Performance investigation of the slaughterhouse wastewater treatment facility: a case of Mwanza City Slaughterhouse, Tanzania

Meat industry generates huge volumes of wastewaters that come from cleaning of slaughterhouse facilities, meat processing, and cleaning of animal carcasses (Bustillo-Lecompte & Mehrab 2015). The volume of these wastewaters being released into the receiving environment has also increased over the years due to increased meat production to meet protein requirements of growing human populations (Emmanuel et al. 2016). Slaughterhouse wastewater contains biodegradable suspensions, colloidal particles, organic matter, fats, and cellulose that usually contribute to elevated levels of chemical oxygen demand (COD) and biochemical oxygen demand (BOD) (Shujun et al. 2015). These materials can eventually reduce the amount of dissolved oxygen (DO) in the receiving aquatic environments (Sunder & Satyanarayan 2013). Thus, slaughterhouse wastewater requires considerable treatment to eliminate environmental contaminants before it is discharged into the receiving aquatic environments (Irshad et al. 2015).

The present study dealt with wastewater treatment processes at the newly installed Mwanza City Slaughterhouse (MCS) wastewater treatment facility in Tanzania. Before the MCS wastewater treatment facility was installed, the effluents used to be released untreated to the nearby receiving waters that empty into Lake Victoria. The Mwanza City Council (MCC) strove to find funds to resolve pollution problems related to industrial effluents that fed into Lake Victoria. The World Bank (WB) funded the construction of the slaughterhouse wastewater treatment facility through the Lake Victoria Environmental Management Project (LVEMP).

The LVEMP then contracted the Nelson Mandela African Institution of Science and Technology (NM-AIST) to design and supervise the construction of a slaughterhouse wastewater treatment facility for MCS. The design flow rate of wastewater into the facility was 130 m³ per day. Other design parameters for influent quality of the MCS wastewater treatment facility, with their values indicated in parentheses, were as follows: pH (7.5), wastewater colour (10750 Pt-Co), total suspended solids (TSS) (9,700 mg/L), BOD₅ (1,200 mg/L), COD (4,500 mg/L), NH₃-N (65 mg/L), (370 mg/L), and faecal coliforms (2 × 10⁷ CFU/100 mL). The mean initial two-month performance after construction of the MCS wastewater treatment facility revealed that effluent contaminant levels were above the TBS regulation (Table 1).

The MCS wastewater treatment facility was designed with several units such as pretreatment unit (screening, oil and fat/grease trap and buffer tank), biodigester unit (stirred batch upflow anaerobic sludge blanket (UASB)), advanced treatment (aeration tank and clarifier) and a polishing step (constructed wetland). Also, the facility has subcomponents such as a biogas holder and a sludge drying bed. This makes it one of the novel systems that has, so far, not been extensively studied. During the study, the MCS wastewater treatment facility was energy positive because the daily energy consumption ranged between 50 and 65 kWh while the daily biogas production ranged from 220 to 250 kWh, if converted into electricity. However, at the time of the study, the biogas produced was not used for power generation because the utilities for power production were yet to be procured and installed. Furthermore, the initial production of biogas from the MCS wastewater treatment facility was below the estimated potential of 200 m³ per day. Therefore, the aim of the present study was to investigate the performance of the MCS wastewater treatment facility by taking into account the factors that affect operational efficiencies.

The MCS wastewater treatment facility is located along Musoma Road, Mahina Ward in Mwanza, Tanzania (Figure 1). The location lies below the equator between latitudes 2° and 4° south and longitude between 32° and 35° east of Greenwich. The city of Mwanza is located on the southern shores of Lake Victoria and has a population of approximately four million according to the national census of 2012.

The system was designed to have the following parameters: feed flow rate, Q = 65 m³/h; the dimensions of the batch stirred biodigester were: diameter of 10 m, total height of 8.50 m with water level height of 6.62 m, a gas collector height of 1.38 m and a free body space of 0.50 m. The biodigester was designed to degrade 4,500 mg/L of influent COD, BOD₅ of 1,200 mg/L, TSS of 9,700 mg/L and of 605 mg/L. The designed hydraulic detention time, τ, was 4 days. The aeration tank has of diameter of 7 m, and height of 6.5 m with water level 6.0 m, aeration volume of 231 m³ and designed hydraulic detention time, τ, of 1.8 days.

The MCS wastewater treatment facility was operated as a semi-batch system, providing an intermittent flow of wastewater into the different units of the plant up to the constructed wetland (Figure 2). The facility was designed to match its operation hours taking into account the fact that slaughter activities normally happen between 3 and 6 AM. The generated wastewater was quickly transferred into the biodigester to minimize biomethanation. Pumping into the biodigester was usually done for 1–2 h. The biodigester was fed from the bottom up. From the biodigester, the wastewater was transferred by gravity to be further treated in the aerated tank. Aeration was done for 12 h, then stopped to allow development of anoxic conditions for another 12 h. This mode of operation assisted in the denitrification of produced in the aerator during the nitrification of ammonium. The aeration tank was designed to receive 85 mg/L of ⁠. Accordingly, the blower was designed to supply 48 kg/h of air for 12 h during the nitrification process.

After the aeration unit, there was a retention tank and a pump that continuously fed the rest of the units. During fieldwork for the present study, the MCS wastewater treatment facility was receiving an average amount of wastewater of 32.7 m³ per day from carcass and meat washing as well as the slaughterhouse floor cleaning due to the small number of animals slaughtered per day. Animals slaughtered at the MCS facility included goats, cattle, and sheep. In the course of the present study, the wastewater samples collected were found to contain high amounts of large solids from undigested offal. The solids were removed from the system using coarse and fine bar screens of 20 mm and 7.5 mm spacing, respectively, at the preliminary treatment stage. The wastewater from the aeration tank was then transferred to the retention tank before being continuously pumped into the clarifier. The clarifier's role was to separate the solids from the aeration tank from the wastewater before the water entered the polishing step. The sludge that accumulated in the clarifier was transferred into the sludge drying beds; later it was used as organic fertilizer for agricultural purposes. The dimensions of the sludge drying bed were: Area of 42 m², four (4) compartments each with a length of 3.75 m, width of 3.75 m (which made a total length of 16.5 m and width of 3.75 m), and height of 1.2 m. Inlet diameter channel was 100 mm, terminating at 300 mm above the sand surface.

Effluents from the clarifier were conveyed by gravity to the constructed wetland (CW), which was used as a polishing unit due to its ability to remove the remaining nutrients, organic matter, and suspended materials from the wastewater. The constructed wetland was divided into two cells each with dimensions of 30 m length, 10 m width and 1 m depth. The wetland effective treatment depth was 0.5 m and granite gravel packing was of 20 mm size with porosity of 0.4. The daily influent to each of the constructed wetland cells was around 16 m³. During the course of the present study, the constructed wetland was observed to have a retention time of around 3.3 days.

Onsite measurements for EC, pH, TDS, temperature, and DO were carried out using a multiparameter probe (Palintest MACRO 900). Wastewater turbidity was measured using a turbidimeter (Palintest 09011150103). Wastewater and biogas volumes were recorded daily using a mass flow totalizer (GFT-110A). Biogas production was recorded daily, whereas biogas composition was analyzed weekly. Biogas composition was determined using a gas analyzer (Geotech BIOGAS 5000). The facility's power consumption was determined using an electrical meter (EDMI EUPR-1232-1100) and a sub-meter (EM 0026-JC).

The daily amount of slaughterhouse wastewater produced at the MCS facility required treatment before discharging into the aquatic environment. In the present study, an analysis of influent and effluent wastewater was carried out. Wastewater sampling and analysis were done for five consecutive months. A performance evaluation period was started after two months of the trial runs, where the plant performance observed was as shown in Table 1. In this period, a total of 112 samples were taken and analyzed.

Duplicate wastewater samples from the influent and effluent of each treatment unit were collected in 500 mL plastic bottles. After collection, the samples for COD, ⁠, and NH₃ analyses were acidified using sulfuric acid to a pH below 2 to deactivate microbial activities while another sample was not acidified, as they could be transferred to reach the NM-AIST laboratory within 24 hours while packed in ice-packed cool boxes and kept at a temperature below 4 °C.

The determination of ⁠, and NH₃ concentrations was done using a spectrophotometer (Hach DR-2800™). The analysis of NH₃ and used the Nessler method as detailed by Jeong et al. (2013). The analysis of was done using the Cadmium reduction method. The was determined using a HI-839800 Thermo-Reactor (HANNA Instruments). These environmental contamination indicators were analyzed as per the standard methods for examinations of water and wastewater (APHA 2012). The titration method was used to measure the soluble volatile fatty acids (VFAs) and alkalinity in the biodigester. The analysis of a five-day was determined through incubation (OxiTop^® IS12). The TSS was determined at a temperature of 105 °C in a drying oven (BINDER GmbH FD 56 E3). The total volatile solids (TVS) and volatile suspended solids (VSS) were quantified following standard methods at a temperature of 550 °C inside a muffle furnace (Cole-Parmer Stable Temp 1,100 °C Box Furnace: CBF Series). The weight of dry solid samples was determined using a weighing balance (CY 204 S/N 15201586). Filtration of the slaughterhouse wastewater was done using a filtration pump (WELCH 2546C-02B) combined with a conical flask (Pyrex 580913 PORO 3). The faecal coliform count was determined in triplicates where the Petri-dishes of MacConkey agar containing 0.1 mL sample on filters (11406 ø 47 AC 1502023 0.45) were inoculated and incubated at 44.5 °C for 24 h before counting.

The present study was carried out for a period of five months, in the months of February to June 2019. The daily amount of wastewater fed into the biodigester at the MCS averaged 32.7 m³. This wastewater volume, produced through animal slaughter and related activities, was a fraction of the design value of 130 m³ per day. The difference was attributable to the lower number of animals that was being slaughtered per day compared to the design capacity. The design capacity was for the facility to slaughter about 750 animals per day. During the course of the present study, only about 250 animals were being slaughtered per day. The effluent from the biodigester and clarifier was fed into the retention tank with a holding capacity of 130 m³ per day installed with a pump that continuously fed the clarifier and the constructed wetland at a rate of 5.42 m³/h.

Wastewater temperature, pH and DO were measured onsite (Table 2). The present study observed that the MCS wastewater treatment facility was operating at a temperature of around 26.3 ± 0.3 °C. This ambient temperature was lower than that recommended in other studies; that is, a mesophilic temperature ranging from 30 to 40 °C for essential enhancement of treatment efficiency as well as biogas production (Tsegaye et al. 2018). Thus, treatment efficiency of the MCS bioreactor can be significantly improved by raising this temperature either using solar heating (Kakaç & Pramuanjaroenkij 2016) or using part of the biogas generated to heat the incoming wastewater before entering the biodigester. At the time of the present study, these improvements were not possible. However, this has remained to be a recommendation for further improvement.

The present study also observed that the MCS wastewater treatment facility operated at a pH of approximately 7.2 ± 0.1 that was within the optimal range for the bioreactors and is usually controlled by the VFA-to-alkalinity ratio. A pH range of 6.5–8.0 is known not to inhibit methanogenic bacteria during biogas production (Reis et al. 2016).

The DO concentrations ranged from 0.29 to 3.82 mg/L (Table 2). DO values of 0.29, 0.24, 3.82 and 1.79 mg/L were recorded in the buffer tank, biodigester outlet, aeration tank and constructed wetland, respectively. In comparison, the anticipated design DO values were 0.21, 5.0 and 2.0 mg/L for the buffer tank, aeration tank and constructed wetland, respectively. The low DO in the aeration tank compared with the anticipated 5 mg/L was probably due to lower air supply from the blower of around 41 kg/h than the anticipated supply of air of 48 kg/h.

DO in anaerobic digesters may be caused by factors such as high mixing rates, high recirculation rate, and too much loss of activated sludge (Kato et al. 1997; Botheju & Bakke 2011). Conklin et al. (2007) studied the influence of DO on anaerobic digestion processes and found that a supply of 3–4 mg/L of DO led to 27% of active methanogenesis. These researchers concluded that a short-term oxygen exposure did not significantly reduce methanogen activity. However, a continuous oxygen exposure was found to negatively affect the methanogenic biomass activity. In spite of the negative influence of oxygen exposure, researchers found no effect on long-term digester performance in terms of the biogas production rate. Therefore, for the MCS facility, it is important to monitor DO loadings into the digester to improve the long-term methanogenic activity of the facility.

The effects of agitation time on the biodigester was investigated for zero to 6 h of agitation. The effects caused by both biodigester agitation duration and influent wastewater volume on the amount of biogas produced at the MCS wastewater treatment facility have been indicated (Table 3). Each of the agitation times was run once per day for 7 days and the biogas produced during that period was recorded daily. With no agitation in the system, and when the average volume of feed was 23 m³/day, the biogas produced was 145 m³ per day. With one hour of constant agitation and an average influent feed of 20 m³ per day, about 230 m³ of biogas was produced. Increasing the hours of agitation to two, at an average influent feed of about 19 m³ per day, continued to lower the volume of biogas produced. Similarly, when agitation time was increased to four hours at an influent volume of about 22 m³ per day, a dramatic reduction in biogas production was observed. A further increase in the number of agitation hours to six at an influent volume of about 22 m³, resulted in a further decrease in biogas production. Thus, the best biogas production occurred when the agitation time of 1 h was applied. It should be noted that before the start of the present study, the biodigester used to be agitated for 4 h per day. Agitation time of 1 h at a rate of 30 rpm was therefore recommended for improved biogas production. Agitation duration and speed have been linked to biogas production in a previous study (Aworanti et al. 2017). It has also been found by other researchers that a gentle biodigester agitation distributes uniformly the substrates to form a uniform suspension of solid and liquid parts, prevents foam formation and improves biogas production through fermentation processes (Lemmer et al. 2013).

The biodigester performance for the removal of important environmental pollutants has been indicated (Table 4). Transformation of N and N-compounds and related mass balance explanations are found in Equations (1)–(7) below.

In the present study, at pH value of 7.2, the NH₃ to ratio was about 0.5 (Table 4). This was consistent with a recent study that investigated ammonia levels in liquid phase during anaerobic digestion (Mutegoa et al. 2020). In wastewaters, ammonia exists, primarily, in two forms: the charged ammonium ion and the uncharged aqueous ammonia. This coexistence is highly pH- and temperature-dependent. The uncharged ammonia component is known to be more toxic than its charged counterpart because of its lipophilicity and ability to traverse biological membranes. At a pH range between 7 and 12, both the charged and uncharged species of ammonia are known to exist in wastewater at varying percentages (Caicedo et al. 2000; Körner et al. 2001; Philippe et al. 2011). Dissolved uncharged ammonia increases with increasing pH and temperature. At pH below 7, virtually, all ammonia is expected to exist as soluble ammonia gas. In the present study, at a pH of 7.2, the measured ammonia concentration was higher than expected and could be considered inhibitive. The cause for this high ammonia concentration is unknown. However, a study by Jeong et al. (2013) pointed out the deficiency of titrimetric methods in estimating the concentration of ammonia species in wastewater, especially when ‘hindering’ ions such as Mg, Cl, and Fe are present in high concentrations. It is possible that ammonia was overestimated in the present study due to the fact that, during the study, there were no apparent toxicity indications in the system as evidenced by the amount of biogas produced. To this point, a recommendation is thus made for a further study to examine the causes of the reported high concentration of ammonia.

Treatment processes in the biodigester were energy-positive, involving simple mechanisms shown in Equations (4)–(7). The expected design and actual performance of the biodigester were satisfactory (Table 4) because, for all parameters, the actual efficiency was lower than the design by an error margin of <15%. Despite these discrepancies in performance, UASB systems, such as the one that was investigated in the present study, are increasingly becoming a promising technology for treating slaughterhouse wastewaters with reported efficiencies ≥84, ≥77% and ≥81% for BOD₅, COD and TSS, respectively (Mittal 2006). This implies that the efficiency of the studied biodigester can be further improved.

In comparison to design, removal efficiencies in the aeration step were satisfactory (Table 5). With the exception of COD, removal efficiencies for each contaminant were either better than the anticipated design values or within a 10% error. The COD removal had a 20% error.

The two compartments of the CW were designed to treat 65 m³ of wastewater per day. In the present study, the CW was observed to treat 42 m³ of wastewater volume per day. The expected effluent quality levels were: COD (60 mg/L), BOD₅ (30 mg/L), TSS (100 mg/L) and (20 mg/L). The actual performance of the constructed wetland was excellent because the maximum error of the actual efficiencies was close to 2% compared to the design efficiencies (Table 6). Other studies on the slaughterhouse wastewater using CWs provided similar results (Vymazal 2010; Paschal et al. 2017). A study conducted in Uganda revealed that a CW could efficiently remove the following contaminants: COD (71%), BOD (71%) and (76%) (Odong et al. 2013). In the present study, the CW removed the measured contaminants with efficiencies >78% (Table 6).

The overall performance of the biodigester-constructed wetland system has been provided (Table 7). The present study shows that the integrated biodigester-CW system performed well in the removal of COD (98.3%), BOD₅ (97.4%), TSS (99.6%), (191.0%), faecal coliform (99.1%), (93.3%), NH₃ (99.1%) and turbidity (99.9%). Levels of NH₃ and were reduced at the aeration stage by the nitrification and denitrification processes, respectively. It could be that the good performance of the biodigester-CW system was due to the presence of intermediate units that were performing complementary treatment tasks. Previous research found that for soluble COD removal, good bioreactor operations were attributed to variations of solids settling in slaughterhouse wastewater (Manjunath et al. 2000). The MCS treatment system was designed to remove 99, 98, and 73% of COD, BOD₅ and ⁠, respectively. The combined biodigester-CW system of the MCS wastewater treatment facility was able to remove all contamination indicators with an efficiency of >97% and produced effluents quality that fell within the TBS limits (Table 7).

Furthermore, the MCS wastewater treatment facility was designed to produce a sludge volume of about 19,400 and about 5,700 m³ per year from the biodigester and aeration tanks, respectively. In the present study, the MCS facility was found to produce a sludge volume of 15,800 m³ and 5,300 m³ per year from the biodigester and aeration tanks, respectively. Sludge produced at the MCS facility is usually applied as fertilizer to boost plant production in nearby agricultural fields.

In the present study, the average biogas production at the MCS facility was 158.2 m³ per day. This high biogas production was probably caused by the presence of organic materials required for anaerobic bacteria as substrates for methanogenesis processes. The substrates present in the slaughterhouse wastewater are known to have adequate nutritional requirements for anaerobic bacteria to form new cells and act as energy sources (Anahita et al. 2019). Degradation of organic materials in the MCS biodigester to produce biogas can be attributed to the consortia of anaerobic bacteria under favourable conditions (Shah et al. 2017). Also, factors such as favourable pH, temperature, and VFA-to-alkalinity ratio are known to stimulate the anaerobic bacteria to digest the liquid and cellulosic material in the slaughterhouse wastewater during the fermentation process (Jain et al. 2015).

The electrical energy consumption for the MCS wastewater treatment facility was observed to range between 50 and 65 kWh per day. Energy-consuming activities included the feeding of slaughterhouse wastewater into the digester, agitation of the biodigester, running of the aeration system, clarifier feeding and facility lighting. The present study found that these activities consumed electricity up to 1,950 kWh per month. This amount of energy was sometimes too costly for the MCC to afford and failed the continuity of operations at the treatment facility. However, the amount of biogas produced per day by the MCS wastewater treatment facility, if converted to electricity, would be enough to power the facility (Figure 3). Uddin et al. (2016) reported that 2.5 kWh electrical energy can be generated from one cubic meter of biogas. Therefore, the daily biogas produced at MCS can satisfy the plant's power requirements if converted to electricity using a biogas-run generator.

The average biogas composition was as follows: CH₄ (70.3%), CO₂ (29.2%), O₂ (0.5%) and other gases of NH₃ (130 ppm) and H₂S (120 ppm) per day (Table 8). Normally, biogas composition is dependent on the feedstock type and the activity of the consortia of anaerobic bacteria involved in the digestion process. The usual biogas composition from anaerobic digestion of organic-rich substrate includes CH₄ (50–75%), CO₂ (25–45%), O₂ (0–2%), NH₃ (0–1%) and H₂S (0–1%) (Shah et al. 2017). Generally, the MCS wastewater treatment facility produced a high amount of biogas. However, the facility has more biogas production potential than it was producing during the course of the present study.

Slaughterhouse wastewater treatment is still a challenge not only to countries in sub-Saharan Africa but also in other developing countries across the world. To solve the energy issues at the study site, the present study is recommending that the MCC should invest in an energy conversion system to benefit from the biogas produced. The MCC should also consider the installation of a heating system, which utilizes the excess heat from the conversion of biogas to electricity, to heat the feedstock into the biodigester. This will result in better performance of the system and increased gas yield. Furthermore, the influent should be fed at a flow rate of 65 m³/h in the integrated biodigester-CW that may be changed to 5.24 m³/h to provide a continuous flow into the facility and maintain its operational processes throughout the day.

In the present study, the biodigester's gentle agitation at 1–2 h per day was found to yield maximum biogas of about 185–231 m³. This estimated amount of biogas would be enough to run the MCS wastewater treatment facility operations with surplus energy that can be supplied to nearby users at a relatively affordable cost.

The authors wish to acknowledge The WISE-Futures African Center of Excellence and The Nelson Mandela African Institute of Science and Technology for their support during the period of this study. Moreover, a grateful recognition goes to the BIOINNOVATE Project and Mwanza City Council administration and staff for all support granted to conduct the study using the MCS facility as a case.

All relevant data are included in the paper or its Supplementary Information.