Results of the start-up and maturation phases of a full-scale, high-rate anaerobic pond bioreactor (HRAPB)® plus improved facultative ponds (IFPs) to treat municipal wastewater are presented (CODt: 759 mg L−1, CODf: 219 mg L−1, S-SO42–: 102 mg L−1, and Cr+: 1,500 μgL−1). The start-up of the HRAPB® comprised, first, the application of a selective pressure increasing up-flow velocity rates. Second, batch stages between successive rates were allowed until 70% of the initial CODf was removed. The IFPs were left in batch and ended when in-pond Chlorophyll-a concentration reached 800 μgL−1. Subsequently, the system underwent gradual maturation and reached effluent concentrations of CODt: 223 mg L−1, CODf: 50 mg L−1, and Cr+: 60 μgL−1. The actual efficiency of the system compared with the expected design efficiency was lower given the characteristics of the influent wastewater biochemical oxygen demand/chemical oxygen demand ratios < 0.4, presence of Cr+ >1,000 μgL−1, and variations in both conductivity (500–4,500 μScm−1) and pH (6.5–10.5 units). Nonetheless, the system exhibited an adaptation state in less than 1.5 months and yielded an ST/SV ratio of 0.46, and specific methanogenic activity of 0.43 g-CH4-CODg−1SV−1d−1 for HRAPB®; the in-pond Chlorophyll-a was on average 1,200 μgL−1 in the IFPs, which demonstrated the robustness of these eco-technologies in tropical conditions.

INTRODUCTION

Wastewater pond technology (WPT) systems are eco-technologies in which wastewater undergoes a mutualistic treatment between algae and bacteria. In these systems, growth rates and microbial activity are positively affected by tropical environmental conditions, such as sunlight radiation intensity, temperature stability, wind velocity, and direction that improve performance in terms of organic matter and nutrient removal (Polprasert et al. 1983; Peña 2002; Mara 2003).

Recent technological developments and innovations have improved the hydrodynamics of these systems (e.g., mixing patterns, contact time between substrate and micro-organisms, and the separation of cellular retention time from hydraulic retention time), which increases the organic volumetric or surface loading rates, while achieving removal efficiencies greater than those of older conventional systems but with reduced area requirements (i.e., development of the high-rate anaerobic pond bioreactor (HRAPB)® and improved facultative ponds (IFPs)).

The HRAPB® combines the performance of an up-flow anaerobic sludge blanket with the simple construction of a conventional anaerobic pond. This bioreactor includes a mix chamber, where anaerobic digestion occurs, and a sedimentation zone for solids retention as shown in Figure A1 in the Appendix (Peña 2002), available online at http://www.iwaponline.com/wst/071/041.pdf. Conversely, the incorporation of baffles or dividing walls in a conventional facultative pond improves its hydrodynamic efficiency and reduces the influence of external factors, such as wind direction and velocity (e.g., active control of short circuits and dead zones) (Ashworth & Skinner 2011). Despite the constructive and operational simplicity of this biological system, its start-up phase is essential so as to ensure good operation and performance conditions.

The main objective of the start-up and maturation phases in any biological treatment or bioremediation system is to fine tune the ecology of the system so as to adapt it to the specific conditions of the substrate and the surrounding environment, with the goal of obtaining an effluent with the best possible quality standards (Guerrero et al. 2012).

Systems based on anaerobic treatment metabolism and processes (e.g., anaerobic lagoons, HRAPB®) require long stabilisation periods because the growth rate of those micro-organisms is lower compared with their aerobic counterparts. Thus, the use of an inoculum to speed up the processes is required (Fernandéz-Güelfo et al. 2010). A deficient start-up leads to biomass wash out from the reactor and hence reduces the resilience of the system to subsequent environmental changes (Van Hulle et al. 2005). The first improved full-scale WPT system using HRAPB® and IFP was built for the small-size city of El Cerrito (p.e. 51,000 inhabitants, in Valle del Cauca, Colombia; N 3°42′05″; W 76°19′44″; 980 m.a.s.l). Thus, this work is aimed at showing the implementation and evaluation of the start-up and maturation processes of the proposed system for municipal wastewater treatment in a tropical small-sized municipality.

METHODOLOGY

Study site

The experiment was performed in a full-scale WPT system in the city of El Cerrito (Colombia). The WPT system comprises two identical parallel lines (HRAPB® + IFP). The system was designed to treat 90 L s−1 or an equivalent population of 51,000 inhabitants. Leather tanning is one of the main economic activities at El Cerrito and it occurs through two big tanneries (i.e., capacities of a few thousand skins per day) down to several small ones (i.e., a few dozen skins per day). The regional environmental authority required all tanneries in town (whether discharging or not into the sewerage network) to realise a plan for pollution load reduction including cleaner production strategies within the production processes, which becomes central to the proper functioning of any biological system given the likely toxic loads of chromium (Cr6+) in the raw sewage. Under this scenario, the improved WPT system was designed in 2004 and built between 2004 and 2010.

Wastewater features

In 2012, when the start-up began, new meat-processing industries and tanneries were also identified and they affected the actual composition of wastewater when compared to its design conditions (Table 1). Concentration values for parameters such as the chemical oxygen demand (COD), biochemical oxygen demand (BOD), pH, and Cr6+ all increased.

Table 1

Comparison of wastewater features 2004 and 2012

Parameters Design (Year 2004) Year 2012 
Flow rate (L s−190a 36.5 ± 5.9 
COD (mg L−1530 759 ± 276 
BOD (mg L−1290 351 ± 99 
TSS (mg L−1260 228 ± 32 
Cr6+ (mg L−10.06 1.45 ± 0.8b 
pH 6.9–9.5 6.2–10.5 
Conductivity (μS cm−1ND 500–4,500 
Parameters Design (Year 2004) Year 2012 
Flow rate (L s−190a 36.5 ± 5.9 
COD (mg L−1530 759 ± 276 
BOD (mg L−1290 351 ± 99 
TSS (mg L−1260 228 ± 32 
Cr6+ (mg L−10.06 1.45 ± 0.8b 
pH 6.9–9.5 6.2–10.5 
Conductivity (μS cm−1ND 500–4,500 

aFlow rate projected to 2024.

bn = 3.

Description of the treatment units

The WPT system at El Cerrito consists of an inlet structure, pumping station, coarse and fine screening, two grit chambers, two HRAPBs®, two IFPs, a single sludge drying bed, and the operations building. Raw wastewater is pumped from the inlet chamber up to the head of the preliminary treatment line, then it flows via gravity towards the HRAPB®. Raw wastewater enters at the bottom of the HRAPB® and gets mixed with the sludge blanket, where microbiological degradation occurs, and biogas is produced and released. Gas collectors placed right at the top of the in-pond mixing chamber fetch the biogas that is sent from there to bio-filters for purification. Meanwhile, the HRAPB's® effluents converge in a collection chamber and are then re-diverted, half for each IFP, whose final effluent is finally discharged to the Sabaletas River. There is also a by-pass connecting the grit chamber with the collection chamber prior to the IFPs. Figure 1 shows the general layout of the WPT system.

Figure 1

General layout of the WPT system at El Cerrito: (A) pumping well; (B) grit chambers; (C) HRAPB® 1; (D) HRAPB® 2; (E) IFP 1; (F) IFP 2; (G) sludge drying beds; (H) operations unit.

Figure 1

General layout of the WPT system at El Cerrito: (A) pumping well; (B) grit chambers; (C) HRAPB® 1; (D) HRAPB® 2; (E) IFP 1; (F) IFP 2; (G) sludge drying beds; (H) operations unit.

The points at (i) HRAPB® 1 inlet; (s) HRAPB® 1 outlet; (1) IFP inlet, (2) one-third of the IFP; (3) two-thirds of the IFP; and (4) IFP outlet refer to the sampling points (SPs) for physico–chemical parameters. Table 2 shows the design and operational conditions for the HRAPBs® and IFPs.

Table 2

Design (year 2004) and average operational conditions (year 2012) for the HRAPBs® and IFPs

Parameters HRAPB® 1 (design) IFP (design)b HRAPB® 1 year 2012 IFP 1 year 2012 IFP 2 year 2012 
Flow rate (L s−145 45 36.5 18.2 18.2 
COD (kg d−12,060 87 2,407 222 289 
BOD (kg d−11,127 328 1,113 349 454 
TSS (kg d−11,011 – 723 – – 
COD/BOD 0.55 – 0.48 – – 
VOL (g BOD m3 d−1800 – 960 – – 
Chamber mixing vol. (m31,148 – 1,148 – – 
SOL (Kg BOD Ha−1 d−1– 331 – 352 458 
Area (Ha) – 0.99 – 0.99 0.99 
Parameters HRAPB® 1 (design) IFP (design)b HRAPB® 1 year 2012 IFP 1 year 2012 IFP 2 year 2012 
Flow rate (L s−145 45 36.5 18.2 18.2 
COD (kg d−12,060 87 2,407 222 289 
BOD (kg d−11,127 328 1,113 349 454 
TSS (kg d−11,011 – 723 – – 
COD/BOD 0.55 – 0.48 – – 
VOL (g BOD m3 d−1800 – 960 – – 
Chamber mixing vol. (m31,148 – 1,148 – – 
SOL (Kg BOD Ha−1 d−1– 331 – 352 458 
Area (Ha) – 0.99 – 0.99 0.99 

aHRAPB® 2 was not monitored due to building mistakes being fixed at the time.

bLength: 174 m; width: 58 m; depth: 1.5 m.

VOL: volumetric organic loading.

SOL: surface organic loading.

Sampling and analytical methods

Wastewater samples were taken at six points along the WPT system, as shown in Figure 1 (A) HRAPB® 1 inlet; (B) HRAPB® 1 outlet; (C) IFP 1 inlet; (D) IFP 2 inlet; (E) IFP 1 outlet; and (F) IFP 2 outlet. The following variables were measured at those points: pH, CODt, CODf, TSS, BODt, and BODf. In addition, for HRAPB® 1, volatile fatty acids (VFAs) and S-SO42– were measured at points (A) and (B), and within the mixing chamber. Samples for buffer index (BI) were taken at 0.5 m from the surface. The following parameters were measured within the IFP: dissolved oxygen (DO), pH, and Chlorophyll-a at two intermediate points and two depths (0.10–0.75 m). For Cr6+, grab samples were collected in the influent and effluent of the WPT system; this parameter was measured three times at the beginning, in the middle and at the end of the start-up process. All physico-chemical parameters were measured in accordance with Standard Methods (2005); Chlorophyll-a data were obtained through a portable laser-induced fluorescence device (AquaFluor® Turner Designs Inc., Sunnyvale, CA, USA).

Figure 2

Behaviour of COD in HRAPB® 1: (a) and (b) behaviour of the total and filtered COD during the batch phase; (c) increase of up-flow velocities; (d) behaviour of the total and filtered COD for different up-flow velocities applied. SP: sampling point.

Figure 2

Behaviour of COD in HRAPB® 1: (a) and (b) behaviour of the total and filtered COD during the batch phase; (c) increase of up-flow velocities; (d) behaviour of the total and filtered COD for different up-flow velocities applied. SP: sampling point.

Inoculum

The inoculum for the HRAPB® was taken from the municipal WPT system at Ginebra (the neighbouring municipality), which has been in operation for 20 years and consists of a conventional anaerobic pond plus facultative pond. The initial seed sludge (i.e., inoculum) characteristics were volatile solids (VS): 84,279 mg L−1, total solids (TS): 140,180 mg L−1, and specific methanogenic activity (SMA): 0.30 g-CH4-DQO g−1VS−1d−1. These figures confirmed the good quality of this anaerobic sludge as inoculum for the HRAPB® start-up. A total volume of 480 m3 of seed sludge was transported from the Ginebra's anaerobic pond, which was equivalent to 40% of the mix chamber volume in HRAPB® 1.

Methane production was calculated for the inoculum, and for the HRAPB® 1 start-up this was determined through direct measurement of biogas concentration in a fixed volume of the inoculum in accordance with the method proposed by Bhatta et al. (2007). The methane concentration was monitored using gas chromatography (Shimadzu GC 14, FID). A Carbowax column 20M (3%) was used, with H3PO4 (1%), and Supelco 80/100. The methane peaks were detected at approximately 90 seconds using Peak Simple 3.0 Software. The operation conditions of the chromatographer were as follows: H2 = 50 psi, air = 50 psi, carrier 1 = 400 psi, and carrier 2 = 60 psi.

Start-up phase

Once the bulk inoculum was added to the mixing chamber of HRAPB® 1, the selective pressure method was used, which consists in gradually increasing the up-flow velocity (Vup) in the mixing chamber. This promotes the washout of poor settling sludge from the mixing chamber (Rodriguez et al. 2001; Yu et al. 2001). Moreover, a fraction of inert remaining solids in the original inoculum was expelled from the mixing chamber. The Vup increase ended when the design Vup of the mixing chamber was reached (0.61 m h−1). Once the selective pressure phase was finished, new samples of the inoculum were taken to measure its VS, TS, and SMA values.

Once the selective pressure step ended, HRAPB® 1 was left in batch for 4 weeks until 60% of the CODt and 70% of the CODf were removed in a sample taken at 0.50 m depth. In addition, alkalinity (ALK) and pH were also measured every 3 days in order to monitor a likely souring of the sludge blanket.

The final start-up phase in continuous flow began with gradual increases of Vup from 0.04 to 0.46 m h−1, which was reached after 5 weeks (these Vup values required hydraulic retention time (HRT) values from 120 h to 10.4 h, respectively). Finally, HRAPB® 1 was left in continuous operation under its design conditions (i.e., flow and Vup) and was also monitored for 6 weeks. The following physico-chemical parameters were measured once per week: VFA, BI, pH, ALK, S-SO42−, and COD.

Meanwhile, for the start-up of IFPs 1 and 2, these ponds were filled with a mixture of rainwater plus untreated wastewater from the grit chamber effluent and effluent from HRAPB® 1 after its final start-up phase. The process of filling the ponds took approximately 3 weeks. Once both IFPs were completely full, the algae community was already assembled and established. At this point, Chlorophyll-a concentrations were greater than 1,200 μgL−1 at 0.10 m depth. Once the levels of Chlorophyll-a were established throughout the pond volume, these were fed continuously with the treated effluent from HRAPB® 1.

RESULTS AND DISCUSSION

Start-up and maturation of HRAPB® 1

The batch stage within the start-up of the HRAPB® exhibited great variability in CODt (512 mg L−1, SD = 68.7; CV = 0.13) and CODf values (207 mg L−1, SD = 60.3; CV = 0.29). By the end of this period, reductions of 63% and 72% for these two parameters were achieved, respectively. Consequently, average concentrations of 187 mg L−1 (SD = 18.2; CV = 0.09) and 55 mg L−1 (SD = 5.8; CV = 0.1) for CODt and CODf values (Figure 2(a) and 2(b)) were achieved. The final stage of the start-up in continuous flow started once the removal of CODf was higher than 60%.

The final stage of the start-up in continuous flow exhibited a 51% reduction in CODt (n = 10, SD = 21.7; CV = 0.43) and a 58% reduction in CODf values (n = 10, SD = 15.8; CV = 0.27) with average effluent concentrations of 390 mg L−1 and 100 mg L−1, respectively (see Figure 2(c) and 2(d) and Table 3).

Table 3

Performance of organic matter removal in HRAPB® 1

 Inlet Outlet % Removal 
  BOD5 CODt CODf BOD5 CODt CODf    
Statistic (mg L−1(mg L−1(mg L−1(mg L−1(mg L−1(mg L−1BOD5 CODt CODf 
n 13 12 14 13 13 12 
Maximum 489 1,295 292 260 608 158 55.3 67.5 79.1 
Minimum 253 426 111 150 263 61 38.7 0.0 10.2 
ŷ 351 759 219 183 381 102 46.8 45.0 54.1 
SD 99.0 276.6 56.0 46,0 95.7 34.4 7.3 21.3 18.7 
CV 0.28 0.36 0.26 0,25 0.25 0.34 0.16 0.47 0.35 
 Inlet Outlet % Removal 
  BOD5 CODt CODf BOD5 CODt CODf    
Statistic (mg L−1(mg L−1(mg L−1(mg L−1(mg L−1(mg L−1BOD5 CODt CODf 
n 13 12 14 13 13 12 
Maximum 489 1,295 292 260 608 158 55.3 67.5 79.1 
Minimum 253 426 111 150 263 61 38.7 0.0 10.2 
ŷ 351 759 219 183 381 102 46.8 45.0 54.1 
SD 99.0 276.6 56.0 46,0 95.7 34.4 7.3 21.3 18.7 
CV 0.28 0.36 0.26 0,25 0.25 0.34 0.16 0.47 0.35 

As shown in Figure 2(d), there was a reduction in the removal efficiencies of CODt and CODf between weeks 6 and 7, which may possibly be related to the fact that those weeks exhibited the highest CODt (1,295 mg L−1), S-SO42− (136 mg L−1), and VFA (216 mg L−1) concentrations during the start-up period. Meanwhile, a BOD/COD ratio of 0.37 was reached, which was below the recommended value for biological treatment. The BI confirmed the atypical functioning of the ecology in HRAPB® 1, since for anaerobic systems, the BI should be approximately 0.4 (Pérez & Torres 2008), whereas at this point during the start-up phase, the BI was 0.57. This value suggests a slight acidification of the system. However, and starting at this point, there was a gradual but small improvement, where BI decreased to 0.53 by the end of the start-up. This improvement was also corroborated with a VFA reduction in the effluent with a mean concentration of 50 mg L−1 (n = 8, SD = 23.8; CV = 0.47) compared to the influent concentration of 161 mg L−1 (n = 8, SD = 46.9; CV = 0.29).

Figure 3

Behaviour of chlorophyll-a and DO vs temperature (h = 0.10 m).

Figure 3

Behaviour of chlorophyll-a and DO vs temperature (h = 0.10 m).

In contrast, the SMA underwent an improvement, going from 0.07 g-CH4-COD g−1 SV−1 d−1 after the selective pressure period, up to 0.43 g-CH4-COD g−1 SV−1 d−1 by the end of the start-up (Table A1 in the Appendix, available online at http://www.iwaponline.com/wst/071/041.pdf). This value of SMA is within the range currently reported in the literature for well functioning high-rate anaerobic reactors treating sewage. Comparing the data obtained during the batch and maturation stages, it may be argued that the removal percentages obtained during the latter phase were within the expected range (difference < 14%), given that HRT in the batch mode was 26 days, whereas for the continuous operation phase, it was just 0.6 days (i.e., 14.4 h).

The sulphate concentration within HRAPB® 1 was reduced to 46% (69 mg L−1), which is most likely associated with the presence of sulphate-reducing bacteria thriving on short-chain VFAs. Despite this bacterial group competing efficiently for substrate (i.e., acetate) with methanogenic Archaea, they are not fully antagonistic and may coexist in anaerobic environments, as demonstrated by the increase in SMA values. Otherwise, hydrogen sulphide production would have overtaken methane generation.

Despite the fact that HRAPB® 1 was designed to remove more than 60% of CODt (provided that tanneries were required to fulfil the cleaner production programme goals), the quality of the effluent showed otherwise since this exhibited higher conductivity and pH levels, COD, and BOD5 levels above the design concentrations, and the presence of Cr6+ (1.5 mg L−1) and sulphate concentration above 100 mg L−1. It remains to be seen if the ecology of the system adapted to these conditions that included eco-toxicological agents such as Cr6+. Another factor that may have affected the full operation and removal efficiency of the WPT system was the absence of hydraulic control systems (e.g., pumping automation, calibrated weirs, and proper controlled flow measurement devices). The latter issue led to HRAPB® 2 being left idle during this research.

Start-up and maturation of IFPs 1 and 2

The start-up process of the IFPs during the batch phase was quick in terms of Chlorophyll-a concentration or algal growth. Once IFP 1 and 2 were completely full, the IFPs already exhibited Chlorophyll-a concentrations greater than 800 μg L−1 and immediately went into the maturation phase with a flow rate of 16 L s−1 per pond, reaching outlet concentrations of CODt and CODf of 216 mg L−1 (n = 13; CV = 0.25) and 67 mg L−1 (n = 13; CV = 0.35) for IFP 1, and 230 mg L−1 (n = 9; CV = 0.22) and 35 mg L−1 (n = 9; CV = 0.62) for IFP 2, respectively (Table 4).

Table 4

Performance of IFP 1 and IFP 2 for organic matter removal

  IFP 1 inlet IFP 1 outlet % Removal of IFP 1 
Statistics BOD5 (mg L−1CODt (mg L−1CODf (mg L−1BOD5 (mg L−1CODt (mg L−1CODf (mg L−1BOD5 CODt CODf 
n 13 13 13 13 13 13 
Maximum 285 950 266 130 337 124 68.0 78.1 85.0 
Minimum 170 314 71 78 155 39 23.5 20.0 14.1 
Ŷ 222 526 156 102 216 67 51.6 55.6 50.7 
CV 0.21 0.33 0.39 0.20 0.25 0.35 0.32 0.29 0.45 
  IFP 2 inlet IFP 2 outlet % Removal of IFP 2 
Statistics BOD5 (mg L−1CODt (mg L−1CODf (mg L−1OD5 (mg L−1CODt (mg L−1CODf (mg L−1BOD5 CODt CODf 
na 
Maximum 341 1,268 326 120 334 63 75.8 74.7 93.0 
Minimum 227 358 147 72 154 52.5 25.1 62.6 
Ŷ 289 618 200 100 230 35 64.4 57.8 79.7 
CV 0.20 0.45 0.27 0.25 0.22 0.62 0.18 0.31 0.14 
  IFP 1 inlet IFP 1 outlet % Removal of IFP 1 
Statistics BOD5 (mg L−1CODt (mg L−1CODf (mg L−1BOD5 (mg L−1CODt (mg L−1CODf (mg L−1BOD5 CODt CODf 
n 13 13 13 13 13 13 
Maximum 285 950 266 130 337 124 68.0 78.1 85.0 
Minimum 170 314 71 78 155 39 23.5 20.0 14.1 
Ŷ 222 526 156 102 216 67 51.6 55.6 50.7 
CV 0.21 0.33 0.39 0.20 0.25 0.35 0.32 0.29 0.45 
  IFP 2 inlet IFP 2 outlet % Removal of IFP 2 
Statistics BOD5 (mg L−1CODt (mg L−1CODf (mg L−1OD5 (mg L−1CODt (mg L−1CODf (mg L−1BOD5 CODt CODf 
na 
Maximum 341 1,268 326 120 334 63 75.8 74.7 93.0 
Minimum 227 358 147 72 154 52.5 25.1 62.6 
Ŷ 289 618 200 100 230 35 64.4 57.8 79.7 
CV 0.20 0.45 0.27 0.25 0.22 0.62 0.18 0.31 0.14 

aThe amount of data for IFP 2 was lower than that for IFP 1 due to the time required to repair a leak in the pipe that connected the flow distribution chamber with IFP 2.

The higher efficiencies in organic matter removal in IFP 2 compared with IFP 1 were probably due to IFP 1 being operated under organic overloading in the first 4 weeks, while a leak in IFP was repaired; this situation affected the global statistics of the results.

The Chlorophyll-a concentration in the IFPs varied along the SPs; however, the Chlorophyll-a concentrations were within the range recommended by Mara (2003), between 500 and 2,000 μg L−1, this in turn insured an increase in DO at the surface, with values exceeding 12 mg L−1 at noon. This was favoured by environmental variables, such as high solar radiation, photoperiods close to 12 hours per day, and water temperatures greater than 25 °C. Figure 3 shows the Chlorophyll-a concentration at the surface, which exhibited a maximum value between 27 and 28 °C. However, as water temperature increased with high solar radiation, the Chlorophyll-a concentration at the surface decreased. This might have been the ecological response to likely photo-inhibition in the photic zone of the IFPs, and motile algae could have moved down the water column. Nevertheless, photosynthetic activity was maintained and even increased, as indicated by the high DO concentrations (Aponte 2013).

Global performance of the system

The total removal of the system in terms of organic matter was around 70% ± 7% for CODt and 77% ± 5% for CODf (see Table A2 in the Appendix, available online at http://www.iwaponline.com/wst/071/041.pdf). CODf removal is important for verifying the dissolved content readily biodegradable by micro-organisms, compared with CODt that can underestimate the real removal of the WPT system since there are always algal TSS present in the effluent. Similarly, there was a remarkable reduction of Cr6+ from 1,500 down to 550 μg L−1. However, total Cr concentrations of 3,000 mg kg−1 VS were measured in a grab sample of the sludge taken from HRAPB® 1. This shows that likely precipitation of metallic oxides under anaerobic conditions plays a role in the removal of this eco-toxicological element in the sludge blanket, but at the same time, it warns against the further use of dry sludge after withdrawal from HRAPB® 1. Behaviour of raw wastewater at the El Cerrito WPT system (i.e., Cr6+ concentrations, pH and conductivity variations, sulphate, COD, and BOD5 concentrations above initial design values) most likely affected the ecology of the WPT system and the quality of by-products (i.e., bio-solids with heavy metals, reduction in methane production, and eventual reduction of removal rates in the effluent).

CONCLUSION

This new and improved WPT technology showed its versatility and adaptation to treating this type of municipal wastewater under tropical environmental conditions. The selection of an inoculum that could adapt to high-load domestic wastewater enabled the start-up process of HRAPB® 1 by the application of selective pressure, and for the IFP by means of a batch filling-rest process. This whole strategy established the ecology and stabilised the system in a short period of only 6 weeks.

COD removals reached 50% for HRAPB® 1 and 70% for the entire system. During the start-up phase, Cr6+ concentrations (1,500 μg L−1) and S-SO42− concentrations (102 mg L−1) did not impair the development of methanogenic Archaea, thus obtaining an SMA average rate of 0.40 g-CH4-DQO g−1VS−1d−1. The same trend was observed with the Chlorophyll-a concentration in the IFP (1,200 μg L−1). Last but not least, the bio-solids generated in HRAPB®1 exhibited high total Cr concentrations that should warn on the potential use of this material, and thus it should be handled and disposed of in safer ways.

REFERENCES

REFERENCES
Aponte
A.
2013
Innovación mediante modelación hidrodinámica y ecológica de lagunas facultativas secundarias para la reducción de la contaminación hídrica en zonas tropicales [Innovation through hydrodynamic and ecological modeling of secondary facultative ponds to reduce water pollution in the tropics]
.
PhD thesis in Sanitary & Environmental Engineering
,
Cinara Institute, Universidad del Valle
,
Cali
,
Colombia
.
Ashworth
J.
Skinner
M.
2011
Waste Stabilization Pond Design Manual
.
Power Water Corporation
,
Australia
.
Bhatta
R.
Tajima
K.
Takusari
N.
Higuchi
K.
Enishi
O.
Kurihara
M.
2007
Comparison of in vivo and in vitro techniques for methane production from ruminant diets
.
Asian Australian Journal of Animal Sciences
20
(
7
),
1049
1056
.
Fernandéz-Güelfo
L. A.
Álvarez-Gallego
C.
Sales Márquez
D.
Romero García
L. I.
2010
Start-up of thermophilic–dry anaerobic digestion of OFMSW using adapted modified SEBAC inoculum
.
Bioresource Technology
101
(
23
),
9031
9039
.
Mara
D. D.
2003
Domestic Wastewater Treatment in Developing Countries
.
Earthscan
,
London
,
England
.
Peña-Varón
M. R.
2002
Advanced Primary Treatment of Domestic Wastewater in Tropical Countries: Development of High-Rate Anaerobic Ponds
.
PhD thesis
,
School of Civil Engineering, University of Leeds
,
UK
.
Pérez
A.
Torres
P.
2008
Índices de alcalinidad para el control del tratamiento anaerobio de aguas residuales de fácilmente acidificables [Alkalinity indices for control of anaerobic treatment of readily acidifiable wastewaters]
.
Revista Ingeniería y Competitividad, Universidad del Valle
10
(
2
),
41
52
.
Polprasert
C.
Dissanayake
M. G.
Thanh
N. C.
1983
Bacteria die-off kinetics in waste stabilization ponds
.
Journal of
Water Pollution Control Federation
55
(
1983
),
285
296
.
Rodriguez
J. A.
Peña
M. R.
Manzi
V.
2001
Application of an innovative methodology to improve the starting-up of UASB reactors treating domestic sewage
.
Water Science Technology
44
(
4
),
295
303
.
Standard Methods for the Examination of Water and Wastewater
2005
American Public Health Association/American Water Works Association/Water Environment Federation
, 21st edn,
Washington, DC
,
USA
.
Van Hulle
S.
Van Den Broeck
S.
Maertens
J.
Villez
K.
Donckels
B.
Schelstraete
G.
Volcke
E.
Vanrolleghem
P.
2005
Construction, start-up and operation of a continuously aerated laboratory-scale SHARON reactor in view of coupling with an Anammox reactor
.
Water SA
31
(
3
),
327
334
.
Yu
H. Q.
Tay
J. H.
Fang
H. P.
2001
The roles of calcium in sludge granulation during UASB reactor start-up
.
Water Research
35
(
4
),
1052
1060
.

Supplementary data