Assessing Urban Lake Performance for Stormwater Harvesting: Insights from Two Lake Systems in Western Sydney, Australia

Abstract

This study examines the impact of catchment characteristics and design on the performance of urban lakes in terms of water quality and stormwater harvesting potential. Two urban lake systems in Western Sydney, Australia, were selected for comparison: Wattle Grove Lake, a standalone constructed lake, and Woodcroft Lake, part of an integrated wetland–lake system. Both systems receive runoff from surrounding residential catchments of differing sizes and land uses. Over a one-year period, continuous monitoring was conducted to evaluate water quality parameters, including turbidity, total suspended solids (TSS), nutrients (total nitrogen and total phosphorus), pH, dissolved oxygen, and biochemical oxygen demand. The results reveal that the lake with an integrated wetland significantly outperformed the standalone lake in terms of water quality, particularly in terms of turbidity and total suspended solids (TSS), achieving up to 70% reduction in TSS at the outlet compared to the inlet. The wetland served as an effective pre-treatment system, reducing pollutant loads before water entered the lake. Despite this, nutrient concentrations in both systems remained above the thresholds set by the Australian and New Zealand Environment and Conservation Council (ANZECC) Guidelines (2000), indicating persistent challenges in nutrient retention. Notably, the larger catchment area and shallow depth of Wattle Grove Lake likely contributed to higher turbidity and nutrient levels, resulting from sediment resuspension and algal growth. Hydrological modelling using the Model for Urban Stormwater Improvement Conceptualisation (MUSIC) software (version 6) complemented the field data and highlighted the influence of catchment size, hydraulic retention time, and lake depth on pollutant removal efficiency. While both systems serve important environmental and recreational functions, the integrated wetland–lake system at Woodcroft demonstrated greater potential for safe stormwater harvesting and reuse within urban settings. The findings from the study offer practical insights for urban stormwater management and inform future designs that enhance resilience and water reuse potential in growing cities.

1. Introduction

As water is essential for life, it is crucial that all measures are taken to sustain, treat, and store this vital resource. Stormwater is defined as the excess water from rainfall. It travels over many surfaces and adsorbs various pollutants. Heavy rain can cause degradation in the water quality as pollutants such as debris, bacteria, nutrients, chemicals, oil, and sediments adhere to the water [1,2,3,4,5]. Aryal et al. [1] identified the possible sources for solids, metals, oil, grease & organics, and nutrients in stormwater. Tsihrintzis and Hamid [3] measured and modelled total kjeldahl nitrogen (TKN) and total suspended solids from an urban catchment. The land use and land cover appear to impact the stormwater quality significantly [2,4,5]. Stormwater enters drains, which ultimately discharge it into a water body, which may be used as a water source for a downstream town or city. In addition, the likelihood of flooding increases with heavy rainfall. One of the objectives of stormwater system design and planning is to retain floodwater and ensure public safety [6].

Urban lakes and wetlands have gained popularity for their role in storing and treating stormwater. They are designed to serve multiple purposes, including storing runoff, promoting physical well-being for visitors of all ages, and enhancing property values in the area [7]. Stormwater can also be harvested for domestic and irrigation purposes within urban areas [8]. Maintaining the stormwater’s high quality is important to enhance its practical use for the above purposes.

Studies conducted by Liu, Egodawatta [9], Miguntanna, Goonetilleke [10], Tong and Chen [11], and Duncan [12] aimed to identify the impact of land use on stormwater quality. Water quality parameters included total suspended solids, nutrients, biochemical oxygen demand, and other relevant factors. In the study conducted by Duncan [12], a meta-analysis (using data from multiple studies) was performed for 21 water quality parameters collected from 508 sites. The author found that there were significant differences in the stormwater quality, such as suspended solids, due to changes in the land uses. Generally, most of the studies that have been conducted thus far have focused on only a few parameters, apart from some studies [13,14] that included a wide range of physicochemical indicators. It is essential that the stormwater monitoring is conducted for at least one year to obtain the rainfall patterns and impacts over the four seasons. For a complete understanding of stormwater quality in a residential area, it is important that physicochemical and biological analyses are performed on samples collected after each storm event.

The main aim of this study is to understand how catchment characteristics and design features influence water quality in two urban lakes. The specific objectives of the study were:

• To evaluate the water quality of two urban lake systems, one with and one without an integrated wetland;
• To assess the impact of catchment characteristics, land use, and lake design on key water quality parameters;
• To compare observed results with the outputs obtained from the Model for Urban Stormwater Improvement Conceptualisation (MUSIC); and
• To identify key design parameters for urban lakes and to evaluate the stormwater harvesting potential.

2. Sites Location

Two lakes situated in urban residential suburbs of Western Sydney, New South Wales, Australia, were studied. Wattle Grove Lake is a standalone stormwater lake. Woodcroft Lake and wetland is a combined system. Both lakes could be potentially used for stormwater harvesting. The two catchments are situated in adjoining suburbs of Western Sydney and share similar climatic conditions. Both catchments have relatively flat topography with gentle slopes, and the underlying soil type in each is predominantly clay loam. These characteristics, along with their comparable urban residential land uses, support a meaningful comparison of their stormwater quality and lake system performance. The site descriptions are shown in

Table 1. Details of urban lakes and wetland.

Parameters	Wattle Grove Lake	Woodcroft Wetland	Woodcroft Lake
Longitude and Latitude	33°56′58.0″ S 150°56′27.3″ E	33°45′15.2″ S 150°52′50.3″ E	33°45′10.9″ S 150°52′48.9″ E
Maximum depth (m)	1.5	1.4	5
Area of water body (ha)	2.5	1.5	3.2
Area of parkland	2.5	4.9
Volume (m3)	40,500	1300	80,000 to 110,000
Catchment area (ha)	95	53

.

2.1. Wattle Grove Lake

Wattle Grove Lake is situated in Wattle Grove, within the Liverpool City Council. It is a constructed urban lake designed to improve stormwater quality and store excess runoff from rainfall within the catchment area (Figure 1). As shown in Table 1, the catchment area contained for Wattle Grove Lake is approximately 95 ha. The catchment area is predominantly residential. The area has around 1022 residential properties. The combined area of the lake and parkland is approximately 2.5 ha. When the lake’s capacity is reached, it discharges the excess water into a creek. To increase flow and promote oxygen in the lake, the local council installed three aerators. Furthermore, fountains were installed to enhance the lake’s aesthetics and facilitate the mixing of its water. Wattle Grove Lake has Phragmites australis and Typha orientalis present as macrophytes. Apart from the environmental aspect, the lake/parkland is designed in a manner that allows for recreational activities such as walking and bird watching. Furthermore, exercise equipment and a children’s park have been installed to promote physical well-being to visitors of all ages. Figure 1 shows the aerial view of the lake. The inlet of the lake is indicated by Point 1, the outlet is indicated by Point 2, and Point 3 shows the additional sampling point used for better statistical reasoning. Points 4 and 5 have autosamplers installed to monitor stormwater quality at the inlet and outlet of the catchment area.

2.2. Woodcroft Lake and Wetland

Woodcroft Lake is situated in Woodcroft, within the Blacktown City Council. It is a constructed integrated system that consists of a wetland connected to an urban lake (Figure 2). Previously, the area was used as a clay quarry. The purpose of this system is similar to that of Wattle Grove Lake, which is to store and treat stormwater. The areas of the lake and wetland are approximately 3.2 ha and 1.5 ha, respectively. The associated parkland has an area of approximately 4.9 ha.

Furthermore, the catchment area is approximately 53 hectares. The catchment area is predominantly residential. The excess runoff from the catchment area drains into the system through the inlet, then passes through the wetland, which acts as a filtering system, before the water reaches the lake. In this wetland, the macrophytes present are Thypha orientalis and Themeda australis. The lake acts as an equalisation basin by which excess water from the lake re-enters the wetland. Additionally, excess water from the wetland exits via the outlet, which is controlled by four syphons. Samples were taken at points illustrated in Figure 2. Similarly to Wattle Grove Lake, the area surrounding the lake has been designed to improve accessibility and promote physical well-being. Furthermore, exercise equipment, walkways and a community hall have been included to encourage interaction with nature and with other community members.

The inlet at WCL is within the Gross Pollutant Trap (GPT) and is shown in Figure 2, labelled as point 8. That is the inlet for the Woodcroft system. The water flows from the wetland (WCW) to the lake (WCL). Finally, it exits through the outlet (Outlet WCL). The syphons indicated in Figure 2, as Point 7, allow for the overflow from the wetland to enter the creek.

3. Methodology

The primary purpose of urban lakes is to treat and store stormwater runoff for potential reuse or recreational purposes; therefore, it is essential that water quality is monitored to assess the efficiency of the treatment system. Water quality not only impacts the lakes but also influences community use and associated benefits. To characterise the condition of the water at the sites, a methodology was developed for collecting and analysing the samples.

The samples were collected bimonthly in polyethylene bottles. In situ readings were taken for parameters such as dissolved oxygen, temperature, electrical conductivity, pH and turbidity immediately after collection. The readings were taken twice for each sample to ensure the reliability of the results. The samples were stored in an cooler box containing ice cubes and cool packs to ensure proper storage conditions during transportation from the site to the laboratory.

3.1. Stormwater Sampling During a Rain Event

Stormwater quality during rain events was also monitored in conjunction with bimonthly analysis. GAMET Autosamplers collected 24 samples over approximately 40 min after a rain event. They were triggered using float switches. These samples were then analysed in the lab. This is to understand the impact of rainfall on various parameters. This monitoring also examines the system’s ability to improve water quality by comparing the inflow and outflow water quality.

Autosamplers were installed at inlets and outlets of the urban lake systems within electrical enclosures and utilised for the sample collection (Figure 3). They were powered by generic 12 V car batteries. The batteries were recharged regularly to ensure ongoing monitoring. Float switches were used to trigger these systems. The arrangement is schematically shown in Figure 3a. The actual automatic sampler is shown in Figure 3b. The switches were installed within plastic boxes to help regulate the water flow and prevent accidental triggering due to turbulent water rather than increasing the flow of water. This system was set up so that the float switches would send a pulse to the autosampler, triggering a pre-set function. The autosamplers collected 1 L samples every 2 min for a total of 24 samples. The samples collected were then analysed in the laboratory. This was to understand the impact of rainfall on various parameters.

To determine pH, electrical conductivity (EC), temperature, and dissolved oxygen (DO), a HACH HQ 40d (HACH Australia) multiprobe device was used. The clarity of water is determined by turbidity. This is impacted by the suspended and dissolved particles in the water. The turbidity was determined using HACH^TM 2100P (HACH Australia). These parameters were measured on-site immediately after the sample was collected.

3.2. Water Quality Parameters Analysed

The Gallery Automated Photometric Analyser (GAPA) (Thermo Scientific Australia)) was used to measure the nitrogen compounds, namely, nitrogen oxides (NO₂⁻ and NO₃⁻) and ammonium (NH₄⁺). Total solids (TS) are made up of total suspended (TSS) and dissolved (TDS) solids. Total solids are influenced by the composition of excess runoff, as well as aquatic organisms and particles, including sediments, algae, and macrophytes. Standard methods, as stated in APHA [15], were followed for measuring TSS. Total Nitrogen (TN) and Total Phosphorus (TP) were analysed utilising the persulfate digestion method and measured using GAPA. All reagents to undertake the tests were obtained from Thermo Fisher Scientific, Australia. The particle size distribution (PSD) of lake water samples was analysed to observe the varying particle sizes at the two sites. Zetasizer and Mastersizer were used to measure the particle size distribution (PSD). This was conducted once in the monitoring period.

3.3. Stormwater Sampling Frequency

For over a year, bimonthly samples were taken at Wattle Grove Lake and Woodcroft Wetland and Lake. Prior to fortnightly samples, samples were taken weekly to understand the general composition of the water. During this time, 5 and 8 sampling points were considered for Wattle Grove Lake, and Woodcroft Lake and Wetland, respectively. After statistical analysis, the points were decreased to 3 for Wattle Grove Lake and 6 for Woodcroft Lake and Wetland (the significance for each location was calculated using the p-test and points were decreased accordingly). The number of sampling sets for each location and the corresponding durations are shown in 3. The samples collected are categorised according to seasons. Additionally, the stormwater inflow and outflow quality were investigated. A total of 16 and 17 stormwater events were monitored at Wattle Grove and Woodcroft, respectively. Actual sample dates are presented in Supplementary Tables S1–S9.

The seasonal groupings used in this study were: Summer (December to February), Autumn (March to May), Winter (June to August), and Spring (September to November).

Table 2. Sampling set for fortnightly samples at Wattle Grove Lake and Woodcroft Lake and Wetland.

Season	Summer (December to February)	Autumn (March to May)	Winter (June to August)	Spring (September to November)
Wattle Grove Lake	21/01/2015– 25/02/2015 (initial weekly sampling) 21/12/2016– 9/02/2017 (9 sets of samples in total)	4/03/2015– 26/05/2015 (initial weekly sampling) 20/04/2016– 18/05/2016, 11/03/2017– 31/05/2017 (13 sets of samples)	02/06/2015– 7/07/2015 (initial weekly sampling) 09/06/2016– 28/08/2016, 24/06/2017– 30/07/2017 (14 sets of samples)	12/09/2016– 30/11/2016 (6 sets of samples)
Woodcroft Wetland and Lake	11/02/2015–26/02/2015 (initial weekly sampling) 13/12/16–18/02/2017 (7 sets of samples)	5/03/2015– 28/05/2015 (initial weekly sampling) 13/04/2016– 11/05/2016, 30/03/2017– 22/05/2017 (18 sets of samples)	4/06/2015– 30/07/2018 (initial weekly sampling) 01/06/2016– 18/08/2016, 26/07/2017– 31/07/2017 (17 sets of samples)	06/09/2016– 27/11/2016 (6 sets of samples)

summarises the sampling set for fortnightly samples at the two sites. To assess the significance of seasonal differences in water quality parameters, one-way ANOVA was performed on selected variables such as turbidity, total suspended solids (TSS), and nutrient concentrations. Statistically significant differences (p < 0.05) were observed for turbidity and TSS at Wattle Grove Lake across seasons. No significant seasonal variation was observed for total nitrogen or phosphorus.

The monitored storm events included a range of rainfall intensities that are typical of Western Sydney’s climate, with several events producing substantial runoff. These events were selected based on rainfall-triggered sampling and reflect the types of storm conditions that frequently affect urban catchments in this region. Although site-specific rainfall data are not detailed in this manuscript, the events captured provide representative insights into how stormwater quality responds to commonly occurring rainfall patterns.

3.4. Stormwater Modelling

Modelling is an essential tool that aids in understanding the design, development, and optimisation of urban stormwater runoff management systems [16]. It is essential, as not all issues in stormwater management are solved with a single solution. Modelling also helps in assessing the impacts of land use and large catchments [17]. To manage stormwater pollution effectively, it is essential to model pollutant concentrations in urban catchments [18]. In this study, the Model for Urban Stormwater Improvement Conceptualisation (MUSIC) software [12] was used to estimate stormwater quantity and quality. MUSIC takes into account the local meteorological data, catchment characteristics, pollutant data, and treatment systems incorporated in the catchment. MUSIC is also widely used by local councils for their catchment modelling purposes. The catchments were modelled, and aspects such as detention times, surface-to-volume ratios, stormwater quality, and harvesting potential were estimated. The modelling would also give an approximate representation of the annual load of nutrients and total suspended solids at each site.

The MUSIC models for Wattle Grove and Woodcroft are illustrated in Figure 4. The catchment’s areas and pervious/impervious percentages for both locations were obtained from Drewitt [19] and are shown in

Table 3. MUSIC input for Wattle Grove Lake Catchment [19].

Catchment Element	Total Area (m2)	Imperviousness (%)
Residential Housing	597,180	50.0
Roads	114,518	100
Parks and Reserves	129,818	0.0
St Marks Coptic College & Wattle Grove Plaza	28,841	82.5
Australis Park & Wattle Grove Public School	53,276	23.5
Wattle Grove Lake	25,443	100
Total	949,076	51.5

and

Table 4. MUSIC input for Woodcroft Lake [19].

Catchment Element	Total Area (m2)	Imperviousness (%)
Residential housing	377,142	70
Roads	60,242	100
Lake Woodcroft	32,450	100
Parks and recreational areas	71,623	0.0
Marayong Public School	29,605	31
Total	571,062	65.3

.

4. Results and Discussions

4.1. General Water Quality

The water quality for the two urban lakes and wetland is summarised in

Table 5. Average ranges for parameters measured in bimonthly sampling taken at Wattle Grove Lake and Woodcroft Lake (ANZECC Guidelines [20] values are also shown).

Water Quality Parameters	Wattle Grove Lake (WGL)	Woodcroft Wetland (WCW)	Woodcroft Lake (WCL)	ANZECC Guidelines (2000) *
Dissolved Oxygen (DO) (mg/L)	6.7–10.7	4.9–10.3	5.4–9.9	6.5–8.5
Electrical Conductivity (EC) (µS/cm)	131–231	149–1387	256.25–485.4	20–30
Turbidity (NTU)	17–86	1 to 33	1.5–12	1–20
pH (H+)	6.4–7.8	6.6–7.9	6.9–7.9	6.5–8.5
Total Nitrogen (TN) (µg/L)	356–1573	15–2509	172.7–2218.3	350
Total Phosphorus (TP) (µg/L)	116–2751	182–1746	21–1765	10
Biochemical Oxygen Demand (BOD) (mg/L)	2.2–8.6	1.5–6.9	1 to 5	-
Chl-a (µg/L)	0–311	2–696	1–166	5
Dissolved Organic Carbon (DOC) (mg/L)	4.9–8.3	5.5–16	4.8–7	-
Total Suspended Solids (TSS)(mg/L)	1–76.3	1–353	1–1116	-

Note: *—Trigger values suggested for Lakes & Reservoirs of South-East Australia.

. All sampling points for each location for each monitoring date were averaged and the upper and lower values are presented. It can be seen that, in general, the lake water quality at both locations does not meet the ANZECC (Australian and New Zealand Environment and Conservation Council) Guidelines [20].

Parameters such as dissolved oxygen, pH, and biochemical oxygen demand are within the guidelines provided. However, the rest of the water quality parameters are outside the recommended value/range. The water quality of Woodcroft Lake appears to be generally better than that of Wattle Grove Lake. Poor water quality in Wattle Grove Lake may be attributed to the catchment size (Table 1) and the presence of aquatic organisms in the lake, such as carp. It is in the nature of carp to disturb the sediments in water bodies to search for food [21,22,23,24,25]. Therefore, the resuspension of sediments is a possible reason for the high turbidity at Wattle Grove Lake.

It can be observed that in the case of Woodcroft, the stormwater quality in Woodcroft Lake (WCL) is superior to that of Woodcroft Wetland (WCW). This indicates that the wetland serves the purpose of retaining nutrients and improving water quality before it reaches the lake. This is also seen in a study conducted by Oberts and Osgood [26]. Although some improvement is seen, the concentrations are still not within the range mentioned in ANZECC Guidelines [20]. A previous study conducted by Natarajan, Hagare, and Maheshwari [27] explored the seasonal variations and the impact of constructed wetlands on improving water quality. They found that the wetland did help to improve the water quality, but not enough to be within the acceptable range recommended by ANZECC Guidelines [20].

4.2. Water Quality of the Stormwater at the Inlets and Outlets

A total of 16 and 17 stormwater events were monitored at Wattle Grove Lake and Woodcroft Lake, respectively. The inlets and outlets of both lakes were monitored to understand the changes in water quality throughout the system. As these lakes were designed to improve water quality through extended detention times and sedimentation, it is essential to explore their efficiency.

The inlet stormwater quality is compared between the two catchments in Figure 5. As shown in Figure 5a, the turbidity and total suspended solids (TSS) concentrations in the stormwater are significantly higher for Wattle Grove than at Woodcroft. It can be seen that the concentration of turbid-causing material entering Wattle Grove Lake is almost double that of entering the Woodcroft system. This indicates that the Wattle Grove catchment contributes relatively higher amounts of turbidity and suspended solids to the lake. On the other hand, as shown in Figure 5b, it was found that the total nitrogen (TN) entering Woodcroft Wetland and Lake was higher than that of Wattle Grove in all seasons. Total phosphorus was found to be generally in the same range, except in summer. The inlet concentrations appear to impact the concentrations in both lakes, particularly with respect to turbidity and TSS, as shown in Figure 6.

Figure 6a compares the turbidity at the inlets and outlets for the lakes and wetland. One of the factors contributing to the high concentration of turbidity-causing material in WGL may be attributed to the size of the catchment, which is approximately twice the size of the WC lake catchment. Moreover, the outlet of Wattle Grove Lake has turbidity almost 50% higher than the inlet of Wattle Grove Lake, which indicates resuspension of sediments and/or algal growth within the lake. This increase in turbidity after a storm event is also seen in the study conducted by Walker [28]. However, in the case of WCL, a significant reduction in the turbidity was observed.

Figure 6b shows the total suspended solids for the inlet and outlets for both lakes. For all seasons, Woodcroft Lake shows improvement in TSS values between the inlet and outlet (up to 70% reduction). This was also observed in a study conducted by Hathaway and Hunt [29]. On the other hand, the TSS results for Wattle Grove Lake indicate that the concentrations are consistently higher at the outlet than at the inlet. This means that similar to turbidity, the WGL contributes to the suspended solids concentration at the outlet. The source for these suspended solids could be either the resuspension of solids or the algal growth within the lake. Resuspension in the case of WGL is possible due to its relatively shallow depth and potential carp activity near the outlet.

Furthermore, the catchment size for Wattle Grove is almost twice that of Woodcroft. This indicates that there is more input to the lake from a larger catchment area, resulting in high turbidity and TSS. The outlet at Woodcroft has the lowest turbidity, indicating that the lake and wetland improve the water quality.

Figure 6c shows the variation in total nitrogen for samples taken at the inlet and outlet at both locations. There is a significant decrease in the TN concentration at Woodcroft Lake most seasons, which may be attributed to the wetland (up to 76%). Similar removal rates were reported by Greenway [30]. The TN at the outlet in Wattle Grove is often higher than the inlet. The reasons may be due to the presence of higher algal content in the lake as well as higher nitrogen content in the sediments. This interaction between nitrogen concentration in the lake and algal growth was observed in a study conducted by Xu, Paerl [31].

Figure 6d displays the total phosphorus (TP) for samples taken after storm events at inlet and outlet at both locations. A reduction in TP is observed in Woodcroft Lake in all seasons except summer. Reduction in TP through the wetland systems was also reported in a study conducted by Martín et al. [32]. A significant decrease is observed in total phosphorus concentrations in Wattle Grove Lake for all seasons except spring. However, the results are not consistent. Further investigations are required to ascertain the exact trend in TP.

Results from the Woodcroft system are generally lower than those from Wattle Grove, indicating the wetland’s ability to retain sediments and pollutants before the stormwater reaches the lake.

4.3. Size of the Particles Suspended in Stormwater

Water samples were taken from both WGL and WCL and analysed using a Zetasizer and a Mastersizer. Zetasizer measures particle sizes under 0.7 µm. Figure 7a presents particle size distribution (PSD) for particles under 0.7 µm for Wattle Grove Lake and Woodcroft Lake. It can be seen that Wattle Grove has a higher percentage of small particles than Woodcroft. At the lower end, about 8% of the particles are of size less than 100 nm for WGL. Whereas in the case of WCL, it is around 2%. This may be a contributing factor to the higher turbidity in the water at Wattle Grove Lake. Wattle Grove Lake has a catchment size of 95 ha, whereas the catchment size of the Woodcroft Lake is 53 ha. The volume of stormwater entering WGL is larger, and therefore, it has a higher concentration of suspended solids, as indicated in the inlet results. The impact of the difference in catchment area is also discussed in the modelling results.

The pollutant loads at the lake inlets are influenced by land use patterns within the catchments. In particular, impervious areas such as roads and commercial buildings contribute to higher runoff volumes and associated pollutant loads, especially sediments and nutrients. For example, the Wattle Grove catchment includes 100% impervious roads and a highly developed commercial zone, contributing to higher sediment loads and nutrient concentrations at the lake inlet. These findings are consistent with earlier studies linking urban land use with degraded stormwater quality (Tong & Chen, 2002 [11]; Goonetilleke et al., 2005 [17]).

Figure 7b displays the particle size distribution (PSD) for particles ranging from 1 µm to 1 mm. As shown in the figure, the stormwater entering WGL appears to contain a smaller fraction of the particles. About 60% of the particles are less than 10 µm in the case of WGL. On the other hand, it is only about 5% in the case of WCL. This reinforces the earlier statement on the increased turbidity (Figure 6a) and TSS (Figure 6b) in the case of WGL due to the presence of finer particles. As stated earlier, higher turbidity and TSS at the inlet could be attributed to the catchment characteristics and the size.

4.4. Stormwater Modelling Results

As outlined in Section 3.4, MUSIC models were set up for both Woodcroft and Wattle Grove catchments and were used to estimate stormwater quality and quantity. The hydraulic retention times were calculated using the MUSIC output for flow for each location, as described in Table 5. It can be seen that the Woodcroft Lake system, on average, allows for a total of up to 143 days of stormwater retention, which is approximately 3.5 times that of Wattle Grove Lake. This promotes sedimentation, leading to improvements in water quality. A higher retention time, combined with a relatively lower surface area-to-volume ratio or higher depth (

Table 6. The calculated hydraulic retention times and surface area-to-volume ratios for all locations.

	Average Hydraulic Retention (Days) *	Surface Area to Volume Ratio (m2/m3)	Average Depth (Volume/Surface Area) (m)
Woodcroft Wetland	1.7	3.38	0.08
Woodcroft Lake	143	0.33	3.44
Wattle Grove Lake	39.4	0.64	1.62

Note: *—calculated using annual stormwater flows.

), is able to promote better settling of particles with minimal resuspension of solids. This could be a reason for better lake water quality in the case of Woodcroft Lake.

The average concentrations of pollutants at the inlet and outlet of each lake system, as determined from field monitoring, are presented in

Table 7. Average inlet and outlet concentrations of pollutants obtained through storm event sampling.

Pollutant	Woodcroft Lake and Wetland			Wattle Grove Lake
	Inlet (GPT)	Outlet of Wetland	Outlet of Lake	Inlet	Outlet
Total nitrogen, µg/L	1810	1276	1061	1150	1302
Total phosphorus, µg/L	594	661	579	755	382
Total suspended solids, mg/L	8	6	4	16	23

. These concentrations were used to calculate the pollution load based on field measurements.

Table 8. Comparison of pollutant loads and removal efficiencies between MUSIC model and field observations for Wattle Grove Lake.

Mean Annual Loads Wattle Grove Lake	Inflow (MUSIC)	Inflow (Field Observations)	Outflow (MUSIC)	Outflow (Field Observations)	% Reduction (MUSIC)	% Reduction (Field Observations)
Flow (ML/yr)	375	-	369	-	2	-
Total suspended solids (kg/yr)	76,800	6000	34,900	8487	55	−41
Total phosphorus (kg/yr)	130	283	79.3	141	39	50
Total nitrogen (kg/yr)	807	431	731	480	10	−11

compares the pollution load and removal efficiencies obtained from MUSIC modelling and field observations for Wattle Grove Lake. It can be seen from the table that under ideal conditions, the Wattle Grove Lake system could potentially remove about half of the TSS that enters the lake from the catchment. Additionally, MUSIC modelling yields a 39% and 10% reduction in total phosphorus and total nitrogen, respectively. On the other hand, the field measurements show an increase in the TSS with negative removal rates and a similar result was obtained for TN. Nevertheless, the field measurements appear to yield a similar % reduction in TP.

As shown in

Table 9. Comparison of pollutant loads and removal efficiencies between MUSIC model and field observations for Woodcroft Wetland.

Mean Annual Loads Woodcroft Wetland	Inflow (MUSIC)	Inflow (Field Observations)	Outflow (MUSIC)	Outflow (Field Observations)	% Reduction (MUSIC)	% Reduction (Field Observations)
Flow (ML/yr)	280	-	258	-	8	-
Total suspended solids (kg/yr)	55,200	2240	6800	1680	88	25
Total phosphorus (kg/yr)	93.2	166	24.4	185	74	−11
Total nitrogen (kg/yr)	613	507	386	357	37	30

, the MUSIC modelling yielded removal rates of 88%, 74% and 37%, respectively, for TSS, TP and TN concentrations for the Woodcroft Wetland. However, the field measurements yielded significantly different pollutant removal rates except in the case of TN. Furthermore, as shown in

Table 10. Comparison of pollutant loads and removal efficiencies between MUSIC model and field observations for Woodcroft Lake.

Mean Annual Loads Woodcroft Lake	Inflow (MUSIC)	Inflow (from Experiments)	Outflow (MUSIC)	Outflow (from Experiments)	% Reduction (MUSIC)	% Reduction (from Experiments)
Flow (ML/yr)	258	-	228	-	12	-
Total suspended solids (kg/yr)	6800	1680	3220	912	53	59
Total phosphorus (kg/yr)	24.4	185	21.3	132	13	21
Total nitrogen (kg/yr)	386	357	258	242	33	52

, the pollutant reductions obtained for TSS, TP, and TN at Woodcroft Lake were somewhat similar between the MUSIC model and field observations.

Table 8, Table 9 and Table 10 appear to indicate that the Woodcroft treatment system yields higher-quality stormwater compared to the Wattle Grove treatment system. There could be several reasons for this. Some possible reasons for the water quality in the WGL include a higher level of turbidity in the stormwater generated by the catchment due to its characteristics, a higher hydraulic loading rate of the WGL, and a smaller average depth (volume/surface area) of the lake. The lower average depth results in the resuspension of sediments accumulated within the lake and the formation of algal blooms. Hunt & Lord [33] recommended that sediments be removed every 5 to 10 years for optimal system functioning.

MUSIC’s prediction of pollution load varied significantly. In some cases, the predictions were similar to field observations; however, in many instances, the predicted values differed considerably from the observed values. This reinforces the need to calibrate the model with additional field data.

4.5. Stormwater Harvesting and Recreational Use of Blue Infrastructure

The two urban lakes considered in this study have the potential to enhance stormwater harvesting within urban areas. Out of the two lakes considered in this study, the water quality in Woodcroft Lake appears to be superior. Hence, the lake can be used for stormwater harvesting and recreational activities within the urban centre. However, the quality of the stormwater needs to be carefully assessed to ensure its suitability for the intended purpose. This would involve ongoing monitoring of the water to ensure that it is safe to use for irrigation of public areas and protocols addressed in the guidelines for stormwater harvesting and reuse [34] are to be followed. The guidelines include the following key elements:

• adhering to planning and regulatory requirements;
• appropriate design of the plumbing and irrigation system;
• treatment of the stormwater;
• sufficient signage on areas irrigated with stormwater;
• scheme commissioning, validation and verification;
• ongoing operation and maintenance; and
• regular auditing and record keeping.

It is noted that this study focused on physicochemical indicators and did not include microbiological parameters such as Escherichia coli or faecal coliforms. These are important for assessing public health risks associated with the reuse of stormwater, especially for irrigation or recreational purposes. Future studies should include microbiological analysis to comprehensively evaluate the suitability of urban lakes for stormwater harvesting.

5. Conclusions

The primary objective of this study was to examine the water quality of urban lakes and evaluate their potential for stormwater harvesting. This was achieved through detailed monitoring of inlet and outlet water quality at two contrasting systems in Western Sydney: Woodcroft Lake (WCL), which incorporates a constructed wetland, and Wattle Grove Lake (WGL), a standalone lake without pre-treatment.

Analysis of inlet water quality revealed that stormwater entering Wattle Grove Lake was highly turbid and carried elevated levels of suspended solids to the lake’s outlet. These conditions are attributed to the larger catchment area, shorter detention time, and a less favourable surface area-to-volume ratio, which together reduce the system’s capacity to settle and retain pollutants.

The study found that the inclusion of a constructed wetland significantly improved stormwater quality, particularly in reducing turbidity, total nitrogen (TN), and total phosphorus (TP). The integrated wetland–lake system at Woodcroft consistently outperformed the standalone system in nutrient and solids retention. However, despite these improvements, nutrient concentrations in both systems remained above the acceptable ranges set by the ANZECC Guidelines [20], indicating that further treatment may be required for safe reuse.

Wattle Grove Lake consistently showed high values for TN and, in most cases, TP, alongside increased turbidity. These findings suggest that the lake receives finer suspended particles, likely due to soil characteristics in the catchment, and may experience internal resuspension from sediment disturbance and algal growth. Resuspension could also be due to the shallow depth of the lake.

A comparison between the MUSIC model and field observations revealed considerable differences in predicted and observed pollution loads. While the model helped estimate average conditions, the discrepancies underline the importance of calibrating modelling tools with local field observations to enhance reliability.

Overall, the findings highlight the potential for urban lakes to support stormwater harvesting and recreational uses. Of the two sites studied, Woodcroft Lake was found to be more suitable for stormwater harvesting due to its better water quality, greater hydraulic retention time, and integrated wetland treatment. These results support the design and retrofitting of urban stormwater infrastructure to incorporate wetlands and enhance catchment management, thereby improving water reuse potential in cities.

Given the likelihood of more intense and unpredictable storm events under future climate scenarios, this framework could be extended by integrating rainfall variability modelling and adaptive infrastructure planning. Such enhancements would support more resilient urban stormwater systems.