In this study, the quality of collected rainwater at a downtown middle school rainwater harvesting system was evaluated by measuring physical, chemical, and microbiological parameters such as pH, dissolved oxygen (DO), chemical oxygen demand (COD), total nitrogen (TN), total phosphorus (TP), NO3, PO4, total coliform (TC), Escherichia coli , and some metals (i.e. Al, Cr, Mn, Zn, Cu, As, Cd, and Pb) (2003 to 2011). The analysis shows that the collected water quality is poor, which presents health, considering the high levels of bacterial indicators detected in the harvested rainwater, i.e. turbidity (1.4 to 15.5 NTU) and E. coli (120 and 35 CFU/100 mL in 2007 to 210 and 60 CFU/100 mL in 2011). This study shows that deteriorating water quality was caused by system contamination due to the absence of maintenance. Based on this study, proper operation and maintenance are generally the simplest and most effective ways of maintaining water quality.

INTRODUCTION

Poor water quality and water shortage are becoming the biggest problems in urban areas (Vialle et al. 2011; Angrill et al. 2012; Rahman et al. 2014). Urban development and industrialization deteriorate water quality and contribute to water shortage (Farreny 2011; Lee et al. 2012). Collected rainwater is safe for consumption if its quality is maintained. It seems to be attractive because collected rainwater is relatively free from contamination compared to other water (e.g. surface water and groundwater) (Liaw et al. 2008; Lee et al. 2012; Rahman et al. 2014). However, its quality can deteriorate during harvesting, storage, and reuse. Dust, leaves, fecal droppings from birds, animals, and insects and contaminated litter on the catchment area can all cause contamination, leading to potential health risks from the reuse of the collected water reuse with poor quality (Vazquez et al. 2003; Evans et al. 2006; Sazakli et al. 2007). Currently, collected rainwater is widely used for toilet flushing and gardening purposes in South Korea. It could be used for showering, laundry, and so on (Kim & Lee 2004; Han & Park 2009). For drinking water purposes, sources of contaminations should be taken into account (Sazakli et al. 2007). Many studies have dealt with collected rainwater quality and have drawn contradictory conclusions (Simmons et al. 2001; Chang et al. 2004; Zhu et al. 2004; Lee et al. 2010; Abbasi & Abbasi 2011; Kwaadsteniet et al. 2013). The condition of the catchment influences the quality of the collected rainwater. Runoff from the catchment surface includes nonpoint pollutants (Kwaadsteniet et al. 2013) and contamination from corrugated roofing materials (e.g. heavy metals, paints, and coatings). Thus, maintenance must be performed for the removal of dirt, leaves, and other accumulated materials. Such cleaning should take place periodically before the start of rainfall. Conveyance systems are required to transfer the collected rainwater to its final point of use via collection devices (e.g. gutters and storage tanks). Corrosive metals may be transferred to the collected rainwater from metal roofs and pipes. Conveyance systems need to be periodically inspected and carefully cleaned to avoid contamination. Storage tanks should be carefully designed so that they offer drinking water with very low health risks. A poorly designed tank can pose high health risks from microbiological contamination. Storage tanks should be checked and cleaned periodically. Inappropriately high temperatures and external pollutants can cause microbial contamination. A secure cover is necessary to prevent mosquitoes from breeding and rodents from drowning. In the case of a rooftop and storage tanks, sunlight can penetrate the tank and cause algae to grow inside. The objective of this study is to show that maintenance may be one of the most important factors to ensure good water quality from rainwater harvesting systems.

MATERIALS AND METHODS

Study area

A rainwater harvesting system was installed at ‘G’ Middle School in South Korea in October 2002 (Figure 1) being the first of its kind in South Korea. The rainwater utilization system is composed of a catchment area (1,713 m2), (1) two treatment facilities including two storage tanks made of corrugated steel (60 m3 each (Φ 2.5 × 13 m)) and having concrete floors and soakaways, (2) and (3) two water supply facilities including two submerged pumps, (4) a filter system, and (5) a small pond for landscaping. In 2002, the quantity of rainwater used was 120 m3/month. From 2002 to 2006, the collected rainwater was used for cleaning, gardening, emergencies, and pond water. However, at present, this system cannot be used as potable water because of a poorly designed tank and a lack of inspection and management. The stored rainwater was contaminated by high concentrations of debris and nutrients. Now, it is used as pond water, with a warning sign saying ‘NOT A SOURCE OF POTABLE WATER-DO NOT DRINK’.
Figure 1

Site description (modified by Han and Park 2005).

Figure 1

Site description (modified by Han and Park 2005).

From 2002 to 2004, the system underwent regular maintenance and was kept secure according to a rainwater harvesting manual. The system was checked for rusting or leaking in gutters or pipes. The managers frequently checked that the gutter screens were not damaged or clogged with leaves or dirt. Also, they cleaned the catchment before rainfall events. The first flush device was cleaned out using drained water from the pipes after each rainfall event. Regularly, the rainwater tank was cleaned and disinfected to prevent slime, algae, bacterial growth and the build-up of sediments.

Sample collection and analysis

In this study, rainwater was collected from two tanks. The samples were collected in 1-L sterilized bottles from 2003 to 2011. The samples were then analyzed at a laboratory. Table 1 shows a summary of the analytical parameters and equipment (APHA 1995). Twenty five samples were analyzed in each year.

Table 1

Review of analytical parameters and methods

 ParameterMethod and equipment
Physical–chemical pH, Turbidity APHA (1995) & Thermo-orion 550A pH meter, HACH 2100P portable turbidimeter 
COD, TSS, TN,PO4 TP APHA (1995)  
NO3, SO4 DIONEX ICS 3000 
Metals APHA (1995) & Perkin-Elmer ELAN 6100 
Micro-biological TC, E. coli ISO method 9308-1 
 ParameterMethod and equipment
Physical–chemical pH, Turbidity APHA (1995) & Thermo-orion 550A pH meter, HACH 2100P portable turbidimeter 
COD, TSS, TN,PO4 TP APHA (1995)  
NO3, SO4 DIONEX ICS 3000 
Metals APHA (1995) & Perkin-Elmer ELAN 6100 
Micro-biological TC, E. coli ISO method 9308-1 

Based on the analysis results for rainwater, descriptive statistics (box plot diagrams) were obtained using SPSS V.12 K software (SPSS Inc., USA). For correlation analysis, a nonparametric Spearman rank correlation analysis was performed to determine possible relationships among duration and physical-chemical and microbiological parameters.

RESULTS AND DISCUSSION

pH and turbidity

Figure 2(a) and 2(b) provide box plot diagrams showing variations in the pH and turbidity, respectively. The average pH was monitored closely in 2003. The system included catchment surfaces and tank flooring of concrete. Further, the catchment was also constructed using concrete. After completion, concrete debris from the catchment and storage tank significantly affected the pH and turbidity of the harvested rainwater. The turbidity increased from an average of 1.4 NTU in 2003 to 15.5 NTU in 2011. Most notably, the turbidity's lowest monitored value was 0.4 NTU (meeting KEOM ≤0.5 NTU). However, the turbidity increased after 2005 because the middle school no longer managed the harvesting system.
Figure 2

Variations in pH, turbidity and CODcr.

Figure 2

Variations in pH, turbidity and CODcr.

Chemical oxygen demand (COD)

Figure 2(c) shows a box plot diagram for variations in COD. In 2003, the research group did not perform measurements because the tank was found to be dirty. The average COD increased gradually by a factor of 9.2 from 2004 to 2011. The COD originated from nonpoint sources in the rooftop catchment and from biologic debris in the storage tank. The water quality rapidly deteriorated after 2009. This means that stored water could become contaminated and may contribute to the spread of pollutants (i.e. bacteria, organic matter, N and P) due to the absence of proper maintenance.

Total nitrogen (TN) and NO3

Figure 3(a) shows the average TN and NO3 concentration monitored in 2004 to be 0.19 mg/L and 0.06 mg/L, respectively. From 2004 to 2011, TN and NO3 concentration sharply increased by factors of 39 and 71 respectively. Although the standard amount for NO3 in drinking water in Korea is 10 mg/L, these measurements show that the water quality deteriorated. The presence of TN and NO3 originated from the catchment due to human and animal activities and indicate that improper maintenance of rainwater harvesting systems affects the water quality adversely.
Figure 3

Variation in TN, NO3, TP and PO4 concentration.

Figure 3

Variation in TN, NO3, TP and PO4 concentration.

Total phosphorus (TP) and PO4

Figure 3(b) shows the average TP and PO4 concentrations that were monitored in 2004 to be 0.01 mg/L and 0 mg/L, respectively. TP and PO4 concentrations increased sharply owing to the cessation of maintenance in 2005. Although the Korean wastewater reclamation and reuse standard (KEOM 2013) amounts for TP and for PO4 are 0.2 mg/L and 0.01 mg/L respectively, the PO4 concentration in harvested rainwater exceeded the baseline 0.01 mg/L after 2005. The changes in TP and PO4 were similar to those of TN and NO3. However, the influence of TP and PO4 on contamination is minimal when compared to TN and NO3.

Al, Cr, Zn, Cu, Mn, As, Cd, and Pb

Rainwater utilization can be hindered by air pollution and by the construction materials used. The analytical metal parameters are Al, Cr, Mn, Zn, Cu, As, Cd, and Pb. The measurement results are shown in Figure 4.
Figure 4

Variations in metal concentrations.

Figure 4

Variations in metal concentrations.

Al

Figure 4 shows also the average Al concentration. Immediately after installing the rainwater utilization facility, the Al concentration ranged widely (20–190 μg/L) (not shown in this paper) because of Al dissolved from the concrete catchment. The average Al concentration of 63.0 μg/L is extremely low compared to the 200 μg/L limit set by the Korean drinking water standards (KEOM 2013).

Zn

Figure 4 shows that the average Zn concentration increased over time. Zn could originate from tank material. In general, the corrugated steel was coated with Zn to prevent corrosion caused by the harvested rainwater. Increasing Zn concentrations could be caused by tank corrosion. In 2011, the average Zn concentration was 517.0 μg/L, which is lower than the 1,000 μg/L limit set by the Korean drinking water standard.

Cu, Pb, Cr, Mn, As, Cd

The average Cu concentration increased sharply over time. Cu could originate from concrete materials and the facility system. However, it is at an extremely low concentration compared with the 1,000 μg/L limit set by the Korean drinking water standard. The average Pb concentration increased slightly over time from 22.8 to 35.9 μg/L. These metals originated from the catchment and storage tank. However, Cr, As, and Cd average concentrations were low compared with the 50 μg/L limit (Cr and As) and 10 μg/L limit (Cd) set by the Korean drinking water standards.

Total coliform (TC) and Escherichia coli (EC)

Figure 5 shows total coliforms and E. coli detected for 5 years (2007–2011). The values of microbiological parameters were not monitored from 2003 to 2005. Total coliforms and E. coli gradually increased from 2007 to 2011 (from 120 CFU/100 mL and 35 CFU/100 mL to 210 CFU/100 mL and 60 CFU/100 mL, respectively). Leaves, fecal droppings from birds, insects and contaminated litter on the catchment area can be reasons for such microbial contamination. The organic substances came from human and animal activities. Notably, the high TC and EC populations in 2009 were caused by a low utilization rate as well as unsanitary conditions. At this time, the harvested rainwater was temporarily not used owing to facility failure. TC and EC concentrations were considerably higher than the Korean wastewater reclamation and reuse standard of 0 CFU/100 mL as well as the Korean drinking water standard of 0 CFU/100 mL.
Figure 5

Variations in total coliform (TC) and Escherichia coli (EC).

Figure 5

Variations in total coliform (TC) and Escherichia coli (EC).

Correlation analysis

Our statistical results suggest that harvested rainwater quality was dependent on both the duration over which the system was left without maintenance and the chemical-microbiological parameters shown in Table 2. Table 2 shows the correlation analysis that we carried out to derive relationships. Duration is significantly related to TC and EC activity, which is taken into account in our Spearman coefficient (ρ = 0.929 for TC and 0.881 for EC). This means that absence of maintenance is significantly related to the increases in TC and EC. In addition, Table 2 shows duration to be strongly related to COD (ρ = 0.99), TN (ρ = 0.98), NO3 (ρ = 0.98), TP (ρ = 0.98), and PO4 (ρ = 0.85). This study confirms that deterioration in water quality was caused by system contamination due to dry deposition from the atmosphere and animal activity exacerbated by the absence of maintenance.

Table 2

Correlation coefficients for duration and chemical, physical, and microbiological parameters (significant at 5% level)

 TurbidityCODcrTNNO3TPPO4TCECDuration
Turbidity 1.00 0.86 0.86 0.86 0.86 0.71 0.79 0.74 0.86 
CODcr  1.00 1.00 1.00 1.00 0.85 0.93 0.88 0.99 
TN   1.00 1.00 1.00 0.85 0.93 0.88 0.98 
NO3    1.00 1.00 0.85 0.93 0.88 0.98 
TP     1.00 0.85 0.93 0.88 0.98 
PO4      1.00 0.73 0.66 0.85 
TC       1.00 0.97 0.93 
EC        1.00 0.88 
Duration         1.00 
 TurbidityCODcrTNNO3TPPO4TCECDuration
Turbidity 1.00 0.86 0.86 0.86 0.86 0.71 0.79 0.74 0.86 
CODcr  1.00 1.00 1.00 1.00 0.85 0.93 0.88 0.99 
TN   1.00 1.00 1.00 0.85 0.93 0.88 0.98 
NO3    1.00 1.00 0.85 0.93 0.88 0.98 
TP     1.00 0.85 0.93 0.88 0.98 
PO4      1.00 0.73 0.66 0.85 
TC       1.00 0.97 0.93 
EC        1.00 0.88 
Duration         1.00 

CONCLUSION

The collected rainwater quality at a downtown middle school was evaluated by measuring physical, chemical, and microbiological parameters. The quality of the collected rainwater definitely depended on maintenance in rainwater harvesting systems. This study can be summarized as follows.

  • (1) Poor harvested rainwater quality indicates the need for improved education on the importance of maintenance in rainwater harvesting systems. Cleaning and hygiene maintenance is a good approach to minimize contamination.

  • (2) Possible contamination (i.e. metals and nutrients) in stored rainwater can be caused by the poor condition of parts of the rainwater harvesting system including the catchment, pipes and tanks.

  • (3) Maintenance and operation has a discernible impact on the quality of harvested rainwater for users

  • (4) Operators should follow a year-round checklist for potable or non-potable rainwater harvesting systems, including the following tasks:

    • Check tank water quality every month

    • Thoroughly clean the dripper and assembly parts every season

    • Check the screen and overflow pipe every month

    • Check particle filters every month

    • Disinfect the stored rainwater every month

    • Check the debris removal device every month

    • Check the cleaning tank every season

    • Clean the catchment every month.

ACKNOWLEDGEMENTS

This research was funded by Korea Environmental Industry & Technology Institute (KEITI), Seoul National University. The authors would like to thank members of the Rainwater Harvesting Center for their help and involvement in the completion of the project.

REFERENCES

REFERENCES
Abbasi
T.
Abbasi
S. A.
2011
Sources of pollution in rooftop rainwater harvesting systems and their control
.
Environmental Science and Technology
41
,
2097
2167
.
APHA
1995
Standard Methods for the Examination of Water and Wastewater
. 19th edn.
Washington, DC
.
Chang
M.
McBroom
M. W.
Scott Beasley
R.
2004
Roofing as a source of nonpoint water pollution
.
J. Environ. Manag
.
73
,
307
315
.
Han
M. Y.
Park
J. S.
2005
Rainwater management in Korea: public involvement and policy development
. In:
Proceedings of 1st International Workshop on Rainwater and Reclaimed Water for Urban Sustainable Water Use
,
Tokyo
,
Japan
.
Han
M. Y.
Park
J. S.
2009
Innovative Rainwater Harvesting and Management in the Republic of Korea
.
Consulting Meeting for development of guidelines four Eco-efficient water infrastructure took place
,
Seoul
,
Korea
.
KEOM
2013
Official Test Methods of Water Quality
. Ed 2013.11.06.
Ministry of the Environment, Korea
.
Kim
R. H.
Lee
S.
2004
Developing technologies for rainwater utilization in buildings for urban area in Korea
. In:
Proceedings of 11th IRCSA
,
Mexico City
,
Mexico
.
Kwaadsteniet
M.
Dobrowsky
P. H.
Van Deventer
A.
Khan
W.
Cloete
T. E.
2013
Domestic rainwater harvesting: microbial and chemical water quality and point-of-use treatment systems
.
Wat. Air. Soil. Pol
.
224
,
1629
.
Liaw
C.
Huang
W.
Tsai
Y.
Chen
J.
2008
Rainwater harvesting for domestic water supply and stormwater mitigation
.
Wat. Sci. Technol
.
52
(
9
),
191
197
.
Rahman
S.
Khan
M. T. R.
Akib
S.
Che Din
N. B.
Biswas
S. K.
Shirazi
S. M.
2014
Sustainability of rainwater harvesting system in terms of water quality
.
Sci. World. J
.
2014
,
10
.
Simmons
G.
Hope
V.
Lewis
G.
Whitmore
J.
Gao
W.
2001
Contamination of potable roof collected rainwater in Auckland, New Zealand
.
Water Res
.
35
(
6
),
1518
1524
.
Vialle
C.
Sablayrolles
C.
Lovera
M.
Jacob
S.
Huau
M. C.
Montrejaud-Vignoles
M.
2011
Monitoring of water quality from roof runoff: interpretation using multivariate analysis
.
Wat. Res
.
45
(
12
),
3765
3775
.