Sediments, nutrients and pesticide residues in event flow conditions in streams of the Mackay Whitsunday Region, Australia C. Mitchell a , J. Brodie b, * , I. White c a Mackay Whitsunday Natural Resource Management Group, Mackay, 4740, Australia b Australian Centre for Tropical Freshwater Research, James Cook University, Townsville, Qld 4811, Australia c Department of Natural Resources and Mines, Mackay, 4740, Australia Abstract The Mackay Whitsunday region covers 9000 km 2 in northeastern Australia. A study of diffuse pollutants during high flow events was conducted in coastal streams in this region. Sampling was conducted in the Pioneer River catchment during a high flow event in February 2002 and in Gooseponds Creek, Sandy Creek and Carmila Creek in March 2003. Concentrations of five herbicides; atra- zine (1.3 lgl 1 ), diuron (8.5 lgl 1 ), 2,4-D (0.4 lgl 1 ), hexazinone (0.3 lgl 1 ) and ametryn (0.3 lgl 1 ) and high concentrations of nutrients (total nitrogen 1.14 mg l 1 , total phosphorus 0.20 mg l 1 ) and suspended sediments (620 mg l 1 ) were measured at Dumble- ton Weir on the lower reaches of the Pioneer River. Drinking water guidelines for atrazine and 2,4-D were exceeded at Dumbleton Weir, low reliability trigger values for ecosystem protection for diuron were exceeded at three sites and primary industry guidelines for irrigation levels of diuron were also exceeded at Dumbleton Weir. Similar concentrations were found in the three smaller streams measured in 2003. Herbicides and fertilisers used in sugarcane cultivation were identified as the most likely major source of the her- bicide residues and nutrients found. Ó 2004 Elsevier Ltd. All rights reserved. Keywords: Nutrients; Pesticide residues; Pioneer river; Mackay; Sugarcane cultivation 1. Introduction The Mackay Whitsunday region is one of fourteen Natural Resource Management regions in Queensland. The major rivers of the region include the Proserpine, OÕConnell and Pioneer and in addition there are many smaller streams which discharge directly to the sea, including importantly for this study, Carmilla, Sandy and Gooseponds Creeks (Fig. 1). The receiving waters for discharge from all these rivers and streams form part of the Great Barrier Reef (GBR) Lagoon and western Coral Sea (Devlin et al., 2001a). The catchments of the streams in this study have high proportions of agricul- tural land uses dominated by sugarcane cultivation and beef grazing with smaller areas of urban use and in some cases considerable areas of native forest. For the combined catchment area of the Pioneer River, Bak- ers Creek, Sandy Creek and Gooseponds Creek (2200 km 2 see Fig. 1) sugarcane occupies 690 km 2 , beef grazing 610 km 2 , forest 670 km 2 , urban 150 km 2 and other uses 83 km 2 . The water quality of QueenslandÕs east coast streams has been of concern for some time (Arthington et al., 1997) and especially the potential for pollution and deg- radation of parts of the Great Barrier Reef (Brodie, 2002; Furnas, 2003). The streams and rivers of the northeast Australian coast form a convenient set for comparative studies with tropical and sub-tropical cli- matic regimes. Many of the rivers have headwaters in natural forest, middle courses in areas of rangeland beef 0025-326X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.marpolbul.2004.10.036 * Corresponding author. Tel.: +61 7 4781 6435; fax: +61 7 4781 5589. E-mail address: [email protected](J. Brodie). www.elsevier.com/locate/marpolbul Marine Pollution Bulletin 51 (2005) 23–36
14
Embed
Sediments, nutrients and pesticide residues in event flow conditions ...
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
www.elsevier.com/locate/marpolbul
Marine Pollution Bulletin 51 (2005) 23–36
Sediments, nutrients and pesticide residues in event flow conditionsin streams of the Mackay Whitsunday Region, Australia
C. Mitchell a, J. Brodie b,*, I. White c
a Mackay Whitsunday Natural Resource Management Group, Mackay, 4740, Australiab Australian Centre for Tropical Freshwater Research, James Cook University, Townsville, Qld 4811, Australia
c Department of Natural Resources and Mines, Mackay, 4740, Australia
Abstract
The Mackay Whitsunday region covers 9000km2 in northeastern Australia. A study of diffuse pollutants during high flow events
was conducted in coastal streams in this region. Sampling was conducted in the Pioneer River catchment during a high flow event in
February 2002 and in Gooseponds Creek, Sandy Creek and Carmila Creek in March 2003. Concentrations of five herbicides; atra-
zine (1.3lg l�1), diuron (8.5lg l�1), 2,4-D (0.4lg l�1), hexazinone (0.3lg l�1) and ametryn (0.3lg l�1) and high concentrations ofnutrients (total nitrogen 1.14mgl�1, total phosphorus 0.20mgl�1) and suspended sediments (620mgl�1) were measured at Dumble-
ton Weir on the lower reaches of the Pioneer River. Drinking water guidelines for atrazine and 2,4-D were exceeded at Dumbleton
Weir, low reliability trigger values for ecosystem protection for diuron were exceeded at three sites and primary industry guidelines
for irrigation levels of diuron were also exceeded at Dumbleton Weir. Similar concentrations were found in the three smaller streams
measured in 2003. Herbicides and fertilisers used in sugarcane cultivation were identified as the most likely major source of the her-
Natural Resource Management regions in Queensland.
The major rivers of the region include the Proserpine,
O�Connell and Pioneer and in addition there are manysmaller streams which discharge directly to the sea,including importantly for this study, Carmilla, Sandy
and Gooseponds Creeks (Fig. 1). The receiving waters
for discharge from all these rivers and streams form part
of the Great Barrier Reef (GBR) Lagoon and western
Coral Sea (Devlin et al., 2001a). The catchments of the
streams in this study have high proportions of agricul-
0025-326X/$ - see front matter � 2004 Elsevier Ltd. All rights reserved.
tions of SS and nutrients in Sandy Creek were moderate
compared to some of the other streams sampled in this
Table 4
Suspended sediment, nutrient species and herbicide concentrations in Goose
Substance Sampling time in March, 2003
23.05 25th 3.0
SS, mgl�1 160 39
TN, mgl�1 1.34 4.6
NOx-N, mgl�1 0.34 2.4
NH3-N, mgl�1 0.047 0.1
DIN, mgl�1 0.387 2.5
DON, mgl�1 0.463 0.8
PN, mgl�1 0.49 1.2
TP, mgl�1 0.25 0.5
PO4-P, mgl�1 0.17 0.3
PP, mgl�1 0.08 0.2
DOP, mgl�1 <0.01 <0
Diuron, lg l�1 0.56 2.5
Atrazine, lg l�1 0.67 1.8
Desethylatrazine, lg l�1 NDR 0.0
Hexazinone, lg l�1 NDR ND
Ametryn, lg l�1 0.71 0.1
2,4-D, lg l�1 0.19 ND
study (Table 6). However, considerable amounts of her-
bicides were detected with diuron concentrations in all
four samples exceeding low-level confidence TV for eco-
system protection of 0.2lg l�1 (ANZECC and ARM-
CANZ, 2000).
Total calculated loads for the events at Dumbleton,Finch Hatton, Gooseponds and Sandy Creek are sum-
marised in Table 7. The relatively large amounts of diu-
ron (470kg) and atrazine (75kg) discharged in the
Pioneer River at Dumbleton are noteworthy.
4. Discussion
Results from Mackay Whitsunday streams can be
compared with other northeastern Australian rivers
including some with limited catchment development
(Jardine and Annan well to the north of Mackay) to
ponds Creek
5 26th 7.05 26th 11.35 26th
0 150 78
5 4.8
3.3 3.2
4 0.14 0.096
4 3.44 3.30
4 0.86 0.904
2 0.70 0.60
7 0.63 0.56
3 0.34 0.35
4 0.25 0.16
.01 0.04 0.05
2.8 5.3
2.1 4.1
9 0.10 0.17
R 0.43 1.0
8 0.12 0.14
R NDR NDR
Table 5
Suspended sediment, nutrient species and herbicide concentrations in Carmilla Creek
Substance Sampling time
8.30 1st 10.30 1st 12.30 1st 14.30 1st 16.30 1st
SS, mgl�1 2 2 3 <1 173
TN, mgl�1 1.94 2.09 2.19 2.16 2.98
NOx-N, mgl�1 1.31 1.63 1.65 1.65 0.88
NH3-N, mgl�1 0.10 0.03 0.02 0.03 0.14
DIN, mgl�1 1.41 1.66 1.67 1.68 1.02
DON, mgl�1 0.42 0.43 0.42 0.37 0.48
PN, mgl�1 0.11 <0.01 0.10 0.11 1.48
TP, mgl�1 0.07 0.03 0.02 0.01 0.30
PO4-P, mgl�1 0.06 0.03 0.02 0.02 0.05
PP, mgl�1 0.03 <0.01 <0.01 <0.01 0.24
DOP, mgl�1 <0.01 <0.01 0.01 <0.01 0.01
Diuron, lg l�1 NDR NDR NDR ND 0.6
Atrazine, lg l�1 NDR NDR NDR NDR NDR
Desethylatrazine, lg l�1 NDR NDR NDR NDR NDR
Hexazinone, lg l�1 NDR NDR NDR NDR NDR
Ametryn, lg l�1 NDR NDR NDR NDR NDR
2,4-D, lg l�1 NDR NDR NDR NDR NDR
Table 6
Suspended sediment, nutrient species and herbicide concentrations at Sandy Creek
Substance Sampling time in March, 2003
12.20 2nd 18.40 2nd 0.35 3rd 7.30 4th
SS, mgl�1 188 151 307 13
TN, mgl�1 1.78 1.38 1.24 1.04
NOx-N, mgl�1 0.44 0.34 0.31 0.20
NH3-N, mgl�1 0.04 0.03 0.03 0.03
DIN, mgl�1 0.48 0.37 0.34 0.23
DON, mgl�1 0.68 0.64 0.67 0.57
PN, mgl�1 0.62 0.37 0.23 0.24
TP, mgl�1 0.30 0.31 0.31 0.30
PO4-P, mgl�1 0.13 0.17 0.21 0.22
PP, mgl�1 0.17 0.17 0.12 0.10
DOP, mgl�1 <0.01 <0.01 <0.01 <0.01
Diuron, lg l�1 0.87 1.1 1.6 0.6
Atrazine, lg l�1 NDR NDR NDR NDR
Desethylatrazine, lg l�1 NDR NDR NDR NDR
Hexazinone, lg l�1 0.1 NDR NDR NDR
Ametryn, lg l�1 NDR NDR NDR NDR
2,4-D, lg l�1 0.87 1.1 1.6 0.6
C. Mitchell et al. / Marine Pollution Bulletin 51 (2005) 23–36 29
large rivers in the Queensland dry tropics with land uses
dominated by rangeland beef grazing (Normanby,
Burdekin, Fitzroy) to those rivers with similar develop-
ment to the Pioneer (urban, cropping and beef grazing
uses) and in similar rainfall regimes (e.g., Johnstone,
Barron, Richmond, Tully and Herbert). Suspended sed-
iment (SS) concentrations in the Pioneer at Dumbleton
during the event were similar to those found in mostother rivers peaking at a concentration of 620mgl�1.
Peak concentrations of SS in dry tropics Queensland riv-
ers are generally considerably higher than these results
with, for example, values closer to 2000–3000mgl�1 in
the Burdekin River (Furnas and Mitchell, 2001). How-
ever, the SS results from the Mackay Whitsunday
streams in the present study are in the same range as
those rivers with similar land uses and rainfall regimes
e.g. Johnstone (100–1300mgl�1) (Hunter et al., 1997),
Herbert (50–800mgl�1) (Mitchell et al., 1997) and Rich-
mond (300–700mgl�1) (Hossein et al., 2002). SS concen-
trations mirror the rise of the hydrograph as hydraulic
power is the principal soil eroding factor and potential
SS in the catchment is virtually inexhaustible in theduration (2–3 days) of this event. Concentrations of
TP and orthophosphate (PO4) were similar to those in
other northeastern Australian rivers. TP concentrations
for the Pioneer at Dumbleton were in the range 160–
500lg l�1, comparable to those found in the Barron
River (30–110lg l�1) (Cogle et al., 2000), Fitzroy River
Table 7
Calculated loads during each event
Loads Dumbleton Finch Hatton Gooseponds Ck Sandy Ck
Flow volumes for the period (ML) 126,000 985 26,000 2100
Loads
TN, tonnes 243 0.68 5.85 30.3
TP, tonnes 44 0.178 1 7.9
Nitrate + nitrite, tonnes 78 0.3
SS, tonnes 41,500 29.5 358 4410
Diuron, kg 470 NDR 6.9 26
Ametryn, kg 22 NDR 0.5 NDR
Atrazine, kg 75 NDR 5.3 NDR
Hexazinone, kg 28 NDR NDR NDR
30 C. Mitchell et al. / Marine Pollution Bulletin 51 (2005) 23–36
(120–640lg l�1) (Furnas, 2003) and Richmond River
(50–600lg l�1) (McKee et al., 2000). In contrast the con-centrations of TN and nitrate plus nitrite (NOx) were
higher in the Pioneer than for most other rivers. NOx
concentrations of 300–900lg l�1 are similar to those
from the Tully River which has nitrate in stormflow in
the range 100–1000lg l�1 (Mitchell et al., 2001; Furnas,2003). Rivers such as the Jardine and Annan have verylow nitrate concentrations (e.g. for the Jardine in flow
conditions in the range 2–13lg l�1, Eyre and Davies,
1996) at all times reflecting the lack of human-influenced
sition). At Dumbleton, NOx showed a strong �first flush�behaviour with the highest concentrations occurring
early in the event and tailing off quickly as the event pro-
gressed. This suggests a limited supply of NOx in thecatchment, derived mostly from applied nitrogen fertil-
iser, which was quickly exhausted, and possibly under-
went a dilution effect late in the hydrograph as low
nitrate water from the upper catchment (with little fertil-
iser use) finally made its way to Dumbleton. On the
other hand dissolved organic nitrogen concentrations
at both Dumbleton and Finch Hatton were quite con-
stant through the event albeit with the Dumbleton con-centrations about double that at Finch Hatton.
Concentrations of many parameters were lower at
Finch Hatton than at Dumbleton as expected. However
elevated concentrations of NOx (760lg l�1) and TN
(1140lg l�1) were detected suggesting a source of nitrateabove the sampling point. The increase in total nitrogen
at Finch Hatton (Table 4) was due entirely to an in-
crease in NOx with particulate nitrogen, ammonia,and dissolved organic nitrogen staying almost un-
changed. The late rise in the NOx concentration com-
pared to the hydrograph at Finch Hatton suggests
that a sub-surface flow of nitrate-rich water may have
been involved. Water at Finch Hatton, even in the peak
of the flow (�7.30am) had quite low suspended solids
(33mgl�1) and virtually no particulate phosphorus.
The rather higher concentrations of nitrate and ortho-phosphate suggest these soluble nutrients may have
arisen from septic systems or animal waste. While the
Finch Hatton sampling site had initially been selected
to reflect a largely unimpacted stream in the region it
was later found to have a considerable rural residential
and small scale tourism infrastructure above the sam-
pling point all served by infiltration septic sewage sys-
tems. Considerable areas of the catchment above the
sampling point have been cleared and riparian vegeta-
tion disturbed. The elevated nitrate concentrations(0.76mgl�1 NO3-N) compared to those normally found
in event runoff from undisturbed rainforest in north
Queensland (e.g. 0.04mgl�1 NO3-N in the Russell–Mul-
grave catchment, Devlin et al., 2001b) reflect this catch-
ment development.
In Gooseponds Creek SS concentrations peaked at
390mgl�1, a lower value that those often seen in larger
rivers in event flow. Similar peak concentrations seenin Sandy Creek (307mgl�1) and Carmilla Creek
(173mgl�1 before the peak) show that levels of ground
cover in these catchments are relatively high and with
the limited rainfall intensity and hydraulic power of
the 2003 events relatively low rates of soil erosion oc-
curred. Over the last decade sugarcane cultivation in
the Mackay Whitsunday region has moved from a sys-
tem where the cane was burned before harvest, and thusno crop residues were retained on the soil between crops,
to one where the cane is harvested green and the cane
leaves are left on the soil as a trash blanket. The older
harvest practice, associated with a high level of tillage,
led to very high soil erosion rates, between 42 and
227tonnesha�1yr�1 (Sallaway, 1979). The modern prac-
tice of green cane harvesting and trash blanketing
(GCTB), associated also with reduced tillage, has re-duced soil erosion rates in sugarcane cultivation to low
values, probably in the range 5–15tonnesha�1yr�1
(Prove and Hicks, 1991; Prove et al., 1995; Rayment,
2003). While some areas of sugarcane cultivation opera-
tions are still erosion prone e.g. headroads and drains,
and many urban development sites in the region are sed-
iment sources, the relatively low SS concentrations
found in the streams draining large areas of sugarcanein the present study show the effectiveness of GCTB as
a soil erosion preventative measure.
C. Mitchell et al. / Marine Pollution Bulletin 51 (2005) 23–36 31
Nutrient concentrations in Gooseponds were high
compared to the other study sites. The high nitrate con-
centrations peaking late in the event may indicate a
combination of surface and sub-surface flows. It has
been suggested that old septic systems, recently replaced
by reticulated sewage collection, in the urban part of theGooseponds catchment may be still leaching nitrates
into sub-surface water flows. However, Gooseponds
catchment also has a large area of sugarcane cultivation
and this may be another source of some of the nitrate.
Concentrations of nutrients in SandyCreek in the 2003
event were all relatively low. In other studies (Wilhelm,
2001) Sandy Creek has been shown to have the highest ni-
trate concentrations in event flows of any of the 11streams analysed in catchments containing significant
sugarcane areas in Queensland. In high discharge periods
Wilhelm (2001) found nitrate concentrations ranging
from 1 to 2.5mgl�1 NO3-N, total nitrogen from 2 to
4mgl�1 and total phosphorus from 0.1 to 0.3mgl�1 in
SandyCreek at the same site sampled in the present study.
The relatively low values found in the present study may
reflect the small scale and low intensity of the event.In Carmilla Creek (Table 5) most samples (first four)
were taken in the baseflow period before the main event.
The relatively high nitrate concentrations in the base-
flow period (1.6mgl�1 NO3-N) may reflect a stable
source of nitrate possibly also associated with sub-sur-
face flow of high nitrate water. Nitrate is known to leach
strongly from soluble nitrogen fertilisers (urea and
ammonium nitrate) used on sugarcane in wet areas ofnorth Queensland (Rasiah and Armour, 2001; Rasiah
et al., 2003; Bohl et al., 2000). Rural residential septic
sewage systems may also be a source of nitrate as shown
in the Johnstone catchment (Hunter and Walton, 1997).
In Carmilla Creek event flow concentrations (sample
five) appear to be rising to the higher concentrations
seen in the other streams in this study except for nitrate
where a dilution of the nitrate rich baseflow with lowernitrate surface flow seems to have occurred.
Nutrient species composition (especially for nitrogen)
in the range of streams sampled in this study follows the
patterns seen in other tropical areas (Lewis et al., 1999;
Downing et al., 1999). Undisturbed tropical landscapes
appear to have high nitrogen loss rates compared to
temperate systems (Downing et al., 1999). Thus tropical
rivers may have higher concentrations of inorganicnitrogen than would be expected from their pristine
state. However the predominant form of N lost from
undisturbed forests, in both tropical and temperate
conditions, is dissolved organic nitrogen (DON) (Lewis
et al., 1999; Perakis and Hedin, 2002). As catch-
ment development proceeds, no matter whether the
development occurs as deforestation, grazing, fertilised
cropping, urbanization or industrial developmentand atmospheric N deposition, proportionally greater
amounts of dissolved inorganic nitrogen (predominantly
nitrate) are exported in rivers and streams (Downing
et al., 1999; Caraco and Cole, 1999; Harris, 2001; Turner
et al., 2003). Runoff from undisturbed catchments in
tropical America in moderate runoff climatic conditions
has volume-weighted mean concentrations of 102lg l�1
NO3-N, 119lg l�1 DON, 86lg l�1 PN and 376lg l�1
TN (Lewis et al., 1999). The limited data available from
north Queensland undisturbed wet tropics streams in
event flows suggest nitrate and PN values are less than
this, perhaps averaging near 50lg l�1 NO3-N and
50lg l�1 PN respectively (Brodie et al., 2003) while the
DON concentrations are similar to tropical America at
150lg l�1 DON. The streams in the present study areobviously not undisturbed and the concentrationsranges of the nitrogen species reflect the degree of distur-
bance and intensity of land use. Nitrate concentrations
in event flow range from 130 to 3300lg l�1 NO3-N,DON from 210 to 900lg l�1 and PN from 170 to
1480lg l�1.These results do show the changes in nitrogen species
composition and concentration anticipated from knowl-
edge of the major intensive land uses in the area i.e. sug-arcane cultivation and urban development. Elevated
concentrations of nitrate and PN are seen compared to
natural forest systems and increases in nitrate and PN
as proportions of the total nitrogen (TN) occur. A mod-
erate increase in DON above natural levels but with
DON as a lower proportion of the TN also occurs. Run-
off in stormwater discharge events from sugarcane fields
in north Queensland can have concentrations of nitratein the range of 500–6000lg l�1 NO3-N (Bramley and
Roth, 2002; Pearson et al., 2003; Faithful and Finlay-
son, in press) and ammonia concentrations can also
reach 5,000lg l�1 NH3-N (Pearson et al., 2003). Nitrate
also leaches to shallow sub-surface waters and ground-
water in high concentrations under sugarcane cultiva-
tion with concentrations similar to that seen in surface
runoff, 1–10mgl�1 NO3-N (Rasiah and Armour, 2001;Rasiah et al., 2003; Biggs et al., 2001). It has been shown
that this nitrate-rich shallow groundwater can eventu-
ally be discharged into adjacent streams (Rasiah et al.,
2003). PN concentrations are typically in the range
100–1,500lg l�1 in stormflow runoff (Bramley and Roth,2002). Urban runoff may also contain similarly elevated
concentrations of nitrate and PN (Chiew and McMa-
hon, 1999). This high concentration water from theareas of catchments under intensive land uses is diluted
with water from low intensity land uses (natural forest,
rangeland grazing, woodlands). The process produces
the characteristic water quality at end-of-catchment sites
where intensive land uses occupy 20–60% of the catch-
ment area as in the present study, with nitrate concen-
trations near 400–2000lg l�1, PN near 500–1000lg l�1
and DON near 300–600lg l�1.Nitrogen to phosphorus molar ratios from the peak
discharge period in each stream are shown in Table 8.
Table 8
Nitrogen to phosphorus molar ratios in peak discharge waters
Stream site TN:TP DIN:PO4
Dumbleton 12:1 17:1
Finch Hatton 22:1 19:1
Gooseponds 18:1 22:1
Sandy 13:1 4:1
Carmilla 22:1 43:1
32 C. Mitchell et al. / Marine Pollution Bulletin 51 (2005) 23–36
Both TN:TP and DIN:PO4 are shown, as while TN:TP
ratios have often been used to predict nutrient limita-
tion, DIN:PO4 may give a more accurate indication
due to the presence of large and often unknown
amounts of non-bioavailable forms of nitrogen (often
DON) in the TN. Plant requirements for N and P arebelieved to occur in the ratio of the intracellular N:P
content of organisms—16:1 for phytoplankton (Redfield
et al., 1963; Harris, 1999). Large deviations from this
ratio in waters indicate a growth-limiting deficiency of
one element (Turner et al., 2003). Waters in the present
study show N:P ratios close to the Redfield ratio with
only one site, Carmilla Creek, showing a considerable
deviation from the Redfield ratio. The Carmilla Creekresults come from only one sample in the event flow per-
iod and may not be completely representative of the N
and P forms in the complete event. With N:P ratios of
these values the event flow waters show no strong ten-
dency to cause either N or P limitation to algal growth.
Concentrations of herbicide residues at Dumbleton
(Table 1) also showed a strong �first flush� behaviour.Diuron had the highest concentration of the herbicidesdetected with a peak concentration of 8.5lg l�1. Thereis little data available on diuron concentrations in river
water for comparison but concentrations of 2.3lg l�1
were detected in the Johnstone River under flow condi-
tions (Hunter et al., 2001). Diuron was not detected
in irrigation channel sediments in the Homebush area
in the Sandy Creek catchment in surveys carried out in
1998 (Muller et al., 2000) but has been detected in Pio-neer River estuarine sediments (Duke et al., 2001; Duke
and Bell, in press). Atrazine concentrations peaked at
1.3lg l�1 at Dumbleton early in the flow event but othervalues at Dumbleton, Gargett (Cattle Creek) and Mia
Mia (mid course Pioneer River) were less than 0.5lg l�1.These concentrations were comparable to those found in
the Fitzroy Basin (Dawson River) during 1993–1999 by
Noble et al. (1997) and Noble and Collins (2000) (gener-ally in the range of 0.1–2.31lg l�1, with one value of
6.5lg l�1) and in the Johnstone River (Hunter et al.,
2001) where concentrations up to 0.7lg l�1 were regu-larly found. Concentrations of 2,4-D (max. 0.4lg l�1
at Dumbleton) are not great when compared to other
river systems for which data exists (e.g. ranges of 0.18–
15.6lg l�1 in the Johnstone River, Hunter et al., 2001).Similar concentrations of herbicides to those found in
this study have also been found in the Mary River sys-
tem (south-east Queensland) by McMahon et al.
(2003) where diuron was the most commonly detected
herbicide, but in relatively low concentrations (0.02–
0.1lg l�1), and simazine found in the highest concentra-tions with three sites with concentrations between 3.2and 4.2lg l�1.The concentrations of herbicide residues and dis-
solved nutrients found in Mackay Whitsunday region
surface waters during these events can also be compared
to concentrations found in recent studies in the Pioneer
basin in groundwater (Baskeran et al., 2002). Ground-
water samples for this study were collected in the lower
Pioneer Valley in April–May 1997. The study showedthat thirty percent of samples were contaminated with
one or more herbicides (ametryn, atrazine, desethylatr-
azine, bromacil, diuron and hexazinone), though none
were present at concentrations exceeding the Drinking
Water Guideline Values (NHMRC, 1996). The concen-
trations found in the groundwater were similar to those
found in the event surface water sampling at Dumble-
ton. In the groundwater diuron was most commonlyfound (nine of 14 bores) at concentrations up to
1.80lg l�1 while atrazine was found in six bores at con-centrations up to 0.12lg l�1.Diuron and to a lesser extent atrazine have the lon-
gest half-lives and are the most soluble of the pesticides
used widely in sugarcane cultivation in Queensland
(Hargreaves et al., 1999). It is thus not unexpected that
these are the pesticides found most commonly in off-farm environments in sugarcane growing regions
whether these are marine sediments (Haynes et al.,
2000a), mangrove sediments (Duke et al., 2001; Duke
and Bell, in press), groundwater (Baskeran et al., 2002)
or, as in the case of the present study, surface water
event flow.
In all four sampling locations ANZECC TVs for TP,
orthophosphate, TN and nitrate were exceeded. In factall concentrations of these four parameters measured
during this study exceed these TVs. This was not entirely
surprising as the TVs are for �slightly disturbed ecosys-tems� (ANZECC and ARMCANZ, 2000) and these re-
sults confirm that the Pioneer River, even in some of
its upper reaches, is more heavily disturbed than
�slightly�. It would also be expected that maximum val-
ues of these parameters would occur under stormflowconditions. It is known from the DNRME statewide
river monitoring data set that, for example, TN gener-
ally exceeded the ANZECC TV in baseflow conditions
at Dumbleton (Brodie, 2004).
A number of the pesticide residue concentrations
were also of some concern. Maximum atrazine concen-
trations at Dumbleton Weir exceeded the ecosystem
health TV. Atrazine and 2,4-D concentrations also ex-ceeded the drinking water guideline value but not the
health value. For drinking water this implies the source
Table 9
Event loads at Dumbleton compared to estimated mean annual loads
Substance 2002 event at Dumbleton Estimated annual mean loads