A SUPPORTING DOCUMENT TO THE SWAN-CANNING CLEANUP PROGRAM ACTION PLAN
November 2001 Swan-Canning Cleanup Program
Seasonal water quality patterns in the Swan River Estuary, 1994–1998, technical report SWAN RIVER TRUST 3rd floor, Hyatt Centre 87 Adelaide Terrace EAST PERTH Western Australia 6004 Telephone: (08) 9278 0400 Fax: (08) 9278 0401 Website: www.wrc.wa.gov.au/srt
While negotiation and economic mechanisms provide external influences, people’s own understanding and attitudes are power- ful determinants of the extent to which they respond to environmental challenges. So the process of guiding the develop- ment of behaviour through education should be an integral part of any strategy for environmental protection.
NSW Government Green Paper, 1996
Printed on recycled paper
ISBN 0-7309-7549-5 ISSN 1444-3279 SWAN RIVER TRUST Swan-Canning Cleanup Program
Seasonal water quality patterns in the Swan River Estuary, 1994–1998, technical report
A supporting document to the Swan-Canning Cleanup Program Action Plan SCCP Report No. 23
November 2001 Acknowledgments
The Aquatic Science Branch (ASB) of the Water and Rivers Commission established and managed this monitoring program since 1994 on behalf of the Swan River Trust through the Swan Canning Cleanup Program.
Several past and present environmental officers from the ASB have made substantial contributions to the collection of the field data, including Kathryn McMahon, Petrina Raitt and Mischa Cousins. Kathryn McMahon and Rachel Spencer performed preliminary data analysis. Logistical support was provided by the Swan River Trust Field Operations unit, David Fardig, Vaughan Smith, Ivan Stork, Barry Boydell and Neil Chapman.
Reference details
The recommended reference for this publication is: Thomson, C.E., Rose, T. and Robb, M. 2001, Seasonal water quality patterns in the Swan River Estuary, 1994-1998, technical report. Swan River Trust, Western Australia.
ii Contents
Acknowledgments ...... ii
1. Introduction ...... 1
2. Objectives ...... 2
3. Background ...... 3
3.1 Previous monitoring programs ...... 3
3.2 Geology and climate ...... 3
3.3 Bathymetry and artificial changes ...... 3
3.4 Tides ...... 4
3.5 Freshwater inflows ...... 5
3.6 Hydrodynamics ...... 5
4. Methods ...... 6
4.1 Sampling sites ...... 6
4.1.1 Estuary (from Blackwall Reach to Upper Swan) ...... 6
4.1.2 Catchment (Upper Swan) ...... 6
4.2 Sampling program...... 6
4.2.1 Physico-chemical data ...... 6
4.2.2 Nutrients ...... 8
4.2.3 Chlorophyll...... 8
5. Results ...... 9
5.1 Physico-chemical parameters...... 9
5.1.1 Freshwater inflows ...... 9
5.1.2 Salinity...... 9
5.1.3 Dissolved oxygen ...... 9
5.1.4 Temperature ...... 12
5.1.5 pH ...... 12
5.1.6 Secchi depth ...... 14
5.2 Water quality data: Summary statistics ...... 14
5.3 Nutrients: Seasonal patterns ...... 14
5.3.1 Nitrogen ...... 14
5.3.2 Phosphorus ...... 17
5.3.3 Silica ...... 19
5.4 Chlorophyll: Seasonal patterns ...... 20 iii 6. Discussion ...... 23
7. Conclusions ...... 26
8. References ...... 27
Appendix A ...... 28
Boxplot explanation ...... 28
Appendix B ...... 28
ANZECC/ARMCANZ 2000 Guidelines ...... 28
Publication feedback form ...... 29
Tables
1. Swan River estuary site details ...... 7
2. Swan River catchment site details ...... 7
3. Water quality: Swan River estuary summary statistics (N = total number of samples, units for all parameters are mgL-1) ...... 15
4. ANZECC water quality default trigger values for estuaries and lowland rivers, southern Western Australia ...... 28
iv Figures
1. Map of Western Australia, showing the Swan River estuary, Canning River and major catchment tributaries ...... 4
2. Water quality sampling sites for the Swan River estuary ...... 6
3. Swan River discharge (Avon and Ellen Brook flows) ...... 10
4. Annual hydrodynamic cycle, 1997 ...... 10
5. Dissolved oxygen (% saturation) monthly medians for the surface and bottom waters of the lower, middle and upper estuary ...... 11
5a. Annual dissolved oxygen dynamics, 1997 ...... 11
6. Monthly median temperatures in the surface and bottom waters of the lower, middle and upper estuary ...... 12
7. Monthly median pH in the surface and bottom waters of the lower, middle and upper estuary ...... 13
8. Monthly median Secchi depths in the lower, middle and upper estuary ...... 13
9. Monthly median concentrations of dissolved inorganic nitrogen (DIN) in the surface and bottom water of the lower, middle and upper estuary ...... 16
+ 10. Monthly median concentrations of NOx and NH4 in the surface and bottom waters of the lower, middle and upper estuary ...... 17
11. Monthly median total nitrogen (TN) concentrations in the surface and bottom water of the lower, middle and upper estuary ...... 18
12. Monthly median TN, DIN and organic-N concentrations in the major catchment inflows to the upper estuary ...... 18
13. Box plots for TN, DIN and organic-N in the major catchment inflows to the middle estuary ...... 19
14. Monthly medians for filterable reactive phosphorus (FRP) in the surface and bottom water of the lower, middle and upper Swan River estuary ...... 20
15. Box plots for total phosphorus (TP) in the surface and bottom water of the lower, middle and upper estuary ...... 21
16. Box plots for TP, FRP and particulate-P in the major catchment inflows to the upper estuary...... 21
17. Box plots for TP, FRP and particulate-P in the major catchment inflows to the middle estuary ...... 22
18. Concentration of silica in the surface, 1 m and bottom waters of the lower, middle and upper Swan River estuary, for 1997 and 1998 data only ...... 22
19. Chlorophyll a monthly medians in the surface and bottom waters of the lower, middle and upper estuary ...... 22
20. DIN plotted against FRP during winter (Jul-Aug) in the three regions of the estuary ...... 23
21. DIN vs FRP during summer (Dec-Feb) in surface and bottom waters ...... 25
v
1. Introduction
By the early 1990s, the Swan-Canning estuarine system • provide surveillance for health alerts and warnings for showed signs of a system under stress. Algal blooms and toxic phytoplankton incidents. fish deaths in the Swan River and toxic blue-green blooms By addressing these objectives, and gaining sufficient in the Canning River focused community attention on the understanding of the system, a detailed Action Plan was deteriorating health of the system. In addition, water developed, entitled, The Swan Canning Cleanup Action samples collected between 1987 and 1992 indicated sub- Plan (SRT, 1999) and implementation commenced, also stantially elevated concentrations of phosphorus and in 1999. nitrogen compared with data measured by CSIRO in the 1940s (e.g. Rochford, 1951 and Spencer, 1956). These The data from the SCCP monitoring program has also symptoms were considered to be an indication of advanc- been made available to various researchers and students ing eutrophication and action was clearly needed to restore for collaborative and independent investigations. For ex- and protect the health of the rivers and estuary. ample, the 1996 data from the routine estuarine monitoring program have been used for the Land-Ocean Interactions The Swan Canning Cleanup Program (SCCP) was in the Coastal Zone (LOICZ) of the International Bio- launched in May 1994 by the State government to investi- sphere-Geosphere Programme (IGBP). This provided gate the causes of these water quality problems and develop information on the flux of dissolved nutrients into and out a program for the effective clean up of the Swan-Canning of the Swan River estuary over an annual cycle (Kalnejais system. The objectives of the first phase of water quality et al. 1999). The data is also being analysed with regard to sampling data (1994–1998) were to: establishing trigger values for estuarine water quality as • assess the current condition of the estuarine system; part of development of ANZECC guidelines by the Water and Rivers Commission. • develop an understanding of the major processes which determine the water quality; and
1 2. Objectives
The objective of this report is to present water quality chlorophyll concentrations. Ultimately, the study of water trends in the Swan River estuary based on the first 4.5 years quality of the Swan-Canning system will lead to increased of the SCCP monitoring data, May 1994 to December understanding of the link between eutrophication and 1998. Based on a program of weekly fixed site monitoring phytoplankton bloom dynamics. This subject will be ad- the most appropriate time scale for analysis is seasonal. dressed in a separate report. The spatial and temporal patterns in the water quality and The SCCP monitoring program also includes both Can- the major processes concerned with nutrient dynamics ning River/estuary and catchment sites, and these are within this system are discussed. This report also de- reported separately in Thomson et al., 2002. scribes the effect of estuarine hydrodynamics on water quality and how physical and chemical factors influence
2 3. Background
3.1 Previous monitoring 3.3 Bathymetry and artificial programs changes
For a number of years, estuarine water quality sampling From the ocean entrance at Fremantle the estuary is a on the Swan-Canning system was carried out on an ad hoc narrow winding channel 8 kilometres in length, cut in the basis often in response to algal bloom events. During Tamala limestone extending through to Blackwall Reach. 1992/93 a detailed study of water quality in relation to The harbour at Fremantle is dredged to 13 metres, but movement of the salt wedge (the ‘Wedge Watch’ project) from the Fremantle railway bridge to Rocky Bay the comprised the first intensive integrated sampling effort channel is only 5 metres deep. This forms a sill that (unpublished WRC data). partially separates the deep water of the harbour from deep water in the upper part of Blackwall Reach and the Catchment tributaries were sampled for nutrient concen- main estuarine basin. The main basin, Melville Water and trations between 1987 and 1992. These data, together with Freshwater Bay, is about 12 kilometres in length and 2 data up to 1998 from the current program have been kilometres wide; it has a maximum depth of 21 metres in analysed for trends in total phosphorus and total nitrogen Mosman Bay and from there shallows progressively to the (Donohue and Jakowyna, 2000). Narrow’s Bridge. There are extensive areas of shallow water, particularly on its southern and eastern shores. 3.2 Geology and climate Perth Water forms a second sill between the main basin and the riverine reaches of the upper estuary; most of it is The Swan-Canning system is a drowned river valley, about 1 metre deep, except for the dredged boat channels. adjacent to a semi-arid hinterland of low relief, draining a The upper estuary is mostly 2 metres to 3 metres deep to catchment area of about 124 000 km2 that has a seasonal, the Middle Swan Bridge, with depths to 5 metres in a few moderate to low rainfall (150-900 millimetres per year). places. The locations of the estuary and major freshwater tributar- Before the dredging of the Fremantle bar between 1892 ies are shown in Figure 1. and 1898, the Swan was mostly fresh and brackish with a The Swan Coastal Plain around Perth is over 100 000 tidal influence probably less than 20 percent of the marine years old. The surface sediments now comprise mainly tide. Since the removal of the bar, the dredging of the large sand with poor nutrient-binding capability. Estuarine con- flood delta to create the Port of Fremantle and subsequent ditions were established about 6 to 15 000 years ago by a dredging for the container terminal, approximately 80-85 sea level rise. The estuarine sediments vary from silty percent of the oceanic tide is transmitted up the estuary sand to black, sulphurous estuarine silt. Sublittoral sandflats (Jennings and Bird, 1967). Examination of historical sa- fringe much of the estuary shoreline. linity data for the Swan suggests that mean salinity has risen considerably and may still be rising. A large amount of fresh water from winter rains accumu- lates underground every year because of the porous, sandy soil. The depth of the groundwater is often less than five metres in low-lying areas. When the underground storag- es overflow, the groundwater flows into river courses. There are areas of contaminated groundwater that may impact on the water quality of the estuarine system, how- ever the quantity of groundwater entering watercourses is substantially less than that of winter surface water runoff.
3 Ellen Brook
Avon RIver
Susannah Brook
Jane Brook
Bennett Brook er Blackadder iv Main Drain R Creek Midland Mundaring Helena River an w Bayswater S Claise Brook Main Drain Main Drain South Belmont Main Drain Helena Reservoir
Mills St Main Drain Canning Yule Brook Bannister Creek Bickley Brook Fremantle River Southern River Canning River
Armadale Canning Reservoir
Wungong
Brook Wungong Reservoir
Figure 1: Map of Western Australia, showing the Swan River estuary, Canning River, major catchment tributaries and catchment water quality monitoring sites
3.4 Tides takes around 1 to 1.5 hours to propagate from the Fremantle Fishing Boat Harbour to Pier 21, a distance of The Swan River Estuary is now permanently open to the approximately 5 kilometres. The tide then takes Indian Ocean and experiences predominantly diurnal tides approximately one hour to pass through the sharp corner with a maximum tidal range of 0.6 – 0.9 metres. The tide at Preston Point at a distance of 1 kilometre. The time may be considered as a progressive wave that propagates taken for the wave to propagate from the fuelling jetty into the estuary. There is a 2.5 to 3 hour time lag between (approximately 400 m upstream of Preston Point) to the tide at the ocean entrance and at the Barrack Street Barrack Street is between 5 and 30 minutes depending on jetty (Lewis and Pattiaratchi, 1989). However, it is the tidal component being analysed (Lewis and Pattiaratchi, interesting to note that most of this lag occurs in the first 8 1989). The tide levels at Barrack Street are approximately kilometres of the estuary. For example, the incoming tide 85 percent of that at Fremantle Fishing Boat Harbour 4 indicating that the attenuation of the tide is approximately 3.6 Hydrodynamics 1 percent per kilometre (Lewis and Pattiaratchi, 1989). As with many shallow, permanently open estuaries, the Although the maximum tidal range is 0.6 metres, the circulation of estuarine water is maintained by the large recorded levels indicate much larger variations of up to variation in water density produced by the salinity differ- 1 metre which coincide with passage of atmospheric pres- ential between freshwater and oceanic water. The wind sure systems (Boardman et al., 1998). Hence, the exact may modify the circulation and become a dominant force behaviour of the estuarine water body to external mete- on occasions, but it is not responsible for the mean circu- orological forcing needs to be investigated further as this lation over extended periods of time, as is the case for forcing could have a large influence on the water circula- oceanic circulation patterns. The estuarine salinity field is tion and flushing characteristics. determined by the balance between advection by the mean In addition, there is the longer term (seasonal) changes to flow and turbulent tidal diffusion. The balance of these sea level which is reflected within the estuary. This amounts forces varies within and between years, however a distinct to an annual variation in the sea level of approximately 25 seasonal pattern characterises the Swan River estuary centimetres due to the passage of the Leeuwin Current. (Spencer 1956; Stephens and Imberger 1996).
Tidal currents dominate the lower estuary with the maxi- Throughout the year the Swan River estuary varies be- mum near-bed tidal currents of the order of 0.8 ms-1 tween a vertically stratified salt wedge estuary and a (Lewis and Pattiaratchi, 1989). These are generated due to slightly stratified estuary where the salinity increases the phase lag in the tidal levels in this part of the estuary. towards the sea but varies little with depth. When the Hence, this part of the estuary is very dynamic with a high vertical stratification is at its strongest, usually in spring, turbulent energy due to seabed friction and shear forces the estuary behaves approximately as a two-layered sys- between the oceanic water and freshwater flow. tem, comprised of a low salinity surface layer and high salinity lower layer which penetrates (as a salt-wedge) into the upper reaches of the Swan River and there is 3.5 Freshwater inflows limited exchange between the two layers. When the water column is completely mixed or with weak vertical stratifi- There is a strongly seasonal inflow of freshwater from the cation, salinity increases towards the sea but varies little catchment, with most rain falling in winter between May with depth and the two-layers can no longer be differenti- and September. Peak runoff occurs during late winter ated. This situation commonly occurs in mid to late summer whereby the saline estuarine water is flushed from the when freshwater flow is minimal and tidal exchange and upper estuary. In the lower estuary the freshwater dis- wind action dominate the hydrodynamics. charge forms a buoyant plume creating a low salinity or brackish surface layer. Occasionally, summer rainfall In winters of high river flow, which may occur once every events result in freshwater inflows during summer; other- 10 years (however, the last large flood was in 1971) the wise there are no river flows during the summer to early whole estuary is fresh, to a depth of about 5 metres. In the autumn months. main basin of the lower estuary, i.e. Freshwater Bay and Melville Water, at depths below 5 metres, it has been The major inflow contributions to the system are from the estimated that the salinity rarely falls below 24 ppt Avon River and Ellen Brook. The Canning River, on (Hodgkin 1987 and confirmed by present study). average constitutes only 7% of the Swan River flow. Groundwater inflows occur throughout the year, although As shown further in this report, the spatial and temporal they are a small fraction of the total freshwater inflows variation in the hydrodynamics has a significant influence (Appleyard, 1992). on water quality in all parts of the Swan Estuary.
5 4. Methods
4.1 Sampling sites Brook, Blackadder Creek, Jane Brook, Helena River, Bayswater Main Drain and South Belmont Main (Table 2). 4.1.1 Estuary (from Blackwall Reach None of the drainage systems that enter into the lower to Upper Swan) Swan are monitored. For the purposes of this study the Swan Estuary is divided into the Lower Estuary, from Blackwall Reach to the 4.2 Sampling program Narrows Bridge, the Middle Estuary from the Causeway to Maylands and the Upper Estuary, from Ron Courtney All parameters were collected at the nine estuary sites on Island to Success Hill. Nine sampling sites are located on a weekly basis, usually on a Monday between 8 a.m. and the Swan River Estuary, three in each region (Figure 2). 1 p.m. The Lower Estuary ranges from 4 to 20 metres in depth with deepest parts being in the channel from Blackwall 4.2.1 Physico-chemical data Reach to Point Walter. Site details including distance upstream of the mouth at Fremantle are listed in Table 1. Vertical profiles of the water column were taken in situ with a Hydrolab Multiprobe Logger (Hydrolab Corpora- 4.1.2 Catchment (Upper Swan) tion Austin Texas, USA). Data was logged at 0.5 m depth intervals from the surface to 5 m and thereafter every 1 m.
The major tributaries of the upper Swan Estuary are In 1994 and 1995 a Hydrolab H2O Datasonde was used to monitored for flow and nutrient concentrations, including collect depth, temperature (°C), salinity (ppt) and dis- the Avon River, Ellen Brook, Bennett Brook, Susannah solved oxygen readings (mgL-1). In 1996 a Hydrolab
T O NK IN
M
ITCHELL
River
Y HW Swan
H W N Y
EASTER
F W
Y
PERTH T EA R G
HWY HWY
G IN N N CA ALBANY STIRLING
H Canning W Y
River
HWY
LEACH
Fremantle
Figure 2: Water quality sampling sites for the Swan River estuary 6 Datasonde 3 was used to collect depth, temperature (°C), and turbidity (NTU) readings. From June 1997 onwards salinity (ppt), conductivity (ms/cm), dissolved oxygen the pH, salinity and oxygen sensors of the Hydrolab (mgL-1 and % saturation), pH (pH units) and turbidity Datasonde probe were calibrated before and after each readings (NTU). In 1997 and 1998 either a Hydrolab day of data collection against standard solutions or in the Datasonde 4 or Hydrolab Datasonde 3 was used to collect case of oxygen against atmospheric pressure. A thorough depth, temperature (°C), salinity (ppt), conductivity maintenance was carried out monthly on all sensors ex- (ms/cm), dissolved oxygen (mgL-1 or % saturation), pH cept the depth sensor, which is factory set. On the Hydrolab
Table 1: Swan River estuary site details
Site Site code Easting Northing Max. Depth Distance from (m) mouth (km)
Lower Blackwall Reach BLA 385040 6456658 18 7 Armstrong Spit ARM 387420 6458148 14.5 10 Narrows Bridge NAR 391039 6462854 6.5 18.5
Middle Nile Street NIL 394558 6463877 4 24 St John’s Hospital STJ 397075 6464275 6 29 Maylands Pool MAY 396692 6465550 4.1 31
Upper Ron Courtney Island RON 399673 6467701 7 35.5 Kingsley KIN 401532 6468358 7 37.5 Success Hill SUC 401479 6470192 6.2 40
Table 2: Swan River catchment site details
Catchment site Code Location
Bayswater Main Drain SWS10 Slade Street, Bayswater South Belmont Main Drain SWS11 Great Eastern Highway/Abernathy Road Avon River SWN4 Walyunga National Park Ellen Brook SWN3 Almeria Parade, The Vines Swan River SWN5 Great Northern Highway Bridge, Upper Swan Bennett Brook SWN1/12 Benara Road, Caversham Susannah Brook SWN6 Great Northern Highway, Herne Hill and River Road Susannah Brook SWN11 Great Northern Highway, Herne Hill and River Road Susannah Brook East SWS9 West Swan Road, Belhus Jane Brook SWN7 Sweeting Bridge, Great Northern Highway Jane Brook SWS8 John Forrest National Park Helena River SWN10 Whiteman Road Bridge Helena River SWN9 Valley Road, Craignish Blackadder Creek SWN8 Francis Street, Middle Swan Henley Brook SWN2 West Swan Road, Belhus
7 Datasonde 3 a salt water cell block was used to determine since 1996. Samples that required analysis for dissolved salinity and conductivity. On the Hydrolab Datasonde 4 species were filtered immediately after collection on board one cell block is sufficient for both fresh and salt water the boat, using a 0.45 mm cellulose nitrate membrane. readings. The conductivity and salinity is compensated TSS and DOC were analysed in the surface samples only. for temperature, and dissolved oxygen readings are com- Infrequently, samples were also analysed for other species pensated for salinity. such as total kjeldahl nitrogen, total organic carbon, ni- trite and selenium. In all cases analyses followed the Light penetration depths were measured at each site using methods of the American Public Health Association a 30 cm diameter, black and white Secchi disk. Cloud (APHA, 1989). cover, tidal influences and wind speed and direction were also recorded as well as any obvious water quality condi- tions. 4.2.3 Chlorophyll
4.2.2 Nutrients At each estuary site, between 0.5–1 L of the surface and bottom water samples were filtered through glass micro Weekly water samples from the surface and bottom of the fibre (Whatman GF/C) filter papers for chlorophyll a, b water column or as an integrated sample over the whole and c and phaeophytin pigment analyses. Filter paper water column are analysed for the major nutrient species. samples were analysed using standard methodology Analyses and audits were undertaken by Australian Envi- (APHA, 1989). ronmental Laboratories (AEL) for the following Data for most of the parameters collected were analysed parameters: for seasonal trends using box and whisker plots on a - - Total nitrogen (TN), nitrate (NO3 ), nitrite (NO2 ), ammo- monthly basis for each of the 3 spatial groupings (Lower, + nium (NH4 ), total phosphorus (TP), filterable reactive Middle and Upper). For an explanation of the symbols phosphorus (FRP), total suspended solids (TSS), silica and coefficients used in box and whisker plot presenta-
(SiO2), dissolved organic carbon (DOC) and alkalinity tions, see Appendix A.
8 5. Results
5.1 Physico-chemical parameters By late spring, the freshwater inflows are significantly reduced and the tidal movement again results in intrusion 5.1.1 Freshwater inflows of saline waters upstream. Gradually, over a typical dry summer the tidal and wind action continue to breakdown The Avon River accounts for more than approximately the stratification and increase the salinity throughout the 81% of the Swan River inflows and Ellen Brook approxi- estuary (Figure 4, December 1 profile). By the end of mately 9%. The discharge for these gauged stations from summer the slightly stratified, March-type profile is re- 1994 to 1998 is shown in Figure 3. The Canning River established. discharge into the lower estuary constitutes approximately After the winter rains subside, there is a typical pattern of 7% of the Swan River inflow volumes. saline water intruding into the upper reaches of the Swan estuary in spring as shown in Figure 4. Typically, the low salinity (< 4 ppt), catchment derived water occurs through- 5.1.2 Salinity out the vertical profile as far downstream as Nile Street in most winters. In relatively wet years, this may occur over Using salinity as a conservative tracer it is possible to a period of months, for example in 1996, the water column describe the relative influence of the tidal exchange and was homogeneous and low in salinity from late July to freshwater inflows on the hydrodynamic behaviour mid October. During high winter flow years, the low throughout the year. salinity layer may extend as far downstream as Blackwall At the end of summer, there is a longitudinal gradient Reach to a depth of 5 m, however high salinity conditions along the length of the estuary, ranging from 22 ppt at prevail in the deeper waters below 5 m. Success Hill, some 40 km from the mouth, to > 36 ppt in the lower estuary (Figure 4). During such times and ac- 5.1.3 Dissolved oxygen cording to hydrodynamic classification schemes, the estuary can be classified as a ‘slightly stratified’ estuary. Surface waters were well oxygenated in all regions and In the absence of freshwater inflows, tidal exchange re- throughout the year (Figure 5). Median values in surface sults in the horizontal transport of salt along the entire waters of the upper Swan were supersaturated (> 100%) length. from December to May, reflecting phytoplankton activity.
In mid-July, after the commencement of the winter rain in The median percent saturation of dissolved oxygen (%DO) the catchment, freshwater enters the upper estuary and in bottom waters for each of the three regions of the forms a buoyant plume, in the upper and middle estuary. estuary, indicates a spatially variable pattern of hypoxic The change from approximately vertical isohalines to events over the annual cycle (Figure 5). In the lower mostly horizontal and the development of two distinct estuary, median values for the bottom waters indicated layers (surface and bottom) leads to the estuary being well-oxygenated conditions (> 70%) from December to characterised as ‘highly stratified’. There is likely to be June (Figure 5). From July on, median values declined strong shear at the interface, because the tidal currents are until a minimum 31% was reached in September. large relative to the freshwater flow. Strong turbulence at In contrast, the median %DO values in the bottom layer the interface results in the vertical flux of salt to the upper for the middle and upper estuary during July to September layer so there is a fairly strong longitudinal gradient in were comparable to the well-oxygenated surface layer salinity in the surface layer. (Figure 5). This is the time when the middle and upper At the end of winter/early spring, the freshwater inflows reaches are most likely to be well flushed with catchment (< 4 ppt) are relatively large compared to the tidal volume derived freshwater inflows and the water column is well- and have almost completely flushed the saline and brack- mixed (or homogeneous) over the depth. In the middle ish waters from the upper and middle estuary. The and upper regions, hypoxic events usually occurred be- isohalines between 4 ppt and 30 ppt are now much more tween October and June. Low median %DOs (< 50%) compressed (see the September 8 profile, Figure 4). This occurred in the middle estuary in June, November and situation is the typical ‘salt wedge’ estuarine condition. December and with greater frequency in the upper estu- Water exchange between the two layers is restricted by ary, between January to June and in December. The the strong stratification and mixing is restricted to the minimum median %DO was 14% in June, in the upper frontal zone, which is governed by intense turbulence. estuary. 9 140
120
100
eek)
80
(GL per w 60
40
Discharge
20
0
7 Jun 95 7 Jun
6 Jun 98 6 Jun
6 Jun 96 6 Jun
7 Jun 94 7 Jun
6 Jun 97 6 Jun
8 Mar 95
7 Mar 98
7 Mar 96
8 Mar 94 7 Mar 97
7 Dec 95
6 Sep 95
6 Dec 98
5 Sep 98
6 Dec 96
5 Sep 96
7 Dec 94
6 Dec 97
6 Sep 94
5 Sep 97
Figure 3: Swan River discharge (Avon and Ellen Brook flows)
BLA ARM NAR NIL STJ MAY RON KIN SUC 0 34 26 36 -5
-10
-15 17 March 1997
0
26 18 30 14 -5
(m) -10
-15 14 July 1997 0
om surface 26 22 1814 10 8 4 -5 30 30
Depth fr -10
-15 8 September 1997 0 22 14 10 30 26 18 8 -5 34
-10 Salinity (ppt)
-15 048 12162024283236 1 December 1997
10 15 20 25 30 35 40 Distance from Fremantle (km) Figure 4: Annual hydrodynamic cycle, 1997 10 200 200 Lower Estuary Surface Middle Estuary Surface Bottom Bottom
ation) ation) 150 150
(% satur (% satur 100 100
xygen xygen
50 50
Dissolved o 0 Dissolved o 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Upper Estuary Surface 200 Bottom
ation) 150
(% satur 100
xygen
50
Dissolved o 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Figure 5: Dissolved oxygen (% saturation) monthly medians for the surface and bottom waters of the lower, middle and upper estuary
BLA ARM NAR NIL STJ MAY RON KIN SUC 0
-5 60
-10
-15 17 March 1997
0
70 60 40 1020 -5
(m) -10
-15 14 July 1997 0
om surface 70 60 20 40 60 70 -5
Depth fr -10
-15 8 September 1997 0
40 80 60 20 -5
60 -10 Dissolved oxygen (%)
-15 0 20 40 60 80 100 120 140 160 1 December 1997
10 15 20 25 30 35 40 Distance from Fremantle (km)
Figure 5a: Annual dissolved oxygen dynamics, 1997 11 The dissolved oxygen profiles corresponding to the salinity 5.1.4 Temperature profiles (Figure 4) illustrate how the hydrodynamics affect The pattern in seasonal variation in water temperature was the oxygen field (Figure 5a). The key feature here is that similar for all estuarine regions. In the lower, middle and areas of extremely low oxygen levels (< 20%) correspond upper estuary surface median temperatures ranged from to the patches of water below areas of severe vertical approximately 13°C in winter to summer highs of 24°C, stratification; for example, from STJ to KIN in July, and 26.5°C and 28°C (respectively) (Figure 6). The bottom around NAR (Narrows Bridge) in September. In these median temperatures ranged from approximately 13–16°C oxygen depleted areas, the severity of the stratification to 24–27°C for all regions. The greatest annual variation was of the order of 6 pptm-1 and higher and they were occurred in the upper estuary with the lower estuary associated with the leading edge of the salt wedge. Another having the least variation. important feature during spring and early summer is that the saline intrusion transports low oxygen waters upstream in the bottom waters (8 September and 1 December 5.1.5 pH profiles). Median pH values in the lower estuary ranged from 7.7 to In March, in the slightly stratified condition, the entire 8.2 (surface) and 7.8 to 8.0 (bottom) (Figure 7). The estuary was well oxygenated, although there was a slight seasonal variation in median values for both the surface longitudinal gradient of decreasing oxygen saturation in and bottom waters was very similar. In the middle estuary, the upstream direction. This can be attributed to the weak the magnitude of surface median pH values (7.6 to 8.1) salinity gradients, which persist in the upper estuary due was comparable to the lower estuary, however bottom to reduced tidal action. values (7.3 to 7.7) were generally lower. A similar pattern The localised, supersaturated (>100%) surface patches at was observed for the upper estuary, however bottom BLA, RON and NIL were most likely caused by phyto- values were lower again ranging from 7.1 to 7.6. It is also plankton bloom activity (Figure 5a, green coloured areas). worth noting that for the middle and upper sites, the closest correspondence in surface and bottom pH medians occurred in August and September, when freshwater flow was at a maximum.
32 32 30 Lower Estuary 30 Middle Estuary 28 28 26 26 (°C) 24 (°C) 24 22 22 20 20 18 18 16 16
Temperature Temperature 14 Surface 14 Surface 12 Bottom 12 Bottom 10 10 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
32 30 Upper Estuary 28 26 (°C) 24 22 20 18 16
Temperature 14 Surface 12 Bottom 10 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Figure 6: Monthly median temperatures in the surface and bottom waters of the lower, middle and upper estuary 12 9.0 9.0 8.8 Lower Estuary 8.8 Surface Middle Estuary 8.6 8.6 Bottom 8.4 8.4 8.2 8.2 8.0 8.0
pH 7.8 pH 7.8 7.6 7.6 7.4 7.4 7.2 Surface 7.2 7.0 Bottom 7.0 6.8 6.8 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
9.0 8.8 Surface Upper Estuary 8.6 Bottom 8.4 8.2 8.0
pH 7.8 7.6 7.4 7.2 7.0 6.8 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Figure 7: Monthly median pH in the surface and bottom waters of the lower, middle and upper estuary
6 6 Lower Estuary Middle Estuary 5 5
4 4
(m) (m) 3 3
sd
sd
Z
Z 2 2
1 1
0 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
6 Upper Estuary 5
4
(m) 3
sd
Z 2
1
0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Figure 8: Monthly median Secchi depths in the lower, middle and upper estuary
13 5.1.6 Secchi depth 5.3 Nutrients: Seasonal patterns Secchi depth is an approximate measure of the depth to 5.3.1 Nitrogen which light penetrates the water column. Low Secchi depths are usually associated with high total suspended In this section we consider the dissolved nitrogen species sediment (TSS) or algal blooms, however a range of (nitrate, nitrite and ammonium) collectively referred to as substances may affect the attenuation of light in the aquatic dissolved inorganic nitrogen (DIN); and total nitrogen environment, for example tannins and gilvin. (TN) which includes dissolved organic nitrogen (e.g. Urea), particulate organic nitrogen (e.g. phytoplankton) and DIN. In terms of Secchi depth, data from the Narrows site most DIN is the most bioavailable component of nitrogen for closely resembled the pattern for the middle estuary and uptake by aquatic plants and is a significant source of therefore, for this parameter it has been grouped with the nitrogen for phytoplankton growth whilst total nitrogen middle estuary sites and the monthly medians for the often reflects phytoplankton density. lower estuary are derived from Blackwall Reach and Armstrong sites only. In the lower estuary, Secchi depths were of the order of 3– 4 m from November to June, and Dissolved inorganic nitrogen from July to October median values were significantly In the lower estuary, the median concentration of DIN less, from 1–2 m (Figure 8). During the month of May at during summer and autumn (December through to May) Blackwall Reach, Secchi depths were often in excess of varied from 0.02–0.03 mgL-1 in both surface and bottom 5m waters (Figure 9). Median concentrations rose in June and The middle and upper regions behaved similarly, with a reached a maximum of 0.61 mgL-1 in July in the surface maximum median Secchi depth of ~1.5 m in May and waters and 0.27 mgL-1 in August in the bottom waters. minimum depths of ~0.8 m from August to December The median concentrations decreased back to minimum (Figure 8). These data show that there was greater light values of approximately 0.02 mgL-1 in October in surface attenuation in the middle and upper estuary in general and waters and by December in the bottom waters. particularly during the winter and spring months when In the middle and upper estuary, the pattern and magni- catchment derived, tannin-stained water is typically present tude of monthly median concentrations of dissolved in all estuary sites. inorganic nitrogen were similar to those for the surface layer of the lower estuary (Figure 9). Compared with the 5.2 Water quality data: lower estuary surface data, there were two notable differ- ences with the middle and upper estuary DIN data. Firstly, Summary statistics the median concentrations in April and May for the mid- dle and upper estuary were slightly higher than those for The total number of samples, the mean, median, mini- the lower estuary surface waters and, in general, there was mum and maximum concentrations of the measured water greater range in values for the middle and upper estuary. quality parameters over the study period are summarised for each of the 3 regions of the Swan River estuary The elevated surface median DIN concentrations observed
(Table 3). For nearly all of the water quality parameters, in winter were almost entirely dominated by the NOx the median values tended to increase from the lower to component (see Figure 10). In contrast, median concen- upper regions. There were two exceptions, where the trations in the bottom waters, from November through to + median values were highest in the middle region, these May, tended to have a more significant NH4 component. + were, ammonium and filterable reactive phosphorus. Apart from the slight winter peak in NH4 in the lower Within each region, the surface median concentrations estuary, there was no discernible seasonality in the monthly + also tended to be higher in the surface sample, again with median NH4 concentrations, however they did tend to be the exception of ammonium, FRP and TP, which were higher for bottom waters than for the surface. higher in the bottom samples for all 3 regions.
14 Table 3: Water quality: Swan River estuary summary statistics (N = total number of samples, units for all parameters are mgL-1)
Parameter Layer N Mean Median Min Max Lower Estuary
NOx surface 553 0.120 0.020 0.0030 2.20 bottom 554 0.050 0.020 0.0030 1.40
NH4 surface 556 0.030 0.020 0.0020 0.36 bottom 557 0.060 0.030 0.0030 0.70 Org-N surface 417 0.440 0.350 0.0200 1.50 bottom 418 0.280 0.240 0.0300 1.60 TN surface 556 0.550 0.370 0.0300 2.90 bottom 557 0.340 0.280 0.0100 3.00 FRP surface 556 0.020 0.020 0.0000 0.11 bottom 557 0.030 0.030 0.0020 0.10 TP surface 556 0.050 0.040 0.0100 0.31 bottom 557 0.050 0.040 0.0100 0.32 Chl a surface 532 0.005 0.002 0.0005 0.44 bottom 533 0.005 0.001 0.0003 0.53
SiO2 surface 293 3.900 3.900 0.0500 14.00 bottom 291 2.890 2.800 0.0500 9.20 DOC surface 142 9.570 6.900 0.5000 44.00 TSS surface 432 9.600 7.000 0.5000 41.00 Middle Estuary
NOx surface 636 0.160 0.030 0.0030 2.80 bottom 636 0.130 0.020 0.0030 2.70
NH4 surface 639 0.070 0.040 0.0030 0.61 bottom 638 0.110 0.080 0.0030 0.98 Org-N surface 498 0.840 0.750 0.1000 7.20 bottom 497 0.720 0.680 0.0900 2.80 TN surface 638 0.980 0.800 0.1100 7.20 bottom 638 0.840 0.710 0.1000 4.40 FRP surface 639 0.040 0.030 0.0020 0.14 bottom 638 0.050 0.040 0.0000 0.41 TP surface 639 0.090 0.080 0.0100 0.93 bottom 638 0.110 0.100 0.0100 0.74 Chl a surface 614 0.013 0.007 0.0002 0.41 bottom 614 0.005 0.002 0.0003 0.55
SiO2 surface 379 5.850 5.700 0.2000 17.00 bottom 372 5.490 5.000 0.6000 23.00 DOC surface 223 14.590 11.000 0.5000 51.00 TSS surface 514 11.800 10.000 1.3000 87.00 Upper Estuary
NOx surface 633 0.160 0.030 0.0030 3.10 bottom 629 0.140 0.020 0.0030 4.90
NH4 surface 634 0.060 0.030 0.0030 1.20 bottom 632 0.120 0.060 0.0030 2.00 Org-N surface 491 0.970 0.860 0.2300 11.00 bottom 487 0.850 0.780 0.2000 4.00 TN surface 636 1.140 0.950 0.3000 11.00 bottom 631 1.000 0.850 0.2300 5.90 FRP surface 634 0.030 0.020 0.0020 0.18 bottom 633 0.050 0.030 0.0020 1.80 TP surface 636 0.100 0.080 0.0100 1.70 bottom 632 0.130 0.100 0.0100 2.60 Chl a surface 609 0.020 0.012 0.0003 0.36 bottom 606 0.007 0.003 0.0003 0.17
SiO2 surface 380 6.850 6.700 0.1000 16.00 bottom 370 6.450 6.250 0.4000 15.00 DOC surface 220 16.280 13.000 1.6000 49.00 TSS surface 508 11.900 9.800 1.0000 130.00
15 2.0 2.0 Lower Estuary Surface Lower Estuary Bottom
1.5 1.5
(mg/L) 1.0 (mg/L) 1.0
DIN DIN
0.5 0.5
0.0 0.0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
*¹ 2.0 2.0 *¹ Middle Estuary Surface Middle Estuary Bottom
1.5 1.5
(mg/L) 1.0 (mg/L) 1.0
DIN
DIN
0.5 0.5
0.0 0.0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
2.0 *² 2.0 *³ Upper Estuary Surface Upper Estuary Bottom
1.5 1.5
(mg/L) 1.0 (mg/L) 1.0
DIN
DIN
0.5 0.5
0.0 0.0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Figure 9: Monthly median concentrations of dissolved inorganic nitrogen (DIN) in the surface and bottom water of the lower, middle and upper estuary (*1 – outliers maximum 3.2 mgL-1, *2 – maximum 3.4 mgL-1, *3 – maximum 3.9 mgL-1)
Total nitrogen the Avon River catchment site (Figure 12). However, maxima were attained in June in the catchment flows, one In general, TN monthly medians were slightly higher in month earlier than observed in the estuarine data (Fig- the surface compared with bottom waters (most pro- ure 9). The correspondence between the estuary freshwater nounced in the lower estuary) and increased from the layer and upper estuary catchment flows is an expected lower to upper estuary (Figure 11). The seasonal variation result given that the upper Swan inflows contribute more followed a similar pattern to DIN with peaks attained than 80% of the total catchment inflows (SRT, 2000). during July and August. DIN concentrations accounted Ellen Brook, which flows from late autumn to midsum- for approximately 50% of TN during winter and some- mer, had the highest total nitrogen values recorded in the what less during summer. current study, from 1.1 to 2.6 mgL-1 (Figure 12). The dominant fraction was organic nitrogen and DIN median Catchment inflows values were lower than the Avon River site. The seasonal pattern for total and dissolved inorganic After the upper Swan, the major additional inflows to the nitrogen observed in estuarine sites was also observed in middle estuary are Bayswater Main Drain and South 16 2.0 2.0 Lower Estuary Surface Lower Estuary Bottom
NH _S NH _B 1.5 4 1.5 4 NOX _S NOX _B
(mg/L) (mg/L)
1.0 1.0
X4 X4 0.5 0.5
NO and NH NO and NH
0.0 0.0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
2.0 2.0 Middle Estuary Surface Middle Estuary Bottom
NH _S NH _B 1.5 4 1.5 4 NOX _S NOX _B
(mg/L) (mg/L)
1.0 1.0
X4 X4 0.5 0.5
NO and NH NO and NH
0.0 0.0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
2.0 2.0 Upper Estuary Surface Upper Estuary Bottom
NH _S NH _B 1.5 4 1.5 4 NOX _S NOX _B
(mg/L) (mg/L)
1.0 1.0
X4 X4 0.5 0.5
NO and NH NO and NH
0.0 0.0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
+ Figure 10: Monthly median concentrations of NOx and NH4 in the surface and bottom waters of the lower, middle and upper estuary
Belmont Main Drain. These drains are both urban and 5.3.2 Phosphorus industrial, and flow all year round. The median TN con- Similarly to nitrogen, phosphorus can be separated into centrations for Bayswater ranged from 1.6–1.9 mgL-1 the bioavailable component, known as phosphate or more with DIN as the dominant fraction (Figure 13). South precisely, filterable reactive phosphorus (FRP), and Belmont total nitrogen medians ranged from 0.8 to 1.2 particulate phosphorus which may be in the form of with the organic-N and DIN fractions approximately equal phytoplankton or bound to sediments. in winter whilst organic-N was dominant in summer (Fig- ure 13). These data illustrate the variable nature of the catchment inflows. Further, the lack of association with winter rainfall suggests the high nutrient concentrations in these urban/industrial tributaries cannot be attributed to diffuse surface runoff, clearly there is another source in this region.
17 4 Lower Estuary Surface 4 Lower Estuary Bottom
3 3
(mg/L) 2 (mg/L) 2
TN TN
1 1
0 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
4 Middle Estuary Surface 4 Middle Estuary Bottom
3 3
(mg/L) (mg/L) 2 2
TN
TN
1 1
0 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
4 Upper Estuary Surface 4 Upper Estuary Bottom
3 3
(mg/L) 2 (mg/L) 2
TN TN
1 1
0 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Figure 11: Monthly median total nitrogen (TN) concentrations in the surface and bottom water of the lower, middle and upper estuary
5 5 Avon River 94-98 TN Ellen Brook 94-98 TN DIN DIN 4 ORG-N 4 ORG-N
3 3
(mg/L) 2 (mg/L) 2 N N
1 1
0 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Figure 12: Monthly median TN, DIN and organic-N concentrations in the major catchment inflows to the upper estuary 18 5 5 Bayswater 94-98 TN South Belmont 94-98 TN DIN DIN 4 ORG-N 4 ORG-N
3 3 (mg/L) (mg/L) 2 2 N N
1 1
0 0 Jan Feb Mar AprMay Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Figure 13: Box plots for TN, DIN and organic-N in the major catchment inflows to the middle estuary
Filterable reactive phosphorus Catchment inflows The seasonal pattern in FRP was quite different to that of The character of the phosphorus concentrations in the dissolved inorganic nitrogen. In the lower estuary, from two major inflows to the upper estuary, Avon and Ellen November to August, the pattern of monthly median Brook are highly contrasting. TP concentrations in Ellen concentrations was similar for both the surface and bot- Brook were an order of magnitude higher than in the tom, ranging from 0.01 to 0.03 mgL-1 (Figure 14). In Avon, and whilst FRP was the dominant fraction in Ellen September and October, median concentrations were Brook, the converse was true for Avon (Figure 16). Ellen higher in the bottom waters; otherwise there was no Brook FRP figures were alarmingly high, median values apparent seasonal variation. 0.2–0.4 mgL-1, substantially higher than all other catchment inflows (SRT 2000 unpublished report), all estuary sites In the middle estuary, the distribution of monthly medians and ten times the ANZECC/ARMCANZ 2000 trigger for FRP was similar for both surface and bottom waters, level (see Appendix B). however median values for bottom waters were, in gen- eral slightly higher than surface values (Figure 14). Median TP concentrations for Bayswater ranged from Compared with the lower estuary, middle estuary medi- 0.07 mgL-1 to 0.13 mgL-1 in June with virtually all being in ans were elevated especially for the months February to the particulate form (Figure 17). Although the flow is May. In addition, there was greater range in the data, negligible in winter compared with Avon and Ellen flows, especially for bottom waters. summer flows could account for TP input at this time, however it is clearly not a direct FRP source to the The pattern for FRP in the upper estuary was very similar estuary. Median TP concentrations in South Belmont to the middle estuary, the major difference being in- were approximately half those of Bayswater, however, creased median values in the bottom waters from February the dissolved (FRP) and particulate forms were compara- to July and greater variation in the data, especially for the ble throughout the year (Figure 17). bottom values (Figure 14).
Total phosphorus 5.3.3 Silica Silica (SiO ) is an essential nutrient element for phyto- The total phosphorus (TP) annual patterns were highly 2 comparable to FRP. The FRP fraction accounted for ap- plankton that have siliceous skeletons, for example diatoms and benthic phytoplankton. Silica is the only form avail- proximately 50% of the total phosphorus. There was a able for diatom growth. general trend of increasing TP median values from the lower to upper estuary; slightly higher values in the bot- In the lower estuary, both the surface and bottom waters tom waters of the middle and upper estuary compared had high silica concentrations in August. The surface with surface values and greater variability in the data water monthly medians ranged from 2 mgL-1 in May to (Figure 15). 7.2 mgL-1 in August (Figure 18). Corresponding data for bottom waters were substantially lower and ranged from 1 mgL-1 in July to 5 mgL-1 in September. 19 0.20 0.20 Lower Estuary Surface Lower Estuary Bottom
0.15 0.15
(mg/L) 0.10 (mg/L) 0.10
FRP FRP 0.05 0.05
0.00 0.00 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
0.20 0.20 Middle Estuary Surface Middle Estuary Bottom 0.15 0.15
(mg/L) 0.10 (mg/L) 0.10
FRP FRP 0.05 0.05
0.00 0.00 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
0.20 0.20 Upper Estuary Upper Estuary Surface Bottom 0.15 0.15
(mg/L) (mg/L) 0.10 0.10
FRP
FRP 0.05 0.05
0.00 0.00 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Figure 14: Monthly medians for filterable reactive phosphorus (FRP) in the surface and bottom water of the lower, middle and upper Swan River estuary [NB. In order to keep scaling uniform, some outliers extend outside plot range, e.g. middle, bottom maximum = 0.4 mgL-1 and upper, bottom maximum =0.7 mgL-1].
In the middle estuary, median silica concentrations for surface and bottom waters ranged from 2 mgL-1 in 5.4 Chlorophyll: November to 8.2 mgL-1 in August. Seasonal patterns In the upper estuary for both surface and bottom waters, There is a clear spatial pattern in chlorophyll a concentra- the distribution of silica data were also similar in magni- tions, from relatively low concentrations in the lower tude and distribution. The surface and bottom waters estuary, maximum median 0.004 mgL-1, to 0.016 mgL-1 in -1 -1 ranged from 3 mgL (December) to 8 mgL (April and the middle and 0.02 mgL-1 in the upper (Figure 19). The August). seasonality also varied between regions. Peak chlorophyll Although a distinct seasonal pattern was not evident in the a values occurred in spring in the lower estuary, in late silica data, maximum median values tended to occur in spring and autumn in the middle and in late spring, sum- late winter/spring in the lower and middle estuary and in mer and autumn in the upper estuary. The lowest late summer/autumn and in winter for the upper estuary. chlorophyll a medians were recorded in July to September Most of the regions recorded the highest concentrations of for both the middle and upper estuary. silica in August. 20 0.5 0.5 Lower Estuary Surface Lower Estuary Bottom 0.4 0.4
0.3 0.3
(mg/L) (mg/L) 0.2 0.2
TP TP
0.1 0.1
0.0 0.0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
0.5 0.5 Middle Estuary Surface Middle Estuary Bottom 0.4 0.4
0.3 0.3
(mg/L) (mg/L) 0.2 0.2
TP TP
0.1 0.1
0.0 0.0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
0.5 0.5 Upper Estuary Surface Upper Estuary Bottom 0.4 0.4
0.3 0.3
(mg/L) (mg/L) 0.2 0.2
TP TP
0.1 0.1
0.0 0.0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Figure 15: Box plots for total phosphorus (TP) in the surface and bottom water of the lower, middle and upper estuary
1.0 1.0 Avon River 94-98 FRP Ellen Brook 94-98 FRP TP TP PART-P 0.8 0.8 PART-P
0.6 0.6 (mg/L) 0.4 (mg/L)
P 0.4 P
0.2 0.2
0.0 0.0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Figure 16: Box plots for TP, FRP and particulate-P in the major catchment inflows to the upper estuary 21 1.0 1.0 Bayswater MD FRP South Belmont MD 94-98 FRP TP TP 0.8 PART-P 0.8 PART-P
0.6 0.6 (mg/L) (mg/L) P
P 0.4 0.4
0.2 0.2
0.0 0.0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Figure 17: Box plots for TP, FRP and particulate-P in the major catchment inflows to the middle estuary
16 16 Lower Estuary 97-98 Surface Middle Estuary 97-98 Surface 14 14 1 m 1 m 12 Bottom 12 Bottom 10 10
(mg/L) 8 (mg/L) 8
24 6 24 6
H SiO H SiO 4 4 2 2 0 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
16 Upper Estuary 97-98 Surface 14 1 m 12 Bottom 10
(mg/L) 8
24 6
Figure 18: Concentration of silica in the surface, H SiO 1 m and bottom waters of the lower, 4 middle and upper Swan River estuary, 2 0 for 1997 and 1998 data only Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
0.02 0.06 Lower Estuary CHLA_S Middle Estuary CHLA_S CHLA_B 0.05 CHLA_B
0.04
(mg/L) (mg/L) 0.01 0.03 a a
Chl Chl 0.02
0.01
0.00 0.00 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
0.06 Upper Estuary CHLA_S 0.05 CHLA_B
0.04
(mg/L) 0.03 a
Figure 19: Chlorophyll a monthly medians in the Chl 0.02
surface and bottom waters of the 0.01 lower, middle and upper estuary (NB: 0.00 different y-axis for lower estuary) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 22 6. Discussion
The Swan River estuary is eutrophic and becomes more Hosja, 1996). Nutrient ratios in the water column are eutrophic with distance upstream. This is generally attrib- another way to determine the potential for nutrient limita- uted to the substantial clearing of the catchment for urban, tion. Phytoplankton generally need a ratio of 16:1 (molar) industrial and agricultural land uses. All water quality nitrogen to phosphorus (Redfield, 1934). If bioavailable parameters in the system have a high degree of spatial and nutrients in the water column are less than this ratio, then temporal variability. The system experiences a wide range nitrogen is considered potentially limiting, or if greater of salinities, occasions of intense vertical stratification than 16:1 phosphorus is potentially limiting. In the fol- and consequent sub-halocline hypoxia, and decreased water lowing discussion on seasonal variations in water quality clarity with distance upstream. The greatest light attenua- dynamics, we consider the nutrient limitation potential of tion is associated with tannin-rich catchment inflows. estuarine waters. Spatially, the lower estuary water quality patterns were in general distinct from the middle and upper estuary. Tem- porally, seasonal differences showed the greatest source Winter dynamics of variation. The most dominant nutrient pattern is the annual, winter- Hydrodynamic and hydrological forcing largely drives time delivery of nitrate-rich waters from the catchment the broadscale temporal and spatial variability within the throughout the upper and middle estuary and extending estuary. The highly seasonal freshwater inflows from down into the surface of the lower estuary. This all but winter rainfall and the tidal saline-water intrusion are the obscures any other pattern in nutrients at this time. Nitrate key factors in determining nutrient delivery and distribu- delivery is clearly related to winter catchment flows (this tion. Provided that the rainfall pattern is comparable to the study and from 1993 to 1994 data, Thompson and Hosja, average in magnitude and distribution, seasonal differ- 1996). The winter season is relatively short and the ‘win- ences are reasonably consistent from year to year. ter’ dynamics generally relate to July and August only.
The concept of nutrient limitation, first applied in studies Phosphate (FRP) concentrations did not display the same of crop plant productivity, is increasingly used in algal winter peak distribution. Although Ellen Brook has a high bloom investigations. The limiting nutrient to the biomass FRP component, with a median midwinter maximum of of phytoplankton is always the nutrient in the lowest 0.4 mgL-1, this contribution would have been diluted concentration relative to the demand for other nutrients. when combined with the Avon flows, which is around In most estuaries around the world nitrogen has been 0.01 mgL-1 FRP and probably adsorbed onto suspended identified as the major nutrient limiting annual produc- sediments. FRP median concentrations were always higher tion, also shown in studies of Wilson Inlet, Moreton Bay, in bottom waters in the upper and middle estuary, which Queensland, and Port Phillip Bay, Victoria (Harris et al., suggests the sediments/groundwater is a greater source of 1996). FRP than freshwater inflows. In the Swan River, the only conclusive study on nutrient Comparing the molar ratio of the major dissolved nutri- limitation showed, through bioassay experiments that ni- ents in winter, the data were either close to, or much trogen had the greatest potential to limit biomass in the greater than the Redfield ratio illustrating the relative Swan during the study period, 1993-1994 (Thompson and abundance of nitrogen (Figure 20).
SURFACE – Winter (Jul - Aug) BOTTOM – Winter (Jul-Aug) 200 200 lower lower middle middle upper 160 upper 160
M) 120 120 M) µ µ ( ( Redfield [16:1]
DIN Redfield [16:1] 80 DIN 80
40 40
0 0 0246810 0246810 FRP (µM) FRP (µM) Figure 20: DIN plotted against FRP during winter (Jul-Aug) in the three regions of the estuary 23 During winter in the middle and upper estuary, chloro- waters below the halocline. As the salt wedge moves phyll a concentrations were at their annual lows and this upstream, the hypoxic bottom layer is also transported region has been shown to have exceptionally low net upstream. Hypoxic conditions result in the release of primary production (Thompson, 1998). Physical condi- bioavailable nutrients, ammonium and phosphate from tions in winter include high light attenuation due to gilvin, the sediments and are occasionally associated with fish minimum temperatures, short day lengths, minimum irra- death incidents. For these reasons, a trial project to diance, and short water residence times. It is therefore safe destratify the salt wedge using a bubble curtain at Maylands to conclude that one or more of these factors counteracts was undertaken in 1997. This proved to be largely unsuc- any potential benefit of abundant, bioavailable nitrogen to cessful because the salt wedge was found to be highly the phytoplankton in the middle and upper estuary at this mobile, moving kilometres in a day rather than hundreds time. of meters as earlier reported, which was partly explained by its dependence on barometric pressure. These features In the winter the short residence times of the water ensure were unaccounted for in the laboratory design experiment that the middle and upper estuary is well flushed, without that lead to the trial (SRT, 1998). stratification and well oxygenated. In contrast the lower estuary becomes stratified and consequently the bottom Many biological processes are partly a function of tem- waters are often oxygen depleted, particularly at the Nar- perature particularly for plants and microbes. For example, rows. High ammonium concentrations observed in the when plant cells uptake nitrate, they must first reduce it to lower estuary bottom waters is a likely consequence of ammonium for assimilation into organic molecules. An these hypoxic conditions. Further downstream, at enzyme, nitrate reductase (NR), mediates this process Blackwall Reach tidal exchange was sufficient to main- within the plant cells.
tain reasonable oxygen levels (> 60%). - Nitrate - Nitrate + NO3 — reductase —> NO2 — reductase ——> NH4 ——> Amino Acids So what is the fate of this high annual load of DIN? Research on a common marine diatom (Skeletonema Nutrient flux and budget studies have shown that most of costatum) has shown that 15°C is the optimal temperature it is exported to the Indian Ocean (Fredericks et al. 1998 for NR activity (Gao et al., 1997; McQueen and Lean, and Kalnejais et al. 1999). This is further supported by 1987). In the Swan River Skeletonema costatum is the water quality monitoring studies showing a nutrient rich dominant spring bloom diatom in the lower estuary and plume in ocean waters in the vicinity of the estuary mouth median water temperatures increase to around 16°C at this (Johannes et al. 1994; Simpson et al. 1996). time.
In the middle and upper estuary spring blooms have a lag Spring dynamics of approximately one month after the lower estuary spring bloom and this reflects increased temperatures occurring In early spring the waters are flush with nutrients, particu- approximately one month later. larly nitrate, which together with changes in physical conditions, fuels spring phytoplankton blooms. The con- Summer dynamics ditions that favour increased phytoplankton production include increased water temperatures at this time and The summer water quality patterns in the lower estuary reduced freshwater flows, which lead to longer residence are quite different to those in the middle and upper regions. times allowing mixing with marine waters and increased In the middle and upper sites FRP concentrations are light penetration. The data also show that DIN concentra- markedly higher and median chlorophyll a concentrations tions decline rapidly, FRP also declines as dissolved reach their annual highs in summer, whilst in the lower nutrients are consumed and converted into phytoplankton estuary their annual lows are reached. In both regions, biomass. physical conditions in summer are conducive to maximum phytoplankton growth. At this time, water temperatures A marked change in the hydrodynamics also occurs in have increased to around 25°C, day lengths are longer, spring when rainfall and catchment runoff has largely irradiance is high and oceanic exchange of waters further ceased and the tidal action creates a saline wedge, which increases water clarity and therefore light penetration. intrudes upstream over the next few months. In early spring the front of the salt wedge, in the lower estuary, is Summer is a highly productive time for phytoplankton in associated with strong vertical stratification and hypoxic the middle and upper estuary. In these areas, DIN 24 SURFACE – Summer (Dec - Feb) BOTTOM – Summer (Dec - Feb) 50 50
lower Redfield [16:1] 40 middle 40 upper Redfield [16:1]
30 30 M) M) µ µ ( ( lower middle
20 DIN 20 DIN upper
10 10
0 0 02468 02468 FRP (µM) FRP (µM)
Figure 21: DIN vs FRP during summer (Dec-Feb) in surface and bottom waters, for the three regions of the estuary concentrations are low in the surface and bottom waters In contrast to the middle and upper estuary summer although ammonium concentrations are slightly higher in dynamics, the lower estuary had very low primary bottom waters. There are two explanations for the higher productivity. Bottom waters were well oxygenated during ammonium concentrations near the sediment surface. the summer months, median DO was ~80% at depth and Groundwater discharge is known to have high ammonium dissolved nutrient concentrations were minimal. The concentrations although the rate at which nutrients are potential influences, which may have brought about these delivered from groundwater is unknown (Turner et al., good water quality conditions, are increased flushing rates, 1998). Another explanation is that anoxic conditions have and possibly more efficient denitrification. Based on blocked the nitrification/denitrification processes. Oxygen 1979–82 data north of the mouth of the Swan River concentrations measured at 0.5 m from the bottom were (Johannes et al. 1994), the marine water, which intrudes consistently low in summer in the middle and upper sites, into the Swan River estuary, could be expected to have medians ranged from 25 to 60%. Although not measured low nitrate concentrations (~0.03 mgL-1). The summer directly, the sediment surface was likely to have been time data (this report), indicate lower estuary nitrate anoxic for extended periods. The pathway for nitrogen in concentrations were close to this value suggesting that the decomposition of organic matter is as follows: there were no groundwater inputs within the lower estuary.
decomposition + nitrification denitrification Flushing undoubtedly plays a significant role in keeping Org-N — —> NH4 — —> NOx — —> N2 } required nutrient concentrations down. In summer the lower estuary O2 In the absence of oxygen, the nitrification/denitrification is completely mixed vertically. Warmer conditions are process is blocked which leads to the build-up of ammo- likely to increase the rate of organic matter decomposition nium. Denitrification is tightly coupled to nitrification and denitrification may therefore also play a role in exporting nitrogen from the estuary as N gas. and is an important mechanism that removes bioavailable 2 nitrogen from the system as nitrogen gas, which diffuses back into the atmosphere. In collaboration with WRC, the Autumn Australian Geological Survey Organisation (AGSO) has Water quality in autumn is marked by a transition to undertaken sediment chamber experiments to measure winter conditions. Sporadic rainfall events create a com- denitrification efficiency. Preliminary results suggest the plex spatial and temporal influence making generalisations denitrification efficiency in the Swan River is very low difficult. For example, the autumn of 1995 recorded a and comparable to the Gippsland lakes, a highly eutrophic period of very high chlorophyll a concentration. This system. peak was followed by a sudden drop in chlorophyll a, Nutrient ratios in summer were generally below the which corresponded to a peak ammonium concentration Redfield ratio indicating the potential for nitrogen limita- caused by the release of ammonium from the breakdown tion (Figure 21). The potential for nitrogen to limit of senescing and decomposing phytoplankton. Autumn in phytoplankton biomass in summer was demonstrated by 1997 also experienced very high chlorophyll a concentra- bioassays (Thompson and Hosja, 1996; Thompson, 1998)). tions corresponding to very high total nitrogen Nitrogen-limitation of phytoplankton biomass is typical concentrations in the surface waters. Occasions of low of many estuaries (e.g. Port Phillip Bay, Moreton Bay, chlorophyll a were experienced, as expected in winter, and Chesapeake Bay (Harris et al., 1996)). especially in winter 1996 during the high flows. 25 7. Conclusions
The key results and comments from analysis of the first material oxygen is consumed, leaving the bottom wa- four years of the Swan-Canning Cleanup Program water ters hypoxic, susceptible to increases in phosphorus quality monitoring between 1994 to 1998 can be summa- release from the sediments and a build-up of ammo- rised as: nium. Collapsing phytoplankton blooms that are consumed by bacteria can also cause low oxygen con- • Hydrodynamic and hydrological forcing drives much centrations. Hypoxic conditions will continue while of the seasonal and spatial water quality dynamics organic loads are substantial and the system has high within the Swan River estuary. productivity due to high dissolved nutrient inputs. • In winter abundant dissolved inorganic nitrogen is de- Issues and areas that require further investigation include: livered to the estuary from catchment runoff. Estuarine phytoplankton activity in winter is limited, not by nutri- • Improved understanding of sediment/groundwater nu- ents, by low light intensity, low temperature, and short trient fluxes is necessary to address management issues. water residence time. Much of the high DIN loads are It is known that groundwater inflow can be high in exported out to sea. ammonium and phosphorus, but to what extent (spa- tially and quantitatively) these nutrients are delivered • The high-nitrate late winter and early spring catchment into the river and estuary is unknown. flows fuel spring phytoplankton blooms; in early spring in the lower estuary and approximately one month later • The temporal and spatial characteristics of the nitrifica- in the middle and upper estuary. The spring blooms tion/denitrification process need to be addressed. The convert DIN to organic-N, which ultimately decom- Australian Geological Survey Organisation (AGSO) is poses and nutrients are recycled back to the water currently undertaking some of this work. column stimulating further phytoplankton activity. • Understanding why N2 fixing cyanobacteria are limited • Summer phytoplankton blooms–restricted to the mid- in this N-limited environment would be useful to better dle and upper estuary–are fuelled by sediment-derived direct management efforts. Much work is being done in nutrients, either from ammonium build-up under an- the marine environment on this issue which have sug- oxic conditions or from groundwater discharge. gested that trace elements, such as molybdenum and Nutrients concentrated in the bottom waters are accessed iron can be limiting. by motile, flagellated phytoplankton, such as dino- • Monitoring the ungauged catchment and after rainfall flagellates, common in the upper region of the Swan in events may provide more information on nutrient loads the warmer months. Mixing events can also transport to the system, particularly as a result of first flush nutrients to the euphotic zone, however the upper estu- events. Some event-based sampling would be useful to ary appears to maintain a weak saline and oxygen explore the potential significance of these areas and vertical stratification throughout summer and therefore times. vertical transport of dissolved nutrients is restricted. Urban drains, such as Bayswater MD had extremely • To fully report on water quality of the Swan-Canning high DIN and maintained flows during summer. These system, the toxicant status should be documented. A have the potential to stimulate localised bloom events. review of existing data has been completed (DA Lord & Associates, 2000) which highlighted the lack of ad- • Relatively low summertime phytoplankton activity in equate temporal and spatial coverage of pollution data. the lower estuary is attributed to nitrogen limitation. A SPIRT grant study, in conjunction with Curtin Uni- Low nitrogen concentrations are sustained by tidal versity, commenced in 2000. This study, using fish as flushing with marine waters and possibly greater biomarkers for toxicants along with parallel sediment denitrification efficiency. analyses for heavy metals and other contaminants was • Low oxygen conditions are an intrinsic feature of the undertaken in spring 2000. Ideally, this project should estuary; different parts of the estuary are affected many continue to expand temporal coverage. months of the year. When bacteria consume organic
26 8. References
APHA 1989. Standard Methods for the Examination of Lewis D.P. and Pattiaratchi, C. 1989. Hydraulic investi- Water and Wastewater, 17th edition. American Public gation of the Swan estuary, Western Australia. University Health Association, Washington, DC. of WA, Centre for Water Research, Report No. WP290DL. ANZECC and ARMCANZ 2000. National Water Quality Management Strategy: Australian and New Zealand McQueen, D. and Lean, D. 1987. ‘Influence of water Guidelines for Fresh and Marine Water Quality. temperature and nitrogen to phosphorus ratios on the dominance of blue-green algae in Lake St George, Appleyard, S. 1992. Estimated nutrient loads discharged Ontario’. Canadian J. of Fisheries and Aq. Sciences 44, into the Swan-Canning Estuary from groundwater. pp. 598-604. Hydrological Report No. 1992/20. Western Australia Geological Survey, Perth, WA. Redfield, A.C. 1934. ‘On the proportions of organic de- rivatives in sea water and their relation to the composition Boardman, B., Hamilton, D. and Robb, M. 1998. Report of plankton’. In: James Johnson Memorial Volume, Ed on monitoring undertaken by the WRC for the 1997 R.J. Daniel, University of Liverpool press, pp. 176-192. Swan River destratification trial. 26 pp. Rochford, D.J. 1951. ‘Studies in Australian estuarine Donohue, R and Jakowyna, B. 2000. SCCP: Trends in hydrology I: Introduction and Comparative features’. total phosphorus and total nitrogen concentrations of Aust. J. of Marine Freshw. Res. 2, pp. 1-116. tributaries to the Swan-Canning Estuary, 1987 to 1998. Swan River Trust Report No. 6. Simpson, C., Burt, J., Cary, J., D’Adamo, N., Masini, R. and Mills, D. 1996. Southern Metropolitan Coastal Fredericks, D.J., Heggie, D.T., Watkins, K. & Longmore, Waters Study (1991-1994) Final Report. Report 17, A. 1998. Nutrient hydrocarbon and agrichemical fluxes Department of Environmental Protection, Perth, WA. in the Swan Canning Estuary. Australian Geological Survey Organisation – Petroleum and Marine Division. Spencer, R.S. 1956. ‘Studies in Australian estuarine hy- Report to WRC, 191pp. drology II: The Swan River’. Aust. J. of Marine Freshw. Res. 7, pp. 193-253. Gao, Y., Smith, J. and Alberte, R.S. 1997. ‘Temperature dependence of nitrate reductase (NR) activity in marine Stephens, R. and Imberger, J. 1996. ‘Dynamics of the phytoplankton: biochemical analysis and ecological Swan River estuary: the seasonal variability’. Mar. implications’. Journal of Phycology 36(2), pp. 304- Freshwater Res., 47, pp. 517-29. 313. Swan River Trust 1999. Swan-Canning Cleanup Pro- Harris, G., Batley, G., Fox, D., Hall, D., Jernakoff, P., gram: Action Plan (SRT/WRC Internal Report). Molloy, R., Murray, A., Newell, B., Parslow, J., Skyring, Swan River Trust 1998. Swan River Destratification Trial, G. and Walker, S. 1996. Port Phillip Bay Environmen- 1997 (SRT/WRC Internal Report). tal Study Final Report. CSIRO, Canberra. Thompson, P.A. 1998. ‘Spatial and temporal factors in- Hodgkin, E.P. 1987. ‘The hydrology of the Swan River fluencing phytoplankton in a salt wedge estuary, the estuary: salinity the master factor’. In: Swan River Swan River, Western Australia’, Estuaries. 21(4B), pp. Estuary: Ecology and Management. John, J. (ed) Curtin 801-817. University. Environmental Studies Group. Report No. 1, pp. 34-44. Thompson, P.A. and Hosja, W. 1996. ‘Nutrient limitation of phytoplankton in the upper Swan River estuary, Jennings, J.N. and Bird, E.C.F. 1967. ‘Regional Western Australia’. Mar. Freshwater Res., 47, pp. geomorphological characteristics of some Australian 659-667. estuaries’. In: Lauff, G.H. (ed.) Estuaries. American Asoc. for the Advancement of Science. Publ. No. 83. Thomson, C.E., Robb, M.E. and Rose, T. 2002. SCCP: Seasonal water quality patterns in the Canning River Johannes, R., Pearce, A., Weibe, C., Crossland, C., estuary, 1995 to 1998. Swan River Trust. Rimmer,D., Smith, D. and Manning, C. 1994. ‘Nutrient characteristics of well mixed coastal waters off Perth, Turner, J.V., Smith, A.J., Linderfelt, W.R. 1998. Swan Western Australia’. Estuarine Coastal and Shelf Sc. 39, River surface water-groundwater interaction: Techni- pp. 273-285. cal supporting document to the SCCP draft Action Plan, CSIRO Land and Water, Perth Laboratory, June Kalnejais, L., McMahon, K. and Robb, M. 1999. Swan 1998. Canning Estuary, Western Australia in Land-Ocean Interactions in the Coastal Zone (LOICZ), Eds Smith, Valiela, I. 1995. Marine Ecological Processes (2nd Ed.) S.V. and Crossland, C.J. LOICZ Reports & Studies Springer-Verlag. No. 12.
27 Appendix A Boxplot explanation
The figures below are produced using boxplots (for exam- values (square box) and the 25 and 75 percentiles. Plots ple using Statistica software). Box plots are an effective also show the outliers and extremes. means of presenting the data. They show the median
Explanatation of box plots:
Extreme values Outliers Outliers Extreme values
Outliers Extremes
Data point is > UBV + o.c. *. (UBV – LBV) Data point value > UBV + 2 * o.c. * (UBV – LBV) OR Data Point is < LBV – o.c. * (UBV – LBV) OR Data point value < LBV – 2 * o.c. * (UBV - LBV) Where o.c. is the outlier coefficient
Appendix B ANZECC/ARMCANZ 2000 Guidelines
The Australian and New Zealand Guidelines for Fresh managers develop site-specific guidelines particularly in and Marine Water Quality (ANZECC/ARMCANZ 2000) the case of sites for which water quality monitoring provide a method to establish site-specific ‘trigger levels’ programs have been undertaken. The WA EPA in using hierarchical decision frameworks that are founded conjunction with the WRC has developed trigger values on a risk analysis approach. They also provide default for the Swan-Canning estuary for a sub-set of the ANZECC trigger values as a basis for the protection of aquatic trigger values with the objective of maintaining recreational ecosystems, applicable to southern, Western Australia values. (see Table 4). These guidelines advocate that water resource
Table 4: ANZECC water quality default trigger values for estuaries and lowland rivers, southern Western Australia (ANZECC/ARMCANZ 2000)
Indicator Lowland Rivers Estuaries Marine (inshore)
-1 NOx (mgL ) 0.15 0.045 0.005 -1 NH4 (mgL ) 0.08 0.04 0.005 TN (mgL-1) 1.2 0.75 0.23 FRP (mgL-1) 0.04 0.005 0.005 TP (mgL-1) 0.065 0.03 0.02 Chl a (mgL-1) 0.003–0.005 0.003 0.0007 DO (% saturation) < 80 < 90 < 90 pH (lower limit) 6.5 7.5 8.0 pH (upper limit) 8.0 8.5 8.4 28 ✁
Publication feedback form
The Water and Rivers Commission welcomes feedback to help us to improve the quality and effectiveness of our publications. Your assistance in completing this form would be greatly appreciated. Please consider each question carefully and rate them on a 1 to 5 scale, where 1 is poor and 5 is excellent (please circle the appropriate number).
Publication title: SEASONAL WATER QUALITY PATTERNS IN THE SWAN RIVER ESTUARY, 1994–1998, TECHNICAL REPORT
How did you rate the quality of information? 12345
How did you rate the design and presentation of this publication? 12345
How can it be improved?
......
......
......
How effective did you find the tables and figures in communicating the data? 12345
How can they be improved? ......
......
......
How did you rate this publication overall? 12345
If you would like to see this publication in other formats, please specify. (E.g. CD)
......
Please cut along the dotted line on the left and return your completed response to:
Publications Coordinator Water and Rivers Commission Level 2, Hyatt Centre 3 Plain Street East Perth WA 6004 Facsimile: (08) 9278 0639
29
3rd floor, Hyatt Centre 87 Adelaide Terrace East Perth Western Australia 6004 Telephone: (08) 9278 0400 Facsimile: (08) 9278 0401 Website: wrc.wa.gov.au/srt
ISBN 0-7309-7549-5 ISSN 1444-3279