ARTICLE IN PRESS
Water Research 38 (2004) 111–127
Sludge accumulation, characteristics, and pathogen inactivation in four primary waste stabilization ponds in central Mexico Kara L. Nelsona,*, Blanca Jime! nez Cisnerosb, George Tchobanoglousc, Jeannie L. Darbyc a
b
Department of Civil and Environmental Engineering #1710, University of California, Berkeley, CA, 94720-1710, USA Hydraulic and Environmental Engineering, Institute of Engineering, National Autonomous University of Mexico, CD Universitaria, A.P. 70-472, Coyoacan 04510 Mexico, DF c Department of Civil and Environmental Engineering, 1 Shields Avenue, University of California, Davis, CA, 95616, USA Received 8 May 2002; received in revised form 14 August 2003; accepted 8 September 2003
Abstract To support the development of safe and feasible sludge management strategies, the accumulation rates of sludge and its characteristics were studied in four primary wastewater stabilization ponds (WSPs) in central Mexico (three facultative and one anaerobic). The accumulation rates and distribution of sludge were determined by measuring the thickness of the sludge layer at 8–40 locations throughout each pond. The average, per capita sludge accumulation rates ranged from 0.021 to 0.036 m3/person/yr. In the anaerobic pond the sludge distribution was uniform throughout the pond, whereas in the three facultative ponds most of the sludge accumulated directly in front of the inlet. To measure the horizontal and vertical variation in the sludge characteristics, sludge cores were collected from 3 to 7 locations in three of the ponds. Each core was divided into 4 sub-samples in which various physical, chemical, and microbiological parameters were measured. In addition, the inactivation of several pathogen indicator organisms was studied in a batch of sludge for 7 months. Based on the microbiological results, it is concluded that reasonable estimates of the inactivation of fecal coliform bacteria, fecal enterococci, F+ coliphage, somatic coliphage, and Ascaris eggs in WSP sludge in central Mexico can be made using first-order rate constants of 0.1, 0.1, 0.01, 0.001, and 0.001 d1, respectively. From the observed changes in the concentrations of total solids and the volatile to fixed solids ratio, empirical equations were developed to describe anaerobic degradation and compression, which are the two most important processes affecting the volume of sludge after its deposition. r 2003 Elsevier Ltd. All rights reserved. Keywords: Waste stabilization pond; Wastewater stabilization pond; Sludge accumulation rate; Sludge distribution; Biosolids; Pathogens
1. Introduction Wastewater stabilization ponds (WSPs) are a simple, low-cost, low-maintenance process for treating waste*Corresponding author. Tel.: +1-510-643-5023; fax: +1510-642-7483. E-mail address:
[email protected] (K.L. Nelson).
water. A typical system consists of several constructed ponds operating in series; larger systems often have two or more series of ponds operating in parallel. Treatment of the wastewater occurs as constituents are removed by sedimentation or transformed by biological and chemical processes. In the bottom of the ponds, a sludge layer forms due to the sedimentation of influent suspended solids as well as algae and bacteria that grow in the
0043-1354/$ - see front matter r 2003 Elsevier Ltd. All rights reserved. doi:10.1016/j.watres.2003.09.013
ARTICLE IN PRESS 112
K.L. Nelson et al. / Water Research 38 (2004) 111–127
pond. Sludge accumulation is greatest in primary ponds and can impact performance by altering the pond’s hydraulics due to a decrease in the pond’s effective volume and changes the shape of the bottom surface [1]. Therefore, periodic sludge removal is usually required and the long-term sustainability of WSP systems is dependent on the safe and effective management of their sludge. Despite the inevitable accumulation of sludge in primary ponds, sludge management is rarely considered as an integral part of pond design. One reason for the lack of attention given to sludge is that little information is available on the accumulation rates, the distribution of sludge within the ponds, and the characteristics of the sludge itself. The accumulation rate of sludge must be known so that the frequency of sludge removal can be determined and integrated into the pond design, maintenance schedule, and budget. Currently, the most common method for estimating sludge accumulation is the empirically determined volumetric, per capita accumulation rate [2–4]. However, the rates currently recommended for design have not been widely validated, and are believed to depend on temperature, among other factors. Thus, more regional data are needed to determine reasonable values for the per capita accumulation rate until models based on pond characteristics are fully developed [5–7]. In addition to knowing the rate of sludge accumulation, it is necessary to know how the sludge is distributed in a pond. The sludge distribution can have a significant impact on the pond’s hydraulics (and consequently the treatment efficiency), the frequency of sludge removal, and the feasible options for sludge removal. The distribution of sludge is primarily a function of the pond configuration [8]. A better understanding of the sludge distribution in ponds could lead to improvements in design to achieve optimal distribution of the sludge. More information is also needed on the characteristics of the sludge itself. The sludge volume and characteristics change with time due to anaerobic degradation, compression, and pathogen inactivation; however, little information has been published about the rates of these processes and the typical characteristics of WSP sludge. Data on the concentrations of pathogens in the sludge layer are needed to estimate the risk that pathogens pose upon removal of the sludge. Information on how the sludge characteristics vary throughout the sludge layer is also needed so that improved models of sludge accumulation as well as sampling protocols for monitoring programs can be developed. The research reported herein was conducted in Mexico, where WSPs are the most common type of wastewater treatment and over 400 systems have been built, most of them since 1980 [9]. To date, however, no studies have been reported on the accumulation rates and characteristics of sludge in Mexican ponds.
Furthermore, the removal of sludge has been undertaken in only a few of the ponds, thus, information is needed to support the development of a sludge management plan and to prevent pond failures [10,11]. The goal of this research was to gather more information on the sludge layer in WSPs to inform improvements in pond design and support the development of safe, effective sludge management practices. The specific objectives were to: 1. Measure the accumulation rate and distribution of sludge in four Mexican WSPs. 2. Evaluate the physical, chemical, and microbiological characteristics of the sludge and their horizontal and vertical variation within the sludge layer, including characterization of compression, anaerobic degradation, and pathogen inactivation.
2. Experimental design and methods Four primary WSPs located in the central highland region of Mexico (B2500 m) were selected for this research (Table 1). Pond depth and the thickness of the sludge layer were measured in all ponds to determine the sludge distribution and average accumulation rate. Sludge cores were extracted from three of the ponds (excluding San Jose de los Laureles); each core was divided into four stratified sub-samples and various physical, chemical, and microbiological parameters were measured. In addition, a batch of sludge from one of the ponds was stored in the laboratory for 7 months and the concentration of pathogen indicator organisms was measured periodically to gather more detailed data on their rates of inactivation. The methods used to collect field samples and the parameters measured are described below. 2.1. Characteristics of field sites The general characteristics of the anaerobic pond and the three facultative ponds are presented in Table 1. The pond in Texcoco treated wastewater from Mexico City, whereas the other three ponds treated wastewater from small communities. The degree of pretreatment varied among the four pond systems. In both Mexicaltzingo and San Jose de los Laureles, the wastewater passed through a biodigester, in which some suspended solids were removed by sedimentation, before entering into the primary pond. In Texcoco, although the only formal pretreatment was a bar screen and grit chamber, the sewer canal that transported wastewater to the pond had an insufficient slope to prevent settling, thus, significant sedimentation of suspended solids occurred before the wastewater was introduced to the primary pond. In Xalostoc, the only pretreatment was a bar screen and
ARTICLE IN PRESS K.L. Nelson et al. / Water Research 38 (2004) 111–127
113
Table 1 Characteristics of the four WSPs Locationa
Mexicaltzingo, Mexico San Jose de los Laureles, Morelos
Population
7280
800
Pond type
Pretreatment
Operation period (yr)
Q b, Mgal/d (L/s)
Surface areab, acres (ha)
Pond depth, ft (m)
Design HRTc, (d)
Average air temperature, F ( C)d Annual
Coldest month
Warmest month
Primary anaerobic
Biodigester
5
0.34 (15)
0.14 (0.06)
8.9 (2.72)
2.5
54 (12.2)
49 (9.4)
58 (14.4)
Primary Facultative
Biodigester
6
0.023 (1)
0.13 (0.05)
5.4 (1.64)
24
70 (21.1)
64 (19.1)
75 (23.6)
Texcoco, Mexico
NAe
Primary Facultative
Settling in sewer
10
2.15 (94)
6.0 (2.4)
5.1 (1.56)
10.6
60 (15.6)
55 (12.9)
65 (18.1)
Xalostoc, Tlaxcala
11,000
Primary Facultative
Grit chamber
15
0.15 (6.4)
1.2 (0.5)
7.8 (2.39)
47
62 (16.4)
57 (13.9)
65 (18.4)
a
Municipality, State. Flow and area are only for the primary pond studied in this research (some systems had more than one primary pond operating in parallel). c HRT=hydraulic residence time=Q/V, where V accounts for the wall slope and is therefore less than the volume calculated directly from the values of surface area and pond depth shown in the table. d Data are mean temperatures, 1951–1980, for the capital city of each state. Source: ‘‘Temperatura media mensual 1951–1980’’, ! Servicio Meterologico Nacional, Mexico. e NA=Not available. Because the wastewater treated in Texcoco is diverted from a large trunk sewer the population served is not known. b
grit chamber. At the time of sampling, all four ponds had been in continuous operation for at least 5 years. 2.2. Measurement of pond depth and thickness of sludge layer Pond depth and sludge thickness were measured at between 8 and 40 points in each pond with a sludge measuring optical gauge (SMOG, Orenco Systems Inc., Sutherlin, OR, USA). The apparatus consisted of a graduated pole with a detection unit at the end on which a light source and light sensor were fixed 1 in apart. An indicator light connected to the light source was on when the detection unit was in the water column and turned off when it entered into the sludge layer. In San Jose and Xalostoc, a rectangular grid was established around the perimeter of the ponds with flagged stakes and each sampling point was located by sighting off the flags. The Texcoco pond was too large to locate the sampling points by visual sighting, so a Global Positioning System (Garmin, Olathe, KS, USA) was used. The sludge thickness data reported for the Mexicaltzingo pond were obtained by another research group using the white towel test [12]. Three-dimensional surface profiles of the sludge distribution in each pond were created with a surface mapping program (Surfer Version 7.00, Golden Software, Inc., Golden, Co.). Grid files were generated from each data set by the point kriging method using exact
interpolation (no smoothing). The bottom surface was inferred from variations in the water depth. Surfer was also used to calculate the total sludge and water volumes for each pond by interpolating the sludge surface using Simpson’s 3/8 rule and integrating. The volumes were then corrected for irregular pond geometry (nonrectangular) and for wall slope. Apparent sludge accumulation rates (mm/yr) were calculated for each pond by dividing the total sludge volume by the pond’s bottom area and the number of years of operation. 2.3. Development of pneumatic apparatus for collecting sludge cores Sludge cores were collected from a rowboat at three to seven locations in each pond, including near the entrance, middle, and exit (Fig. 1). A sludge coring apparatus was developed specifically for this research because commercially available coring devices were found to be inadequate. Typically, coring devices used to sample lake, river, and ocean sediments have a sediment catch that is forced open as the column is pushed through the sediment. The pond sludge, however, was not dense enough to force open the sediment catch without causing significant vertical disturbance of the sludge layers. Also, such corers are primarily designed to retain sediment and do not completely seal at the bottom after the core is collected, allowing liquid to drain out by gravity. Because the parameters
ARTICLE IN PRESS K.L. Nelson et al. / Water Research 38 (2004) 111–127
114
(a)
(b)
(c)
(d) Fig. 1. Sludge distribution in the four ponds (a) Mexicaltzingo (raw data collected by Lloyd et al. [12]) (b) San Jose de los Laureles, (c) Texcoco, and (d) Xalostoc. The horizontal locations of the inlet and outlet structures are indicated by arrows. The locations at which sludge cores were collected are shown with crosses.
ARTICLE IN PRESS K.L. Nelson et al. / Water Research 38 (2004) 111–127
measured in this research were dependent on dissolved constituents and very small particles, it was necessary to completely seal the bottom of the core before it was raised from the sludge layer. Several previous researchers have used a small diameter PVC pipe to obtain sludge cores from ponds; the pipe was screwed into the compacted clay lining or soil at the pond bottom to form a plug that sealed the bottom of the core [13–16]. However, this method would not work in a pond with an artificial liner, such as concrete (Mexicaltzingo) or a geomembrane (San Jose de los Laureles). Furthermore, the ability of the plug to support the weight of the overlying sludge/water column limits the pipe diameter, thereby limiting the sample size as well as increasing the potential for contamination of lower layers by upper layers as the small-diameter pipe is pushed into the sludge. Thus, a rotating stainless steel door activated by a pneumatic cylinder was developed for this research to completely seal the core even in very thick sediments (30% total solids) and at water depths of 3 m without collecting material from the pond lining; a pressurized air tank was necessary to operate the corer. A diagram of the coring apparatus is provided in Fig. 2. Sample ports were located such that a minimum of 4 sub-samples could be obtained from each sludge core greater than 150 mm in height. Aluminum pipes were attached to the top of the corer to permit sludge collection at water depths greater than 0.8 m. After a core was collected, overlying water was removed using a large syringe. A sample was also collected with the syringe at the water–sludge interface. The remaining sludge samples were collected in bottles starting with the top layer by allowing the sludge to flow out the sampling ports. A specially designed plunger was used to force thick sludge out of the ports.
2.4. Estimation of sludge age To investigate the effect of age on sludge characteristics, a method for estimating the age of each sludge core sub-sample was needed. Sludge age is similar to relative depth, but while both parameters account for spatial variation in the rate of deposition to the sludge layer, only sludge age accounts for compression and biodegradation. Thus, at each core location it was assumed that fixed solids were deposited to the sludge layer at a constant rate throughout the operational lifetime of the pond, and that once in the sludge layer they were conserved. The age of the sludge at any depth was calculated to be directly proportional to the mass fraction of fixed solids that had been deposited above that depth, and the average age of each sample was calculated by averaging the age at the top and the bottom of the sample. The following equation was
115
Fig. 2. Diagram of pneumatic apparatus used to collect sludge cores (not to scale).
developed: mass of fixed solids in sample j and samples above it Aj ¼ years of operation total mass of fixed solids "P ¼
j Pi¼1 4 i¼1
in sludge core # FSi Y; FSi
ð1Þ
where j is the sub-sample of interest, j=1, 2, 3, 4, Aj the age of sludge at bottom of sub-sample j, FSi the mass of fixed solids in sub-sample i, i=1,2,y,j, and Y is the number of years the pond had been in operation. 2.5. Measurement of horizontal and vertical changes in sludge layer characteristics In each sludge core sub-sample, several physical, chemical, and microbiological parameters were measured (Table 2). Temperature, pH, and oxidation
ARTICLE IN PRESS 116
K.L. Nelson et al. / Water Research 38 (2004) 111–127
Table 2 Description of the parameters measured and the methods used in the sub-samples from the sludge cores Parameter
Methoda
Ponds sampled
Temperature pH Oxidation reduction potential (ORP) Total solids (TS) Volatile solids (VS) Fixed solids (FS) VS/FS Helminth eggs Fecal coliform Fecal enterococci Somatic coliphage F+ coliphage
Temperature probe pH electrode Platinum ORP electrode Gravimetric (2540 Gb) ‘‘ ‘‘ ‘‘ US EPA, [17] Multiple tube fermentation, direct method (9221 E.2) Spread plate (9215 C) with mEnterococcus agar (9230 C.2.c) Double agar layer [18], naldixic acid resistant E.coli CN 13 Double agar layer [18], streptomycin/ampicillin—resistant E. coli Famp
Mexicaltzingo, Texcoco, and Xalostoc ‘‘ ‘‘ ‘‘ ‘‘ ‘‘ ‘‘ Xalostoc only ‘‘ ‘‘
a
Numbers in parenthesis refer to Standard Methods [19]. A 50-mL syringe with the tip cut off was used to measure the initial sample volume to allow measurement of the solids’ concentrations on a weight per volume (w/v) as well as a weight per weight (w/w) basis. b
reduction potential (ORP) were measured immediately after extrusion from the core. Then, the samples were transported in a cooler to the laboratory and stored at 4 C. Total, volatile, and fixed solids, as well as all microbiological analyses, except helminth eggs, were performed on well-homogenized samples within 24 h of sample collection. Determination of helminth eggs was performed within one week of sample collection in most cases; no significant change in the viability of the eggs was expected during this time because the eggs can survive for years at low temperatures [20]. Care was taken during enumeration of the helminth eggs to minimize exposure to diethyl ether [21]. Fecal coliform bacteria and helminth eggs were enumerated in all sludge core samples. In the Xalostoc pond, three additional indicator organisms were measured—fecal enterococci, somatic coliphage, and F+ coliphage. For analysis of fecal coliform bacteria and fecal enterococci, the first dilution consisted of blending 10 g of sample with 90 mL of buffered dilution water (9050C 1a [19] at high speed for 1 min. For enumeration of somatic and F+ coliphages, the host bacterial strains were E. coli CN13 (resistant to naldixic acid) and E. coli Famp (resistant to streptomycin and ampicillin), respectively (obtained from Dr. Mark Sobsey, Dept. of Environmental Science and Engineering, University of North Carolina). To prepare samples for coliphage analysis, an initial elution was performed by vigorously mixing 5 g of sample with 5 mL of 0.1% Tween 80 for 1 min in a 50-mL centrifuge tube. The sample was left to sit for 10 min and mixed again for 1 min.Then, 5 mL of chloroform was added, the sample was mixed vigorously for 3 min, and centrifuged for 20 min at 2000 g. Serial dilutions were prepared from the supernatant.
2.6. Measurement of indicator organism inactivation in batch test Approximately 3 L of sludge was removed from the surface of the sludge layer near the inlet of the Xalostoc pond. The sludge was stored in the laboratory at ambient temperature in a 4 L container; the lid was closed, but not sealed, to allow gas to escape. Concentrations of fecal coliform bacteria, fecal enterococci, somatic coliphage, and F+ coliphage were measured periodically for 7 months. The enumeration methods were the same as those used for the sludge cores. Over the duration of the batch experiment, the temperature in the sludge ranged from 11 C to 16 C, which was similar to the range of temperatures measured in the Xalostoc sludge layer.
3. Results and discussion 3.1. Sludge distribution and rate of accumulation In all three facultative ponds, the distribution of sludge was very uneven (Figs. 1b–d), whereas in the anaerobic pond it was fairly uniform (Fig. 1a). In the facultative ponds, the maximum sludge thickness occurred near the single pond inlet; higher accumulation also occurred in some of the corners. In the anaerobic pond, the more even sludge distribution was attributed to two factors. First, instead of one inlet the pond had five, so the incoming solids were distributed over a larger surface area. Second, the hydraulic residence time (HRT) was much shorter (2 d compared with more than 11 d), thus, the overflow rate in the pond was much
ARTICLE IN PRESS K.L. Nelson et al. / Water Research 38 (2004) 111–127
higher and solids were carried further into the pond before settling to the bottom. In this research, from 8% to 25% of the ponds’ volumes were occupied by solids, resulting in proportional decreases in the design HRT (Table 3). It is likely that the effective HRTs in the facultative ponds were even further reduced by the formation of preferential flow paths and dead zones. The results from this research contribute to a growing body of evidence demon-
117
strating that in facultative ponds with single inlets, the majority of sludge accumulates directly in front of the inlet [8,13,22,23]. More information is needed on alternative inlet configurations that would distribute the sludge over a larger area, such as installing additional inlet pipes or increasing the inlet velocity or direction. The sludge accumulation rates were determined both on a per capita basis and as the average annual net increase in sludge thickness (Table 4), because both are
Table 3 Sludge thickness, volume, and resulting decrease in the hydraulic retention times (HRT) of the four ponds Pond System
Operation period, (yr) a
Mexicaltzingo San Jose Texcoco Xalostoc a
5 6 10 15
Sludge thickness (m) Avg.
Max.
Min.
Sludge vol./total vol(%)
0.67 0.15 0.36 0.34
0.81 0.38 1.11 1.95
0.61 0.05 0.06 0.07
25.3 8.2 14.4 13.2
HRT without HRT with sludge (d) sludge (d)
2.5 24 10.6 47
1.9 22 9.0 41
Raw data collected by Lloyd and Vorkas [12].
Table 4 Mean concentrations of solids in the sludge layer and measured accumulation rates in the ponds from this research compared to those reported in the literaturea Pond location
Pond type
No. of ponds sampled
Operation Mean TS period (yr) (g/L)
Mean VS/FS
Accumulation rate 3
m /pers yr
mm/yr
Reference
This research Mexicaltzingo San Jose Texcoco Xalostoc
Anaerobic Facultative Facultative Facultative
1 1 1 1
5 6 10 15
171 NA 112 166
0.63 NA 0.57 0.67
0.022 0.036 NAb 0.021
119 21 21 19
Literature values Columbia
Anaerobic
2
NA
NA
Pen˜a et al. [1]
Anaerobic
2
Facultative 1 NA 1 Facultative 12
2.5 10 3–10
77 53 NA NA 15–85
Mississippi, USA Utah, USA
Facultative 15
0.5–7
35–192
NA
15–51
Middlebrooks et al. [13]f
Facultative
7 13
59 77
0.62 NA 1.5 NA 0.29– 0.94 0.11– 0.59 2.23 4
Teles et al. [24]
NE Brazil France France
172 NA 39 187 54–136
0.055 0.040 0.023 0.026 NA NA 0.12
NA
SE Brazil
2.6 5 NA
NA
6.85 8.1
Schneiter et al. [16]g
a
2
Ayres et al. [25]c Carr!e and Baron [15]d Carr!e et al. [23]e
NA=not available. Because the wastewater treated in Texcoco was diverted from a large trunk sewer the population served is not known. c Values were calculated by averaging data reported for 15 sludge cores collected throughout the pond. d Value was calculated using raw data reported for two sludge cores, one near the inlet and one near the outlet. e Values for solids concentrations were presumably calculated from sludge cores. f Data are averages of sludge cores collected along two perpendicular axes in each pond. g Sampling protocol accounted for variation of sludge characteristics with location and depth. b
ARTICLE IN PRESS K.L. Nelson et al. / Water Research 38 (2004) 111–127
used in design. The average, per capita accumulation rates ranged from 0.021 to 0.036 m3/person/yr in the Mexican ponds; these rates are similar to values that have been measured in Columbia and Brazil but are significantly lower than rates measured in France. The higher accumulation rates in France may be partly due to colder temperatures, although this cannot be confirmed because the locations and temperatures of the French ponds were not reported. Other factors affecting the per capita sludge accumulation rate may include sewer inputs by industry, stormwater, and infiltration. Although published values determined from field data are few, a value of 0.04 m3/person/yr is often recommended for designing anaerobic ponds with average temperatures above 20 C [2,3,26,27]. It is concluded from the results of this research that 0.04 m3/person/yr is a reasonable estimate of the average rate of sludge accumulation in both facultative and anaerobic ponds in the central region of Mexico, even in regions with average temperatures well below 20 C. Additional field data on accumulation rates are needed from different regions and pond types. In contrast to the per capita rate of sludge accumulation, the measured average annual increase in sludge thickness varied widely, both in the ponds from this research (19–119 mm/yr) and among values reported in the literature (7–85 mm/yr) (Table 4). This variation is expected because the depth accumulation is strongly affected by the pond loading rate and treatment efficiency and is thus specific to each pond. For example, although the per capita accumulation rates were similar in the Mexicaltzingo and Xalostoc ponds, the thickness of the sludge layer increased over five times faster in Mexicaltzingo because the detention time was much shorter. 3.2. Horizontal and vertical variation of sludge characteristics The horizontal variation of sludge characteristics was studied by comparing the average values of each parameter in sludge cores collected near the inlet, middle, and outlet of each pond. For studying the vertical variation, the appropriate independent variable was either depth within the sludge layer or sludge age, depending on the parameter. The total solids concentration was found to be dependent on the depth within the sludge layer, whereas the remaining parameters were either dependent on the sludge age or constant throughout the sludge layer. 3.2.1. Estimation of sludge age The relationship between depth in the sludge layer and the estimated sludge age (Eq. (1)) is shown in Fig. 3. To account for differences in the operation periods and the thickness of the sludge layer, the normalized values
1
0.8 Relative depth
118
0.6
0.4 Mexicaltzingo Texcoco Xalostoc
0.2
0
0
0.2
0.4 0.6 Relative age
0.8
1
Fig. 3. Relationship between relative depth in the sludge layer and relative sludge age in the three ponds.
are plotted. If the sludge depth had increased by the same amount each year, then the relationship would be a straight line with a slope of one. However, as shown in the figure, the newer sludge occupied more volume than the older sludge. The figure can also be interpreted as showing the increase in sludge depth with time; the slope then represents the relative accumulation rate, which decreases with time. Potential sources of error in estimating the sludge age include changes in the concentration of fixed solids in the influent wastewater, production rate of algae and bacteria, hydraulic flow pattern, sludge settling zones, and resuspension or lateral movement of sludge over the operational lifetime of the pond. Also, the production of fixed solids through new cell growth in the sludge layer was not accounted for. Historical data were not available for any of the ponds, but according to interviews with plant operators and municipal authorities, no major changes occurred in any of the sewer networks that fed the ponds since operation was initiated, nor were there periods when the ponds were non-operational. 3.2.2. Total solids concentration The total solids (TS) concentration was found to be correlated with depth in the sludge layer (R2=0.84), with values increasing from around 3 g/L at the sludge/ water interface to over 300 g/L in the deepest sludge (Fig. 4a). It is a significant finding that the compression of sludge solids appeared similar throughout all three ponds. Furthermore, no relationship was found between TS and variables describing the qualitative characteristics of the sludge, such as age or VS/FS ratio. The resulting regression equation can be used to estimate the
ARTICLE IN PRESS K.L. Nelson et al. / Water Research 38 (2004) 111–127 350
119
TS, g/L = 417(Depth, m) 0.57
300
TS, g/L
250 200 150 100
Mexicaltzingo Texcoco Xalostoc
50 0
(a)
(b) 25
Temperature, °C
20 15 10 5 0 0
(c)
0.1
0.2
0.3
0.4
0.5
0.6
Depth in sludge layer, m
0.7
Inlet
(d)
Middle
Outlet
Location in pond
Fig. 4. Profiles of the total solids concentration (a and b) and temperature (c and d) as a function of depth in the sludge layer and distance from the pond inlet. In (a), the data points represent individual sub-samples from the sludge cores, whereas average values are reported in (b), (c), and (d).
solids concentration of the sludge if the thickness of the sludge layer is known. This information is needed to develop more accurate methods for estimating the rate of sludge accumulation, because the volume (and therefore depth) occupied by a given mass of sludge is a function of the TS concentration (which is an approximate measure of the sludge density). Knowing the TS concentration may also help to determine the most appropriate method of sludge removal. In the facultative ponds, the concentration of total solids (TS) decreased towards the pond outlet, whereas the concentration in the anaerobic pond was similar throughout (Fig. 4b). The higher TS concentration near the inlet in the facultative ponds is primarily a reflection of the greater thickness of the sludge layer, which causes more compression, but may also be affected by a higher fraction of higher density silts and sand that settle out near the inlet. The mean TS concentrations in the ponds from this research are reported in Table 4, and fall within the upper limit of values that have been reported in the literature. In general, the low TS values that have been reported occurred in sludge with a high concentration of volatile solids (high VS/FS ratio). 3.2.3. Temperature At the time the ponds were sampled, the temperature was fairly constant throughout the sludge layer (Figs. 4c
and d), although it was usually significantly different from the temperature in the overlying water layer and the air temperature (not shown). Over the course of one sampling day, the water temperature often varied more than 5 C while the temperature of the sludge layer remained constant. The average air temperatures during the months that the ponds were sampled were 10.8 C, 14.8 C, and 18.3 C in the Mexicaltzingo, Texcoco, and Xalostoc regions, respectively. There was much less variation in the average temperatures in the sludge layer of the ponds, which were 16.3 C, 16.7 C, and 17.9 C, respectively. The buffering of the sludge temperature can be explained because the rate of heat transfer depends on the degree of mixing, which is expected to be lower in the sludge than in the water or air. It is important to point out that the similarity in average sludge temperatures in the three ponds at the time of sampling was coincidental, and does not indicate that the sludge temperatures were similar at other times of the year. The Xalostoc pond, for example, was sampled again 3 months after the initial sampling and the sludge temperature had dropped to 14.3 C (the average air temperature dropped to 14.1 C). More research is needed on the fluctuations of temperature in the sludge layer so that the implications on temperature-dependent microbial processes, such as anaerobic degradation and pathogen inactivation, can be defined.
ARTICLE IN PRESS K.L. Nelson et al. / Water Research 38 (2004) 111–127
120
3.2.4. Anaerobic degradation (VS/FS, pH, and ORP) The ratio of volatile to fixed solids (VS/FS) was used as an approximate measure of the fraction of organic to inorganic matter in the WSP sludge. The decrease in the VS/FS ratio with sludge age was similar in all three ponds, and the slope of the line was interpreted as the rate of anaerobic degradation (Fig. 5a). The initial degradation rate was several orders of magnitude greater than the long-term degradation rate, particularly in cores with a high initial VS/FS ratio (individual core data not shown). It is believed that the decrease in degradation rate occurred as the rapid degradation of easily hydrolyzable organic matter was replaced by the continual, slow degradation of recalcitrant organic matter that required considerable
processing before it could be hydrolyzed and degraded by anaerobic bacteria. This trend has been identified previously, although not quantitatively [15,28,29]. Within the first year, about 30% of the VS was degraded in the facultative ponds and about 25% in the anaerobic pond, on average, although these figures may be underestimated because it was difficult to make an accurate measure of the initial VS/FS ratio. Better methods for collecting fresh sludge are needed so that the decrease in VS during the first few weeks after deposition can be quantified [29]. The long-term, first-order inactivation rate constants (after the first year) were determined from the slope of the linear regression fitted to the log-transformed VS/FS
1.25
Mexicaltzingo Texcoco Xalostoc
VS/FS
1
0.75
0.5
0.25
0
(a)
(b) 9
pH
8
7
6
5
(c)
(d) 50
ORP, mV
0
-50
-100
-150
-200 0
(e)
2
4
6
8
10
Estimated sludge age, yrs
12
Inlet
14
(f)
Middle
Outlet
Location in pond
Fig. 5. Profiles of the volatile to fixed solids ratio (a and b), pH (c and d), and oxidation reduction potential (e and f) as a function of sludge age and distance from the pond inlet.
ARTICLE IN PRESS K.L. Nelson et al. / Water Research 38 (2004) 111–127
data, discarding the initial measurement. The values were 0.122, 0.061, and 0.042 yr1 in the Mexicaltzingo, Texcoco, and Xalostoc sludge, respectively. These values are believed to be the first reported on the long-term degradation rates in WSP sludge, and are useful for estimating the reduction in sludge mass with long-term storage in the pond. In all three ponds the VS/FS ratio in the oldest sludge was approximately 0.5, suggesting that some organic matter may not degrade for many years, if ever. In all of the ponds, the VS/FS ratio increased towards the outlet (Fig. 5b). A possible explanation is that the denser, inorganic solids settled closer to the inlet, whereas the lighter, organic solids settled out near the outlet (predominately bacteria and algae in the facultative ponds). A similar increase in the VS/FS ratio towards the outlet was observed in a primary facultative WSP in northeast Brazil [25]. The mean VS/FS ratio in the ponds from this research is reported in Table 4; the values are also compared to others reported in the literature. The pH was constant throughout the sludge layer in all three ponds, with average values between 6.8 and 6.9 (Figs. 5c and d). The neutral pH suggests that methanogenesis was occurring, otherwise a build-up of fatty acids, the products of acidogenesis, would eventually overcome the buffer capacity of the sludge and cause the pH to drop. The transport of fatty acids to the overlying water column may also help to maintain stable conditions in the sludge layer. Although the neutral pH measured throughout the sludge layer suggests that methanogenesis was occurring, the oxidation reduction potentials (ORP) measured in the sludge layer suggested the dominance of sulfatereducing reactions (Fig. 5e). However, ORP was measured in the sludge samples after extrusion from the sludge corer, thus, it is possible that the measured values were higher than the actual values due to oxygen exposure upon sampling. In addition, the 2-mm platinum electrode may have been too large to detect microsites in which methanogenesis occurred. In future studies, it is suggested that ORP be measured directly in the undisturbed sludge layer, with a smaller electrode. The values measured in this research are higher than those reported by Carr!e et al. [15] for pond sludge in France, in which minimum values of –300 mV were found. In both of the facultative ponds the ORP was positive in the overlying water, as expected, and dropped sharply at the surface of the sludge layer. In contrast, in the anaerobic pond the ORP in the overlying water was similar to the ORP at the surface of the sludge layer. No significant change in oxidation reduction potential (ORP) was observed in the facultative ponds from the inlet to the outlet (Fig. 5f); no comparison could be made in the anaerobic pond because measurements were taken at only one location.
121
3.2.5. Helminth eggs The concentration of total helminth eggs was constant with sludge age in Xalostoc, but increased with age in Mexicaltzingo and decreased in Texcoco (Fig. 6a). Apparently, fewer eggs were deposited in the sludge layer during recent years in Mexicaltzingo. In Texcoco, on the other hand, the most likely explanation for the observed trend is that eggs were physically destroyed in the aging sludge because the percentage of total eggs that were viable actually increased with sludge age, whereas in the other two ponds it decreased (Fig. 6c). One reason that eggs could have been destroyed in Texcoco and not in the other two ponds is that the Texcoco pond treated wastewater from Mexico City, whereas the other two ponds treated wastewater from small communities. Thus, compounds could have been present in the Texcoco sludge, such as from industrial discharges, that caused a destruction of the eggs; the high salinity of the Texcoco wastewater may also have affected the eggs. The concentration of helminth eggs decreased dramatically towards the outlet of the facultative ponds, but was similar throughout the anaerobic pond (Fig. 6b). This pattern reflects the settling conditions in the ponds; the concentration of helminth eggs follows approximately the distribution of sludge. The percentage of viable helminth eggs also decreased significantly towards the outlet in the facultative ponds; the reported values are for the sub-samples from the top of the cores (Fig. 6d). A similar pattern was observed in a primary pond in northeast Brazil [30]. One reason for this trend could be that non-viable eggs have a lower density than viable eggs and therefore take longer to settle out (the density was not measured, however, in this research). Another reason could be that older, inactivated eggs were resuspended and a net movement of sludge toward the outlet occurred. More than 85% of the helminth eggs isolated from the sludge were Ascaris sp.; the remaining eggs were Trichuris, Hymenolepis, and Toxocara. sp. (Table 5). The species distribution is a function of the prevalence of infection in the community, as well as the settling patterns in the pond. Only the inactivation rate of Ascaris is considered here, because the eggs of Ascaris were more resistant than those of Trichuris and Toxocara (data not shown), and because the viability of Hymenolepis could not be determined by the method used. To measure the inactivation rate, it was assumed that no transport of eggs occurred within the sludge layer. A change in the concentration of viable Ascaris eggs per gram fixed solids was chosen as the most accurate measure of inactivation; basing the concentration on fixed rather than total solids eliminated any bias introduced by degradation of the sludge. Although it was expected that the inactivation rate varied with
ARTICLE IN PRESS K.L. Nelson et al. / Water Research 38 (2004) 111–127
122
Total helminth eggs, #/g FS
600 500 400 300
Mexicaltzingo Texcoco Xalostoc
200 100 0
(a)
(b)
Viable helminth eggs, %
100
80
60
40
20
0
Viable Ascaris eggs, #/g FS
(c)
(d) 10
3
10
2
10
1
10
0
10
-1
Fecal coliform bacteria, MPN/g TS
(e)
(f) 10
8
10
7
10
6
10
5
10
4
103 102 101 100 0
(g)
2
4
6
8
10
12
Estimated sludge age, yrs
Inlet
14
(h)
Middle
Outlet
Location in pond
Fig. 6. Profiles of the concentration of total helminth eggs (a and b), percentage of viable helminth eggs (c and d), concentration of viable Ascaris eggs (e and f), and concentration of fecal coliform bacteria (g and h) as a function of sludge age and distance from the pond inlet.
location in the pond, with seasonal changes in temperature, and with the concentration of acids, ammonia, and predatory microorganisms in the sludge layer [31], the
goal of this analysis was to determine a rough estimate of the average inactivation rate constant throughout the whole pond and over its entire lifetime.
ARTICLE IN PRESS K.L. Nelson et al. / Water Research 38 (2004) 111–127 Table 5 Species composition (%) of helminth eggs in the sludge layer of the three ponds Pond
Ascaris
Trichuris
Hymenolepis
Toxocara
Mexicaltzingo Texcoco Xalostoc Average
93.5 85.3 87.4 88.7
3.5 13 2.2 5.4
0.2 1.7 9.8 4.5
2.8 0.4 0.6 1.3
The inactivation rate constants were determined from the slope of the linear regression fitted to the log-transformed concentration of viable Ascaris eggs (Fig. 6e). The values for all three ponds were similar, ranging from 0.0007 to 0.001 d1 (Table 6). It is concluded that a first-order rate constant of 0.001 d1 is reasonable for estimating the average inactivation of Ascaris eggs in the sludge layer of WSPs in central Mexico. In a study reported separately, a similar mean inactivation rate was measured in dialysis chambers loaded with Ascaris eggs and stored in the Texcoco sludge layer for 14 months [32]. These values are the first known report of Ascaris egg inactivation rates in WSP sludge. 3.2.6. Indicator organisms In contrast to the helminth eggs (Fig. 6b), there was no clear trend in the concentration of fecal coliform bacteria (Fig. 6h) and the other indicator organisms (data not shown) from the inlet to the outlet. However, in all three ponds the concentration of fecal coliform bacteria decreased with sludge age (Fig. 6g) and the pattern was similar for somatic coliphage,1 F+coliphage, and fecal enterococci, although the rate of decrease varied for the different organisms (Fig. 7a). The inactivation rates of the four indicators were determined from the sludge core data assuming firstorder kinetics based on the initial concentration of the organisms at the sludge/water interface and the mean concentration of organisms in each sludge core, because these values were measured with a high-degree of confidence. The relationship between the mean and 1 At the time the experiment was initiated, beef extract was not available for preparation of the phage eluent, so a solution of 0.1% Tween 80 was used. At a later date, a comparison was undertaken in which triplicate sludge samples were eluted using either 0.1% Tween 80 or 3 g of beef extract in 0.1% Tween 80. The concentrations of F+ and somatic phages were 40–50% lower in the samples eluted with only Tween 80. Thus, the concentrations of phages measured in this study are likely to be at least 50% lower than the actual concentrations in the sludge. However, it is believed that the underestimation of the actual phage concentration did not have a significant effect on calculation of the inactivation rates.
123
Table 6 First-order rate constants for the inactivation of Ascaris eggs and indicator organisms. The values for the indicator organisms were determined by two methods—sludge cores and a batch test k (d1)
Organism
Sludge cores Ascaris eggs Mexicaltzingo Texcoco Xalostoc
Batch test
0.0009 0.0007 0.0010
Indicator organisms (Xalostoc) Somatic coliphage 0.0016 F+coliphage 0.016 Fecal coliform 0.13 Fecal enterococci 0.26
0.0074 0.037 0.16 0.20
initial concentration is:
Cm ¼
Co
R t1
kt t e R0t1 t0 dt
dt
Co =k ekt1 ekto ; ¼ t1 t0
ð2Þ
where Cm is the mean (geometric) concentration of organism measured in the entire sludge core, No./g FS, Co the concentration of organism measured at the sludge water interface, No./g FS, k the first-order inactivation rate constant, d1, t0 the age of sludge at top of core (sludge/water interface), d, and t1 is the age of sludge at bottom of core, d. The concentration of indicator organisms in the batch of sludge that was stored in the laboratory also decreased with time (Fig. 7b). The inactivation rate constants in the batch experiment were calculated by fitting a straight line to the log-transformed concentrations (No./g FS) of the four organisms, discarding points in the tailing region. The constants calculated from both the sludge cores and the batch experiment are reported in Table 6. Based on the close agreement between the two independent measures of inactivation, it is concluded that values of 0.1, 0.01, and 0.001 d1 are conservative estimates of the first-order rate constants for fecal coliform (and fecal enterococci), F+ coliphage, and somatic coliphage, respectively, under the conditions of this research. Fecal coliform bacteria and fecal enterococci are commonly used as indicators of enteric bacteria inactivation, and F+ coliphage may be an adequate indicator of enteric virus inactivation [33–36]. Thus, the results from this study provide strong evidence that in the sludge layer of WSPs in central Mexico, most enteric bacterial pathogens are inactivated within several months, whereas the inactivation of viral pathogens may take several years.
ARTICLE IN PRESS K.L. Nelson et al. / Water Research 38 (2004) 111–127
124
sludge when the sub-samples were extruded from the corer. Because the initial concentrations of organisms were very high (106–107) and the inactivation rates rapid compared to the age of the sludge, even a small degree of contamination would have obscured the actual decrease in concentration with sludge age. The method used to calculate the inactivation rate in the sludge cores (Eq. (2)), however, was not affected by the apparent contamination because it was based on the mean concentration of the entire sludge core. In the case of helminth eggs, the apparent contamination would have had a minimal impact on the measured concentrations because the relative change in concentration from the youngest to the oldest sludge was much lower. Based on similar logic, it is reasoned that there was no significant impact from contamination on the remaining parameters measured in this research. It should be emphasized, however, that when using sludge cores, as with any field sampling method, the results should be interpreted carefully.
Somatic coliphage F+ coliphage Fecal coliform bacteria Fecal enterococci 0
Log (C/Co)
-1
1.3 x 10
5
-2 6.6 x 103 -3
1.8 x 103
-4
9.3 x 10
2
-5 0
2
(a)
4 6 8 10 12 Estimated sludge age, yrs
14
0
Log (C/Co)
-1
4.0 x 10
5
-2 1.9 x 10
3
-3 1.3 x 102
-4
1.4 x 10
3
-5 0 (b)
1
2
3 4 5 Time, months
6
7
8
Fig. 7. Relative concentrations of indicator organisms measured in the (a) sludge cores and (b) a batch test of sludge from Xalostoc. The final concentrations (organisms/g TS) are also reported.
3.2.7. Contamination of sludge core samples If Figs. 7a and b are compared, there is a clear discrepancy between the concentrations of the four indicator organisms remaining in the sludge cores and those in the batch test (note the different time scales). Based on the determined rate constants (Table 6), the concentrations of the four indicator organisms measured in the older sludge core sub-samples could not have been due to surviving organisms; the most likely explanation is that the older sludge was contaminated by the newer
3.2.8. Summary and implications for sludge management The characteristics of the WSP sludge varied more in the vertical than horizontal direction. In fact, in the anaerobic pond there were no significant trends in the horizontal direction, which is consistent with its even distribution of sludge. Analogously, the parameters that varied horizontally in the facultative ponds were those that were a function of sedimentation (TS, VS/FS, concentration and viability of helminth eggs), with the denser constituents accumulating near the pond inlet where the thickness of the sludge layer was greatest. In all ponds, the sludge was transformed by compression, anaerobic degradation, and inactivation of the microorganisms such that significant variation of the sludge characteristics was observed in the vertical direction. One implication of the observed horizontal and vertical variations is that monitoring of the sludge layer cannot be achieved by taking measurements in one location or collecting grab samples. Rather, it is recommended that a protocol similar to the one used in this research be followed, in which sludge cores are collected from several representative locations throughout the pond. Within the first few months to 1 year in the sludge layer, significant stabilization of the organic matter and inactivation of helminth eggs and indicator organisms occurred. Thus, a significant improvement in sludge quality may be achieved if a pond is taken out of operation for a period before the sludge is removed. This option, however, requires that ponds are operated in parallel such that the remaining ponds can accommodate 100% of the inflow. Although the rates are expected to vary from region to region as a function of temperature, if sludge is removed during normal pond operation the recently deposited sludge will exert a dominant influence on the sludge characteristics.
ARTICLE IN PRESS K.L. Nelson et al. / Water Research 38 (2004) 111–127
125
Table 7 Meana and maximum concentrations of helminth eggs and indicator organisms in the sludge layers of the three ponds and maximum values of helminth eggs and fecal coliform bacteria allowed by the USEPAa and Mexican governmentb in biosolids that are to be land applied Location
This research Mexicaltzingo Texcoco Xalostoc Regulationc US EPA Class A US EPA Class B Mexico Class A Mexico Class B
Total helminth eggs, eggs/g TS
Viable helminth eggs, eggs/g TS
Mean
Max.
Mean
Max.
129 49 277
184 273 657
25 25 48
55 169 257
Somatic coliphage, pfu/g TS
F+coliphage, pfu/g TS
Fecal coliform bacteria, MPN/g TS
Fecal enterococci, cfu/g TS
Mean
Mean
Mean
Max.
Mean
1.3 105 5.7 104 3.1 104
1.2 107 1.5 107 4.4 107 3.4 103 7.9 106
Max.
Max.
5.3 105 4.2 106 1.2 104 1.3 106
0.25d No limite 10 35
Max.
1 103 1 103 1 103 2 106
a
Geometric mean for bacteria and virus, arithmetic mean for helminth eggs. USEPA [37]. c INE [38]. Also stipulates a Salmonella concentration p3 and p300 MPN/g TS for Class A and B biosolids, respectively. d The actual standard requires measurement of 4 g TS, such that the concentration is o1 viable egg/4 g TS. e No regulations exist for the concentration of Somatic coliphage, F+coliphage, or Fecal enterococci. b
Depending on the sludge removal process, evaluating the risk posed by the pathogens in the sludge may require determination of both the maximum and average concentrations of the pathogens and/or indicator organisms. The concentrations of helminth eggs and fecal coliform bacteria measured in this research exceeded the values allowed by the US EPA for both Class A and B biosolids and the Mexican government for Class A biosolids. The average values in the Mexicaltzingo and Texcoco sludge met the Mexican requirements for Class B biosolids (Table 7).
4. Conclusions and recommendations The results from this research on the characteristics of WSP sludge and the rates of the most important transformation processes in the sludge layer—compression, anaerobic degradation, and pathogen inactivation—can be used to evaluate sludge removal and treatment options. Based on the estimated degradation and inactivation rates, a significant improvement in sludge quality could be achieved by taking a pond out of operation for a period of time before removing the sludge. Because most of the sludge characteristics measured (TS, VS/FS, helminth eggs, indicator organisms) varied significantly both horizontally and vertically in the sludge layer of the facultative ponds, and vertically in the anaerobic pond, future efforts to characterize WSP sludge should include the collection of core samples from representative locations through-
out the pond.The specific conclusions from this research include: 1. Given the range in per capita sludge accumulation rates from 0.021 to 0.036 m3/person/yr measured in this research, a value of 0.04 m3/person/yr is a reasonable estimate of the rate of sludge accumulation in both anaerobic and facultative ponds in the central region of Mexico. 2. Although the per capita sludge accumulation rates were similar in the facultative and anaerobic ponds, the distribution of the sludge was dramatically different. In the anaerobic pond with multiple inlets and shorter HRT the sludge distribution was uniform throughout the pond, whereas in the three facultative ponds with single inlets and longer HRTs, most of the sludge accumulated directly in front of the inlet. 3. The two main processes that affect the volume of the sludge after it is deposited—compression and anaerobic degradation—were similar in all three ponds. A regression equation relating the total solids concentration to the thickness of the sludge layer was developed that can be used to evaluate different processes for sludge removal. The rate of anaerobic degradation decreased significantly after the first year, after which the long-term, first-order inactivation rate constant ranged from 0.042 to 0.122 yr1 in the different ponds. 4. Using two independent methods (sludge cores and batch experiment), the inactivation rates of four indicator organisms were estimated. The results
ARTICLE IN PRESS K.L. Nelson et al. / Water Research 38 (2004) 111–127
126
provide strong evidence that most bacterial pathogens are inactivated within several months in the sludge layer, whereas the inactivation of viral pathogens may take several years, depending on the initial concentrations; the inactivation of Ascaris eggs was even slower. Reasonable estimates of the inactivation of fecal coliform bacteria, F+coliphage, and Ascaris eggs in WSP sludge in central Mexico can be made using first-order rate constants of 0.1, 0.01, and 0.001 d1, respectively. The rates are expected to be dependent on temperature, among other factors, and may vary significantly outside this region. In terms of the average concentrations of helminth eggs and fecal coliform bacteria, the sludge in the Mexicaltzingo and Texcoco ponds complied with the Mexican standards for Class B biosolids, but the average concentration of viable helminth eggs in the Xalostoc sludge exceeded the value allowed for reuse or disposal of biosolids.
Acknowledgements The authors thank the Engineering Institute at the National Autonomous University of Mexico, Mexico City, for providing laboratory facilities, office space, and institutional support for this research. In addition, we thank Bill Sluis for designing and building the sludge corer, Eric Van Dusen, Leopoldo Sanabria Olmos, and Peter Nelson for their cheerful endurance collecting field samples, German Salgado Vel!asquez, Catalina Maya ! Elly Natty S!anchez Rodr!ıguez, and Adrianna Rendon, Romero Rosales for their dedicated laboratory help, and Mark Sobsey for donating the coliphage host strains. Financial support from the Fulbright Foundation and the University of California Institute for Mexico and the United States (UC MEXUS) was invaluable.
References [1] Pen˜a MR, Mara DD, Sanchez A. Dispersion studies in anaerobic ponds: implications for design and operation. Water Sci Technol 2000;42(10):273–82. [2] Mara DD, Alabaster GP, Pearson HW, Mills SW. Waste stabilization ponds: a design manual for eastern Africa., Lagoon Technology International and Overseas Development Administration, Leeds, England, 1992. [3] Mara D, Pearson H. Design manual for waste stabilization ponds in Mediterranean countries. Lagoon Technology International, Leeds, England, 1998. ! y mantenimiento [4] Oakley SM. Manual: Disen˜o, operacion ! en Centroam!erica (Manual: para lagunas de estabilizacion Design, operation and maintenance for stabilization ponds in Central America), AGISA/AIDIS, ERIS/USAC, INFOM, UNICEF, OPS/OMS, CARE, Guatemala City, Guatemala, 1998.
[5] Saqqar M, Pescod MB. Modelling sludge accumulation in anaerobic wastewater stabilization ponds. Water Sci Technol 1995;31(12):185–90. [6] Nelson KL. Development of a mechanistic model of sludge accumulation in primary wastewater stabilization ponds. Proceedings of the 5th International Specialist Group Conference on Waste Stabilization Ponds, Auckland, NZ, 2002. [7] Papadopoulos A, Parisopoulos G, Papadopoulos F, Karteris A. Sludge accumulation pattern in an anaerobic pond under Mediterranean climatic conditions. Water Res 2003;37(3):634–44. [8] Franci, R. Gerenciamento do lodo de lagoas de estabilizacao nao mecanizadas (Management of sludge from non-mechanized stabilization ponds). Programa de Pesquisa em Saneamento Basico (PROSAB), Rio de Janeiro, 1999. [9] Jim!enez B, Ramos J, Gonz!alez A, Mendoza H. Manual de disen˜o de lagunas (Design manual for ponds), 2nd ed. Mexico City: Comision Nacional de Agua; 1996, ISBN 968-36-5704-4. [10] Escalante V, Moeller G, Rivas A. Pond evaluation and upgrading in Mexico. Water Sci Technol 2000;42(10): 43–50. [11] Oakley SM, Pocasangre A, Flores C, Monge J, Estrada M. Waste stabilization pond use in Central America: the experiences of el salvador, guatemala, honduras and nicaragua. Water Sci Technol 2000;42(10):51–8. [12] Lloyd B, Vorkas C. Technical report on the evaluation of the Mexicaltzingo WSP system. Technical Report No. 5 Under DFID Research Contract No. R6871, Govt. of the United Kingdom, Center for Environmental Health Engineering, University of Surrey, England, Surrey, England, 1999. [13] Middlebrooks EJ, Panagiotou AJ, Williford HK. Sludge accumulation in municipal sewage lagoons. Water Sewage Works 1965;112(1):63–8. [14] Parker CD, Skerry GP. Function of solids in anaerobic lagoon treatment of wastewater. J Water Pollut Control Federation 1968;40(2):192–204. [15] Carr!e J, Baron D. Effects of maturation on the characteristics of wastewater stabilization pond sludges. Water Sci Technol 1987;19(12):169–75. [16] Schneiter RW, Middlebrooks EJ, Sletten RS, Reed SC. Sludges from cold region lagoons. Water Environ Res 1993;65(2):146–55. [17] USEPA. Control of pathogens and vector attraction in sewage sludge. EPA/625/R-92/013, US Environmental Protection Agency, Washington, DC, 1992. [18] Adams MH. Bacteriophages. New York: Interscience; 1959. [19] Eaton AD, Clesceri LS, Greenberg AE. Standard methods for the examination of water and wastewater. American Public Health Association, American Water Works Association, and Water Environment Federation, 1995. [20] Feachem RG, Bradley DJ, Garelick H, Mara DD. Sanitation and disease: health aspects of excreta and wastewater management. New York: Wiley; 1983. [21] Nelson KL, Darby JL. Inactivation of viable Ascaris eggs by reagents during enumeration. Appl Environ Microbiol 2001;67(12):5453–9.
ARTICLE IN PRESS K.L. Nelson et al. / Water Research 38 (2004) 111–127 [22] Schneiter RW, Middlebrooks EJ, Sletten RS. Cold region wastewater lagoon sludge accumulation. Water Res 1983;17:1201–6. [23] Carr!e J, Laigre MP, Legeas M. Sludge removal from some wastewater stabilization ponds. Water Sci Technol 1990;22(3–4):247. [24] Teles CR, Costa ANd, Gon9calves R.F. Sludge production in primary stabilization ponds and its use to forest species cultivation in the southeast region of Brazil. Proceedings of the Specialized Conference on Disposal and Utilization of Sewage Sludge: Treatment Methods and Application Modalities. International Association on Water Quality, Athens, Greece, 1999. [25] Ayres RM, Lee DL, Mara DD, Silva SA. The accumulation, distribution and viability of human parasitic nematode eggs in the sludge of a primary facultative waste stabilization pond. Trans Roy Soc Trop Med Hyg 1993;87: 256–8. [26] Gloyna EF.Waste stabilization ponds. World Health Organization, Geneva, 1971. [27] Arthur JP. Notes on the design, operation of waste stabilization ponds in warm climates of developing countries. World Bank Technical Paper No. 7, The World Bank, Washington, DC, 1983. [28] Bryant CWJ, Rich LG. Stabilization of organic carbon and nitrogen in consolidating benthal deposits. Water Sci Technol 1984;17:929–40. [29] Iwema A, Carr!e J, Minot D. Sedimentation and digestion on pond bottoms—an attempt to establish a short-term material balance. Water Sci Technol 1987;19(12):153–9. [30] Stott R, Ayres R, Lee D, Mara D. An experimental evaluation of potential risks to human health from parasitic nematodes in wastewaters treated in wastewater
[31]
[32]
[33]
[34]
[35] [36]
[37]
[38]
127
stabilization ponds and used for crop irrigation. Research Monograph No. 6, University of Leeds, Leeds, UK, 1994. Reimers RS, Bowman DD, Schafer PL, Tata P, Leftwich B, Atique MM. Factors affecting lagoon storage disinfection of biosolids. WEF/AWWA/CWEA Joint Residual and Biosolids Management Conference Biosolids 2001, San Diego, CA, USA, 2001. Nelson KL, Darby JL. Determination of the inactivation rate of Ascaris eggs in wastewater stabilization pond sludge using dialysis chambers and sludge cores. Water Environ Res 2002;74(4):362–9. IAWPRC, and Study Group on Health Related Water Microbiology. Bacteriophages as model viruses in water quality control. Water Res 1991;25(5):529–45. Woody MA, Cliver DO. Effects of temperature and host cell growth phase on replication of F-specific RNA coliphage Qb. Appl Environ Microbiol 1995;61(4):1520–6. Armon R, Kott Y. Bacteriophages as indicators of pollution. Crit Rev Environ Sci Technol 1996;26(4):299–335. Nasser A, Oman S. Quantitive assessment of the inactivation of pathogenic and indicator viruses in natural water sources. Water Res 1999;33(7):1748–52. USEPA. 40 CFR Part 257. Standards for the use and disposal of sewage sludge; final rules. Federal Reg 1993;58(32):9248. INE. Proyecto de Norma Oficial Mexicana NOM-004ECOL-2000. Proteccion ambiental. Lodos y biosolidos— especificaciones y limites maximos permisibles de contaminantes para su aprovechamiento y disposicion final. (Sludge and biosolids—specifications and maximum allowable concentrations of contaminants for their beneficial reuse and final disposal), Instituto Nacional de Ecolog!ıa, Mexico City, Mexico, 2000.
