Transformation of ammonium to nitrate upon sewage discharge to sub-surface environment exposes about 65 million households in rural and urban India to risks of drinking nitrate contaminated groundwater. Building on earlier research, a twin pit is modified in Mulbagal town, Karnataka, to remove nitrate in pit toilet sewage and is functional for nearly one year. The first pit serves as an anaerobic chamber, while the second pit facilitates aerobic reactions in the upper half and is equipped with a bio-barrier in its lower half. Quality of treated sewage is monitored by soil water samplers installed adjacent to the pit. After anaerobic digestion in pit 1, sewage flows into the aerobic chamber (upper half of pit 2), where COD/N ratio of 1.49 to 1.73 facilitates aerobic conversion of ammonium to nitrite and nitrate ions. Annamox reactions in a bio-barrier chamber (lower half of pit 2) reduce ammonium and nitrite concentrations, while denitrification reactions in the bio-barrier remove nitrite and nitrate from pit toilet sewage. Besides nitrate, the modified twin pit reduces COD (chemical oxygen demand), ammonium, and thermotolerant coliform levels in the discharged sewage.
The anaerobic character of pit toilet sewage conserves N as ammonium; upon release of leachate to the sub-surface, ammonium ions transform to nitrate under conditions of adequate oxygen availability and reduced competition from organic carbon and contaminate the groundwater (Campos et al. 1999; Carrera et al. 2004; Ling & Chen 2005; Dzwairo et al. 2006; Lu et al. 2008; van Voorthuizen et al. 2008; Graham & Polizzotto 2013; Nyenje et al. 2013; Rao et al. 2013). Nitrate originating from pit toilet leachate has the potential to expose about 65 million households in rural and urban India to risks of drinking nitrate contaminated groundwater. Bio-barriers composed of sand and organic substrate have been earlier employed for nitrate removal in laboratory column experiments (Hunter 2009; Cameron & Schipper 2010; Liu et al. 2013; Rao & Malini 2015).
Building on earlier research, a twin pit toilet is modified (Rao et al. 2017) to incorporate anaerobic, aerobic and bio-barrier chambers to reduce nitrate loads in discharged sewage. The pour flush toilet connected to the modified twin pit serves four to eight people (users vary depending on house guests) and is located in Mulbagal town Karnataka (13.165494 N, 78.395192 E). Sullage does not enter the modified twin pit and is separately discharged into a municipal drain. The sub-surface profile at the pit toilet location comprises a few meters (0–6 m) of reddish-brown sandy clay-silt, followed by sandy deposit of variable thickness which is underlain by soft, disintegrated weathered rock. Mulbagal town has a geographical area of 8.5 square km and a population of about 60,000. This study reports the efficacy of the modified twin pit for nitrate removal from pit toilet leachate after nearly one year of functioning.
DESCRIPTION OF MODIFIED TWIN PIT
The first pit serves as the anaerobic chamber, while the second pit facilitates aerobic reactions in the upper half and is equipped with bio-barrier (bio-barrier chamber) in the lower half. Unlike a conventional twin pit, the walls and base of the anaerobic chamber and aerobic chamber (upper half of pit 2) are sealed to prevent liquid flow into surrounding soil (Figure 1). Air-exchange is facilitated in the aerobic chamber by keeping the upper surface open to the atmosphere and installation of a water circulation pump (Figure 1). A mixture of air-dried cattle manure, sand, and gravel is used as bio-barrier media for nitrate/nitrite reduction. Cattle manure serves as an affordable organic C source; sand particles act as media for attached bacterial growth, while gravel improves the permeability of the barrier. Three suction-based soil-water samplers (air entry value of soil-water sampler cup = 50 kPa) are positioned to intersect the flow path of treated sewage emerging from the bio-barrier. A schematic of the modified twin pit is shown in Figure 1 and construction details of the modified twin pit are available in Rao et al. (2017).
FUNCTIONING OF MODIFIED TWIN PIT
The anaerobic chamber (pit 1) can hold 1,220 liters of sewage which allows 20 days of storage based on average sewage input of 60 liters/day. About 60 cm of free board is allowed for gas collection above the liquid surface in the anaerobic chamber (Figure 1). Dimensions of the anaerobic chamber (diameter = 120 cm and depth = 168 cm, Figure 1) are larger than recommended for twin pit pour flush toilets (diameter = 105 cm, depth = 100 cm; GOI 2012) which allows flow of the clearer, anaerobically digested liquid fraction for aerobic treatment. A concrete dome covers the anaerobic chamber and a vent pipe facilitates escape of gases produced during anaerobic digestion. IS 2470 (Part 1 1985) recommends de-sludging of septic tanks between 0.5 and 1 year; accordingly, accumulated sludge in the anaerobic chamber was discarded after 260 days. According to Bounds’ (1997) equation, 840 liters of sludge would have accumulated in the anaerobic chamber after 0.71 years (260 days). Gases generated during anaerobic decomposition were not analyzed.
The aerobic chamber (upper-half of pit 2) holds 210 liters of anaerobically digested liquid, which allows 3.5 days of storage (average daily input = 60 liters). The overflow from the aerobic chamber (upper half of pit 2) drains into the bio-barrier chamber (lower half of pit 2). The bio-barrier chamber can hold 550 liters of aerated sewage and represents 9 days of input (average of 60 liters/day). Liquid from the bio-barrier chamber percolates through the bio-barrier and is discharged into soil beneath pit 2. Liquid captured by soil-water samplers measure the quality of sewage treated by the modified twin pit.
FIELD SAMPLING AND ANALYSIS
Samples were collected from the anaerobic chamber, aerobic chamber, bio-barrier chamber, and two of the three soil-water samplers – termed L1 and L3 (soil water sampler L2 did not function as its tubing was damaged during back-filling) and shipped and stored (at 4 °C) prior to laboratory analysis. The anaerobic chamber sample was collected from the clear zone (Figure 1). Liquid in the aerobic chamber is continuously agitated by a re-circulation pump and the sample is therefore representative. The bio-barrier chamber sample is representative of liquid standing above the bio-barrier, before percolation. The samples were analyzed for chemical oxygen demand (COD) using silver sulfate-sulfuric acid digestion method (APHA 1998). Ammonium, nitrate, and nitrite concentrations in the samples were determined using ion-chromatograph (Dionex ICS 2000). Bicarbonate concentration was determined using Auto-titrator (Metrohm, Titrino Plus 877). The pH of the samples was measured using electrode technique. Thermotolerant coliforms were analyzed for samples collected from soil-water samplers during the fifth and sixth months using spread plating method on membrane-fecal coliform agar at 45 °C (APHA 1998). Redox potential (Eh) of the raw and treated sewage samples was measured using Eh meter. The dissolved oxygen (DO) of raw sewage was measured using a DO probe. An untreated sewage sample was obtained from the collection chamber, that is located prior to the modified twin pit, and analyzed for the above parameters. Monitoring of the modified twin pit commenced from April 2017 to the time of writing. At the time of revising the paper, the modified twin pit has been monitored for 355 days and will continue until December 2018.
PERFORMANCE OF MODIFIED TWIN PIT
Over the period of study, raw sewage (without settleable solids) was characterized by DO of 0 mg/L, Eh (redox potential) of −198 to −246 millivolts, pH of 7.3 to 7.5, COD of 1,700–2,600 mg/L and ammonium ion concentrations of 370 to 560 mg/L.
During the period of study (355 days), pH of samples from the anaerobic chamber ranged between 7.7 and 8.25. The slightly alkaline pH of anaerobic chamber liquid is attributed to bicarbonate formed during sewage digestion. The pH of aerobic chamber liquid mostly ranged between 7.2 and 8.3, while pH of liquid collected in soil-water samplers ranged between 7.3 and 8.2.
Figure 2 plots variation of COD with time for samples collected from the anaerobic chamber, aerobic chamber, and soil-water samplers (L1 and L3). Anaerobic reactions reduce COD concentration of sewage to range between 430 and 755 mg/L at steady state (after 95 days of operation). Considering average COD values for raw and treated sewage, anaerobic treatment causes a 72% reduction in COD. Sewage in the anaerobic chamber was characterized by Eh of −315 to −345 millivolts and DO of 0 mg/L, underlining the anaerobic character of reactions in the chamber. Kujawa-Roeleveld & Zeeman (2006), Barros et al. (2008), and de Graaff et al. (2010) report 65 to 91% reduction in COD of black wastewater upon anaerobic digestion.
After 145 days, COD levels in the aerobic chamber attained steady state and values ranged between 170 and 390 mg/L. Sewage in the aerobic chamber was characterized by Eh of 150 to 170 millivolts and DO of 2 to 4 mg/L, underlining the aerobic nature of reactions in this chamber. Samples collected in the soil-water samplers exhibit steady state from the beginning of sample collection and values range between 70 and 190 mg/L. Filtration of suspended carbon particles and metabolization of organic carbon by the consortium of bacteria in the bio-barrier matrix possibly contribute to reduction in COD upon the passage of liquid.
Ammonium ion concentrations range from 370 to 560 mg/L in untreated sewage and remain unaltered in the anaerobic chamber (Figure 3). Reduction in ammonium concentration occurred after a water re-circulation pump was installed in the aerobic chamber (around 150 days) and values ranged between 50 and 150 mg/L. Decrease in ammonium concentration was accompanied with concomitant increase in nitrate (10–150 mg/L) and nitrite (50–550 mg/L) concentrations (Figure 4(a) and 4(b)). Absence of nitrate and nitrite concentrations at 180, 242, and 302 days (Figure 4(a) and 4(b)) arise due to non-functioning of the re-circulation pump. Available COD and ammonium concentrations in the clear zone of the anaerobic chamber after 95 days (Figures 2 and 3) suggest COD/N ratio of 1.49 to 1.73 is favorable for aerobic treatment. Earlier researchers (Charmot-Charbonnel et al. 1999; Carrera et al. 2004; Zhang et al. 2015) have also examined the influence of COD/N ratio (0.5 to 3.4) and observed an exponential decrease of nitrification rate as influent COD/N ratio increased; stable and high nitrite accumulation ratios around 90% were obtained at COD/N ratios of 0.5 and 1.0. Peak ammonium concentrations in the aerobic chamber (260 to 350 mg/L, Figure 3) coincide with non-functioning of the re-circulation pump during those periods (180, 242, and 302 days).
Laboratory experiments had earlier demonstrated (Rao et al. 2017) that microbial degradation of organic matter in cattle manure led to acetic acid formation. Researchers have reported the role of VFA as carbon source in microbial denitrification (Gayle et al. 1989; Paul & Beauchamp 1989; Ganaye et al. 1996; Elefsiniotis & Li 2006). The bio-barrier installed in the lower half of pit 2 successfully removes nitrite and nitrate by denitrification reaction as both anions were absent in soil-water samplers between 145 and 355 days (Figure 4(a) and 4(b)).
Thermotolerant coliform measurements were performed for various samples at 152 and 180 days of functioning of the modified twin pit. Untreated sewage has a thermotolerant coliform count of 1 × 106 CFU/mL. At 152 days, anaerobic and aerobic treatment reduced the count to 2 × 104 CFU/mL and 5 × 103 CFU/mL, respectively. Subsequent passage through the bio-barrier did not further reduce thermotolerant coliform counts; values of 2.1 × 104 and 3.2 × 103 CFU/mL were recorded in the soil-water samplers. At 180 days, the anaerobic chamber sample exhibited a value of 1.38 × 105 CFU/mL. Comparatively, the aerobic chamber and soil water sampler exhibited values of 8.6 × 103, 4.7 × 103, and 8 × 103 CFU/mL, respectively. Depletion of organic carbon during anaerobic and aerobic reactions are presumably responsible for the decay of thermotolerant coliforms. Anaerobic reactions decrease the thermotolerant coliforms by one log cycle, while aerobic reactions reduce the thermotolerant coliform counts by 2.5 cycles. Barros et al. (2008) similarly report 2.1 log cycle reduction in thermotolerant coliforms upon anaerobic and constructed wetland treatment of domestic wastewater.
A modified twin pit that performs anaerobic, aerobic, anammox, and denitrification reactions to reduce nitrate load in discharged sewage is functional in Mulbagal town, Karnataka, for nearly one year. The first pit serves as an anaerobic chamber that reduces COD level by 72%. The second pit facilitates aerobic reactions in the upper half and its lower half is equipped with a bio-barrier. COD/N ratio of 1.49 to 1.73 is favorable for ammonium oxidation in the aerobic chamber. Annamox reactions reduce ammonium and nitrite concentrations in the bio-barrier chamber. Passage of liquid through the bio-barrier removes nitrite and nitrate by denitrification reactions. Consumption of organic carbon in the anaerobic and aerobic reactions reduced the thermotolerant coliform counts by 2.5 log cycles. Sewage treatment by modified twin pit besides removing nitrate, reduces COD, ammonium, and thermotolerant coliform levels.
The authors acknowledge Arghyam Foundation, Bangalore, and Ministry of Drinking Water & Sanitation, Government of India for funding the project.