Reuse of recycled wastewater is one of the most feasible unconventional urban water sources in the United Arab Emirates (UAE). The extraction and reuse of greywater might affect the water and sewer pipe networks by changing flow characteristics. Therefore, the research question is how much greywater can be reused without affecting the existing water and sewer pipe networks? A residential complex located in Al Ain, UAE, was investigated in this regard. Water pipe network, water consumption and sewer flow data were collected and scenario modelling was conducted using the EPANET software. It was found that 100% capture and reuse of greywater has no impact on the upstream water distribution network, but reduces the recommended design flow in the downstream sewer network. The modelling results show that about 50% of generated greywater (about 95 litres per capita per day) can be harvested without affecting the upstream and downstream water and sewer flow characteristics, respectively.

INTRODUCTION

Use of water efficient appliances, reuse of wastewater/greywater and harvesting and reuse of rainwater/stormwater are the commonly applied urban water demand management practices. Water demand reduction leads to a reduction of wastewater flow and subsequently increases the wastewater strength. A reduction of 30% to 55% of water demand can reduce about 15% to 16% of wastewater flow (DeZellar & Maier 1980). As the flow decreases, the concentration of biochemical oxygen demand (BOD) and total suspended solids (TSS) is reported to increase by 25% to 40%. Parkinson et al. (2005) found that adoption of 6 litres and 4.5 litres capacities of toilet flush instead of 9 litres increased the TSS, BOD, chemical oxygen demand and ammonia nitrogen (NH4-N) concentrations by 10% and 24%, respectively. According to Marleni et al. (2012), there are three sewer problems associated with changes of wastewater quantity and quality. These are sewer blockage, formation of odour and corrosion of sewer pipes. Previous studies informed that the occurrence of sewer blockage is determined by the wastewater flow velocity and its quality (Arthur et al. 2008).

The United Arab Emirates (UAE) is one of the most water-scarce countries, but their water consumption rate is significantly high. Water consumption in traditional villa-type detached houses in the Abu Dhabi Emirate ranges between 270 and 1,760 litres per capita per day (lpcd) with an average of more than 600 lpcd (Environment Agency – Abu Dhabi 2015). The security of its urban water supply requires an augmentation of water supply sources and reduction of water demand. While treated sewage effluent (TSE) is used for landscape irrigation in the Abu Dhabi Emirate, the TSE is not reused for domestic non-potable purposes because of the high expenses of the dual reticulation system. However, the decentralized greywater reuse scheme is comparatively less expensive than the centralized treated wastewater scheme. In a recent study, Chowdhury et al. (2015) explored the potential of decentralized greywater reuse for water demand reduction in the study area. The average greywater generation rate in the study area was found to be 192 lpcd, which was about 69% of the average internal water consumption. The sources of greywater in the study area are showers (49%), ablution (18%), laundry (10%) and wash basins (23%) (Chowdhury et al. 2015). It was shown that the generated greywater is sufficient to fulfil the non-potable water demand in houses. Therefore, the research question is how much greywater can be extracted for reuse without affecting the existing water and sewer pipe networks. The objectives of the study were to estimate the total mains water consumption, and investigate the water end uses and the quantity of downstream sewer flow. This study only focused on the quantitative analysis of greywater.

FACTORS AFFECTING GREYWATER REUSE

There are many challenges to reusing greywater. Presence of salts in treated greywater is an issue and presents a barrier to their reuse for irrigation. Stevens et al. (2011) assessed the environmental risk posed to Australian and New Zealand ecosystems by the presence of powdered laundry detergents in greywater used for garden irrigation. They observed an increased alkalinity, sodicity and salinity in plants and soils irrigated by treated greywater. The presence of phosphorus and boron was also identified in some detergents that was considered to be a high risk for some plants. This would be especially the case in very low rainfall areas such as the UAE.

In addition to chemical hazards, there are some social challenges to greywater reuse. Some of these challenges are lack of public awareness, public perception, unavailability of regulations and standards, and lack of social and political willingness (Khatun & Amin 2011). Greywater has low quality in comparison with municipal water, so there are many social challenges to reusing it on-site (McNeill et al. 2009). Social acceptance is the primary issue for utilizing this unconventional resource. Without public acceptance, reuse of greywater cannot take a step forward (Nitivattananon & Sa-nguanduan 2013). Social attitude has a great influence as well. In particular, water and wastewater have a religious value in some cultures. A study conducted in Palestine showed that about 75% of the residents of the West Bank did not agree to accept treated wastewater for irrigation purposes because of their local social–cultural values (Halalsheh et al. 2008). On the other hand, in Jordan, more than 50% of farmers agreed to accept treated wastewater for irrigation purposes. A comprehensive greywater reuse education and awareness plan can raise this acceptability level (McNeill et al. 2009). The cost of greywater treatment varies according to its utilization and the type of treatment selection. The payback period of greywater treatment systems depends on the volume of treated greywater and the savings from municipal water charges (Li et al. 2010). Usually, the greywater treatment system used for irrigation purposes exhibits a shorter payback period.

STUDY AREA AND DATA

A residential complex located in Al Ain, UAE, was considered in this study. The complex occupies 100 villa-type detached houses. Figure 1 shows an image of the complex. Municipal water consumption data were collected from February 2013 to January 2014. Appropriate approval was taken from the Al Ain Distribution Company (AADC) and the UAE University (UAEU) for inspection and collection of water records from the study area. The collected water meter reading data were compared with the AADC district meter reading data for verification purposes. A structured questionnaire survey was performed to gather demographic data on dwellers. Sewer flow records were also collected for the same period. Hydraulic modelling software, EPANET (US EPA), was applied to analyse the hydraulic effects of greywater extraction on the existing water network. For this purpose, the water distribution network's (DN) geographic information system (GIS) data were collected from the AADC. The historical records of water inflow data for the study area, residential water meter readings, distribution pipe diameters and elevation were collected and used in the hydraulic modelling tasks. Different greywater reuse or extraction scenarios were performed in the modelling work.
Figure 1

Location of study area (image taken from Google Earth Pro on 14 August 2016).

Figure 1

Location of study area (image taken from Google Earth Pro on 14 August 2016).

The consumed water volume was divided by the estimated population of the complex and by the numbers of days per year in order to estimate the per capita per day water consumption. Comparison was made between the calculated actual water consumption in each house and the guided consumption prescribed by local regulations and the Vision 2030. Tables 1 and 2 show this comparison, partially extracted from the Environment Agency Abu Dhabi (EAD) and the Regulations and Supervision Bureau (RSB). The Vision 2030 sets the priority threshold target for domestic water consumption (Table 1). The RSB, an independent body, supervises the water quality and supply standards in the Abu Dhabi Emirate, and provides the water demand and sizing criteria for water distribution companies in the Emirate (Table 2).

Table 1

Recommended water consumption guidelines according to the Abu Dhabi Environment Vision 2030 (EAD 2012), which provides guidelines to preserve the environment and water resources in the Emirate

Priority outcome Measure Baseline 2010 Threshold 2030 
3.1.1 Moderate average domestic water consumption Domestic water consumption in L/capita/day 586 <340 
3.1.2 Moderate average daily domestic outdoor water use in villas and shabiat Domestic outdoor water consumption (villa and shabiat) in L/capita/day 756 <340 
3.1.3 Maximum use of recycled water for amenities plantation % of total water consumed for amenities plantation 35% 67% 
Priority outcome Measure Baseline 2010 Threshold 2030 
3.1.1 Moderate average domestic water consumption Domestic water consumption in L/capita/day 586 <340 
3.1.2 Moderate average daily domestic outdoor water use in villas and shabiat Domestic outdoor water consumption (villa and shabiat) in L/capita/day 756 <340 
3.1.3 Maximum use of recycled water for amenities plantation % of total water consumed for amenities plantation 35% 67% 

Described under Priority Area 3: Efficient Management and Conservation of Water Resources, and Priority 3.1: Integrated and Efficient Use of Water Resources.

Table 2

Water demand and sizing criteria according to the RSB (RSB 2009)

Type of premises and consumption categories Description Estimates of daily consumption rate (Imperial gallons) Estimates of daily consumption rate (litre) 
Villa and shabiat Per capita 77 350 
Villa and shabiata General services 250/450 1,100/2,000 
Villa and shabiatb Per bedroom 110 500 
Villa and shabiat Per capita 77 350 
Type of premises and consumption categories Description Estimates of daily consumption rate (Imperial gallons) Estimates of daily consumption rate (litre) 
Villa and shabiat Per capita 77 350 
Villa and shabiata General services 250/450 1,100/2,000 
Villa and shabiatb Per bedroom 110 500 
Villa and shabiat Per capita 77 350 

aGeneral services mean water used for internal gardening and general cleaning purposes for a standard-sized shabiat and villa. It may increase or decrease according to the type and size of villa. However, the estimated rate of consumption for calculating additional quantities shall be between 1 and 1.5 gallons per square metre per day.

bFor the shabiat and villa category, a reduction factor may be applied for every additional bedroom, according to the Distribution Company's own criteria.

HYDRAULIC MODELLING

The primary purpose in performing the hydraulic modelling of the existing water network in the study area is to identify the effects of estimated greywater reuse, as reduction in total water consumption and change in water flow rate. The hydraulic modelling was carried out using the EPANET software (www.epa.gov). The water network pipes and their characteristics were collected from the AADC. Three scenarios were performed in order to estimate the effects of greywater reuse on the pipe and sewer network (Table 3). The first scenario represents the actual water consumption. The second and third scenarios consider a reduction of total water consumption by 5% and 10%, respectively. It was found that the indoor water consumption was only 15% of the total water consumption. At the 192 lpcd greywater generation rate, theoretically, a maximum of 8% of municipal water could be saved if all the generated greywater was harvested and reused (Chowdhury et al. 2015). The adopted scenarios in Table 3 identified the pressure and velocity drop and rise on the existing water DN.

Table 3

Description of water consumption and greywater reuse scenarios

Scenario Actual yearly consumption (m³) % Reduction Annual consumption (m³) used for modelling 
1st 867,434 0% 867,434 
2nd 5% 824,062 
3rd 10% 780,691 
Scenario Actual yearly consumption (m³) % Reduction Annual consumption (m³) used for modelling 
1st 867,434 0% 867,434 
2nd 5% 824,062 
3rd 10% 780,691 

The first scenario (Table 3) was based on the actual water consumption of 867,434 m³ (no reduction), which was estimated at the consumers’ connection points linearly converted into a flow rate of 27.51 litres/second (L/sec). The total wastewater produced by indoor water activities was estimated at 127,131 m³ annually, which was 15% of the total municipal water consumption (867,434 m³). In the literature, it was found that about 69% of wastewater is greywater (Chowdhury et al. 2015), and its generation rate was found to be about 192 lpcd. Therefore, a maximum possible water-saving target for the study area was about 8% of total water consumption (indoor and outdoor) and considering the 69% of wastewater is greywater, a maximum of 10% of total water could be saved. Hence, the other two scenarios (the second and third scenarios) assumed in this study were reductions of total water consumption by 5% and 10%, respectively.

In order to simulate the hydraulic model for the existing water network, several hydraulic and physical parameters were required as input parameters in the model. These parameters were water flow rate, pipe size and slope, pipe material and operational condition of the existing water supply system. These parameters were obtained from the AADC. Figure 2 shows the water distribution mains in the study area. The pipe lengths and pipe sizes are essential for adequate simulation of the existing water DN. The total pipe lengths of the existing water network were calculated to be 8,042.23 metres. The DN pipe of size 150 mm, 200 mm and 300 mm has a total length of 7,539.95 metres, 422.8 metres and 79.48 metres, respectively. Ductile iron (DI) pipes were used in the DN. Table 4 shows different DN pipe segments (pipe IDs are shown in Figure 2) and their corresponding sizes, lengths and their flow rate scenarios. Similar to the pipe length, the elevation data of existing pipelines were also collected from the AADC, supplied as the built profile drawings. The laid pipelines’ elevation details at each junction were recorded and marked with a unique junction ID (Figure 2) in order to be used in the hydraulic modelling.
Table 4

Details of pipes, their ID and junction elevation (Source: AADC)

  Junction demand, litre/second (L/sec)
 
  
Node ID Elevation (m) Scenario-1 Scenario-2 Scenario-3 Remarks 
DMA 241.232 0.000 0.000 0.000 Boundary node 
J-1 241.525 0.000 0.000 0.000 T-junction 
J-2 242.522 0.000 0.000 0.000 Reducer 200 × 150 
J-3 242.533 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-4 243.459 0.000 0.000 0.000 Elbow 
J-5 243.751 0.000 0.000 0.000 Elbow 
J-6 240.845 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-7 240.839 0.000 0.000 0.000 Reducer 200 × 150 
J-8 239.261 0.000 0.000 0.000 Elbow 
J-9 238.716 0.000 0.000 0.000 Elbow 
J-10 237.496 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-11 237.810 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-12 237.881 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-13 238.330 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-14 241.174 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-15 241.174 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-16 241.431 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-17 241.471 0.000 0.000 0.000 Elbow 
J-18 242.765 0.000 0.000 0.000 Elbow 
J-19 243.165 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-20 242.790 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-21 242.700 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-22 245.508 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-23 245.618 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-24 245.618 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-25 242.199 1.72 L/sec 1.63 L/sec 1.55 L/sec  
  Junction demand, litre/second (L/sec)
 
  
Node ID Elevation (m) Scenario-1 Scenario-2 Scenario-3 Remarks 
DMA 241.232 0.000 0.000 0.000 Boundary node 
J-1 241.525 0.000 0.000 0.000 T-junction 
J-2 242.522 0.000 0.000 0.000 Reducer 200 × 150 
J-3 242.533 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-4 243.459 0.000 0.000 0.000 Elbow 
J-5 243.751 0.000 0.000 0.000 Elbow 
J-6 240.845 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-7 240.839 0.000 0.000 0.000 Reducer 200 × 150 
J-8 239.261 0.000 0.000 0.000 Elbow 
J-9 238.716 0.000 0.000 0.000 Elbow 
J-10 237.496 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-11 237.810 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-12 237.881 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-13 238.330 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-14 241.174 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-15 241.174 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-16 241.431 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-17 241.471 0.000 0.000 0.000 Elbow 
J-18 242.765 0.000 0.000 0.000 Elbow 
J-19 243.165 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-20 242.790 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-21 242.700 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-22 245.508 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-23 245.618 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-24 245.618 1.72 L/sec 1.63 L/sec 1.55 L/sec  
J-25 242.199 1.72 L/sec 1.63 L/sec 1.55 L/sec  
Figure 2

Layout of EPANET junctions and pipes.

Figure 2

Layout of EPANET junctions and pipes.

The water consumption pattern is one of the essential input data requirements for time series analysis in the EPANET software. The hourly consumption pattern provides the water requirement at each node at any time during the day, which helps to manage the operation of the water distribution system. The hourly water consumption pattern was collected for the month of August 2013 from the AADC. Significant variations among different months were not generally observed for diurnal variation of water consumption. Figure 3 shows the hourly water consumption pattern used in the EPANET modelling. Figure 3 shows two distinct diurnal patterns of water consumption, from 6:00 am to 5:00 pm the average hourly flow is 105 m³, whereas from 6:00 pm to 5:00 am the average hourly flow is 95 m³. This indicates that water consumption in daytime is about 11% higher than the consumption at night. The pressure at the pumping station was kept at the 1.5 bars.
Figure 3

Diurnal pattern of water consumption and pressure record (Source: AADC).

Figure 3

Diurnal pattern of water consumption and pressure record (Source: AADC).

The Hazen–Williams equation was used in the EPANET software for modelling the water network and to estimate flow friction loss due to the physical properties of the pipe. The equation simply shows the relation between mean velocity of water in a pipe and the geometric properties of the pipe and slope of the energy line: V = kCR0.63S0.54, where V is velocity (m/s), k is a conversion factor (k = 0.849 for SI units), C is the roughness coefficient, R is the hydraulic radius (m) and S is the slope of the energy line (head loss per length of pipe). The coefficient, C = 130, was considered for new DI pipes in the study area. The existing pressure of 1.5 bars at the source was entered in the fixed grade node or boundary node in addition to their elevation. The elevation of all junctions was entered according to the AADC built profile drawings and the demand for every junction was entered according to Table 4. After successfully running the model for hydraulic simulation, time series graphs for pressure and flow velocity were recorded. The tabulated results were also prepared for evaluation of the optimal selection.

RESULTS AND DISCUSSION

The average plot and building size of houses in the complex is about 2,000 m2 and 600 m2, respectively. This indicates that about 70% of plot area is kept as open land area and more than 80% of houses occupy garden and amenity plantations in the open spaces. The estimated median family size was found to be ten persons per household. The average water consumption was found to be about 10,608 m3/year. It was found that only 20% of supply water is discharged into sewer pipes, and the remaining 80% is used for outdoor purposes (irrigation to plantation and car washing). Statistically significant (at 95% significance level) correlation was not identified between family size and annual water consumption per house. This is because water consumption is mostly influenced by outdoor water uses, which is independent of family size. From the model analysis, it was found that if greywater is reused at a 190 lpcd rate, about 30% of sewer flow is reduced, but total water consumption (indoor and outdoor) is reduced by less than 10%. Currently, only 8% of generating TSE is used for street landscape irrigation and the rest is discharged to the Arabian Gulf. Therefore, from a sewer flow perspective, the establishment of TSE connection pipes to every house/complex may serve better performance than the greywater reuse scheme.

Simulation results – scenario no. 1

Scenario-1 represents the existing water consumption and demand situation. This was conducted in order to obtain the existing nodal pressure and pipe flow velocities. The results show that the nodes J-24 and J-10 exhibit the lowest pressure of 9.96 metres at 5:00 am and the highest pressure of 18.28 metres at 7:00 pm, respectively (Figures 4(a) and 4(b)). The pipe segments P-15 and P-5 exhibit the lowest velocity of 0.02 m/s at 12:00 am and the highest velocity of 0.87 m/s at 5:00 am, respectively (Figures 5(a) and 5(b)). For all nodes, it was found that the pressure starts to drop at around 4 am and increases at around 4 pm. The reason is that flow velocity increases during this time period and the flow velocity and pressure are inversely related.
Figure 4

(a) Simulated pressure (metres) at nodes J-6 to J-10 (scenario-1) (from top to bottom, the graph represents nodes J-10, J-9, J-8, J-7 and J-6, respectively; nodes J-7 and J-6 coincided); (b) simulated pressure (metres) at nodes J-21 to J-25 (scenario-1) (from top to bottom, the graph represents nodes J-25, J-21, J-22, J-23 and J-24, respectively; nodes J-23 and J-24 coincided).

Figure 4

(a) Simulated pressure (metres) at nodes J-6 to J-10 (scenario-1) (from top to bottom, the graph represents nodes J-10, J-9, J-8, J-7 and J-6, respectively; nodes J-7 and J-6 coincided); (b) simulated pressure (metres) at nodes J-21 to J-25 (scenario-1) (from top to bottom, the graph represents nodes J-25, J-21, J-22, J-23 and J-24, respectively; nodes J-23 and J-24 coincided).

Figure 5

(a) Simulated velocity (m/s) detail in pipes P-1 to P-5 (scenario-1) (from top to bottom, the graph represents links P-5, P-4, P-2, P-1 and P-3, respectively); (b) simulated velocity (m/s) detail in pipes P-11 to P-15 (scenario-1) (from top to bottom, the graph represents links P-14, P-13, P-11, P-12 and P-15, respectively).

Figure 5

(a) Simulated velocity (m/s) detail in pipes P-1 to P-5 (scenario-1) (from top to bottom, the graph represents links P-5, P-4, P-2, P-1 and P-3, respectively); (b) simulated velocity (m/s) detail in pipes P-11 to P-15 (scenario-1) (from top to bottom, the graph represents links P-14, P-13, P-11, P-12 and P-15, respectively).

Simulation results – scenario no. 2

Scenario-2 was simulated based on the 5% reduction in the current water consumption or 50% extraction and reuse of generated greywater for every villa. The purpose was to understand the changes in nodal pressure and pipe velocities in comparison to the existing water consumption scenario-1. The simulated nodal pressures and pipe velocities show that the nodes J-24 and J-10 exhibit the lowest pressure of 10.02 metres at 5:00 am and the highest pressure of 18.29 metres at 5:00 pm, respectively (Figures 6(a) and 6(b)). The pipe segments P-15 and P-5 exhibit the lowest velocity of 0.02 m/s at 12:00 am and the highest velocity of 0.82 m/s at 5:00 am, respectively (Figures 7(a) and 7(b)).
Figure 6

(a) Simulated pressure (m) detail at nodes J-6 to J-10 (scenario-2) (from top to bottom, the graph represents nodes J-10, J-9, J-8, J-7 and J-6, respectively; nodes J-7 and J-6 coincided); (b) simulated pressure (m) detail at nodes J-21 to J-25 (scenario-2) (from top to bottom, the graph represents nodes J-25, J-21, J-22, J-23 and J-24, respectively; nodes J-23 and J-24 coincided).

Figure 6

(a) Simulated pressure (m) detail at nodes J-6 to J-10 (scenario-2) (from top to bottom, the graph represents nodes J-10, J-9, J-8, J-7 and J-6, respectively; nodes J-7 and J-6 coincided); (b) simulated pressure (m) detail at nodes J-21 to J-25 (scenario-2) (from top to bottom, the graph represents nodes J-25, J-21, J-22, J-23 and J-24, respectively; nodes J-23 and J-24 coincided).

Figure 7

(a) Simulated velocity (m/s) detail in pipes P-1 to P-5 (scenario-2) (from top to bottom, the graph represents links P-5, P-4, P-2, P-1 and P-3, respectively); (b) simulated velocity (m/s) detail in pipes P-11 to P-15 (scenario-2) (from top to bottom, the graph represents links P-14, P-13, P-11, P-12 and P-15, respectively).

Figure 7

(a) Simulated velocity (m/s) detail in pipes P-1 to P-5 (scenario-2) (from top to bottom, the graph represents links P-5, P-4, P-2, P-1 and P-3, respectively); (b) simulated velocity (m/s) detail in pipes P-11 to P-15 (scenario-2) (from top to bottom, the graph represents links P-14, P-13, P-11, P-12 and P-15, respectively).

Simulation results – scenario no. 3

Scenario-3 was simulated based on the 10% reduction in the current water consumption or 100% extraction and reuse of generated greywater for every villa (which is equivalent to 69% of indoor water consumption). It was observed that the nodes J-24 and J-10 exhibit the lowest pressure of 10.07 metres at 5:00 am and the highest pressure of 18.36 metres at 7:00 pm, respectively (Figures 8(a) and 8(b)). The pipe links P-15 and P-5 exhibit a velocity of 0.02 m/s at 12:00 am and the highest velocity of 0.78 m/s at 5:00 am, respectively (Figures 9(a) and 9(b)).
Figure 8

(a) Simulated pressure (m) detail at nodes J-6 to J-10 (scenario-3) (from top to bottom, the graph represents nodes J-10, J-9, J-8, J-7 and J-6, respectively; nodes J-7 and J-6 coincided); (b) simulated pressure (m) detail at nodes J-21 to J-25 (scenario-3) (from top to bottom, the graph represents nodes J-25, J-21, J-22, J-24 and J-23, respectively; nodes J-24 and J-23 coincided).

Figure 8

(a) Simulated pressure (m) detail at nodes J-6 to J-10 (scenario-3) (from top to bottom, the graph represents nodes J-10, J-9, J-8, J-7 and J-6, respectively; nodes J-7 and J-6 coincided); (b) simulated pressure (m) detail at nodes J-21 to J-25 (scenario-3) (from top to bottom, the graph represents nodes J-25, J-21, J-22, J-24 and J-23, respectively; nodes J-24 and J-23 coincided).

Figure 9

(a) Simulated velocity (m/s) detail in pipes P-1 to P-5 (scenario-3) (from top to bottom, the graph represents links P-5, P-4, P-2, P-1 and P-3, respectively); (b) simulated velocity (m/s) detail in pipes P-11 to P-15 (scenario-3) (from top to bottom, the graph represents links P-14, P-13, P-11, P-12 and P-15).

Figure 9

(a) Simulated velocity (m/s) detail in pipes P-1 to P-5 (scenario-3) (from top to bottom, the graph represents links P-5, P-4, P-2, P-1 and P-3, respectively); (b) simulated velocity (m/s) detail in pipes P-11 to P-15 (scenario-3) (from top to bottom, the graph represents links P-14, P-13, P-11, P-12 and P-15).

The modelling results indicate that a reduction in water consumption increases the pressure and drops the maximum velocity in the water network, as presented in Figure 10, but the lowest velocity remains same, as presented in Figure 11. At 5% reduction (scenario-2), the existing pressure was improved by 0.60% at the minimum pressure location (J-24) and by 0.055% at the maximum pressure location (J-10). Furthermore, at 10% reduction (scenario-3), the pressure was improved by 1.11% at the same low pressure point and 0.44% at the same maximum pressure node. It is to be noted that the increase in water network pressure due to extraction of greywater has no influence on the water line strength. This is because the AADC service lines have a design pressure capacity of 100 m and the distribution lines and fittings can allow up to 1,600 m of design pressure. Similarly, the velocity was reduced by 5.75% at the maximum point (P-5) and the lowest velocity remained the same as before (P-15). At the 10% reduction of water consumption (scenario-3), the flow velocity at the maximum flow pipe (P-5) was reduced by 10.4%, but the lowest velocity was not found to be affected. The improved pressure on the water DN may help to detect the water leakages in the network and a reduction in high velocity may lessen the erosion of pipe linings. Furthermore, this approach improves the minimum residual pressure in the network and helps to maintain the RSB recommended pressure of 12.5 m in the network.
Figure 10

Node pressure detail for three different scenarios.

Figure 10

Node pressure detail for three different scenarios.

Figure 11

Flow velocity detail for three different scenarios.

Figure 11

Flow velocity detail for three different scenarios.

Downstream sewer conditions

Sewer flow comes from indoor water consumption and it was found that about 15% to 20% of the total water supply discharges to the sewer network. It is obvious that extraction of greywater reduces municipal water consumption, which will not only help to decrease water consumption but also alleviate the sewer load on treatment plants. On the other hand, a decline in wastewater discharge may affect the sewer flow and can increase the risk of sewer lines clogging. Wastewater treatment plants may experience dense sewer loads that require frequent removal of sludge (Friedler & Hadari 2006). The main purpose in assessing the impacts of greywater reuse on the sewer lines and treatment plant was to review different aspects of flow reduction and the building up of dense sludge in the sewer network. The Abu Dhabi Sewerage Services Company (ADSSC) wastewater treatment plant was visited to obtain the required information. Informal consultation was conducted with the plant operation engineers in order to evaluate their impacts on the sewer lines.

After the extraction of greywater, the wastewater treatment plants may receive a similar load with more concentrated substances due to less flow, and the boiler may consume the same amount of electricity as the load is same. But, the hydraulic retention time (HRT) in the aeration tank might be more than the usual HRT. Furthermore, some equipment may become oversized due to a reduction in the flow volume. The sewer line's HRT can be increased and sewer-solid content may settle on the pipe due to long HRT and the smaller portion of sewer water. Moreover, prolonged HRT provides a favourable condition for anaerobic bacterial growth. An aerobic process that occurs over a long time produces odour in the pipeline. The effect can be evaluated by estimating the average daily generation of black water and kitchen wastewater during a 24-hour period. Chowdhury et al. (2015) estimated that the average daily water consumption for toilet flush and dishwashing is about 39 litres/person and 44 litres/person in the study area. This indicates that a person generates about 83 litres of wastewater daily and discharges into the sewer network.

The ADSSC guidelines for the design of sewer lines recommend a design flow of 180 litres per capita for residential houses. Hence, it can be concluded that if 100% of available greywater is captured and reused, the average daily sewer flow rate will be about 50% (83 ≈ 90 litres/capita/day) of the design flow rate. As a result, extraction of 100% greywater, which is 192 lpcd, can effect the sewer line flow. However, extraction and reuse of 50% of available greywater fulfils the design flow rate in the sewer network. In addition, there is a possibility of odour and the building up of dense sludge in the sewer pipes, which can be minimized by proper operation and maintenance of the sewer network.

CONCLUSION

Urban water management is one of the major challenges in the UAE. The per capita water consumption in the UAE is excessively high, which is because of outdoor water uses. In the study area comprising 100 traditional villas, municipal water is used in outdoor plantations in almost every residential villa. It was identified that about 20% of the total supplied water is returned back to the downstream sewer system, which indicates that about 80% of the supplied water is used for outdoor activities. Because of this significantly high outdoor water use, water consumption was found to be independent of the number of family members. Residential water supply in the UAE is solely dependent on expensive desalinated water. Without reduction of water demand and augmentation by alternative water sources, there is no doubt that the urban water supply in the UAE will be unsustainable in the future. Since rainfall and runoff are not an abundant source of supply, the only feasible unconventional water source is recycled wastewater. This study explored the potential of greywater reuse, and in particular, how much greywater can be extracted without affecting the upstream water distribution and downstream sewer networks.

The modelling results indicate that the extraction of greywater has positive impacts on the water DN, such as that the improved pressure and reduced high velocity in the pipes may be favourable for identifying leakage and reducing pipe erosion, respectively. Furthermore, this may improve the minimum residual pressure in the pipes and be advantageous for maintaining the RSB recommended pressure of 12.5 m in the network. However, 100% extraction of greywater affects the downstream sewer network by reducing the design flow rate. It was shown that about 50% of greywater (about 95 lpcd) can be extracted without affecting the design flow in the sewer network. This study considered the quantitative aspect of greywater reuse and concluded that 50% of greywater can be recycled and reused. A comprehensive assessment considering quality, treatment, cost, health risk assessment and social acceptability is particularly essential for its application.

ACKNOWLEDGEMENTS

The research is funded by the UAEU-SEED research grant (31N183) entitled ‘sewer network problems associated with urban water source management practices’ awarded to the first author.

REFERENCES

REFERENCES
Arthur
S.
Crow
H.
Pedezert
L.
2008
Understanding blockage formation in combined sewer networks
.
Water Management
161
(
4
),
215
221
.
Chowdhury
R. K.
El-Shorbagy
W.
Ghanma
M.
El-Ashkar
A.
2015
Quantitative assessment of residential water end uses and greywater generation in the city of Al Ain
.
Water Science and Technology: Water Supply
15
(
1
),
114
123
.
DeZellar
J. T.
Maier
W. J.
1980
Effects of water conservation on sanitary sewers and wastewater treatment plants
.
Journal of the Water Pollution Control Federation
52
(
1
),
76
88
.
EAD
2012
Advancing Sustainable Groundwater Management in Abu Dhabi
.
Environment Agency, Abu Dhabi
,
UAE
.
Environment Agency Abu Dhabi
2015
A Water Budget Approach for the Emirate of Abu Dhabi, Policy Brief
.
Environment Agency
,
Abu Dhabi
,
UAE
.
Halalsheh
M.
Dalahmeh
S.
Sayed
M.
Suleiman
W.
Shareef
M.
Mansour
M.
Safi
M.
2008
Greywater characteristics and treatment options for rural areas in Jordan
.
Bioresource Technology
99
,
6635
6641
.
Khatun
A.
Amin
M. R.
2011
Greywater reuse: a sustainable solution for water crisis in Dhaka, Bangladesh
. In:
Proceedings of the 4th Annual Paper Meet and 1st Civil Engineering Congress
,
Noor
M. A.
Saiful Amin
A. F. M.
Chowdhury
S. R.
Bhuiyan
M. A. R.
Kakoli
S. T. N.
(eds).
Civil Engineering Division, Institution of Engineers
,
Bangladesh
, pp.
427
434
.
Marleni
N.
Gray
S.
Sharma
A.
Burn
S.
Muttil
N.
2012
Impact of water source management practices in residential areas on sewer networks – a review
.
Water Science and Technology
65
(
4
),
624
642
.
Nitivattananon
V.
Sa-nguanduan
N.
2013
Domestic water reuse situation in the context of middle-income countries: a case analysis
.
Frontiers in Environmental Engineering
2
(
1
),
7
17
.
Parkinson
J.
Schutze
M.
Butler
D.
2005
Modelling the impacts of domestic water conservation on the sustainability of the urban sewerage system
.
Journal of the Chartered Institution of Water and Environmental Management
19
(
1
),
49
56
.
RSB
2009
Guide to Water Supply Regulation 2009
.
Regulation and Supervision Bureau
,
Abu Dhabi Emirate
,
UAE
.
Stevens
D.
Dillon
P.
Page
D.
Warne
M.
Ying
G. G.
2011
Assessing environmental risks of laundry detergents in greywater used for irrigation
.
Journal of Water Reuse and Desalination
1
(
2
),
61
77
.