Swimming pools are heavy consumers of potable water and energy in cities. Despite this, the lack of monitoring requirements produces a lack of references for their water consumption. This paper aims to fill this gap by presenting a case-study water consumption analysis of a public swimming pool in the city of Bologna (Italy). After upgrading the water fixtures and filters in summer 2012, consumption and attendance at the Cá Selvatica swimming pool were monitored. With an average daily attendance of 88.6 people/day, taking into account both personal and technical consumption, an overall per capita consumption of about 96.1 l/person/day was determined. The water demand for personal uses only (showers, toilets, etc.) was found to be about 44.6 l/person/day. These values can be used to assess retrofitting benefits and water consumption benchmarks.

ABBREVIATIONS

     
  • OW/D

    overall average water consumption per day

  •  
  • OW/P

    overall average water consumption per person per day

  •  
  • TW/D

    theoretical technical water consumption per day

  •  
  • TW/P

    theoretical technical water consumption per person per day

  •  
  • PW/D

    personal average water consumption per day

  •  
  • PW/P

    personal average water consumption per person per day

INTRODUCTION

Water consumption and water management in cities has become an increasingly central issue in the debate on global sustainability, especially considering the growing global scarcity of water. In this context, Italy is one of the European countries with the largest water footprint (total volume of freshwater used to produce the goods and services consumed by the individual or community or produced by business – The Water Footprint Network). It is higher than the average value of the European Union by 25%, equating to 1,836 m3 per capita per year. On a global level, the Italian water footprint is 66% higher than the world average, which is 1,385 m3 per capita per year (WWF Italia 2014).

Most empirical studies carried out on urban water demand have traditionally focused on residential users (Wong & Mui 2007; Fan et al. 2013) leaving aside service (wholesale and retail trade, repair services, accommodation and food activities, entertainment and recreation activities, etc.) (Proença et al. 2010) and industrial users (Gao et al. 2008) that can account for an even larger percentage of the water consumed in cities (Arbués et al. 2010). The consequence is a lack of empirical evidence and benchmarks to use when talking about water consumption in non-residential buildings.

Among non-residential buildings that result in high water consumption in cities, we find public swimming pools. It is recognized that public swimming pools are high consumers of potable water resources and energy in municipalities; they deal with large quantities of water and constitute high frequency facilities, where personal needs are grouped. Swimming pool water is an expensive commodity given the fact that a large body of water needs to be continually pumped, treated, filtered, backwashed, and then heated to temperatures of 26–30 °C (79–86 °F) (Seneviratne 2007).

Water-efficient fixtures represent important means of water conservation in swimming pools (Seneviratne 2007) and minimizing overall water needs results not only in the saving of a precious resource, but also in minimizing energy usage and costs. Excessive flows and leaks in public shower areas can be a significant source of water waste as many users attend the facility, whereas installing low-flow fixtures can have multiple benefits for costs and safety (Australian Government 2006).

Despite this evidence, the lack of literature and references about water consumption in swimming pools is evident. Understanding the advantages of renovation is not an easy task, and this affects decision makers in both the public and private sectors. The absence of empirical studies is also due to a lack of monitoring regulations and obligations. In Italy, for example, there is regulation concerning sanitary aspects of the construction, maintenance and control of swimming pools (Liguori et al. 2014), and strong hygienic surveillance requirements have been established for the water quality aspects (Dallolio et al. 2013). However, there are no requirements for water quantity, which neglects the importance of water consumption monitoring and management in swimming pools, and its impact on public expense and on resource conservation. Water consumption is not even included in the aspects addressed during swimming pool inspections. In fact, during inspections routinely conducted by Local Health Authorities in Italian swimming facilities (Liguori et al. 2014), the only focus is on the sanitary and hygienic aspects. The lack of requirements to focus on other aspects during these inspections is evidence that the reporting of water consumption falls to the wayside.

Metering utilities to monitor water consumptions are not only important for water conservation goals but also help building owners and facility managers to ensure that the building is functioning properly. By installing sub-meters and data loggers, base flow can be detected and recorded (Seneviratne 2007) to ensure that a building performs the way it was designed to, therefore avoiding leakages and losses of water and money.

The aim of this study is to analyze a water-retrofitting project in a public swimming pool in Italy, and provide a reference on water consumption in a non-residential building. This research stems from studies conducted by the CIRI Building and Construction – Fluid Dynamics unit and the DICAM Department of the University of Bologna on the analysis of per capita water consumption of non-residential users in the city of Bologna, Italy (Farina et al. 2013). This case study project started in 2012 on the occasion of the renovation of the municipal swimming pool Cà Selvatica, which was granted to the sports club Rari Nantes Bologna. With the support of the sports club and the design studio Ricerca e Progetto – Galassi, Mingozzi and Associates, the research study focused on the installation of low water consumption technologies, and subsequently on water consumption in the facility through monitoring and data analysis.

MATERIALS AND METHODS

Study framework

The Cà Selvatica swimming pool is a small-sized (12 ×5.5 ×1.2 m, 80 m3 of water) covered pool located in via Cà Selvatica 11/2, Bologna (Italy) on the ground floor of a public school building. After a seismic event occurred in May 2012, the swimming pool facility was declared inhabitable. The sports club and private society Rari Nantes Bologna (RNB) was selected after a public notice by Bologna's Municipality to perform the restoration of the damaged area, and manage the facility for the following 15 years. The RNB club considered the renovation of the damaged areas an opportunity not only to re-establish structural safety, but also to upgrade the water and sanitation devices.

Renovation phase

The renovated area (Section 3 in Figure 1) occupies about 100 m2 and is composed of two almost symmetrical areas, one dedicated to women and one to men. The fixture retrofit program addressed the water closets, showers, faucets, and filters servicing the swimming pool. Two additional water sub-meters were installed at the water intake.
Figure 1

Space plan and new water pipelines (existing pipes in the swimming pool area are not shown here as not being part of the renovation); new water meters represent the monitoring points.

Figure 1

Space plan and new water pipelines (existing pipes in the swimming pool area are not shown here as not being part of the renovation); new water meters represent the monitoring points.

The old equipment was replaced and new devices were installed in October 2012 as shown in Table 1. Two sand filters and the water quality sampling system in the basement were also replaced: those filtering tanks and machines perform continuous water cleaning and sampling necessary to assure that the water in the pool is sanitarily safe. Hot water for the pool is withdrawn cold from the public network and heated on site.

Table 1

Low consumption fixtures installed during the retrofit of water devices

FixtureLocationNo. of unitsConsumption
WC Male and female restroom 3–6 l/flush 
Bidet Female restroom 5 l/min 
Washbasin Male and female restroom 5 l/min 
Swimsuit washbasin Male and female restroom 5 l/min 
Showers Male and female shower room 16 6 l/min 
Common showers Pool entrance 10 l/min 
Feet washbasin Between shower rooms 6 l/min 
 
FixtureLocationNo. of unitsConsumption
WC Male and female restroom 3–6 l/flush 
Bidet Female restroom 5 l/min 
Washbasin Male and female restroom 5 l/min 
Swimsuit washbasin Male and female restroom 5 l/min 
Showers Male and female shower room 16 6 l/min 
Common showers Pool entrance 10 l/min 
Feet washbasin Between shower rooms 6 l/min 
 

Data collection

In January 2013 the swimming pool opened again to the public and the data collection activity started in March 2013. Three main categories were addressed:

  • water consumption: monitoring of the meters;

  • number of users: pool attendance in time; and

  • use of the services: survey on how and when people use the water devices.

Water consumption

Before the renovation, school and pool consumptions were only measured together through a common water flow meter located inside the school building. Unfortunately, because the meter was combining the consumption of both the swimming pool and school, it was impossible for the study to understand the previous and actual consumptions for the swimming pool only. To keep track of only the swimming pool's water consumptions, during the refurbishment work two additional individual water meters were installed (Figure 1). This sub-metering was intended both for this study and for helping the managing company track consumption in the future.

The experimental estimation of the swimming pool water consumption was carried out continuously for 11 weeks from 18th March 2013 to 31st May 2013. The data recording was done at 15-min intervals thorough image capture with a webcam, and digitalization of the two volumetric water meter screens. The study discarded the festivity days when the facility was closed. Also excluded from the study were the weekends, as they present variations in activity schedules, due to private classes and birthday parties. Because of some data acquisition problems in the 11-week period, only 22 weekdays presented a full 24-hour reading, with no interruptions.

Number of users

The digitalization of the daily attendance register was very important to link the hourly water consumption to the hourly pool attendance. Users were not homogeneous during the weekdays:

  • 6:00 am–9:00 am: only technical consumption and water used for cleaning;

  • 9:00 am–1:00 pm: the pool was used by the public school for physical education activities a few days a week;

  • 1:00 pm–10:00 pm: the pool was managed by the RNB club and offered water aerobics, swimming classes, and lap swimming;

  • 10:00 pm–6:00 am: the pool was closed, but the filters and the automatic water quality analysis were still active and contributed to the technical consumption.

After the daily attendance analysis, the average presence per day (24 h) in the Cà Selvatica swimming pool was found to be about 88.6 people/day (76% female, 24% male) (Table 3). The type of courses offered, largely focused on water aerobics, determined the observed prevalence of female users. An additional source of variability in daily users came from the adjacent school using the pool a few mornings a week for physical education classes.

Use of the services

To be able to predict the expected consumption per user it was necessary to have more information about the users' consumption habits (as in Proença et al. 2010). For this purpose a survey was done, and people from different courses and of different age and sex gave answers about their habits in using water fixtures. In particular the survey asked for information about the user and the frequency and duration of fixture usage. The results (Table 2) have been used to estimate the theoretical per capita water consumption for personal uses, which resulted in 47.0 l/person/day.

Table 2

Results of the behavior survey on water usage: hypothetical l/person/day

Low consumption water fixturesUsage (users' behavior)aConsumption (water)bUsage (time)aVolume (water × time × usage%)c
%l/min – l/flushminl/person/day
Shower 89% 37.4 
Bidet 3% 0.4 
WC 67% 3.0–6.0 
Washbasin 22% 2.2 
Swimsuit washbasin 32% 3.2 
Foot washbasin 14% 0.8 
   Total 47.0 
Low consumption water fixturesUsage (users' behavior)aConsumption (water)bUsage (time)aVolume (water × time × usage%)c
%l/min – l/flushminl/person/day
Shower 89% 37.4 
Bidet 3% 0.4 
WC 67% 3.0–6.0 
Washbasin 22% 2.2 
Swimsuit washbasin 32% 3.2 
Foot washbasin 14% 0.8 
   Total 47.0 

aInformation from survey. Column 2 gives the % of time people used the fixture when they were in the pool. It was calculated by averaging and calculating the % of the declared behavior in the surveyed sample (always = 100%, often = 75%, seldom = 25%, never = 0% of the time).

bInformation from field measurements.

cHypothetical water use per person per day for personal uses taking into account behavioral patterns declared in the survey.

Data analysis

The goal of the data analysis phase was to link the water consumption gathered from the meter-readings to the swimming pool attendances, therefore creating usage patterns and understanding the actual per capita consumption. In Figure 2, two examples of daily water consumption are shown. Figure 3 represents the overall water consumption per person per day (daily consumption/daily users; consumption includes technical and personal water usage); it shows box plots for the 22 weekdays in the studied sample.
Figure 2

(a) A day of the week with school classes in the morning and RNB courses in the afternoon. (b) A day without morning classes (hours: 1:00 pm–10:00 pm).

Figure 2

(a) A day of the week with school classes in the morning and RNB courses in the afternoon. (b) A day without morning classes (hours: 1:00 pm–10:00 pm).

Figure 3

Boxplot of water consumption for the 22 weekdays analyzed.

Figure 3

Boxplot of water consumption for the 22 weekdays analyzed.

In the analysis phase of water consumption from water meter readings, it is important to underline that those water quantities include not only water for personal uses like showers, toilets, faucets, etc. They also include the water necessary for ‘technical consumption’: periodical water cleaning, automatic chemical sampling for pool water quality (pH, chlorine, etc.) and compensation for pool water losses; in fact, cold water from the public network is used to recharge the pool for the water losses due to sampling (72 l/cycle) and people (refilling usually when level is low and daily at 10:00 pm). Hot water for the pool is withdrawn cold from the public network and heated on site. The fact that hot and cold water were metered separately made it possible to isolate the technical consumption per day. In fact, when there is just cold water consumption this can be attributed to technical consumption only, using hot water as proxy for user presence. The only day in the sample which presented zero hot water consumption was Sunday, 19 May. It was used to estimate the technical consumption per day and the 4.03 m3 of cold water used on that day were considered the benchmark for the hypothetical daily technical water consumption (TW/D).

RESULTS AND DISCUSSION

The per capita water consumption during the selected 22 weekdays of complete and reliable monitoring (March to May 2013), is shown in Table 3. Difference in daily users is due to the fact that on a few mornings a week the school students have physical education at the pool.

Table 3

Consumption and attendance for the 22 weekdays of complete and reliable monitoring

Monday
Tuesday
Wednesday
Thursday
Friday
m3/dayPeople/daym3/dayPeople/daym3/dayPeople/daym3/dayPeople/daym3/dayPeople/day
8.58 124 9.12 121 9.72 85 11.79 136 9.57 128 
6.48 70 9.40 105 7.37 84 6.95 77 8.19 84 
9.51 133 6.63 113 7.61 51 7.41 115 7.25 58 
6.18 66 7.42 87   6.76 78   
6.50 47 7.18 90   6.25 60   
6.64 37         
Average Average Average Average Average 
7.32 80 7.95 103 8.23 73 7.83 93 8.34 90 
OW/D Overall average consumption per day m3/day 7.8 Overall water per day (Σ consumption) ÷ No. of days = 7.8 m3/day 
Average users/day Person/day 88.6 Pool attendants from presence registers (Σ users) ÷ No. of days = 88.6 person/day 
OW/P Overall average consumption per person per day l/person/day 96.1 Overall water per person per day Avg (consumption/person/day) = 96.1 l/person/day 
TW/D Theoretical technical consumption per day m3/day 4.03 Technical water per day From days with only cold water consumption 
TW/P Theoretical technical consumption per person per day l/person/day 51.5 Technical water per user per day [Avg (4.03 ÷ users/day)] × 1000 = 51.5 l/person/day 
PW/D = OW/D – TW/D m3/day 3.8 Personal water per day Avg (consumption/day − TW/day) = 3.8 m3/day 
PW/P Personal consumption per person per day l/person/day 44.6 Personal water per day Avg (PW/D ÷ users/day) = 44.6 l/person/day 
Monday
Tuesday
Wednesday
Thursday
Friday
m3/dayPeople/daym3/dayPeople/daym3/dayPeople/daym3/dayPeople/daym3/dayPeople/day
8.58 124 9.12 121 9.72 85 11.79 136 9.57 128 
6.48 70 9.40 105 7.37 84 6.95 77 8.19 84 
9.51 133 6.63 113 7.61 51 7.41 115 7.25 58 
6.18 66 7.42 87   6.76 78   
6.50 47 7.18 90   6.25 60   
6.64 37         
Average Average Average Average Average 
7.32 80 7.95 103 8.23 73 7.83 93 8.34 90 
OW/D Overall average consumption per day m3/day 7.8 Overall water per day (Σ consumption) ÷ No. of days = 7.8 m3/day 
Average users/day Person/day 88.6 Pool attendants from presence registers (Σ users) ÷ No. of days = 88.6 person/day 
OW/P Overall average consumption per person per day l/person/day 96.1 Overall water per person per day Avg (consumption/person/day) = 96.1 l/person/day 
TW/D Theoretical technical consumption per day m3/day 4.03 Technical water per day From days with only cold water consumption 
TW/P Theoretical technical consumption per person per day l/person/day 51.5 Technical water per user per day [Avg (4.03 ÷ users/day)] × 1000 = 51.5 l/person/day 
PW/D = OW/D – TW/D m3/day 3.8 Personal water per day Avg (consumption/day − TW/day) = 3.8 m3/day 
PW/P Personal consumption per person per day l/person/day 44.6 Personal water per day Avg (PW/D ÷ users/day) = 44.6 l/person/day 

First, the overall daily water consumption per person, including personal and technical water is calculated. Subtracting the technical consumption (TW/D, Table 3), previously established as 4.03 m3/day, it is possible to calculate the average ‘personal’ water used by the average attendant each day (PW/P, Table 3).

Now we can compare this actual result with the theoretical 47.0 l/person/day of water for personal uses estimated previously by means of a survey and of a calculation based on the users' behavior in the fixture usage (Table 2). The two values are consistent and this validates the formula used in Table 2 to calculate the theoretical per capita water consumption. Therefore we can use the same formula to establish hypothetical water consumption before the renovation, with obsolete high usage water fixtures installed. To calculate this new value, low-flow consumption in Table 2 (column 3) was replaced with the consumption of the old fixtures, measured before the refurbishment work. The result is a water consumption of about 69.3 l/person/day for personal care (Table 4).

Table 4

Water consumption per person per day with old high-consumption fixtures

High consumption water fixturesUsage (users' behavior)Consumption (water)Usage (time)Volume (water ×time ×usage%)
%l/min – l/flushMinl/person/day
Shower 89% 10 62.3 
Bidet 3% 0.5 
WC 67% 10 3.0 
Washbasin 22% 3.5 
Swimsuit washbasin 32% 0.0 
Foot washbasin 14% 0.0 
   Total 69.3 
High consumption water fixturesUsage (users' behavior)Consumption (water)Usage (time)Volume (water ×time ×usage%)
%l/min – l/flushMinl/person/day
Shower 89% 10 62.3 
Bidet 3% 0.5 
WC 67% 10 3.0 
Washbasin 22% 3.5 
Swimsuit washbasin 32% 0.0 
Foot washbasin 14% 0.0 
   Total 69.3 

If only the filters were upgraded, without replacing the old fixtures with high water consumption, 22.3 l/person per day would have been lost (Table 5).

Table 5

Hypothetical water savings obtained by installing new low-flow fixtures

New high performance filter consumption 51.5 l/person/day (TW/person in Table 3 Hypothetical OW/day   
Old high water fixture consumption 69.3 l/person/day (Table 469.3 + 51.5 = (old fix. + new filters) 120.8 l/person/day   
New low water fixture consumption 47.0 l/person/day (Table 247.0 + 51.5 = (new fix. + new filters) 98.5 l/person/day Saved water 22.3 l/person/day (−18%) 
New high performance filter consumption 51.5 l/person/day (TW/person in Table 3 Hypothetical OW/day   
Old high water fixture consumption 69.3 l/person/day (Table 469.3 + 51.5 = (old fix. + new filters) 120.8 l/person/day   
New low water fixture consumption 47.0 l/person/day (Table 247.0 + 51.5 = (new fix. + new filters) 98.5 l/person/day Saved water 22.3 l/person/day (−18%) 

Unfortunately, these water savings are only theoretical because of the lack of monitoring requirements in the past; the historical consumption for the pool is not retrievable and water was billed jointly for the swimming pool and the school. In addition, no sub-monitoring was in place before the current study.

CONCLUSIONS

To assess per capita water consumption after the installation of low-flow fixtures and new water filters in the public swimming pool Cá Selvatica, water consumption and attendance evidence were collected from March 2013 to May 2013, resulting in 22 weekdays of complete and reliable data. After the new devices were installed, the overall per capita consumption detected was 96.1 l/person/day (OW/P in Table 3), comprehensive of system and user consumption for the average 88.6 pool attendants/day. A further step was taken then to isolate the consumption linked to personal uses only, by subtracting from the overall consumption the portion due to technical consumption only (water filtering and sampling). The result was a 44.6 l/person/day water need for personal uses only (PW/P in Table 3). These collected data were used to validate a theoretical per capita consumption calculation, based on fixture flows in time and on users' behavior. The first was measured on site, while the second was established by means of a survey on usage patterns. The result was a theoretical per capita consumption of about 47.0 l/person/day for personal uses only, which was consistent with the 44.6 l/person/day found through actual data analysis. Replacing the low fixture flows in Table 3 with the high consumption flows of the fixtures installed previously, the theoretical water usage resulted in about 69.4 l/ person/day. To understand the benefits of investing in water fixture refurbishment a comparison of water consumptions with high and low-flow fixtures was done: if only the filters were upgraded, without replacing the high water-consumption old fixtures, 22.3 l/person per day would have been lost, resulting in 18% higher consumption than achieved with the investment in low-flow technologies. In the Cá Selvatica swimming pool, with an average daily attendance of 88.6 people/day and a 200-day yearly operating period, this would have resulted in 395 m3/year more water consumed and lost because of high consumption fixtures. Unfortunately this water-saving value is only theoretical due to the lack of empirical consumption values from before the renovation of the swimming pool. Nevertheless both empirical and theoretical results in this study are consistent with the benchmarks for public swimming pools defined in the report by Seneviratne (2007) where typical water usage for swimming pools in Sydney was found to be around 60 l/visitor/day (no. of visitors <500), reduced to 40 l/visitor/day if the daily users were more than 500 persons/day. Last but not least, in environments that are inhabited by young children and students, such as the Cá Selvatica swimming pool, which is used both by a school and a sports club, there is a big opportunity to educate the users with exemplary performance actions and communication campaigns. In a situation where there is heavy use of water as in swimming pools, forecasting water consumption can contribute to evident water and money savings, which benefit the general public as in the case of the Cá Selvatica swimming pool.

ACKNOWLEDGEMENTS

This study was possible because of the willingness and financial support of the sports club Rari Nantes Bologna and was carried out as part of the University of Bologna CIRI EC and DICAM research on water consumption. The help of multiple people from the RNB club and the design studio Ricerca e Progetto – Galassi, Mingozzi and Associates, the contribution of the UNIBO intern Michele Di Florio and the feedback from Warren A. Randle were fundamental to making this manuscript achievable.

REFERENCES

REFERENCES
Arbués
F.
García-Valiñas
M. Á.
Villanúa
I.
2010
Urban water demand for service and industrial use: the case of Zaragoza
.
Water Resources Management
24
(
1
),
4033
4048
.
Australian Government
2006
Water Efficiency Guide: Office and Public Buildings
.
Department of the Environment and Heritage, Focus Press, South Strathfield, NSW, Australia
.
Dallolio
L.
Belletti
M.
Agostini
A.
Teggi
M.
Bertelli
M.
Bergamini
C.
Chetti
L.
Leoni
E.
2013
Hygienic surveillance in swimming pools: assessment of the water quality in Bologna facilities in the period 2010–2012
.
Microchemical Journal
110
,
624
628
.
Fan
L.
Liu
G.
Wang
F.
Geissenc
V.
Ritsemac
C. J.
Tongb
J.
2013
Water use patterns and conservation in households of Wei River Basin, China
.
Resources Conservation Recycling
74
,
45
53
.
Farina
M.
Maglionico
M.
Pollastri
M.
Stojkov
I.
2013
Water consumption in public schools for the city of Bologna, Italy
.
Water Science and Technology: Water Supply
13
(
2
),
257
264
.
Gao
L.
Zongguo
W.
Bin
D.
Chao
Z.
Jining
C.
2008
An analysis of industrial water conservation potential and selection of key technologies based on the IWCPA model
.
Resources Conservation Recycling
52
,
1141
1152
.
Liguori
G.
Capelli
G.
Carraro
E.
Di Rosa
E.
Fabiani
L.
Leoni
E.
Marensi
L.
Napoli
C.
Pasquarella
C.
Romano Spica
V.
Canossa
C.
Dallolio
L.
Di Onofrio
V.
Gallè
F.
Giampaoli
S.
2014
A new checklist for swimming pools evaluation: a pilot study
.
Microchemical Journal
112
,
181
185
.
Proença
L. C.
Ghisi
E.
2010
Water end-uses in Brazilian office buildings
.
Resources Conservation Recycling
54
,
489
500
.
Seneviratne
M.
2007
Swimming pools
. In:
A Practical Approach to Water Conservation for Commercial and Industrial Facilities
.
Elsevier Science, Oxford, UK
, pp.
291
299
.
WWF Italia
2014
L'impronta idrica dell'Italia
. .