You are on page 1of 7

Science of the Total Environment 472 (2014) 363369

Contents lists available at ScienceDirect

Science of the Total Environment

journal homepage:

Water quality assessment of an unusual ritual well in Bangladesh and impact of mass bathing on this quality
H. Zabed a,b, A. Suely c, G. Faruq b,, J.N. Sahu d

Department of Microbiology, University of Chittagong, Chittagong 4331, Bangladesh Institute of Biological Sciences, Faculty of Science, University of Malaya, Kuala Lumpur 50603, Malaysia Department of Zoology, University of Chittagong, Chittagong 4331, Bangladesh d Department of Petroleum and Chemical Engineering, Faculty of Engineering, Institut Teknologi Brunei, Tungku Gadong, P.O. Box 2909, Brunei Darussalam
b c

H I G H L I G H T S Ritual well water was analyzed in three seasons and during mass bathing. Water fullled the standards for bathing water in all seasons. Physicochemical parameters uctuated unusually during mass bathing. Microbial indicator concentration increased after the mass bathing. Some parameters and indicators showed signicant relationship among them.

a r t i c l e

i n f o

a b s t r a c t
A sacred ritual well with continuously discharging of methane gas through its water body was studied for physicochemical and microbiological quality in three seasons and during ritual mass bathing. Most of the physicochemical parameters showed signicant seasonal variations (P b 0.05) and a sharp uctuation during mass bathing. Dissolved oxygen (DO) was found negatively correlated with temperature (r = 0.384, P b 0.05), biochemical oxygen demand (BOD) (r = 0.58, P b 0.001) and ammonia (r = 0.738, P b 0.001), while BOD showed positive correlation with chemical oxygen demand (COD) (r = 0.762, P b 0.001) and ammonia (r = 0.83, P b 0.001). Simple regression analysis also yielded signicant linear relationship in DO vs. temperature (r2 = 0.147, P b 0.05), DO vs. ammonia (r2 = 0.544, P b 0.001) and BOD vs. DO (r2 = 0.336, P b 0.001). A total of eight microbial indicators were studied and found that all of them increased unusually during mass bathing comparing with their respective seasonal values. Total coliforms (TC) were found positively correlated with fecal coliforms (FC) (r = 0.971), FC with Escherichia coli (EC) (r = 0.952), EC with intestinal enterococci (IE) (r = 0.921), fecal streptococci (FS) with IE (r = 0.953) and Staphylococcus aureus (SA) with Pseudomonas aeruginosa (PA) (r = 0.946), which were signicant at P b 0.001. Some regression models showed signicant linear relationship at P b 0.001 with r2 value of 0.943 for FC vs. TC, 0.907 for EC vs. FC, 0.869 for FS vs. FC, 0.848 for IE vs. EC and 0.909 for IE vs. FS. The overall results found in this study revealed that well water is suitable for bathing purpose but the religious activity considerably worsen its quality. 2013 Elsevier B.V. All rights reserved.

Article history: Received 8 September 2013 Received in revised form 5 November 2013 Accepted 10 November 2013 Available online xxxx Keywords: Mass bathing Ritual well Physicochemical parameters Microbiological indicators Bathing water quality

1. Introduction The Chandra Nath Hill, located at Sitakunda Upazila (sub-district) under Chittagong district in Bangladesh, is one of the holiest places in the country to the Hindu community since very past for the presence
Abbreviations: BOD, biochemical oxygen demand; CFU, colony forming unit; COD, chemical oxygen demand; DO, dissolved oxygen; EC, Escherichia coli; FC, fecal coliforms; FS, fecal streptococci; IE, intestinal enterococci; MPN, most probable number; PA, Pseudomonas aeruginosa; SA, Staphylococcus aureus; TC, total coliforms; TVBC, total viable bacterial count. Corresponding author. Tel.: +60 3 79675805; fax: +60 379675908. E-mail address: (G. Faruq). 0048-9697/$ see front matter 2013 Elsevier B.V. All rights reserved.

of ancient temples on the hill top and several ritual activities. In Barabkunda (a place under Sitakunda) section of this hill (Longitude 22.582612 N and Latitude 91.701748 E), there is another holly place where four old and almost broken temples are located adjacently (Fig. 1), which were mysteriously built on the unapproachable hill top in an unknown period of time though the current priest of the temple demands that these were made about 350 years ago. In the underground of the middle temple, a rectangular natural well (Length, breadth and depth are approximately 3.5, 1.5 and 2.5 m respectively) is located where ground water along with methane gas is continuously discharging to the surface of its water body. But the ow of water is not strong enough resulting in a very slow streaming toward the back side


H. Zabed et al. / Science of the Total Environment 472 (2014) 363369

Fig. 1. (a) Upazila (sub-district) map of Bangladesh indicating (green area) the location of Sitakunda Upazila (source: (b) Sitakunda Upazila map indicating (green circle) Barabkunda (source: and (c) approximate location of the temples and studied well at Barabkunda on the hill (source:

hill. In one side of the well, methane gas is so strongly discharging that there is a continuous re burning over the water all round the year as long as anybody wishes to switch off it (Fig. 2). This made the place sacred to the Hindus, eventually pilgrims and devotees from across the country visit here for their ritual activities. The place is located approximately 45 km away from the Dhaka-Chittagong Highway at Barabkunda town and walking through the hills and forest is the only way to get to it. For this reason only once in every year (10 Falgun i.e., 23 February) they get together here to worship the re and to have a ritual bath in the well that washes out their sins as they believe. They call this mass bathing event Tirtho Snan (Dedicated Bathing). Approximately 510 thousand or more people gather here on this occasion and a good number (about two-third) of them take a bath or make their body wet with the well water. Bathing or recreational water quality evaluation is to some extent relying on the physicochemical parameters of the water. Several parameters such as temperature, pH, DO, BOD, COD, total alkalinity, chloride, ammonia etc. have been studied in different surface water systems used for bathing purposes by different researchers to determine the water quality that concluded with some deviation in these parameters for some anthropogenic or natural reasons (Arvanitidou et al., 2005;

Bhatnagar and Devi, 2012; Debels et al., 2005; Kumar et al., 2012; Ouyang et al., 2006). Microorganisms are commonly used as indicators of bathing or drinking water quality assessment for their wide distribution, abundance and diversication in nature (Baghel et al., 2005; Okpokwasili and Akujobi, 1996). The widely used indicators include mainly fecal contaminants such as TC and FC (Noble et al., 2003; Sood et al., 2008) or more specically EC (Mansilha et al., 2009). In addition, FS and IE are considered the better indicators as they can stay alive in water longer time than FC or EC (Gleeson and Gray, 2002; Zhang et al., 2013). Presence of these fecal contaminants in bathing water poses the risk for various water borne diseases (Moresco et al., 2012). Considering this health risk, several regulatory organizations, countries and regions prescribed fecal indicators as the standards for bathing water assessment though there is the lacking of universal agreement to use the most appropriate polluting indicators (EEC, 1976, 2006; USEPA, 1986). However, evaluation of fecal contaminants does not necessarily assess the risk of non enteric pathogens such as SA and PA which have alarmingly been detected from different surface and recreational water systems (Casanovas-Massana and Blanch, 2013; Plano et al., 2011; Suzuki et al., 2013; Tirodimos et al., 2010). PA from such contaminated water

Fig. 2. (a) Stairs of the middle temple to go to the well and (b) ritual well showing continuous re burning over water due to the discharging of the methane gas; lighting is articial as the well is located in underground and no sunlight can reach there (Courtesy by Fawaz Rob).

H. Zabed et al. / Science of the Total Environment 472 (2014) 363369 Table 1 Collection time and total number of the samples. Sampling periods Summer Rainy season Winter During mass bathing Date May 29, 2007 June 26, 2007 July 23, 2007 August 25, 2007 Dec 12, 2007 Jan 07, 2008 Feb 23, 2008 Sampling time 12:5013:25 14:2514:55 13:1513: 50 12:4013:15 11:3012:10 13:4514:30 10:1510:35 15:2515:55 Number of samples 5 5 5 5 5 5 5 5 Total number of samples 10 10 10 10


meter (Model HI 110, Hanna Instruments, USA). Other physicochemical parameters such as DO, BOD, COD, chloride, total alkalinity and ammonia were measured in the laboratory using standard titrimetric methods as described in standard methods for the examination of water and wastewater (APHA, 1998). 2.3. Microbiological analysis 2.3.1. Enumeration of TVBC TVBC of the collected samples were determined by serially diluting and inoculating the nutrient agar (beef extract 3.0 g/L, bacto peptone 5.0 g/L, sodium chloride 5.0 g/L, bacto agar 15.0 g/L; pH 7.4 0.2) plates with 0.1 mL of each dilution separately. The inoculated plates were incubated at 37 C for 48 h and thereafter the colonies on wellspaced plates were counted and expressed as log10 CFU/100 mL. 2.3.2. Enumeration of TC, FC and EC TC were enumerated by most probable number (MPN) method. Samples were inoculated into lactose broth (LB) (peptone 5.0 g/L, beef extract 3.0 g/L, lactose 5.0 g/L; pH 6.9 0.2) tubes and incubated aerobically at 37 1 C for 48 h. Production of acid and/or gas indicated a positive presumptive test for TC. Total number of positive tubes were counted and taken into consideration for calculation of MPN using the standard chart (Collins et al., 1989). FC were determined by inoculating the EC broth (Hi-media, M127) with a loopful of positive MPN tubes from the above experiment and incubating at 44.5 1 C for 48 h in shaking incubator. Production of acid and/or gas after 48 h indicated a positive presumptive test for FC. To enumerate the EC, a loopful from positive EC broth tubes were inoculated on EMB agar (Oxoid, CM0069) plates and incubated at 37 1 C for 24 h. Dark centered at colonies with or without metallic sheen were observed for EC. Suspicious colonies from EMB plate were transferred to nutrient agar slants and incubated at 37 1 C for 24 h. Colonies from nutrient agar slant was then identied by Gram's staining, IMViC tests that included Indole, methyl red (MR), VogesProskauer (VP) and citrate tests and re-inoculated into LB to conrm gas production. Gram's staining was performed using the laboratory method (Collins and Lyne, 1984; Bartholomew and Mittwer, 1952). Indole, MR, VP and citrate tests were done by inoculating tryptophan broth (bacto tryptone 10.0 g/L, L-tryptophan 1.0 g/L, sodium chloride 5.0 g/L; pH 7.5 0.2) MR-VP broth (bacto peptone 7.0 g/L, dextrose 5.0 g/L, K 2 HPO 4 5.0 g/L; pH 6.9 0.2) and citrate broth (sodium citrate 3.0 g/L, sodium ammonium phosphate 1.5 g/L, magnesium sulfate 0.2 g/L, KH2PO4 1.0 g/L; pH 6.7 0.2) respectively using the laboratory methods (Collins and Lyne, 1984; SAB, 1957). IMViC test results shown as ++ or + were considered as a positive

may cause skin, ear and eye infections to the bathers (Mena and Gerba, 2009), while SA is responsible for skin and soft tissue infection as well as some systemic infections such as sepsis, pneumonia, osteomyelitis and endocarditis (Noskin et al., 2007; Plano et al., 2013). It is a growing concern that anthropogenic activities are continuously polluting the global water resources (Agarwal et al., 2006). Mass bathing which is an age old ritual activity in Indian subcontinent is another human activity that certainly degrades the water quality by uctuating different physicochemical parameters and microbiological indicators (Bhatnagar and Sangwan, 2009; Kulshrestha and Sharma, 2006; Sinha et al., 1991). In the present study, TVBC, TC, FC, EC, FS, IE, SA and PA were studied along with some physicochemical parameters in summer, rainy season, winter and during mass bathing to assess the normal seasonal conditions of the water and impact of mass bathing on this natural well water. 2. Materials and methods 2.1. Collection of samples A total of 40 water samples were collected in summer, rainy season, winter and during mass bathing. Sampling date, time and quantity were given in Table 1. Five sampling points were selected in the well to collect the samples each time (Fig. 3). All samples were collected in sterilized glass bottles from 15 to 30 cm below the surface of the water, stored and transported in a cold box kept below 4 C and analyzed within 56 h of sampling (APHA, 1998). 2.2. Physicochemical analysis Temperature of the collected samples was measured on site with a digital thermometer (Shenzhen Tonglixing Technology, Model: TL8027) and pH was determined in the laboratory with an electric pH

Fig. 3. Schematic diagram of the well showing ve sampling points (SP) from which water samples were collected each time. Here, arrow () denotes the discharging of methane gas and length of arrow denotes the strength of the ow of gas. Fire burning occurred in SP-1. Methane gas discharging in sampling point 4 and 5 is almost absent.


H. Zabed et al. / Science of the Total Environment 472 (2014) 363369

for EC (Kulshrestha and Sharma, 2006) and for the respective EC broth tubes. All counts were expressed as log10 MPN/100 mL. 2.3.3. Enumeration of FS and IE Water samples were ltered through 0.45 membrane lters and inoculated into KF streptococcal agar (Oxoid CM0701) for enumeration of FS (Kenner et al., 1961) and mEI agar (Difco, cat. no. 214885) for IE (Rhodes and Kator, 1997). Thereafter, the inoculated media were incubated at 35 1 C for 48 h and 41 0.5 C for 24 h respectively. All counts were expressed as log10 CFU/100 mL. 2.3.4. Enumeration of SA and PA SA was enumerated using the standard procedure (HPA, 2004). Culture media used for this purpose were Mannitol salt agar with 0.005% sodium azide (beef extract 1.0 g/L, peptone 10.0 g/L, D-mannitol 10.0 g/L, sodium chloride 75.0 g/L, sodium azide 50 mg/L, phenol red 25 mg/L, agar 15.0 g/L; pH 7.4 0.2), DNase Agar (Oxoid CM0321) and Columbia blood agar (Oxoid CM0331). For enumeration of PA, collected water samples were ltered through 0.45 membrane lters and inoculated into m-PA-C agar plates (BD; cat no.: 298153) followed by incubation at 42 0.05 C aerobically for 48 h (Levin and Cabelli, 1972). All counts were expressed as log10 CFU/100 mL. 2.4. Statistical analysis Geometric mean of the obtained data was measured and expressed as mean SE (Standard Error). Physicochemical and microbiological data obtained in three seasons and during mass bathing were compared using one-way analysis of variance (ANOVA). Tukey post-hoc test was conducted to compare the variations among different seasons and during mass bathing. Pearson's product moment correlation and simple regression analyses were performed among different physicochemical parameters and microbiological indicators to analyze the relationship among them. All the statistical analyses were conducted using SPSS 21 (IBM Corporation, Armonk, NY, USA). 3. Results and discussion 3.1. Physicochemical study Temperature, pH, DO, BOD, COD, total alkalinity, chloride and ammonia were measured in summer, rainy season, winter and during mass bathing as an attempt of physicochemical study of the well water. Geometric mean values of these parameters are given in Table 2.

Table 2 Geometric mean values of different physicochemical parameters of the collected samples (Mean Standard Error; minimum and maximum values). Parameters Temperature (C) Summer Rainy season 29.06 0.15 (28.529.8) 7.81 0.03 (7.67.9) 3.63 0.06 (3.43.9) 2.06 0.06 (1.82.3) 4.7 0.12 (4.15.3) 77.84 0.62 (75.881.4) 15.05 0.21 (13.916.3) 0.07 0.0 (0.60.1) Winter 21.09 0.59 (19.223.3) 8.13 0.03 (88.3) 4.16 0.09 (3.74.5) 1.88 0.08 (1.52.2) 5.47 0.09 (4.95.8) 79.53 0.48 (76.481.6) 18.39 0.24 (17.419.7) 0.09 0.0 (0.070.12) During mass bathing 23.84 0.2 (23.124.6) 8.87 0.04 (8.79) 2.8 0.06 (2.53.1) 3.88 0.22 (2.94.7) 6.31 0.19 (5.57.3) 88.83 0.66 (86.492.3) 22.01 0.33 (20.823.7) 0.2 0.01 (0.140.26)

32.09 0.09 (31.732.7) pH 8.4 0.05 (8.28.6) DO (mg/L) 3.19 0.05 (33.5) BOD (mg/L) 1.49 0.08 (1.21.9) COD (mg/L) 5.11 0.08 (4.75.4) Total alkalinity (mg/L) 82.69 0.43 (80.384.3) Chloride (mg/L) 16.06 0.14 (15.316.7) Ammonia (mg/L) 0.11 0.01 (0.070.14)

DO, dissolved oxygen; BOD, biological oxygen demand; COD, chemical oxygen demand.

Temperature and pH are two important parameters in an aquatic environment that regulate the biological processes of this ecosystem. In the present study, average water temperature recorded in summer, rainy season, winter and during mass bathing were 32.09, 29.06, 21.09 and 23.84 C respectively (Table 2), presenting signicant uctuations at P b 0.001. pH values were found within the prescribed limit for bathing water (6.0 9.0) described in Turkish bathing water directive 2006 (Aydin et al., 2010; TBWD, 2006) in all seasons and during mass bathing (Table 2). The minimum and maximum pH were found 7.6 and 9.0 respectively, indicating the alkaline nature of the well water. This alkaline nature was a situation representing that water was well buffered and in high trophic status (Bhatnagar and Sangwan, 2009). Signicant (P b 0.001) variations in pH were found in all the four sampling time periods with a sharp increase during mass bathing as compared to seasonal variations possibly for adding different ingredients such as soap or soap based products, ashes, sindoor (red powder worn by married Hindu women along the parting of their hair and traditionally made from lime or alum) etc. into the water during bathing or as a part of worship activities that ultimately increased the pH of water. Mathur et al. (2007) also reported that variation in pH in an aquatic environment was mainly dependent upon the input materials in the water bodies. DO, BOD, COD and ammonia are four important and interrelating parameters in the aquatic environment. Fluctuations were found in these quality attributes in different seasons showing statistically signicance at 95% condence level with some exceptions. Fluctuations in BOD from summer to winter and winter to rainy season, in COD from summer to rainy season and in ammonia from winter to summer and winter to rainy season were not found signicant (P N 0.05). Highest DO was recorded in winter, while the lowest value was found during mass bathing (Table 2). Sood et al. (2008) also reported increased DO in winter season while studying with Gangetic river system in India. On the other hand, minimum BOD was found in summer and that of COD and ammonia was found in rainy season, whereas the maximum values of all of them were recorded during mass bathing. Signicant (P b 0.01) decrease in DO and increase in BOD, COD and ammonia was found during this occasion due to pilgrim's activities. Bhatnagar and Sangwan (2009) reported that mass bathing decreased the DO and increased the BOD due to add soaps, detergents and other ingredients into water by the pilgrims. They further explained that adding ashes to the water from the worship or urination caused the increase in ammonia content in water. Several other researchers explained the reason of increased BOD, COD and ammonia during mass bathing as this activity caused the entry of organic, domestic and biodegradable wastes by pilgrims that ultimately lead to the microbial decomposition utilizing oxygen for their respiration and releasing CO2 in turn (Bhatnagar and Devi, 2012; Kulshrestha and Sharma, 2006). An important ndings of the present study was that DO values were found lower than the standard (N 6.5 mg/L) as described by Lai et al. (2013). Constant discharging of methane gas which ultimately dissolved in the water may be the main cause of this lower DO (Pham and Bolton, 2012). Besides, there is little chance of growing photosynthetic organisms that could evolve oxygen and accumulate in water as no sunlight can reach there. Total alkalinity and chloride are another two physicochemical parameters, of which the former refers to the presence and level of certain components such as bicarbonate, carbonate and hydroxide, while the latter indicates the hardness of water and presence of sodium, magnesium and calcium (Bhatnagar and Sangwan, 2009). In the present study, geometric mean values of total alkalinity in summer, rainy season and winter were found 82.69, 77.84 and 79.53 mg/L, while that of chloride were 16.06, 15.05 and 18.39 mg/L respectively (Table 2). Signicant seasonal variations were observed in both parameters at P b 0.01 except the insignicant (P N 0.05) change in total alkalinity from rainy season to winter. However, both of these parameters increased suddenly and signicantly (P b 0.001) during mass bathing and exceeded the respective values found in three seasons. This increase was probably caused by

H. Zabed et al. / Science of the Total Environment 472 (2014) 363369


adding different organic materials from worship and other pilgrim's activities (Bhatnagar and Sangwan, 2009). 3.2. Microbiological study A total of eight microbial indicators i.e., TVBC, TC, FC, EC, FS, IE, SA and PA were studied in three seasons and during mass bathing. All data were transformed to log10 values for better interpretation. Geometric mean values of the converted data are given in Table 3 along with standard error (SE), minimum and maximum values. Total viable and active bacterial counts give the indication of microbial load in the bathing water (Baghel et al., 2005). In the present study, TVBC in summer, rainy season and winter were recorded in the range of 4.835.06 log10 CFU/100 mL (Table 3), where the highest count was found in rainy season followed by summer and winter. One of the possible reasons for higher count in rainy season was the higher humidity that helped increasing in settling down of airborne bacteria (Arundel et al., 1986) into water from the surrounding atmosphere. No signicant (P N 0.05) seasonal variations were found in TVBC, probably for noninterference of natural changes on the well water as it is situated inside the temple and ow of constant methane gas from the ground and/or dissolved methane in the surrounding water. However, these viable bacterial counts increased signicantly (P b 0.001) and unusually during mass bathing reaching the maximum value of 9.90 log10 CFU/ 100 mL. This increase in bacterial count was caused by the introduction of more bacteria by pilgrims themselves or dispersion of the sediments, sands and rocks of the well during this activity that had been shown slightly higher bacterial count than water (data not shown). Increased bacterial count in surface water due to ritual activities and mass bathing was also reported in previous study with the sacred Himalayan Rivers in India (Semwal and Akolkar, 2006). The commonly used fecal indicators are the members of coliform group (Enterobacteriaceae) such as TC, FC and EC, and non coliform group such as FS and IE which are widely used to assess the water quality. In the present study, TC and FC found in summer, rainy season and winter complied with the respective guide values prescribed in 76/160/EEC (2.7 log10 CFU/100 mL for TC and 2.0 log10 CFU/100 mL for FC) (EEC, 1976), but exceeded during mass bathing (Table 3). EC complied with the excellent quality standards (2.4 log10 CFU/100 mL) of the directive 2006/7/EC (EEC, 2006) in all seasons and during mass bathing. TC, FC and EC did not show any statistically signicant (P N 0.05) variation in different seasons but increased signicantly (P b 0.001) during mass bathing. Standard guide value for FS in

Table 3 Geometric mean values of different microbiological indicators of the collected samples (Mean Standard Error; minimum and maximum values). Parameters TVBC (log10 CFU/100 mL) TC (log10 MPN/100 mL) FC (log10 MPN/100 mL) EC (log10 MPN/100 mL) FS (log10 CFU/100 mL) IE (log10 CFU/100 mL) SA (log10 CFU/100 mL) PA (log10 CFU/100 mL) Summer 4.93 0.02 (4.865.01) 2.06 0.05 (1.852.34) 1.60 0.08 (1.151.95) 0.77 0.11 (0.301.08) 0.95 0.07 (0.601.26) 0.59 0.11 (01.0) 0.63 0.11 (01.04) 0.92 0.06 (0.701.26) Rainy season 5.01 0.1 (4.935.06) 2.11 0.07 (1.782.45) 1.70 0.06 (1.411.95) 0.89 0.08 (0.31.15) 1.17 0.05 (0.91.36) 0.84 0.05 (0.481.08) 0.76 0.06 (0.481.15) 1.02 0.04 (0.781.15) Winter 4.89 0.02 (4.834.96) 1.85 0.09 (1.432.23) 1.33 0.10 (0.851.78) 0.63 0.11 (0.31.23) 0.94 0.05 (0.781.18) 0.57 0.11 (00.9) 0.59 0.09 (00.95) 0.79 0.07 (0.300.95) During mass bathing 9.17 0.21 (8.439.90) 3.05 0.07 (2.73.2) 2.65 0.06 (2.452.95) 2.03 0.07 (1.782.45) 2.60 0.04 (2.452.73) 1.96 0.72 (1.602.23) 2.05 0.05 (1.852.32) 2.09 0.04 (1.902.26)

recreational water is 2.0 log10 CFU/100 mL (EEC, 1976) and good quality standard for IE is 2.3 log10 CFU/100 (EEC, 2006). Average counts of both indicators found in our study complied with the above standards in all seasons and only FS exceeded the respective standard value during mass bathing (Table 3). FS showed signicant seasonal variations at P b 0.05 except between summer and winter (P N 0.05), while IE did not show any signicant changes in different seasons (P N 0.05). However, both FS and IE increased signicantly (P b 0.001) during mass bathing from their respective seasonal values. Higher fecal indicators in water due to mass bathing practice were also described in several other studies while studying with different water bodies (Kulshrestha and Sharma, 2006; Telang et al., 2009). Although human activities are very limited in this ritual well before mass bathing occasion, presence of several fecal indicators in all seasons may raise a question about their sources. It may be explained that these fecal indicators were possibly introduced into well water from non-human sources (Anderson et al., 1997) such as monkeys, wild chicken, jackals, rats, lizards, reptiles, wild pigeon etc. or they may survive here for long time after once introduced into the water (Anderson et al., 2005; Davies et al., 1995) from humans through mass bathing or other ritual activities. SA and PA are two important non enteric pathogens isolated from several bathing water sources in addition to the fecal indicators that cause possible health risk to the bathers (Fleisher et al., 1996). In our study, a total of 30 samples were collected in three seasons (before mass bathing) and found that all of them were positive for both SA and PA though the lowest and highest counts were recorded (log10 CFU/100 mL) 0 and 1.15 for SA and 0.3 and 1.26 for PA (Table 3). Yoshpe-Purer and Golderman (1987) reported the presence of SA and PA in 60.7 and 44.4% samples respectively out of 628 collected from 34 beaches. Conversely, higher percentage (100%) of both of these pathogens found in our study may be due to the single and small sampling location where diversication for sample collection was not possible. Besides, unusual condition of the well water leads these bacteria to struggle for their survival, where PA is more efcient than SA giving the higher presence of PA than SA. No signicant (P N 0.05) variation was found in the count of these bacteria in different seasons which was also similar to the ndings of Yoshpe-Purer and Golderman (1987). The geometric mean of SA in summer, rainy season and winter were found 0.63, 0.76 and 0.59 log10 CFU/100 mL and that of PA were 0.92, 1.02 and 0.79 log10 CFU/100 mL respectively (Table 3), while these values increased to 2.05 log10 CFU/100 mL for SA and 2.09 log10 CFU/100 mL for PA during mass bathing that were statistically signicant at P b 0.001 with seasonal variations. Mohammed et al. (2012) reported the higher load of SA and PA in nutrient rich area of sandy beach than from less nutrient sea water, but in this study, there was no such nutrient differences in the well water rather there was methane discharging through the water body, and hence mass bathing activities was the main reason of unusual increase in these pathogens during this occasion. 3.3. Correlation and regression analyses Pearsons product moment correlation analysis was done using overall data (N = 40) found in this study. Results of this analysis showed signicant and strong correlation among some important physicochemical parameters and microbiological indicators at 0.1 (**) to 5% (*) level of signicance. It was found that DO was negatively correlated with temperature (r = 0.384*), BOD (r = 0.58**), COD (r = 0.4*) and ammonia (r = 0.738**). On the other hand, BOD showed strong positive correlations with COD (r = 0.762**) and ammonia (r = 0.83**). Similarly, pH and ammonia were positively correlated with each other (r = 0.85**) as well as with total alkalinity (r = 0.863** and 0.894** respectively) and with chloride (r = 0.816** and 0.826** respectively). Among fecal indicators, TC showed 97.1% correlation with FC, 93.0% with EC, 93.2% with FS and 88.0% with IE, and all r values were signicant at P b 0.001. Besides, FC was correlated with EC (r = 0.952**), FS

TVBC, total viable bacterial count; TC, total coliforms; FC, fecal coliforms; EC, Escherichia coli; FS, fecal streptococci; IE, intestinal enterococci; SA, Staphylococcus aureus; PA, Pseudomonas aeruginosa.


H. Zabed et al. / Science of the Total Environment 472 (2014) 363369

(r = 0.932**) and IE (r = 0.909**); FS with IE (r = 0.953**) and EC with IE (r = 0.921**). In addition, strong positive correlation was also found between SA and PA (r = 0.946**) which again correlated with TC (r = 0.897** and 0.932** respectively), FC (r = 0.891** and 0.924** respectively) and FS (r = 0.933** and 0.961** respectively). Positive correlation among different fecal indicators in aquatic environment is a familiar scenario and the relationships among these indicators found in our study ultimately drew this known picture. Based on the ndings from the study with different bathing or recreational water systems by several researchers (Lui et al., 2013; Noble et al., 2003), it can be concluded that fecal indicators correlate among them in such ecosystem. Furthermore, correlation between non fecal pathogens (SA and PA) and fecal indicators (TC, FC and FS) in bathing water was also reported in a good number of previous works (Efstratiou et al., 1998; oli and Krstulovi, 1994; Yoshpe-Purer and Golderman, 1987), from which present work also extracts the support. Physicochemical and microbiological data were also analyzed using some regression models to show the linear relationship among them. As temperature has a negative impact on DO, a model was analyzed considering DO as dependent and temperature as independent variables but the value of r2 was found only 0.147 (P b 0.05) revealing a weak linear relationship (14.7%) between these two parameters. This happened possibly for the inuence of other factors such as ammonia or discharging methane of the well on the DO content in addition to temperature. Further analysis using DO vs. ammonia model showed a signicant (P b 0.001) r2 value of 0.544. Signicant relationship was also found between DO and BOD (r2 = 0.336, P b 0.001) using the model BOD vs. DO. Five simple regression models were applied in different fecal indicators to study the relationship among them. An r2 value was found 0.943 for FC vs. TC, 0.907 for EC vs. FC, 0.869 for FS vs. FC, 0.848 for IE vs. EC and 0.909 for IE vs. FS with F-test yielding P value b 0.001 for each r2 value. 4. Conclusion Water samples studied in different seasons revealed that almost all of the physicochemical parameters and microbiological indicators are in good status, expressing its suitability for bathing purpose. The seasonal uctuation scenario in different parameters and indicators was not much more distinct probably for the well water not being directly inuenced by natural changes. On the other hand, signicant and even unusual uctuations were observed in almost all of these quality attributes during mass bathing because of taking holy bath by a good number of pilgrims in a single occasion revealing that water quality was deteriorated during this religious activity. The purpose of this present investigation is not to draw a picture of horror and discourage religious activities but to develop base line data on physicochemical and microbiological aspects of this sacred natural well that may need attention for its upkeep in near future. To keep the pollution minimum in this holy well during mass bathing we recommend that pilgrims should be given a speech on the basic concept of protecting water quality by the priest and they should be properly cleaned and maintained good personal hygiene prior to taking bath, water may be withdrawn from the well and used in bathing outside of it, urination or other unhygienic activities by the small children should be avoided around the well, water should be disinfected in intervals during and after this occasion. Above all, government should take necessary steps for proper management of this ritual well by improving its infrastructure and conducting comprehensive and scheduled monitoring programs. Acknowledgments The authors are thankful to the Department of Microbiology, University of Chittagong, Bangladesh for conducting this research and University of Malaya, Malaysia for providing nancial support from the project UM.C/HIR/MOHE/ENG/20. We also express our gratitude to Mr. Fawaz

Rob for his cordial cooperation providing the photographs of the temples and well.

Agarwal T, Khillare PS, Shridhar V. PAHs contamination in bank sediment of the Yamuna River, Delhi, India. Environ Monit Assess 2006;123:15166. Anderson S, Turner S, Lewis G. Enterococci in the New Zealand environment: implications for water quality monitoring. Water Sci Technol 1997;35:32531. Anderson KL, Whitlock JE, Harwood VJ. Persistence and differential survival of fecal indicator bacteria in subtropical waters and sediments. Appl Environ Microbiol 2005;71: 30418. APHA (American Public Health Association). Standard methods for the examination of water and wastewater. 20th ed. Washington, DC: American Public Health Association, American Water Works Association, Water Pollution Control Federation; 1998. Arundel AV, Sterling EM, Biggin JH, Sterling TD. Indirect health effects of relative humidity in indoor environments. Environ Health Perspect 1986;65:35161. Arvanitidou M, Kanellou K, Vagiona DG. Diversity of Salmonella spp. and fungi in northern Greek rivers and their correlation to fecal pollution indicators. Environ Res 2005;99: 27884. Aydin A, Kahraman T, Aydin H. The microbiological and physicochemical parameters with trace metal pollution of coastal bathing water in Dardanelles and Thracian Sea. J Anim Vet Adv 2010;9:107782. Baghel VS, Gopal K, Dwivedi S, Tripathi RD. Bacterial indicators of fecal contamination of the Gangetic river system right at its source. Ecol Indic 2005;5:4956. Bartholomew JW, Mittwer T. The gram stain. Bacteriol Rev 1952;16:129. Bhatnagar A, Devi P. Applications of correlation and regression analysis in assessing lentic water quality: a case study at Brahmsarovar Kurukshetra, India. Int J Environ Sci 2012;3:81320. Bhatnagar A, Sangwan P. Impact of mass bathing on water quality. Int J Environ Res 2009;3:24752. Casanovas-Massana A, Blanch AR. Characterization of microbial populations associated with natural swimming pools. Int J Hyg Environ Health 2013;216:1327. Collins CH, Lyne MP. Microbiological methods. 5th ed. London: Butterworth; 1984. Collins CH, Lyne PM, Grange JM. Collins & Lyne's microbial methods. 6th ed. London: Butterworth; 1989. Davies CM, Long J, Donald M, Ashbolt NJ. Survival of fecal microorganisms in marine and freshwater sediments. Appl Environ Microbiol 1995;61:188896. Debels P, Figueroa R, Urrutia R, Barra R, Niell X. Evaluation of water quality in the Chilln River (Central Chile) using physicochemical parameters and a modied water quality index. Environ Monit Assess 2005;110:30122. EEC. Council Directive of 8 December 1975 concerning the quality of bathing water (76/160/EEC), Brussels. Off J Eur Communities 1976;L/031:17. EEC. Directive 2006/7/EC of the European Parliament and of the Council of 15 February, 2006 concerning the management of bathing water quality and repealing Directive 76/160/EEC. Off J Eur Union 2006;L64/37. [4.3.2006]. Efstratiou M, Mavridou A, Richardson S, Papadakis J. Correlation of bacterial indicator organisms with Salmonella spp., Staphylococcus aureus and Candida albicans in sea water. Lett Appl Microbiol 1998;26:3426. Fleisher J, Kay D, Salmon R, Jones F, Wyer M, Godfree A. Marine waters contaminated with domestic sewage: nonenteric illnesses associated with bather exposure in the United Kingdom. Am J Public Health 1996;86:122834. Gleeson C, Gray N. The coliform index and waterborne disease. London: Chapman and Hall; 2002. HPA. Enumeration of Staphylococcus aureus by membrane ltration. National standard method W 10 issue 3. Health Protection Agency; 2004 []. Kenner BA, Clark HF, Kabler PW. Fecal streptococci I. Cultivation and enumeration of streptococci in surface waters. Appl Microbiol 1961;9:1520. Kulshrestha H, Sharma S. Impact of mass bathing during Ardhkumbh on water quality status of river Ganga. J Environ Biol 2006;27:43740. Kumar S, Tripathi V, Garg S. Physicochemical and microbiological assessment of recreational and drinking waters. Environ Monit Assess 2012;184:26918. Lai YC, Tu YT, Yang CP, Surampalli RY, Kao CM. Development of a water quality modeling system for river pollution index and suspended solid loading evaluation. J Hydrol 2013;478:89101. Levin M, Cabelli V. Membrane lter technique for enumeration of Pseudomonas aeruginosa. Appl Microbiol 1972;24:86470. Lui DV, Lui D, Peut D, Miovi V, Glad M, Bilajac L, et al. Evaluation of equivalence between different methods for enumeration of fecal indicator bacteria before and after adoption of the new bathing water directive and risk assessment of pollution. Mar Pollut Bull 2013;73:2527. Mansilha CR, Coelho CA, Heitor AM, Amado J, Martins JP, Gameiro P. Bathing waters: new directive, new standards, new quality approach. Mar Pollut Bull 2009;58:15625. Mathur P, Agarwal S, Nag M. Assessment of physicochemical characteristics and suggested restoration measures for Pushkar Lake, Ajmer Rajasthan (India). Proceedings of Taal 2007: The 12th World Lake Conference, 1518. 2007. p. 1529. Mena K, Gerba C. Risk assessment of Pseudomonas aeruginosa in water. In: Whitacre DM, editor. Reviews of environmental contamination and toxicology. United States: Springer; 2009. p. 71115. Mohammed RL, Echeverry A, Stinson CM, Green M, Bonilla TD, Hartz A, et al. Survival trends of Staphylococcus aureus, Pseudomonas aeruginosa, and Clostridium perfringens in a sandy South Florida beach. Mar Pollut Bull 2012;64:12019.

H. Zabed et al. / Science of the Total Environment 472 (2014) 363369 Moresco V, Viancelli A, Nascimento MA, Souza DSM, Ramos APD, Garcia LAT, et al. Microbiological and physicochemical analysis of the coastal waters of southern Brazil. Mar Pollut Bull 2012;64:408. Noble RT, Moore DF, Leecaster MK, McGee CD, Weisberg SB. Comparison of total coliform, fecal coliform, and Enterococcus bacterial indicator response for ocean recreational water quality testing. Water Res 2003;37:163743. Noskin GA, Rubin RJ, Schentag JJ, Kluytmans J, Hedblom EC, Jacobson C, et al. National trends in Staphylococcus aureus infection rates: impact on economic burden and mortality over a 6-year period (19982003). Clin Infect Dis 2007;45:113240. Okpokwasili GC, Akujobi TC. Bacteriological indicators of tropical water quality. Environ Toxicol Water Qual 1996;11:7781. Ouyang Y, Nkedi-Kizza P, Wu QT, Shinde D, Huang CH. Assessment of seasonal variations in surface water quality. Water Res 2006;40:380010. Pham MPT, Bolton DW. Dissolved methane concentrations in well water in the Appalachian plateau physiographic province of Maryland. Maryland Department of Natural Resources, Resource Assessment Service, Maryland Geological Survey; 2012 [administrative report no. 13-02-01]. Plano LR, Garza AC, Shibata T, Elmir SM, Kish J, Sinigalliano CD, et al. Shedding of Staphylococcus aureus and methicillin resistant Staphylococcus aureus from adult and pediatric bathers in marine waters. BMC Microbiol 2011;11:5. Plano LR, Shibata T, Garza AC, Kish J, Fleisher JM, Sinigalliano CD, et al. Human-associated methicillin-resistant Staphylococcus aureus from a subtropical recreational marine beach. Microb Ecol 2013:113. Rhodes M, Kator H. Enumeration of Enterococcus sp. using a modied mE method. J Appl Microbiol 1997;83:1206. SAB (Society of American Bacteriologists). Manual of microbiological methods. New York: McGraw-Hill Book Co. Inc.; 1957.


Semwal N, Akolkar P. Water quality assessment of sacred Himalayan rivers of Uttoranchal. Curr Sci 2006;91:48696. Sinha AK, Pande DP, Srivastava RK, Srivastava P, Srivastava KN, Kumar A, et al. Impact of mass bathing on the water quality of the Ganga River at Haudeshwarnath (Pratapgarh), India a case study. Sci Total Environ 1991;101:27580. oli M, Krstulovi N. Presence and survival of Staphylococcus aureus in the coastal area of Split (Adriatic Sea). Mar Pollut Bull 1994;28:696700. Sood A, Singh KD, Pandey P, Sharma S. Assessment of bacterial indicators and physicochemical parameters to investigate pollution status of Gangetic river system of Uttarakhand (India). Ecol Indic 2008;8:70917. Suzuki Y, Kajii S, Nishiyama M, Iguchi A. Susceptibility of Pseudomonas aeruginosa isolates collected from river water in Japan to antipseudomonal agents. Sci Total Environ 2013;450451:14854. TBWD. Turkish bathing water directives, 2006, Ankara, Turkey. Doc no. 09/01/2006-No: 26048; 2006. Telang S, Saxena Y, Chaturvedi A. Effect of mass bathing on the water quality of Narmada River at district Hoshangabad, (M.P.) India. Curr World Environ 2009;4: 2113. Tirodimos I, Arvanitidou M, Dardavessis T, Bisiklis A, Alexiou-Daniil S. Prevalence and antibiotic resistance of Pseudomonas aeruginosa isolated from swimming pools in northern Greece. East Mediterr Health J 2010;16:7837. USEPA. Ambient water quality criteria for bacteria. Washington DC: US Environmental Protection Agency; 1986 [EPA440/5-84-002]. Yoshpe-Purer Y, Golderman S. Occurrence of Staphylococcus aureus and Pseudomonas aeruginosa in Israeli coastal water. Appl Environ Microbiol 1987;53:113841. Zhang W, Wang J, Fan J, Gao D, Ju H. Effects of rainfall on microbial water quality on Qingdao no. 1 bathing beach, China. Mar Pollut Bull 2013;66:18590.