The effectiveness of domestic wastewater treatment ...
THE EFFECTIVENESS OF DOMESTIC WASTEWATER TREATMENT TECHNOLOGIES IN THE CONTEXT OF NEW CONSTRAINTS IMPOSED BY LIFESTYLE CHANGES IN NORTH AMERICAN FAMILIES
Article written by
Mr Roger Lacasse Scientific and Technical Director of Premier Tech Environnement
THE EFFECTIVENESS OF DOMESTIC WASTEWATER TREATMENT TECHNOLOGIES IN THE CONTEXT OF NEW CONSTRAINTS IMPOSED BY LIFESTYLE CHANGES IN NORTH AMERICAN FAMILIES By Roger Lacasse1 Abstract: Family lifestyles have undergone major changes over the last 10 to 20 years. People travels more, eat out more often and, in many families, both parents work outside the home. The growing divorce rate also had an impact on home occupancy. The occupancy of a blended family home can also vary from week to week due to shared custody arrangements. Such lifestyle changes all have an impact on the flow of wastewater from individual dwellings. Nowadays, intermittent (stop/start) and peak flows can be observed not only in secondary or seasonal dwellings, but increasingly in main residences. The testing protocols currently being used for technology certification in the U.S., Canada, and Europe include testing under stress conditions, but are not representative of the new constraints imposed by changes in family lifestyles. For example, the NSF (National Sanitation Foundation) protocol established during the 1970’s does not provide for any testing during peak/overload conditions, and simulation for the two working parents’ condition takes place during only one of the 26 weeks of testing (4%). In an effort to submit wastewater treatment technologies to testing that is more representative of the new constraints imposed by current lifestyles, protocols must evolve in such a way as to ensure that short-term (6- to 12-month) technology evaluations are representative of their long-term performance under actual usage conditions. This paper presents a review of the various existing standards, the performance observed in real-life situations, the evolution observed in the last two years in Europe and Canada, and the changes needed to protocols so that certified technologies can ensure protection of the environment and water resources under conditions imposed by current lifestyles. Lifestyle Changes Over the past two decades, family lifestyles have undergone major changes. People travel more, eat out more often and, in many families, both parents work outside the home. According to the U.S. Census, in 1975, both parents worked outside the home in 45% of families. In 2007 (U.S. Census, 2007), the percentage was around 66%. Figure 1 provides a comparison of the rate of participation of US mothers in the labor force from 1975 to 2006, according to the age of youngest child (Bureau of Labor Statistics, 2006). During that 31-year period, we note an increase of 22% to 26% in the mothers’ participation rate. In 1970 in Canada, both parents worked outside the home in 31% of families. In 1990, this percentage had risen to more than 70%, and this percentage still applies today (Statistics Canada, 2005). In France, the proportion of women and men living as couples and working outside the home in 2007 corresponds to 77.4% and 94%, respectively (INSEE, 2007). The growing divorce rate has also had an impact on home occupancy, and the occupancy of a blended-family home can vary from week to week due to shared custody arrangements. Based on US Census statistics (U.S. Census, 2005), only 58% of men married during the period 1975-1979 stayed married 20 years, compared to 76% of men married in 1955-1959.
1 Roger Lacasse, Technical and Scientific Director, Premier Tech Environnement, 1, ave Premier, Rivière-duLoup, Québec Canada, G5R 6C1. [email protected]
Premier Tech Environnement
2/16
The effectiveness in the context of new constraints 2009-05-21
Figure 1. Labor force participation rates of mothers in the USA
Marital longevity for women who married during these same time periods also fell. As well, married-couple family units dropped from 78.2% in 1950 to 55.2% in 1993 (U.S. Census, 1940-1993). Such lifestyle changes over the past 20 to 30 years are having an impact on the flow of wastewater from individual dwellings. In more than two thirds of North American families, both parents are working outside the home and the wastewater production is concentrated during two periods of the day: in the morning and in the evening. As previously mentioned, the occupancy of a blended-family home can also vary from week to week due to shared custody arrangements (high variation of home occupancy from week to week). Nowadays, intermittent (stop/start) and peak flows can be observed not only in secondary or seasonal dwellings, but increasingly in main residences. As example, preliminary results coming from flow monitoring at blended-family home indicates that weekly average flow rate is up to 3 times higher when the children are at home.
Technology evaluation and certification As previously presented, over the past 20 to 30 years, we’ve observed major lifestyle changes that are having a significant impact on flow from single dwellings (daily flow pattern, peak flow, zero or low flow period, etc.). This reality raises some interesting questions. Do we assume that, over the short term (6 to 12 months), existing standards can be representative of the new “real life” conditions under which systems will be used? How can we make sure that certified onsite wastewater treatment systems are performing under real life conditions? In an effort to gain a better understanding of the different testing conditions, the following paragraphs provide a review and analysis of the three existing standards.
Premier Tech Environnement
3/16
The effectiveness in the context of new constraints 2009-05-21
ANSI/NSF Standard 40 The NSF Certification Program for wastewater treatment units originated over 30 years ago at the request of the regulatory community. It was at this time that the alternative onsite residential wastewater treatment industry began to grow, meeting the need for an effective onsite wastewater treatment solution. ANSI/NSF Standard 40 was developed in the late 70s to evaluate/certify residential wastewater treatment systems with rated capacities between 400 to 1,500 gallons per day. Standard 40 is not restrictive in the type of treatment technology. Any system can be evaluated. As presented on the NSF International Web site: “The standard includes a wide range of product evaluation methods and criteria for residential treatment systems. Most notably is the ability of the treatment system to produce an acceptable quality of effluent. This is demonstrated during a six month (26 week) test where wastewater of required strength is subjected to the system at the rated capacity of the system as evenly dosed at periods prescribed by the standard. Stress sequences are included to simulate wash day, working parent, power outage, and vacation conditions. The effluent criteria required of a Class I system is based on the U.S. EPA secondary effluent treatment requirements for municipal treatment facilities.” For a residential wastewater treatment system to achieve Class I effluent, it must produce an effluent that meets the EPA (Environmental Protection Agency) guidelines for secondary effluent discharge: •
CBOD5: Each 30-day average of effluent samples shall not exceed 25 mg/L and each 7-day average of effluent samples shall not exceed 40 mg/L;
•
TSS: Each 30-day average of effluent samples shall not exceed 30 mg/L and each 7-day average of effluent samples shall not exceed 45 mg/L;
•
pH: Individual effluent values remain between 6.0 and 9.0.
The previous criteria are applicable over a 26-week period, divided into 3 parts: 16 weeks at design loading, 7.5 weeks at stress loading and 2.5 weeks at design loading. During design loading periods, the system is dosed 7 days per week according to the following pattern: •
6 a.m. to 9 a.m. – 35% of daily rated capacity
•
11 a.m. to 2 p.m. – 25% of daily rated capacity
•
5 p.m. to 8 p.m. – 40% of daily rated capacity
The influent and effluent 24-hour composite samples are collected 5 days per week. The stress-loading sequence includes (1) Laundry Day stress, (2) Working Parent stress, (3) Power/Equipment Failure stress, and (4) Vacation stress, defined as: 1- Laundry Day stress consists of 3 laundry days over a 5-day period (each separated by a 24hour period). The system is fed at a design load of 3 loads of laundry (previous dosing pattern is applied).
Premier Tech Environnement
4/16
The effectiveness in the context of new constraints 2009-05-21
2- Working Parent stress: For 5 consecutive days, the system is dosed at 40% of its daily hydraulic capacity between 6:00 a.m. and 9:00 a.m. Between 5:00 p.m. and 8:00 p.m., the system is dosed at the remaining 60% of its hydraulic capacity (including one load of laundry). 3- Power/Equipment Failure stress: Power to the system is turned off and dosing discontinued for 48 hours. 4- Vacation stress: Dosing is discontinued for 8 consecutive days (power continues to be supplied to the system). During the stress test sequences, 24-hour composite samples are collected on the day each stress is initiated. There is no sampling during the other stress days. Influent and effluent 24hour composite samples are also collected 24 hours after completion of laundry day, working parent, and vacation stress for 6 consecutive days (at design loading) and 48 hours after the completion of the power/equipment failure stress for 5 consecutive days (at design loading). Each stress is followed by 7 consecutive days of dosing at design rate capacity before the next stress test begins. According to the NSF 40 testing protocol described previously, certification of a residential wastewater treatment system under Class I only shows that the system can produce a secondary treatment effluent quality under the controlled conditions of the testing protocol. We should be very careful in interpreting results from NSF Standard 40 testing. Based on existing NSF testing reports, there are approximately 93 sampling days under design loading conditions and 23 sampling days during the 4 stress test periods, but the performance of the system during the 11 days when the system is fed but not sampled according to the protocol are not taken into account, that is, the 4 days during the laundry day stress, the 4 days during the working parent stress, the 2 days after the power/equipment failure stress and the 1 day after the vacation stress period. Furthermore, under the NSF testing protocol, a wastewater treatment system might produce average effluent concentrations of TSS and CBOD5 that are below 10 mg/L (overall average of the results obtained during the 26-week testing period) and have a performance that is barely in line with the limit values for weekly/monthly averages during the stress test periods (25/40 mg/L for CBOD5 and 30/45 mg/L for TSS). As presented previously, we know that, under real-life conditions, flow and organic loading from a residence can vary a great deal, depending on the family’s activities and the occupancy of the house: •
In real life, every week includes laundry days. So, what happens in the field if the performance of a residential wastewater treatment system is not well suited for this condition? Note that, according to NSF protocol, influent and effluent 24-hour composite samples are collected only on the day the laundry stress is initiated. There is no sampling over the next 4 days. The system could actually receive NSF 40 Class I certification even if the weekly average during the days following the laundry day stress is 40 mg/L for CBOD5 and 45 mg/L for TSS.
•
In more than 66% of North American families, both parents work outside the home. What happens in the field if the performance of a residential wastewater treatment system certified under Class I of NSF Standard 40 is not well suited for working-parent conditions (wastewater produced only early in the morning and during the evening)? Note
Premier Tech Environnement
5/16
The effectiveness in the context of new constraints 2009-05-21
that, according to the NSF protocol, influent and effluent 24-hour composite samples are collected only on the day the working-parent stress is initiated. There is no sampling over the next 4 days. The system could actually receive NSF 40 Class I certification even if the weekly average during the days following the working parent stress is 40 mg/L for CBOD5 and 45 mg/L for TSS. •
Family lifestyle changes (people travel more, many families have secondary homes and the divorce rate is higher) affect house occupancy. In newly formed families, house occupancy may vary from week to week because of shared custody or other similar arrangements. As a result, intermittent and peak flows are generated not only in secondary or seasonal homes, but increasingly in permanent homes as well. What happens in the field if the performance of a residential wastewater treatment system certified under Class I of NSF Standard 40 is not well suited for rest periods (zero flow periods) like vacation stress condition) or intermittent flow? Note that, according to the NSF protocol, influent and effluent 24-hour composite samples are collected for 6 days, starting 24 hours after the end of the rest period. The system could actually receive NSF 40 Class I certification even if the weekly average during the days following the vacation stress is 40 mg/L for CBOD5 and 45 mg/L for TSS.
Based on the statements presented above, it is clear that the results of many of the studies performed in the field show that NSF 40 certified ATU (Aerobic Treatment Unit) technologies under Class I do not, in real life, perform at the same level as that observed during testing on the NSF bench. Sexton and al. (2000) present the results of a survey of home ATUs operating in 6 West Virginia counties. The West Virginia Bureau of Public Health requires that each and every ATU system meet NSF Standard 40 Class I. Of the 419 systems examined, 85 were sampled for BOD5 and TSS. The results obtained indicate that 48% of measured samples exceeded the monthly average of 30 mg/L for TSS, and 69% of BOD5 results exceeded the monthly average of 30 mg/L. Maintenance deficiencies, like septic solids in the aeration chamber, aerator malfunctions or floating solids in the settling chamber, impact performance, but systems with no maintenance deficiencies exceeded TSS and/or BOD5 31% of the time. These results may suggest that the testing protocol does not properly represent the real life conditions related to the new lifestyles.
NQ 3680-910 Standard This BNQ (Bureau de la Normalisation du Québec) Standard, which has been applied in Province of Quebec since 2006, introduced different classes of treatment according to regulation requirements: Class I (primary TSS ≤ 100 mg/L), Class II (secondary TSS ≤ 30 mg/L and CBOD5 ≤ 30 mg/L), Class III (advanced secondary TSS and CBOD5 ≤ 15 mg/L and fecal coliforms ≤ 50,000 counts/100 mL, Class IV (tertiary with phosphate removal (TSS and CBOD5 ≤ 15 mg/L, fecal coliforms ≤ 50,000 counts/100 mL and Total P ≤ 1.0 mg/L) and Class V (tertiary with disinfection TSS and CBOD5 ≤ 15 mg/L, fecal coliforms ≤ 200 counts/100 mL). The first 6-month testing period corresponds to the ANSI/NSF 40 protocol, including stress tests (a reciprocity agreement exists between the BNQ and the NSF), and the second 6-month testing period assures that certified onsite wastewater treatment systems are tested under all four North American season conditions (warm and cold conditions). During this additional 6-month testing period, the system is dosed 7 days per week according to the NSF dosing pattern (3 periods of 3 hours per day), with a minimum of 10 sampling days. During this second testing period, 80% of the results must comply with the limits for the class for which technology is tested. Premier Tech Environnement
6/16
The effectiveness in the context of new constraints 2009-05-21
The testing protocol also includes an annual field audit to assure that certified technologies under NQ 3680-910 Class I to V are in compliance with the limits prescribed for the different classes. The audit is performed by a third party entity under BNQ responsibility on 1% of installed systems (minimum of 5 systems per year and maximum of 10). To meet the requirements of the standard and ensure a wastewater treatment system maintains its “certified” status, 80% of the results obtained in the field each year must comply with the limits of the certified class. The first technologies were certified under Standard NQ 3680-910 at the end of 2005, and, from 2006 to 2008, annual audits were performed on the certified technologies. For example, for the Ecoflo® Biofilter, results obtained in the field are in line with those obtained on the BNQ testing platform (Table 1), and the annual audit results also agree with the Ecoflo® Biofilter’s field performance obtained during Premier Tech’s voluntary sampling program (Premier Tech Environment, 2006).
Table 1.
Comparison of field and platform performance for Ecoflo Biofilter Standard NQ 3680- 910 Parameter
TSS (mg/L)
CBOD5 (mg/L)
Fecal coliforms (counts/100 mL)
Class III limits
15
15
50,000
BNQ testing platform 2 2 1,250 (average)1 BNQ testing platform 100% 100% 99% (% of results ≤ class limits) BNQ annual audit 4 5 256 (average)2 BNQ annual audit 100% 95% 100% (% of results ≤ class limits) PTE voluntary monitoring program 4 5 1073 (average)3 PTE voluntary monitoring program 99% 98% 94% (% of results ≤ class limits) 1 Average of 120 samples for TSS and CBOD5 and geometric average of 398 samples for fecal coliforms 2 Average of 19 samples 3 Average of 205 samples for TSS and CBOD5 and geometric average of 194 samples for fecal coliforms
EN 12566-3 Standard (2004) The European Standard EN-12566-3 (CEN/TC 165, 2004) for onsite wastewater treatment certification is based on 38 weeks of testing after approximately 4 weeks of system start-up. The system is fed 7 days per week according to the following pattern: • • • • • •
30% of the daily volume for 3 hours 15% of the daily volume for 3 hours 0% for 6 hours 40% of the daily volume for 2 hours 15% of the daily volume for 3 hours 0% for 7 hours
Premier Tech Environnement
7/16
The effectiveness in the context of new constraints 2009-05-21
The testing protocol is divided into 10 sequences with different sampling frequencies (Table 2).
Table 2.
Testing protocol of Standard EN 12566-3
Sequence
Description
% of design flow rate
Duration (weeks)
Sampling days
1
Start-up period
100
approx. 4
0
2
Design flow
100
6
4
3
Low flow
50
2
2
4
Design flow + power failure (system fed)
100
6
5
5
Vacation
0
2
0
6
Design flow
100
6
3
7
Design + 2-day peak flow
100 + 150 (2-day)
2
2
8
Design flow + 3 days of power failure (system fed)
100
6
5
9
Low flow
50
2
2
10
Design flow
100
6
3
Based on the above review and analysis of the three existing testing standards in North America and Europe, some changes in the testing protocol have been observed over the past 5 years. The testing period for BNQ and European Standards is based on 12 and 9 months, respectively, to better reflect seasonal variations (warm and cold conditions), compared to the 6-month period for the NSF Standard. As a condition for certification renewal, the BNQ Standard has introduced an annual audit to control technology performance not only on the testing platform, but also in the field. Also, as presented at Figure 2, the European Standard has introduced a dosing protocol that better reflects the lifestyle of the majority of families, where both parents are working outside the home. However, considering the new constraints imposed by current lifestyles (more peak flow, etc.), the existing protocols must be revised to ensure that short-term technology evaluations are representative of their long-term performance under actual usage conditions, and that certified technologies will perform not only on the testing platform under controlled conditions, but also in the field.
Premier Tech Environnement
8/16
The effectiveness in the context of new constraints 2009-05-21
30
25
% of daily volume/h
NSF dosing protocol NSF working parent EU dosing protocol
20
Series1 Series2
15
10
5
0 1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 hours
Figure 2. Comparison of dosing protocols
Evolution of testing protocols and standards In an effort to submit wastewater treatment technologies to testing that is more representative of the new constraints, in the past two years, we observed some improvements in testing protocols and standards in both Europe and Canada. New testing protocol in Europe In order to test residential wastewater treatment technologies under conditions that better reflect the conditions observed in individual applications (peak flows, variable flows, etc.) and today’s lifestyles, in 2006-2007, a comparative study was performed by the CSTB (Building Scientific and Technical Center) and monitored by an independent group of experts (Veolia Water, 2007). This group of experts developed a testing protocol that compared and evaluated eight (8) different wastewater treatment technologies on the CSTB testing platform located in Nantes, France. The new protocol is based on a 40-week testing period that includes the following steps: • • • • • • • • •
Four (4) weeks for system start-up (stabilizing of the biological process) at design flow rate; Twelve (12) weeks at design flow rate; Four (4) weeks at design flow rate from Monday to Thursday, and at 2 times design flow rate for the other 3 days. This condition reflects typical high occupancy during week-ends (a frequent condition in secondary homes); Three (3) weeks at 2 times design flow rate (overloading conditions for high-occupancy conditions); Three (3) weeks at zero flow to simulate a vacation period; Two (2) weeks at design flow rate from Monday to Thursday, and at 2 times design flow rate for the other 3 days; Four (4) weeks at design flow rate; Two (2) weeks at 50% of design flow rate (low occupancy conditions); Six (6) weeks at design flow rate with 3 days of power failure (one every 2 weeks).
Premier Tech Environnement
9/16
The effectiveness in the context of new constraints 2009-05-21
During all of the previous periods, the wastewater treatment technologies are fed according to the EN 12566-3 dosing protocol. Phase 1 of this evaluation project began in February 2006 and was completed in November 2006. New testing phases are presently ongoing. The eight (8) wastewater treatment technologies tested were representative of the different treatment processes existing in the market: • • • • • • • •
Sand filter sized according to the French regulation Reduced size sand filter using a gravelless distribution device (plastic channel with geotextile interior) Constructed wetland Zeolith-based media filter Organic-based media biofilter Textile-based media biofilter Suspended growth ATU Attached growth ATU
The capacity of the tested systems corresponds to 5-6 inhabitant equivalents, and 24-hour composite samples were collected twice a week to analyze for TSS, COD, CBOD5, ammonia, and pH at the system inlet (inlet of the septic tank), septic tank effluent and treatment unit effluent. Table 3 provides the raw wastewater characteristics at the inlet of the system showing that the wastewater used was representative of domestic applications with low dilution. According to the protocol, minimum CBOD5 concentrations of 300 mg/L were required to reflect sufficient organic loading on the tested system. Table 4 provides a summary of the results (average ± standard deviation) obtained during Phase 1 of the evaluation project. An analysis of the results provided above shows significant performance differences among the eight types of technology tested when the protocol is closer to actual lifestyles (e.g.: peak flow and load), and sampling is also done during the stress conditions. On average, the organic-based media biofilter is the system that performs best with effluent concentrations of 7 and 5 mg/L for TSS and CBOD5 respectively, and provides the highest stability, with a standard deviation of 3 mg/L. Average results with sand filter effluent were also below 10 mg/L for TSS and CBOD5, but with slightly more variation. It is important to note that the installed sand filter was under drained, and results would likely be different if the sand filter had not been drained before the infiltration of treated effluent into native soil. The draining layer at the bottom of every filter (sand, peat, coconut, foam, etc.) has a major impact on the hydrodynamic conditions of the filter (saturation level, aeration, etc.) and on the filter’s performance. Three other biofiltration technologies achieved average performance, below 15 mg/L for TSS and CBOD5, but with more variation, especially with a reduced size sand filter using a gravelless distribution device. It is interesting to note that the first five technologies are all based on a biofiltration process. Results obtained with ATUs indicated higher average effluent concentrations and significantly less treatment stability. For an ATU with attached growth, average effluent concentrations were in line with the secondary treatment level (TSS = 25 mg/l and BOD5 = 30 mg/L). However, ATU with suspended growth provided a notably lower quality of effluent along with fairly poor stability of the treatment performance. It is interesting to note that previous results obtained under testing conditions representing typical residential flow and real-life load variations are close to the ATU performance reported by Sextone and al (2000) for systems monitored under field conditions. A total of 85 ATU systems were Premier Tech Environnement
10/16
The effectiveness in the context of new constraints 2009-05-21
sampled for BOD5 and TSS. The results obtained indicate that 48% of measured samples exceeded the monthly average of 30 mg/L in TSS, and 69% of BOD5 results exceeded the monthly average of 30 mg/L.
Table 3.
Raw Wastewater Characteristics (at septic tank inlet)
Parameter
Units
Average
Median
Minimum
CBOD5
mg/L
TSS
mg/L
COD
mg/L
Maximum
330*
290
110
780
304*
267
91
1092
722*
645
250
1990
*Average of 42 samples
To illustrate the impact of stress conditions occurring in typical single dwellings with today’s lifestyles (design and peak flows, zero flow period, power outage, etc.) on wastewater treatment performance, Figures 3, 4 and 5 compare the quality (TSS content) of the treated effluent of three different types of technologies to the treated effluent quality produced by the organic-based media biofilter that obtained the best performance during the testing period.
Table 4. Treated Effluent Characteristics during Phase 1 Technology Organic-based biofilter
Mean concentration of effluent* (± standard deviation) TSS (mg/L) CBOD5 (mg/L) COD (mg/L) 7±3 5±3 54 ± 19
Sand filter
7±6
6±4
44 ± 26
Textile-based biofilter
13 ± 11
8±3
59 ± 26
Zeolith Filter
14 ± 9
11 ± 5
85 ± 25
Reduced sand filter size with gravelless distribution system
15 ± 10
13 ± 9
75 ± 42
Attached growth ATU
16 ± 12
19 ± 9
80 ± 34
Suspended growth ATU**
40 ± 30
45 ± 33
189 ± 98
Constructed wetland*** 36 ± 10 60 ± 23 168 ± 51 * Average of 40 samples ** Sludge removal was required after the first 6 months of the operation period. *** The initial system has been replaced by a new version of a constructed wetland (monitoring is ongoing).
Figure 3 presents TSS variations in suspended-growth ATU treated effluent. Under all conditions tested, results clearly show low performance under peak flow conditions and the need for at least 12 weeks of operation at the design flow rate before the process is stabilized (TSS = 25 mg/L). Also, sludge had to be pumped out after only 33 weeks of operation, and it is clear that the treated effluent from such a system (suspended-growth ATU) requires an additional treatment step (filtration) before it is infiltrated into native soil. By comparison, the organic-based media biofilter provided high stability (TSS ≤ 10 mg/L) under all conditions tested.
Premier Tech Environnement
11/16
The effectiveness in the context of new constraints 2009-05-21
Figure 4 provides the TSS effluent concentrations produced by an attached growth ATU. Two different testing periods are of particular significance: 1) 12 weeks are required to stabilize the attached growth ATU. As shown in Figure 4, the TSS concentrations at the ATU effluent gradually decrease to below 25 mg/L after 12 weeks; 3) Power or blower failure conditions (aeration system turns off and system feeding is maintained) had a major impact on the performance of the ATU, with TSS concentrations close to 40 mg/L at the treated effluent. Although the results obtained with attached growth ATUs are better than those for suspended growth ATUs, results clearly show that attached growth ATUs are not well suited for the intermittent flow conditions and variable organic loading observed in secondary or seasonal homes and in an increasing number of permanent homes. Here again, the treated effluent from such systems requires an additional treatment step (filtration) before infiltration in native soil.
Figure 3. Comparison of the Treatment Performance between an Organic-based Biofilter and a Suspended Growth ATU
Premier Tech Environnement
12/16
The effectiveness in the context of new constraints 2009-05-21
Figure 4. Treatment Performance Comparison between an Organic-based Biofilter and an Attached Growth ATU
Finally, Figure 5 provides the treated TSS effluent concentrations from an under drained sand filter. The performance of both types of biofilter is very stable except after the rest period (zero flow conditions) for the sand filter (TSS near 40 mg/L after the zero flow conditions). This result illustrates the main difference between multiple porosity level filtering media, such as organic-based media (e.g.: peat or coconut) or synthetic-based media (e.g.: foam) and single porosity media, such as sand. In fact, sand has only one level of porosity, which corresponds to the space between the grains of sand. Under zero flow conditions, sand media drains gradually and the dryness significantly impacts the survival of the micro-organisms in the media. This explains the decreased performance of the sand filter when the system is fed following a rest period. For multiple porosity level media, water is retained not only between the media particles, but also inside each media grain. Under zero flow conditions, this type of media stays wet, which ensures the survival of sufficient numbers of micro-organisms, providing a consistently high performance under all conditions, even when the system is fed after a rest period. Results show that a biofilter based on multiple porosity levels media (e.g.: peat) is particularly well suited to the intermittent flow conditions observed in secondary or seasonal homes and in an increasing number of permanent homes.
Premier Tech Environnement
13/16
The effectiveness in the context of new constraints 2009-05-21
Figure 5. Comparison of the Treatment Performance between an Organic-based Biofilter and an Under Drained Sand Filter
New BNQ Standard in Canada Early in 2007, some Canadian provincial authorities asked the BNQ to develop a new BNQ Standard regarding the certification of onsite wastewater treatment systems, applicable across all provinces and territories in Canada. A technical committee composed of representatives of different groups (regulators, manufacturers, universities, associations, and customers) from the majority of the 12 provinces and territories held a first meeting in March 2007. The existing Quebec Standard (NQ 3680-910) was the starting point for the committee’s work. By the summer of 2008, a draft version of the new BNQ Standard 3680-600 was published for public hearing for 60 days (BNQ, 2008). The Standard is now completed and has been submitted for approval by the Standard Council of Canada. Approval is expected in the spring of 2009. The new BNQ Standard 3680-600 corresponds to the standard now in existence in Province of Quebec (12- month testing period with the first 6 months corresponding to the ANSI/NSF 40 protocol, including stress tests), but includes some major improvements that better reflect the new constraints imposed by current lifestyles (working parents, etc.) and new environmental requirements: •
During the stress test sequences, 24-hour composite samples are collected on all stress days (not only on the day each stress is initiated, as prescribed in the ANSI/NSF 40 protocol);
•
A change in the dosing pattern for the second 6-month testing period (Annex B) to reflect conditions where both parents are working outside the home: 40% of the daily volume applied in the morning (between 6 a.m. and 9 a.m.) and 60% in the evening (between 5 p.m. and 8 p.m.), five days per week. During the two other days, the system is fed according to the ANSI/NSF 40 dosing pattern (3 periods of 3 hours per day). Then, the
Premier Tech Environnement
14/16
The effectiveness in the context of new constraints 2009-05-21
system is tested under working parent conditions for 27 weeks (1 week during the first 6month testing period and 26 weeks during Annex B) out of 52 (52% of the time as opposed to 4% of the time with Standard ANSI/NSF 40); •
Influent and effluent sampling one day per week for a total of 26 sampling days during the second 6-month testing period (Annex B). At the present time, the existing BNQ Standard 3680-910 requires a minimum of 10 samples over the Annex B testing period;
•
Addition of treatment classes corresponding to current and future needs and related to the dispersal of treated effluent in sensitive areas (disinfection, phosphorus or nitrogen removal). The definitions of the different classes are presented at Table 5 and the standard allows for different combinations of “Basic treatment” (B classes) with other classes (D, P or N).
Table 5. Treatment Classes defined in the new BNQ Standard 3680-600 Treatment classes
Basic treatment (B)
Disinfection (D)
P removal (P)
N removal (N)
TSS*
CBOD5*
Fecal coliforms or E. Coli*
Total P*
Total N
B-I
100
150
B-II
30
25
B-III
15
15
B-IV
10
10
D-I
50,000
D-II
200
D-III
ND (median < 10)
P-I
1.0
P-II
0.3
N-I
50%
N-II
75%
* All values are in mg/L except for fecal coliforms or E. coli, which is measured in counts/100 mL.
Conclusion Over the past 20 to 30 years, we have observed major lifestyle changes that have a significant impact on the wastewater flow from single dwellings (daily flow pattern, peak flow, zero or low flow period, etc.). A review of the existing standards (NSF, BNQ or European) clearly shows that improvements are needed to better reflect current lifestyles and ensure the certification of wastewater treatment systems performing under real-life conditions. We noted that the existing BNQ and European Standards provide some major improvements compared to the NSF Standard (for example, a longer testing period - up to 12 months for the BNQ Standard) to ensure certification of the technology under conditions for all four North American seasons, and a dosing pattern that is closer to current lifestyles (European Standard), where both parents work outside the home in more than 66% of families. In addition, the major study undertaken at the CSTB site in France showed the importance of submitting treatment technologies to more challenging protocols that reflect current lifestyles, in order to make sure that certified technologies perform under the new “real life” conditions. Premier Tech Environnement
15/16
The effectiveness in the context of new constraints 2009-05-21
Finally, the evolution of (changes in) the standard for onsite wastewater treatment system certification is ongoing in Canada and 2009 will see the publication of a new and improved BNQ Standard that will provide the possibility of certifying treatment technologies for more and more challenging environmental conditions, applicable at the national level.
References Bureau de Normalisation du Québec (BNQ), 2005. Wastewater Treatment – Stand-alone treatment systems for isolated single dwelling – Advanced secondary treatment system (class III) - Biofilter Ecoflo® ST-650. Performance report for Annex A. Bureau de Normalisation du Québec (BNQ), 2005. Wastewater Treatment – Stand-alone treatment systems for isolated single dwelling – Advanced secondary treatment system (class III) - Biofilter Ecoflo® ST-650. Performance report for Annex B. Bureau de Normalisation du Québec (BNQ), 2008. Onsite Residential Wastewater Treatment Technologies. Final Draft Standard D 3680-600. CEN/TC 165, Small Wastewater Treatment Systems for up to 50 PT – Part 3: Packaged and/or site assembled domestic wastewater treatment plants, November 2004. Bureau of Labor Statistics, Chart 6-3, 2006 INSEE, Activité, emploi et chômage selon le type de ménage et le nombre d’enfants, 2007. Premier Tech Environment, 2006. Voluntary monitoring program – 1995 to 2006. Sextone, A.,Bissonette G., Flemeing K., Kinner K., hench K., Bozicevitch T., Cooley B. and Winant E. 2000 A survey of home aerobic treatment systems operating in six West Virginia Counties. Small Flows Quaterly, fall 2000,Vol.1, Nb 4, pp. 38-46. Statistics Canada, Population Survey, 2005 and earlier. US Census, Population Profile of United States, Family and Living Arrangements in 2005. U.S. Census Bureau, Current Population Survey, March and Annual Social and Economic Supplements, 2007 and earlier. U.S. Census Bureau, Table A-1, Average Population per Household and Family: 1940-1993. Veolia Water (2007), Etude comparative de 8 filières de traitement: résultats et évaluations. Conference presented in Cahors (France), October 25th, 2007,
Premier Tech Environnement
16/16
The effectiveness in the context of new constraints 2009-05-21
Article written by
Mr Roger Lacasse Scientific and Technical Director of Premier Tech Environnement
THE EFFECTIVENESS OF DOMESTIC WASTEWATER TREATMENT TECHNOLOGIES IN THE CONTEXT OF NEW CONSTRAINTS IMPOSED BY LIFESTYLE CHANGES IN NORTH AMERICAN FAMILIES By Roger Lacasse1 Abstract: Family lifestyles have undergone major changes over the last 10 to 20 years. People travels more, eat out more often and, in many families, both parents work outside the home. The growing divorce rate also had an impact on home occupancy. The occupancy of a blended family home can also vary from week to week due to shared custody arrangements. Such lifestyle changes all have an impact on the flow of wastewater from individual dwellings. Nowadays, intermittent (stop/start) and peak flows can be observed not only in secondary or seasonal dwellings, but increasingly in main residences. The testing protocols currently being used for technology certification in the U.S., Canada, and Europe include testing under stress conditions, but are not representative of the new constraints imposed by changes in family lifestyles. For example, the NSF (National Sanitation Foundation) protocol established during the 1970’s does not provide for any testing during peak/overload conditions, and simulation for the two working parents’ condition takes place during only one of the 26 weeks of testing (4%). In an effort to submit wastewater treatment technologies to testing that is more representative of the new constraints imposed by current lifestyles, protocols must evolve in such a way as to ensure that short-term (6- to 12-month) technology evaluations are representative of their long-term performance under actual usage conditions. This paper presents a review of the various existing standards, the performance observed in real-life situations, the evolution observed in the last two years in Europe and Canada, and the changes needed to protocols so that certified technologies can ensure protection of the environment and water resources under conditions imposed by current lifestyles. Lifestyle Changes Over the past two decades, family lifestyles have undergone major changes. People travel more, eat out more often and, in many families, both parents work outside the home. According to the U.S. Census, in 1975, both parents worked outside the home in 45% of families. In 2007 (U.S. Census, 2007), the percentage was around 66%. Figure 1 provides a comparison of the rate of participation of US mothers in the labor force from 1975 to 2006, according to the age of youngest child (Bureau of Labor Statistics, 2006). During that 31-year period, we note an increase of 22% to 26% in the mothers’ participation rate. In 1970 in Canada, both parents worked outside the home in 31% of families. In 1990, this percentage had risen to more than 70%, and this percentage still applies today (Statistics Canada, 2005). In France, the proportion of women and men living as couples and working outside the home in 2007 corresponds to 77.4% and 94%, respectively (INSEE, 2007). The growing divorce rate has also had an impact on home occupancy, and the occupancy of a blended-family home can vary from week to week due to shared custody arrangements. Based on US Census statistics (U.S. Census, 2005), only 58% of men married during the period 1975-1979 stayed married 20 years, compared to 76% of men married in 1955-1959.
1 Roger Lacasse, Technical and Scientific Director, Premier Tech Environnement, 1, ave Premier, Rivière-duLoup, Québec Canada, G5R 6C1. [email protected]
Premier Tech Environnement
2/16
The effectiveness in the context of new constraints 2009-05-21
Figure 1. Labor force participation rates of mothers in the USA
Marital longevity for women who married during these same time periods also fell. As well, married-couple family units dropped from 78.2% in 1950 to 55.2% in 1993 (U.S. Census, 1940-1993). Such lifestyle changes over the past 20 to 30 years are having an impact on the flow of wastewater from individual dwellings. In more than two thirds of North American families, both parents are working outside the home and the wastewater production is concentrated during two periods of the day: in the morning and in the evening. As previously mentioned, the occupancy of a blended-family home can also vary from week to week due to shared custody arrangements (high variation of home occupancy from week to week). Nowadays, intermittent (stop/start) and peak flows can be observed not only in secondary or seasonal dwellings, but increasingly in main residences. As example, preliminary results coming from flow monitoring at blended-family home indicates that weekly average flow rate is up to 3 times higher when the children are at home.
Technology evaluation and certification As previously presented, over the past 20 to 30 years, we’ve observed major lifestyle changes that are having a significant impact on flow from single dwellings (daily flow pattern, peak flow, zero or low flow period, etc.). This reality raises some interesting questions. Do we assume that, over the short term (6 to 12 months), existing standards can be representative of the new “real life” conditions under which systems will be used? How can we make sure that certified onsite wastewater treatment systems are performing under real life conditions? In an effort to gain a better understanding of the different testing conditions, the following paragraphs provide a review and analysis of the three existing standards.
Premier Tech Environnement
3/16
The effectiveness in the context of new constraints 2009-05-21
ANSI/NSF Standard 40 The NSF Certification Program for wastewater treatment units originated over 30 years ago at the request of the regulatory community. It was at this time that the alternative onsite residential wastewater treatment industry began to grow, meeting the need for an effective onsite wastewater treatment solution. ANSI/NSF Standard 40 was developed in the late 70s to evaluate/certify residential wastewater treatment systems with rated capacities between 400 to 1,500 gallons per day. Standard 40 is not restrictive in the type of treatment technology. Any system can be evaluated. As presented on the NSF International Web site: “The standard includes a wide range of product evaluation methods and criteria for residential treatment systems. Most notably is the ability of the treatment system to produce an acceptable quality of effluent. This is demonstrated during a six month (26 week) test where wastewater of required strength is subjected to the system at the rated capacity of the system as evenly dosed at periods prescribed by the standard. Stress sequences are included to simulate wash day, working parent, power outage, and vacation conditions. The effluent criteria required of a Class I system is based on the U.S. EPA secondary effluent treatment requirements for municipal treatment facilities.” For a residential wastewater treatment system to achieve Class I effluent, it must produce an effluent that meets the EPA (Environmental Protection Agency) guidelines for secondary effluent discharge: •
CBOD5: Each 30-day average of effluent samples shall not exceed 25 mg/L and each 7-day average of effluent samples shall not exceed 40 mg/L;
•
TSS: Each 30-day average of effluent samples shall not exceed 30 mg/L and each 7-day average of effluent samples shall not exceed 45 mg/L;
•
pH: Individual effluent values remain between 6.0 and 9.0.
The previous criteria are applicable over a 26-week period, divided into 3 parts: 16 weeks at design loading, 7.5 weeks at stress loading and 2.5 weeks at design loading. During design loading periods, the system is dosed 7 days per week according to the following pattern: •
6 a.m. to 9 a.m. – 35% of daily rated capacity
•
11 a.m. to 2 p.m. – 25% of daily rated capacity
•
5 p.m. to 8 p.m. – 40% of daily rated capacity
The influent and effluent 24-hour composite samples are collected 5 days per week. The stress-loading sequence includes (1) Laundry Day stress, (2) Working Parent stress, (3) Power/Equipment Failure stress, and (4) Vacation stress, defined as: 1- Laundry Day stress consists of 3 laundry days over a 5-day period (each separated by a 24hour period). The system is fed at a design load of 3 loads of laundry (previous dosing pattern is applied).
Premier Tech Environnement
4/16
The effectiveness in the context of new constraints 2009-05-21
2- Working Parent stress: For 5 consecutive days, the system is dosed at 40% of its daily hydraulic capacity between 6:00 a.m. and 9:00 a.m. Between 5:00 p.m. and 8:00 p.m., the system is dosed at the remaining 60% of its hydraulic capacity (including one load of laundry). 3- Power/Equipment Failure stress: Power to the system is turned off and dosing discontinued for 48 hours. 4- Vacation stress: Dosing is discontinued for 8 consecutive days (power continues to be supplied to the system). During the stress test sequences, 24-hour composite samples are collected on the day each stress is initiated. There is no sampling during the other stress days. Influent and effluent 24hour composite samples are also collected 24 hours after completion of laundry day, working parent, and vacation stress for 6 consecutive days (at design loading) and 48 hours after the completion of the power/equipment failure stress for 5 consecutive days (at design loading). Each stress is followed by 7 consecutive days of dosing at design rate capacity before the next stress test begins. According to the NSF 40 testing protocol described previously, certification of a residential wastewater treatment system under Class I only shows that the system can produce a secondary treatment effluent quality under the controlled conditions of the testing protocol. We should be very careful in interpreting results from NSF Standard 40 testing. Based on existing NSF testing reports, there are approximately 93 sampling days under design loading conditions and 23 sampling days during the 4 stress test periods, but the performance of the system during the 11 days when the system is fed but not sampled according to the protocol are not taken into account, that is, the 4 days during the laundry day stress, the 4 days during the working parent stress, the 2 days after the power/equipment failure stress and the 1 day after the vacation stress period. Furthermore, under the NSF testing protocol, a wastewater treatment system might produce average effluent concentrations of TSS and CBOD5 that are below 10 mg/L (overall average of the results obtained during the 26-week testing period) and have a performance that is barely in line with the limit values for weekly/monthly averages during the stress test periods (25/40 mg/L for CBOD5 and 30/45 mg/L for TSS). As presented previously, we know that, under real-life conditions, flow and organic loading from a residence can vary a great deal, depending on the family’s activities and the occupancy of the house: •
In real life, every week includes laundry days. So, what happens in the field if the performance of a residential wastewater treatment system is not well suited for this condition? Note that, according to NSF protocol, influent and effluent 24-hour composite samples are collected only on the day the laundry stress is initiated. There is no sampling over the next 4 days. The system could actually receive NSF 40 Class I certification even if the weekly average during the days following the laundry day stress is 40 mg/L for CBOD5 and 45 mg/L for TSS.
•
In more than 66% of North American families, both parents work outside the home. What happens in the field if the performance of a residential wastewater treatment system certified under Class I of NSF Standard 40 is not well suited for working-parent conditions (wastewater produced only early in the morning and during the evening)? Note
Premier Tech Environnement
5/16
The effectiveness in the context of new constraints 2009-05-21
that, according to the NSF protocol, influent and effluent 24-hour composite samples are collected only on the day the working-parent stress is initiated. There is no sampling over the next 4 days. The system could actually receive NSF 40 Class I certification even if the weekly average during the days following the working parent stress is 40 mg/L for CBOD5 and 45 mg/L for TSS. •
Family lifestyle changes (people travel more, many families have secondary homes and the divorce rate is higher) affect house occupancy. In newly formed families, house occupancy may vary from week to week because of shared custody or other similar arrangements. As a result, intermittent and peak flows are generated not only in secondary or seasonal homes, but increasingly in permanent homes as well. What happens in the field if the performance of a residential wastewater treatment system certified under Class I of NSF Standard 40 is not well suited for rest periods (zero flow periods) like vacation stress condition) or intermittent flow? Note that, according to the NSF protocol, influent and effluent 24-hour composite samples are collected for 6 days, starting 24 hours after the end of the rest period. The system could actually receive NSF 40 Class I certification even if the weekly average during the days following the vacation stress is 40 mg/L for CBOD5 and 45 mg/L for TSS.
Based on the statements presented above, it is clear that the results of many of the studies performed in the field show that NSF 40 certified ATU (Aerobic Treatment Unit) technologies under Class I do not, in real life, perform at the same level as that observed during testing on the NSF bench. Sexton and al. (2000) present the results of a survey of home ATUs operating in 6 West Virginia counties. The West Virginia Bureau of Public Health requires that each and every ATU system meet NSF Standard 40 Class I. Of the 419 systems examined, 85 were sampled for BOD5 and TSS. The results obtained indicate that 48% of measured samples exceeded the monthly average of 30 mg/L for TSS, and 69% of BOD5 results exceeded the monthly average of 30 mg/L. Maintenance deficiencies, like septic solids in the aeration chamber, aerator malfunctions or floating solids in the settling chamber, impact performance, but systems with no maintenance deficiencies exceeded TSS and/or BOD5 31% of the time. These results may suggest that the testing protocol does not properly represent the real life conditions related to the new lifestyles.
NQ 3680-910 Standard This BNQ (Bureau de la Normalisation du Québec) Standard, which has been applied in Province of Quebec since 2006, introduced different classes of treatment according to regulation requirements: Class I (primary TSS ≤ 100 mg/L), Class II (secondary TSS ≤ 30 mg/L and CBOD5 ≤ 30 mg/L), Class III (advanced secondary TSS and CBOD5 ≤ 15 mg/L and fecal coliforms ≤ 50,000 counts/100 mL, Class IV (tertiary with phosphate removal (TSS and CBOD5 ≤ 15 mg/L, fecal coliforms ≤ 50,000 counts/100 mL and Total P ≤ 1.0 mg/L) and Class V (tertiary with disinfection TSS and CBOD5 ≤ 15 mg/L, fecal coliforms ≤ 200 counts/100 mL). The first 6-month testing period corresponds to the ANSI/NSF 40 protocol, including stress tests (a reciprocity agreement exists between the BNQ and the NSF), and the second 6-month testing period assures that certified onsite wastewater treatment systems are tested under all four North American season conditions (warm and cold conditions). During this additional 6-month testing period, the system is dosed 7 days per week according to the NSF dosing pattern (3 periods of 3 hours per day), with a minimum of 10 sampling days. During this second testing period, 80% of the results must comply with the limits for the class for which technology is tested. Premier Tech Environnement
6/16
The effectiveness in the context of new constraints 2009-05-21
The testing protocol also includes an annual field audit to assure that certified technologies under NQ 3680-910 Class I to V are in compliance with the limits prescribed for the different classes. The audit is performed by a third party entity under BNQ responsibility on 1% of installed systems (minimum of 5 systems per year and maximum of 10). To meet the requirements of the standard and ensure a wastewater treatment system maintains its “certified” status, 80% of the results obtained in the field each year must comply with the limits of the certified class. The first technologies were certified under Standard NQ 3680-910 at the end of 2005, and, from 2006 to 2008, annual audits were performed on the certified technologies. For example, for the Ecoflo® Biofilter, results obtained in the field are in line with those obtained on the BNQ testing platform (Table 1), and the annual audit results also agree with the Ecoflo® Biofilter’s field performance obtained during Premier Tech’s voluntary sampling program (Premier Tech Environment, 2006).
Table 1.
Comparison of field and platform performance for Ecoflo Biofilter Standard NQ 3680- 910 Parameter
TSS (mg/L)
CBOD5 (mg/L)
Fecal coliforms (counts/100 mL)
Class III limits
15
15
50,000
BNQ testing platform 2 2 1,250 (average)1 BNQ testing platform 100% 100% 99% (% of results ≤ class limits) BNQ annual audit 4 5 256 (average)2 BNQ annual audit 100% 95% 100% (% of results ≤ class limits) PTE voluntary monitoring program 4 5 1073 (average)3 PTE voluntary monitoring program 99% 98% 94% (% of results ≤ class limits) 1 Average of 120 samples for TSS and CBOD5 and geometric average of 398 samples for fecal coliforms 2 Average of 19 samples 3 Average of 205 samples for TSS and CBOD5 and geometric average of 194 samples for fecal coliforms
EN 12566-3 Standard (2004) The European Standard EN-12566-3 (CEN/TC 165, 2004) for onsite wastewater treatment certification is based on 38 weeks of testing after approximately 4 weeks of system start-up. The system is fed 7 days per week according to the following pattern: • • • • • •
30% of the daily volume for 3 hours 15% of the daily volume for 3 hours 0% for 6 hours 40% of the daily volume for 2 hours 15% of the daily volume for 3 hours 0% for 7 hours
Premier Tech Environnement
7/16
The effectiveness in the context of new constraints 2009-05-21
The testing protocol is divided into 10 sequences with different sampling frequencies (Table 2).
Table 2.
Testing protocol of Standard EN 12566-3
Sequence
Description
% of design flow rate
Duration (weeks)
Sampling days
1
Start-up period
100
approx. 4
0
2
Design flow
100
6
4
3
Low flow
50
2
2
4
Design flow + power failure (system fed)
100
6
5
5
Vacation
0
2
0
6
Design flow
100
6
3
7
Design + 2-day peak flow
100 + 150 (2-day)
2
2
8
Design flow + 3 days of power failure (system fed)
100
6
5
9
Low flow
50
2
2
10
Design flow
100
6
3
Based on the above review and analysis of the three existing testing standards in North America and Europe, some changes in the testing protocol have been observed over the past 5 years. The testing period for BNQ and European Standards is based on 12 and 9 months, respectively, to better reflect seasonal variations (warm and cold conditions), compared to the 6-month period for the NSF Standard. As a condition for certification renewal, the BNQ Standard has introduced an annual audit to control technology performance not only on the testing platform, but also in the field. Also, as presented at Figure 2, the European Standard has introduced a dosing protocol that better reflects the lifestyle of the majority of families, where both parents are working outside the home. However, considering the new constraints imposed by current lifestyles (more peak flow, etc.), the existing protocols must be revised to ensure that short-term technology evaluations are representative of their long-term performance under actual usage conditions, and that certified technologies will perform not only on the testing platform under controlled conditions, but also in the field.
Premier Tech Environnement
8/16
The effectiveness in the context of new constraints 2009-05-21
30
25
% of daily volume/h
NSF dosing protocol NSF working parent EU dosing protocol
20
Series1 Series2
15
10
5
0 1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 hours
Figure 2. Comparison of dosing protocols
Evolution of testing protocols and standards In an effort to submit wastewater treatment technologies to testing that is more representative of the new constraints, in the past two years, we observed some improvements in testing protocols and standards in both Europe and Canada. New testing protocol in Europe In order to test residential wastewater treatment technologies under conditions that better reflect the conditions observed in individual applications (peak flows, variable flows, etc.) and today’s lifestyles, in 2006-2007, a comparative study was performed by the CSTB (Building Scientific and Technical Center) and monitored by an independent group of experts (Veolia Water, 2007). This group of experts developed a testing protocol that compared and evaluated eight (8) different wastewater treatment technologies on the CSTB testing platform located in Nantes, France. The new protocol is based on a 40-week testing period that includes the following steps: • • • • • • • • •
Four (4) weeks for system start-up (stabilizing of the biological process) at design flow rate; Twelve (12) weeks at design flow rate; Four (4) weeks at design flow rate from Monday to Thursday, and at 2 times design flow rate for the other 3 days. This condition reflects typical high occupancy during week-ends (a frequent condition in secondary homes); Three (3) weeks at 2 times design flow rate (overloading conditions for high-occupancy conditions); Three (3) weeks at zero flow to simulate a vacation period; Two (2) weeks at design flow rate from Monday to Thursday, and at 2 times design flow rate for the other 3 days; Four (4) weeks at design flow rate; Two (2) weeks at 50% of design flow rate (low occupancy conditions); Six (6) weeks at design flow rate with 3 days of power failure (one every 2 weeks).
Premier Tech Environnement
9/16
The effectiveness in the context of new constraints 2009-05-21
During all of the previous periods, the wastewater treatment technologies are fed according to the EN 12566-3 dosing protocol. Phase 1 of this evaluation project began in February 2006 and was completed in November 2006. New testing phases are presently ongoing. The eight (8) wastewater treatment technologies tested were representative of the different treatment processes existing in the market: • • • • • • • •
Sand filter sized according to the French regulation Reduced size sand filter using a gravelless distribution device (plastic channel with geotextile interior) Constructed wetland Zeolith-based media filter Organic-based media biofilter Textile-based media biofilter Suspended growth ATU Attached growth ATU
The capacity of the tested systems corresponds to 5-6 inhabitant equivalents, and 24-hour composite samples were collected twice a week to analyze for TSS, COD, CBOD5, ammonia, and pH at the system inlet (inlet of the septic tank), septic tank effluent and treatment unit effluent. Table 3 provides the raw wastewater characteristics at the inlet of the system showing that the wastewater used was representative of domestic applications with low dilution. According to the protocol, minimum CBOD5 concentrations of 300 mg/L were required to reflect sufficient organic loading on the tested system. Table 4 provides a summary of the results (average ± standard deviation) obtained during Phase 1 of the evaluation project. An analysis of the results provided above shows significant performance differences among the eight types of technology tested when the protocol is closer to actual lifestyles (e.g.: peak flow and load), and sampling is also done during the stress conditions. On average, the organic-based media biofilter is the system that performs best with effluent concentrations of 7 and 5 mg/L for TSS and CBOD5 respectively, and provides the highest stability, with a standard deviation of 3 mg/L. Average results with sand filter effluent were also below 10 mg/L for TSS and CBOD5, but with slightly more variation. It is important to note that the installed sand filter was under drained, and results would likely be different if the sand filter had not been drained before the infiltration of treated effluent into native soil. The draining layer at the bottom of every filter (sand, peat, coconut, foam, etc.) has a major impact on the hydrodynamic conditions of the filter (saturation level, aeration, etc.) and on the filter’s performance. Three other biofiltration technologies achieved average performance, below 15 mg/L for TSS and CBOD5, but with more variation, especially with a reduced size sand filter using a gravelless distribution device. It is interesting to note that the first five technologies are all based on a biofiltration process. Results obtained with ATUs indicated higher average effluent concentrations and significantly less treatment stability. For an ATU with attached growth, average effluent concentrations were in line with the secondary treatment level (TSS = 25 mg/l and BOD5 = 30 mg/L). However, ATU with suspended growth provided a notably lower quality of effluent along with fairly poor stability of the treatment performance. It is interesting to note that previous results obtained under testing conditions representing typical residential flow and real-life load variations are close to the ATU performance reported by Sextone and al (2000) for systems monitored under field conditions. A total of 85 ATU systems were Premier Tech Environnement
10/16
The effectiveness in the context of new constraints 2009-05-21
sampled for BOD5 and TSS. The results obtained indicate that 48% of measured samples exceeded the monthly average of 30 mg/L in TSS, and 69% of BOD5 results exceeded the monthly average of 30 mg/L.
Table 3.
Raw Wastewater Characteristics (at septic tank inlet)
Parameter
Units
Average
Median
Minimum
CBOD5
mg/L
TSS
mg/L
COD
mg/L
Maximum
330*
290
110
780
304*
267
91
1092
722*
645
250
1990
*Average of 42 samples
To illustrate the impact of stress conditions occurring in typical single dwellings with today’s lifestyles (design and peak flows, zero flow period, power outage, etc.) on wastewater treatment performance, Figures 3, 4 and 5 compare the quality (TSS content) of the treated effluent of three different types of technologies to the treated effluent quality produced by the organic-based media biofilter that obtained the best performance during the testing period.
Table 4. Treated Effluent Characteristics during Phase 1 Technology Organic-based biofilter
Mean concentration of effluent* (± standard deviation) TSS (mg/L) CBOD5 (mg/L) COD (mg/L) 7±3 5±3 54 ± 19
Sand filter
7±6
6±4
44 ± 26
Textile-based biofilter
13 ± 11
8±3
59 ± 26
Zeolith Filter
14 ± 9
11 ± 5
85 ± 25
Reduced sand filter size with gravelless distribution system
15 ± 10
13 ± 9
75 ± 42
Attached growth ATU
16 ± 12
19 ± 9
80 ± 34
Suspended growth ATU**
40 ± 30
45 ± 33
189 ± 98
Constructed wetland*** 36 ± 10 60 ± 23 168 ± 51 * Average of 40 samples ** Sludge removal was required after the first 6 months of the operation period. *** The initial system has been replaced by a new version of a constructed wetland (monitoring is ongoing).
Figure 3 presents TSS variations in suspended-growth ATU treated effluent. Under all conditions tested, results clearly show low performance under peak flow conditions and the need for at least 12 weeks of operation at the design flow rate before the process is stabilized (TSS = 25 mg/L). Also, sludge had to be pumped out after only 33 weeks of operation, and it is clear that the treated effluent from such a system (suspended-growth ATU) requires an additional treatment step (filtration) before it is infiltrated into native soil. By comparison, the organic-based media biofilter provided high stability (TSS ≤ 10 mg/L) under all conditions tested.
Premier Tech Environnement
11/16
The effectiveness in the context of new constraints 2009-05-21
Figure 4 provides the TSS effluent concentrations produced by an attached growth ATU. Two different testing periods are of particular significance: 1) 12 weeks are required to stabilize the attached growth ATU. As shown in Figure 4, the TSS concentrations at the ATU effluent gradually decrease to below 25 mg/L after 12 weeks; 3) Power or blower failure conditions (aeration system turns off and system feeding is maintained) had a major impact on the performance of the ATU, with TSS concentrations close to 40 mg/L at the treated effluent. Although the results obtained with attached growth ATUs are better than those for suspended growth ATUs, results clearly show that attached growth ATUs are not well suited for the intermittent flow conditions and variable organic loading observed in secondary or seasonal homes and in an increasing number of permanent homes. Here again, the treated effluent from such systems requires an additional treatment step (filtration) before infiltration in native soil.
Figure 3. Comparison of the Treatment Performance between an Organic-based Biofilter and a Suspended Growth ATU
Premier Tech Environnement
12/16
The effectiveness in the context of new constraints 2009-05-21
Figure 4. Treatment Performance Comparison between an Organic-based Biofilter and an Attached Growth ATU
Finally, Figure 5 provides the treated TSS effluent concentrations from an under drained sand filter. The performance of both types of biofilter is very stable except after the rest period (zero flow conditions) for the sand filter (TSS near 40 mg/L after the zero flow conditions). This result illustrates the main difference between multiple porosity level filtering media, such as organic-based media (e.g.: peat or coconut) or synthetic-based media (e.g.: foam) and single porosity media, such as sand. In fact, sand has only one level of porosity, which corresponds to the space between the grains of sand. Under zero flow conditions, sand media drains gradually and the dryness significantly impacts the survival of the micro-organisms in the media. This explains the decreased performance of the sand filter when the system is fed following a rest period. For multiple porosity level media, water is retained not only between the media particles, but also inside each media grain. Under zero flow conditions, this type of media stays wet, which ensures the survival of sufficient numbers of micro-organisms, providing a consistently high performance under all conditions, even when the system is fed after a rest period. Results show that a biofilter based on multiple porosity levels media (e.g.: peat) is particularly well suited to the intermittent flow conditions observed in secondary or seasonal homes and in an increasing number of permanent homes.
Premier Tech Environnement
13/16
The effectiveness in the context of new constraints 2009-05-21
Figure 5. Comparison of the Treatment Performance between an Organic-based Biofilter and an Under Drained Sand Filter
New BNQ Standard in Canada Early in 2007, some Canadian provincial authorities asked the BNQ to develop a new BNQ Standard regarding the certification of onsite wastewater treatment systems, applicable across all provinces and territories in Canada. A technical committee composed of representatives of different groups (regulators, manufacturers, universities, associations, and customers) from the majority of the 12 provinces and territories held a first meeting in March 2007. The existing Quebec Standard (NQ 3680-910) was the starting point for the committee’s work. By the summer of 2008, a draft version of the new BNQ Standard 3680-600 was published for public hearing for 60 days (BNQ, 2008). The Standard is now completed and has been submitted for approval by the Standard Council of Canada. Approval is expected in the spring of 2009. The new BNQ Standard 3680-600 corresponds to the standard now in existence in Province of Quebec (12- month testing period with the first 6 months corresponding to the ANSI/NSF 40 protocol, including stress tests), but includes some major improvements that better reflect the new constraints imposed by current lifestyles (working parents, etc.) and new environmental requirements: •
During the stress test sequences, 24-hour composite samples are collected on all stress days (not only on the day each stress is initiated, as prescribed in the ANSI/NSF 40 protocol);
•
A change in the dosing pattern for the second 6-month testing period (Annex B) to reflect conditions where both parents are working outside the home: 40% of the daily volume applied in the morning (between 6 a.m. and 9 a.m.) and 60% in the evening (between 5 p.m. and 8 p.m.), five days per week. During the two other days, the system is fed according to the ANSI/NSF 40 dosing pattern (3 periods of 3 hours per day). Then, the
Premier Tech Environnement
14/16
The effectiveness in the context of new constraints 2009-05-21
system is tested under working parent conditions for 27 weeks (1 week during the first 6month testing period and 26 weeks during Annex B) out of 52 (52% of the time as opposed to 4% of the time with Standard ANSI/NSF 40); •
Influent and effluent sampling one day per week for a total of 26 sampling days during the second 6-month testing period (Annex B). At the present time, the existing BNQ Standard 3680-910 requires a minimum of 10 samples over the Annex B testing period;
•
Addition of treatment classes corresponding to current and future needs and related to the dispersal of treated effluent in sensitive areas (disinfection, phosphorus or nitrogen removal). The definitions of the different classes are presented at Table 5 and the standard allows for different combinations of “Basic treatment” (B classes) with other classes (D, P or N).
Table 5. Treatment Classes defined in the new BNQ Standard 3680-600 Treatment classes
Basic treatment (B)
Disinfection (D)
P removal (P)
N removal (N)
TSS*
CBOD5*
Fecal coliforms or E. Coli*
Total P*
Total N
B-I
100
150
B-II
30
25
B-III
15
15
B-IV
10
10
D-I
50,000
D-II
200
D-III
ND (median < 10)
P-I
1.0
P-II
0.3
N-I
50%
N-II
75%
* All values are in mg/L except for fecal coliforms or E. coli, which is measured in counts/100 mL.
Conclusion Over the past 20 to 30 years, we have observed major lifestyle changes that have a significant impact on the wastewater flow from single dwellings (daily flow pattern, peak flow, zero or low flow period, etc.). A review of the existing standards (NSF, BNQ or European) clearly shows that improvements are needed to better reflect current lifestyles and ensure the certification of wastewater treatment systems performing under real-life conditions. We noted that the existing BNQ and European Standards provide some major improvements compared to the NSF Standard (for example, a longer testing period - up to 12 months for the BNQ Standard) to ensure certification of the technology under conditions for all four North American seasons, and a dosing pattern that is closer to current lifestyles (European Standard), where both parents work outside the home in more than 66% of families. In addition, the major study undertaken at the CSTB site in France showed the importance of submitting treatment technologies to more challenging protocols that reflect current lifestyles, in order to make sure that certified technologies perform under the new “real life” conditions. Premier Tech Environnement
15/16
The effectiveness in the context of new constraints 2009-05-21
Finally, the evolution of (changes in) the standard for onsite wastewater treatment system certification is ongoing in Canada and 2009 will see the publication of a new and improved BNQ Standard that will provide the possibility of certifying treatment technologies for more and more challenging environmental conditions, applicable at the national level.
References Bureau de Normalisation du Québec (BNQ), 2005. Wastewater Treatment – Stand-alone treatment systems for isolated single dwelling – Advanced secondary treatment system (class III) - Biofilter Ecoflo® ST-650. Performance report for Annex A. Bureau de Normalisation du Québec (BNQ), 2005. Wastewater Treatment – Stand-alone treatment systems for isolated single dwelling – Advanced secondary treatment system (class III) - Biofilter Ecoflo® ST-650. Performance report for Annex B. Bureau de Normalisation du Québec (BNQ), 2008. Onsite Residential Wastewater Treatment Technologies. Final Draft Standard D 3680-600. CEN/TC 165, Small Wastewater Treatment Systems for up to 50 PT – Part 3: Packaged and/or site assembled domestic wastewater treatment plants, November 2004. Bureau of Labor Statistics, Chart 6-3, 2006 INSEE, Activité, emploi et chômage selon le type de ménage et le nombre d’enfants, 2007. Premier Tech Environment, 2006. Voluntary monitoring program – 1995 to 2006. Sextone, A.,Bissonette G., Flemeing K., Kinner K., hench K., Bozicevitch T., Cooley B. and Winant E. 2000 A survey of home aerobic treatment systems operating in six West Virginia Counties. Small Flows Quaterly, fall 2000,Vol.1, Nb 4, pp. 38-46. Statistics Canada, Population Survey, 2005 and earlier. US Census, Population Profile of United States, Family and Living Arrangements in 2005. U.S. Census Bureau, Current Population Survey, March and Annual Social and Economic Supplements, 2007 and earlier. U.S. Census Bureau, Table A-1, Average Population per Household and Family: 1940-1993. Veolia Water (2007), Etude comparative de 8 filières de traitement: résultats et évaluations. Conference presented in Cahors (France), October 25th, 2007,
Premier Tech Environnement
16/16
The effectiveness in the context of new constraints 2009-05-21
