- Email: [email protected]
Recycled water — A source of potable water: City of San Diego health effects study
~
Pergamon
Wat. Sci. Tech. Vol.33, No. to-11, pp. 285-296, 1996. Copyright © 1996 lAWQ. Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved. 0273-1223/96 $15'00 + 0·00
PH: S0273-1223(96)00431-3
RECYCLED WATER - A SOURCE OF POTABLE WATER: CITY OF SAN DIEGO HEALTH EFFECTS STUDY Adam W. Olivieri*, Don M. Eisenberg*, Robert C. Cooper**, George Tchobanoglolls*** and Paul Gagliardot * Western Consortium for Public Health, 1410 Jackson St., Oakland, CA 94612, USA ** University ofCalijornia, School of Public Health, Warren Hall, Berkeley,
CA 94720, USA *** Department of Civil Environmental Engineering, University ofCalijornia at Davis, Davis, CA 95616, USA t City ofSan Francisco, Water Utilities Department, 600 B Street, San Diego, CA 92101, USA
ABSTRACT During the late 1980's and early 1990's the City of San Diego built and operated a 1,900 m3/d (0.5 Mgal/d) wastewater pilot facility to evaluate the efficacy of a unique combination of treatment methods for the reclamation of wastewater to a quality acceptable for human consumption. The primary objective of the study was to investigate if the City's advanced wastewater treatment system could reliably reduce the contaminants of concern to levels such that the health risks posed by an assumed potable use are no greater than those associated with the present water supply. The results of this substantial research effort indicated that the processes involved can reliably produce a water of equal or better quality than that of the present raw water supply. Copyright © 1996 IAWQ. Published by Elsevier Science Ltd.
KEYWORDS Advanced wastewater treatment; biomonitoring; chemical analyses; enteric viruses; epidemiology; fish mutagenic; genetic toxicity; health effects plant; hyacinth ponds; microbiological agents; reliability; reverse osmosis; risk assessment.
HEALTH EFFECT STUDY OBJECTIVES (HES) The HES was designed to determine possible health risks of direct reuse of reclaimed wastewater (Eisenberg et at., 1987). The primary objective of the Health Effects Study (HES) was to investigate if the City's advanced wastewater treatment system (Aqua IT) could reliably reduce contaminants of public health concern to levels such that the health risks posed by an assumed potable use of the treated water are no greater than those associated with the present water supply. The Western Consortium for Public Heaith entered into a contract with the City of San Diego in July of 1985 to conduct a Heath Effects Study (RES). In order to provide overall guidance on the project to the City, the City of San Diego formed a Technical Advisory Committee (TAC) and a Health Advisory Committee (HAC) made up of representatives from the California Department of Health Services, the University of California at Berkeley, San Francisco, Davis, and Los Angeles, the San Diego Health Department, and the California Water Resources Control Board. The HAC provided guidance and review on the technical aspects of the project and performed the final assessment of the health risks associated with the use of a treated wastewater 285 JWST 33-10/11-K
286
A. W. OLIVIERI et al.
as a potable water supply. The technical work was carried out by staff of the Western Consortium and a group of subcontractors that included professors and staff from the California Universities noted above, San Diego State University, and the virus and microbial laboratories of the California Department of Health Services.
WASTEWATER TREATMENT SYSTEM (AQUA II) The pilot wastewater treatment system (Aqua II) that was located in Mission Valley, San Diego California and was a unique and innovative system, utilizing channels containing water hyacinths for secondary treatment, followed by an advanced wastewater treatment (AWT) system designed to upgrade the secondary effluent water to a quality equivalent to raw water for potable reuse. The total capacity of the water hyacinth ponds was approximately 590 m3/d (156,000 gal/d) and the capacity of the AWT facility was approximately 189 m3/d (50,000 gal/d). The Water Hyacinth System, when operated within appropriate loading parameters, has produced secondary effluent Biological Oxygen Demand (BOD) and Total Suspended Solids (TSS) concentrations consistently below 10 mg/L. Experimentation with the design and operation of the water hyacinth ponds (Tchobanoglous et al., 1987) lead to a system that is: • Able to treat a wastewater containing a high concentration of sulfate without the development of odors or other nuisance conditions; • Able to produce a treated effluent that corresponds to what would be expected from an advanced secondary treatment process; • Extremely stable with respect to effluent quality; • Robust with respect to the ability to withstand plant upsets; and • Simple to operate. A more detailed discussion of the evolution of aquatic portion of the wastewater treatment pilot plant is contained in a separate paper (Tchobanoglous et al., 1987). This effluent serves as the feed water to the Advanced Wastewater Treatment Plant. The AWT plant contains a full range of advanced wastewater and water treatment unit processes, including coagulation, filtration, reverse osmosis, air stripping, carbon absorption, and final disinfection.
STUDY ELEMENTS The study included elements for identifying, characterizing, and quantifying infectious disease agents and potentially-toxic chemicals, and for screening mutagenicity and bioaccumulation of the chemical mixtures present, in both the treated wastewater and the City's untreated potable water supply. The untreated water supply was represented by raw water entering the Miramar water treatment plant. The study also included a reliability analysis, using data from the technical performance evaluation of unit processes in the demonstration water reclamation plant. The evaluation included conducting an epidemiology study to collect baseline data of the San Diego population and conducting a chemical risk assessment on both waters. A summary of the results from each of the study elements is presented below. Infectious Disease Agents Data sufficient to make an informed judgement as to the public health risk of infectious disease from agents in both the reclaimed water and the untreated raw water supply were collected. The infectious agents that were measured included representative bacterial, viral, and parasitic human pathogens.
Microbiological indicators of water quality. There are a number of microbial diseases which can be transmitted via the water source. Routine monitoring for all of the pathogenic microorganisms is, at present
'lfl,7
Recycled water - a source of potable water
time, impractical because: (a) they are present in low numbers relative to the normal aquatic microbial population and (b) present methods for the isolation, identification and enumeration are poorly developed and tedious to perform. Because of these difficulties, the environmental and public health engineers rely upon indicator organisms whose presence can be determined and quantified easily, and whose occurrence can be taken as an indication of the presence of human or animal wastes which have the potential to contain disease• producing microorganisms. Indicator organisms that included total and fecal coliform and fecal streptococci were used for routine water quality monitoring. A statistical summary of the raw data for the number (MPN) of indicator bacteria found in the three water sources is presented in Table 1. As anticipated, the RWW sam pIes yielded the highest total and fecal coliform and fecal streptococci counts of the three TABLE 1. SUMMARY STATISTICS FOR INDICATORS RWW, MIRAMAR AND AQUA II WATER (MPN/l00ML) Indicator
Water Source
Sample Size
Log Avg.
Log S.D.
Min.
Max.
Total Coliform
RWW
212
7.37
0.31
6.11
8.20
MIRAMAR
428
1.51
0.82
0.3
3.7
Aqua II
413
0.4
0.28
0.3
1.9
RWW
212
7.05
0.32
6.32
8.20
MIRAMAR
425
0.5
0.36
0.3
2.48
413
0.3
0
0.3
0.3
RWW
196
6.48
0.34
5.6
7.38
MIRAMAR
425
0.95
0.77
0
3.2
Aqua II
411
0.3
0
0.3
0.3
Fecal Coliform
Aquall Fecal Streptococcus
a
a For statistical
purposes values of less than 2.2 MPNII 00 ml were recorded as 2.2 MPN/IOOml. The majority of the samples were <2.2 MPN/ml.
waters, and were in the range expected for raw domestic wastewater. Both the Aqua IT and Miramar waters contained low coliform and enterococci values. The Aqua II values were consistently lower than those seen in Miramar samples. It should be noted that the results of this study reflect water quality of the Aqua II without disinfection. As shown in Table I, the Aqua II product water has a much higher microbiological quality than Miramar water based on indicator organism data (i.e., total and fecal coliform and fecal streptococci).
Enteric viruses. The potential presence of viruses in wastewater and reclaimed water are of concern to public health officials and other decision makers. For this reason, a great deal of effort was expended during the HES to examine for the presence and numbers of enteric viruses in the influent and effluent of the Aqua II plant (Cooper et al., 1991b). The raw wastewater (RWW) influent to the plant and the final effluent were monitored for virus from June 1987 through April 1990. For purposes of comparison virus sampling was performed on Miramar raw water in parallel with the Aqua II plant sampling. Samples of RWW, Aqua II, and Miramar waters were collected by project staff on site. One liter of composite RWW samples were collected, while in the case of Miramar and Aqua II samples of up to 3,800 L (1000 gallons) were collected and concentrated by passing through parallel spun fiberglass-acrylic filters. All samples were concentrated further using protein precipitation methods (Standard Methods 1985). Viruses were enumerated by plaque assay using Buffalo Green Monkey Kidney cells. Resultant plaques were enumerated and subcultured. During the monitoring period, 203,588 L (53.576 gallons) of Aqua II effluent and 163,506 L (43,028 gallons) of Miramar water were concentrated for viruses. No viruses were detected, in samples as large as
288
A. W. OLIVIERI et al.
3,800 L (l,000 gallons), in either Aqua II product water or the Miramar raw water indicating there is ?o difference in the virus concentration found in the two waters. In the RWW, the percentage of composite samples containing detectable virus was 56; thus, slightly less than half the samples were virus negative. The median virus concentration in RWW was 9 Pfu/L, ranging from nondetected to 17,300 Pfu/L. The number of RWW samples that had no detectable viruses present per liter of wastewater was surprising. A review of the laboratory data indicated that on many occasions the number of virus plaques per liter were found to increase upon dilution of the sample. This observed increase is most likely do to the inhibitory effect associated with the undiluted wastewater.
Enteric bacteriaL pathogens. Each of the three water sources: San Diego raw wastewater (RWW); Aqua II reclaimed water; and the associated community raw drinking water (Miramar) was examined for the presence of the bacterial pathogen SaLmoneLla spp. One liter samples of RWW and one to ten liters of Aqua II and Miramar waters were concentrated by filtration and were examined. Results of the two year monitoring program for SaLmonella spp. are summarized in Table 2. It can be seen that of the three waters TABLE 2. SUMMARY OF SaLmonella spp. MPN RESULTS OBSERVED IN AQUA II, MIRAMAR, AND RWW WATER SOURCES Water Type
Number of Samples
Volume Filtered (L)
RWW
84
0.76
Miramar
55
10.10
Aqua II
51
10.10
a
Salmonella Concentrationa MPNIL
Median values. b Range of values 0 - >28,000 MPN method. e Lowest level of detection using MPN method. a
investigated, only the raw wastewater (84 samples) was found to contain detectable levels of SaLmoneLla. The median value for in situ SaLmonella was 13/L. There was a great deal of variation from sample to sample with a range from less than 3 to >20,000/L. None were detected in either the Aqua II or Miramar water (55 and 51 samples; respectively). These values are similar to those generally found in domestic wastewater in the U.S., which reportedly range from non-detected to 28,000/100 mL.
Enteric Parasites: A large portion of this aspect of the microbial pathogen study was dedicated to methods development. The three test waters were monitored for parasites once every two weeks during the first year and once a week thereafter. Because of the extensive methods development required, not all samples were analyzed using the same procedures. Detailed discussion of the results of the sampling and filtering methods development studies is contained in a separate report (Copper et al., 1991c). Non-pathogenic protozoa and metazoa of various types were present in all waters examined. Generally, there was little difference between the type of organisms identified in the Aqua II and Miramar samples. More ciliates were observed overall in the Aqua IT than in the Miramar, but more algae and metazoa were observed in the Miramar water. There were no Giardia Lamblia cysts identified in either the Aqua II or the Miramar water samples although they were commonly recovered from the raw wastewater. The number of Giardia present was about 300 cysts per liter. Based on the recovery rate for Giardia LambLia cysts from raw wastewater, this would imply that the actual concentration range of Giardia would have been as high as 5000/L of RWW. No Giardia were observed in the Aqua IT product water, indicating that the entire treatment process was capable of achieving a greater than 99.9 percent removal rate. No other enteric parasites of public health significance were observed to be present in the Aqua IT or Miramar water.
Recycled water - a source of potable water
289
Chemical Agents Chemical screening and monitoring of the Aqua II and Miramar waters was carried out over a three year period. The objectives of this element were twofold; one to determine which chemicals are present and their concentrations,and; two to identify which of these chemicals are known to be of public health concern in using the Aqua II water as a municipal raw water source. What follows is a brief summary of the work as reported by Froines et al. (1991).
Results of chemical analysis. Miramar and Aqua II waters were analyzed for inorganics, purgeable organics, extractable organics, pesticides, Polychlorinated Biphenyls (PCBs), and chlorinated dibenzodio• xins/dibenzofurans. Most contaminant levels in the reclaimed water were extremely low; well below conven• tional detection limits. Results of the Inductively Coupled Plasma Atomic Emission Spectroscopy (lCP-AES) analysis indicated that boron, calcium, magnesium, manganese, molybdenum, phosphorus, sodium and strontium occurred in measurable quantities in both Miramar and Aqua II waters. Arsenic was identified in the Miramar water by Atomic Absorption Spectroscopy (AA) analysis at very low concentrations, mean value of 1.5 ug/L and not detected in the Aqua II water. A comparison of the Aqua II and Miramar mean water quality against a range United States surface and groundwater quality is shown in Table 3. Results of the purgeable organic analyses indicated that trihalomethanes (THMs) were the compounds detected at concentrations averaging 10 ug/L in the Miramar water. These were byproducts of upstream pre• chlorination. TABLE 3. COMPARISON OF AQUA II AND MIRAMAR WATER QUALITY WITH U.S. RAW WATER SUPPLIES METALS (mg/L) 3
U.S. b Groundwaters
U.S. Surface b Watersb
c
0.001-0.100
0.001-0.020
c
0.001-0.058
0.001-0.120
0.003
0.0008-0.050
0.0002-0.650
0.009
0.001-0.100
0.0004-0.200
0.002-0.100
0.0002-1.560
0.001-0.130
0.0013-0.500
0.001-0.061
0.0001-0.047
0.0027-1.3767
0.001-8.600
Constituent
Awr
Miramar
Arsenic
O.OO03 c
0.0017
Cadmium
2E-05
c
8E-05
Chromium
0.008
Copper
0.005
Lead
0.0004
Nickel
0.008
Selenium
O.OOOl
Zinc
0.025
c
0.0005
c
0.003 c
0.0007 0.017
c
Notes: 3 Aqua II and Miramar values are arithmetic mean concentrations. Data generated by University of California at Los Angeles during the chemical monitoring phase of the HES project. b U.S. groundwater and surface water values are a range of arithmetic mean concentrations. c Arsenic, cadmium, lead and selenium were analyzed using the AA method. Maximum contaminant level Tests were conducted to identify the THM formation potential in both waters (Standard Methods, 1989). These indicated that the Miramar water had ten times the THM potential of the Aqua IT water. The only extractable organic compounds measured above detection limits in either Aqua II or Miramar water were phthalates: Bis(2-ethylhexyl) phthalate (BEHP), diethyl phthalate and di-n-octyl phthalate. BEHP represents the only significant compound of public health concern and was identified in the Aqua II water at an average concentration of < 0.5 ug/L (geometric mean) or 7.6 ug/L (arithmetic mean) and in Miramar at
N
-.0
c
TABLE 4. AQUA II PERFORMANCE SUMMARY (All units are mg/L unless otherwise noted) Physical 87-89
RAW cone.
Primary Effluent cone.
% removal
Secondary Effluent cone.
% removal
Tertiary Effluent cone.
% removal
CBOD TSS COD TOC TS Turb (NTUI
212 209 427 71 1008 96
142 108 270 56 898 76
33 48 37 21 11 21
9 17 48 16 815 11
63 44 52 56 8 68
NA 7 24 8 804 0.7
Physical 90-92 CBOD TSS COD TOC TS Turb (NTU)
179 211 371 68 1180 69
132 129 282 57 1083 79
26 39 24 16 8 0
11 18 57 17 976 17
68 53 61 59 9 75
NA 1.4 22 6.5 997 0.4
8 9 15 0 24
6 17 11 1 4
9.8 2.9 12.7 172 185
54 0 0 2 2
9 3.6 2.2 169 192
3 0 74 2 0
Nutrients Ammonia - N Nitrate - N Phosphate - P Sulfate Chloride Metals Arsenic Cadmium Chromium Copper Lead Manganese Mercury Nickel Selenium Silver Zinc Boron Calcium Iron Magnesium Sodium 1.
2. 3. 4.
5 6 11 1 11
A WT Effluent cone. NA 2.7 15 1.1 81 NA
% removal
2 2 10
72
NA 1.3 15 1 254 NA
0 2 8 61
1.1 0.6 1.6 3.1 16
32 0 4 94 87
Total Plant
% removal 96 99 96 98 92 99+
94 99+ 96 99
78 99+
:> 24.8 0.12 14.1 177 195
23.3 0.1 12.6 176 188
96 0 89 98 92
~
0 r
<:
m
C! 0.0025 0.0028 0.0167 0.103 0.0293 0.097 0.0012 0.0212 0.0049 0.0078 0.1088 0.263 67.71 0.795 29.78 127.3
0.0026 0.0024 0.0142 0.094 0.0298 0.094 0.001 0.0185 0.0047 0.0057 0.095 0.251 70.52 0.68 29.69 120.8
0 14 15 9 0 3 17 13 4 27 13 5 0 14 0 5
0.0025 0.0018 0.0083 0.029 0.0173 0.06 0.0011 0.0158 0.0025 0.0067 0.0241 0.26 64.4 0.496 27.09 130.7
0 21 35 63 41 35 0 13 45 0 65 0 5 23 9 0
0.0017 0.0015 0.004 0.023 0.0097 0.118 0.001 0.0193 0.0036 0.0031 0.021 0.256 65.25 0.134 29.92 136.1
32 11 26 6 26 0 0 0 0 33 3 0 0 46 0 0
0.0016 0.001 0.0018 0.017 0.0032 0.015 0.001 0.0044 0.0034 0.0045 0.0084 0.23 3.61 0.039 1.62 11.3
4 18 13 6
22 46 0 54 0 0 12 8 90 12 86 86
36 64 89 83 89 85 17 79 31 42 92 13 95 95 95 91
Percent removals are the percentage of the total constituent removal due to the individual treatment. Treatment removals sum to equal the total plant removal Values which evaluate to negative removals are reported as zero. NA: Not Available Physical and Nutrient concentration values are arithmetic means. Any nondetected observations were assumed to be present at the corresponding detectton limit.
Metal concentration values are geometric means determIned through problt analysis.
~
~
Recycled water - a source of potable water
291
<0.5 ug/L (geometric mean) to 4.2 ug/L (arithmetic mean). No PCBs, pesticides or dioxinlfurans were identified above detection limits in either Aqua II or Miramar waters (Table 4). The average total organic carbon (TOC) concentration found in Aqua II water was 1.37 ug/L, while TOC concentration averaged 9.83 ug/L in Miramar water.
Results of genetic toxicity testing. Genetic toxicity testing was used as a cost effective bioassay for screening the relative toxicity of the two waters being investigated. Four separate types of widely used bioassay systems were employed to assess genetic toxicity and potential cancer-eausing effects of the Aqua II and the Miramar waters. These included the Ames Assay, Micronucleus Test, 6-Thioguanine Resistance Assay, and Cellular Transformation Assay. Mutagenic and genotoxic activity has frequently been observed in studies of source and treated drinking waters. Consistent with these observations, organic extracts from both the AWT and the Miramar water sources exhibit some mutagenic activity. In addition, some genotoxic activity may also be associated with fractions which were not efficiently extracted by the resin columns used in these studies. The genotoxic effects detected in this study were observed primarily in the Ames test but some indications of potential mutagenic activity were also observed in the other bioassays. Based on the Ames test data, the tested Miramar extracts exhibited somewhat higher mutagenic activity than the comparable Aqua II extracts. The reason for this difference in observed mutagenicity is unknown but may reflect differences in composition of the original source waters or the result of the use of disinfectants on Miramar water. These data from short-term bioassay results of organic extracts of both Aqua II and Miramar water indicate that water from the Aqua II facility appears to exhibit less genotoxic or mutagenic activity than the low levels observed in water from the raw water entering the Miramar water treatment plant. GeMS identification of chemical substances. Synthetic resins were used to concentrate trace organics from
Aqua II, Miramar and reagent (blank) water. A total of 71 resin extracts were analyzed by gas chromatography-mass spectrometry in an effort to characterize the component mixture. The resin extract for each sample was fractionated by High Performance Liquid Chromatography (HPLC) into four fractions with one of the fractions fractionated again into three subfractions resulting in a total of six fractions for each sample. Analysis revealed the presence of many classes of organic compounds in low to trace amounts in the waters. Several herbicides were found in four of the five Miramar water samples. Only one compound found in the monthly Miramar water samples, 2-hexenal, was predicted to be an active mutagen and reported positive in the Ames testing. Fish Biomonitoring To complement information from other HES activities, fish biomonitoring experiments were expected to provide information on chronic exposure to trace contaminants contained in the effluent from the Aqua II plant and the untreated raw city water that accumulate in tissue but are not known to be active in genetic toxicity screening bioassays (de Peyster et al., 1991). Juvenile fathead minnows (Pimephales promelas) were exposed to Aqua II water, Miramar water, or charcoal filtered San Diego tap water (acclimation or control water) in flow-through aquaria. Three 28-day bioaccumulation experiments were completed at each site under standardized conditions following EPA and ASTM bioconcentration guidelines. In addition, swim speed and optomotor response tests were conducted. A total of 120 fish tissue samples and 60 aquarium water samples were collected for analysis of 69 base/neutral/acid extractable organics, 27 pesticides/PCBs, and 26 inorganics by investigators at theUniversity of California at Los Angeles (UCLA). Quadruplicate samples of fish tissue and either duplicate or triplicate samples of water were collected and analyzed for each sampling time (Days 0, 7, 14, and 28 of the 28-day experiments). The concentrations of most analytes in these samples were below the detection limit. Statistical comparisons were made only if at least 75 % of the samples exceeded the detection limit for a given analyte. Chemical analysis of fish tissue and water in the 28-day bioconcentration experiments revealed no statistically significant differences or between the fish exposed to Aqua II water and those exposed to Miramar water. The fish also displayed no significant difference in survival, growth, or swim speed after up to 28 days of continuous whole body exposure to test waters. This was a consistent finding for all three tests
292
A. W. OLIVIERI e/ al.
conducted on each of the three waters. Plant Reliability Evaluating the risk associated with the AWT plant effluent includes considering the ability of the treatment system to consistently achieve the level of treatment that was used as the basis for the health risk assessment. A reliability analysis was conducted to determine the reliability of the mechanical systems, characterize the treatment process variability, and present the probabilities of effluent characteristics meeting specific observed concentrations. To address these objectives, the reliability analysis included the review and analysis of plant performance and mechanical systems data. Summary statistics of the water quality data show.:d very effective treatment, with virtually complete removal of biological indicator organisms, and removals on the order of 87 to 98 % for general parameters, "conventional" pollutants, and for most inorganic ions (Table 5). The treatment plant was slightly less effective at removing metal ions, with observed removals on the order of 74 to 91 %. The results are contained and discussed in detail in the final HES report (Cooper et al., 1992a) and a separate plant reliability report (Cooper et al., 1992b). The levels of treatment obtained at the Aqua II plant were sufficient to meet all existing drinking water standards for all of the data reviewed in this study. The results of the statistical analysis and the mechanical reliability analysis indicate that compliance with Maximum Contaminant Levels (MCLs) can be maintained at the Aqua II demonstration AWT plant, on a long-term basis, with low probability of exceeding MeL concentrations. Results of the mechanical reliability analysis indicate that, over a two and one-half year period, the critical equipment are operational nearly 100% of the time. In addition, the m~chanical reliability analysis confirms statistically what is known practically about the Aqua II Plant: observed equipment failures do not cause a significant interruption in operation of the plant and impact on the effluent quality produced. Chemical and Virus Seeding Studies Chemical and virus seeding studies were conducted at the Aqua II plant. The Aqua II plant was spiked with both organic and inorganic constituents that could potentially enter a wastewater treatment plant as a large pulse flow. A summary of the results of this work is shown in Table 6. The ponds removed 60-99% of the inorganic and organic seed, and the advanced treatment processes removed >98.5 - >99.9% of the seed. Overall, the Aqua II plant removed >99.5 - >99.997% of the seeded chemical constituents (Froines et at., 1991). Four virus seeding studies performed on the Aqua II plant indicated a very high virus removal capability. As shown in Table 7, none of the seeded virus (attenuated vaccine strain Poliovirus 2) were recovered in the final effluent indicating a removal efficiency of as great as eight orders of magnitude (99.999999%). This removal rate is considerably greater than that reported in either the Pamona virus study (SDL, 1977) the Monterey study (Engineering Science, 1987) which used filtration and chlorination of secondary effluent but not reverse osmosis without final disinfection. Epidemiology The principal purpose of the epidemiology component was to develop a baseline of pertinent morbidity and mortality data for the City of San Diego so that a basis of comparison is available if direct potable reuse of municipal wastewater becomes a reality. The health of the residents was evaluated by characterizing thereproductive health and analyzing vital statistics of San Diego County. In addition, a neural tube defects survey was preformed to establish baseline prevalence rates in California and in San Diego. The epidemiological component provided baseline data to facilitate monitoring for actual health effects if the city decides to proceed with potable reuse of the type of treated effluent studied in this project (Molg~d et at., 1990).
TABLE 5. PLANT PERFORMANCE STATISTICS FOR MECHANICAL RELIABILITY SUMMARIZED BY PLANT UNIT Number of Occurances (Sum=M)
Number of Failures (Sum =R)
Mean Time to Repair MTTR(hr)
Mean Time Between Failures MTBF(yr)
Overall 90% CL Limit (yr)
Operating Availability AVO
Availa bility AVI
Headworks
393
28
13.6
0.352
0.280
0.9920
0.9956
Primary System
18
13
5.1
0.924
0.666
0.9977
0.9994
Secondary System
423
54
2.2
1.294
1.093
0.9975
0.9998
Operation Area
Inherent
~ (") '< (")
Package Plant
134
106
1.0
0.377
0.333
0.9989
0.9997
Ultraviolet I
143
8
6.7
5.362
3.577
0.9988
0.9999
Reverse Osmosis
414
144
2.2
0.393
0.354
0.9970
0.9993
Ultraviolet II
422
7
4.4
7.740
5.043
0.9969
0.9999
Aeration Tower
56
55
0.4
0.136
0.115
0.9989
0.9996
~G
Carbon Tower
13
9
l.l
1.304
0.889
0.9999
0.9999
"'Cl 0
Product Water
79
78
0.5
0.177
0.153
0.9996
0.9997
~
Caltrans. Equip.
32
29
1.3
0.396
0.315
0.9995
0.9996
G
Co ~
~ 00;
I ~ (II
0
0
....,
g. ~
~
~
a Mean Time to Repair: Number of calendar hours (amount of downtime) during which repairs due to failures were being made. b Mean Time Between Failures: Based on the chi-distribution such that MTBF = (2l/x 2 (0.5;2r+2». MTBF provides an estimate of the mean operating time between failures of all equipment in the unit process. T = total operating years (versus years in service) of all equipment. Example: if MTBF = 4 years for unit process A for which there are three pieces of equipment which operate 60% of the time, then in actual time, one can expect a failure to occur in 2.22 years. cOverall 90% Confidence Limit: The true value of MTBF is at least as great as the calculated 90% confidence limit. d Operating Availability: Calculated from the operating history of each unit, and is a measure of the fraction of the calendar time that the unit was operating. e Inherent Availability: A measure of the fraction of time that the component can be expected to be operational, excluding preventive maintenance downtime.
~
~
N
~
-l'>-
TABLE 6. CHEMICAL REMOVAL EFFICIENCY BY HYACINTH PONDS AND THE AQUA II FACILITY - SPIKE STUDY Spike
Pond Effluent
% Removal
(~gIL)
(~gIL)
by pond
Chromium
182
74
Tetrachloroethylene
900
Tetrachlorethane
Compound
Aqua II Effluent
% Removal by
Overall
(~)
Aqua IIa
59.3
<1
>98.6
>99.5
29
96.8
<0.3
99.9
>99.997
224
53
76.3
<.1
>99.8
>99.996
Trichlorobenzene
813
130
84.0
<.3
>99.8
>99.996
Tetrachlorobenzene
140
2
98.6
<.03
>98.5
>99.98
Lindane
70
12
82.9
<.03
>99.7
>99.96
a Tertiary treatment processes including package plant (coagulation/filtration), reverse osmosis, aeration tower, and carbon tower
TABLE 7. VIRUS REMOVAL EFFICIENCY OF THE AQUA II WATER RECLAMATION PLANT
» ~
ol ' =2
m CZ ~
Virus in Pond Effluent
~
Run Number
Expecteda (pfu/L)
Observed (pfu/L)
% Virus Removed in Ponds
Virus in Aqua II Effluent (pfu/L)
PUS-lu
4.7 X 104
1.9 X 102
99.6
NDb
>99.9999
PUS-2
9.6 X 104
1.2 X 104
87.5
NDc
>99.99999
PUS-3
1.0 X 10 5
4.0 X 103
96.1
ND
>99.99999
PUS-4
5.6 X 104
3.7 X 103
93.5
ND
>99.99999
• The number expected if no virus removal occurred (based on Lithium Tracer study). ND = none detected in 3800 liters. The occurrences of 0.2 pfuIL in this sample was due to cross contamination. d Calculation based upon 3800 liter sample.
b
C
Percent Virus Removal d
Recycled water - a source of potable water
295
Chemical Risk Assessment The chemical risk estimates were driven by Arsenic and THMs present in the Miramar water (THMs apparently from upstream chlorination of the raw water supply), and by bis-ethyl hexyl phthalate (BEHP) present in the Aqua II water. Arsenic accounts for approximately 80 % and THMs account for approximately 20 % of the risk estimate in Miramar water. The concentration of arsenic in Miramar water is well within applicable federal and state standards, approximately 2 % of the existing drinking water standards. The concentration of THMs in the Miramar water is 10% of the existing drinking water standards. The water from the Aqua II system was analyzed for arsenic and does not have enough arsenic to be reliably detected. Risk calculations, using EPA upper confidence level estimates, indicate that the Aqua II water, if used directly as potable water, would represent an estimated lifetime risk of 3.2 X 10-6 (.03 excess cancers/lO,ooo people) versus 1.2 X 10-4 (1.2 excess cancers/l0,000 people) for Miramar water. These estimates indicate that the Aqua II risk estimate is approximately 40 times less than the estimated risk associated with the untreated Miramar water. Results of the noncarcinogenic risk indicated that, for both waters, non• carcinogenic chemicals at observed low concentrations would not be anticipated to result in any significant health risk.
HEALTH ADVISORY COMMITTEE CONCLUSION Based on the above research, the overall conclusion reached by the Health Advisory Committee was that the health risk associated with the use of the Aqua II water as a raw water supply is less than or equal to that of the use of the existing raw water supply as represented by the Miramar source (Cooper et a/., 1992a).
FUTURE INVESTIGATIONS The positive performance and operational results obtained from the Aqua II pilot treatment facility along with the results of the health effects study provided the City with the necessary technical support to proceed with 3 the construction of a full scale (Aqua Ill), 3,800 m /d (1.0 Mgal/d), treatment facility. Aqua III is located in the San Pasqual area of San Diego County, California and includes the following unit operations and process: headworks consisting of bar screens and grit removal units; primary treatment consisting of rotary drum filters and a screw press for additional solids dewatering; aquatic treatment consisting of 24 water hyacinth ponds, operating in a step feed mode with aeration and effluent recirculation; filtration facilities consisting of chemical feed systems for ferric chloride and hydrated lime, a solids contact clarifier, and dual media filters; and advanced water treatment facilities consisting of ultraviolet disinfection, reverse osmosis, air stripping, carbon adsorption, and final disinfection with sodium hypochlorite. Starting in January 1995 these facilities were placed in operation with the intent of investigating and documenting full scale plant performance and reliability. Various health effects studies are also underway to verify pilot plant results. At the present time the treated water from Aqua m is being used for landscape irrigation. Future plans call for use of the treated water to recharge the aquifer underlying the Hodges Basin. The San diego region is also actively considering potable reuse of AWT treated effluent by discharge to a reservoir at a scale greater than 76,000 m3/d (20 Mgal/d). Feasibility studies have been completed and conditional approvals have been obtained from health authorities. Any decision to actually implement such a project will be subject to additional studies, public acceptance, and cost evaluations.
ACKNOWLEDGEMENT The authors wish to acknowledge the assistance of the treatment plant operations staff, in particular Jack
296
A. W. OLIVIERI et al.
Swerlein and Abel Hernandez, and the City of San Diego Laboratory staff, specifically John Chappin, for their invaluable help in making this project successful.
REFERENCES Cooper, RC., Olivieri, A.W., Eisenberg, D.M., Danielson, RE., Donohue, R, and Pettegrew L.A. (1991a). "Microbiological Indicators of Water Quality." Western Consortium For Public Health, Berkeley, CA. Cooper, RC., Olivieri, A.W., Eisenberg, D.M., Danielson, RE., Donohue, R, and Pettegrew L.A. (1991b). "Enteric Viruses" Western Consortium For Public Health, Berkeley, CA. Cooper, RC., Olivieri, A.W., Eisenberg, D.M., Danielson, RE., Donohue, R, and Pettegrew L.A. (1991c). "Bacterial Pathogens and Parasites." Western Consortium For Public Health, Berkeley, CA. Cooper, R.C., Olivieri, A.W., Eisenberg, D.M., Pettegrew L.A., Danielson, RE., Fairchild, W.A., and Sanchez, L.A. (1992a). "The City of San Diego Total Resource Recovery Project: Health Effects Study." Western Consortiumfor Public Health, Berkeley, CA. Cooper, RC., Olivieri, A.W., Eisenberg, D.M., Pettegrew L.A., Danielson, R E., Fairchild, W.A., and Sanchez, L.A. (1992b). "San Diego Aqua II Pilot Plant Reliability Report." Western Consortium for Public Health, Berkeley, CA. de Peyster, A, Rudnicki, R, Slymen, D.J., and Kimball, A., (1991). Biomonitoring Program - Final Report." Graduate School of Public Health, San Diego State University, San Diego, CA. Eisenberg, D.M., Olivieri, A.W., and Cooper RC. (1987) "Evaluation of Potential Health Risk of Direct Potable Reuse of Reclaimed Wastewater." Proc., Environmental Engineering Specialty Conference, Orlando, FL. Engineering Science (1987). "Monterey Wastewater Reclamation Study for Agriculture" Berkeley, CA. Froines, J. (1991). "Evaluation of Potential Health Risks Presented by Chemical Agents Associated with the San Diego Total Resource Recovery Program - Final Report." University of California, School of Public Health, Los Angeles, CA. Molgaard, C.A., Golbeck, A.L., and Elder, J.P. (1990). "The Epidemiology of Human Reproduction in San Diego County - Final Report." Graduate School of Public Health, San Diego State University, San Diego, CA. Sanitation Districts of Los Angeles (1991). "Pomona Virus Study - Final Report." Los Angeles, CA. Standard Methods for the Examination of Water and Wastewater. (1985). American Public Health Association, Washington, D. C. Standard Methods for the Examination of Water and Wastewater. (1989). American Public Health Association, Washington, D. C. Tchobanoglous, G.T., Maitski, F., Thompson, K., and Chadwick, T.H. (1987). "Evolution and Performance of City of San Diego Pilot Scale Aquatic Wastewater Treatment System Using Water Hyacinths" Proc., 60th Annual Conference of the WPCF, Philadelphia, PA. U.S. EPA. (1982). "Evaluation and Documentation of Mechanical Reliability of Conventional Wastewater Treatment Plant Components". Center for Envir. Res. Inform., Cincinnati, Ohio.
Pergamon
Wat. Sci. Tech. Vol.33, No. to-11, pp. 285-296, 1996. Copyright © 1996 lAWQ. Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved. 0273-1223/96 $15'00 + 0·00
PH: S0273-1223(96)00431-3
RECYCLED WATER - A SOURCE OF POTABLE WATER: CITY OF SAN DIEGO HEALTH EFFECTS STUDY Adam W. Olivieri*, Don M. Eisenberg*, Robert C. Cooper**, George Tchobanoglolls*** and Paul Gagliardot * Western Consortium for Public Health, 1410 Jackson St., Oakland, CA 94612, USA ** University ofCalijornia, School of Public Health, Warren Hall, Berkeley,
CA 94720, USA *** Department of Civil Environmental Engineering, University ofCalijornia at Davis, Davis, CA 95616, USA t City ofSan Francisco, Water Utilities Department, 600 B Street, San Diego, CA 92101, USA
ABSTRACT During the late 1980's and early 1990's the City of San Diego built and operated a 1,900 m3/d (0.5 Mgal/d) wastewater pilot facility to evaluate the efficacy of a unique combination of treatment methods for the reclamation of wastewater to a quality acceptable for human consumption. The primary objective of the study was to investigate if the City's advanced wastewater treatment system could reliably reduce the contaminants of concern to levels such that the health risks posed by an assumed potable use are no greater than those associated with the present water supply. The results of this substantial research effort indicated that the processes involved can reliably produce a water of equal or better quality than that of the present raw water supply. Copyright © 1996 IAWQ. Published by Elsevier Science Ltd.
KEYWORDS Advanced wastewater treatment; biomonitoring; chemical analyses; enteric viruses; epidemiology; fish mutagenic; genetic toxicity; health effects plant; hyacinth ponds; microbiological agents; reliability; reverse osmosis; risk assessment.
HEALTH EFFECT STUDY OBJECTIVES (HES) The HES was designed to determine possible health risks of direct reuse of reclaimed wastewater (Eisenberg et at., 1987). The primary objective of the Health Effects Study (HES) was to investigate if the City's advanced wastewater treatment system (Aqua IT) could reliably reduce contaminants of public health concern to levels such that the health risks posed by an assumed potable use of the treated water are no greater than those associated with the present water supply. The Western Consortium for Public Heaith entered into a contract with the City of San Diego in July of 1985 to conduct a Heath Effects Study (RES). In order to provide overall guidance on the project to the City, the City of San Diego formed a Technical Advisory Committee (TAC) and a Health Advisory Committee (HAC) made up of representatives from the California Department of Health Services, the University of California at Berkeley, San Francisco, Davis, and Los Angeles, the San Diego Health Department, and the California Water Resources Control Board. The HAC provided guidance and review on the technical aspects of the project and performed the final assessment of the health risks associated with the use of a treated wastewater 285 JWST 33-10/11-K
286
A. W. OLIVIERI et al.
as a potable water supply. The technical work was carried out by staff of the Western Consortium and a group of subcontractors that included professors and staff from the California Universities noted above, San Diego State University, and the virus and microbial laboratories of the California Department of Health Services.
WASTEWATER TREATMENT SYSTEM (AQUA II) The pilot wastewater treatment system (Aqua II) that was located in Mission Valley, San Diego California and was a unique and innovative system, utilizing channels containing water hyacinths for secondary treatment, followed by an advanced wastewater treatment (AWT) system designed to upgrade the secondary effluent water to a quality equivalent to raw water for potable reuse. The total capacity of the water hyacinth ponds was approximately 590 m3/d (156,000 gal/d) and the capacity of the AWT facility was approximately 189 m3/d (50,000 gal/d). The Water Hyacinth System, when operated within appropriate loading parameters, has produced secondary effluent Biological Oxygen Demand (BOD) and Total Suspended Solids (TSS) concentrations consistently below 10 mg/L. Experimentation with the design and operation of the water hyacinth ponds (Tchobanoglous et al., 1987) lead to a system that is: • Able to treat a wastewater containing a high concentration of sulfate without the development of odors or other nuisance conditions; • Able to produce a treated effluent that corresponds to what would be expected from an advanced secondary treatment process; • Extremely stable with respect to effluent quality; • Robust with respect to the ability to withstand plant upsets; and • Simple to operate. A more detailed discussion of the evolution of aquatic portion of the wastewater treatment pilot plant is contained in a separate paper (Tchobanoglous et al., 1987). This effluent serves as the feed water to the Advanced Wastewater Treatment Plant. The AWT plant contains a full range of advanced wastewater and water treatment unit processes, including coagulation, filtration, reverse osmosis, air stripping, carbon absorption, and final disinfection.
STUDY ELEMENTS The study included elements for identifying, characterizing, and quantifying infectious disease agents and potentially-toxic chemicals, and for screening mutagenicity and bioaccumulation of the chemical mixtures present, in both the treated wastewater and the City's untreated potable water supply. The untreated water supply was represented by raw water entering the Miramar water treatment plant. The study also included a reliability analysis, using data from the technical performance evaluation of unit processes in the demonstration water reclamation plant. The evaluation included conducting an epidemiology study to collect baseline data of the San Diego population and conducting a chemical risk assessment on both waters. A summary of the results from each of the study elements is presented below. Infectious Disease Agents Data sufficient to make an informed judgement as to the public health risk of infectious disease from agents in both the reclaimed water and the untreated raw water supply were collected. The infectious agents that were measured included representative bacterial, viral, and parasitic human pathogens.
Microbiological indicators of water quality. There are a number of microbial diseases which can be transmitted via the water source. Routine monitoring for all of the pathogenic microorganisms is, at present
'lfl,7
Recycled water - a source of potable water
time, impractical because: (a) they are present in low numbers relative to the normal aquatic microbial population and (b) present methods for the isolation, identification and enumeration are poorly developed and tedious to perform. Because of these difficulties, the environmental and public health engineers rely upon indicator organisms whose presence can be determined and quantified easily, and whose occurrence can be taken as an indication of the presence of human or animal wastes which have the potential to contain disease• producing microorganisms. Indicator organisms that included total and fecal coliform and fecal streptococci were used for routine water quality monitoring. A statistical summary of the raw data for the number (MPN) of indicator bacteria found in the three water sources is presented in Table 1. As anticipated, the RWW sam pIes yielded the highest total and fecal coliform and fecal streptococci counts of the three TABLE 1. SUMMARY STATISTICS FOR INDICATORS RWW, MIRAMAR AND AQUA II WATER (MPN/l00ML) Indicator
Water Source
Sample Size
Log Avg.
Log S.D.
Min.
Max.
Total Coliform
RWW
212
7.37
0.31
6.11
8.20
MIRAMAR
428
1.51
0.82
0.3
3.7
Aqua II
413
0.4
0.28
0.3
1.9
RWW
212
7.05
0.32
6.32
8.20
MIRAMAR
425
0.5
0.36
0.3
2.48
413
0.3
0
0.3
0.3
RWW
196
6.48
0.34
5.6
7.38
MIRAMAR
425
0.95
0.77
0
3.2
Aqua II
411
0.3
0
0.3
0.3
Fecal Coliform
Aquall Fecal Streptococcus
a
a For statistical
purposes values of less than 2.2 MPNII 00 ml were recorded as 2.2 MPN/IOOml. The majority of the samples were <2.2 MPN/ml.
waters, and were in the range expected for raw domestic wastewater. Both the Aqua IT and Miramar waters contained low coliform and enterococci values. The Aqua II values were consistently lower than those seen in Miramar samples. It should be noted that the results of this study reflect water quality of the Aqua II without disinfection. As shown in Table I, the Aqua II product water has a much higher microbiological quality than Miramar water based on indicator organism data (i.e., total and fecal coliform and fecal streptococci).
Enteric viruses. The potential presence of viruses in wastewater and reclaimed water are of concern to public health officials and other decision makers. For this reason, a great deal of effort was expended during the HES to examine for the presence and numbers of enteric viruses in the influent and effluent of the Aqua II plant (Cooper et al., 1991b). The raw wastewater (RWW) influent to the plant and the final effluent were monitored for virus from June 1987 through April 1990. For purposes of comparison virus sampling was performed on Miramar raw water in parallel with the Aqua II plant sampling. Samples of RWW, Aqua II, and Miramar waters were collected by project staff on site. One liter of composite RWW samples were collected, while in the case of Miramar and Aqua II samples of up to 3,800 L (1000 gallons) were collected and concentrated by passing through parallel spun fiberglass-acrylic filters. All samples were concentrated further using protein precipitation methods (Standard Methods 1985). Viruses were enumerated by plaque assay using Buffalo Green Monkey Kidney cells. Resultant plaques were enumerated and subcultured. During the monitoring period, 203,588 L (53.576 gallons) of Aqua II effluent and 163,506 L (43,028 gallons) of Miramar water were concentrated for viruses. No viruses were detected, in samples as large as
288
A. W. OLIVIERI et al.
3,800 L (l,000 gallons), in either Aqua II product water or the Miramar raw water indicating there is ?o difference in the virus concentration found in the two waters. In the RWW, the percentage of composite samples containing detectable virus was 56; thus, slightly less than half the samples were virus negative. The median virus concentration in RWW was 9 Pfu/L, ranging from nondetected to 17,300 Pfu/L. The number of RWW samples that had no detectable viruses present per liter of wastewater was surprising. A review of the laboratory data indicated that on many occasions the number of virus plaques per liter were found to increase upon dilution of the sample. This observed increase is most likely do to the inhibitory effect associated with the undiluted wastewater.
Enteric bacteriaL pathogens. Each of the three water sources: San Diego raw wastewater (RWW); Aqua II reclaimed water; and the associated community raw drinking water (Miramar) was examined for the presence of the bacterial pathogen SaLmoneLla spp. One liter samples of RWW and one to ten liters of Aqua II and Miramar waters were concentrated by filtration and were examined. Results of the two year monitoring program for SaLmonella spp. are summarized in Table 2. It can be seen that of the three waters TABLE 2. SUMMARY OF SaLmonella spp. MPN RESULTS OBSERVED IN AQUA II, MIRAMAR, AND RWW WATER SOURCES Water Type
Number of Samples
Volume Filtered (L)
RWW
84
0.76
Miramar
55
10.10
Aqua II
51
10.10
a
Salmonella Concentrationa MPNIL
Median values. b Range of values 0 - >28,000 MPN method. e Lowest level of detection using MPN method. a
investigated, only the raw wastewater (84 samples) was found to contain detectable levels of SaLmoneLla. The median value for in situ SaLmonella was 13/L. There was a great deal of variation from sample to sample with a range from less than 3 to >20,000/L. None were detected in either the Aqua II or Miramar water (55 and 51 samples; respectively). These values are similar to those generally found in domestic wastewater in the U.S., which reportedly range from non-detected to 28,000/100 mL.
Enteric Parasites: A large portion of this aspect of the microbial pathogen study was dedicated to methods development. The three test waters were monitored for parasites once every two weeks during the first year and once a week thereafter. Because of the extensive methods development required, not all samples were analyzed using the same procedures. Detailed discussion of the results of the sampling and filtering methods development studies is contained in a separate report (Copper et al., 1991c). Non-pathogenic protozoa and metazoa of various types were present in all waters examined. Generally, there was little difference between the type of organisms identified in the Aqua II and Miramar samples. More ciliates were observed overall in the Aqua IT than in the Miramar, but more algae and metazoa were observed in the Miramar water. There were no Giardia Lamblia cysts identified in either the Aqua II or the Miramar water samples although they were commonly recovered from the raw wastewater. The number of Giardia present was about 300 cysts per liter. Based on the recovery rate for Giardia LambLia cysts from raw wastewater, this would imply that the actual concentration range of Giardia would have been as high as 5000/L of RWW. No Giardia were observed in the Aqua IT product water, indicating that the entire treatment process was capable of achieving a greater than 99.9 percent removal rate. No other enteric parasites of public health significance were observed to be present in the Aqua IT or Miramar water.
Recycled water - a source of potable water
289
Chemical Agents Chemical screening and monitoring of the Aqua II and Miramar waters was carried out over a three year period. The objectives of this element were twofold; one to determine which chemicals are present and their concentrations,and; two to identify which of these chemicals are known to be of public health concern in using the Aqua II water as a municipal raw water source. What follows is a brief summary of the work as reported by Froines et al. (1991).
Results of chemical analysis. Miramar and Aqua II waters were analyzed for inorganics, purgeable organics, extractable organics, pesticides, Polychlorinated Biphenyls (PCBs), and chlorinated dibenzodio• xins/dibenzofurans. Most contaminant levels in the reclaimed water were extremely low; well below conven• tional detection limits. Results of the Inductively Coupled Plasma Atomic Emission Spectroscopy (lCP-AES) analysis indicated that boron, calcium, magnesium, manganese, molybdenum, phosphorus, sodium and strontium occurred in measurable quantities in both Miramar and Aqua II waters. Arsenic was identified in the Miramar water by Atomic Absorption Spectroscopy (AA) analysis at very low concentrations, mean value of 1.5 ug/L and not detected in the Aqua II water. A comparison of the Aqua II and Miramar mean water quality against a range United States surface and groundwater quality is shown in Table 3. Results of the purgeable organic analyses indicated that trihalomethanes (THMs) were the compounds detected at concentrations averaging 10 ug/L in the Miramar water. These were byproducts of upstream pre• chlorination. TABLE 3. COMPARISON OF AQUA II AND MIRAMAR WATER QUALITY WITH U.S. RAW WATER SUPPLIES METALS (mg/L) 3
U.S. b Groundwaters
U.S. Surface b Watersb
c
0.001-0.100
0.001-0.020
c
0.001-0.058
0.001-0.120
0.003
0.0008-0.050
0.0002-0.650
0.009
0.001-0.100
0.0004-0.200
0.002-0.100
0.0002-1.560
0.001-0.130
0.0013-0.500
0.001-0.061
0.0001-0.047
0.0027-1.3767
0.001-8.600
Constituent
Awr
Miramar
Arsenic
O.OO03 c
0.0017
Cadmium
2E-05
c
8E-05
Chromium
0.008
Copper
0.005
Lead
0.0004
Nickel
0.008
Selenium
O.OOOl
Zinc
0.025
c
0.0005
c
0.003 c
0.0007 0.017
c
Notes: 3 Aqua II and Miramar values are arithmetic mean concentrations. Data generated by University of California at Los Angeles during the chemical monitoring phase of the HES project. b U.S. groundwater and surface water values are a range of arithmetic mean concentrations. c Arsenic, cadmium, lead and selenium were analyzed using the AA method. Maximum contaminant level Tests were conducted to identify the THM formation potential in both waters (Standard Methods, 1989). These indicated that the Miramar water had ten times the THM potential of the Aqua IT water. The only extractable organic compounds measured above detection limits in either Aqua II or Miramar water were phthalates: Bis(2-ethylhexyl) phthalate (BEHP), diethyl phthalate and di-n-octyl phthalate. BEHP represents the only significant compound of public health concern and was identified in the Aqua II water at an average concentration of < 0.5 ug/L (geometric mean) or 7.6 ug/L (arithmetic mean) and in Miramar at
N
-.0
c
TABLE 4. AQUA II PERFORMANCE SUMMARY (All units are mg/L unless otherwise noted) Physical 87-89
RAW cone.
Primary Effluent cone.
% removal
Secondary Effluent cone.
% removal
Tertiary Effluent cone.
% removal
CBOD TSS COD TOC TS Turb (NTUI
212 209 427 71 1008 96
142 108 270 56 898 76
33 48 37 21 11 21
9 17 48 16 815 11
63 44 52 56 8 68
NA 7 24 8 804 0.7
Physical 90-92 CBOD TSS COD TOC TS Turb (NTU)
179 211 371 68 1180 69
132 129 282 57 1083 79
26 39 24 16 8 0
11 18 57 17 976 17
68 53 61 59 9 75
NA 1.4 22 6.5 997 0.4
8 9 15 0 24
6 17 11 1 4
9.8 2.9 12.7 172 185
54 0 0 2 2
9 3.6 2.2 169 192
3 0 74 2 0
Nutrients Ammonia - N Nitrate - N Phosphate - P Sulfate Chloride Metals Arsenic Cadmium Chromium Copper Lead Manganese Mercury Nickel Selenium Silver Zinc Boron Calcium Iron Magnesium Sodium 1.
2. 3. 4.
5 6 11 1 11
A WT Effluent cone. NA 2.7 15 1.1 81 NA
% removal
2 2 10
72
NA 1.3 15 1 254 NA
0 2 8 61
1.1 0.6 1.6 3.1 16
32 0 4 94 87
Total Plant
% removal 96 99 96 98 92 99+
94 99+ 96 99
78 99+
:> 24.8 0.12 14.1 177 195
23.3 0.1 12.6 176 188
96 0 89 98 92
~
0 r
<:
m
C! 0.0025 0.0028 0.0167 0.103 0.0293 0.097 0.0012 0.0212 0.0049 0.0078 0.1088 0.263 67.71 0.795 29.78 127.3
0.0026 0.0024 0.0142 0.094 0.0298 0.094 0.001 0.0185 0.0047 0.0057 0.095 0.251 70.52 0.68 29.69 120.8
0 14 15 9 0 3 17 13 4 27 13 5 0 14 0 5
0.0025 0.0018 0.0083 0.029 0.0173 0.06 0.0011 0.0158 0.0025 0.0067 0.0241 0.26 64.4 0.496 27.09 130.7
0 21 35 63 41 35 0 13 45 0 65 0 5 23 9 0
0.0017 0.0015 0.004 0.023 0.0097 0.118 0.001 0.0193 0.0036 0.0031 0.021 0.256 65.25 0.134 29.92 136.1
32 11 26 6 26 0 0 0 0 33 3 0 0 46 0 0
0.0016 0.001 0.0018 0.017 0.0032 0.015 0.001 0.0044 0.0034 0.0045 0.0084 0.23 3.61 0.039 1.62 11.3
4 18 13 6
22 46 0 54 0 0 12 8 90 12 86 86
36 64 89 83 89 85 17 79 31 42 92 13 95 95 95 91
Percent removals are the percentage of the total constituent removal due to the individual treatment. Treatment removals sum to equal the total plant removal Values which evaluate to negative removals are reported as zero. NA: Not Available Physical and Nutrient concentration values are arithmetic means. Any nondetected observations were assumed to be present at the corresponding detectton limit.
Metal concentration values are geometric means determIned through problt analysis.
~
~
Recycled water - a source of potable water
291
<0.5 ug/L (geometric mean) to 4.2 ug/L (arithmetic mean). No PCBs, pesticides or dioxinlfurans were identified above detection limits in either Aqua II or Miramar waters (Table 4). The average total organic carbon (TOC) concentration found in Aqua II water was 1.37 ug/L, while TOC concentration averaged 9.83 ug/L in Miramar water.
Results of genetic toxicity testing. Genetic toxicity testing was used as a cost effective bioassay for screening the relative toxicity of the two waters being investigated. Four separate types of widely used bioassay systems were employed to assess genetic toxicity and potential cancer-eausing effects of the Aqua II and the Miramar waters. These included the Ames Assay, Micronucleus Test, 6-Thioguanine Resistance Assay, and Cellular Transformation Assay. Mutagenic and genotoxic activity has frequently been observed in studies of source and treated drinking waters. Consistent with these observations, organic extracts from both the AWT and the Miramar water sources exhibit some mutagenic activity. In addition, some genotoxic activity may also be associated with fractions which were not efficiently extracted by the resin columns used in these studies. The genotoxic effects detected in this study were observed primarily in the Ames test but some indications of potential mutagenic activity were also observed in the other bioassays. Based on the Ames test data, the tested Miramar extracts exhibited somewhat higher mutagenic activity than the comparable Aqua II extracts. The reason for this difference in observed mutagenicity is unknown but may reflect differences in composition of the original source waters or the result of the use of disinfectants on Miramar water. These data from short-term bioassay results of organic extracts of both Aqua II and Miramar water indicate that water from the Aqua II facility appears to exhibit less genotoxic or mutagenic activity than the low levels observed in water from the raw water entering the Miramar water treatment plant. GeMS identification of chemical substances. Synthetic resins were used to concentrate trace organics from
Aqua II, Miramar and reagent (blank) water. A total of 71 resin extracts were analyzed by gas chromatography-mass spectrometry in an effort to characterize the component mixture. The resin extract for each sample was fractionated by High Performance Liquid Chromatography (HPLC) into four fractions with one of the fractions fractionated again into three subfractions resulting in a total of six fractions for each sample. Analysis revealed the presence of many classes of organic compounds in low to trace amounts in the waters. Several herbicides were found in four of the five Miramar water samples. Only one compound found in the monthly Miramar water samples, 2-hexenal, was predicted to be an active mutagen and reported positive in the Ames testing. Fish Biomonitoring To complement information from other HES activities, fish biomonitoring experiments were expected to provide information on chronic exposure to trace contaminants contained in the effluent from the Aqua II plant and the untreated raw city water that accumulate in tissue but are not known to be active in genetic toxicity screening bioassays (de Peyster et al., 1991). Juvenile fathead minnows (Pimephales promelas) were exposed to Aqua II water, Miramar water, or charcoal filtered San Diego tap water (acclimation or control water) in flow-through aquaria. Three 28-day bioaccumulation experiments were completed at each site under standardized conditions following EPA and ASTM bioconcentration guidelines. In addition, swim speed and optomotor response tests were conducted. A total of 120 fish tissue samples and 60 aquarium water samples were collected for analysis of 69 base/neutral/acid extractable organics, 27 pesticides/PCBs, and 26 inorganics by investigators at theUniversity of California at Los Angeles (UCLA). Quadruplicate samples of fish tissue and either duplicate or triplicate samples of water were collected and analyzed for each sampling time (Days 0, 7, 14, and 28 of the 28-day experiments). The concentrations of most analytes in these samples were below the detection limit. Statistical comparisons were made only if at least 75 % of the samples exceeded the detection limit for a given analyte. Chemical analysis of fish tissue and water in the 28-day bioconcentration experiments revealed no statistically significant differences or between the fish exposed to Aqua II water and those exposed to Miramar water. The fish also displayed no significant difference in survival, growth, or swim speed after up to 28 days of continuous whole body exposure to test waters. This was a consistent finding for all three tests
292
A. W. OLIVIERI e/ al.
conducted on each of the three waters. Plant Reliability Evaluating the risk associated with the AWT plant effluent includes considering the ability of the treatment system to consistently achieve the level of treatment that was used as the basis for the health risk assessment. A reliability analysis was conducted to determine the reliability of the mechanical systems, characterize the treatment process variability, and present the probabilities of effluent characteristics meeting specific observed concentrations. To address these objectives, the reliability analysis included the review and analysis of plant performance and mechanical systems data. Summary statistics of the water quality data show.:d very effective treatment, with virtually complete removal of biological indicator organisms, and removals on the order of 87 to 98 % for general parameters, "conventional" pollutants, and for most inorganic ions (Table 5). The treatment plant was slightly less effective at removing metal ions, with observed removals on the order of 74 to 91 %. The results are contained and discussed in detail in the final HES report (Cooper et al., 1992a) and a separate plant reliability report (Cooper et al., 1992b). The levels of treatment obtained at the Aqua II plant were sufficient to meet all existing drinking water standards for all of the data reviewed in this study. The results of the statistical analysis and the mechanical reliability analysis indicate that compliance with Maximum Contaminant Levels (MCLs) can be maintained at the Aqua II demonstration AWT plant, on a long-term basis, with low probability of exceeding MeL concentrations. Results of the mechanical reliability analysis indicate that, over a two and one-half year period, the critical equipment are operational nearly 100% of the time. In addition, the m~chanical reliability analysis confirms statistically what is known practically about the Aqua II Plant: observed equipment failures do not cause a significant interruption in operation of the plant and impact on the effluent quality produced. Chemical and Virus Seeding Studies Chemical and virus seeding studies were conducted at the Aqua II plant. The Aqua II plant was spiked with both organic and inorganic constituents that could potentially enter a wastewater treatment plant as a large pulse flow. A summary of the results of this work is shown in Table 6. The ponds removed 60-99% of the inorganic and organic seed, and the advanced treatment processes removed >98.5 - >99.9% of the seed. Overall, the Aqua II plant removed >99.5 - >99.997% of the seeded chemical constituents (Froines et at., 1991). Four virus seeding studies performed on the Aqua II plant indicated a very high virus removal capability. As shown in Table 7, none of the seeded virus (attenuated vaccine strain Poliovirus 2) were recovered in the final effluent indicating a removal efficiency of as great as eight orders of magnitude (99.999999%). This removal rate is considerably greater than that reported in either the Pamona virus study (SDL, 1977) the Monterey study (Engineering Science, 1987) which used filtration and chlorination of secondary effluent but not reverse osmosis without final disinfection. Epidemiology The principal purpose of the epidemiology component was to develop a baseline of pertinent morbidity and mortality data for the City of San Diego so that a basis of comparison is available if direct potable reuse of municipal wastewater becomes a reality. The health of the residents was evaluated by characterizing thereproductive health and analyzing vital statistics of San Diego County. In addition, a neural tube defects survey was preformed to establish baseline prevalence rates in California and in San Diego. The epidemiological component provided baseline data to facilitate monitoring for actual health effects if the city decides to proceed with potable reuse of the type of treated effluent studied in this project (Molg~d et at., 1990).
TABLE 5. PLANT PERFORMANCE STATISTICS FOR MECHANICAL RELIABILITY SUMMARIZED BY PLANT UNIT Number of Occurances (Sum=M)
Number of Failures (Sum =R)
Mean Time to Repair MTTR(hr)
Mean Time Between Failures MTBF(yr)
Overall 90% CL Limit (yr)
Operating Availability AVO
Availa bility AVI
Headworks
393
28
13.6
0.352
0.280
0.9920
0.9956
Primary System
18
13
5.1
0.924
0.666
0.9977
0.9994
Secondary System
423
54
2.2
1.294
1.093
0.9975
0.9998
Operation Area
Inherent
~ (") '< (")
Package Plant
134
106
1.0
0.377
0.333
0.9989
0.9997
Ultraviolet I
143
8
6.7
5.362
3.577
0.9988
0.9999
Reverse Osmosis
414
144
2.2
0.393
0.354
0.9970
0.9993
Ultraviolet II
422
7
4.4
7.740
5.043
0.9969
0.9999
Aeration Tower
56
55
0.4
0.136
0.115
0.9989
0.9996
~G
Carbon Tower
13
9
l.l
1.304
0.889
0.9999
0.9999
"'Cl 0
Product Water
79
78
0.5
0.177
0.153
0.9996
0.9997
~
Caltrans. Equip.
32
29
1.3
0.396
0.315
0.9995
0.9996
G
Co ~
~ 00;
I ~ (II
0
0
....,
g. ~
~
~
a Mean Time to Repair: Number of calendar hours (amount of downtime) during which repairs due to failures were being made. b Mean Time Between Failures: Based on the chi-distribution such that MTBF = (2l/x 2 (0.5;2r+2». MTBF provides an estimate of the mean operating time between failures of all equipment in the unit process. T = total operating years (versus years in service) of all equipment. Example: if MTBF = 4 years for unit process A for which there are three pieces of equipment which operate 60% of the time, then in actual time, one can expect a failure to occur in 2.22 years. cOverall 90% Confidence Limit: The true value of MTBF is at least as great as the calculated 90% confidence limit. d Operating Availability: Calculated from the operating history of each unit, and is a measure of the fraction of the calendar time that the unit was operating. e Inherent Availability: A measure of the fraction of time that the component can be expected to be operational, excluding preventive maintenance downtime.
~
~
N
~
-l'>-
TABLE 6. CHEMICAL REMOVAL EFFICIENCY BY HYACINTH PONDS AND THE AQUA II FACILITY - SPIKE STUDY Spike
Pond Effluent
% Removal
(~gIL)
(~gIL)
by pond
Chromium
182
74
Tetrachloroethylene
900
Tetrachlorethane
Compound
Aqua II Effluent
% Removal by
Overall
(~)
Aqua IIa
59.3
<1
>98.6
>99.5
29
96.8
<0.3
99.9
>99.997
224
53
76.3
<.1
>99.8
>99.996
Trichlorobenzene
813
130
84.0
<.3
>99.8
>99.996
Tetrachlorobenzene
140
2
98.6
<.03
>98.5
>99.98
Lindane
70
12
82.9
<.03
>99.7
>99.96
a Tertiary treatment processes including package plant (coagulation/filtration), reverse osmosis, aeration tower, and carbon tower
TABLE 7. VIRUS REMOVAL EFFICIENCY OF THE AQUA II WATER RECLAMATION PLANT
» ~
ol ' =2
m CZ ~
Virus in Pond Effluent
~
Run Number
Expecteda (pfu/L)
Observed (pfu/L)
% Virus Removed in Ponds
Virus in Aqua II Effluent (pfu/L)
PUS-lu
4.7 X 104
1.9 X 102
99.6
NDb
>99.9999
PUS-2
9.6 X 104
1.2 X 104
87.5
NDc
>99.99999
PUS-3
1.0 X 10 5
4.0 X 103
96.1
ND
>99.99999
PUS-4
5.6 X 104
3.7 X 103
93.5
ND
>99.99999
• The number expected if no virus removal occurred (based on Lithium Tracer study). ND = none detected in 3800 liters. The occurrences of 0.2 pfuIL in this sample was due to cross contamination. d Calculation based upon 3800 liter sample.
b
C
Percent Virus Removal d
Recycled water - a source of potable water
295
Chemical Risk Assessment The chemical risk estimates were driven by Arsenic and THMs present in the Miramar water (THMs apparently from upstream chlorination of the raw water supply), and by bis-ethyl hexyl phthalate (BEHP) present in the Aqua II water. Arsenic accounts for approximately 80 % and THMs account for approximately 20 % of the risk estimate in Miramar water. The concentration of arsenic in Miramar water is well within applicable federal and state standards, approximately 2 % of the existing drinking water standards. The concentration of THMs in the Miramar water is 10% of the existing drinking water standards. The water from the Aqua II system was analyzed for arsenic and does not have enough arsenic to be reliably detected. Risk calculations, using EPA upper confidence level estimates, indicate that the Aqua II water, if used directly as potable water, would represent an estimated lifetime risk of 3.2 X 10-6 (.03 excess cancers/lO,ooo people) versus 1.2 X 10-4 (1.2 excess cancers/l0,000 people) for Miramar water. These estimates indicate that the Aqua II risk estimate is approximately 40 times less than the estimated risk associated with the untreated Miramar water. Results of the noncarcinogenic risk indicated that, for both waters, non• carcinogenic chemicals at observed low concentrations would not be anticipated to result in any significant health risk.
HEALTH ADVISORY COMMITTEE CONCLUSION Based on the above research, the overall conclusion reached by the Health Advisory Committee was that the health risk associated with the use of the Aqua II water as a raw water supply is less than or equal to that of the use of the existing raw water supply as represented by the Miramar source (Cooper et a/., 1992a).
FUTURE INVESTIGATIONS The positive performance and operational results obtained from the Aqua II pilot treatment facility along with the results of the health effects study provided the City with the necessary technical support to proceed with 3 the construction of a full scale (Aqua Ill), 3,800 m /d (1.0 Mgal/d), treatment facility. Aqua III is located in the San Pasqual area of San Diego County, California and includes the following unit operations and process: headworks consisting of bar screens and grit removal units; primary treatment consisting of rotary drum filters and a screw press for additional solids dewatering; aquatic treatment consisting of 24 water hyacinth ponds, operating in a step feed mode with aeration and effluent recirculation; filtration facilities consisting of chemical feed systems for ferric chloride and hydrated lime, a solids contact clarifier, and dual media filters; and advanced water treatment facilities consisting of ultraviolet disinfection, reverse osmosis, air stripping, carbon adsorption, and final disinfection with sodium hypochlorite. Starting in January 1995 these facilities were placed in operation with the intent of investigating and documenting full scale plant performance and reliability. Various health effects studies are also underway to verify pilot plant results. At the present time the treated water from Aqua m is being used for landscape irrigation. Future plans call for use of the treated water to recharge the aquifer underlying the Hodges Basin. The San diego region is also actively considering potable reuse of AWT treated effluent by discharge to a reservoir at a scale greater than 76,000 m3/d (20 Mgal/d). Feasibility studies have been completed and conditional approvals have been obtained from health authorities. Any decision to actually implement such a project will be subject to additional studies, public acceptance, and cost evaluations.
ACKNOWLEDGEMENT The authors wish to acknowledge the assistance of the treatment plant operations staff, in particular Jack
296
A. W. OLIVIERI et al.
Swerlein and Abel Hernandez, and the City of San Diego Laboratory staff, specifically John Chappin, for their invaluable help in making this project successful.
REFERENCES Cooper, RC., Olivieri, A.W., Eisenberg, D.M., Danielson, RE., Donohue, R, and Pettegrew L.A. (1991a). "Microbiological Indicators of Water Quality." Western Consortium For Public Health, Berkeley, CA. Cooper, RC., Olivieri, A.W., Eisenberg, D.M., Danielson, RE., Donohue, R, and Pettegrew L.A. (1991b). "Enteric Viruses" Western Consortium For Public Health, Berkeley, CA. Cooper, RC., Olivieri, A.W., Eisenberg, D.M., Danielson, RE., Donohue, R, and Pettegrew L.A. (1991c). "Bacterial Pathogens and Parasites." Western Consortium For Public Health, Berkeley, CA. Cooper, R.C., Olivieri, A.W., Eisenberg, D.M., Pettegrew L.A., Danielson, RE., Fairchild, W.A., and Sanchez, L.A. (1992a). "The City of San Diego Total Resource Recovery Project: Health Effects Study." Western Consortiumfor Public Health, Berkeley, CA. Cooper, RC., Olivieri, A.W., Eisenberg, D.M., Pettegrew L.A., Danielson, R E., Fairchild, W.A., and Sanchez, L.A. (1992b). "San Diego Aqua II Pilot Plant Reliability Report." Western Consortium for Public Health, Berkeley, CA. de Peyster, A, Rudnicki, R, Slymen, D.J., and Kimball, A., (1991). Biomonitoring Program - Final Report." Graduate School of Public Health, San Diego State University, San Diego, CA. Eisenberg, D.M., Olivieri, A.W., and Cooper RC. (1987) "Evaluation of Potential Health Risk of Direct Potable Reuse of Reclaimed Wastewater." Proc., Environmental Engineering Specialty Conference, Orlando, FL. Engineering Science (1987). "Monterey Wastewater Reclamation Study for Agriculture" Berkeley, CA. Froines, J. (1991). "Evaluation of Potential Health Risks Presented by Chemical Agents Associated with the San Diego Total Resource Recovery Program - Final Report." University of California, School of Public Health, Los Angeles, CA. Molgaard, C.A., Golbeck, A.L., and Elder, J.P. (1990). "The Epidemiology of Human Reproduction in San Diego County - Final Report." Graduate School of Public Health, San Diego State University, San Diego, CA. Sanitation Districts of Los Angeles (1991). "Pomona Virus Study - Final Report." Los Angeles, CA. Standard Methods for the Examination of Water and Wastewater. (1985). American Public Health Association, Washington, D. C. Standard Methods for the Examination of Water and Wastewater. (1989). American Public Health Association, Washington, D. C. Tchobanoglous, G.T., Maitski, F., Thompson, K., and Chadwick, T.H. (1987). "Evolution and Performance of City of San Diego Pilot Scale Aquatic Wastewater Treatment System Using Water Hyacinths" Proc., 60th Annual Conference of the WPCF, Philadelphia, PA. U.S. EPA. (1982). "Evaluation and Documentation of Mechanical Reliability of Conventional Wastewater Treatment Plant Components". Center for Envir. Res. Inform., Cincinnati, Ohio.