ba/Ca ratios
and Y concentrations (Fig. 5A–B). Therefore, the coral
Ba/Ca ratios and Y concentrations after 1870 may be
responding specifically to increased land-use in the
Bowen and upper Burdekin sub-catchments. This finding
is important for the management of soil erosion in the
Burdekin River catchment as catchment specific sources
can be identified to prioritise remedial works.
Coral Y concentrations may be a more reliable indicator of catchment land-use changes and related soil
erosion than Ba/Ca ratios. As with the Ba/Ca ratio, coral
Y concentrations fluctuate on a decadal-scale with
Table 3
Burdekin load data and suspended sediment concentrations for the
2004/05 wet season (modified from Bainbridge et al., 2006)
Burdekin
sub-catchment
Catchment load
(million tonnes)
Peak suspended
sediment conc. (mg/L)
Upper Burdekin
Bowen
Suttor
Belyando
Cape
End of catchment
1.7
0.34
0.06⁎
0.06⁎
0.11
2.7
1960
8700
320
300
314
1630
⁎ estimate.
S.E. Lewis et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 246 (2007) 471–487
relatively subdued levels coinciding with periods of
major drought and associated loss of stock in the
Burdekin River catchment (Fig. 5A–B, 6).
6.3. Manganese concentrations
The manganese (Mn) record shows a more complex
evolution and is not correlated with coral Ba/Ca ratio or
Y concentration. The extremely low coral Th concentrations (b1 ppb) suggest that there has been no significant incorporation of geological detritus into the
coral (Wyndham et al., 2004). In addition, the chemically pre-treated coral Mn record displays an identical
trend and concentrations to the untreated biennial resolution analysis conducted at the ACQUIRE laboratory
(Fig. 4A). Therefore we consider this coral trace element
record to be genuine and most likely lattice-bound,
although the presence of MnO/OH coatings cannot be
discounted (D. Sinclair, pers comm., 2006). In any case,
the presence of these coatings does not detract from our
main findings.
Previous investigations have linked elevated coral
Mn concentrations to a wide range of processes including ash fallout from fire (Abram et al., 2003), release
483
from bottom sediments (Shen et al., 1992; Alibert et al.,
2003; Wyndham et al., 2004), volcanic eruptions (Shen
et al., 1991) and upwelling events (Shen et al., 1991).
None of these studies have reported such high Mn
concentrations as those reported here, and none of the
above explanations can fully explain systematic trends
in coral Mn levels on a decadal-scale. The initial coral
Mn increase in 1855–1856 directly follows the opening
of Port Curtis (Gladstone) and the very first sheep run in
the Suttor and Belyando sub-catchments (Fig. 5C) by
Jeremiah Rolfe in the vicinity of Mistake Creek in 1854
(O'Donnell, 1989). Therefore, we turn to land-use changes within the Burdekin catchment as the most likely
explanation for the Mn signature of the Magnetic Island
coral.
The dramatic increase in coral Mn after the 1853–1854
growth bands coincides with the very first occupation of
sheep in the southern Burdekin catchment where land was
acquired in the Peak Downs/Clermont area in 1854 (Fox,
1919–1923; Bode, 1984; O'Donnell, 1989). This was
immediately followed by the rapid expansion of sheep
runs in the Belyando and Suttor sub-catchments with
much of the area being sub-divided for sheep and cattle
farming by 1861 (Cunningham, 1895; Bolton, 1963). The
Fig. 6. Mn and Y concentrations in the coral core from 1813–1986. Yttrium concentration, like Ba is interpreted to be an erosion indicator, steadily
increasing after 1860–70. Manganese defines a different history. Elevated Mn concentrations coincide with major land settlement in the Burdekin
catchment. The initial Mn spike in 1855–1856 is related to the establishment of the first sheep run in the southern Burdekin catchment. The peak Mn
concentration coincides with the end of the American Civil War and the climax of rapid expansion of the sheep industry in the Burdekin catchment.
The second major Mn peak in 1883–1884 coincides with and may thus be related to the expansion of the cattle industry or the beginning of the sugar
cane industry on the lower Burdekin catchment. The return of Mn concentrations to pre-1850 levels coincides with the Federation Drought, which
decimated sheep and cattle number throughout the catchment. An increase in Mn after WWII may be related to the further development of the cattle
industry. Reproducibility of Mn by our method is shown for the (slightly heterogeneous) G.S.J. coral standard JCp-1. Also shown for comparison, are
average and standard deviation of Mn in a mid-Holocene coral recovered from Nelly Bay, Magnetic Island.
484
S.E. Lewis et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 246 (2007) 471–487
rate of land take-up was in part fuelled by the land act of
1860, which required that the area be stocked within the
first year of purchase (Bolton, 1963; Thorpe, 1996). For
coral Mn to respond so swiftly to land-use changes, the
grazing of the landscape by sheep must have immediately
triggered Mn release from that area of the catchment
(Fig. 5C).
Because no undisturbed, regularly burnt area has
remained it is not possible to study the chemistry, mineralogy and structure of pre-1850 soils in coastal Queensland. However, clues may be obtained from the nature of
the underlying lithologies. The Peak Downs/Clermont
area coincides with one of the largest basaltic provinces
in Queensland: the Tertiary-age Peak Range Volcanics.
Because basaltic soils are frequently Mn-rich (1750 ppm:
Faure, 1998) and are commonly 2–3 times more enriched
than granitic and most sedimentary rocks (e.g. Faure,
1998), it seems likely that this huge area of basalt bedrock
supplied the Mn, which made its way to the inner GBR
after European settlement. Other basaltic provinces in the
upper Burdekin provide another potential Mn source, but
the initial coral Mn rise in 1854–1856 coincides exclusively with settlement in the Peak Downs region. The
earliest statistics for the upper Burdekin show that in 1860
only 2 sheep and 121 cattle were grazed in this region
compared to ∼300,000 sheep and ∼13,000 cattle in the
Belyando sub-catchment. In addition, the timing of the
initial Mn increase occurs around a decade prior to
settlement in the local (Townsville/Magnetic Island) area.
Therefore on the basis of these empirical observations, we
argue that the source of the enriched coral Mn concentrations was derived from the southern parts of the Burdekin
catchment.
It is proposed that Mn became further concentrated in
coastal Queensland soils over 40,000 yr because of the
build up of ash from Aboriginal burning (Griffiths,
2002), and that native vegetation became accustomed to
the biogeochemical balance provided by this ash. We
conjecture that the natural tendency of ash to increase
soil pH helped to build a substantial, but transient Mn
reservoir because of the relative immobility of Mn under
alkaline pH (Chirenje and Ma, 2002).
The peak Mn concentration in 1865–1866 coincides
with the end of the American Civil War and the drop in
wool price caused by the lifting of the cotton embargo to
Great Britain, a period that also marks the collapse of the
sheep industry in coastal Queensland (Fig. 5C). Agricultural efforts turned instead to cattle grazing and the
establishment of the sugarcane industry with strong
expansion between 1870 and 1895, at which time the
area experienced its worst drought on record (the “Federation Drought”, 1895–1902), an event that drastically
reduced livestock numbers (Fig. 3) and saw Mn concentrations in the coral returning to the low pre-1850
levels (Fig. 6). The 20th century witnessed another
significant Mn peak coinciding with the period immediately following World War II (WW-II), during which time
agriculture, including the cattle industry, intensified in
part due to land release to war veterans and recent European immigrants, but mainly due to the increased mechanisation of land clearing. The cattle industry particularly
became prominent in the upper Burdekin sub-catchment,
which appears to be the most recently settled land in the
Burdekin. Cattle numbers also rose significantly in
the Suttor and Bowen sub-catchments following World
War II and the increase in coral Mn concentrations may
coincide with Mn release from soils covering large basalt
provinces in these three sub-catchments (Fig. 2). The
period between 1945–1953 also saw the development of
the failed tobacco industry where large tracts of land in the
lower Burdekin sub-catchment were cleared by the government following a major drought during the World
War II years (Kerr, 1994). This extensive clearing in the
lower reaches of the Burdekin is another possible reservoir for Mn; however, the banks of the lower Burdekin are
typically at higher elevations than the surrounding landscape and runoff from this lower catchment is commonly
from local creek systems rather than from the Burdekin
River (J. Brodie personal communication, 2005). Lowest
Mn concentrations are found during the major drought
between 1958–1968 (Young, 2000) and coincide with a
drastic decline of sheep numbers in the Burdekin catchment (Fig. 5C).
The post 1868 drop off in sheep numbers was more
severe in the coastal districts and it appears that the Mn
concentration in the coral responded most closely to the
history of this part of the catchment. Regardless, the
swiftness of the initial rise in Mn after the introduction of
sheep to the region and its short-lived character suggest
the original existence of a large, but transient reservoir of
Mn in the soil cover of the Burdekin catchment, which
rapidly became depleted (Fig. 5C). The timing of the
second and third spikes in coral Mn concentration exactly
corresponds to the growth of cattle herds, which replaced
sheep (Figs. 3B, 6). However, the magnitude of Mn
response declined with each stockage and the observation
that growing herds after the Federation Drought never
produced a comparable response in terms of Mn levels
confirm the transient nature of this topsoil Mn reservoir.
6.4. Environmental significance of the coral Mn record
Identification of the environmental parameters responsible for the historic release of an apparently transient Mn
S.E. Lewis et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 246 (2007) 471–487
reservoir will remain tentative because no land conservation measures protected any comparable soils in eastern
Australia from grazing. Nonetheless, it is remarkable that
the fluctuations of Mn concentration in this coral appear to
be so decoupled from the Y concentration and Ba/Ca ratio
records, both in timing and in amplitude. If the proposal
that Ba/Ca (McCulloch et al., 2003) and Y concentrations
(Sinclair, 2005; this study) of near shore GBR seawater
are proxies for fine sediment release holds up, then the
much more dramatic response of Mn cannot relate to
physical erosion and soil loss but to release of Mn, dissolved in run-off.
Although the surface seawater inventory of Mn,
which is concentrated in plant ash, can be influenced by
wildfires (Abram et al., 2003), the fluctuations found in
this study are not related to bush fires. Namely, across
the entire 130-year growth period of the mid-Holocene
specimen that was analysed for comparison, there is no
significant deviation in Mn concentration from the mean
(0.86 ± 0.23 ppm; Table 2; Fig 6). The mid-Holocene
mean, which corresponds to a time of certain Aboriginal
inhabitation, is within error identical to the pre-1854
mean of the recent coral (0.60 ± 0.13 ppm), which argues
against wildfire influences from Aboriginal back-burning
practices.
Manganese distribution in soils reflects a complex
balance between mineralogical, biological, textural (grain
size) and physicochemical (redox, state, pH) parameters
(Post, 1999), most of which would have been affected by
sheep grazing. Unlike native tetrapods, sheep not only
disturb the soil texture with their hooves but also by
cropping plants low and commonly pulling shrubs and
grasses out of the soil by their roots. Exposure of disrupted
soil cover to air, the efficient recycling of topsoil via sheep
faeces (the volume of which would have initially overwhelmed indigenous dung beetles) and the loss of pHbuffering ash resulted in conditions under which Mn was
rapidly lost to run-off. The paucity of carbonate in the
catchment soils may have also contributed to the reduced
Mn-holding capacity of the disturbed soils (Mania et al.,
1989). The combination of these factors would have
effectively mobilised Mn, making it more available to
plants and possibly even toxic to the pre-existing native
vegetation. It may be speculated that the rapid loss of
native vegetation was triggered not only by the more
obvious physical impacts of grazing and seed dispersal
via the introduced sheep, but also by more subtle changes
to the existing micronutrient balance, which is strongly
related to Mn-mineral surface chemistry (Post, 1999).
In the modern inner Great Barrier Reef, Mn is consumed rapidly in the water column by phytoplankton
(Alibert et al., 2003), but the large release of dissolved
485
Mn in the 19th century was evidently sufficient to reach
certain near-shore reefs such as those surrounding Magnetic Island. However, Mn was not necessarily transported solely as a dissolved (Mn II) phase through the
catchment because Mn is commonly photo-chemically
reduced from Mn (IV) in particulate matter to soluble
Mn (II) species in the estuarine zone or at the ocean's
surface (Morel and Price, 2003). Further studies of
corals from similar proximity to significant catchments
will need to be investigated for Mn, Y and Ba/Ca to
refine our understanding of the details of land-use practices on near-shore seawater quality and its consequences for the health of the reef.
7. Conclusions
We investigated Ba/Ca ratios, Y and Mn concentrations in a coral from Magnetic Island, Australia. We
found that coral Y concentrations are positively correlated with cattle numbers in the catchment of the Burdekin River, the most significant source of fine grained
sediment into the GBR lagoon. Yttrium concentration
may be a more reliable proxy of soil erosion than the
coral Ba/Ca ratio, which appears to be influenced by
additional factors. The increase in coral Y and Ba/Ca
after 1870 may specifically be related to a rise in cattle
grazing within the upper Burdekin and Bowen subcatchments which, presently contribute the vast majority
of suspended sediments to the Burdekin River plume.
The 1870 to 1986 Y concentration time series shows a
significant linear trend (r2 = 0.426), which highlights the
persistent and problematic increase of sediment release
into the inner GBR lagoon.
Coral Mn concentrations followed a more complex
pattern and were linked to changes in land-use in the
Burdekin River catchment, particularly in sub-catchments containing large outcrops of geologically young
basalt. Mn concentrations significantly increased in the
1855–1856 coral record coinciding with the first stockage of sheep in the Belyando and Suttor sub-catchments,
the first land to be settled in the Burdekin catchment.
The coral Mn record demonstrates the effects of drought
on the marine environment (via reduced stock numbers)
and the influence (via wool price) of a Northern Hemisphere political event (the American Civil War) on a
coral reef in the Southern Hemisphere. Most importantly, however, it underlines that the pre-European
ecosystem in eastern Queensland relied not only on the
more obvious effects of burning undercover vegetation
but also on the subtle consequences of burning on soil
chemistry and physics. Regardless of whether the
Aboriginal inhabitants were aware that burning aided
486
S.E. Lewis et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 246 (2007) 471–487
micronutrient balance and soil pH, this new coral record
is testimony to the ignorance of European settlers of this
fragile equilibrium, which by 1900 had already been
permanently damaged, resulting in the unprecedented
loss of mammal species (Lunney, 2001), and increasingly intense bushfires.
Acknowledgements
The ICP-MS data for this study greatly benefited
from the expertise of Alan Greig. They were obtained
when the third author was employed at the University of
Queensland. We thank Prof. David Bowman and Dr. Jon
Brodie who provided informal reviews of the manuscript. SL is supported by a School of Earth Sciences
scholarship with a CRC Reef top up scholarship. Funding for the project was provided by CRC Reef. Sallie
Webster from the JCU library is acknowledged for her
help locating references and Andrea Sorbello and Gregory Lewis are thanked for assisting in the compilation
of the statistical data. The constructive comments of
Daniel Sinclair and two anonymous reviewers are gratefully acknowledged.
References
Abram, N.J., Gagan, M.K., McCulloch, M.T., Chappell, J., Hantoro,
W.S., 2003. Coral reef death during the 1997 Indian Ocean dipole
linked to Indonesian wildfires. Science 301, 952–955.
Alibert, C., Kinsley, L., Fallon, S.J., McCulloch, M.T., Berkelmans, R.,
McAllister, F., 2003. Source of trace element variability in Great
Barrier Reef corals affected by the Burdekin flood plumes. Geochim.
Cosmochim. Acta 67, 231–246.
Bainbridge, Z., Brodie, J., Lewis, S., Faithful, J., Duncan, I., Furnas, M.,
Post, D., 2006. Event based Water Quality Monitoring in the Burdekin Dry Tropics Region- 2004–2005 Wet Season. ACTFR Report
No. 06/01. Australian Centre for Tropical Freshwater Research,
James Cook University, Townsville. www.actfr.jcu.edu.au.
Beale, E., 1970. Kennedy Workbook: A Critical Analysis of the Route
of E.B. Kennedy's 1848 Exploration of Cape York Peninsula with a
Transcript of Fragments of his Journal. Wollongong University
College, Wollongong.
Belperio, A.P., 1983. Terrigenous sedimentation in the central Great
Barrier Reef lagoon: a model from the Burdekin Region. BMR
J. Aust. Geol. Geophys. 8, 179–190.
Bode, A.M., 1984. The Pioneers Went These Ways. A.M. Bode, Torrens
Creek, QLD.
Bolton, G.C., 1963. AThousand Miles Away: A History of North Queensland to 1920. The Jacaranda Press, Brisbane.
Bowman, D., 1998. Tansley review no. 101. The impact of Aboriginal
landscape burning on the Australian biota. New Phytol. 140,
385–410.
Brodie, J., McKergow, L.A., Prosser, I.P., Furnas, M., Hughes, A.O.,
Hunter, H., 2003. Sources of sediment and nutrient exports to the
Great Barrier Reef World Heritage Area. ACTFR Report No. 03/
11. Australian Centre for Tropical Freshwater Research, James
Cook University, Townsville. 208 pp. http://www.actfr.jcu.edu.au/
Publications/PDFs/03%2711%20Sources%20of %20sediment%
20and%20nutrient%20runoff %20to%20GBRWHA%201.pdf.
Campbell, J., 1936. The early settlement of Queensland. The Bibliographical Society of Queensland Transactions Second Series no 1,
Brisbane.
Chirenje, T., Ma, L.Q., 2002. Impact of high-volume wood-fired boiler
ash amendment on soil properties and nutrients. Commun. Soil
Sci. Plant Anal. 33, 1–17.
Cunningham, M.W., 1895. The Pioneering of the River Burdekin.
Argus Office, Rockhampton.
Eggins, S.M., Woodhead, J.D., Kinsley, L.P.J., Mortimer, G.E.,
Sylvester, P., McCulloch, M.T., Hergt, J.M., Handler, M.R.,
1997. A simple method for the precise determination of N40 trace
elements in geological samples by ICP-MS using enriched isotope
internal standardisation. Chem. Geol. 134, 311–326.
Esslemont, G., Russei, R.A., Maher, W.A., 2004. Coral record of
harbour dredging: Townsville, Australia. J. Mar. Syst. 52, 51–64.
Faure, G., 1998. Principles and Applications of Geochemistry: A Comprehensive Textbook for Geology Students, 2nd ed. Prentice Hall,
New Jersey.
Fensham, R.J., Fairfax, R.J., Butler, D.W., Bowman, D.M.J.S., 2003.
Effects of fire and drought in a tropical eucalypt savanna colonized
by rain forest. J. Biogeogr. 30, 1405–1414.
Flannery, T., 1994. The Future Eaters: An Ecological History of the
Australasian Lands and People. Grove Press, New York.
Fox, M.J., 1919–1923. The History of Queensland: Its People and
Industries, an Historical and Commercial Review, Descriptive and
Biographical Facts, Figures and Illustrations, an Epitome of Progress
V. 1–3. States Publishing Company, Adelaide.
Furnas, M., 2003. Catchments and Corals: Terrestrial Runoff to the Great
Barrier Reef. Australian Institute of Marine Science, Townsville.
Griffiths, T., 2002. How many trees make a forest? Cultural debates
about vegetation change in Australia. Aust. J. Bot. 50, 375–389.
Hendy, E.J., Gagan, M.K., Alibert, C.A., McCulloch, M.T., Lough, J.M.,
Isdale, P.J., 2002. Abrupt decrease in tropical Pacific sea surface
salinity at end of little ice age. Science 295, 1511–1514.
Hoyle, J., Elderfield, H., Gledhill, A., Greaves, M., 1984. The behaviour of the rare earth elements during mixing of river and sea
waters. Geochim. Cosmochim. Acta 48, 143–149.
Isdale, P.J., Daniel, E., 1989. The design and deployment of a new
lightweight submarine fixed drilling system for the acquisition and
preparation on coral cores. Mar. Technol. Soc. J. 23, 3–8.
Jones, R., 1969. Fire stick farming. Aust. Nat. Hist. 225 September issue.
Kamber, B.S., Greig, A., Schoenberg, R., Collerson, K.D., 2003. A
refined solution to Earth's hidden niobium: implications for evolution
of continental crust and mode of core formation. Precambrian Res.
126, 289–308.
Kerr, J.D., 1994. Black Snow and Liquid Gold: A History of the
Burdekin Shire. Burdekin Shire Council, Ayr, Queensland.
King, B., McAllister, F., Wolanski, E., Done, T., Spagnol, S., 2001. River
plume dynamics in the Central Great Barrier Reef. In: Wolanski, E.
(Ed.), Oceanographic Processes of Coral Reefs: Physical and Biological Links in the Great Barrier Reef. CRC Press, Boca Raton,
pp. 145–160.
Knight, J.J., 1895. In the Early Days: History and Incident of Pioneer
Queensland. Sapsford and Co., Brisbane.
Lea, D.W., Shen, G.T., Boyle, E.A., 1989. Coralline barium records
temporal variability in equatorial Pacific upwelling. Nature 340,
373–376.
Leichhardt, L., 1847. Journal of an Overland Expedition in Australia:
from Moreton Bay to Port Essington, a Distance of Upwards of 3000
Miles, During the Years 1844–1845. T. & W. Boone, London.
S.E. Lewis et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 246 (2007) 471–487
Lewis, S.E., 2006. Environmental trends in the GBR lagoon and
Burdekin River catchment during the mid-Holocene and since European settlement using Porites coral records, Magnetic Island,
Queensland. Unpublished PhD Thesis, School of Earth and Environmental Sciences, James Cook University, Townsville, Australia.
Lough, J.M., Barnes, D.J., Devereux, M.J., Tobin, B.J., Tobin, S.,
1999. Variability in growth characteristics of massive Porites on
the Great Barrier Reef. Technical Report No. 28. CRC Reef
Research Centre, Townsville.
Lunney, D., 2001. Causes of the extinction of native mammals of the
Western Division of New South Wales: an ecological interpretation
of the nineteenth century historical record. Rangeland J. 23, 44–70.
Luo, Y.-R., Byrne, R.H., 2004. Carbonate complexation of yttrium and
rare earth elements in natural waters. Geochim. Cosmochim. Acta
68, 691–699.
McCulloch, M.T., Fallon, S., Wyndham, T., Hendy, E., Lough, J.,
Barnes, D., 2003. Coral record of increased sediment flux to the
inner Great Barrier Reef since European settlement. Nature 421,
727–730.
Mania, J., Chauve, P., Remy, F., Verjus, P., 1989. Evolution of iron and
manganese concentrations in presence of carbonates and clays in
the alluvial groundwaters of the Ognon (Franche-Comté, France).
Geoderma 44, 219–227.
Menghetti, D., 1992. Ravenswood: Five Heritage Trails. Department
of History and Politics, James Cook University, Townsville.
Mitchell, T.L., 1848. Journal of an Expedition into the Interior of
Tropical Australia: in Search of a Route from Sydney to the Gulf of
Carpentaria. Longman, Brown, Green and Longmans, London.
Morel, F.M.M., Price, N.M., 2003. The biogeochemical cycles of trace
metals in the ocean. Science 300, 944–947.
Moss, A.J., Rayment, G.E., Rellly, N., Best, E.K., 1992. A Preliminary
Assessment of Sediment and Nutrient Exports form Queensland
Coastal Catchments. Department of Environment and Heritage
Report. Queensland Government.
O'Donnell, D., 1989. A History of Clermont and District: Belyando
Shire. Belyando Shire Council, Clermont, Queensland.
Okai, T., Suzuki, A., Kawahata, H., Terashima, S., Imai, N., 2002.
Preparation of a New Geological Survey of Japan Geochemical
Reference Material: Coral JCp-1. Geostand. Newsl. 26, 95–100.
Post, J.E., 1999. Manganese oxide minerals: crystal structures and
economic and environmental significance. Proc. Natl. Acad. Sci.
96, 3447–3454.
487
Prosser, I.P., Moran, C.J., Lu, H., Scott, A., Rustomji, P., Stevenson, J.,
Priestly, G., Roth, C.H., Post, D., 2002. Regional patterns of
erosion and sediment transport in the Burdekin River catchment.
CSIRO Land and Water Technical Report No 5/02. February.
Shen, G.T., Boyle, E.A., 1987. Lead in corals: reconstruction of
historical fluxes to the surface ocean. Earth Planet. Sci. Lett. 82,
289–304.
Shen, G.T., Sanford, C.L., 1989. Trace element indicators of climate
variability in reef-building corals. In: Glynn, P.W. (Ed.), Global
Ecological Consequences of the 1982–83 El Niño Southern
Oscillation. Elsevier Oceanography Series, vol. 52, pp. 255–283.
Shen, G.T., Boyle, E.A., Lea, D.W., 1987. Cadmium in corals as a tracer
of historical upwelling and industrial fallout. Nature 328, 794–796.
Shen, G.T., Campbell, T.M., Dunbar, R.B., Wellington, G.M., Colgan,
M.W., Glynn, P.W., 1991. Paleochemistry of manganese in corals
from the Galapagos Islands. Coral Reefs 10, 91–100.
Shen, G.T., Linn, L.J., Campbell, T.M., 1992. A chemical indicator of
trade wind reversal in corals from the Western Tropical Pacific.
J. Geophys. Res. 97, 12689–12697.
Sinclair, D.J., 2005. Non-river flood barium signals in the skeletons of
corals from coastal Queensland, Australia. Earth Planet. Sci. Lett.
237, 354–369.
Sinclair, D.J., McCulloch, M.T., 2004. Coral record low mobile barium
concentrations in the Burdekin River during the 1974 flood: evidence for limited Ba supply to rivers? Palaeogeogr. Palaeoclimatol.
Palaeoecol. 214, 155–174.
Statistics of the Colony of Queensland/Statistics of the State of
Queensland, 1860−1974. Bureau of Census and Statistics,
Queensland Office, Brisbane.
Thorpe, B., 1996. Colonial Queensland: Perspectives on a Frontier
Society. University of Queensland Press, Brisbane.
Wyndham, T., McCulloch, M., Fallon, S., Alibert, C., 2004. Highresolution coral records of rare earth elements in coastal seawater:
biogeochemical cycling and a new environmental proxy. Geochim.
Cosmochim. Acta 68, 2067–2080.
Young, A., 2000. Environmental Change in Australia since 1788, 2nd ed.
Oxford University Press, Melbourne.