Environ Sci Pollut Res DOI 10.1007/s11356-017-9668-z

ADVANCES IN ENVIRONMENTAL BIOTECHNOLOGY AND ENGINEERING 2016

Predicting attenuation of solar radiation (UV-B, UV-A and PAR) in waste stabilization ponds under Sahelian climatic conditions Ynoussa Maiga 1 & Joseph Wethé 2 & Aboubakar Sidiki Ouattara 1 & Alfred S. Traoré 1

Received: 27 February 2017 / Accepted: 27 June 2017 # Springer-Verlag GmbH Germany 2017

Abstract Because of its importance in pathogen removal and algal productivity in waste stabilization ponds, sunlight penetration was measured in microcosms and in situ under Sahelian climatic conditions. The different wavelengths were detected using a submersible radiometer equipped with three sensors: UV-B (311 nm), UV-A (369 nm) and photosynthetically available radiation (PAR, 400–700 nm). UV-B was more attenuated than UV-A and PAR. Facultative pond was more light-attenuating than maturation pond. The mean euphotic depths for UV-B were 0.20 and 0.31 m, respectively, in the facultative and maturation ponds; PAR penetrated deeper with mean euphotic depths of 0.27 and 0.42 m, respectively. The mean Secchi depths were 0.16 and 0.10 m in the maturation and facultative ponds waters, respectively. In view of the reported results, the contribution of the deeper sections of ponds to pathogen removal mediated by sunlight seems negligible. Therefore, when designing WSPs, these findings should be considered to increase the penetration of damaging wavelengths in order to ensure efficient microbial removal. For more pathogen elimination, downstream shallow ponds could be considered. The paper also shows how suspended solids, turbidity, and Secchi depth are related to the attenuation coefficients and euphotic depths. The developed models could be used to predict light penetration and then algal growth and Responsible editor: Philippe Garrigues * Ynoussa Maiga [email protected]

1

Laboratory of Microbiology and Microbial Biotechnology, University Ouaga 1 Pr Joseph KI-ZERBO, 03 BP 7021, Ouagadougou 03, Burkina Faso

2

International Institute for Water and Environmental Engineering, 01 BP 594, Ouagadougou 01, Burkina Faso

pathogen removal mediated by sunlight in waste stabilization ponds located in Sahelian climate. Keywords Pathogen . Sahelian climate . Sunlight attenuation . Waste stabilization pond

Introduction Waste stabilization ponds (WSPs) are widely used in Sahelian countries because of their flexibility and adaptability to the socio-economic context of these regions (Mara 2004). In developing countries, effluents coming from WSPs are reused for agricultural purposes because of water scarcity (IWMI and IDRC 2010) and its nutrients. Pathogen removal is therefore an important objective during wastewater treatment. Nearly half of sunlight is in the visible range (400–700 nm) which is also the photosynthetically available radiation (PAR). The solar UV spectrum is conventionally divided into UV-B (300–320 nm) and UV-A (320–400 nm). Previous studies have shown that sunlight exposure has a detrimental effect on pathogens diluted in wastewaters (Maiga et al. 2009a, b; Sun et al. 2016), and that disinfection is influenced by optical characteristics of the waters (Davies-Colley et al. 1995). Indeed, UV-B is strongly absorbed by DNA causing damage to the microorganisms (Liltved and Landfald 2000). UV-A has been shown to be implicated in both damage and repair of DNA (Muela et al. 2000). UV-A and PAR are implicated in the disinfection through photooxidation mechanism. Direct damage of DNA caused by UV-A through the direct generation of abasic sites, single-strand break, double-strand break and pyrimidine dimers has also been reported (Jiang et al. 2009). Recently, Sun et al. (2016) have reported direct DNA damage of phiX174 virus mediated by long wavelength (UVA and visible light). Further, previous researchers have

Environ Sci Pollut Res

reported that plant photosynthesis was linearly related to daily incident PAR exposure (Kadir and Nelson 2014), and that PAR drives photosynthesis of algae (Rosati and Dejong 2003). Algal growth raises the pH and dissolved oxygen (DO) concentration in WSPs, leading to bacterial inactivation through photooxidation (Bitton 2005). Moreover, pond depth has an influence on sunlight disinfection of wastewater (Maiga et al. 2009b). Furthermore, it has been reported that sunlight is involved in photocatalytic degradation of recalcitrant organic compounds such as phenols (Ahmed et al. 2010). Light penetration into treatment ponds is therefore of fundamental importance in the functioning of WSPs (Curtis et al. 1994) and an essential step in the modelling of the disinfection. Despite the contribution of light in WSPs efficiency, the optical characteristics of treatment ponds have been studied only fairly (Curtis et al. 1994; Heaven et al. 2005; DaviesColley et al. 2005). This situation is more critical in Sahelian regions like Burkina Faso, where more than 300 days per year can be expected to be sunny (Kenfack 2006). Indeed, sunlight inactivation of pathogens diluted in wastewater has been studied in this Sahelian region (Maiga et al. 2009a, b). However, because of lack of data addressing the WSPs optics, apparent attenuation coefficients have been estimated from equations provided by Davies-Colley and Vant (1987) and Kirk (1984) using wastewater filtrates. In addition, light penetration into water depends on both the water and its constituents and upon the conditions of illumination (Booth and Morrow 1997), so that it is not possible to safely extrapolate the results reported in temperate climate to the WSPs located in Sahelian regions. Therefore, the overall aim of this investigation was to characterize the fundamental aspects of WSPs optics in a Sahelian region, in order to determine the contributing wavelengths in the disinfection occurring in the ponds. The specific objectives were to:

from the campus and was composed of three ponds in series (anaerobic, facultative and maturation ponds). In addition, an experimental device, composed of two rectangular concrete microcosms (0.70 m × 0.40 m) with a water column of 1 m, was used. The microcosms were filled with wastewater collected from the entry of the maturation and facultative ponds. In situ underwater light measurement In order to follow the diurnal variation of surface and underwater irradiance, in situ measurements were conducted at the surface and in the first 20 cm water column at the entry of the facultative and maturation ponds. All the measurements were conducted between 8.00 and 17.00 h at a frequency of 1 h. The effect of reflection will be considered as 10% of the incident irradiance as suggested by Craggs et al. (2004). Moreover, underwater profiles were carried out at the entry of both ponds until reaching undetectable values. These in situ measurements were performed in unmixed ponds. The profiles were made when the sun was not hidden by clouds and as close to solar noon as possible (12.00 h), to reduce the changes in solar irradiance associated with changing solar elevation. UV-visible measurements (311, 369 and 400–700 nm) were made using a submersible UV-visible radiometer UV203-3 (Macam Photometrics Ltd., Livingston, Scotland) at the water-air interface (0 m) and underwater at 0.05, 0.1, 0.15, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9 and 1.0 m. Microcosms measurements

Methods

To predict light attenuation, wastewaters collected from facultative or maturation ponds were used on separate days for irradiance and Secchi depth measurements. For each sampling day, both microcosms were filled in the morning (8.00 h) using the same wastewater. Close to midday, one of the microcosms was used for underwater measurements as described above; just before, the microcosm was mixed and a sample collected for suspended solids (SS), humic substances and turbidity measurements. The second microcosm was conserved unmixed for Secchi depth measurement which is a simple alternative measure of light penetration. A blackwhite Secchi disc, 20 cm in diameter, attached to a graduated stick, was lowered into the water until no longer visible. The depth at which this occurs is known as the Secchi depth (Davies-Colley et al. 2005).

Study site

Evaluation of light attenuation parameters

This study was conducted in the pilot scale WSP of the International Institute for Water and Environmental Engineering (2iE) located at Ouagadougou (12° N, 2.3° O) in Burkina Faso. The pilot plant treated domestic wastewater

The attenuation of irradiance with depth in optically homogenous waters is based on the equation:

(i) Determine the irradiance attenuation coefficients in the UV-B, the UV-A and the PAR ranges of sunlight (ii) Determine the euphotic layers of UV-B, UV-A and PAR ranges of sunlight (iii) Predict light attenuation through relationships between the attenuation parameters and the wastewater characteristics

Iz ¼ I0e−Kd:Z

ð1Þ

Environ Sci Pollut Res

IZ and I0 are the downward irradiance at a depth Z and just below the surface, respectively; Kd (m−1) is the vertical attenuation coefficient (Brown 1984; Kirk 1994). In this study, the irradiance measured at 0.05 m will be considered as the irradiance just below the surface (I0). The irradiance attenuation coefficients were estimated by the slope of the linear regression of the natural logarithm of measured irradiance versus depth (Davies-Colley et al. 1995; Huovinen et al. 2003). In the present study, all the exponential fits presented a coefficient of determination (R2) higher than 0.95. The euphotic depth (Zeu), the depth at which PAR is reduced to 1% of its value at the water surface, is evaluated as (Brown 1984; Kirk 1994): Zeu ¼ Z1% ¼

4:6 Kd

ð2Þ

Using this equation, the 1% attenuation depth of the UV light will be considered as the euphotic depth in the UV range. Wastewater characteristics assessment SS were assessed by filtration of midday samples (100 ml) through a pre-dried and weighed GFC filter (Whatman, UK), in accordance with the procedures in Standard Methods for the Examination of Water and Wastewater (APHA 1998). Turbidity was measured using a Palintest photometer 7500 (Gateshead NE11, ONS, UK). Moreover, each midday sample was filtered (0.2 μm membrane), and the absorbance of the filtrate was measured between 300 and 740 nm with a Biomate 3 spectrophotometer (Thermo Spectronic, Rochester, NY, USA). The absorption coefficient (aλ) for the dissolved fraction is calculated as: aλ ¼

2:303Aλ ; r

ð3Þ

where Aλ is the absorbance and r the path length (Kirk 1994). The amount of the humic substances was expressed by the absorption coefficient at 380 nm (a380) (Reinart and Pedusaar 2008).

Results and discussion Underwater radiation field The measurements conducted in situ showed that the diurnal variation of solar irradiance in the maturation and facultative ponds followed the incident intensity with the highest values obtained between 11.00 and 13.00 h (Fig. 1). The incident radiation recorded at midday ranged from 1.6 to 2.48 W m−2 for UV-B and from 12.1 to 19.9 W m−2 for UV-A. PAR values

were high ranging from 234.3 to 335.4 W m−2. UV-A irradiance and PAR were available through the first 20 cm from 8.00 to 17.00 h in both ponds, the values being low at the morning and the late afternoon (Fig. 1c–f). Consequently, in both ponds, pathogen inactivation mediated by sunlight should be more efficient from 11 h to the afternoon. This direct effect is enhanced by the synergistic action of pH and DO. Indeed, the afternoon hours are marked by establishment of drastic conditions of pH and DO (Maiga et al. 2009b) because of high PAR levels which promote algal growth. Three mechanisms were previously suggested to describe sunlightmediated inactivation in WSPs: (i) direct UV-B damage to DNA; this mechanism is independent of oxygen, (ii) oxygen-dependent indirect damage caused by UV-B and endogenous sensitizers and (iii) oxygen-dependent indirect damage caused by exogenous sensitizers (Davies-Colley et al. 2000). UV-B irradiance was reduced to very low values after 15.00 h in the facultative pond (Fig. 1b) probably because of low incident values and high attenuation. Therefore, the adverse effect of UV-B radiation to microorganisms in this pond could be negligible after 15.00 h. The solar radiation was differentially attenuated in the water column (Fig. 2). In both ponds, the measured underwater spectra demonstrated a wavelength dependent attenuation of solar radiation with depth (Fig. 2a, b). At solar noon, UV-B did not penetrate more than 0.3 and 0.15 m in the maturation and facultative pond, respectively. Bolton et al. (2011) showed that UV-B penetration in a facultative pond was limited to the first 0.08 m. Our results showed that UV-A was detected no longer than 0.7 m in the maturation pond and 0.4 m in the facultative pond. PAR penetrated more deeply in both ponds than UV-A and UV-B, with detectable values at 0.9 m in the maturation pond and 0.6 m in the facultative pond (Fig. 2). Sunlight wavelengths penetrated more deeply in the maturation pond compared to the facultative pond probably because of the differences in the wastewater characteristics. Attenuation characteristics The behaviour of sunlight in the treatment ponds is described by the attenuation characteristics presented in Table 1. The penetration of sunlight in water is quantified by Kd, the probability of disappearance of a photon per unit depth in the water column (Davies-Colley et al. 1995). Sunlight wavelengths are more attenuated in the facultative pond than the maturation pond. Indeed, the largest Kd (i.e., the strongest attenuation) was observed in the facultative pond for both microcosms and in situ measurements (Table 1). For example, from the microcosms measurements, Kd.UV-A were 12.15 m−1 in the maturation pond water and 20.49 m−1 in the facultative pond water, which presented the highest turbidity. Short wavelengths were attenuated more rapidly in the facultative pond water, for both microcosms and in situ

Environ Sci Pollut Res incident 0.1

Incident 0.1 m

Adjusted incident 0.2 m

1000 100

a) UV-B, MP

10 1 0.1 0.01

100

1 0.1 0.01 8h 9h 10h 11h 12h 13h 14h 15h 16h 17h Period of the day (h)

8h 9h 10h 11h 12h 13h 14h 15h 16h 17h Period of the day (h) 1000

100

UV-A Irradiance (W/m2)

UV-A irradiance (W/m2)

1000

10 1 0.1

c) UV-A, MP

0.01

100 10 1 0.1 0.01

8h 9h 10h 11h 12h 13h 14h 15h 16h 17h Period of the day (h)

8h 9h 10h 11h 12h 13h 14h 15h 16h 17h Period of the day (h) 1000

1000

100

100

PAR irradiance (W/m2)

PAR irradiance (W /m2)

d) UV-A, FP

0.001

0.001

10 1 0.1

e) PAR, MP

0.01

0.001 8h 9h 10h 11h 12h 13h 14h 15h 16h 17h Period of the day (h)

0.001

b) UV-B, FP

10

0.001

0.001

0.00001 0

Adjusted incident 0.2 m

1000 UV-B irradiance (W/m2)

UV-B irradiance (W/m2)

Fig. 1 Diurnal variation of UV-B (a, b), UV-A (c, d) and PAR (e, f) (logarithmic scale) measured on the air (incident), underwater (0.1 and 0.2 m) and the adjusted incident irradiance for reflection (10%) in the maturation pond (MP) and facultative pond (FP). Data obtained from in situ measurements in the maturation and facultative ponds

Irradiance (W/m2) 0.1

10

0.00001 0

1000

0.1

0.1

0.2

0.2

10 1 0.1

f) PAR, FP

0.01

0.001 8h 9h 10h 11h 12h 13h 14h 15h 16h 17h Period of the day (h)

0.001

Irradiance (W/m2) 0.1

10

1000

UV-B

0.3

0.3 Depth (m)

Depth (m)

UV-B

0.4 0.5 0.6

0.4

0.6 0.7

0.7

UV-A

0.5

PAR

UV-A

0.8

a) Maturation pond

1

0.8

b) Facultative pond

0.9

0.9 PAR

1

Fig. 2 Depth profile of irradiance (logarithmic scale) in the maturation pond (a) and the facultative pond (b) for UV-B (peak at 311 nm), UV-A (peak at 369 nm) and PAR (400–700 nm). Profiles made at 12.00 h local time. Data obtained from in situ measurements in the maturation and facultative ponds

Environ Sci Pollut Res Table 1

Average values of optical variables in the maturation pond and facultative pond Maturation pond

Facultative pond

Optical variables

Microcosma n = 13

In situb n = 11

Microcosma n=3

In situb n=2

Kd.UV-B (m−1)

16.22 ± 3.81

15.56 ± 3.46

25.32 ± 5.14

22.49 ± 2.26

Kd.UV-A (m−1) Kd.PAR (m−1) Zeu.UV-B (m) Zeu.UV-A (m) Zeu.PAR (m)

12.15 ± 2.46 12.70 ± 2.33 0.30 ± 0.07 0.40 ± 0.09 0.37 ± 0.06

10.94 ± 1.75 11.06 ± 1.63 0.31 ± 0.07 0.43 ± 0.07 0.42 ± 0.06

20.49 ± 2.85 19.49 ± 4.51 0.19 ± 0.04 0.23 ± 0.03 0.24 ± 0.05

19.05 ± 1.96 16.89 ± 0.55 0.20 ± 0.02 0.24 ± 0.02 0.27 ± 0.01

Secchi depth (m) Turbidity (NTU)

0.16 ± 0.03 119.62 ± 36.60

ND ND

0.10 ± 0.01 251.67 ± 40.41

ND ND

SS (mg/L) a380 (m−1)

67.08 ± 26.02 9.21 ± 1.04

ND ND

120.81 ± 39.97 8.44 ± 1.04

ND ND

a absorption coefficient, ND not determined, NTU nephelometric turbidity unit, SS suspended solids, ± standard deviation a

Data collected from mixed microcosm except Secchi depth (measured in unmixed microcosm)

b

Data collected from unmixed ponds. All data were collected on clear sky days at 12. 00 h

measurements, the most attenuated wavelengths being UV-B, after that, the UV-A and finally the PAR. The average attenuation coefficients for UV-B, UV-A and PAR from the facultative pond in situ measurements were 22.49, 19.05 and 16.89 m−1, respectively (Table 1). Heaven et al. (2005), working on waste stabilization ponds in UK, reported Kd.PAR values of 4.8 to 13.7 m−1 with suspended solids varying from 42 to 172 mg L−1. The attenuation coefficients (Kd) from the mixed microcosms were slightly higher than the values measured in situ (Table 1), probably because of non-uniform distribution of the water characteristics in situ (sedimentation). However, for a given type of wastewater and wavelength, statistical analyses (t test at α = 0.05) showed that there were no significant differences between Kd values when microcosms and in situ measurements are compared. From the microcosms measurements, the 1% penetration depths estimated for UV-B ranged from 0.18 to 0.41 m in the maturation pond water and 0.15 to 0.22 m in the facultative pond water. The corresponding UV-A penetration depths varied from 0.28 to 0.56 m in the maturation pond water and 0.2 to 0.26 m in the facultative pond water. PAR penetrated from 0.27 to 0.49 m and from 0.19 to 0.28 m in the maturation and facultative pond waters, respectively. The mean euphotic depths for PAR were 0.24 m in the facultative pond water and 0.37 m in the maturation pond water (Table 1). This contrast can find an explanation in the characteristics of the wastewaters. Indeed, the average turbidity is around two times higher in the facultative pond water than the maturation pond water. The mean euphotic depths for PAR measured in situ were 0.42 and 0.27 m in the facultative and maturation pond, respectively (Table 1). Davies-Colley et al. (2005) estimated a

median euphotic depth of 0.35 m for maturation ponds treating sewage (turbidity of 19 NTU), while Sukias et al. (2001) reported more restricted value of 0.11 m in facultative ponds treating dairy cattle wastes (turbidity of 140 NTU). High rate algal ponds presented euphotic depths of 0.18 m (sewage, turbidity of 53 NTU) and 0.07 m (dairy cattle wastes, turbidity of 120 NTU) (Davies-Colley et al. 1995). In contrast, very high euphotic depths of 3.5, 7 and 15 m were reported in marine waters for UV-B, UV-A and PAR, respectively (Helbling et al. 2005). There is variation in the depth to which different wavelengths penetrated wastewater as revealed by the Zeu values (Table 1). PAR penetrated more deeply in facultative pond than UV-A, while UV-B is more strongly attenuated with depth. This finding is similar to previous studies which found that the short wavelengths of the solar UV-B, with lower intensity than longer wavelengths UV-A, were attenuated the most intensely underwater (Huovinen et al. 2003). In the maturation pond, PAR penetrated deeper than UV-A and UV-B (Fig. 2), but Zeu.UV-A was sometimes higher than Zeu.PAR. This situation may find an explanation in the characteristics of the wastewater and the incident light intensity. Indeed, Curtis et al. (1994) found that relatively little UV light can penetrate the water in lagoons because of the absorbance of gilvin, algae and inanimate particulate matter. Differences in algal populations caused the differences in light attenuation seen between ponds in the PAR range. The different wavelengths were attenuated at different rates, the longer penetrating deeply than the shorter ones. In addition, in some wastewaters, the spectral variation of the attenuation coefficient exhibited a peak at 470 nm rather than showing a steady increase with decreasing wavelength (Curtis et al. 1994).

Environ Sci Pollut Res

Sunlight-mediated disinfection is widely recognized as the main contributor for disinfection of pathogens diluted in water (Davies-Colley et al. 2000; Maiga et al. 2009b; Kadir and Nelson, 2014). In view of the reported results (availability of sunlight expressed by Zeu values in Table 1), sunlightmediated inactivation occurring in WSPs may be concentrated to the shallow sections of maturation and facultative ponds. The contribution of the deeper sections of these ponds to microbial mortality mediated by sunlight could be negligible compared to that of shallower sections because of low light availability. Indeed, Mayo (1989) studied coliform decay along the pond depth by incubating wastewater in 600 ml fully transparent bottles and found the 90% inactivation to be 21, 90 and 150 h for bottles left above the water level and placed at 0.15 and 1 m, respectively. Greater mortality rates were obtained in the sections near the water surface; this mortality decreased as depth increased. Therefore, the author concluded that disinfection increased with increasing direct solar radiation. Recently, Dias and von Sperling (2017) improved this methodology by using quartz vessels (that allowed solar radiation to pass through without being distorted, reflected or attenuated) instead of plastic bottles (block most of the wavelengths including UV radiation). They incubated contaminated wastewater in both illuminated and non-illuminated (aluminium wrapped flask) quartz flasks at 0.1, 0.2 and 0.3 m in maturation pond, and noticed (i) the high contribution of sunlight to microbial inactivation and (ii) the same finding of the mortality decreasing with increasing depth in illuminated flasks. Moreover, the influence of physicochemical factors (pH, DO), which depend on algal growth and therefore on PAR availability, should be greater in the upper sections compared to the deeper sections of these ponds. The low light penetration in the facultative pond water compared to the maturation pond water is confirmed by the Secchi depth measurements with average values of 0.10 and 0.16 m, respectively. This difference should be related to the water characteristics, since Secchi depths of 0–2 m have been reported from humic lakes with values rising to 5–10 m in clear lakes (Lindell et al. 1996). In addition, Davies-Colley et al. (2005) reported Secchi depths of 0.12 m in maturation pond treating sewage and 0.025 m in facultative ponds treating dairy cow waste in New Zealand. Prediction of sunlight attenuation in wastewater treatment ponds The irradiance attenuation coefficient (Kd) is useful for estimating the mean irradiance over a given water depth in order to assess the effect of sunlight on diluted pathogens. Therefore, the following section describes how Kd can be estimated using the water characteristics. The light absorption which takes place in natural waters is attributable to four components of the aquatic system: the

water itself, gilvin (dissolved yellow humic matter), tripton (inanimate particulate matter) and algae (Kirk 1994). In order to have these components uniformly distributed throughout the water column, only the data collected from the microcosms were used for the predictions. To predict PAR, UV-A and UV-B attenuation, the following relations were developed based on turbidity (ɳ) measurements:
Kd:UV‐B ¼ 0:0713 η þ 7:6246 R2 ¼ 0:73; n ¼ 16 ð4Þ
Kd:UV‐A ¼ 0:0602 η þ 5:0244 R2 ¼ 0:86; n ¼ 16 ð5Þ
ð6Þ Kd:PAR ¼ 0:0542 η þ 6:146 R2 ¼ 0:83; n ¼ 16 The attenuation coefficients of the three fractions of sunlight increased with turbidity. This finding has been noted between PAR and turbidity in lakes and reservoirs and estimated by Kd.PAR = C.ɳ with C varying from 0.15 to 0.5 (Brown 1984). Turbidity can be a good predictor of attenuation in ponds as suggested by Bolton et al. (2011) who found UV-A and UV-B to be well correlated to turbidity with R2 values of 0.72 and 0.732, respectively. In our study, PAR and UV-A were well correlated to turbidity than UV-B. Nevertheless, the R2 values were at least the same or higher than that of Bolton et al. (2011). Turbidity is a parameter related to other water quality parameters, especially SS, but it can hardly be considered as a surrogate measurement method for these parameters (Joannis et al. 2008). For this reason and assuming SS ( in the equation) as amount of algae and tripton, the following relationships have been developed. Because the water column is well mixed, we assumed that the SS are uniformly distributed over the pond depth:
ð7Þ Kd:UV‐B ¼ 0:129 ϕ þ 7:922 R2 ¼ 0:69; n ¼ 16
ð8Þ Kd:UV‐A ¼ 0:107 ϕ þ 5:427 R2 ¼ 0:79; n ¼ 16
ð9Þ Kd:PAR ¼ 0:101 ϕ þ 6:134 R2 ¼ 0:83; n ¼ 16 Using data collected from a lake (SS ≤ 140 mg L−1), Stefan et al. (1983) found the relationship between PAR attenuation coefficient and SS to be Kd:PAR ¼ 0:043ϕ þ 1:97

ð10Þ

Moreover, Heaven et al. (2005), working on waste stabilization ponds in the UK, noticed a reasonable correlation between Kd.PAR values and SS concentrations (42–172 mg L−1) in the studied ponds. The relationship between Kd.PAR and SS is as follows: Kd:PAR ¼ 0:056 ϕ þ 2:931

with

R2 ¼ 0:74

ð11Þ

Environ Sci Pollut Res

In order to compare the equations, PAR attenuation coefficients are estimated from Eq. 10 (Ks) and Eq. 11 (Kh), respectively, using our SS data (26 to 152 mg L−1); when our Kd.PAR is plotted against the estimated Ks and then Kh values, the following relationships are developed: Kd:PAR ¼ 2:62 Ks

with

R2 ¼ 0:82

ð12Þ

Kd:PAR ¼ 1:91 Kh

with R2 ¼ 0:82

ð13Þ

Kd.PAR is slightly equal to or greater than 2 times the estimated attenuation coefficients from Eqs. 10 and 11, respectively; the differences between Eqs. 9 and 10 and Eqs. 9 and 11 highlight that, in addition to SS, other factors contribute to the attenuation of sunlight in waters. Secchi depth is a measure of water clarity which is affected by algae, soil particles and other materials suspended in the water. It can be easily evaluated by lowering a disc attached to a graduated stick into the water until no longer visible. For this reason, a relation was developed between the attenuation coefficient of PAR and Secchi depth (ρ): Kd:PAR ¼

1:899 þ 0:66 ρ


R2 ¼ 0:83; n ¼ 16

ð14Þ

Brown (1984), assessing the relationship between Secchi depth and percentage of light penetration in lakes, has reported that Secchi depth corresponded to 10 or 20% of surface light over a wide range of Secchi depths (1–15 m). Zeu.PAR is useful when assessing light availability for algal growth which influences physicochemical characteristics (pH, DO) of the ponds and therefore pathogen removal. It can be estimated using Secchi depth as:
ð15Þ Zeu:PAR ¼ 2:297 ρ R2 ¼ 0:80; n ¼ 16 Brown (1984), studying the relationship between Zeu.PAR and Secchi depth in lakes, found that defining the euphotic zone as three of four times the Secchi depth is a good estimate. In order to consider more water characteristics to assess light availability in the ponds, multiple regressions were performed between euphotic depths and the water characteristics (turbidity, SS and Secchi depth). Due to the weak coefficient of the adjusted models (around 0.6 for UV-A and 0.5 for UVB), only the equations related to the euphotic depth of PAR have been considered. The resulting equations are as follows: Zeu:PAR ¼ 54:87−0:068 η−0:126 ϕ Zeu:PAR ¼ 1:32 ρ−0:042 η þ 21:11 Zeu:PAR ¼ 1:75 ρ−0:076 ϕ þ 15:37


R2 aj ¼ 0:82; n ¼ 16
R2 aj ¼ 0:79; n ¼ 16
R2 aj ¼ 0:80; n ¼ 16

Zeu:PAR ¼ 1:42 ρ−0:022 η−0:059 ϕ þ 22:03

ð16Þ ð17Þ ð18Þ


R2 aj ¼ 0:79; n ¼ 16 ð19Þ

PAR attenuation is closely related to SS more than that of UV-B and UV-A as shown by the linear regression analyses

and the reported coefficients (Eqs.7, 8, 9). An explanation of this situation could be the fact that SS is a measurement of the amounts of organic and inorganic particles in the water including algae (Curtis et al. 1994). Algae are the main components of SS in WSPs effluents (Mara 2004). Since algae are photosynthetic, they contain large amounts of pigments, which capture light energy for photosynthesis. This phenomenon can induce a high pH in the ponds (especially in maturation ponds); the pH can rise to >9.4, which is critical for faecal bacterial die-off in ponds (Mara 2004). In contrast, previous studies have suggested that UVattenuation in waters is strongly influenced by dissolved organic carbon (Curtis et al. 1994; Huovinen et al. 2003). The good correlation found between SS and Kd.PAR suggests that the variation in light penetration into maturation and facultative ponds could be almost attributed to differences in SS concentrations (algal content). SS content is a good predictor of Kd.PAR as well as turbidity and Secchi depth based on the R2 values reported in Eqs. 6, 9 and 14. For example, a good linear relationship was found between the reciprocal of the Secchi depth (m) and Kd.PAR (m−1) (R2 = 0.83), enabling Kd.PAR to be derived using common monitoring measurements. From the multiple regression analyses (Eqs. 16, 17, 18 and 19), it appears that increasing turbidity and/or SS causes a decrease in Zeu.PAR values. In that case, turbid waters are likely to act as protective screen for pathogens diluted in wastewater, because of high light attenuation. Therefore, when designing treatment ponds intended for pathogen removal, the pond depth should be managed taking into account this finding. Zeu.PAR and Secchi depth are positively correlated; then, when Secchi depth increases, the euphotic depth of PAR also increases. R2 value (0.83) shows that SS and turbidity taken together (Eq. 16) are the best predictors of Zeu.PAR in the studied ponds. To be able to predict disinfection in waste stabilization ponds from climate station reports of irradiance, these values should be adjusted to take into account not only the attenuation of the biologically active components of sunlight (using Kd values provided by this study) but also the reflection effect. Thus, Eq. 1 could be improved by using the climate station reports of irradiance instead of the irradiance just beneath the water surface. In this case, the incident irradiance should be reduced by a factor accounting for the reflection. Craggs et al. (2004) have considered the reflected fraction to be 10% of the incident irradiance at the water surface.

Conclusions This study showed that light availability in the ponds varies throughout the day and depends on the wastewater characteristics. Facultative pond was appreciably more lightattenuating than maturation pond.

Environ Sci Pollut Res

UV-B has an acute adverse effect on pathogens diluted in waters. It is theoretically capable of damaging microorganisms in wastewater treatment ponds; however, its detrimental effect on pathogens seems reduced, because the intensity at the wavelength 311 nm in the UV-B becomes rapidly very low. Therefore, to increase the treatment efficiency, the UV light effect needs to be optimized. The reported results showed that the deeper sections of the ponds are not illuminated. Therefore, when designing WSPs, these findings should be considered to allow more damaging wavelengths to penetrate into the ponds, which in turn will ensure efficient microbial removal. For more pathogen elimination, downstream shallow ponds could be considered. The relationships developed using SS, turbidity or Secchi depth are useful to predict light penetration, and then, algal growth and pathogen inactivation mediated by sunlight in maturation and facultative ponds, located in Sahelian developing countries, where wastewaters have almost the same characteristics.

References Ahmed S, Rasul MG, Martens WN, Brown R, Hashib MA (2010) Heterogeneous photocatalytic degradation of phenols in wastewater: a review on current status and developments. Desalination 261:3– 18. doi:10.1016/j.desal.2010.04.062 American Public Health Association/American Water Works Association/Water Environment Federation (1998) Standard methods for the examination of water and wastewater, 20th edn. USA, Washington DC Bitton G (2005) Wastewater microbiology, 3rd edn. John Wiley & Sons, Hoboken Bolton NF, Cromar NJ, Buchanan NA, Fallowfield HJ (2011) Variations in sunlight attenuation in waste stabilization ponds and environmental waters. In: 9th IWA specialist Conference on Waste Stabilization Ponds. Adelaide, Australia. 1-3 August 2011 Booth CR, Morrow JH (1997) The penetration of UV into natural waters. Photochem Photobiol 65:254–257 Brown R (1984) Relationships between suspended solids, turbidity, light attenuation and algal productivity. Lake Reserv Manage 1:198–205. doi:10.1080/07438148409354510 Craggs RJ, Zwart A, Nagels JW, Davies-Colley RJ (2004) Modelling sunlight disinfection in a high rate pond. Ecol Eng 22:113–122. doi:10.1016/j.ecoleng.2004.03.001 Curtis TP, Mara DD, Dixo NGH, Silva SA (1994) Light penetration in waste stabilization ponds. Water Res 28:1031–1038. doi:10.1016/ 0043-1354(94)90188-0 Davies-Colley RJ, Vant WN (1987) Absorption of light by yellow substances in freshwater lakes. Limnol Oceanogr 32:416–425 Davies-Colley RJ, Hickey CW, Quinn JM (1995) Organic matter, nutrients and optical characteristics of sewage lagoon effluents. NZJ Mar Freshwat Res 29:235–250 Davies-Colley RJ, Donnison AM, Speed DJ (2000) Towards a mechanistic understanding of pond disinfection. Water Sci Technol 42:149–158 Davies-Colley RJ, Craggs RJ, Park J, Nagels JW (2005) Optical characteristics of waste stabilization ponds: recommendations for monitoring. Water Sci Technol 5:153–161

Dias DFC, von Sperling M (2017) Vertical profiling and modelling of Escherichia coli decay in a shallow maturation pond operating in a tropical climate. Environ Technol In press. doi:10.1080/09593330. 2017.1310936 Heaven S, Blanks CJ, Zotova EA (2005) Light attenuation parameters for waste stabilization ponds. Water Sci Technol 51:143–152 Helbling WE, Barbieri ES, Marcoval AM, Gonçalves RJ, Villafane VE (2005) Impact of solar ultraviolet radiation on marine phytoplankton of Patagonia, Argentina. Photochem Photobiol 81: 807–818 Huovinen PS, Penttilä H, Soimasuo MR (2003) Spectral attenuation of solar ultraviolet radiation in humic lakes in central Finland. Chemosphere 51:205–214 IWMI and IDRC (International Water Management Institute and International Development Research Centre) (2010) Wastewater irrigation and health: assessing and mitigating risk in low-income countries. Drechsel P, Scott CA, Raschid-Sally L, Redwood M, Bahri A (ed), Earthscan, London Jiang Y, Rabbi M, Kim M, Ke C, Lee W, Clark RL, Mieczkowski PA, Marszalek PE (2009) UVA generates pyrimidine dimers in DNA directly. Biophys J 96:1151–1158. doi:10.1016/j.bpj.2008.10.030 Joannis C, Ruban G, Gromaire MC, Bertrand-Krajewski JL, Chebbo G (2008) Reproducibility and uncertainty of wastewater turbidity measurements. Water Sci Technol 57:1667–1673 Kadir K, Nelson KL (2014) Sunlight mediated inactivation mechanisms of Enterococcus faecalis and Escherichia coli in clear water versus waste stabilization pond water. Water Res 50:307–314 Kenfack S (2006) Helio-photocatalytic enhancement of the biodegradation of biorecalcitrant pollutants in water: physico-chemical and technical aspects, Thesis no. 3550, EPFL, Lausanne Kirk JTO (1994) Light and photosynthesis in aquatic ecosystems, 2nd edn. Cambridge University Press, Cambridge Kirk JTO (1984) Dependence of relationship between inherent and apparent optical properties of water on solar altitude. Limnol Oceanogr 29:350–356 Liltved H, Landfald B (2000) Effects of high intensity light on ultravioletirradiated and non-irradiated fish pathogenic bacteria. Water Res 34: 481–486 Lindell MJ, Graneli HW, Tranvik LJ (1996) Effects of sunlight on bacterial growth in lakes of different humic content. Aquat Microb Ecol 11:135–141 Maiga Y, Denyigba K, Wethe J, Ouattara AS (2009a) Sunlight inactivation of Escherichia coli in waste stabilization microcosms in a sahelian region (Ouagadougou, Burkina Faso). J Photochem Photobiol B: Biol 94:113–119 Maiga Y, Wethe J, Denyigba K, Ouattara AS (2009b) The impact of pond depth and environmental conditions on sunlight inactivation of E. coli and enterococci in wastewater in a warm climate. Can J Microbiol 55:1364–1374 Mara D (2004) Domestic wastewater treatment in developing countries. Earthscan, London Mayo AW (1989) Effect of pond depth on bacterial mortality rate. J Environ Eng 115:964–977 Muela A, Garcias-Bringas JM, Arana I, Barcina I (2000) The effect of simulated solar radiation on Escherichia coli: the relative roles of UV-B, UV-A and photosynthetically active radiation. Microbial Ecol 39:65–71 Reinart A, Pedusaar T (2008) Reconstruction of the time series of the underwater light climate in a shallow turbid lake. Aquatic Ecol 42: 5–15 Rosati A, Dejong TM (2003) Estimating photosynthetic radiation use efficiency using incident light and photosynthesis of individual leaves. Ann Bot 91:869–877 Stefan HG, Gardoni JJ, Schiebe FR, Cooper CM (1983) Model of light penetration in a turbid lake. Water Resour Res 19:109–120

Environ Sci Pollut Res Sukias JPS, Tanner CC, Davies-Colley RJ, Nagels JW, Wolters R (2001) Algal abundance, organic matter and physico-chemical characteristics of dairy farm facultative ponds. Implications for treatment performance. NZJ Agric Res 44:279–296

Sun CX, Kitajima M, Gin KYH (2016) Sunlight inactivation of somatic coliphage in the presence of natural organic matter. Science of the Total Environ 541:1–7. doi:10.1016/j.scitotenv.2015.08.136