Methane emissions from stabilization ponds for municipal wastewater treatment in Mexico

Abstract

Wastewater treatment (WWT) is applied for environmental protection and water reclamation. However, this activity has been identified as a source of methane (CH₄), contributing to climate change. WWT (municipal and industrial) is estimated to produce 8 to 11% to overall CH₄ emissions. In order to apply effective mitigation strategies in the water sector, a more precise inventory of CH₄ emission should be accomplished. The application of the Tier 1 methodology in the Guidelines for National Greenhouse Gas Inventories of the Intergovernmental Panel on Climate Change (IPCC) results in rough estimations, as default emission factors are used. An effort should be done for determining actual emission factors for the more representative treatment processes in a given region or country. In this work, a detailed inventory of CH₄ emissions form municipal wastewater management in Mexico was obtained, based on the Tier 1 IPCC methodology. In addition, on-site CH₄ emission measurements in five stabilization ponds (SP) were realized. The total CH₄ emissions generated by municipal WWT in Mexico were 600.4 Gigagrams (Gg) for year 2010. Also, the IPCC (theoretical) estimations showed that CH₄ emissions were overestimated if compared with the results obtained in the five sampled facilities. The on-site emission factors obtained showed wide variation as they are specific to each sampled system and to their particular environmental and operating conditions. In Mexico, the value of 0.431 kg CH₄/kg BOD removed (0.110 kg CH₄/m³ treated water) may be used for SP with good operational practices.

Keywords:

• Emission factors
• methane emissions
• stabilization ponds
• wastewater treatment

1. Introduction

Emission of greenhouse gases (GHG) is one of the major worldwide environmental challenges nowadays, due to its potential consequences on climate change. Therefore, the implementation of reliable strategies in order to mitigate those emissions or to adapt to the predicted changes are of great importance. Wastewater management has been identified as an important source of methane (CH₄) (El-Fadel & Massoud 2001) being part of the anthropogenic methane production from wastes. It is estimated that 8 to 11% of global methane comes from wastewater treatment (WWT) (Fayez & Al-ghazzawi 2011). Methane, a short-lived climate forcer, has a significant overall contribution to global warming considering that over a 100-year time frame, it is 34 times more effective at trapping heat in the atmosphere than carbon dioxide (Myhre et al. 2013). As a result, a significant reduction in CH₄ emissions is a high priority to counteract the effects of climate change in the coming years (UNEP 2011).

However, in spite of the importance of assessing actual CH₄ emissions from WWT, there are uncertainties in GHG inventories, mainly due to the lack of reliable information. The imprecision results from the volume of wastewater that is treated and not treated, the operating conditions of the facilities and the methodologies for estimation of methane emissions based on default emission factors. When these factors are used, local conditions, such as treatment process and environmental data regarding wastewater and sanitation coverage are not properly considered (Flores-Alsina et al. 2011). Thus, it is necessary to assess CH₄ emissions from actual, local factors as well as to establish a correlation between emission levels and the operating conditions of the WWT facilities in a given region, considering their specific environmental conditions.

The accepted methodology for estimating methane emissions is based on the emission factors proposed by the Intergovernmental Panel on Climate Change (IPCC 2006). However, the lack of in situ measurements of methane emissions is one of the main weaknesses in the implementation of GHG emission inventories (Wang et al. 2011). Therefore, due to the lack of quantitative on-site data of CH₄ emissions from wastewater treatment plants (WWTP) and the uncertainty and variability of CH₄ emission estimation using default factors (Bousquet et al. 2006), a systematic measurement by direct methods is required. In order to do that, actual emission factors based on the conditions of specific treatment facilities must be determined. This would provide real data for the implementation of effective mitigation strategies and policies regarding GHG emissions from the water sector.

México and Latin America have adopted several treatment processes for sewage management. Based on the number of facilities, stabilization ponds (SP) are the most applied WWT process in the region (Noyola et al. 2012). SP comprise an arrangement in series, basically anaerobic, facultative and maturation ponds, or just two of these elements. SP technology consists largely of the interactions of bacteria and algae in suspension, highly dependent on environmental conditions such as temperature, wind speed and light intensity (Yánez 1993). This technology has many advantages for developing countries, such as simple operation and very low energy requirements, resulting in low operational costs (Konnerup et al. 2009).

Anaerobic ponds (AP) are efficient at removing organic carbon from wastewater. The use of an in-front AP in a SP system allows a reduction in the required surface, without energy need. AP converts organic carbon principally into methane and carbon dioxide (Picot et al. 2003). In AP, mainly in the settled sludge, anaerobic microorganisms break down organic matter and produce different metabolic compounds, methane among them. AP are commonly 2.5 to 5 m deep, and are the smallest and the deepest units of the pond system with a hydraulic retention time between 3 and 6 days.

Facultative ponds (FP) are the second step of a SP system. However, they can be used also as primary pond, follow by a maturation pond (two ponds in series). The hydraulic retention time of these units should be within a range of 5–30 days, with depth between 1 and 2.5 m (Yánez 1993).

SP are considered by the US Environmental Protection Agency as a source of CH₄ emissions (Todd et al. 2011); anaerobic pons are responsible for most of the GHG production in a pond system, with more than 90% of its total emission (Hernandez-Paniagua et al. 2013). Considering their relative importance for WWT management in developing countries, further research on the factors that influence methane emissions from SP is required.

Mitigation of CH₄ emissions has been an important environmental issue in Mexico, under its Federal Law on Climate Change and the national commitments (30% GHG emission reduction for year 2020 and 50% for year 2050, with the year 2000 baseline). The aim of this study was to develop an inventory of CH₄ emissions for the municipal WWT sector in Mexico, including theoretical and field measurements on CH₄ production in the SP process. The study involved five SP facilities, which were chosen in order to cover representative plant capacity, geographical distribution and the variability of the environmental conditions of each region.

2. Methodology

A methane inventory for the Mexican municipal WWT facilities was calculated using official data from municipal treatment plants and theoretical process data. Field measurements were done in five municipal WWTP (SP) in order to obtain methane emission factors and estimation of the annual production.

2.1. Calculation of theoretical methane emissions from municipal WWTP in Mexico

CH₄ emissions for the country were estimated from the 2010 baseline year, using the IPCC Guidelines for national inventories of GHG (Tier 1 method, IPCC 2006). This methodology quantifies CH₄ emissions for national inventories, using default values for the emission factors, which is considered a good practice in countries with limited data. The procedure was applied for each of the 2186 municipal WWTP reported for that year in the National Inventory of Municipal Water and Wastewater Treatment Plants (CONAGUA 2011). Table 1 shows specific parameters used for calculating CH₄ emissions generated by municipal treatment facilities in Mexico; organic matter is reported as biochemical oxygen demand (BOD).

Table 1. Parameters used to calculate theoretical CH₄ emissions from sewage treatment in Mexico (Year 2010).

Parameter	Value	Units
Population^a	112,336,538	Person
Wastewater generated	235	m^3/s
Collected wastewater^b	208	m^3/s
Treated wastewater	94	m^3/s
% of treated flow^c	45.1	%
Generated BOD^d	2,460,170,182	kg BOD/year
BOD that enters WWTP^e	709,998,981	kg BOD/year
BOD removed by WWTP	606,001,912	kg BOD/year
Removal efficiency	85.35	%

^a CONAPO 2007. http://www.conapo.gob.mx/

^b Per-capita wastewater generation 180 L/inhab/day (CONAGUA 2010).

^c Percentage of collected wastewater in municipal sewage systems being treated.

^d Per-capita generation BOD of 60 g/inhab/day (CONAGUA 2010).

^e [BODin] 242 mg/L obtained from a representative sample of 124 municipal WWTP.

2.2. SP evaluated to determine experimental CH4 emissions

Field measurements were realized in five municipal SP in Mexico: Torreón (TOR), Los Mochis (MOC), Irapuato (IRA), Coatzacoalcos (COA) y Comitán (COM). The field sampling was carried out during the dry season (February to May 2014). It is important to note that ambient temperature is one of the main factors influencing methane production, thus, Mexico was divided into three regions (north, central and south) in order to take into account these temperature variations. The selection of the SP considered in this study was based on regional distribution and representative treatment (flow) capacity. The location of these treatment facilities is presented in Figure 1.

Figure 1. Geographical location of the evaluated stabilization ponds.

System configuration and sampling points are shown in Figure 2. TOR and MOC consist of two ponds in series (anaerobic and facultative), COA and COM with a three ponds in series (the last unit, a maturation pond, was not sampled) and IRA with just an anaerobic step. The TOR facility had two different pond systems working in parallel. The whole set of sampling points were as follows: 32 (TOR); 50 (MOC); 17 (COM); 16 (COA) and 15 (IRA), depending on the size of each facility.

Figure 2. Configuration schemes of the evaluated stabilization ponds: (1) Torreón (TOR), (2) Los Mochis (MOC), (3) Comitán (COM), (4) Coatzacoalcos (COA) and (5) Irapuato (IRA). Anaerobic units are in gray, facultative units in white. Sampling points are represented by asterisks. Arrows indicate the flow direction.

Wastewater quality data were obtained directly from operational records, provided by the operation staff. Wastewater quality (BOD, chemical oxygen demand (COD), total suspended solids (TSS), nitrogen, phosphorus and pH), operational parameters, general information, characteristics and dimensions of the evaluated SP systems are shown in Tables 2 and 3.

Table 2. General information of evaluated stabilization ponds with wastewater quality parameters (average values).

	TOR		MOC		IRA		COA		COM
Location	25°30′43.40″N 103°20′9.78″O		25°43′30.79″N 109° 4′24.87″O		20°38′42.24″N 101°21′17.12″O		18° 6′55.77″N 94°29′0.61″O		16°14′26.42″N 92° 5′46.43″O
State	Coahuila1900		Sinaloa1100		Guanajuato700		Veracruz 340		Chiapas210
Installed capacity (L/s)
Treated flow (L/s)	1357 ± 128		1036 ± 191		775 ± 65		220 ± 60		160 ± 14
Years in operation	10		14		22		12		10
Final disposal/reclamation	Discharge into water bodies		River discharge		Agricultural irrigation		River discharge		Agricultural irrigation
Parameter	Influent	Effluent	Influent	Effluent	Influent	Effluent	Influent	Effluent	Influent	Effluent
BOD5 (mg/L)	344 ± 36	88 ± 7	61 ± 15	36 ± 17.5	241 ± 45	108 ± 15	61 ± 15	41 ± 15	278 ± 35	140 ± 39
COD (mg/L)	644 ± 55	270 ± 20	195 ± 48	145 ± 40	686 ± 230	273 ± 150	NA	NA	NA	NA
TSS (mg/L)	285 ± 63	89 ± 12	162 ± 92	70 ± 17	198 ± 72	75 ± 24	NA	168 ± 79	537 ± 40	40 ± 1.5
N Total (mg/L)	42 ± 2	39 ± 4	22 ± 4	17 ± 4.6	41 ± 15	31 ± 13	NA	NA	18 ± 3	NA
P Total (mg/L)	10 ± 3	8 ± 0.8	4.2 ± 0.8	4.3 ± 0.7	9 ± 4	7.7 ± 2.7	NA	NA	4.1 ± 0.5	NA
Temperature (°C)	32.4 ± 1.7	29 ± 3	27 ± 3	27.6 ± 3	30 ± 2	30 ± 2	31 ± 2	31 ± 2	24 ± 2	24 ± 2
pH	7.26 ± 0.05	8.3 ± 0.12	7.4 ± 0.2	8.1 ± 0.3	7.54 ± 0.16	7.34 ± 0.27	NA	7.0 ± 0.3	7.1 ± 0.0	7.7 ± 0.3
BOD removal (%)	75	40	55	33	50

NA: Not available

Table 3. Dimensions of the evaluated stabilization ponds.

WWTP	Anaerobic ponds (AP)					Facultative ponds (FP)
	Nr of ponds	Retention time (day)	Depth (m)	Area (m^2)	Volume (m^3)	Nr of ponds	Retention time (day)	Depth (m)	Area (m^2)	Volume (m^3)
TOR	4	6	3	55,640	166,380	4	12	2	136,000	272,000
	2	6	3	38,080	114,240	2	12	2	104,000	208,000
MOC	8	5	2	79,500	159,000	12	12	1.5	53,530	80,295
IRA	1	4	2	123,930	247,860	–		–	–	–
COA	2	3	3.5	16,400	57,400	2	6	1.9	20,000	38,000
COM	3	4	4.5	8,625	38,813	3	7	1.5	52,900	79,350

2.2.1. Calculation of theoretical methane emissions from the five SPs

As mentioned, CH₄ emissions from the five systems were estimated using the Tier 1 IPCC Guidelines (IPCC 2006). Table 4 shows specific parameters used for calculating the theoretical CH₄ emissions from the five municipal WWTP evaluated. The default emission factors were obtained using a maximum CH₄ production capacity (B_o) of 0.6 kg CH₄/kg DBO removed (recommended default value). The methane correction factor (MCF) was either 0.8 or 0.5, depending on the characteristics of the sampled treatment facilities. The flow values used for the theoretical calculation correspond to those reported in the National Inventory of Municipal Water and Wastewater Treatment Plants (CONAGUA 2011). The influent BOD concentration was assumed to be the representative value obtained from 124 WWTP in Mexico.

Table 4. Values used to estimate theoretical methane emissions from the evaluated stabilization ponds.

Parameter	Unit	TOR	MOC	IRA	COA	COM
Treated flow	L/s	1550	900	700	250	160
Influent BOD^a	kg BOD /year	11,731,392	6,811,776	5,298,048	1,892,160	1,210,982
Removed BOD ^b	kg BOD /year	10,264,968	5,960,304	4,635,792	1,655,640	1,059,610
Theoretical removal	%	87.5	87.5	87.5	87.5	87.5
CH4 correction factor (MCF)^.^c	–	0.8	0.8	0.5	0.5	0.8

^a [BODin] 242 mg/L obtained from a representative sample of 124 municipal WWTP.

^b [BODeff] 30 mg/L according to Mexican discharge standards in NOM-001-SEMARNAT-1996 (SEMARNAT 1997).

^c The methane correction factors are the recommended default values in Tier 1 method (IPCC 2006). An MCF of 0.5 was chosen based on expert judgement.

2.2.2. On-site measurement of methane emissions from the five SP

Methane sampling in SP is a complex task, considering that emissions come from an extensive area and that methane fluxes are very sensitive to environment disturbances (Detto et al. 2011). The static chamber method was used for CH₄ measurements based on the considerations of previous research (Czepiel et al. 1993; Duchemin & Lucotte 1999; Hartman 2003; Schmitdt 2004; Yacob et al. 2006). This method considers the linear rate of gas accumulation in the chambers over time (Vincent et al. 2000). The method offers operational advantages as the equipment and procedures are simpler and less expensive, enabling the deployment of multiple chambers over the same time period. Also, the absence of inlet and outlet flowing gases minimizes potential disturbances of the natural flux conditions (Hartman 2003).

The design and construction of the static chamber was done according to the US EPA Flux Chamber Method EPA/600/8-8E/008 (Klenbusch 1986). Figure 3 shows a picture of the static chamber, consisting of a floating reservoir of clear plexiglass with a total volume of 38.3 L. The dome was covered with anti-reflective film, with the aim of reducing temperature variations in the chamber. Five chambers were assembled, each having the following dimensions at its base: inside diameter (0.4 m), height 0.178 m), surface (0.125 m²); and a dome height of 0.10 m).

Figure 3. A view of the static chamber used in this study (five similar units).

The quantification of in situ CH₄ emissions was conducted as follows:

(1)	The five chambers were repeatedly distributed along each pond in order to cover the largest possible surface (Figure 2), anchoring them to prevent wind drift during the whole sampling period for each facility (3 to 5 days, depending on the number of sampling points). A minimal use of two chambers is recommended for large-scale treatment (Baker et al. 2003).
(2)	The wall of the static chambers must dive to a minimum depth between 2 and 10 cm (Duchemin & Lucotte 1999). In order to take advantage of an increased chamber volume, a 5 cm depth was chosen, resulting in a working volume of 31.8 L.
(3)	Once the static chambers were placed at the defined sampling points, methane sampling and measuring were done at intervals of 30 min for two hours (T0 = start, T1 = 30 min, T2 = 60 min, T3 = 90 min and T4 = 120 min). This allowed the accumulation of gases inside the chamber so they could be quantified (Duchemin & Lucotte 1999; Deborde et al. 2010; Silva-Vinasco & Valverde-Solís 2011).
(4)	The methane gas concentrations was determined with a portable biogas analyzer (BIOGAS 5000, Fonotest, Spain), connected to the static chamber by a sampling tube (Yacob et al. 2006).
(5)	The temperature at each measurement was registered using a thermocouple HI 92804C (Weishampel & Kolka 2008).

Methane fluxes were calculated by linear regression based on the change of the concentration versus time for each of the samples, using the following equation:(1)

Where:

Slope resulting from a linear regression analysis, considering the increase of concentration with respect to time (mg/m³hr)

Area = Surface area covered by the chamber (0.125 m²)

A determination coefficient (r²) higher than 0.85 was taken for accepting or discarding the experimental results.

3. Results and discussion

3.1. Theoretical methane emissions by WWTP in Mexico

In 2010, there were 2186 municipal WWTP in Mexico, with a total treated flow of 93.6 m³/s, representing 45% of collected municipal wastewater at national level (CONAGUA 2011). Using the Tier 1 methodology of the IPCC with this data and the estimations in Table 1, the municipal WWT in Mexico produced 600.4 Gigagrams (Gg) of CH₄.

It is worth noticing that the limited treatment coverage of Mexico (45% of collected sewage, 40% of total sewage) results in a high fraction of methane emissions coming from raw sewage that is discharge to the environment (71%). In such case, the mitigation possibilities are clear and focus on increasing the treatment capacity of the country, meeting at the same time the existing environmental regulations. This should be done based on sound decisions on selecting treatment technologies that may have lower environmental impact, meeting technical and economic criteria (Noyola et al. 2012).

3.2. Methane fluxes from the five SP

The methane fluxes of each of the selected facilities are shown in Table 5. As expected, AP in all cases produced the largest amount of methane, if compared with the following FP.

Table 5. CH₄ fluxes from the five stabilization pond systems.

WWTP	Anaerobic ponds		Facultative ponds
	Nr. of ponds	CH4 flux	Nr. of ponds	CH4 flux
		(mg CH4/m^2 h)		(mg CH4/m^2 h)
TOR	4	1,157 ± 430	4	127 ± 21
	2	2,440 ± 1,018	2	156 ± 29
MOC	8	235 ± 62	12	139 ± 19
IRA	1	366 ± 104	–	–
COA	2	348 ± 113	2	124 ± 30
COM	3	878 ± 304	3	192 ± 27

The CH₄ fluxes ranged from 348 to 2,440 mg CH₄/m²h in AP and 124 to 192 in FP. As seen in previous studies, CH₄ fluxes showed spatial variation with the largest emissions in the AP and the lowest in the FP (data not shown), as reported by Toprak (1995), with values of 1,450 and 541 mg CH₄/m²h for AP and FP, respectively. The spatial variation of CH₄ emissions in a given facility is the result of the combined effects of physical, chemical and biochemical processes, influenced by wastewater variation (flow and concentration), operational practices and climatic conditions (Johansson et al. 2004; Kone et al. 2010).

Paing et al. (2000) and Picot et al. (2003) worked in an AP fed with municipal wastewater, finding a value of 2,035 and 2,970 mg CH₄/m² h, respectively; similar to that obtained in one of the subsystems in the TOR location (2,440 mg CH₄/m² h). COM system produced a similar CH₄ flux (878 mg CH₄/m² h) than the highest value reported by Czepiel et al. (1993), with 842 mg CH₄/m² h for AP.

Lower fluxes were measured in the AP of the MOC, IRA and COA systems (235, 366, 348 mg CH₄/m² h, respectively) but they are higher than the values reported by Parra et al. (2010) and Wang et al. (2011) with a maximum methane flow of 152 and 143 mg CH₄/m² h, respectively. The first authors reported a low operating temperature (12 to 13 °C) and the second ones a low concentration of influent COD (200 mg/L).

Table 6 presents the CH₄ fluxes obtained in this study and those published from other authors using similar systems. The data shows clear differences in the methane fluxes obtained at each evaluated facility. This behavior may be explained by the variation of the influent BOD (organic load) and the operating temperature, among other possible factors. Discarding the lowest value obtained by Parra et al. (2010) in Bolivia under low temperatures, the flux range is between 55 and 2,970 mg CH₄/m² h.

Table 6. Data from other studies determining CH₄ flux from municipal stabilization ponds.

Systems	Location	Flow treated (L/s)	COD (mg/L)	Temperature (°C)	CH4 flux (mg CH4/m^2 h)	Reference
Stabilization ponds	Mexico	160–1,357	195–686	24–32	124–2,440	This study
Stabilization ponds	Bolivia	317	1,336	13–14	5–152	Parra et al. (2010)
Stabilization ponds	Portugal	15.5	699	20	541–1,450	Toprak (1995)
Anaerobic ponds	England	46.3	–	10–20	175–842	Czepiel et al. (1993)
Anaerobic ponds	France	10.2	685	12–24	55–2,035	Paing et al. (2000)
Anaerobic ponds	China	3,472	200	12−24	89–143	Wang et al. (2011)
Anaerobic ponds	France	Population served 13,800	589	18	358–2,970	Picot et al. (2003)

Methane production in ponds depends primarily on the quantity of degradable organic material that enters the system (organic volumetric load), the temperature and the type of pond. A temperature increase will result in a higher CH₄ production rate (Gupta & Singh 2012). This is especially important in uncontrolled systems, such as SP systems, where ambient changes may result in wide variations in metabolic activity and methane production.

3.3. Methane emissions in terms of CO2 eq

Figure 4 shows the results obtained using the default IPCC emission factor (theoretical calculation based on Tier 1 method and the data in Table 4) and the values obtained experimentally based on on-site measurements (equivalent to the IPCC Tier 2 method). The data is presented as CO₂ equivalent using a Global Warming Potential (GWP) factor of 34, as reported in the fifth Evaluation Report of IPCC (Myhre et al. 2013).

Figure 4. Methane emissions in CO2 eq. for the five stabilization ponds evaluated, accordingly to IPCC methodology and on-site data.

The Tier 1 IPCC methodology appears to overestimate CH₄ emissions from the five sampled treatment facilities, a finding that is consistent with the results of Monteith et al. (2005) and El-Fadel and Massoud (2001). A reason for this result may be the assumption taken by the theoretical estimation, in the sense that under anaerobic conditions, the total organic fraction removed is converted to methane, based on a stoichiometric calculation. This approach considers an ideal behavior of the systems and it is a basis for developing methane emission inventories when no actual data on emission factors is available (Tier 1). As an ideal consideration, it does not take into account multiple factors that have great impacts on the system performance, such as: degree of decomposition (methane conversion), nutrient limitation, biological inhibition, physicochemical interactions, process operating conditions, mainly temperature, and the specific local urban-geographical context, among others (El-Fadel & Massoud 2001). In contrast, the use of actual emission factors based on in situ measurements may estimate the values of a specific system with good precision, but as a unique facility; its emissions will vary when compared to other similar systems even in the same location or region. The on-site data obtained in this study shows this behavior. In such case, a national inventory base on few actual emission factors will also result in a rough approximation to the real value. A comprehensive sampling campaign based on statistical significant sample is necessary in order to have country specific quality data, based on plant-specific emissions data from large or representative WWT facilities. This will enable to carry out emission inventories based on the Tier 3 method of the IPCC (2006).

Further into the differences shown in Figure 4, the BOD removal efficiencies should be considered in the discussion. In the Tier 1 IPCC methodology, a CH₄ emission factor is chosen as a default value for each treatment processes, as well as a general average influent BOD concentration for all the treatment facilities, in this case 242 mg/L, according to a representative value obtained from 142 WWTP in Mexico. Also, a common BOD removal efficiency is used for all facilities of the same technology (in this case, 87.5% for SP, Table 4). As expected, in this study, influent actual on-site values differ in a range of 61 to 344 mg BOD/L) and the removal efficiencies between 33 and 75% (Table 2). In the case of the MOC and COA systems, the difference between the theoretical yearly methane emissions and those measured on-site is mainly due to the low actual average influent BOD (61 mg/L) and the resulting low BOD removal efficiencies attained by those treatment facilities (40 and 33%, respectively). The IRA system, although with a similar influent BOD concentration as the national average, also resulted in lower experimental emissions, due to its limited BOD removal efficiency (55%).

On the other hand, the TOR and COM facilities showed similar theoretical and experimental emission values. In the first case, the influent BOD was higher (344 mg/L) than the one used in the theoretical calculation, but the removal efficiency was lower (75%), a combination that led to similar methane emissions. In the second case, the actual influent BOD was 278 mg/L, just 15% higher than the BOD used for the theoretical emissions, but the measured BOD removal efficiency was lower (50%) than the theoretical one. It is clear that the application of good practices in the operation of the treatment facilities becomes a relevant criteria, considering their impact on BOD removal efficiencies and CH₄ emissions. In this sense, the TOR system applied the best practices among the 5 sampled treatment facilities, with a trained operational staff and good management.

3.4. Methane emission factors for the five SP

Emission factor is the quantity of GHG emitted from a unit of source or activity associated with that emission (Gupta & Singh 2012). In addition, specific (actual) emission factors may be considered as performance indicators for WWT systems. In the case of WWT inventories, emissions factors are usually expressed as kg CH₄ emitted per kg BOD removed or per m³ of treated water. The CH₄ emission factors in Table 7 are based on the on-site measurements already presented and the system monitoring data, and as such, integrates environmental and operational aspects of each facility.

Table 7. Methane emission factors based on on-site measurements.

WWTP	kg CH4/kg BOD rem	kg CH4/m^3 treated water
TOR	0.431	0.110
MOC	2.561	0.064
IRA	0.122	0.016
COA	1.044	0.021
COM	0.669	0.092

Emission factors based on BOD removal should not be higher than the maximum CH₄ producing capacity (0.6 kg CH₄/kg DBO removed, IPCC 2006) or 0.25 kg CH₄/kg COD removed, based on stoichiometric calculations. Thus, emission factors related to organic matter removal obtained for MOC and COA facilities are not useful for the purpose of this work because they reflect specific operating conditions encountered at the sampling period, that apparently were not representative of long term operation. As mentioned, those two facilities were fed with a very low BOD concentration (61 mg/L) during the measurement period. As a result, the daily methane production did not correspond to the influent mass of BOD in that specific period, being less dependent of short-time influent variation of organic matter concentration. Also, the emission factor based on the BOD removed by the COM facility is higher than the maximum value, a result that may have the same explanation, even if in that case the BOD is in the usual range for municipal sewage (278 mg/L) with lower removal efficiency than expected (50%). The emission factor obtained for the TOR system (0.431 kg CH₄/kg BOD removed) may be taken as a representative value for SPs in Mexico, considering the operational good practices applied in that facility. As a reference value, an emission factor can be calculated (0.36 kg CH₄/kg BOD removed) from the results of Parra et al. (2010) under lower temperature conditions (13 to 14 °C) than those registered at the TOR facility (around 30 °C)

On the other side, emission factors based on treated unit volume (m³) may be a more representative number, as the short term variations on the influent wastewater characterization have a lower impact on methane flux in SPs. This assumption is based on the low organic volumetric rate applied in these systems, being a simple process with no real control measures and slow response to influent changes. Even so, the resulting emission factors shows a wide range variation (0.016 to 0.110 kg CH₄/m³ treated water). For comparison purposes, a value of 0.141 kg CH₄/m³ treated water may be calculated from the results of Parra et al. (2010). On that pond system, the influent sewage concentration was very high (1,336 mg COD/L; 503 mg BOD/L) with 79% removal efficiency, values that may explain the different emission factor obtained for the TOR system, the facility with the best practices.

4. Conclusions

The theoretical values of methane emissions from SP using the Tier 1 IPCC methodology produce an overestimation regarding actual emissions obtained in the field. In this study, an increase factor higher than 3 was determined for the IRA and COA systems, while it was just 1.05 for the facility with the best operational practices (TOR).

The on-site emission factors obtained are specific to each sampled system, considering that each facility has particular environmental and operating conditions. In order to determine reliable regional, process-specific emission factors, a field measurement campaign based on statistical representative sample should be undertaken.

The emission factor obtained in the TOR pond system (0.45 kg CH4/kg BOD removed) may be considered in SP facilities with good operational practices in Mexico. For those working below the expected operational standards, the used of Tier 1 method should be maintained.

Acknowledgement

The first author is grateful to the Mexican National Council of Science and Technology (CONACyT) for a PhD scholarship. The authors would like to thank Jonathan Fernandez for technical assistance.

Disclosure statement

No potential conflict of interest was reported by the authors.

Funding

This work was supported by the project SLCF-2013 Mexico, funded by the Global Environment Facility (GEF) and coordinated by the Molina Center for Energy and the Environment, under contract GFL-4C58 of UNEP. Additional funding was obtained from the Internal Research Fund from the Instituto de Ingeniería UNAM.

Additional information

Funding

This work was supported by the project SLCF-2013 Mexico, funded by the Global Environment Facility (GEF) and coordinated by the Molina Center for Energy and the Environment, under contract GFL-4C58 of UNEP. Additional funding was obtained from the Internal Research Fund from the Instituto de Ingeniería UNAM.