- Email: [email protected]
Seasonal and diurnal variations of temperature, pH and dissolved oxygen in advanced integrated wastewater pond system® treating tannery effluent
ARTICLE IN PRESS
Water Research 38 (2004) 645–654
Seasonal and diurnal variations of temperature, pH and dissolved oxygen in advanced integrated wastewater pond systems treating tannery effluent I. Tadessea,*, F.B. Greenb, J.A. Puhakkaa a
Institute of Environmental Engineering and Biotechnology, Tampere University of Technology, P.O. Box 541, Tampere 33101, Finland b Lawrence Berkeley National Laboratory, One Cyclotron Road, MS 70A-3317F, Berkeley, CA 94720, USA Received 6 January 2003; received in revised form 9 September 2003; accepted 2 October 2003
Abstract Seasonal and diurnal fluctuations of pH, dissolved oxygen (DO) and temperature were investigated in a pilot-scale advanced integrated wastewater pond system (AIWPSs) treating tannery effluent. The AIWPSs was comprised of advanced facultative pond (AFP), secondary facultative pond (SFP) and maturation pond (MP) all arranged in series. The variations of pH, DO and temperature in the SFP and MP followed the diurnal cycle of sunlight intensity. Algal photosynthesis being dependent on sunlight radiation, its activity reached climax at early afternoons with DO saturation in the SFP and MP in excess of over 300% and pH in the range of 8.6–9.4. The SFP and MP were thermally stratified with gradients of 3–5 C/m, especially, during the time of peak photosynthesis. The thermal gradient in the AFP was moderated by convective internal currents set in motion as a result of water temperature differences between the influent wastewater and contents of the reactor. In conclusion, the AFP possessed remarkable ability to attenuate process variability with better removal efficiencies than SFP and MP. Hence its use as a lead treatment unit, in a train of ponds treating tannery wastewaters, should always be considered. r 2003 Elsevier Ltd. All rights reserved. Keywords: Tannery effluent; Photosynthesis; Seasonal and diurnal variations; Stratification; Advanced integrated wastewater pond system
1. Introduction Wastewater ponds are natural systems whose biochemical and hydrodynamic processes are influenced by meteorological factors such as sunshine, wind, temperature, rainfall and evaporation [1]. Sun is the driving force in the purification process of pond systems as is in any natural water body. Wastewater ponds, however, differ greatly from natural water bodies, e.g. lakes and oceans in nutrient loading, oxygen demand, depth, size, water residence time, material residence time and flow pattern [2]. Variations in meteorological factors often *Corresponding author. P.O. Box 101204, Addis Ababa, Ethiopia. Fax: +251-1-634997. E-mail address: [email protected] (I. Tadesse).
trigger fluctuations in water quality parameters, such as, temperature, pH and dissolved oxygen (DO) both seasonally and diurnally. Photosynthetic oxygenation, which is essential for aerobic oxidation of the waste organics, varies diurnally with light intensities with peaks occurring between 1300 and 1500 h in the tropics [3]. It is not uncommon to find dissolved oxygen supersaturation between 300% and 400% on the top water layers of algal ponds on warm sunny afternoons in the tropics [4–6]. The pH of algal ponds increases with photosynthesis as algae continue consuming carbon dioxide faster than it can be produced by bacterial respiration. As CO2 diffusion from the atmosphere is minimal, primarily due to elevated surface water temperatures, the CO2 deficit during peak photosynthesis is met from the dissociation
0043-1354/$ - see front matter r 2003 Elsevier Ltd. All rights reserved. doi:10.1016/j.watres.2003.10.006
ARTICLE IN PRESS I. Tadesse et al. / Water Research 38 (2004) 645–654
646
of bicarbonate ions. This bicarbonate dissociation with concomitant consumption of CO2 by algae increases the concentration of hydroxyl ions in the water column causing the pH to rise to well over 10 [7–9]. Temperature has a pronounced effect on both biochemical and hydrodynamic processes of pond systems. During the daylight hours, solar radiation heats up the top water layer thus causing thermal stratification with a consequence of the warmer and lighter water overlaying the cooler and denser deeper water. Such density stratification reduces the performances of pond systems by increasing short-circuiting and disrupting the internal diffusive and advective mass transfer mechanisms. Thermal stratification also localizes algae into bands of 10–15 cm width that move up and down through the water column in response to changes in light intensity [10]. After dusk photosynthesis ceases and the rapid cooling on the surface with intense wind during the night forces the denser top water layer to sink into the bottom causing mixing in the water column. By early morning the algal ponds are destratified with no marked gradient in their water column quality. At dawn, the warming up of the top water layers continues until it reaches maximum by early afternoon and distinct density layers are established and the cycle repeats itself. Wind plays an important role in the vertical mixing of the pond contents by destroying thermal stratification. Understanding the seasonal and diurnal changes of water quality parameters gives insight into the process mechanism involved and helps devising short and longterm operational strategies. The aim of this study was to 1. Combined tannery wastewater 2. Screening & settling 3. Horizontal settling tanks 4. Feed pump 5. Submerged fermentation pit
examine the variations of pH, DO and temperature within the AIWPSs reactors and verify the occurrence of any adverse stratification in any of the reactors.
2. Material and methods 2.1. Study site The pilot-scale AIWPSs Facility was installed in Modjo tannery, Ethiopia 75 km south of the capital Addis Ababa along the major high way connecting the capital to the Red Sea port of Djibouti. Modjo town is located 8 350 North and 39 100 East with an altitude of 1825 m above mean sea level [11]. The climatic condition is that of tropical rain with the mean annual rainfall varying between 680 and 2000 mm. The mean air temperature recorded, at the site, during the course of the experiment varied between 15 C and 30 C. 2.2. Pilot-scale AIWPS The pilot-scale AIWPSs was comprised of an advanced facultative pond (AFP) followed by secondary facultative pond (SFP) and maturation pond (MP) (Fig. 1). Pretreatment was provided by passing the raw combined tannery wastewater through successive bar screens and detaining the screened wastewater for 1 day in a two-chamber horizontal settling tank (HST) (No.3, Fig. 1). The physical dimensions of the AIWPSs reactors were as given in Table 1.
6. Advanced facultative pond 10. Return pump 7. Secondary facultative pond 11. Optional oxygenated 8. Maturation pond water recirculation 9. Treated effluent SP =Sampling points
Fig. 1. Schematic flow diagram of the pilot-scale AIWPSs facility built at Modjo Tannery, Ethiopia [12]. Table 1 Physical dimensions of the AIWPSs reactors [12] Reactor
Reactor geometry
Areal dimensions (m) (m)
Reactor depth Water column Mid-depth (m) depth (m) area (m2)
Liquid volume (m3)
Advanced facultative pond Submerged fermentation pit Secondary facultative pond Maturation pond
Square Cylindrical Rectangular Rectangular
33 +1.6 3.2 1.6 21
3.5 2 1.8 1.3
23 4 8 2
3.3 2 1.5 1
9 2 5 2
ARTICLE IN PRESS I. Tadesse et al. / Water Research 38 (2004) 645–654
2.3. Sampling and analytical methods Sample collection was based on the techniques outlined by Pearson et al. [13] and the analyses of organics (BOD5 and COD), nutrients (NH3–N and Ortho-P) and other pertinent parameters were conducted according to standard methods for the examination of water and wastewater [14]. Dissolved oxygen was monitored using Hach model 16046 portable dissolved oxygen meter and pH using an Inventron pH/mV meter fitted with a combined pH electrode. Mid-depth temperatures were measured by polystyrene-float suspended maximum and minimum thermometer while electrical conductivity was tested by a WTW-8120 Weilheim conductivity meter. 2.4. Tanning process and waste characteristics On average, Modjo tannery processed 25 t of skin to wet-blue and crust leather and produced 250 m3 effluent daily during the time of the study. All wastewater streams emerging from the different process halls were diverted into one underground channel from where all was conveyed in an open ditch to the nearby Modjo River. The tannery was close to EPA [15] subcategory I in terms of its hair pulping and chrome tanning, retanning and wet-finishing processes. Unlike EPA’s subcategory I, Modjo tannery was processing exclusively sheep and goatskins. Mean composition of the raw tannery wastewater, in terms of major pollutants of interest, were as given in Table 3. 2.5. Operating conditions The volume of inflow to the AIWPSs facility was regulated by varying the speed of the feed peristaltic
647
pump against a tabulated calibration curve. Feed flows of 0.500, 0.800 and 1.275 m3 were applied during phase 1, 2 and 3 of the experiment, respectively. Table 2, summarizes the feed volumes and the corresponding mean organic loading rates applied during the three phases of the experiment.
3. Results and discussions 3.1. Performances of the reactors The AIWPSs reactors achieved high pollutant removals as shown in Table 3. Among the AIWPSs reactors, the highest removal for 3+ BOD5, COD, Ortho-P, SO2 occurred in the 4 and Cr AFP. Much of the suspended solids (50–90%) was removed in the HST within 1-day retention time. The BOD5 removal of 24–38% obtained in the HST is in conformity with performances of unaided settling tanks that are generally reported to achieve a BOD5 removal of 25–40% [18]. The effluent ammonia–nitrogen concentration of the AFP was higher than its influent due to anaerobic ammonification of organic nitrogen in the lower fermentation pit. An increase in sulphate concentration in the final effluent could be due to the oxidation of sulphide back to sulphate in oxygen-rich shallow water column of the MP. The suspended solids from the two algal ponds (SFP and MP) also showed an increase in concentration and that was due to the production of more algal biomass. From the filtered effluent BOD5 values of the SFP and MP reported in Tadesse [17], algae was found to constitute 50–60% of the final effluent BOD5. The observation was close to the findings by Mara et al. [10] who reported up to 65%
Table 2 Mean applied organic loading rates [16] Feed phase
Parameters
Units
AIWPSs reactors AFP
SFP
MP
Feed I
Feed flow Hydraulic retention time Surface BOD5 loading Volumetric BOD5 loading
l/d days kg/ha-d g/m3-d
500 8 5560 280
500 16 250 15
500a 4 210 20
Feed II
Feed flow Hydraulic retention time Surface BOD5 loading Volumetric BOD5 loading
l/d days kg/ha-d g/m3-d
800 5 8300 415
800 10 895 55
800 3 1480 150
Feed III
Feed flow Hydraulic retention time Surface BOD5 loading Volumetric BOD5 loading
l/d days kg/ha-d g/m3-d
1275 3 13640 680
1275 6 1020 65
1275 2 1590 160
a
Net evaporative losses were negligible and hence not accounted for.
a
ARTICLE IN PRESS I. Tadesse et al. / Water Research 38 (2004) 645–654
648
Table 3 Mean removal performances of the AIWPS reactors (complied from Tadesse et al. [12,16,17] Parameter
Feed phase
Raw tannery wastewater (mg/l)
Effluent
Overall removal (%)
HST (mg/l)
AFP (mg/l)
SFP (mg/l)
MP (mg/l)
SS
I II III
5071 3219 2517
418(92) 821(74) 1173(53)
339(19) 227(72) 158(87)
279(18) 357(57) 176(11)
386(38) 343(4) 143(19)
92 89 94
BOD5
I II II
3580 3075 2810
2225(38) 2075(33) 2142(24)
250(89) 560(73) 396(82)
85(66) 370(34) 251(37)
50(41) 236(36) 170(32)
99 92 94
COD
I II II
11157 9650 7937
4544(59) 7405(23) 5840(26)
669(85) 1382(81) 2359(60)
463(31) 877(37) 887(62)
364(21) 756(14) 813(8)
97 92 90
NH3-N
I II II
152 136 156
135(11) 130(4) 137(12)
156(16) 224(72) 242(77)
58(63) 100(55) 120(50)
20(66) 79(21) 89(26)
87 42 43
Ortho P
I II II
12 9 16
9(25) 8.9(1) 12(25)
8(11) 1.95(78) 2.7(77)
7(12.5) 0.75(62) 1.2(56)
7(0) 0.75(0) 1(17)
42 92 94
SO2 4
I II II
840 1279 840
624(26) 609(52) 765(9)
115(82) 78(87) 54(93)
150(35) 90(15) 26(52)
155(3) 120(33) 40(54)
82 91 95
Cr3+
I II II
46.00 2.50 12.35
13(72) 1.02(59) 6.72(46)
0.66(95) 0.43(58) 1.38(80)
0.33(50) 0.24(44) 0.93(30)
0.31(6) 0.21(13) 0.78(16)
99 92 94
Values in parentheses are % removals. HST=horizontal settling tank, AFP=advanced facultative pond, SFP=secondary facultative pond, MP=maturation pond.
algal BOD5 in the final effluents of their Brazilian experimental facultative ponds. Whether regulators should include algal BOD5 or not is still controversial, although it is often included by default. Proponents of pond systems argue that algae discharged with the pond effluent are rapidly consumed by zooplankton in the receiving watercourse and in daylight hours they produce oxygen, so they have little chance to exert their BOD5 [19]. 3.2. Variations of air and mid-depth temperatures The air and mid-depth temperatures of the AIWPSs reactors recorded throughout the experiment were as shown in Fig. 2. Ambient air temperature was taken on the sampling deck built around the AFP 2.75 m above ground level, while mid-depth temperatures were taken at 1.15, 0.75 and 0.5 m below the water surfaces of the AFP, SFP and MP, respectively. Except during MayAugust 1997 and December 1998 to January 1999, the air temperature was found to be higher than the reactors
mid-depth water temperatures by an average of 1.4– 2.3 C. During May to August 1997, the mid-depth water temperature of the MP was higher than the ambient air temperature by an average of 4 C. The reactors volumes had an important influence on the middepth temperatures. The MP, with the smallest liquid volume among the reactors, followed the air temperature more closely than the AFP and SFP. The mean COD removal in the AFP was above 80% during the initial and intermediate phases and decreased to 60% during phase 3 (Table 3). The exceptionally low mean mid-depth water temperature of 17 C experienced in the AFP during November– December 1998 (Fig. 2) could, among other factors, have contributed to the decline in COD removal efficiency during phase 3. 3.3. Diurnal variations The peaks in the diurnal variations of temperature, pH and DO observed in the top water layers of
ARTICLE IN PRESS I. Tadesse et al. / Water Research 38 (2004) 645–654 Air
30
AFP
SFP
Phase 1
Temperature (°C)
649 MP
Phase 2
Phase 3
25
20
3F
20 J
31 D
6 J 99
4D
26 D
27 N
6N
20 N
30 O
16 O
7A
7O
9J
24 J
4J
25 J
28 M
4D
4 M 98
29 N
9O
28 O
9A
26 S
8J
2A
26 M
18 M
14 M 97
15
Date
Fig. 2. Variations in the air and mid-depth temperatures of the AIWPSs reactors.
SFP
MP
Mean effluent temperature (˚C)
Mean effluent temperature (˚C)
AFP 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15
Phase 1
8
9
10
11
12
13
14
15
16
17
18
Mean effluent temperature (˚C)
AFP
SFP
SFP
MP
Phase 2
8
Hours of the day 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14
AFP
33 32 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 9
10
11
12 13 14 Hours of the day
15
16
17
18
MP
Phase 3
7
8
9
10
11
12
13
14
15
16
17
18
Hours of the day
Fig. 3. Diurnal variations of temperature.
AFP, SFP and MP occurred between 1300 and 1500 h (Figs. 3–5). In the SFP and MP, DO followed the temperature and pH variations. This was somewhat expected, because photosynthetic oxygenation was high when solar insolation and consequently the temperature were high. Early afternoon DO concentration in the top upper most water layer of the MP was over 20 mgO2 l1 throughout the experiment (Fig. 5). This was a 300%
supersaturation over the saturation DO level of 6.9 mgO2 l1 occurring at the prevailing water temperature. Picot et al. [20] also observed DO levels of up to 350% saturation in their high rate algal pond system during the early afternoons. As the feed organic loading on the SFP and MP increased from phase 1 to phase 3 (a four-fold increase in all), the length of time during which DO was above the saturation gradually decreased. For
ARTICLE IN PRESS I. Tadesse et al. / Water Research 38 (2004) 645–654
650 AFP
9.6
SFP
MP
AFP
9
SFP
MP
9.4 8.8 Mean effluent pH
Mean effluent pH
9.2 9 8.8 Phase 1
8.6 8.4
8.6 8.4 Phase 2 8.2
8.2 8
8
7.8 8
9
10
11
12
13
14
15
16
17
18
AFP
8.8
7.8 8
Hours of the day
SFP
9
10
11
12 13 14 Hours of the day
15
16
17
18
MP
Mean effluent pH
8.7 8.6 8.5
Phase 3
8.4 8.3 8.2 8.1 7
8
9
10
11
12 13 14 Hours of the day
15
16
17
18
Fig. 4. Diurnal variations of pH.
example, during phase 1 both SFP and MP had sustained supersaturation concentrations for over 6 h. However, during phases 2 and 3, the duration over which the MP was above the DO saturation level was reduced to only 4 and 5 h, respectively (Fig. 5). Similar situation was also observed for SFP except during phase 3 when its DO level dropped below the saturation level. Soon after 1700 h, DO in the two algal ponds was below 2 mg O2 l1. There was a net oxygen carryover of more than 3 mg O2 l1 from the night into the next morning in both algal ponds during phase 1. This was due to the low organic loading that resulted in limited respiration during the night. The superficial DO concentration of AFP remained around 0.2 mg O2 l1 throughout the experiment. The pH in the upper water layers of the SFP and MP was above 8.4 during much of day with peaks (pH 9.4) coinciding with the time of peak photosynthesis (Fig. 4). The nighttime pH in SFP and MP dropped to 8.4–8.6 mainly as a result of the absence of photosynthesis and the respiration of organic wastewater constituents by algae and bacteria, which released CO2. It appears from Fig. 3 that MP had the warmest top water layer. There was a marked drop in surface water temperature in the AFP during phase 3 (Fig. 3) that possibly brought about a 25% reduction in the COD removal efficiency (Table 3).
3.4. Vertical variations Vertical variations in DO, pH and temperature (Figs. 6–8) were monitored in all AIWPSs reactors to delineate oxic and anoxic layers and verify whether stratification was taking place. Depth variations in DO, pH and temperature within the AFP were minimal, as shown in Fig. 6. This low thermal gradient is usually ascribed to mixing due to rising biogas bubbles [21]. In this study, intense biogas mixing was far from being the cause for low thermal gradient, as the overall biogas production was found to be low [22]. Convective water movement, due mainly to temperature difference up to 4 C between the incoming wastewater and the AFP’s content, may have a more pronounced effect on the low thermal gradient than the rising biogas bubbles. Torres et al. [23] also observed the effects of such thermal currents in their studies in deep Spanish ponds. The thermal gradient of 2 C in a water column of 3.30 m also fulfils the 0.6 C/m gradient limit set by Pires and Kellner [24] for the occurrence of thermal stratification in ponds of tropical region, quoted by Kellner and Pires [25]. The AFP was anaerobic with DO less than 0.2 mg O2 l1 throughout its depth. The pH varied between 8 and 8.4 at the surface to 7.2 at the bottom, which was conducive for anaerobic digestion of the waste in the fermentation
ARTICLE IN PRESS I. Tadesse et al. / Water Research 38 (2004) 645–654 AFP
SFP
MP -1
18 16 14 12 10 8
O 2 saturation level (6.9 mg/l)
6 4 Phase 1
2
SFP
MP
20 18 16 14 12 10 8
O 2 saturation level (6.9 mg/l)
6 4
Phase 2
2 0
0 8
9
10
11
12 13 14 Hours of the day
AFP
22
15
16
17
SFP
7
18
8
9
10
11
12
13
14
15
16
17
18
Hours of the day
MP
20
-1
Mean dissolved oxygen (mg O2 l )
AFP
22
20
Mean effluent dissolved oxygen (mg O2 l )
-1
Mean effluent dissolved oxygen (mg O2 l )
22
651
18 16
Phae 3
14 12 10 8
O 2 saturation level (6.9 mg/l)
6 4 2 0 7
8
9
10
11
12
13
14
15
16
17
18
Hours of the day
Fig. 5. Diurnal variations of dissolved oxygen. O 2 (mg l ), pH & Temp. (° C) 4
8
12
16
20
-1
-1
0
24
4
8
12
16
20
0
5
5
10
10
10
20
20
20
30 40
O2
pH
60 80 100 120 140 200
40
O2
pH
60 80 100 120 140 200
Phase 1
320
°C
4
8
12
16
20
30 40 60
pH O2
°C
80 100 120 140 200 260
260
260 320
˚C
30
Depth below water surface (cm)
5
Depth below water surface (cm)
Depth below water surface (cm)
0
O 2 (mg l ), pH, Temp. (° C)
O 2 (m g l ), pH & Tem p. (° C)
-1
320
Phase 2
Phase 3
Fig. 6. Vertical variations of DO, pH and temperature in the advanced facultative pond of the AIWPS.
pit. This near-to-neutral pH of 7.2 at the bottom was predominant up to 40 cm above the sludge-water interface. The DO profile for SFP (Fig. 7) demonstrates that the oxypause (depth where oxygen depletion occurred) gradually decreased as organic loading
increased from phase 1–3. At the beginning of the experiment (phase 1, Fig. 7), nearly one-third of liquid depth of SFP was above 2 mg O2 l1 with a maximum (>20 mg O2 l1) occurring in the top 15 cm of the pond water column during the early afternoons.
ARTICLE IN PRESS I. Tadesse et al. / Water Research 38 (2004) 645–654
652 -1
-1
O 2 (mg l ), pH & Temp. (° C) 4
8 12 16 20 24 28
0
4
O 2 (mg l ), pH & Temp. (° C)
8 12 16 20 24 28
0
5
5
10
10
10
15
15
20 25 O2
pH
30
°C
40 60 80
Depth below water surface (cm)
5
Depth below water surface (cm)
Depth below water surface (cm)
0
-1
O 2 (mg l ), pH & Temp. (° C)
20 25 O2
°C
pH
30 40 60 80
20 25 O2
30
°C
pH
40 60 80 100
120
120
120
140
140
140
Phase 1
8 12 16 20 24
15
100
100
4
Phase 2
Phase 3
Fig. 7. Vertical variations of DO, pH and temperature in the secondary facultative pond of the AIWPS.
-1
-1
O 2 (mg l ), pH & Temp. (˚C) 0
4
-1
O 2 (mg l ), pH & Temp. (˚C)
8 12 16 20 24 28
O 2 (mg l ), pH & Temp. (˚C)
0 4 8 12 16 20 24 28 32
0
5
5
5
10
10
10
O2
pH
30
50
70
°C
15
20 O2
°C
pH
30
50
15
20 O2
pH
°C
30
50
90
90
Phase 1
8 12 16 20 24 28
70
70
90
Depth below water surface (cm)
20
Depth below water surface (cm)
Depth below water surface (cm)
15
4
Phase 2
Phase 3
Fig. 8. Vertical variations of DO, pH and temperature in the maturation pond of the AIWPS.
Algal–bacteria respiration during the night depleted the water column oxygen to the extent that the morningtime DO (except during phase 1, Fig. 5) was nearly zero. The situation was, however, reversed to supersaturation by early afternoon when photosynthesis reached its peak. The thermal behaviour of SFP was characterized by brief thermocline (at 20 cm depth) usually occurring during the late afternoons. During this brief period of
stratification there was a 5–8 C difference between the surface and bottom water layers. The vertical pH in the SFP during peak photosynthesis varied from 9 near the water surface to 8.2 at the bottom. The MP appeared to be oxygenated to nearly half of its liquid depth during phase 1 (Fig. 8). Successive high organic loading rates gradually diminished the oxygenated depth from 45 cm during phase 1 to only 15 cm
ARTICLE IN PRESS I. Tadesse et al. / Water Research 38 (2004) 645–654
during phase 3 (Fig. 8). The relatively shallow water column (1 m) of the MP permitted sufficient light penetration thereby allowing photosynthesis to occur at deeper layers. There was a period of thermal stratification during peak photosynthesis with variations of 3–5 C between the top and bottom water layers. A pH gradient of 1 unit/m (i.e., 9.4 at the surface to 8.4 at the bottom) was also common in the MP at the time of peak photosynthesis. 3.5. Programming of effluent discharge Knowledge of the diurnal variations of water quality parameters help in devising realistic discharge programmes. Picot et al. [20] proposed a daytime effluent discharge scheme for NH3–N and PO4–P. Daytime discharge is also preferred from public health consideration as releasing effluent from the surface water layer, where photosynthesis is the highest, has the lowest pathogen concentration. But discharging effluent from the layers of highest photosynthesis has the apparent disadvantage of releasing high algal solids with the effluent. As a trade-off between the solids concentration and bacteriological quality of effluent, Mara and Pearson [26] recommended 60 cm take-off point from facultative ponds and 50 cm from maturation points. It is evident from these proposals that the actual discharge programme and the depth from which the effluent is withdrawn depend on the quality of the raw wastewater, intended use of the treated effluent and the discharge limits applicable to the particular place area. In setting acceptable effluent discharging programme for tanneries, considerations should also be given to the manufacturing schedule and volume of wastewater produced. In large tanneries, where leather manufacturing is carried out round-the-clock and the presence of excreted human pathogens in the raw wastewater is minimal, a continuous release of the treated effluent may be considered.
4. Conclusions 1. The study demonstrates that advanced integrated wastewater pond system preceded with simple pretreatment can adequately treat raw combined tannery wastewater. 2. In advanced integrated wastewater pond system, seasonal and diurnal variations of pH, temperature and dissolved oxygen are more significant in the secondary facultative and maturation ponds than in the primary advanced facultative pond. 3. The secondary facultative and maturation ponds show a thermal stratification gradient of 3–5 C/m, especially, during the time of peak photosynthesis
653
and diurnal variations in pH, temperature, and dissolved oxygen follow the pattern of the daily cycle of the sunlight intensity. 4. In the primary advanced facultative pond, convective and vertical thermal currents contribute more to low thermal gradient than mixing by rising biogas bubbles. 5. A continuous discharging of treated effluent may be necessary for tanneries that operate round-the-clock and where their raw effluents contain minimal excreted human pathogens.
Acknowledgements We wish to express our gratitude to the United Nations Industrial Development Organization, UNIDO, Vienna, Austria, which financed [96/024/ML-DP/ ETH/93/005] the pilot experimental work carried out in Modjo tannery as part of its development assistance to the leather industry in Ethiopia. Financial support also came from the Academy of Finland and Tampere University of Technology doctoral school in the form of study grant to the first author. We are also grateful to Modjo Tanner for providing all necessary assistance during the conduct of the research.
References [1] WHO. Wastewater stabilization ponds: principles of planning and practice. WHO-EMRO Technical Publication No. 10. Alexandria, Egypt.WHO-EMRO; 1987. 138p. [2] Gu R, Stefan H. Stratification dynamics in wastewater stabilization ponds. Water Res 1995;29(8):1909–23. [3] Pescod MB, Mara DD. Design, operation and maintenance of wastewater stabilization ponds. In: Pescod MB, Arar A, editors. Treatment and use of sewage effluent for irrigation. UK: Butterworths; 1988. p. 93–115. [4] Vonshak A, Abeliovich A, Boussiba S, Richmond A. Production of Spirulina biomass: effects of environmental factors and population density. Biomass 1982;2(175). [5] Richmond A. Outdoor mass cultures of microalgae. In: Richmond A, editor. Handbook of microalgal mass culture. Florida: CRC Press; 1986. p. 285–330. [6] Picot B, Bahlaoui A, Moersidik S, Baleux B, Bontoux J. Comparison of the purifying efficiency of high rate algal pond with stabilization pond. Water Sci Technol 1992; 25(12):197–206. [7] Azov Y. Effect of pH on inorganic carbon uptake in algal cultures. Appl or Environ Microbiol 1982;43:1300–6. [8] Buhr HO, Miller SB. A dynamic model of the high rate algal bacterial wastewater treatment pond. Water Res 1983;17:29–37. [9] Mara DD, Pearson HW. Design manual for waste stabilization ponds in Mediterranean Countries, Leeds, England: Lagoon Technology International; 1998. 113p.
ARTICLE IN PRESS 654
I. Tadesse et al. / Water Research 38 (2004) 645–654
[10] Mara DD, Pearson HW, Silva SA. Brazilian stabilizationpond research suggests low-cost urban applications. World Water, July 1983. [11] EMA. National Atlas of Ethiopia. Addis Ababa: Ethiopian Mapping Authority; 1988. 76p. [12] Tadesse I, Isoaho SA, Green FB, Puhakka JA. Transformation of sulphur in advanced integrated wastewater pond systems treating tannery effluent. J Am Leather Chem Assoc 2003;98(6):203–15. [13] Pearson HW, Mara DD, Bartone CR. Guidelines for the minimum evaluation of the performance of full-scale waste stabilization ponds. Water Res 1987;21(9):1067–75. [14] APHA, AWWA and WEF. Standard methods for the examination of waste and wastewater. 18th ed. American Public Health Association, American Water Works Association and Water Environment Federation, New York, 1992. [15] EPA. Development document for effluent limitations guidelines, standards for the leather tanning and finishing point source category. Cincinnati, USA: US Environmental Protection Authority; 1982. [16] Tadesse I, Isoaho SA, Green FB, Puhakka JA. Removal of organics and nutrients from tannery effluent by advanced integrated wastewater pond systems technology. Water Sci Technol 2003;48(2):307–14. [17] Tadesse I. Tannery effluent treatment in advanced integrated wastewater pond system. Doctoral thesis. Tampere, Finland: Tampere University of Technology, 2003.
[18] Metcalf & Eddy. Waste engineering: treatment disposal reuse, Singapore: McGraw-Hill; 1991. [19] Mara DD. Waste stabilization ponds: effluent quality requirements and implications for process design. Water Sci Technol 1996;33(7):23–31. [20] Picot B, Moersidik S, Casellas C, Bontoux J. Using diurnal variations in a high rate algal pond for management pattern. Water Sci Technol 1993;28(19):169–75. [21] Pena MR, Madera CA, Mara DD. Feasibility of waste stabilization pond technology for small municipalities in Colombia. Water Sci Technol 2000;42(10–11): 273–82. [22] Tadesse I, Green FB, Puhakka JA. The role of sulphidogenesis in anaerobic treatment phase of tannery wastewater treatment in advanced integrated wastewater pond system. Biodegradation 2003;14:219–27. [23] Torres JJ, Soler A, S!aez J, Llorens M. Hydraulic performance of a deep stabilization pond fed at 3.5 m depth. Water Res 2000;34(3):1042–9. [24] Pires EC, Kellner E. Thermal stratification model for wastewater stabilization ponds. 15th Brazilian Congress of ! ! Mechanical Engineers. Aguas de Lindoia, S*ao Paulo, Brazil, November 22–26, 1999 (in Portuguese). [25] Kellner E, Pires EC. The influence of thermal stratification on the hydraulic behaviour of waste stabilization ponds. Water Sci Technol 2000;45(01):41–8. [26] Mara DD, Pearson HW. Designing against pond horrors. World Water, April 1987.
Water Research 38 (2004) 645–654
Seasonal and diurnal variations of temperature, pH and dissolved oxygen in advanced integrated wastewater pond systems treating tannery effluent I. Tadessea,*, F.B. Greenb, J.A. Puhakkaa a
Institute of Environmental Engineering and Biotechnology, Tampere University of Technology, P.O. Box 541, Tampere 33101, Finland b Lawrence Berkeley National Laboratory, One Cyclotron Road, MS 70A-3317F, Berkeley, CA 94720, USA Received 6 January 2003; received in revised form 9 September 2003; accepted 2 October 2003
Abstract Seasonal and diurnal fluctuations of pH, dissolved oxygen (DO) and temperature were investigated in a pilot-scale advanced integrated wastewater pond system (AIWPSs) treating tannery effluent. The AIWPSs was comprised of advanced facultative pond (AFP), secondary facultative pond (SFP) and maturation pond (MP) all arranged in series. The variations of pH, DO and temperature in the SFP and MP followed the diurnal cycle of sunlight intensity. Algal photosynthesis being dependent on sunlight radiation, its activity reached climax at early afternoons with DO saturation in the SFP and MP in excess of over 300% and pH in the range of 8.6–9.4. The SFP and MP were thermally stratified with gradients of 3–5 C/m, especially, during the time of peak photosynthesis. The thermal gradient in the AFP was moderated by convective internal currents set in motion as a result of water temperature differences between the influent wastewater and contents of the reactor. In conclusion, the AFP possessed remarkable ability to attenuate process variability with better removal efficiencies than SFP and MP. Hence its use as a lead treatment unit, in a train of ponds treating tannery wastewaters, should always be considered. r 2003 Elsevier Ltd. All rights reserved. Keywords: Tannery effluent; Photosynthesis; Seasonal and diurnal variations; Stratification; Advanced integrated wastewater pond system
1. Introduction Wastewater ponds are natural systems whose biochemical and hydrodynamic processes are influenced by meteorological factors such as sunshine, wind, temperature, rainfall and evaporation [1]. Sun is the driving force in the purification process of pond systems as is in any natural water body. Wastewater ponds, however, differ greatly from natural water bodies, e.g. lakes and oceans in nutrient loading, oxygen demand, depth, size, water residence time, material residence time and flow pattern [2]. Variations in meteorological factors often *Corresponding author. P.O. Box 101204, Addis Ababa, Ethiopia. Fax: +251-1-634997. E-mail address: [email protected] (I. Tadesse).
trigger fluctuations in water quality parameters, such as, temperature, pH and dissolved oxygen (DO) both seasonally and diurnally. Photosynthetic oxygenation, which is essential for aerobic oxidation of the waste organics, varies diurnally with light intensities with peaks occurring between 1300 and 1500 h in the tropics [3]. It is not uncommon to find dissolved oxygen supersaturation between 300% and 400% on the top water layers of algal ponds on warm sunny afternoons in the tropics [4–6]. The pH of algal ponds increases with photosynthesis as algae continue consuming carbon dioxide faster than it can be produced by bacterial respiration. As CO2 diffusion from the atmosphere is minimal, primarily due to elevated surface water temperatures, the CO2 deficit during peak photosynthesis is met from the dissociation
0043-1354/$ - see front matter r 2003 Elsevier Ltd. All rights reserved. doi:10.1016/j.watres.2003.10.006
ARTICLE IN PRESS I. Tadesse et al. / Water Research 38 (2004) 645–654
646
of bicarbonate ions. This bicarbonate dissociation with concomitant consumption of CO2 by algae increases the concentration of hydroxyl ions in the water column causing the pH to rise to well over 10 [7–9]. Temperature has a pronounced effect on both biochemical and hydrodynamic processes of pond systems. During the daylight hours, solar radiation heats up the top water layer thus causing thermal stratification with a consequence of the warmer and lighter water overlaying the cooler and denser deeper water. Such density stratification reduces the performances of pond systems by increasing short-circuiting and disrupting the internal diffusive and advective mass transfer mechanisms. Thermal stratification also localizes algae into bands of 10–15 cm width that move up and down through the water column in response to changes in light intensity [10]. After dusk photosynthesis ceases and the rapid cooling on the surface with intense wind during the night forces the denser top water layer to sink into the bottom causing mixing in the water column. By early morning the algal ponds are destratified with no marked gradient in their water column quality. At dawn, the warming up of the top water layers continues until it reaches maximum by early afternoon and distinct density layers are established and the cycle repeats itself. Wind plays an important role in the vertical mixing of the pond contents by destroying thermal stratification. Understanding the seasonal and diurnal changes of water quality parameters gives insight into the process mechanism involved and helps devising short and longterm operational strategies. The aim of this study was to 1. Combined tannery wastewater 2. Screening & settling 3. Horizontal settling tanks 4. Feed pump 5. Submerged fermentation pit
examine the variations of pH, DO and temperature within the AIWPSs reactors and verify the occurrence of any adverse stratification in any of the reactors.
2. Material and methods 2.1. Study site The pilot-scale AIWPSs Facility was installed in Modjo tannery, Ethiopia 75 km south of the capital Addis Ababa along the major high way connecting the capital to the Red Sea port of Djibouti. Modjo town is located 8 350 North and 39 100 East with an altitude of 1825 m above mean sea level [11]. The climatic condition is that of tropical rain with the mean annual rainfall varying between 680 and 2000 mm. The mean air temperature recorded, at the site, during the course of the experiment varied between 15 C and 30 C. 2.2. Pilot-scale AIWPS The pilot-scale AIWPSs was comprised of an advanced facultative pond (AFP) followed by secondary facultative pond (SFP) and maturation pond (MP) (Fig. 1). Pretreatment was provided by passing the raw combined tannery wastewater through successive bar screens and detaining the screened wastewater for 1 day in a two-chamber horizontal settling tank (HST) (No.3, Fig. 1). The physical dimensions of the AIWPSs reactors were as given in Table 1.
6. Advanced facultative pond 10. Return pump 7. Secondary facultative pond 11. Optional oxygenated 8. Maturation pond water recirculation 9. Treated effluent SP =Sampling points
Fig. 1. Schematic flow diagram of the pilot-scale AIWPSs facility built at Modjo Tannery, Ethiopia [12]. Table 1 Physical dimensions of the AIWPSs reactors [12] Reactor
Reactor geometry
Areal dimensions (m) (m)
Reactor depth Water column Mid-depth (m) depth (m) area (m2)
Liquid volume (m3)
Advanced facultative pond Submerged fermentation pit Secondary facultative pond Maturation pond
Square Cylindrical Rectangular Rectangular
33 +1.6 3.2 1.6 21
3.5 2 1.8 1.3
23 4 8 2
3.3 2 1.5 1
9 2 5 2
ARTICLE IN PRESS I. Tadesse et al. / Water Research 38 (2004) 645–654
2.3. Sampling and analytical methods Sample collection was based on the techniques outlined by Pearson et al. [13] and the analyses of organics (BOD5 and COD), nutrients (NH3–N and Ortho-P) and other pertinent parameters were conducted according to standard methods for the examination of water and wastewater [14]. Dissolved oxygen was monitored using Hach model 16046 portable dissolved oxygen meter and pH using an Inventron pH/mV meter fitted with a combined pH electrode. Mid-depth temperatures were measured by polystyrene-float suspended maximum and minimum thermometer while electrical conductivity was tested by a WTW-8120 Weilheim conductivity meter. 2.4. Tanning process and waste characteristics On average, Modjo tannery processed 25 t of skin to wet-blue and crust leather and produced 250 m3 effluent daily during the time of the study. All wastewater streams emerging from the different process halls were diverted into one underground channel from where all was conveyed in an open ditch to the nearby Modjo River. The tannery was close to EPA [15] subcategory I in terms of its hair pulping and chrome tanning, retanning and wet-finishing processes. Unlike EPA’s subcategory I, Modjo tannery was processing exclusively sheep and goatskins. Mean composition of the raw tannery wastewater, in terms of major pollutants of interest, were as given in Table 3. 2.5. Operating conditions The volume of inflow to the AIWPSs facility was regulated by varying the speed of the feed peristaltic
647
pump against a tabulated calibration curve. Feed flows of 0.500, 0.800 and 1.275 m3 were applied during phase 1, 2 and 3 of the experiment, respectively. Table 2, summarizes the feed volumes and the corresponding mean organic loading rates applied during the three phases of the experiment.
3. Results and discussions 3.1. Performances of the reactors The AIWPSs reactors achieved high pollutant removals as shown in Table 3. Among the AIWPSs reactors, the highest removal for 3+ BOD5, COD, Ortho-P, SO2 occurred in the 4 and Cr AFP. Much of the suspended solids (50–90%) was removed in the HST within 1-day retention time. The BOD5 removal of 24–38% obtained in the HST is in conformity with performances of unaided settling tanks that are generally reported to achieve a BOD5 removal of 25–40% [18]. The effluent ammonia–nitrogen concentration of the AFP was higher than its influent due to anaerobic ammonification of organic nitrogen in the lower fermentation pit. An increase in sulphate concentration in the final effluent could be due to the oxidation of sulphide back to sulphate in oxygen-rich shallow water column of the MP. The suspended solids from the two algal ponds (SFP and MP) also showed an increase in concentration and that was due to the production of more algal biomass. From the filtered effluent BOD5 values of the SFP and MP reported in Tadesse [17], algae was found to constitute 50–60% of the final effluent BOD5. The observation was close to the findings by Mara et al. [10] who reported up to 65%
Table 2 Mean applied organic loading rates [16] Feed phase
Parameters
Units
AIWPSs reactors AFP
SFP
MP
Feed I
Feed flow Hydraulic retention time Surface BOD5 loading Volumetric BOD5 loading
l/d days kg/ha-d g/m3-d
500 8 5560 280
500 16 250 15
500a 4 210 20
Feed II
Feed flow Hydraulic retention time Surface BOD5 loading Volumetric BOD5 loading
l/d days kg/ha-d g/m3-d
800 5 8300 415
800 10 895 55
800 3 1480 150
Feed III
Feed flow Hydraulic retention time Surface BOD5 loading Volumetric BOD5 loading
l/d days kg/ha-d g/m3-d
1275 3 13640 680
1275 6 1020 65
1275 2 1590 160
a
Net evaporative losses were negligible and hence not accounted for.
a
ARTICLE IN PRESS I. Tadesse et al. / Water Research 38 (2004) 645–654
648
Table 3 Mean removal performances of the AIWPS reactors (complied from Tadesse et al. [12,16,17] Parameter
Feed phase
Raw tannery wastewater (mg/l)
Effluent
Overall removal (%)
HST (mg/l)
AFP (mg/l)
SFP (mg/l)
MP (mg/l)
SS
I II III
5071 3219 2517
418(92) 821(74) 1173(53)
339(19) 227(72) 158(87)
279(18) 357(57) 176(11)
386(38) 343(4) 143(19)
92 89 94
BOD5
I II II
3580 3075 2810
2225(38) 2075(33) 2142(24)
250(89) 560(73) 396(82)
85(66) 370(34) 251(37)
50(41) 236(36) 170(32)
99 92 94
COD
I II II
11157 9650 7937
4544(59) 7405(23) 5840(26)
669(85) 1382(81) 2359(60)
463(31) 877(37) 887(62)
364(21) 756(14) 813(8)
97 92 90
NH3-N
I II II
152 136 156
135(11) 130(4) 137(12)
156(16) 224(72) 242(77)
58(63) 100(55) 120(50)
20(66) 79(21) 89(26)
87 42 43
Ortho P
I II II
12 9 16
9(25) 8.9(1) 12(25)
8(11) 1.95(78) 2.7(77)
7(12.5) 0.75(62) 1.2(56)
7(0) 0.75(0) 1(17)
42 92 94
SO2 4
I II II
840 1279 840
624(26) 609(52) 765(9)
115(82) 78(87) 54(93)
150(35) 90(15) 26(52)
155(3) 120(33) 40(54)
82 91 95
Cr3+
I II II
46.00 2.50 12.35
13(72) 1.02(59) 6.72(46)
0.66(95) 0.43(58) 1.38(80)
0.33(50) 0.24(44) 0.93(30)
0.31(6) 0.21(13) 0.78(16)
99 92 94
Values in parentheses are % removals. HST=horizontal settling tank, AFP=advanced facultative pond, SFP=secondary facultative pond, MP=maturation pond.
algal BOD5 in the final effluents of their Brazilian experimental facultative ponds. Whether regulators should include algal BOD5 or not is still controversial, although it is often included by default. Proponents of pond systems argue that algae discharged with the pond effluent are rapidly consumed by zooplankton in the receiving watercourse and in daylight hours they produce oxygen, so they have little chance to exert their BOD5 [19]. 3.2. Variations of air and mid-depth temperatures The air and mid-depth temperatures of the AIWPSs reactors recorded throughout the experiment were as shown in Fig. 2. Ambient air temperature was taken on the sampling deck built around the AFP 2.75 m above ground level, while mid-depth temperatures were taken at 1.15, 0.75 and 0.5 m below the water surfaces of the AFP, SFP and MP, respectively. Except during MayAugust 1997 and December 1998 to January 1999, the air temperature was found to be higher than the reactors
mid-depth water temperatures by an average of 1.4– 2.3 C. During May to August 1997, the mid-depth water temperature of the MP was higher than the ambient air temperature by an average of 4 C. The reactors volumes had an important influence on the middepth temperatures. The MP, with the smallest liquid volume among the reactors, followed the air temperature more closely than the AFP and SFP. The mean COD removal in the AFP was above 80% during the initial and intermediate phases and decreased to 60% during phase 3 (Table 3). The exceptionally low mean mid-depth water temperature of 17 C experienced in the AFP during November– December 1998 (Fig. 2) could, among other factors, have contributed to the decline in COD removal efficiency during phase 3. 3.3. Diurnal variations The peaks in the diurnal variations of temperature, pH and DO observed in the top water layers of
ARTICLE IN PRESS I. Tadesse et al. / Water Research 38 (2004) 645–654 Air
30
AFP
SFP
Phase 1
Temperature (°C)
649 MP
Phase 2
Phase 3
25
20
3F
20 J
31 D
6 J 99
4D
26 D
27 N
6N
20 N
30 O
16 O
7A
7O
9J
24 J
4J
25 J
28 M
4D
4 M 98
29 N
9O
28 O
9A
26 S
8J
2A
26 M
18 M
14 M 97
15
Date
Fig. 2. Variations in the air and mid-depth temperatures of the AIWPSs reactors.
SFP
MP
Mean effluent temperature (˚C)
Mean effluent temperature (˚C)
AFP 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15
Phase 1
8
9
10
11
12
13
14
15
16
17
18
Mean effluent temperature (˚C)
AFP
SFP
SFP
MP
Phase 2
8
Hours of the day 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14
AFP
33 32 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 9
10
11
12 13 14 Hours of the day
15
16
17
18
MP
Phase 3
7
8
9
10
11
12
13
14
15
16
17
18
Hours of the day
Fig. 3. Diurnal variations of temperature.
AFP, SFP and MP occurred between 1300 and 1500 h (Figs. 3–5). In the SFP and MP, DO followed the temperature and pH variations. This was somewhat expected, because photosynthetic oxygenation was high when solar insolation and consequently the temperature were high. Early afternoon DO concentration in the top upper most water layer of the MP was over 20 mgO2 l1 throughout the experiment (Fig. 5). This was a 300%
supersaturation over the saturation DO level of 6.9 mgO2 l1 occurring at the prevailing water temperature. Picot et al. [20] also observed DO levels of up to 350% saturation in their high rate algal pond system during the early afternoons. As the feed organic loading on the SFP and MP increased from phase 1 to phase 3 (a four-fold increase in all), the length of time during which DO was above the saturation gradually decreased. For
ARTICLE IN PRESS I. Tadesse et al. / Water Research 38 (2004) 645–654
650 AFP
9.6
SFP
MP
AFP
9
SFP
MP
9.4 8.8 Mean effluent pH
Mean effluent pH
9.2 9 8.8 Phase 1
8.6 8.4
8.6 8.4 Phase 2 8.2
8.2 8
8
7.8 8
9
10
11
12
13
14
15
16
17
18
AFP
8.8
7.8 8
Hours of the day
SFP
9
10
11
12 13 14 Hours of the day
15
16
17
18
MP
Mean effluent pH
8.7 8.6 8.5
Phase 3
8.4 8.3 8.2 8.1 7
8
9
10
11
12 13 14 Hours of the day
15
16
17
18
Fig. 4. Diurnal variations of pH.
example, during phase 1 both SFP and MP had sustained supersaturation concentrations for over 6 h. However, during phases 2 and 3, the duration over which the MP was above the DO saturation level was reduced to only 4 and 5 h, respectively (Fig. 5). Similar situation was also observed for SFP except during phase 3 when its DO level dropped below the saturation level. Soon after 1700 h, DO in the two algal ponds was below 2 mg O2 l1. There was a net oxygen carryover of more than 3 mg O2 l1 from the night into the next morning in both algal ponds during phase 1. This was due to the low organic loading that resulted in limited respiration during the night. The superficial DO concentration of AFP remained around 0.2 mg O2 l1 throughout the experiment. The pH in the upper water layers of the SFP and MP was above 8.4 during much of day with peaks (pH 9.4) coinciding with the time of peak photosynthesis (Fig. 4). The nighttime pH in SFP and MP dropped to 8.4–8.6 mainly as a result of the absence of photosynthesis and the respiration of organic wastewater constituents by algae and bacteria, which released CO2. It appears from Fig. 3 that MP had the warmest top water layer. There was a marked drop in surface water temperature in the AFP during phase 3 (Fig. 3) that possibly brought about a 25% reduction in the COD removal efficiency (Table 3).
3.4. Vertical variations Vertical variations in DO, pH and temperature (Figs. 6–8) were monitored in all AIWPSs reactors to delineate oxic and anoxic layers and verify whether stratification was taking place. Depth variations in DO, pH and temperature within the AFP were minimal, as shown in Fig. 6. This low thermal gradient is usually ascribed to mixing due to rising biogas bubbles [21]. In this study, intense biogas mixing was far from being the cause for low thermal gradient, as the overall biogas production was found to be low [22]. Convective water movement, due mainly to temperature difference up to 4 C between the incoming wastewater and the AFP’s content, may have a more pronounced effect on the low thermal gradient than the rising biogas bubbles. Torres et al. [23] also observed the effects of such thermal currents in their studies in deep Spanish ponds. The thermal gradient of 2 C in a water column of 3.30 m also fulfils the 0.6 C/m gradient limit set by Pires and Kellner [24] for the occurrence of thermal stratification in ponds of tropical region, quoted by Kellner and Pires [25]. The AFP was anaerobic with DO less than 0.2 mg O2 l1 throughout its depth. The pH varied between 8 and 8.4 at the surface to 7.2 at the bottom, which was conducive for anaerobic digestion of the waste in the fermentation
ARTICLE IN PRESS I. Tadesse et al. / Water Research 38 (2004) 645–654 AFP
SFP
MP -1
18 16 14 12 10 8
O 2 saturation level (6.9 mg/l)
6 4 Phase 1
2
SFP
MP
20 18 16 14 12 10 8
O 2 saturation level (6.9 mg/l)
6 4
Phase 2
2 0
0 8
9
10
11
12 13 14 Hours of the day
AFP
22
15
16
17
SFP
7
18
8
9
10
11
12
13
14
15
16
17
18
Hours of the day
MP
20
-1
Mean dissolved oxygen (mg O2 l )
AFP
22
20
Mean effluent dissolved oxygen (mg O2 l )
-1
Mean effluent dissolved oxygen (mg O2 l )
22
651
18 16
Phae 3
14 12 10 8
O 2 saturation level (6.9 mg/l)
6 4 2 0 7
8
9
10
11
12
13
14
15
16
17
18
Hours of the day
Fig. 5. Diurnal variations of dissolved oxygen. O 2 (mg l ), pH & Temp. (° C) 4
8
12
16
20
-1
-1
0
24
4
8
12
16
20
0
5
5
10
10
10
20
20
20
30 40
O2
pH
60 80 100 120 140 200
40
O2
pH
60 80 100 120 140 200
Phase 1
320
°C
4
8
12
16
20
30 40 60
pH O2
°C
80 100 120 140 200 260
260
260 320
˚C
30
Depth below water surface (cm)
5
Depth below water surface (cm)
Depth below water surface (cm)
0
O 2 (mg l ), pH, Temp. (° C)
O 2 (m g l ), pH & Tem p. (° C)
-1
320
Phase 2
Phase 3
Fig. 6. Vertical variations of DO, pH and temperature in the advanced facultative pond of the AIWPS.
pit. This near-to-neutral pH of 7.2 at the bottom was predominant up to 40 cm above the sludge-water interface. The DO profile for SFP (Fig. 7) demonstrates that the oxypause (depth where oxygen depletion occurred) gradually decreased as organic loading
increased from phase 1–3. At the beginning of the experiment (phase 1, Fig. 7), nearly one-third of liquid depth of SFP was above 2 mg O2 l1 with a maximum (>20 mg O2 l1) occurring in the top 15 cm of the pond water column during the early afternoons.
ARTICLE IN PRESS I. Tadesse et al. / Water Research 38 (2004) 645–654
652 -1
-1
O 2 (mg l ), pH & Temp. (° C) 4
8 12 16 20 24 28
0
4
O 2 (mg l ), pH & Temp. (° C)
8 12 16 20 24 28
0
5
5
10
10
10
15
15
20 25 O2
pH
30
°C
40 60 80
Depth below water surface (cm)
5
Depth below water surface (cm)
Depth below water surface (cm)
0
-1
O 2 (mg l ), pH & Temp. (° C)
20 25 O2
°C
pH
30 40 60 80
20 25 O2
30
°C
pH
40 60 80 100
120
120
120
140
140
140
Phase 1
8 12 16 20 24
15
100
100
4
Phase 2
Phase 3
Fig. 7. Vertical variations of DO, pH and temperature in the secondary facultative pond of the AIWPS.
-1
-1
O 2 (mg l ), pH & Temp. (˚C) 0
4
-1
O 2 (mg l ), pH & Temp. (˚C)
8 12 16 20 24 28
O 2 (mg l ), pH & Temp. (˚C)
0 4 8 12 16 20 24 28 32
0
5
5
5
10
10
10
O2
pH
30
50
70
°C
15
20 O2
°C
pH
30
50
15
20 O2
pH
°C
30
50
90
90
Phase 1
8 12 16 20 24 28
70
70
90
Depth below water surface (cm)
20
Depth below water surface (cm)
Depth below water surface (cm)
15
4
Phase 2
Phase 3
Fig. 8. Vertical variations of DO, pH and temperature in the maturation pond of the AIWPS.
Algal–bacteria respiration during the night depleted the water column oxygen to the extent that the morningtime DO (except during phase 1, Fig. 5) was nearly zero. The situation was, however, reversed to supersaturation by early afternoon when photosynthesis reached its peak. The thermal behaviour of SFP was characterized by brief thermocline (at 20 cm depth) usually occurring during the late afternoons. During this brief period of
stratification there was a 5–8 C difference between the surface and bottom water layers. The vertical pH in the SFP during peak photosynthesis varied from 9 near the water surface to 8.2 at the bottom. The MP appeared to be oxygenated to nearly half of its liquid depth during phase 1 (Fig. 8). Successive high organic loading rates gradually diminished the oxygenated depth from 45 cm during phase 1 to only 15 cm
ARTICLE IN PRESS I. Tadesse et al. / Water Research 38 (2004) 645–654
during phase 3 (Fig. 8). The relatively shallow water column (1 m) of the MP permitted sufficient light penetration thereby allowing photosynthesis to occur at deeper layers. There was a period of thermal stratification during peak photosynthesis with variations of 3–5 C between the top and bottom water layers. A pH gradient of 1 unit/m (i.e., 9.4 at the surface to 8.4 at the bottom) was also common in the MP at the time of peak photosynthesis. 3.5. Programming of effluent discharge Knowledge of the diurnal variations of water quality parameters help in devising realistic discharge programmes. Picot et al. [20] proposed a daytime effluent discharge scheme for NH3–N and PO4–P. Daytime discharge is also preferred from public health consideration as releasing effluent from the surface water layer, where photosynthesis is the highest, has the lowest pathogen concentration. But discharging effluent from the layers of highest photosynthesis has the apparent disadvantage of releasing high algal solids with the effluent. As a trade-off between the solids concentration and bacteriological quality of effluent, Mara and Pearson [26] recommended 60 cm take-off point from facultative ponds and 50 cm from maturation points. It is evident from these proposals that the actual discharge programme and the depth from which the effluent is withdrawn depend on the quality of the raw wastewater, intended use of the treated effluent and the discharge limits applicable to the particular place area. In setting acceptable effluent discharging programme for tanneries, considerations should also be given to the manufacturing schedule and volume of wastewater produced. In large tanneries, where leather manufacturing is carried out round-the-clock and the presence of excreted human pathogens in the raw wastewater is minimal, a continuous release of the treated effluent may be considered.
4. Conclusions 1. The study demonstrates that advanced integrated wastewater pond system preceded with simple pretreatment can adequately treat raw combined tannery wastewater. 2. In advanced integrated wastewater pond system, seasonal and diurnal variations of pH, temperature and dissolved oxygen are more significant in the secondary facultative and maturation ponds than in the primary advanced facultative pond. 3. The secondary facultative and maturation ponds show a thermal stratification gradient of 3–5 C/m, especially, during the time of peak photosynthesis
653
and diurnal variations in pH, temperature, and dissolved oxygen follow the pattern of the daily cycle of the sunlight intensity. 4. In the primary advanced facultative pond, convective and vertical thermal currents contribute more to low thermal gradient than mixing by rising biogas bubbles. 5. A continuous discharging of treated effluent may be necessary for tanneries that operate round-the-clock and where their raw effluents contain minimal excreted human pathogens.
Acknowledgements We wish to express our gratitude to the United Nations Industrial Development Organization, UNIDO, Vienna, Austria, which financed [96/024/ML-DP/ ETH/93/005] the pilot experimental work carried out in Modjo tannery as part of its development assistance to the leather industry in Ethiopia. Financial support also came from the Academy of Finland and Tampere University of Technology doctoral school in the form of study grant to the first author. We are also grateful to Modjo Tanner for providing all necessary assistance during the conduct of the research.
References [1] WHO. Wastewater stabilization ponds: principles of planning and practice. WHO-EMRO Technical Publication No. 10. Alexandria, Egypt.WHO-EMRO; 1987. 138p. [2] Gu R, Stefan H. Stratification dynamics in wastewater stabilization ponds. Water Res 1995;29(8):1909–23. [3] Pescod MB, Mara DD. Design, operation and maintenance of wastewater stabilization ponds. In: Pescod MB, Arar A, editors. Treatment and use of sewage effluent for irrigation. UK: Butterworths; 1988. p. 93–115. [4] Vonshak A, Abeliovich A, Boussiba S, Richmond A. Production of Spirulina biomass: effects of environmental factors and population density. Biomass 1982;2(175). [5] Richmond A. Outdoor mass cultures of microalgae. In: Richmond A, editor. Handbook of microalgal mass culture. Florida: CRC Press; 1986. p. 285–330. [6] Picot B, Bahlaoui A, Moersidik S, Baleux B, Bontoux J. Comparison of the purifying efficiency of high rate algal pond with stabilization pond. Water Sci Technol 1992; 25(12):197–206. [7] Azov Y. Effect of pH on inorganic carbon uptake in algal cultures. Appl or Environ Microbiol 1982;43:1300–6. [8] Buhr HO, Miller SB. A dynamic model of the high rate algal bacterial wastewater treatment pond. Water Res 1983;17:29–37. [9] Mara DD, Pearson HW. Design manual for waste stabilization ponds in Mediterranean Countries, Leeds, England: Lagoon Technology International; 1998. 113p.
ARTICLE IN PRESS 654
I. Tadesse et al. / Water Research 38 (2004) 645–654
[10] Mara DD, Pearson HW, Silva SA. Brazilian stabilizationpond research suggests low-cost urban applications. World Water, July 1983. [11] EMA. National Atlas of Ethiopia. Addis Ababa: Ethiopian Mapping Authority; 1988. 76p. [12] Tadesse I, Isoaho SA, Green FB, Puhakka JA. Transformation of sulphur in advanced integrated wastewater pond systems treating tannery effluent. J Am Leather Chem Assoc 2003;98(6):203–15. [13] Pearson HW, Mara DD, Bartone CR. Guidelines for the minimum evaluation of the performance of full-scale waste stabilization ponds. Water Res 1987;21(9):1067–75. [14] APHA, AWWA and WEF. Standard methods for the examination of waste and wastewater. 18th ed. American Public Health Association, American Water Works Association and Water Environment Federation, New York, 1992. [15] EPA. Development document for effluent limitations guidelines, standards for the leather tanning and finishing point source category. Cincinnati, USA: US Environmental Protection Authority; 1982. [16] Tadesse I, Isoaho SA, Green FB, Puhakka JA. Removal of organics and nutrients from tannery effluent by advanced integrated wastewater pond systems technology. Water Sci Technol 2003;48(2):307–14. [17] Tadesse I. Tannery effluent treatment in advanced integrated wastewater pond system. Doctoral thesis. Tampere, Finland: Tampere University of Technology, 2003.
[18] Metcalf & Eddy. Waste engineering: treatment disposal reuse, Singapore: McGraw-Hill; 1991. [19] Mara DD. Waste stabilization ponds: effluent quality requirements and implications for process design. Water Sci Technol 1996;33(7):23–31. [20] Picot B, Moersidik S, Casellas C, Bontoux J. Using diurnal variations in a high rate algal pond for management pattern. Water Sci Technol 1993;28(19):169–75. [21] Pena MR, Madera CA, Mara DD. Feasibility of waste stabilization pond technology for small municipalities in Colombia. Water Sci Technol 2000;42(10–11): 273–82. [22] Tadesse I, Green FB, Puhakka JA. The role of sulphidogenesis in anaerobic treatment phase of tannery wastewater treatment in advanced integrated wastewater pond system. Biodegradation 2003;14:219–27. [23] Torres JJ, Soler A, S!aez J, Llorens M. Hydraulic performance of a deep stabilization pond fed at 3.5 m depth. Water Res 2000;34(3):1042–9. [24] Pires EC, Kellner E. Thermal stratification model for wastewater stabilization ponds. 15th Brazilian Congress of ! ! Mechanical Engineers. Aguas de Lindoia, S*ao Paulo, Brazil, November 22–26, 1999 (in Portuguese). [25] Kellner E, Pires EC. The influence of thermal stratification on the hydraulic behaviour of waste stabilization ponds. Water Sci Technol 2000;45(01):41–8. [26] Mara DD, Pearson HW. Designing against pond horrors. World Water, April 1987.