Skip to main content
  • Research article
  • Open access
  • Published:

Comparative analysis of solar pasteurization versus solar disinfection for the treatment of harvested rainwater

Abstract

Background

Numerous pathogens and opportunistic pathogens have been detected in harvested rainwater. Developing countries, in particular, require time- and cost-effective treatment strategies to improve the quality of this water source. The primary aim of the current study was thus to compare solar pasteurization (SOPAS; 70 to 79 °C; 80 to 89 °C; and ≥90 °C) to solar disinfection (SODIS; 6 and 8 hrs) for their efficiency in reducing the level of microbial contamination in harvested rainwater. The chemical quality (anions and cations) of the SOPAS and SODIS treated and untreated rainwater samples were also monitored.

Results

While the anion concentrations in all the samples were within drinking water guidelines, the concentrations of lead (Pb) and nickel (Ni) exceeded the guidelines in all the SOPAS samples. Additionally, the iron (Fe) concentrations in both the SODIS 6 and 8 hr samples were above the drinking water guidelines. A >99% reduction in Escherichia coli and heterotrophic bacteria counts was then obtained in the SOPAS and SODIS samples. Ethidium monoazide bromide quantitative polymerase chain reaction (EMA-qPCR) analysis revealed a 94.70% reduction in viable Legionella copy numbers in the SOPAS samples, while SODIS after 6 and 8 hrs yielded a 50.60% and 75.22% decrease, respectively. Similarly, a 99.61% reduction in viable Pseudomonas copy numbers was observed after SOPAS treatment, while SODIS after 6 and 8 hrs yielded a 47.27% and 58.31% decrease, respectively.

Conclusion

While both the SOPAS and SODIS systems reduced the indicator counts to below the detection limit, EMA-qPCR analysis indicated that SOPAS treatment yielded a 2- and 3-log reduction in viable Legionella and Pseudomonas copy numbers, respectively. Additionally, SODIS after 8 hrs yielded a 2-log and 1-log reduction in Legionella and Pseudomonas copy numbers, respectively and could be considered as an alternative, cost-effective treatment method for harvested rainwater.

Background

Several countries around the world utilise alternative water sources, such as rainwater harvesting (RWH) and surface water, to meet the increasing water demand and augment available water supplies. Rainwater harvesting in particular has been identified by the South African government as an alternative and sustainable water source that could provide water directly to households [1, 2]. Rainwater is considered a pure water source, however, during the harvesting process, it can become polluted with microorganisms and atmospheric particles such as, organic and inorganic matter (e.g. heavy metals and dust) [24]. Depending on the roof maintenance, leaves, animal faecal matter (which may contain chemicals such as phosphorous, nitrogen and trace elements) [4] and other debris particles, may also wash into the rainwater storage tank after a rain event and negatively affect the microbial quality of the tank water [46].

It has thus been concluded that stored harvested rainwater is not suitable for potable purposes due to the microbial quality in particular not complying with drinking water standards as established by the Department of Water Affairs and Forestry (DWAF) [7] and World Health Organization (WHO) [8] and it was recommended that harvested rainwater should be treated before utilisation as a primary water source [5, 9]. In developing countries, particularly, researchers seek cost- and time-effective treatment methods in order to improve the quality of harvested rainwater, for utilisation as a potable water source and for other domestic activities [6]. Solar disinfection (SODIS) and solar pasteurization (SOPAS) systems have been considered as efficient and cost-effective treatment methods for harvested rainwater [1, 6].

A SODIS system is based on the effect of ultra-violet (UV) light and heat from the sun, which inactivates microorganisms [6, 10]. A very simple example of a SODIS system is outlined by Amin and Han [1] and Amin et al. [6] where a transparent container is filled with harvested rainwater, placed onto a reflective surface and is exposed to direct sunlight for at least 6 to 8 hrs. Advantages of this system include cost-effectiveness and due to its simplicity it can be implemented worldwide [11]. Recent studies have also shown that SODIS improves the microbial quality of harvested rainwater [1, 6], although certain microorganisms and endospores may persist. Furthermore, the turbidity of the water may decrease the efficiency of the system due to the systems’ dependence on direct UV radiation penetration. Although the SODIS system is easier to implement than the SOPAS system, the efficiency of both systems decreases with cloudy weather conditions [1, 6, 10] and both systems may not improve the chemical quality of the harvested rainwater [10, 12].

A SOPAS system relies on the thermal effect (at least 70 °C), without UV radiation to inactivate microbes [13]. An example of a simple SOPAS system is the contemporary solar geyser, where water fills the borosilicate glass tubes, which is exposed to solar radiation. The energy which is obtained from solar radiation is transferred to the water which effectively heats up [14]. In addition, the time needed to treat water will decrease with an increase in temperature. Thus, the time required to treat water will decrease with a factor of 10 for every 10 °C increase in temperature above 50 °C [15]. This system is considered a cost-effective treatment method that is not influenced by the turbidity of the water [16, 17]. Research has also indicated that microbes will be inactivated when the water reaches a temperature of 55 °C or higher [6, 18, 19]. In a study conducted by Dobrowsky et al. [17], an Apollo™ SOPAS system (manufactured in China) successfully reduced the bacterial indicator counts in the rainwater samples pasteurized at the temperature ranges of 72 to 74 °C, 78 to 81 °C, and 90 to 91 °C, to below the detection limit (≥99.9%). Furthermore, Legionella spp. and Pseudomonas spp. were detected at the higher pasteurization temperatures (>78 °C), using the Polymerase Chain Reaction (PCR), however the viability of these organisms at temperatures higher than 72 °C was not confirmed. In a follow up study, Reyneke et al. [20] then utilised ethidium monoazide bromide quantitative polymerase chain reaction (EMA-qPCR) to verify that viable Legionella spp. were detected in solar pasteurized rainwater samples (>70 °C).

Legionellosis is a lung infection caused by Legionella spp. where the bacterium enters the lungs by inhalation of aerosolized contaminated water. It is well known that Legionella can proliferate at high temperatures [17, 21], however the growth temperature for Legionella is between 25 °C and 45 °C with an optimum temperature of 36 °C [22]. In a recent study conducted by Reyneke et al. [20] the research group showed that Legionella spp. are viable at temperatures higher than 70 °C. Numerous Pseudomonas spp. are associated with water environments as well as heated water sources such as hot tubs, physiotherapy and hydrotherapy pools and whirlpools [23, 24]. This is one of the most common opportunistic pathogens associated with nosocomial infections in individuals with a vulnerable immune system [23]. It normally enters the human body through a skin wound or during surgery where it is then taken up into the blood stream leading to bacteraemia that could cause pneumonia, endocarditis, osteomyelitis, gastrointestinal infections, urinary tract infections and is a leading cause of septicaemia [24, 25]. Pseudomonas is generally spread through contaminated water that comes into contact with a human host, or surgical equipment and the hands of hospital personnel that transfer it to a patient in the case of nosocomial infections [23].

Results obtained by Dobrowsky et al. [17] and Reyneke et al. [20] however, also indicated that significant concentrations of iron (Fe), aluminium (Al), lead (Pb) and nickel (Ni) may have been leaching from the 100 L stainless steel storage tank of the Apollo™ SOPAS system, which may have negatively affected the chemical quality of the treated rainwater. In the current study a new PhungamanziTM SOPAS system, which was designed and manufactured in South Africa and which consists of a 125 L high grade polyethylene storage tank, was utilised for the solar pasteurization of rainwater. The primary aim of the current study was to conduct a comparative analysis of the new SOPAS system versus SODIS for the treatment of rainwater. The treatment times of the SODIS systems included 6 and 8 hrs, while the treated rainwater for the SOPAS system was collected at different temperature ranges (70 to 79 °C; 80 to 89 °C; and 90 °C and above). To monitor the general microbial quality of the rainwater, indicator bacterial counts, including, Escherichia coli (E. coli), enterococci and faecal coliforms as well as the heterotrophic plate count (HPC), were determined using culture based methods. Chemical analysis was also performed (monitoring the concentration of cations and anions) in order to determine whether the treatment methods utilised alter the chemical quality of the rainwater. Finally, the efficiency of the two treatment methods in reducing the level of viable Legionella spp. and Pseudomonas spp. in roof harvested rainwater was analysed utilising EMA-qPCR. Ethidium monoazide bromide is a nucleic acid binding dye that can be used to bind to the deoxyribonucleic acid (DNA) of cells (after photoactivation) with damaged and permeable membranes (non-viable cells). The binding of the dye to the DNA prevents PCR amplification of the DNA and thereby leads to a strong signal reduction during qPCR as only the DNA from intact (viable) cells will be amplified [20, 26].

Methods

Description of the sampling site

A RWH system was installed on Welgevallen Experimental farm, Stellenbosch University (GPS co-ordinates: 33° 56′ 36.19″S, 18° 52′ 6.08″E), South Africa. The roof used as the catchment area was constructed from asbestos, while the gutter system leading to the polyethylene rainwater tank (2 000 L tank installed on a metal stand) was constructed from Chrysotile (white asbestos) (Fig. 1a). Furthermore, the sampling site is surrounded by trees and is located next to a dairy farm. However, no tree branches obstructed the catchment area.

Fig. 1
figure 1

a The SOPAS system utilised in the current study was connected to a RWH tank installed on a metal stand. A: Untreated RWH tank (capacity: 2 000 L), B: Inlet pipe leading into the SOPAS tank, C: High grade polyethylene tank (capacity: 125 L), D: 10 × High borosilicate glass collector tubes, E: Outlet pipe and water collection point. b The SODIS system with two polyethylene terephthalate (PET) bottles containing harvested rainwater. The SODIS system was constructed from a black polyethylene material that was lined with a reflective aluminium surface. The system was covered with a transparent Perspex lid to increase insulation

Solar pasteurization system

The PhungamanziTM solar pasteurization system (manufactured in South Africa) was donated to Stellenbosch University by Crest Organization, Stellenbosch. This SOPAS system was connected to the 2 000 L polyethylene RWH tank, which was installed on a metal stand so that rainwater was able to flow from the rainwater storage tank into the SOPAS system in a passive manner (Fig. 1a). The water from the RWH tank flowed through the system components (Fig. 1a) as follows; water flowed from the RWH tank (A) through a pipe (B) into the high grade polyethylene tank (C) of the solar system, which has a 125 L storage capacity. The water then moved through the high borosilicate glass cylinders (D) in order to capture heat. Due to the thermo-siphoning effect, as the water was heated, the water moved into the main storage tank. The pasteurized water was then collected from the outlet tap (E).

Solar disinfection system

Two SUNSTOVE 2000™ solar oven systems (Sunstove Organization, South Africa), were placed on the rooftop of the JC Smuts building (33° 55′ 51.7″S 18° 51′ 55.3″E) at Stellenbosch University, South Africa, for the solar disinfection of the rainwater samples. As indicated in Fig. 1b, the solar oven has a very simplistic design, with the inside of the system constructed from a reflective aluminium plate and a black polyethylene material enclosing the system. In addition, in order to trap solar radiation, the inner section of the system was covered with a transparent Perspex lid.

Sample collection

For both the SOPAS and SODIS systems, water samples were collected from July 2015 till October 2015, with a sampling event conducted one to four days after a rain event. Throughout the sampling period, for the SOPAS system, untreated rainwater (collected directly from RWH tank A) and solar pasteurized rainwater samples were collected in sterile 5 L polypropylene containers, respectively. Solar pasteurized samples were collected at the temperature ranges of 70 to 79 °C; 80 to 89 °C; and 90 °C and above. A MadgeTech TC101A thermocouple temperature Data Logger (MadgeTech, Inc.) was installed inside the SOPAS system in order to monitor the temperature of the treated rainwater for one month (01/08/2015 to 31/08/2015). The temperature data was obtained from the log tagger and analysed using the Data Logger Software Ver. 4.1.5 (Madge Tech, Inc.).

The SODIS treatment of rainwater was performed five times and for each sampling occasion, four sterile transparent 2 L polyethylene terephthalate (PET) bottles were filled to three-quarter capacity with roof harvested rainwater, obtained from the RWH tank A (Fig. 1a). Space was left in each bottle for aeration purposes and directly after collection each bottle was shaken for approximately 10 s in order to oxygenate the water [6, 27]. Two PET bottles were placed on the base of each respective SODIS system (Fig. 1b) and the one SODIS system was exposed to direct sunlight for 6 hrs, while the second SODIS system was exposed to direct sunlight for 8 hrs [28]. Furthermore, for each sampling occasion an untreated rainwater sample was also collected from tank A in a 5 L PET bottle.

The pH and temperature of each water sample was measured on site, using a hand-held pH meter (Milwaukee Instruments, Inc., USA) and mercury thermometer (ALLA® France, France), respectively. The daily temperature and rainfall data were obtained from the South African Weather Services (personal communication) and the solar irradiation data was obtained from the Stellenbosch Weather Services, Stellenbosch University, Faculty of Engineering (http://weather.sun.ac.za/).

Chemical analysis

The chemical quality, including cation and anion concentrations of untreated and pasteurized (SOPAS) rainwater samples, collected for the various temperatures (cations: 71 °C, 86 °C and 93 °C) was determined. In addition, the chemical quality of untreated and SODIS rainwater samples collected after 6 hrs of treatment (cations: 70 °C and 89 °C) and 8 hrs of treatment (cations: 63 °C and 86 °C), were also analysed. For the determination of cation and metal ion concentrations, Falcon™ 50 mL high-clarity polypropylene tubes (Corning Life Sciences, USA) containing polyethylene caps were pre-treated with 1% nitric acid before sampling. The cation and metal ion concentrations [aluminium (Al), chromium (Cr), copper (Cu), iron (Fe), manganese (Mn), vanadium (V), and zinc (Zn), amongst others] were then determined using inductively coupled plasma atomic emission spectrometry (ICP-AES) [29]. This analysis was completed by the Central Analytical Facility (CAF), Stellenbosch University.

Furthermore, the anion analyses [SOPAS: untreated and 71 °C; SODIS untreated and treated at 6 hrs (52 °C; 70 °C and 89 °C) and 8 hrs (63 °C and 86 °C)] of the samples were performed by PathCare Reference Laboratory (PathCare Park, Cape Town, South Africa). All anions including, chloride, fluoride, nitrate and nitrite, phosphate and sulphates were measured utilising a Thermo Scientific Gallery™ Automated Photometric Analyser. The turbidity [​Nephelometric Turbidity Units (NTU)] of selected (untreated and treated) water samples was also determined by PathCare Reference Laboratory (PathCare Park, Cape Town, South Africa).

Microbial analysis of treated and untreated rainwater samples

Enumeration of traditional indicator bacteria in rainwater samples

A serial dilution was prepared (10−1–10−3) for each rainwater sample collected during the sampling period [SOPAS (untreated and pasteurized samples) and SODIS (untreated and treated samples)] and using the spread plate method, 100 μL of the undiluted rainwater sample and each dilution (10−1 –10−3) was cultured in duplicate onto Slanetz and Bartley Agar (Oxoid, Hampshire, England) that was incubated for 44 - 48 hrs at 36 ± 2 °C, m-FC Agar (Merck, Darmstadt, Germany) that was incubated for 22 – 24 hrs at 35 ± 2 °C and R2A Agar (Oxoid, Hampshire, England) that was incubated for 72 – 96 hrs at 35 ± 2 °C, to enumerate enterococci, faecal coliforms and HPC, respectively.

For each sample, E. coli was enumerated by filtering a total volume of 100 mL (undiluted) through a sterile GN-6 Metricel® S-Pack Membrane Disc Filter (Pall Life Sciences, Michigan, USA) with a pore size of 0.45 μm and a diameter of 47 mm, at a filtration flow rate of approximately ≥ 65 mL/min/cm2 at 0.7 bar (70 kPa), in duplicate. The membrane filters were then incubated on Membrane Lactose Glucuronide Agar (MLGA) (Oxoid, Hampshire, England) at 35 ± 2 °C for 18 - 24 hrs.

Rainwater concentration, EMA treatment and DNA extraction

For each sampling event, 1 L rainwater sample [SOPAS (untreated and pasteurized samples) and SODIS (untreated and treated samples)] was concentrated as outlined in Reyneke et al. [20]. The concentrated rainwater samples utilised for Legionella spp. detection were treated with 2.5 μg/mL ethidium monoazide bromide (EMA) as previously described by Delgado-Viscogliosi et al. [30]. The same parameters were then utilised for the detection of Pseudomonas spp. in the concentrated rainwater samples. Following the addition of EMA, the samples were incubated on ice for 10 min followed by a 15 min halogen light exposure (keeping the samples on ice to avoid over-heating during the photoactivation step). The EMA treated samples were then washed with 1 mL NaCl (0.85%) followed by centrifugation (16 000 × g for 5 min). The DNA extractions were completed using the Soil Microbe DNA MiniPrep™ Kit (Zymo Research, USA) as per manufacturer’s instructions by first re-suspending the obtained pellet in the lysis solution and transferring the mixture to the ZR BashingBead™ Lysis Tubes.

Quantitative PCR for the detection of Legionella and Pseudomonas spp.

Following the EMA treatment and DNA extractions, EMA-qPCR was performed on a LightCycler®96 (Roche Applied Science, Mannheim, Germany) using the FastStart Essential DNA Green Master Mix (Roche Applied Science, Mannheim, Germany). To a final reaction volume of 20 μL, the following were added: 10 μL FastStart Essential DNA Green Master Mix (2x), 5 μL template DNA (diluted by 10 fold) and 0.4 μL of each primer (final concentration 200 nM) as previously described by Herpers et al. [31] for Legionella spp. and by Roosa et al. [32] for Pseudomonas spp.

For Legionella spp., the primers LegF (5′–CTAATTGGCTGATTGTCTTGAC–3′) and LegR (5′–CAATCGGAGTTCTTCGTG–3′) were utilised to amplify a 259 bp product of the 23S rRNA gene [31]. The amplification conditions for Legionella spp. were as follows: initial denaturation at 95 °C for 10 min, followed by 50 cycles of denaturation at 95 °C for 15 s, annealing at 60 °C for 15 s and extension at 72 °C for 11 s.

For Pseudomonas spp., the primers PS1 (5′-ATGAACAACGTTCTGAAATTC-3′) and PS2 (5′-CTGCGGCTGGCTTTTTCCAG-3′) were utilised to amplify a 249 bp product of the Pseudomonas lipoprotein oprI gene [33]. The amplification conditions for Pseudomonas spp. were as follows: initial denaturation at 95 °C for 10 min, followed by 50 cycles of denaturation at 94 °C for 30 s, annealing at 58 °C for 30 s and extension at 72 °C for 30 s.

The standard curves for the Legionella spp. qPCR assays were produced by amplifying the 23S rRNA gene of Legionella pneumophila ATCC 33152, using primers LegF and LegR. In addition, the standard curves for the Pseudomonas spp. qPCR assays were produced by amplifying the lipoprotein oprI gene of P. aeruginosa ATCC 27853, using primers PS1 and PS2. The PCR products were then purified using the DNA Clean & Concentrator™-5 Kit (Zymo Research) and were verified by DNA sequencing followed by quantifying the DNA in triplicate using the NanoDrop® ND-1000 (Nanodrop Technologies Inc., Wilmington, Delaware, USA). A serial 10-fold dilution (Legionella spp.: 108 to 101; Pseudomonas spp.: 109 to 100) of the PCR products was prepared in order to generate the standard curves, where the regression coefficient (R 2) was kept higher than 0.98 and 1.00 for Legionella and Pseudomonas spp., for each experiment, respectively. For Legionella spp. and Pseudomonas spp. detection, a concentration of 1.00 × 108 and 1.00 × 109 gene copies/μL was prepared for the dilution with the highest copy number, respectively, while a concentration of 1.00 × 101 and 1.00 × 100 gene copies/μL was prepared for the dilution with the lowest copy number. The standard curves were generated by plotting quantitative cycle (Cq) values versus the log concentrations of standard DNA, as previously described by Chen and Chang [34], for determining the copy number of the 23S rRNA gene in Legionella spp. and the copy number of the lipoprotein oprI gene in Pseudomonas spp.in all samples analysed. Melt curve analysis was included for both Legionella and Pseudomonas spp. SYBR green real-time PCR assays in order to verify the specificity of the primer set by ramping the temperature from 65 to 97 °C at a rate of 0.2 °C/s with continuous fluorescent signal acquisition at 5 readings/°C.

The determination of bacterial removal efficiency of the treatment systems

The bacterial removal efficiency of each treatment system (SOPAS and SODIS) was obtained by comparing the bacterial counts obtained from the samples collected before treatment and the average bacterial counts obtained from samples collected after treatment. The percentage reduction was calculated using Eq. 1 [35].

$$ \mathrm{Percentage}\ \mathrm{reduction} = 100\ \hbox{-} \left(\mathrm{Survivor}\ \mathrm{count}/\mathrm{Initial}\ \mathrm{count}\right) \times 100 $$
(1)

Statistical analysis

The statistical software package Statistica™ Ver. 11.0 (Stat Soft Inc., Tulsa, USA) was used for the evaluation of the microbial analysis and the temperature of the collected rainwater samples (untreated, pasteurized and disinfected). To test the significance of the data set, an ANOVA analysis was performed for evenly distributed data while for non-evenly distributed data, a spearman rank order correlation was performed. A significant level of 5% was used as a standard in the hypothesis tests [36], while in all tests a p-value of <0.05 was considered statistically significant.

Results

Physico-chemical parameters for water samples collected from SOPAS and SODIS treatment systems

The temperature of the solar pasteurized water samples collected throughout the sampling period (n = 6) ranged from 71 °C (July 2015) to the highest temperature of 93 °C (October 2015). The temperature of the SODIS samples were also monitored after 6 hrs and 8 hrs of treatment, respectively, with the temperature of the 6 hr samples (n = 5) ranging from 52 °C (July 2015) to 89 °C (October 2015) and the temperature of the 8 hr SODIS samples (n = 5) ranging from 63 °C (August 2015) to 86 °C (October 2015). For both the SOPAS and the SODIS treatment, the highest total monthly rainfall over the sampling period was recorded in July 2015 (174.4 mm), which then decreased to 67.6 mm in August 2015, increased to 78.2 mm in September and then decreased to the lowest rainfall recorded in October 2015 (10.0 mm).

For the SODIS treatment, an overall average daily ambient temperature of 24.3 °C was recorded during the sampling period, with the lowest temperature of 17.2 °C recorded during July 2015 and the highest temperature of 29.7 °C recorded during October 2015. The temperature of the untreated water samples (collected directly from the RWH tank), averaged 20.2 °C, with the lowest temperature measured as 17.2 °C (July 2015) and the highest temperature measured as 25.2 °C (October 2015). In addition, an overall average pH of 8.0 was recorded for the untreated water samples, while an overall pH of 8.1 was recorded for the solar disinfected water samples after 6 hrs and 8 hrs of treatment, respectively.

For the SOPAS treatment, an overall average daily ambient temperature of 25.5 °C was recorded during the sampling period, with the lowest temperature of 17.2 °C recorded during July 2015 and the highest temperature of 30.6 °C recorded during October 2015. Similarly, the temperature of the untreated water samples (collected directly from the RWH tank), averaged 24.7 °C, with the lowest temperature measured as 19 °C (July 2015) and the highest temperature measured as 29.0 °C (October 2015). In addition, an overall average pH of 8.0 was recorded for the untreated water samples, while an overall pH of 7.6 was recorded for the solar pasteurized water samples.

Furthermore, a data logger probe was used to measure the water temperature inside the SOPAS system for a period of one month (01/08/2015 to 31/08/2015) (results not shown). An overall average ambient temperature of 21.1 °C was obtained with the lowest temperature recorded as 7.4 °C and the highest temperature recorded as 39.0 °C. In addition, the water temperature inside the SOPAS system had an overall average of 56.9 °C during the monitored month which ranged from 40.1 °C to 82.9 °C. Solar irradiation data was obtained from Stellenbosch Weather Service (Engineering Facility) and ranged from 0.01 W/m2 to 881.37 W/m2 with an overall average of 297.27 W/m2. A direct positive correlation between the ambient temperature and solar irradiation (R = 0.69; p < 0.05) and the temperature of the water inside the system (R = 0.20; p < 0.05) was also obtained.

Chemical analysis of untreated and treated rainwater samples

Chemical analysis of the SOPAS rainwater samples

Untreated and solar pasteurized water samples (71 °C) collected during the first sampling event were analysed for their anion concentrations (results not shown). All anion concentrations of the untreated water sample and the solar pasteurized water sample were within the drinking water guidelines as stipulated by Australian Drinking Water Guidelines (ADWG) [37], DWAF [7] and South African National Standards (SANS) 241 [38]. A previous study conducted by Dobrowsky et al. [17] also indicated that there was no significant difference between the anion concentrations in the untreated and solar pasteurized water samples (55 to 91 °C). Anion analyses were thus not conducted on the untreated and solar pasteurized rainwater samples collected during the remainder of the sampling period. The turbidity of the untreated and pasteurized rainwater samples was also measured and according to DWAF [7], SANS 241 [38] and ADWG [37], the turbidity should not exceed 1.00 NTU. For both the untreated and solar pasteurized water sample, the turbidity was measured as 0.00 NTU, thus the turbidity complied with the respective drinking water guidelines.

The metal ions and cation concentrations were determined for pasteurized water samples collected at 71 °C, 86 °C and 93 °C and the corresponding unpasteurized samples (Table 1). The concentrations of the metal ions and cations in the untreated and SOPAS treated rainwater samples were below the recommended guidelines as stipulated by ADWG [37], DWAF [7] and SANS 241 [38], with the exception of Pb and Ni. However, while all the before and after SOPAS treatment samples were within the stipulated guidelines for Fe concentrations, the concentration of Fe in the before treatment sample (172.92 μg/L), collected with the corresponding 71 °C SOPAS sample, exceeded the DWAF [7] drinking water guideline of <100 μg/L. The Fe concentration in the SOPAS treatment sample collected at 71 °C then decreased significantly (p < 0.05) to 29.19 μg/L.

Table 1 Cation and metal ion concentrations of the untreated water samples and the corresponding solar pasteurized water samples collected at various temperatures compared to the recommended drinking water guidelines

In addition, while Ni was within the SANS 241 [38] drinking water guideline in all the water samples analysed, it was detected above the drinking water guideline (<20 μg/L) according to ADWG [37] for all three samples collected after pasteurization (71 °C, 30.00 μg/L; 86 °C, 26.46 μg/L; and 93 °C, 25.59 μg/L). Furthermore, Pb was detected above the drinking water guideline stipulated by ADWG [37], DWAF [7] and SANS 241 [38] for all three samples collected after pasteurization (71 °C, 86 °C and 93 °C) with a concentration of 74.12 μg/L, 26.30 μg/L and 19.67 μg/L recorded, respectively. It should however, be noted that both the Ni and Pb concentrations decreased with an increase in SOPAS temperature.

Chemical analysis of the SODIS rainwater samples

All anion concentrations of the SODIS rainwater samples [untreated and treated at 6 hrs (anions: 52 °C; 70 °C and 89 °C) and 8 hrs (anions: 63 °C and 86 °C)], were within the drinking water guidelines as stipulated by ADWG [37], DWAF [7] and SANS 241 [38] (results not shown). In addition, there was no significant (p > 0.05) increase in the anion concentrations after treatment. While the turbidity measurements of all the water samples before and after treatment, were within the 1.00 NTU recommended guideline [7, 37, 38], the turbidity of samples collected during the first sampling event in August 2015, were not within the drinking water guidelines. It should however be noted that the untreated water sample had a turbidity of 1.90 NTU, which already exceed the drinking water guidelines. After 6 hrs of treatment by SODIS (70 °C), the turbidity increased to 2.14 NTU, while the sample treated for 8 hrs (63 °C) had a turbidity of 2.09 NTU.

The metal ions and cation concentrations were measured for representative SODIS sampling events [6 hrs (70 °C and 89 °C) and 8 hrs (63 °C and 86 °C) after treatment] and their corresponding untreated water sample (Table 2). Similar to the results obtained for the SOPAS treated water samples, the concentrations of all the metal ions and cations, in the untreated and SODIS rainwater samples were within the recommended guidelines as stipulated by ADWG [37], DWAF [7] and SANS 241 [38]. However, the concentrations of Fe in the untreated and treated (6 and 8 hrs) samples were significantly (p < 0.05) higher compared to the drinking water guidelines as stipulated by ADWG [37], DWAF [7] and SANS 241 [38]. The first untreated sample had an Fe concentration of 571.26 μg/L, which increased to 729.71 μg/L after 6 hrs of treatment (70 °C) and then decreased to a concentration of 645.39 μg/L after 8 hrs of treatment (63 °C). Similarly, an Fe concentration of 112.60 μg/L was recorded in the untreated sample corresponding to the temperature ranges of 89 °C (6 hrs of treatment) and 86 °C (8 hrs of treatment), with the Fe concentration increasing to 1015.32 μg/L (6 hrs) and decreasing to 505.35 μg/L after 8 hrs.

Table 2 Cation and metal ion concentrations of the untreated water samples and the corresponding SODIS treated water samples collected after 6 and 8 hrs compared to the recommended drinking water guidelines

Indicator bacterial counts in untreated and treated rainwater samples

Indicator bacteria detected in untreated and SOPAS rainwater samples

For each untreated water sample and the corresponding pasteurized sample collected at various temperatures ranging from 71 °C to 93 °C, water samples were analysed for the presence of indicator bacteria including E. coli, HPC, enterococci and faecal coliforms (Table 3). Enterococci and faecal coliforms were not detected in any of the untreated as well as the pasteurized rainwater samples. However, the HPC for the untreated water samples ranged from a minimum of 7.05 × 106 CFU/100 mL to a maximum of 7.4 × 107 CFU/100 mL and were reduced to below the detection limit (<1 CFU/mL) after pasteurization for all temperature ranges (71 °C to 93 °C).

Table 3 Indicator counts for solar pasteurized water samples and the corresponding untreated water samples collected at various temperatures

Escherichia coli were also detected in all the untreated water samples with a minimum of 2 CFU/100 mL to a maximum of 3 CFU/100 mL recorded. Similarly, E. coli counts were reduced to below the detection limit after pasteurization (71 °C to 93 °C). For the untreated rainwater samples, both the HPC and the E. coli counts exceeded the drinking water guidelines as stipulated by the DWAF [7]. However, after pasteurization a >99% reduction in indicator counts was observed for all the pasteurized rainwater samples and the counts were within the DWAF [7] standards.

Indicator bacteria detected in untreated and SODIS rainwater samples

For each untreated water sample and the corresponding solar disinfected water sample, collected at various temperatures ranging from 52 °C to 89 °C and 63 °C to 86 °C treated for 6 hrs and 8 hrs, respectively, water samples were analysed for the presence of indicator bacteria including E. coli, HPC, enterococci and faecal coliforms. Similar to results obtained for the SOPAS samples, enterococci and faecal coliforms were not detected in any of the untreated as well as both the 6 hr and 8 hr disinfected water samples. However, the HPC in all the untreated water samples ranged from a minimum of 7.05 × 106 CFU/100 mL to a maximum of 9.95 × 107 CFU/100 mL and was reduced to below the detection limit (< 1 CFU/mL) after 6 hrs of disinfection (Table 4). Escherichia coli were also detected in all the untreated water samples (6 hrs of treatment) with counts ranging from a minimum of 2 CFU/100 mL to a maximum of 4 CFU/100 mL. The E. coli counts were then also reduced to below the detection limit after 6 hrs of disinfection.

Table 4 Indicator counts for solar disinfected water samples collected after 6 hrs of treatment and the corresponding untreated water samples collected at various temperatures

The HPC in all the untreated water samples corresponding to 8 hrs of SODIS treatment ranged from a minimum of 1.45 × 107 CFU/100 mL to a maximum of 9.95 × 107 CFU/100 mL and was reduced to below the detection limit (< 1 CFU/mL) after 8 hrs of disinfection (Table 5). Escherichia coli were also detected in all the untreated water samples (8 hrs of treatment) with counts ranging from a minimum of 2 CFU/100 mL to a maximum of 13 CFU/100 mL. The E. coli counts were then also reduced to below the detection limit (< 1 CFU/mL) after 8 hrs of disinfection.

Table 5 Indicator counts for solar disinfected water samples collected after 8 hrs of treatment and the corresponding untreated water samples collected at various temperatures

For the untreated rainwater samples (6 hrs and 8 hrs), both the HPC and E. coli counts exceeded the drinking water guidelines stipulated by the DWAF [7]. However, after both 6 hrs and 8 hrs of SODIS treatment a significant (p < 0.05) reduction (>99%) in indicator counts was observed and all counts were within the DWAF [7] guidelines.

Quantitative PCR for the detection of Legionella spp.

Quantitative PCR for the detection of viable Legionella spp. in SOPAS samples

The presence of viable Legionella cells in the untreated and corresponding treated SOPAS samples were determined using qPCR assays in conjunction with the EMA pre-treatment. A standard curve was constructed with a linear range of quantification from 108 to 101 gene copies per µL using the LightCycler®96 software Ver. 1.1.0.1320 (Roche Diagnostics International Ltd). A qPCR efficiency of 1.86 (93%) was obtained, with a linear regression coefficient (R 2) value of 0.98. Using the standard curve, viable Legionella copy numbers were quantified in the untreated and corresponding solar pasteurized water samples collected at various temperatures and are represented as 23S rRNA gene copies per mL (Fig. 2a).

Fig. 2
figure 2

Viable (a) Legionella spp. and (b) Pseudomonas spp. gene copy numbers recorded in corresponding untreated and solar pasteurized rainwater samples collected at various temperatures. The overall average of Legionella and Pseudomonas spp. gene copy numbers of the untreated rainwater samples is indicated by a dotted line, while the overall average of Legionella and Pseudomonas spp. gene copy numbers of the treated rainwater samples is indicated by a dashed line. Error bar: SE (1SD) of duplicate samples analysed

A significant reduction (p < 0.05) in viable Legionella copy numbers after solar pasteurization of the rainwater samples collected at all temperature ranges (70 to 79 °C, 80 to 89 °C and 90 °C and above) was obtained (Fig. 2a). For the temperature range of 70 to 79 °C, an average of 1.74 × 105 copies/mL was observed for the untreated water samples, which decreased to an average of 6.15 × 103 copies/mL for the pasteurized water samples. For the temperatures ranging from 80 to 89 °C, an average of 4.79 × 105 copies/mL was observed for the untreated water samples, compared to an average of 4.57 × 104 copies/mL obtained for the pasteurized water. Lastly, for the temperatures 90 °C and above, an average of 6.49 × 105 copies/mL for the untreated water samples was obtained, which decreased to an average of 8.92 × 103 copies/mL for the pasteurized water samples.

At the lowest (70 to 79 °C) and highest (90 °C and above) pasteurization temperature ranges, a percentage reduction of 99.97% and 96.83% was observed (2-log reduction) in Legionella copy numbers, respectively, while the lowest percentage reduction (89.76%) in copy numbers was observed for the 80 to 89 °C temperature range (1-log reduction).

Quantitative PCR for the detection of viable Legionella spp. in SODIS samples

The same standard curve as described for the quantification of Pseudomonas copy numbers in th e untreated and SOPAS treated samples, was utilised to quantify viable Legionella copy numbers per mL for the untreated and corresponding solar disinfected water samples after 6 and 8 hrs (various temperatures recorded), respectively.

The results obtained for the qPCR assays showed that there was a reduction in viable Legionella copy numbers after SODIS treatment for 6 hrs (Fig. 3a). The lowest percentage reduction (24.46%) in Legionella copy numbers was observed for a solar disinfected sample with a temperature of 68 °C, where Legionella copy numbers decreased from 1.56 × 107 copies/mL for the untreated sample to 1.18 × 107 copies/mL for the solar disinfected sample. The highest percentage reduction (74.09%) in copy numbers was observed for a solar disinfected sample with a temperature of 75 °C, where Legionella copy numbers decreased from 1.76 × 105 copies/mL for the untreated sample to 4.56 × 104 copies/mL for the solar disinfected sample. A significant (p < 0.05) reduction (72.6%) in Legionella copy numbers was also observed at 89 °C, where Legionella copy numbers of 4.13 × 104copies/mL were observed for the untreated sample and then decreased to 1.13 × 104 copies/mL after SODIS at 6 hrs.

Fig. 3
figure 3

Viable (a) Legionella spp. and (b) Pseudomonas spp. gene copy numbers recorded in corresponding untreated and solar disinfected (for 6 hrs) rainwater samples collected at various temperatures. Viable (c) Legionella spp. and (d) Pseudomonas spp. gene copy numbers recorded in corresponding untreated and solar disinfected (for 8 hrs) rainwater samples collected at various temperatures. The overall average of Legionella and Pseudomonas spp. gene copy numbers of the untreated rainwater samples is indicated by a dotted line, while the overall average of Legionella and Pseudomonas spp. gene copy numbers of the treated rainwater samples is indicated by a dashed line. Error bar: SE (1SD) of duplicate samples analysed

The results obtained for the qPCR assays, indicated that overall there was a 2-log reduction in viable Legionella copy numbers (except 63 °C sample) after SODIS of 8 hrs for the rainwater samples collected at temperatures ranging from 67 to 86 °C (Fig. 3c). The lowest percentage reduction (50.07%) in copy numbers was observed for a SODIS temperature of 86 °C, where 4.12 × 104 copies/mL was observed in the untreated sample compared to the solar disinfected sample where 2.06 × 104 copies/mL was recorded. The highest percentage reduction (99.97%) in copy numbers was observed for a solar disinfected temperature of 76 °C, where 1.56 × 107 copies/mL was observed in the untreated sample compared to the solar disinfected sample where 4.03 × 103 copies/mL was recorded. For the temperatures of 67 °C and 72 °C, a percentage reduction in copy numbers of 75.43% and 75.41% was recorded, respectively. However, an increase in Legionella spp. copy numbers was observed for the solar disinfected sample with a temperature of 63 °C, where 3.32 × 105copies/mL was observed in the untreated sample compared to 9.54 × 105 copies/mL recorded in the solar disinfected sample.

Quantitative PCR for the detection of Pseudomonas spp.

Quantitative PCR for the detection of viable Pseudomonas spp. in SOPAS samples

The quantification of viable Pseudomonas cells in the untreated and corresponding treated SOPAS samples was determined using qPCR assays in conjunction with the EMA pre-treatment. A standard curve was constructed with a linear range of quantification from 109 to 100 gene copies per μL using the software LightCycler®96 Version 1.1.0.1320 (Roche Diagnostics International Ltd). A qPCR efficiency of 1.83 (92%) was obtained, with a linear regression coefficient (R 2) value of 1.00. Using the standard curve, viable Pseudomonas copy numbers were quantified in the untreated and corresponding solar pasteurized (treated) water samples collected at various temperatures and are represented as Pseudomonas lipoprotein oprI gene copies per mL (Fig. 2b).

A significant reduction (p < 0.05) in viable Pseudomonas copy numbers after solar pasteurization of the rainwater samples collected at all temperature ranges (70 to 79 °C, 80 to 89 °C and 90 °C and above) was obtained (Fig. 2b). For the temperature range of 70 to 79 °C, an average of 2.07 × 107 copies/mL was observed for the untreated water samples, which decreased to an average of 1.13 × 105 copies/mL for the pasteurized water samples. For the temperatures ranging from 80 to 89 °C, an average of 4.37 × 107 copies/mL was observed for the untreated water samples, compared to an average of 1.84 × 105 copies/mL obtained for the pasteurized water. Lastly, for the temperatures ranging from 90 °C and above, an average of 3.57 × 107 copies/mL for the untreated water samples was obtained, which decreased to an average of 7.31 × 104 copies/mL for the pasteurized water samples. It should however be noted that while an average of 2.45 × 107 Pseudomonas copies/mL was observed in the untreated water sample (collected with the 93 °C SOPAS sample), no amplification of the oprI gene was recorded in the 93 °C pasteurized water sample resulting in a Cq value below detection limit obtained (not presented on Fig. 2b).

For the pasteurization temperature ranges of 70 to 79 °C and 80 to 89 °C a reduction of 99.45% and 99.58% was observed in Pseudomonas copy numbers, respectively, thus a 2-log reduction was observed for both these temperature ranges. In addition, the greatest percentage reduction of 99.80% (3-log reduction) in copy numbers was observed for the 90 °C and above temperature range.

Quantitative PCR for the detection of viable Pseudomonas spp. in SODIS samples

The same standard curve as described for the quantification of Pseudomonas copy numbers in th e untreated and SOPAS treated samples, was utilised to quantify viable Pseudomonas copy numbers per mL for the untreated and corresponding solar disinfected water samples after 6 and 8 hrs (various temperatures recorded), respectively.

The results obtained for the qPCR assays showed that there was a reduction in viable Pseudomonas copy numbers after SODIS treatment for 6 hrs (Fig. 3b). The lowest percentage reduction (5.53%) in Pseudomonas copy numbers was observed for a solar disinfected sample with a temperature of 52 °C, where Pseudomonas copy numbers decreased from 1.63 × 107 copies/mL for the untreated sample to 1.54 × 107 copies/mL for the solar disinfected sample. The highest percentage reduction (93.73%) in copy numbers was observed for a solar disinfected sample with a temperature of 89 °C, where Pseudomonas copy numbers decreased from 6.90 × 107 copies/mL for the untreated sample to 4.33 × 106 copies/mL for the solar disinfected sample yielding a 1-log reduction.

The results obtained for the qPCR assays, indicated that there was an overall 1-log reduction in viable Pseudomonas copy numbers after SODIS of 8 hrs (Fig. 3d) for the rainwater samples collected at temperatures ranging from 63 to 86 °C. The lowest percentage reduction (14.37%) in copy numbers was observed for a SODIS temperature of 63 °C, where 3.73 × 107 copies/mL was observed in the untreated sample compared to the solar disinfected sample where 3.19 × 107 copies/mL was recorded. The highest percentage reduction (96.12%) in copy numbers was observed for a solar disinfected temperature of 86 °C, where 6.90 × 107 copies/mL was observed in the untreated sample compared to the solar disinfected sample where 2.68 × 106 copies/mL was recorded.

Discussion

The efficiency of two solar based treatment systems (SOPAS and SODIS) were evaluated for the treatment of roof harvested rainwater. Numerous chemical and microbial parameters were investigated in order to determine which system effectively improved the overall quality of the harvested rainwater to within drinking water guidelines. Chemical analysis of the solar pasteurized and corresponding untreated rainwater samples then indicated that all cation (with the exception of Pb and Ni) and anion concentrations were within the drinking water guidelines as stipulated by the ADWG [37], DWAF [7] and SANS 241 [38]. Nickel and Pb were detected in all three pasteurization water samples (71 °C, 86 °C and 93 °C) analysed at concentrations exceeding the drinking water guidelines. Although the SOPAS system has a storage tank constructed from high grade polyethylene, it contains SABS approved Ni coated dezincification resistant (DZR) brass connector points utilised for mounting purposes. Nickel could have thus leached from the Ni coated brass metal during exposure to high temperatures in the SOPAS system. However, only long term exposure to Ni at high concentrations may be toxic to humans as the concentration of beta-microglobulin increases in the kidneys [37]. In addition, the Pb detected could have leached from the surface of the polyethylene storage tank into the water, as the high grade polyethylene storage tank is treated with Pb (personal communication, Crest Organization) which acts as a stabilizer and is often used to treat polyethylene surfaces exposed to high temperature [39]. Significantly high concentrations of Pb have a severe effect on the human central nervous system and results in the interference with calcium metabolism (bone formation), red blood cell production and contributes to kidney failure [37].

For the SODIS system, chemical analysis revealed that the cation (with the exception of Fe) and anion concentrations, were also within the drinking water guidelines as stipulated by the ADWG [37], DWAF [7] and SANS 241 [38]. It should however, be noted that the untreated water samples had iron concentrations which exceeded the drinking water guidelines. These concentrations then increased in the SODIS samples treated for 6 and 8 hrs, respectively. Suib [40] indicated that the synergistic effect of solar photons and hydrogen peroxide generates hydroxide inside microbial cells by Fenton’s reaction, causing Fe and hydrogen peroxide to flow through the cell membrane. Furthermore, when cells are irradiated with near UV photons, an increase in ferrous (Fe2+) iron occurs due to increased membrane permeability, resulting in an increased Fe concentration in the surrounding environment. As SODIS uses both heat and UV to treat the water samples, this phenomenon could have been observed in the treated water samples.

Numerous studies have indicated that the microbial quality of harvested rainwater does not comply with drinking water guidelines [18, 41, 42]. The untreated rainwater, SOPAS and SODIS rainwater samples were thus analysed for the presence of the indicator bacteria E. coli, HPC, enterococci and faecal coliforms. Escherichia coli and HPC were detected in all the untreated water samples collected for SOPAS analysis, and were effectively reduced (>99%) to below the detection limit in all the samples collected at the various temperature ranges (71 °C to 93 °C). These results correlate with a study conducted by Dobrowsky et al. [17], where the research group showed that indicator counts in solar pasteurized water were reduced to below the detection level at temperatures of 72 °C and above. Similar to the results obtained for the SOPAS system, the E. coli and HPC counts recorded in the untreated water samples were also above the drinking water guidelines as stipulated by DWAF [7] and were reduced to below the detection limit after 6 and 8 hrs of SODIS treatment, with a minimum final temperature of 52 °C and 63 °C recorded, respectively. A study conducted by Berney et al. [43] showed that SODIS with strong irradiation conditions of up to 6 hrs disrupts a sequence of basic cellular functions in E. coli that leads to cell death. Overall the results thus indicate that the SOPAS system and SODIS systems (6 and 8 hrs of treatment), successfully reduced indicator bacteria numbers by >99%, at a minimum temperature of 71 °C for the SOPAS system and 52 °C for the SODIS system. These results correlate to a study conducted by Spinks et al. [18] where the research group suggested that a minimum temperature of 55 °C was sufficient to eliminate enteric pathogenic bacteria in water samples.

A poor correlation between indicator microorganisms and opportunistic bacteria has however, been reported [4446] as previous studies have shown that opportunistic bacteria, such as Legionella and Pseudomonas spp. amongst others, persist in roof harvested rainwater when low indicator counts are recorded [17, 42]. Oliver [47] then indicated that opportunistic pathogenic bacteria such as Legionella spp. are able to enter a viable but non-culturable state and therefore in the current study, EMA-qPCR assays were utilised to test for the presence and viability of these organisms in solar pasteurized and solar disinfected treated rainwater samples. Although conventional PCR can effectively be utilised as a presence/absence indicator of a particular gene or organism, it cannot be used to indicate the viability of the organism detected. In contrast, EMA-qPCR can be used to analyse for the presence and the viability of an organism and is considered a beneficial method for the detection and quantification of intact microorganisms [20, 48].

The EMA-qPCR assays indicated that a significant (p < 0.05) reduction (94.70%) in viable Legionella copy numbers was obtained after SOPAS and yielded a 2-log reduction overall. For the SODIS system, Legionella copy numbers also decreased in samples treated for 6 and 8 hrs, respectively. In addition, treatment after 8 hrs yielded a greater decrease (75.22%) in copy numbers (2-log reduction) in comparison to treatment for 6 hrs (maximum of 1-log reduction for the various temperatures), where a 50.60% reduction was observed. However, an increase in copy numbers was obtained for one solar disinfection sample (63 °C) treated for 8 hrs. It is known that Legionella can form associations with protozoa where they exist as intracellular parasites and are able to proliferate at temperatures from 50 °C to 65 °C due to the presence of heat shock proteins [49, 50]. Legionella spp. are therefore able to out compete other organisms and survive at these high temperatures (>90 °C) [51]. Moreover, a study conducted by Vervaeren et al. [50] showed that L. pneumophila is able to proliferate in heat treated water (up to a temperature of 70 °C). According to a study conducted by Hussong et al. [52] viable but non-culturable Legionella spp. also regain culturability and remain pathogenic when favourable conditions arise.

The EMA-qPCR assays for Pseudomonas yielded similar results to those obtained for Legionella. A reduction of 99.61% (3-log reduction) in viable Pseudomonas copy numbers was obtained after SOPAS treatment. In addition, SODIS treatment after 8 hrs yielded a greater reduction of 58.31% in viable copy numbers of Pseudomonas spp. in comparison to treatment for 6 hrs (47.27%). It is hypothesized that samples treated for 8 hrs were exposed to UV irradiation for an extended time period resulting in a greater microbial reduction. Furthermore, it is well known that Pseudomonas can enter a viable but non-culturable state [53] and results obtained in the current study indicated that Pseudomonas spp. remain viable at a temperature of 89 °C after treatment by SODIS. Several studies [5456] have thus utilised the addition of a photocatalytic material, to enhance the effect of microbial inactivation over a wide range of microorganisms and thereby increase the efficiency of a SODIS system. Titanium dioxide (TiO2) is considered the most suitable photocatalyst due to the lack of toxicity and chemical and photochemical stability, however further research is needed to determine to potability of TiO2 treated water [57].

Results obtained in the current study indicated that SOPAS treatment yielded a greater reduction in viable Legionella and Pseudomonas spp. (94.70% and 99.61%, respectively) copy numbers, compared to SODIS treatment after 6 (50.60% for Legionella spp. and 47.27% for Pseudomonas spp.) and 8 hrs (75.22% for Legionella spp. and 58.31% for Pseudomonas spp.). While not significant, treatment with SOPAS yielded a lower reduction in viable Legionella copy numbers compared to Pseudomonas copy numbers. It is hypothesized that Legionella spp. may have been able to persist due to: the presence of heat shock proteins to protect them from high temperatures; associations with amoebae species; and the formation of biofilms [58].

Conclusions and future research

Based on the indicator count analysis, treatment of harvested rainwater with both SOPAS and SODIS improved the microbial quality of rainwater and the water could be utilised for irrigation and domestic purposes such as cooking, laundry and washing. The SOPAS system can however, effectively treat larger volumes of rainwater in comparison to the SODIS system and based on the EMA-qPCR results obtained in the current study, SOPAS was the most effective for the reduction of viable Legionella and Pseudomonas spp. copy numbers in harvested rainwater. However, depending on the material utilised to construct the storage tank, metals and chemicals may leach into the water when temperatures higher than 71 °C are achieved inside the SOPAS system. In contrast, SODIS systems function as batch culture systems and are more cost-effective and easier to operate and maintain. Future research should however, focus on up-scaling SODIS systems to allow for the efficient treatment of larger volumes of rainwater.

Abbreviations

ANOVA:

Analysis of variance

Ave.:

Average

BDL:

Below detection limit

bp:

Base pairs

CAF:

Central analytical facility

CFU:

Colony forming units

Cq :

Quantitative cycle

DNA:

Deoxyribonucleic acid

DWAF:

Department of Water Affairs and Forestry

DZR:

Dezincification resistant

EMA:

Ethidium monoazide

EMA-qPCR:

Ethidium monoazide bromide quantitative polymerase chain reaction

HPC:

Heterotrophic plate count

Hrs:

Hours

ICP-AES:

Inductively coupled plasma atomic emission spectrometry

m-FC:

Membrane filter faecal coliform

MLGA:

Membrane lactose glucuronide agar

NTU:

Nephelometric turbidity unit

PCR:

Polymerase chain reaction

PET:

Polyethylene terephthalate

qPCR:

Quantitative PCR

R 2 :

Linear regression coefficient

R2A:

Reasoner’s 2A agar

rRNA:

Ribosomal ribonucleic acid

RWH:

Rainwater harvesting

SANS:

South African National Standards

SD:

Standard deviation

SODIS:

Solar disinfection

SOPAS:

Solar pasteurization

USEPA:

United States Environmental Protection Agency

UV:

Ultra-violet

WHO:

World Health Organization

References

  1. Amin M, Han M. Roof-harvested rainwater for potable purposes: application of solar collector disinfection (SOCO-DIS). Water Res. 2009;43(20):5225–35.

    Article  CAS  PubMed  Google Scholar 

  2. De Kwaadsteniet M, Dobrowsky PH, Van Deventer A, Khan W, Cloete TE. Domestic rainwater harvesting: microbial and chemical water quality and point-of-use treatment systems. Water Air Soil Poll. 2013;224(7):1–19.

    Article  Google Scholar 

  3. Helmreich B, Horn H. Opportunities in rainwater harvesting. Desalination. 2009;248(1):118–24.

    Article  CAS  Google Scholar 

  4. Abbasi T, Abbasi SA. Sources of pollution in rooftop rainwater harvesting systems and their control. Crit Rev Env Sci Tec. 2011;41(23):2097–167.

    Article  Google Scholar 

  5. Dobrowsky PH, Mannel D, De Kwaadsteniet M, Prozesky H, Khan W, Cloete TE. Quality assessment and primary uses of harvested rainwater in Kleinmond, South Africa. Water SA. 2014;40(3):401–6.

  6. Amin MT, Nawaz M, Amin MN, Han M. Solar disinfection of Pseudomonas aeruginosa in harvested rainwater: a step towards potability of rainwater. PloS One. 2014;9(3):1–10.

    Article  Google Scholar 

  7. Department of Water Affairs and Forestry (DWAF). South African Water Quality Guidelines, 2nd Edition Volume 1: Domestic Water Use. Pretoria: CSIR Environmental Services; 1996. http://www.iwa-network.org/filemanager-uploads/WQ_Compendium/Database/Selected_guidelines/041.pdf. Accessed 28 Sept 2015.

  8. World Health Organization (WHO). Guidelines for Drinking-Water Quality, 4th Edition. Geneva: World Health Organization; 2011. http://apps.who.int/iris/bitstream/10665/44584/1/9789241548151_eng.pdf. Accessed 18 Sept 2015.

  9. Dobrowsky PH, De Kwaadsteniet M, Cloete TE, Khan W. Distribution of indigenous bacterial pathogens and potential pathogens associated with roof-harvested rainwater. Appl Environ Microbiol. 2014;80(7):2307–16.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  10. McGuigan KG, Conroy RM, Mosler H, Du Preez M, Ubomba-Jaswa E, Fernandez-Ibanez P. Solar water disinfection (SODIS): A review from bench-top to roof-top. J Hazard Mater. 2012;235:29–46.

    Article  PubMed  Google Scholar 

  11. Safapour N, Metcalf RH. Enhancement of solar water pasteurization with reflectors. Appl Environ Microbiol. 1999;65(2):859–61.

    CAS  PubMed  PubMed Central  Google Scholar 

  12. Islam MF, Johnston RB. Household Pasteurization of drinking-water: The Chulli Water-treatment System. J Health Popul Nutr. 2006;24(3):356–62.

    PubMed  PubMed Central  Google Scholar 

  13. Sommer B, Marinõ A, Solarte Y, Salas ML, Dierolf C, Valiente C, et al. SODIS - an emerging water treatment process. J Water SRT-Aqua. 1997;46:127–37.

    Google Scholar 

  14. Raveendhra D, Faruqui S, Saini P. Transformer less FPGA Controlled 2-Stage isolated grid connected PV system. PESTSE 2014. 2014;doi:10.1109/PESTSE.2014.6805304.

  15. Feachem RE, Bradley DJ, Garelick H, Mara DD. Sanitation and disease: Health Aspects of Excreta and Wastewater Management. Washington DC: John Wiley & Sons; 1983.

    Google Scholar 

  16. Burch JD, Thomas KE. Water disinfection for developing countries and potential for Solar Thermal Pasteurization. Sol Energy. 1998;64:87–97.

    Article  Google Scholar 

  17. Dobrowsky PH, Carstens M, De Villiers J, Cloete TE, Khan W. Efficiency of a closed-coupled solar pasteurization system in treating roof harvested rainwater. Sci Total Environ. 2015;536:206–14.

    Article  CAS  PubMed  Google Scholar 

  18. Spinks AT, Dunstan RH, Harrison T, Coombes P, Kuczera G. Thermal inactivation of water-borne pathogenic and indicator bacteria at sub-boiling temperatures. Water Res. 2006;40(6):1326–32.

    Article  CAS  PubMed  Google Scholar 

  19. El Ghetany HH, Abdel Dayem A. Numerical simulation and experimental validation of a controlled flow solar water disinfection system. Desalin Water Treat. 2010;20(1-3):11–21.

    Article  Google Scholar 

  20. Reyneke B, Dobrowsky PH, Ndlovu T, Khan S, Khan W. EMA-qPCR to monitor the efficiency of a closed-coupled solar pasteurization system in reducing Legionella contamination of roof-harvested rainwater. Sci Total Environ. 2016;553:662–70.

    Article  CAS  PubMed  Google Scholar 

  21. Bédard E, Fey S, Charron D, Lalancette C, Cantin P, Dolcé P, Laferrière C, Déziel E, Prévost M. Temperature diagnostic to identify high risk areas and optimize Legionella pneumophila surveillance in hot water distribution systems. Water Res. 2015;71:244–56.

    Article  PubMed  Google Scholar 

  22. Yee RB, Wadowsky RM. Multiplication of Legionella pneumophila in unsterilized tap water. Appl Environ Microbiol. 1982;43(6):1330–4.

    CAS  PubMed  PubMed Central  Google Scholar 

  23. Giamarellou H. Prescribing guidelines for severe Pseudomonas infections. J Antimicrob Chemother. 2002;49(2):229–33.

    Article  CAS  PubMed  Google Scholar 

  24. Mena KD, Gerba CP. Risk assessment of Pseudomonas aeruginosa in water. Rev Environ Contam T. 2009;201:71–115.

    CAS  Google Scholar 

  25. Lyczak JB, Cannon CL, Pier GB. Establishment of Pseudomonas aeruginosa infection: lessons from a versatile opportunist. Microb Infect. 2000;2(9):1051–60.

    Article  CAS  Google Scholar 

  26. Fittipaldi M, Nocker A, Codony F. Progress in understanding preferential detection of live cells using viability dyes in combination with DNA amplification. J Microbiol Methods. 2012;91(2):276–89.

    Article  CAS  PubMed  Google Scholar 

  27. Reed RH. Solar inactivation of faecal bacteria in water: the critical role of oxygen. Lett Appl Microbiol. 1997;24(4):276–80.

    Article  CAS  PubMed  Google Scholar 

  28. Burton A. Purifying drinking water with Sun, Salt, and limes. Environ Health Perspect. 2012;120(8): 10.1289/ehp.120-a305.

  29. Saleh MA, Ewane E, Jones J, Wilson BL. Monitoring Wadi El Raiyan lakes of the Egyptian desert for inorganic pollutants by ion-selective electrodes, ion chromatography and inductively coupled plasma spectroscopy. Ecotoxicol Environ Saf. 2000;45(3):310–6.

    Article  CAS  PubMed  Google Scholar 

  30. Delgado-Viscogliosi P, Solignac L, Delattre JM. Viability PCR, a Culture-Independent Method for Rapid and Selective Quantification of Viable Legionella pneumophila Cells in Environmental Water Samples. Appl Environ Microbiol. 2009;75(11):3502–12.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  31. Herpers BL, De Jongh BM, Van Der Zwaluw K, Van Hannen EJ. Real-time PCR assay targets the 23S-5S spacer for direct detection and differentiation of Legionella spp. and Legionella pneumophila. Clin Microbiol Rev. 2003;41(10):4815–6.

    Article  CAS  Google Scholar 

  32. Roosa S, Wauven CV, Billon G, Matthijs S, Wattiez R, Gillan DC. The Pseudomonas community in metal-contaminated sediments as revealed by quantitative PCR: a link with metal bioavailability. Res Microbiol. 2014;165:647–56.

    Article  CAS  PubMed  Google Scholar 

  33. Bergmark L, Poulsen PHB, Abu Al-Soud W, Norman A, Hansen LH, Sørensen SJ. Assessment of the specificity of Burkholderia and Pseudomonas qPCR assays for detection of these genera in soil using 454 pyrosequencing. FEMS Microbiology Letters. 2012;333:77–84.

    Article  CAS  PubMed  Google Scholar 

  34. Chen NT, Chang CW. Rapid quantification of viable Legionellae in water and biofilm using ethidium monoazide coupled with real‐time quantitative PCR. J Appl Microbiol. 2010;109(2):623–34.

    CAS  PubMed  Google Scholar 

  35. Brözel VS, Cloete TE. Resistance of bacteria from cooling waters to bactericides. J Ind Microbiol. 1991;8(4):273–6.

    Article  Google Scholar 

  36. Dunn OJ, Clark VA. Applied statistics: analysis of variance and regression. New York: Wiley; 1974.

    Google Scholar 

  37. NHMRC, NRMMC. Australian Drinking Water Guidelines Paper 6, National Water Quality Management Strategy. Canberra: National Health and Medical Research Council, National Resource Management Ministerial Council Commonwealth of Australia; 2011.

    Google Scholar 

  38. South African Bureau of Standards (SABS). South African National Standards (SANS) 241: Drinking Water Quality Management Guide for Water Services Authorities. Annexure 1. ISBN 0-626-17752-9. 2005.

  39. Pedersen K. Biofilm development on stainless steel and PVC surfaces in drinking water. Water Res. 1990;24(2):239–43.

    Article  CAS  Google Scholar 

  40. Suib SL. New and future development in catalysis. Oxford: Elsevier. ISBN: 978—444-53872-2. 2013.

  41. Ahmed W, Richardson K, Sidhu JPS, Jagals P, Toze S. Inactivation of fecal indicator bacteria in a roof‐captured rainwater system under ambient meteorological conditions. J App Microbiol. 2014;116(1):199–207.

    Article  CAS  Google Scholar 

  42. Ahmed W, Vieritz A, Goonetilleke A, Gardner T. Health risk from the use of roof-harvested rainwater in Southeast Queensland, Australia, as potable or non-potable water, determined using quantitative microbial risk assessment. Appl Environ Microbiol. 2010;76:7382–91.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  43. Berney M, Weilenmann HU, Ihssen J, Bassin C, Egli T. Specific growth rate determines the sensitivity of Escherichia coli to thermal, UVA, and solar disinfection. Appl Environ Microbiol. 2006;72:2586–93.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  44. Harwood VJ, Levine AD, Scott TM, Chivukula V, Lukasik J, Farrah SR, Rose JB. Validity of the indicator organism paradigm for pathogen reduction in reclaimed water and public health protection. Appl Environ Microbiol. 2005;71(6):3163–70.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  45. Savichtcheva O, Okabe S. Alternative indicators of faecal pollution: relations with pathogens and conventional indicators, current methodologies for direct pathogen monitoring and future application perspectives. Water Res. 2006;40(13):2463–76.

    Article  CAS  PubMed  Google Scholar 

  46. Wilkes G, Edge T, Gannon V, Jokinen C, Lyautey E, Medeiros D, et al. Seasonal relationships among indicator bacteria, pathogenic bacteria, Cryptosporidium oocysts, Giardia cysts, and hydrological indices for surface waters within an agricultural landscape. Water Res. 2009;43(8):2209–23.

    Article  CAS  PubMed  Google Scholar 

  47. Oliver JD. Recent findings on the viable but non-culturable state in pathogenic bacteria. FEMS Microbiol Rev. 2010;34(4):415–25.

    Article  CAS  PubMed  Google Scholar 

  48. Postollec F, Falentin H, Pavan S, Combrisson J, Sohier D. Recent advances in quantitative PCR (qPCR) applications in food microbiology. Food Microbiol. 2011;28(5):848–61.

    Article  CAS  PubMed  Google Scholar 

  49. Fields BS, Benson RF, Besser RE. Legionella and Legionnaires’ Disease: 25 Years of Investigation. Clin Microbiol Rev. 2002;15(3):506–26.

    Article  PubMed  PubMed Central  Google Scholar 

  50. Vervaeren H, Temmerman R, Devos L, Boon N, Verstraete W. Introduction of a boost of Legionella pneumophila into a stagnant-water model by heat treatment. FEMS Microbiol Ecol. 2006;58(3):583–92.

    Article  CAS  PubMed  Google Scholar 

  51. United States Environmental Protection Agency (USEPA). Legionella: Drinking Water Health Advisory. Report number: EPA-822-B-01-005. 2001. https://www.epa.gov/sites/production/files/2015-10/documents/legionella-report.pdf Accessed 28 Aug 2015.

  52. Hussong D, Colwell RR, O’Brien M, Weiss E, Pearson AD, Weiner RM, et al. Viable Legionella pneumophila not detectable by culture on agar media. Biotechnol. 1987;9(5):947–50.

    Article  Google Scholar 

  53. Dwidjosiswojo Z, Richard J, Moritz MM, Dopp E, Flemming HC, Wingender J. Influence of copper ions on the viability and cytotoxicity of Pseudomonas aeruginosa under conditions relevant to drinking water environments. Int J Hyg Environ Health. 2011;214(6):485–92.

    Article  CAS  PubMed  Google Scholar 

  54. McCullagh C, Robertson JM, Bahnemann DW, Robertson PK. The application of TiO2 photocatalysis for disinfection of water contaminated with pathogenic micro-organisms: a review. Res Chem Intermed. 2007;33(3-5):359–75.

    Article  CAS  Google Scholar 

  55. Dalrymple OK, Stefanakos E, Trotz MA, Goswami DY. A review of the mechanisms and modelling of photocatalytic disinfection. Appl Catal B. 2010;98(1):27–38.

    Article  CAS  Google Scholar 

  56. Malato S, Fernández-Ibáñez P, Maldonado MI, Blanco J, Gernjak W. Decontamination and disinfection of water by solar photocatalysis: recent overview and trends. Catal Today. 2009;147(1):1–59.

    Article  CAS  Google Scholar 

  57. Byrne JA, Fernandez-Ibañez PA, Dunlop PS, Alrousan D, Hamilton JW. Photocatalytic enhancement for solar disinfection of water: a review. Int J Photoenergy. 2011;2011:1–12.

    Article  Google Scholar 

  58. Rakić A, Štambuk-Giljanović N. Physical and chemical parameter correlations with technical and technological characteristics of heating systems and the presence of Legionella spp. in the hot water supply. Environ Monit Assess. 2016;188(2):1–12.

    Google Scholar 

Download references

Acknowledgements

The authors would like to thank Jacques de Villiers from the Crest Organization (Stellenbosch, Western Cape) for providing the PhungamanziTM solar pasteurization system. The South African Weather Services is thanked for providing the daily ambient temperature and rainfall data.

Funding

The Water Research Commission (WRC; Project K5/2368//3) and the National Research Foundation of South Africa (Grant number: 90320) funded this project. Opinions expressed and conclusions arrived at, are those of the authors and are not necessarily to be attributed to the National Research Foundation.

Availability of data and materials

The datasets during and/or analysed during the current study is available from the corresponding author on reasonable request.

Authors’ contributions

AS and WK conceived and designed the experiments. AS performed the experiments and analysed the data. PD, TN and BR co-supervised the experimental procedures. WK acquired funding for the study and contributed reagents/materials/analysis tools. AS and WK wrote the paper. All authors edited the drafts of the manuscript and approved the final version of the manuscript.

Competing interests

The authors declare that they have no competing interests.

Consent for publication

Not applicable.

Ethics approval and consent to participate

Not applicable.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Wesaal Khan.

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Strauss, A., Dobrowsky, P.H., Ndlovu, T. et al. Comparative analysis of solar pasteurization versus solar disinfection for the treatment of harvested rainwater. BMC Microbiol 16, 289 (2016). https://doi.org/10.1186/s12866-016-0909-y

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12866-016-0909-y

Keywords