The study examines the impact of evaporation on the fate of ammonium-N reactions in blackwater-contaminated soils. During evaporation, ammonia (g) volatilization is the preferred route of NH4-N transformation and nitrate formation is initiated thereafter. Ammonia volatilization ceased at residual blackwater contents of 16–40% owing to loss of air-void connectivity. Experimental results indicated that owing to ammonia volatilization and reduced blackwater content only 23–35% of initial NH4-N concentration was transformed to NO3-N. This study also predicted the nitrate accumulation in Mulbagal town aquifer due to blackwater discharge from pit toilets. The prediction indicated that the permissible (45 mg/L) nitrate concentration in the aquifer may have been breached several decades ago, exposing the populace to prolonged drinking water contamination.
Ammonium-N in pit toilet blackwater is a major source of groundwater nitrate pollution (Pujari et al. 2012; Graham & Polizzotto 2013; Nyenje et al. 2013; Rao et al. 2013). Ammonium-N tends to form nitrate in shallow aquifers that support oxidizing conditions and exist as ammonium ions in deep aquifers that favour a reducing environment (Hammoud et al. 2018). Ammonia volatilization and adsorption of ammonium ions by soil particles are the other pathways of ammonium-N reactions in blackwater-infiltrated soils. The nitrate concentration (50 mg/L) in shallow contaminated aquifers can reach the permissible limit in less than 2 years (Templeton et al. 2015). Physical, chemical, and biological attenuation and denitrification reactions in the unsaturated soil zone can reduce nitrate concentration in contaminated aquifers (Butler et al. 2003; Graham & Polizzotto 2013; Nyenje et al. 2013; Lee et al. 2014; Rao & Malini 2014). Factors of groundwater depth, soil pH, soil water content, bacterial population and redox conditions influence the ammonium reactions in blackwater-contaminated soils (Stark & Firestone 1995; Jacks et al. 1999; Sahrawat 2008; Nyenje et al. 2013; Ma et al. 2016; Hammoud et al. 2018).
In semi-arid and tropical climate regions, the pit-toilet leach pits are located in the active soil zone (Dzwairo et al. 2006) that experience changes in moisture content from interactions with the atmosphere (Fredlund et al. 2012). Reduction in blackwater content of the contaminated soils during the dry season can favour (i) transport of gases, (ii) greater interactions between solutes in blackwater and soil particles, and (iii) bacterial cell dehydration that retards the enzyme activity of the nitrifying bacteria (Stark & Firestone 1995; Fredlund et al. 2012). This study examines the impact of evaporation on the fate of ammonium-N reactions in blackwater-contaminated soils. Results of laboratory studies with artificially contaminated soils are used to predict the decadal growth of nitrate concentration in an aquifer exposed to blackwater contamination.
Five-litre (L) batches of blackwater samples were collected at 40-day intervals during December 2016 to November 2017. The samples were collected from the inspection chamber of a single-household, pit toilet in Mulbagal town, Karnataka (distance from Bangalore = 95 km). The blackwater sample was transported to the laboratory at 4 °C and refrigerated. The various samples were used individually. A 5 L batch of blackwater was consumed in 32–35 days after which another batch was collected from the field. Residually derived red soil from the Indian Institute of Science campus, Bangalore, was used in the study. The soil has coarse (particle size: 4.75 mm to 0.075 mm) and fine (particle size <0.075 mm) contents of 54% and 46% respectively. The liquid and plastic limits of the soil correspond to 33% and 21% respectively. Sewage sludge was used as microbial inoculum source and was obtained from a 150 KLD (kilolitres/day) sequencing batch reactor (SBR) of a wastewater treatment plant in Bangalore.
Preparation of artificially contaminated compact specimens
A mixture of 1.2 L blackwater + 0.2 L sewage sludge was anaerobically digested at 37 °C for 16 days. This period was selected as it simulated the contact period between blackwater and sewage sludge in a single household leach pit. After 16 days the supernatant fraction was decanted without disturbing the settled sludge. The water contents of the supernatant fraction before (100%) and after digestion remained unchanged.
As the leach pit was functional, the collection of soil samples from beneath the pit was not possible. Artificially contaminated soils were hence prepared in the laboratory. Known volumes of 16-day-digested blackwater was mixed with dry soil batches to prepare contaminated soils having blackwater contents of 20%, 30% and 50% respectively. The blackwater-contaminated soils were hand-pressed in a circular and flat-bottomed Petri dish (inner diameter = 76 mm) to a thickness of 16 mm. The blackwater content of 20% nearly equals the plastic limit (21%) of the red soil, while the blackwater content of 30% is less than its liquid limit (33%). The blackwater content of 50% is greater than the soil's liquid limit. The 20% and 30% blackwater content specimens are representative of contaminated soils in the vicinity of the leach pit. The 50% specimen illustrates the fate of NH4-N reactions at a large blackwater content. The liquid limit is the boundary water content between liquid and plastic states of the soil and the plastic limit is the boundary between plastic and semi-solid states (Fredlund et al. 2012).
Restricted evaporation of artificially contaminated compact specimens
After t days of evaporation, three 10 g portions were sliced from each compact specimen to monitor the changes in pH, NH4-N and NO3-N contents. Each 10 g slice was extracted with 50 mL of 0.01 M KCl solution by agitating the slurry on a rotary shaker for 2 h. The centrifuged soil mass was subjected to two additional extractions. Thus 150 mL of 0.01 M KCl extractant was collected for each slice. The ammonium and nitrate concentrations in each 150 mL lot were determined. The average ion concentrations of the three slices were used (the average values with standard deviation are provided in Appendix Table A1, Supplementary Data). A portion of the unsliced soil was used to measure soil pH.
Methods of chemical analysis
Electrical conductivity (EC), pH, redox potential (Eh) and dissolved oxygen (DO) contents of the blackwater samples were measured in the field using portable pH, EC, Eh and DO meters (Eutech instruments). All other parameters were measured in the laboratory. TKN (total Kjeldahl nitrogen) and COD (chemical oxygen demand) concentrations of the blackwater samples were determined in the laboratory using unfiltered samples. The COD concentrations were determined using the silver sulfate–sulfuric acid digestion method and TKN values were determined (APHA 1998) using the TKN analyser (Pelican KEL Plus, India). Other parameters were measured for filtered samples (0.45 μm filter, Merck Millipore). The nitrate and ammonium ion concentrations were determined using an ion-chromatograph (Dionex ICS 2000). The pH of the soil extracts was measured using a table-top pH meter (Hanna pH 213) and the ammonium and nitrate concentrations were measured as described earlier.
RESULTS AND DISCUSSION
The redox potentials of the blackwater samples ranged from −189 to −298 mV (median = −255 mV) and DO was absent. The samples were characterized by NH4-N concentrations of 173–538 mg/L (median = 321 mg/L). The NO3-N concentrations ranged from 0.87 to 1.63 mg/L indicating minimal ammonium conversion due to the anaerobic nature of blackwater. The COD concentrations in the blackwater samples ranged from 1,056 to 3,557 mg/L (median = 1,788 mg/L) and TKN concentrations ranged from 199 to 619 mg/L (median = 372 mg/L). The variability in COD and N concentrations in blackwater arises from differences in amounts of flushing water and concentration of faeces and urine in blackwater (van Voorthuizen et al. 2008; Florentino et al. 2019). Despite its dominance in blackwater, ammonia has not been reported to accumulate appreciably in groundwater that is exposed to pit toilet discharge (Dzwairo et al. 2006; Graham & Polizzotto 2013).
Ammonium-N transformation reactions
Figure 1 plots the changes in NH4-N, NO3-N and pH with evaporation period for the 20%, 30% and 50% compact specimens. Mixing red soil with 50% blackwater content imposed an NH4-N concentration of 29 mg/100 g of soil. The NH4-N concentration reduced to 6.0 and 1.6 mg/100 g soil after 4 and 16 days of evaporation respectively. The NO3-N concentration decreased up to 2 days and thereafter increased from 0.18 to 6.7 mg/100 g soil after 16 days.
The reduction in pH of the 30% compact specimen (7.86–6.95) between 1 and 8 days is attributed to the release of H+ ions from NO3− formation (Equations (5) and (6)). The changes in NH4-N, NO3-N and pH with evaporation period of the 20% compact specimen are also included in Figure 1.
Figure 2 illustrates that ammonia volatilization and nitrate formation increase with reduction in water content before attaining equilibrium. Ammonia volatilization reached equilibrium at residual blackwater contents of 40%, 21% and 16% for the 50%, 30% and 20% compact specimens respectively. Apparently, the loss of connectivity of air-voids halted further ammonia volatilization. Nitrate formation became apparent (>0%) after ammonia volatilization ceased and maximum NO3-N formation (23%, 23%, 35%) is noted at residual blackwater contents of 11%, 4% and 2% for the 50%, 30% and 20% compact specimens respectively. Initiation of nitrification after ammonia volatilization implies that NH3 (g) formation is the preferred path of NH4-N transformation in blackwater-contaminated soils during evaporation. After NH3 (g) volatilization reached equilibrium, 29%, 58% and 76% of initial NH4-N concentration was available in the 50%, 30% and 20% compact specimens for possible NO3-N formation (from Figure 2).
The inability of larger NO3-N formation is ascribed to the low water content (4% and 2%, Figure 2) that apparently caused bacterial cell dehydration and retarded enzyme activity of the nitrifying bacteria (Stark & Firestone 1995). In addition, oxygen deficiency at low water content would inhibit the nitrifiers (Kim 2016). According to Henriksen & Kemp (1988) nitrifiers cease to work at O2 concentrations of 1–6 μM.
Prediction of nitrate contamination of groundwater from blackwater
The discharged blackwater can release 40 mg NO3-N/L (23% of 173 mg NH4-N/L) to 188 mg NO3-N/L (35% of 538 mg NH4-N/L) to the sub-surface soil. Therefore, 60 litres of blackwater discharged daily from a single household leach pit has the potential to release 2,400–11,280 mg of NO3-N into the subsurface soil.
Fractured granite gneiss, granites and schists constitute the 300 m thick aquifer in Mulbagal town (Hegde 2017). The median water-table depth (−13.7 m) in Mulbagal town (Rao et al. 2013) confers an aquifer thickness of 286 m (300 m minus 13.7 m). The 8.5 square km aquifer of Mulbagal town can potentially store 2.43*109 litres of groundwater.
The cumulative accumulation of nitrate in the Mulbagal town aquifer from 1851 to 2011 is calculated (Appendix Table A2) from a knowledge of:
cumulative number of dwellings in each decade (1851–2011);
cumulative NO3 discharged to the aquifer by blackwater in each decade from all the dwellings;
volume of groundwater contained in the aquifer.
The calculations for cumulative nitrate accumulation involved the following assumptions:
Nitrate produced from the discharged blackwater migrated to the aquifer without loss.
Blackwater discharge is the only source of nitrate contamination.
The decadal population growth in Mulbagal town was in accordance with Equation (8).
The number of persons in each dwelling was constant throughout the computation period.
The volume of water available in the aquifer was constant throughout the computation period.
At 23% conversion of NH4-N to NO3-N, (lower envelope, Figure 3) the results suggest that the nitrate concentration in Mulbagal aquifer may have breached the permissible limit for drinking water (45 mg/L for India) in 1991.
At 35% conversion (upper envelope), the results indicate that the permissible limit in groundwater may have been breached in 1921. The nitrate concentrations inside Mulbagal town were measured in 2010 (inner town series: ITS, Rao et al. 2013) and the data points (n = 31) are plotted in Figure 3. About 39% of data points plot in the vicinity of the 35%-conversion envelope and 26% plot in the vicinity of the 23%-conversion envelope. The remaining (35%) points are scattered between the two envelopes (Figure 3). The good agreement (65% of the data agree with the 23% or 35% conversion models) between the measured and predicted values indicate that the assumptions for nitrate accumulation in Mulbagal aquifer from blackwater discharge are reasonable. Besides variability in initial blackwater content and extent of moisture loss, other major factors that could have contributed to the scatter (35% of data points) of the measured values are:
variability in the volume of water in the aquifer;
transport of nitrate from the contaminated aquifer;
variability in population and number of persons in each household between 1851 and 1951;
contribution to NH4-N from other anthropogenic sources (animal and other livestock manure, fertilizers, plant debris, solid waste dumps);
partial denitrification during nitrate migration through the vadose zone;
nitrate utilization by plants for their growth.
Reduction in blackwater content from evaporation influences the fate of NH4-N reactions in blackwater-contaminated soils. Nitrate formation is initiated after ammonia volatilization attains equilibrium. The results indicated that NH3(g) formation, is the preferred route of NH4-N transformation in blackwater-contaminated soils. Connectivity of air-voids is essential for NH3 volatilization and the presence of adequate soil water content (>4%) is essential for microbial oxidation of ammonium to nitrate ions. The results indicate that 23–35% of the initial NH4-N concentration in blackwater can transform to NO3-N, depending on the initial blackwater content of the soil and the extent of blackwater evaporation.
Based on the number of dwellings, the NO3-N produced from the discharged blackwater and the volume of groundwater, nitrate accumulation is predicted in the Mulbagal town aquifer. The predicted values are in good agreement with the measured concentrations. The predictions also indicate that at the lower (23%) NH4-N to NO3-N conversion, the nitrate concentration in the aquifer may have breached the drinking water limit (45 mg/L) in 1991. In comparison, at the higher (35%) conversion, the permissible limit may have been breached in 1921. The significance of the findings is that the affected populace may have been consuming nitrate-contaminated water for several decades. To eliminate the risk of groundwater nitrate contamination, pit toilets capable of removing ammonia by volatilization, adsorption or denitrification should be developed.
The authors acknowledge Arghyam Foundation, Bangalore, and the Ministry of Drinking Water & Sanitation, Government of India for funding the project.
The Supplementary Material for this paper is available online at https://dx.doi.org/10.2166/ws.2019.145.