Distribution and spatial variability of sludges in a wastewater stabilization pond system without desludging for a long period of time

Ecological Engineering 50 (2013) 5–12

Contents lists available at SciVerse ScienceDirect

Ecological Engineering journal homepage: www.elsevier.com/locate/ecoleng

Distribution and spatial variability of sludges in a wastewater stabilization pond system without desludging for a long period of time ˜ ∗ , J.J. Salas-Rodríguez R. Bouza-Deano Center for New Water Technologies (CENTA) Foundation, Autovía Sevilla-Huelva, Km. 28, 41820 Seville, Spain

a r t i c l e

i n f o

Article history: Received 28 December 2011 Received in revised form 19 July 2012 Accepted 23 July 2012 Available online 12 October 2012 Keywords: Stabilization ponds Sludge accumulation rate Desludging Sludge characteristics Multivariate analysis

a b s t r a c t This paper describes the sludge accumulation and its characteristics after several years of operation (15–20 years) of an urban wastewater treatment system using stabilization ponds (anaerobic, facultative, and maturation ponds) in southern Spain. The rate of sludge accumulation (0.011 m3 /person year for anaerobic pond) was lower than previously reported by other authors (0.04 m3 /person year) due to the effects of degradation and consolidation after this long period of time, and the vertical distributions of humidity and volatile solids indicate a consolidation and mineralization of sludge with depth. Confirmation of this fact by new experimental data might require a reconsideration of desludging times in such systems. Principal component analysis revealed some specific features of the data structure, and three principal components were identified which collectively accounted for 91.1% of the total variance. Principal components analysis results were confirmed by cluster analysis. Three clusters of variables were detected, corresponding to the three previously identified components. These results confirm that there are clear differences in the physico-chemical properties of sludges deposited in each pond. Vertical distributions of parameters indicate a consolidation and mineralization of sludge with depth. This supports the hypothesis of consolidation, mineralization, and volume reduction of sludge after long periods of time. © 2012 Elsevier B.V. All rights reserved.

1. Introduction Due to the European Directive 91/271/CEE concerning the purification of wastewater, since the year 2005 almost all wastewater must be treated with appropriate technologies. In small municipalities (<2000 equiv. inhabitants) possible solutions for wastewater treatment depend on suitably trained staff to run the facilities and on their energy requirements. The selection of the best treatment system in each case must be supported by six criteria (EPA, 1997): (i) processes which require a minimum number of operators should be selected; (ii) equipment requiring minimal maintenance should be selected; (iii) there should be effective operation over a wide range of levels of flow and load; (iv) energy consumption should be minimal; (v) facilities where potential failures of equipment and processes cause the minimum deterioration of effluent quality should be chosen; and (vi) there should be maximum integration of the treatment system into the environment.

∗ Corresponding author. Tel.: +34 954 75 90 20; fax: +34 954 75 52 95. ˜ E-mail address: [email protected] (R. Bouza-Deano). 0925-8574/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ecoleng.2012.07.010

Wastewater stabilization ponds (WSPs) are a simple low-cost, low-maintenance process for treating urban wastewater that is recommended for small populations where vast stretches of land are available (>7 m2 /equiv. inhabitant). A typical system consists of several constructed ponds operating in series. Treatment of the wastewater occurs as constituents are removed by sedimentation and/or transformed by biological and chemical processes. WSPs reproduce self-purification phenomena that occur naturally in watercourses (Salas et al., 2007). In a way, the system could be defined as “a compartmentalized river”, which simulates the first stage of anaerobic conditions that occur in channels when there is a discharge with a strong organic biodegradable content, whereas later stages are similar to situations that occur downstream of the spill in the process of naturally recovering the initial conditions of the water body. A sludge layer forms in the bottom of the ponds due to the sedimentation of influent suspended solids as well as algae and bacteria that grow in the pond. The accumulation of sewage sludges from urban wastewater treatment is a growing environmental problem. During the period 1992–2000 the production of sludge in the EU reached around eight million tons of dry waste matter per year (Magoarou, 2000), while a similar amount was produced in the USA in 2000 (Englande and Reimers, 2001). The generation of sludge in the anaerobic stage is

6

R. Bouza-Dea˜ no, J.J. Salas-Rodríguez / Ecological Engineering 50 (2013) 5–12

estimated at 30–40 litres/capita equiv. per year (Mara, 1976). Due to its high residence time in the ponds, the sludge is stabilized, and volatile-mineral average relationships are around 40–60% (Salas et al., 2007). With the system of common impoundments, anaerobic ponds (AP), facultative ponds (FP) and maturation ponds (MP), it is only necessary to carry out the removal of sludge every 5–10 years in the first stage of the life of the facility. This sludge accumulation is greatest in primary ponds and can affect performance by reducing the effective pond volume and ˜ shortening hydraulic residence time (Schneiter et al., 1984; Pena et al., 2000). Saqqar and Pescod (1995) studied sludge accumulation in Jordan; Goncalves (2002) reported data on sludge accumulation in Brazil; and Papadopoulos et al. (2003) studied sludge accumulation patterns in anaerobic ponds under Mediterranean climatic conditions in France and northern Greece. Nelson and Yang (2004) reported data on sludge accumulation characteristics in Mexico and stated that more regional data are needed to determine the sludge accumulation rate, sludge distribution, and sludge characteristics. There are still gaps in the existing data of ponds operated for periods of over 10 years without sludge disposal. Data collection in this regard will facilitate the understanding of the evolution of the accumulated sludge and the need for periodic removal. In addition the study of the spatial variation of characteristics among the various ponds allows further information about the sedimentation of sludge and the processes occuring in. The aim of this paper is to present and analyze the sludge accumulation pattern in the Center for New Water Technologies (CENTA) Foundation Experimental Center Stabilization Lagoons unit and its physico-chemical characteristics after 20 years of operations without desludging. Multivariate analysis tools are used to study the spatial variations in the physico-chemical characteristics of the sludge.

2. Materials and methods 2.1. Stabilization pond system The system studied was located at the R&D Wastewater Treatment Plant (37◦ 21 38 N, 6◦ 20 4 W) of the CENTA Foundation, located in the town of Carriónde los Céspedes, Seville, Spain. The experimental plant receives untreated wastewater from Carrión de los Céspedes, with approximately 2500 people. Part of the influent is directed to the system investigated after the pretreatment unit. Table 1 shows average values for raw wastewater. The studied system consists of three stabilization ponds: one anaerobic pond (AP), a facultative pond (FP), and two maturation ponds (MIP and MIIP) operated in series. The general characteristics

Table 1 Characteristic influent parameters. Parameter

Medium

Maximum

Minimum

pH Electrical conductivity (␮S/cm at 20 ◦ C) Total suspended solids (mg/L) Chemical oxygen demand (mg O2 /L) Biochemical oxygen demand (mg O2 /L) Ammonium (mg N/L) Phosphate (mg phosphate/L)

7.8 1821 340 904 471 86 40

8.0 2003 562 1280 708 111 60

7.5 1690 209 524 281 32 6

of these ponds are presented in Fig. 1, and the unit performance is shown in Table 2. 2.2. Sampling and analysis The bathymetric and sampling campaigns were carried out in winter 2010 after 15 years of operation of the anaerobic pond and 20 years of operation of the facultative and maturation ponds. The accumulation rates and distribution of sludge were determined by measuring the thickness of the sludge layer at several locations in each pond using a grid according to the size of the pond. In order to characterize accumulation and mineralization rates during its operation periods, 12 sampling points were located throughout the system, two in the anaerobic pond (influent and effluent), four in the facultative pond (influent, effluent, centre right, centre left), and three in each maturation pond (influent, centre, and effluent). A home-made sludge-core sampler with a transparent plexiglass tube with a diameter of 10 cm and length of 3.5 m was used. The tube was opened at its base and supplied with a movable lid which adheres well to the base of the tube, preventing any flow when it is closed. The lid was connected to a long chain hanging freely outside the tube. The procedure of sludge-core sampling consisted of: (i) lowering the tube vertically into the pond (at the point of sampling) until it reached the bottom; (ii) strongly pulling the chain to close the bottom of the tube, holding the core sample undisturbed in the tube; (iii) up the tube; (iv) pouring the tube contents into a polyethylene container; (v) refrigeration of containers until laboratory analysis. Samples from the anaerobic pond were divided into three different sub-samples in order to study the vertical differences in the properties of sludge. Laboratory analyses of parameters including pH, ORP, total solids (TS), fixed solids (FS), volatile solids (VS), Kjeldahl nitrogen, and ammonium nitrogen were performed on grab samples in

Fig. 1. Diagram of the system with dimensions and sampling points.

R. Bouza-Dea˜ no, J.J. Salas-Rodríguez / Ecological Engineering 50 (2013) 5–12

7

Table 2 System performance. Parameter

Total suspended solids BOD5 COD N-NH4 + N P Fecal coliforms

Reduction (%) Anaerobic pond

Facultative pond

Maduration ponds

Stabilization ponds system

50–60 40–50 40–50 – 5–10 0–5 0.2–0.5 ␮log

0–70 60–80 55–75 20–60 30–60 0–30 2.2 ␮log

35–40 25–40 20–35 15–25 15–50 30–45 0.7–1.3 ␮log

40–80 75–85 70–80 30–70 40–80 30–60 3–4 ␮log

accordance with the analytical methods described in Standard Methods (2005). Water content (WC) was also measured. The total concentration of elements (As, Ca, K, Na, Mg, Cd, Co, Cr, Cu, Fe, Hg, Mn, Ni, P, Pb, S, and Zn) was determined by ICPAES after digestion of the samples with aqua regia in a microwave digester in accordance with USEPA method 3051A. These elements have been analyzed in order to get as much information as possible about sludges mineral composition and to characterize sludge for its possible re-use. A Milestone Ethos 900 microwave instrument with an MDR-300/10S rotor and Teflon TFM vessels was used for total elements microwave digestion. A Varian 720-ES simultaneous axial view spectrometer with a CCD solid-state detector was used to determine the total content of elements (As, Ca, K, Na, Mg, Cd, Co, Cr, Cu, Fe, Hg, Mn, Ni, P, Pb, S, and Zn). All the reagents used were of analytical grade. Deionized water was obtained from a Millipore Milli Q system. Certified aqueous standards of the elements (Panreac, Barcelona, Spain) were used for ICP-AES.

2.3. Multivariate data analysis In this study, multivariate chemometric techniques were performed using the commercial software Statistica 10.0 (StatSoft, 2010). Principal component analysis (PCA) is one of the approaches most commonly applied to the study of data structures in environmetrics. It is aimed at finding and interpreting hidden complex and casually determined relationships between dataset features. This is accomplished by studying the data structure in a reduced dimension while retaining the maximum amount of variability present in the data. More precisely, a matrix of pairwise correlations among parameters is decomposed into eigenvectors, which, in turn, are sorted in descending order of their corresponding eigenvalues. Mathematically, PCA normally involves three major steps: (1) the standardization of measurements to ensure that they have equal weights in the analysis by autoscaling the data to produce new variables, where the mean is equal to zero and the standard deviation is equal to the unit; (2) calculation of the covariance matrix by identifying the eigenvalues and their corresponding eigenvectors; and (3) the elimination of components that account for only a small proportion of the variation in datasets. Cluster analysis (CA) complements PCA nicely. It was used to search for natural groupings among objects and to discover latent structures present in the data. Analyzed parameters were sorted into groups, or clusters, so that the degree of association between members of different clusters was strong. Prior to CA, the descriptor variables were block standardized by range to avoid effects of scale or units on the distance measurements. Hierarchical agglomerative CA was performed on the normalized data set with Ward’s method, using Euclidean distances as a measure of similarity.

3. Results and discussion 3.1. Sludge accumulation and distribution After 15 years of operation the total volume of sludge in the anaerobic pond was measured as 41.78 m3 , representing 55.7% of the total pond volume. This percentage is slightly higher than that (one-third full of sludge) recommended by Mara (2004) for anaerobic pond desludging (2.5 years are calculated for this pond according Mara (2004) recommendations). The volumes of sludge in the facultative pond and the two maturation ponds were measured and represented 11.1%, 25.7%, and 21.3% of the pond volumes, respectively, after 20 years of operation. With an organic load equivalent to 248 inhabitants and a medium BOD5 of 425 mg O2 /L, the sludge accumulation rate for the anaerobic pond was calculated to be 0.011 m3 /person year (0.031 L/person day), which is significantly lower than the value of 0.04 m3 /person year recommended when designing anaerobic ponds in warm climates (20 ◦ C) or 0.1 m3 /person year recommended for winter temperatures (10 ◦ C) (Mara and Pearson, 1998). Other authors’ data are 0.04 m3 /person year reported by ˜ et al. (2000) Mara (2004), 0.05 m3 /person year reported by Pena in Columbia, and 0.052 m3 /person year reported by Goncalves (2002) in Brazil. Those values are less than the median value of 0.12 m3 /person year observed in France by Carré et al. (1990) in 12 primary ponds under oceanic climatic conditions. The value obtained in this work is very similar to the values reported by Picot et al. (2003) in France and by Konate et al. (2011) in Burkina Faso of 0.007–0.02 m3 /person year. This low rate of sludge generation could be due to the higher temperature in the Mediterranean region during the warmest month, which could contribute to better degradation of sludge or a combination of degradation and consolidation of the sludge as reported by Nelson and Yang (2004). It should be remembered that decomposition of the settled sludge in primary anaerobic ponds occurs over a very long time (normally five years), which allows the solids which have a slow biodegradability rate to be decomposed, and hence the sludge volume will be further reduced. Thus the data obtained after 15 years of operation show one of the lowest ratios reported in the literature to date, which could be due to the effect of degradation and consolidation of the sludge in the lagoon, which, after long periods of time, causes the decomposition of organic matter and moisture loss from the lower layers of sludge, which will reduce its volume over time. Confirmation of this fact by new experimental data might require a reconsideration of desludging times in such systems. The accumulation rates of the facultative pond and the two maturation ponds were 0.027 m3 /capita year, 0.015 m3 /capita year, and 0.009 m3 /capita year, respectively. The sludge distribution was very uniform on the facultative and maturation ponds. Anaerobic pond bathymetry (Fig. 2) shows a higher concentration of sludge in the output, due to input

8

R. Bouza-Dea˜ no, J.J. Salas-Rodríguez / Ecological Engineering 50 (2013) 5–12

Fig. 2. Sludge distribution in the anaerobic pond after 15 years of operation.

configuration and the short distance between the entrance (located at 150 cm depth) and output. Considering the 15-year period, those values result in an annual linear accumulation of sludge in the anaerobic pond of 8.87 cm/year. 3.2. Sludge characteristics Data on sludge characteristics are shown in Table 3 after 15 years of operation in the case of the anaerobic pond and after 20 years in the other cases. These ponds operated with a mean organic charge of 425 g BOD5 /m3 day. The varying values of water content (76.0–90.0%) show that the sludge consistency ranged from liquid to pasty. The highest values for water content were observed in the first 25 cm depth starting at the top of the pond (Table 4). The average values found for fixed solids of more than 50% indicated that the sludge was well digested. As can be seen, these materials have a high percentage of fixed solids, with increases occurring during the process. The lower values for fixed solids were observed in the first 25 cm depth starting at the top of the sludge (Table 4), corresponding to the most recently sedimented sludge. Most of the parameters analyzed show a similar profile of concentration changes (high Pearson’s correlations; Table 5) among different ponds in the system, with higher concentrations in the anaerobic and facultative ponds and lower concentrations in the maturation ponds as is the case for the organic matter content (Fig. 3, upper). The pH and ORP are virtually constant throughout the system, and other elements such as potassium (Fig. 3, middle) and manganese. There are two elements, iron and cobalt (Fig. 3, bottom), that show a different behaviour from the rest, increasing their concentration in the sludge in the maturation ponds.

The results of the PCA analysis show that the first three principal factors have eigenvalues greater than unity (a criterion used to determine the number of factors retained (Yu et al., 1998): the first component (F1) accounted for about 63.8% of the total variance, the second component (F2) accounted for about 20.6% of the total variance, and the third component (F3) accounted for about 6.6%f of the total variance of the dataset. Table 6 gives the loadings for the three first components, and square cosines are presented in Fig. 4. A variable is increasingly well represented by a component as the corresponding value of the square cosine approaches the unit. Almost all variables are well represented by the first two components. Only pH, Na, and ORP could have been better represented by a different component. These three first components reduced the dimensionality of the total data from 25 to 3 (an 88.0% reduction) and resulted in the loss of 8.9% of the information contained in the dimensions. Organic matter, TKN, N-NH4 , P, Ca, Mg, Cu, S, Cr, Cu, Hg, Ni, Pb, and Zn are the variables that primarily contributed to the first

3.3. Multivariate methods A starting data matrix, with columns representing the different samplings (cases) and rows corresponding to the measured parameters (variables), was constructed for PCA and CA. These 12 observations were partitioned into the following groups: two observations in the anaerobic pond, four in the facultative pond, and three in each of the maturation ponds.

Fig. 3. Box–Whisker plot for organic matter (top), potassium (middle), and cobalt (bottom).

R. Bouza-Dea˜ no, J.J. Salas-Rodríguez / Ecological Engineering 50 (2013) 5–12

9

Table 3 Sludge characteristics. Parameter

Units

Anaerobic

pH ORP Water content Total solid Organic matter Inorganic matter Total nitrogen Ammonium Total phosphorus Ca Mg Na K Fe Cu Mn S Zn As Cd Co Cr Hg Ni Pb Chlorophyll

– mV % % % d.m. % d.m. g N/kg d.w. mg N/g d.w. g/kg d.w. g/kg d.w. g/kg d.w. g/kg d.w. g/kg d.w. g/kg d.w. mg/kg d.w. mg/kg d.w. g/kg d.w. mg/kg d.w. mg/kg d.w. mg/kg d.w. mg/kg d.w. mg/kg d.w. mg/kg d.w. mg/kg d.w. mg/kg d.w. mg/kg

6.6 ± −335.3 ± 83.3 ± 16.7 ± 35.0 ± 65.0 ± 15.6 ± 4.0 ± 14.7 ± 81.0 ± 7.0 ± 0.9 ± 3.0 ± 15.0 ± 378.3 ± 175.7 ± 11.7 ± 997.3 ± 9.8 ± 2.0 ± 4.0 ± 54.9 ± 5.7 ± 28.0 ± 113.1 ± –

Facultative 0.2 20.5 5.6 5.6 1.4 1.4 1.9 0.7 1.9 4.2 0.7 0.2 0.1 0.3 18.2 4.8 1.3 206.8 2.2 0.2 0.2 10.6 3.3 3.7 45.0

eigenvector, which can be interpreted as organic matter and associated elements. The second eigenvector was mainly related to K, Co, Mn, and Fe (some oligoelements). The third eigenvector does not represent any variables. In order to elucidate the differences in physico-chemical characteristics in the sludge analyzed in the different ponds, all observations were represented in the planes F1 through F2 (Fig. 5). It is possible to differentiate three groups of points corresponding to the three types of ponds studied. The maturation pond samples present positive values of F1, the organic matter component, while

6.6 −317.0 86.8 13.3 36.8 63.3 21.5 2.5 16.6 71.7 6.9 0.7 3.4 18.2 316.8 213.8 13.2 751.3 10.5 1.5 4.2 48.9 5.8 22.7 75.9 37.6

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.0 12.1 1.7 1.7 1.5 1.5 1.8 0.3 1.2 1.9 0.2 0.1 0.2 1.0 18.9 21.4 1.4 42.4 0.8 0.1 0.2 1.4 0.6 1.3 3.8 41.9

Maturation I 6.5 −360.3 78.7 21.3 18.3 81.7 10.9 1.2 11.4 41.4 4.7 0.7 3.0 17.4 84.3 175.3 8.4 213.7 5.4 0.8 4.5 30.2 1.8 15.3 25.0 44.9

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

Maturation II

0.1 41.7 3.8 3.8 2.1 2.1 2.9 0.2 3.0 6.0 0.6 0.2 0.5 4.6 16.6 33.7 2.3 61.7 1.3 0.8 1.1 6.7 0.3 3.3 5.4 11.0

6.6 −334.0 82.7 17.3 16.3 83.7 10.7 1.3 11.4 41.9 4.7 0.5 3.1 17.1 77.7 188.3 9.6 180.7 5.1 0.3 4.8 29.7 1.7 16.3 26.4 54.9

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.1 92.5 2.5 2.5 3.1 3.1 3.0 0.2 2.3 4.9 0.8 0.1 0.5 3.3 15.9 34.9 1.8 44.9 1.8 0.1 0.9 5.2 0.4 3.1 6.2 32.9

the anaerobic and facultative ponds have negative values of this factor. While the facultative pond samples have values close to zero for the second factor, F2, in the case of the anaerobic pond samples these values are clearly negative. Selection of these two eigenvectors permitted direct data evaluation via a two-dimensional plot (Fig. 5) and illustrates the variability observed throughout the sludge deposits in each pond. To confirm the associations between the variables in the dataset, CA was performed on the measured chemical variables. The search for natural grouping among the variables was a complementary

Table 4 Sludge characteristics at different depths in anaerobic pond. Input

pH ORP Water content Total solid Organic matter Inorganic matter Total nitrogen Ammonium Total phosphorus Ca Mg Na K Fe Cu Mn S Zn As Cd Co Cr Hg Ni Pb

– mV % % % d.w. % d.w. mg N/kg d.w. mg N/g d.w. mg/kg d.w. g/kg d.w. g/kg d.w. g/kg d.w. g/kg d.w. g/kg d.w. mg/kg d.w. mg/kg d.w. g/kg d.w. mg/kg d.w. mg/kg d.w. mg/kg d.w. mg/kg d.w. mg/kg d.w. mg/kg d.w. mg/kg d.w. mg/kg d.w.

Output

Upper

Medium

Bottom

Upper

Medium

Bottom

6.4 −328.0 89.0 11.0 36.0 64.0 17.2 4.9 15.3 79.9 7.8 1.2 3.2 15.3 375.0 178.0 10.0 880.0 8.3 1.6 4.3 69.7 1.3 32.6 68.5

6.5 −312.0 84.0 16.0 35.0 65.0 16.1 3.9 15.6 77.4 6.8 0.8 2.8 14.7 380.0 174.0 11.3 834.0 9.8 1.9 3.9 54.3 8.0 27.5 86.0

6.6 −313.0 77.0 23.0 33.0 67.0 13.1 3.3 12.5 87.0 6.4 0.7 3.0 15.2 353.0 178.0 13.2 1165.0 13.3 2.1 4.0 46.9 7.2 23.7 160.1

6.4 −356.0 90.0 10.0 37.0 63.0 17.5 4.8 16.2 77.2 7.8 1.2 3.1 15.1 398.0 174.0 10.9 910.0 7.2 2.3 4.2 64.0 1.8 31.4 96.0

6.6 −346.0 83.0 17.0 35.0 65.0 16.4 4.1 16.4 79.2 6.9 0.8 3.0 15.3 399.0 182.0 11.6 855.0 9.2 2.0 4.1 53.8 8.6 28.7 89.3

6.8 −357.0 77.0 23.0 34.0 66.0 13.4 3.2 12.2 85.4 6.3 0.7 2.8 14.5 365.0 168.0 13.2 1340.0 11.1 2.1 3.7 40.9 7.4 24.0 178.8

PH

ORP

HUM

DM

pH ORP HUM DM OM IM TKN N-NH4 As Ca K Na Mg Cd Co Cr Cu Fe Hg Mn Ni P Pb S Zn

1 0.11 0.26 −0.26 0.30 −0.30 0.12 0.13 0.18 0.27 −0.12 0.14 0.16 −0.03 −0.46 0.03 0.08 −0.35 0.16 −0.07 0.14 0.29 0.14 0.18 0.08

OM

1 0.16 −0.16 0.04 −0.04 0.10 0.700 0.08 0.16 −0.26 −0.11 0.09 −0.26 −0.25 −0.02 −0.10 −0.31 −0.01 −0.11 −0.06 0.15 −0.07 0.19 −0.14

1 −1.00 0.82 −0.82 0.92 0.70 0.86 0.62 0.67 0.06 0.84 0.35 0.41 0.67 0.65 0.56 0.80 0.77 0.69 0.85 0.70 0.94 0.63

1 −0.82 1 0.82 −1.00 −0.92 0.95 −0.760 0.76 −0.86 0.92 −0.62 0.79 −0.67 0.49 −0.06 0.39 −0.84 0.89 60.35 0.54 50.41 0.16 −0.67 0.81 −0.65 0.82 −0.56 0.32 −0.80 0.92 −0.77 0.53 −0.69 0.78 −0.85 0.89 −0.070 0.79 −0.94 0.84 −0.63 0.81

IM

TKN

N-NH4

As

Ca

K

1 −0.95 −0.76 −0.92 −0.79 70.49 −0.39 −0.89 −0.54 −0.16 −0.81 −0.82 −0.32 −0.92 70.53 −0.78 −0.89 −0.79 −0.84 −0.81

1 0.76 0.92 0.71 0.58 0.27 0.91 0.54 0.31 0.80 0.80 0.49 0.92 0.65 0.78 0.90 0.78 0.92 0.78

1 0.77 0.87 0.29 0.49 0.88 0.72 0.28 0.94 0.92 0.13 0.86 0.38 0.97 0.73 0.99 0.80 0.93

1 0.83 0.49 0.41 0.93 0.59 0.19 0.83 0.82 0.36 0.91 0.57 0.77 0.88 0.78 0.89 0.79

1 0.18 1 0.65 −0.28 0.88 0.46 0.59 −0.05 0.06 0.81 0.86 0.33 0.83 0.30 −0.03 0.93 0.80 0.42 0.30 0.93 0.83 0.31 0.82 0.50 0.85 0.30 0.76 0.57 0.82 0.29

Na

Mg

Cd

Co

Cr

Cu

Fe

Hg

Mn

Ni

P

Pb

S

Zn

1 0.44 0.71 −0.39 0.59 0.61 −0.38 0.43 −0.24 0.56 0.37 0.50 0.17 0.63

1 0.67 0.27 0.93 0.92 0.32 0.97 0.50 0.89 0.85 0.90 0.93 0.88

1 −0.12 0.79 0.85 −0.05 0.75 −0.07 0.78 0.40 0.76 0.48 0.85

1 0.23 0.17 0.83 0.19 0.78 0.22 0.21 0.26 0.37 0.17

1 0.99 0.19 0.92 0.34 0.97 0.71 0.95 0.78 0.98

1 0.17 0.94 0.29 0.97 0.71 0.95 0.78 0.99

1 0.30 0.86 0.15 0.32 0.15 0.45 0.15

1 0.42 0.89 0.79 0.90 0.89 0.91

1 0.33 0.66 0.36 1 0.66 0.29

1 0.70 0.98 0.78 0.98

1 0.73 0.88 0.96

1 0.80 0.69

1 0.73

1

Variable

−0.179 −0.030 −0.798 0.798 −0.978 0.978 −0.898 −0.854 −0.980 −0.961 −0.508 −0.518 −0.997 −0.750 0.122 −0.975 −0.962 −0.219 −0.975 −0.524 −0.918 −0.850 −0.913 −0.895 −0.941 63.87

F1

−0.356 −0.557 0.470 −0.470 −0.117 0.117 0.226 −0.327 0.093 −0.217 0.853 −0.576 0.045 −0.441 0.897 −0.057 −0.261 0.950 −0.162 0.773 −0.087 0.174 −0.314 0.395 −0.288 20.62

F2

−0.567 −0.665 −0.256 0.256 −0.082 0.082 −0.194 0.253 −0.035 0.020 −0.010 0.411 0.052 0.231 0.365 0.163 0.030 0.142 −0.096 −0.173 0.290 −0.234 0.173 −0.112 0.122 6.61

F3

Table 6 Loadings of the first three eigenvectors, F1, F2 and F3, in the total data set.

pH ORP Humidity Dry matter Organic matter Inorganic matter TKN N-NH4 Ac Ca K Na Mg Cd Co Cr Cu Fe Hg Mn Ni P Pb S Zn Eigenvalue (%)

way to study the latent structure of the data and permitted the comparison of CA results with those provided by the PCA. When CA was applied, the dendrogram (Fig. 6) showed three different clusters identified as A, B, and C. Cluster A corresponds with factor F1 for PCA, cluster B corresponds with F2, and cluster C with F3. There was adequate agreement between the results obtained by unsupervised PCA and CA to confirm the conclusions drawn over the complete dataset. These results confirm that there are clear differences in the physico-chemical properties of the sludges deposited in each pond. To elucidate variations in sludge quality at different depths in the anaerobic pond, a PCA analysis was also realized on this data.

Fig. 4. The square cosines for all the variables in components F1 and F2.

R. Bouza-Dea˜ no, J.J. Salas-Rodríguez / Ecological Engineering 50 (2013) 5–12

Variable

10

Table 5 Pearson’s correlations between different variables. Values in bold present significantly correlations (alpha = 0.05).

R. Bouza-Dea˜ no, J.J. Salas-Rodríguez / Ecological Engineering 50 (2013) 5–12

11

Table 7 Loadings of the first three eigenvectors, F1, F2 and F3, in the anaerobic pond depth data set.

Fig. 5. A two-dimensional plot of the observations in F1 and F2.

This analysis shows that the first four principal factors have eigenvalues greater than unity: the first component (F1) accounted for about 70.4% of the total variance of the dataset, the second (F2) accounted for about 12.2%, the third (F3) accounted for about 8.9%, and the fourth (F4) accounted for about 5.9%. There is a clear decrease in humidity down the column of sludge, and the same is found for the organic matter concentration and its main components such TKN, P, Mg, Na, K, Fe, Cu, Co, Cr, and Ni. On the other hand, Ca, S, Zn, As, Hg, and Pb increased down the column as did the inorganic solid concentration. All of this is reflected in

Fig. 6. The dendrogram obtained by applying Ward’s method.

Variable

F1

F2

F3

F4

pH ORP Water content Total solid Organic matter Inorganic matter Total nitrogen Ammonium Total phosphorus Ca Mg Na K Fe Cu Mn S Zn As Cd Co Cr Hg Ni Pb Eigenvalue (%)

0.892 0.118 −0.986 0.986 −0.924 0.924 −0.975 −0.992 −0.863 0.834 −0.967 −0.911 −0.716 −0.542 −0.715 −0.390 0.962 0.829 0.906 0.294 −0.863 −0.967 0.729 −0.989 0.905 17.60

0.239 −0.741 0.110 −0.110 0.376 −0.376 0.125 −0.002 0.081 −0.283 0.028 0.072 −0.422 −0.665 0.402 −0.729 0.061 0.191 −0.401 0.532 −0.409 −0.172 −0.030 0.013 0.165 3.05

0.044 0.276 −0.046 0.046 −0.065 0.065 0.180 −0.119 0.474 −0.456 −0.252 −0.397 −0.483 −0.108 0.416 0.316 −0.070 −0.515 −0.038 −0.099 −0.201 −0.142 0.666 −0.064 −0.363 2.23

0.145 −0.480 −0.091 0.091 −0.011 0.011 −0.008 −0.008 0.153 0.092 −0.022 −0.074 0.270 0.501 0.387 0.457 0.234 0.031 −0.016 0.576 0.215 −0.123 0.137 −0.037 0.134 1.48

the first factor, F1. The other three factors (F2, F3, and F4) do not show a clear correlation with any particular parameter (Table 7). To elucidate the differences in the physico-chemical characteristics of the anaerobic pond depth samples, all observations were represented in the planes F1 through F2 (Fig. 7). The upper samples present negative values from F1 and the bottom samples present positive values. This demonstrates that only this eigenvector provides evidence of the different sludge characteristics at different depths due to the process of maturation and consolidation of sludge.

Fig. 7. A two-dimensional plot of different depth samples in the anaerobic pond for observations in F1 and F2.

12

R. Bouza-Dea˜ no, J.J. Salas-Rodríguez / Ecological Engineering 50 (2013) 5–12

4. Conclusions The present work presents results concerning the quantity and quality of sludge produced during a long period of operation (15 years for the anaerobic pond and 20 years for the facultative and maturation ponds). In this study the rate of sludge accumulation in a Mediterranean climate was estimated at 0.011 m3 /capita year (0.031 L/capita day)in the anaerobic pond, 0.027 m3 /capita year in the facultative pond, and 0.015 and 0.009 m3 /capita year, respectively, in the maturation ponds, working with an organic load equivalent to 248 inhabitants and a medium BOD5 of 425 mg O2 /L. This low rate of sludge generation could be due to a combination of degradation and consolidation. So, long periods of time may cause the decomposition of organic matter and moisture loss from the lower layers of sludge, which will reduce its volume over time. Confirmation of this fact by new experimental data might require a reconsideration of desludging times in such systems. For each sample, pH, ORP, ST, organic matter, TKN, N-NH4 , total phosphorus, and several elements were analyzed. Mean data from the different ponds were compared in order to detect spatial differences. Organic matter and related elements present higher concentrations in the anaerobic pond and lower ones in the facultative and maturation ponds. The average values of more than 50% found for fixed solids indicated that the sludge was well digested. The higher values for organic solids and humidity were observed in the first 25 cm depths starting at the top of the sludge, corresponding to the most recently sedimented sludges, which have the lowest digestion times and compactness. An effort was made to extract more information from the datasets through the use of multivariate analysis techniques. PCA and CA revealed some specific features of the data structure and three principal components were identified with collectively accounted for 91.1% of the total variance. The first component was identified as the organic matter and related elements, most of the oligoelements are grouped on the second component and the third component did not represent any variables. The results of PCA were confirmed with CA. Three clusters of variables were detected, corresponding to the three previously identified components. These results confirm that there are clear differences in the physico-chemical properties of the sludges deposited in each pond. The vertical distributions of humidity (it decrease with depth), total solids, and fixed solids (they increase with depth) indicate a consolidation and mineralization of sludge with depth. This supports the hypothesis of a breakdown, compaction, and volume reduction of sludge after long periods of time, which could require a reconsideration of desludging times in such systems.

References Carré, J., Laigre, M.P., Legeas, M., 1990. Sludge removal from some wastewater stabilization ponds. Water Sci. Technol. 23 (3/4), 247. 1991. Off. J. Eur. Commun. L 135, 0040–0052, 30/05/1991. Englande, A.J., Reimers, R.S., 2001. Biosolids management in sustainable development status and future direction. In: Proceedings of IWA Specialised Conference on Sludge. Entering the 3rd Millennium, Taipei, Taiwan, pp. 38–45. EPA, 1997. Response to Congress on Use of Decentralized Wastewater Treatment Systems, EPA832-R-97-001b. Environmental Protection Agency Office of Wastewater Management, Washington, DC. Goncalves, R.F., 2002. Recycled sludge thickening and digestion pond from physicochemical upgrading process of facultative pond effluent. In: Proceedings of 5th International IWA Specialist Group Conference on Waste Stabilization Ponds, Pond Technology for the New Millennium, Auckland, New Zeeland. Konate, Y., Maiga, A.H., Wethe, J., Casella, C., Picot, B., 2011. Sludge accumulation in stabilization ponds in the Soudano-Sahelian climate of Burkina Faso. In: Smallwat11 3rd International Congress in Wastewater in Small Communities, Sevilla, Spain. Magoarou, P., 2000. Urban waste water in Europe: What about the sludge? In: Problem Around Sludge. Workshop Proceedings, EUR 19657 EN, Stresa, Italy, pp. 9–16. Mara, D.D., 1976. Sewage Treatment in Hot Climates. John Wiley & Sons, London. Mara, D.D., 2004. Domestic Wastewater Treatment in Developing Countries. Earthscan Publications, London. Mara, D., Pearson, H., 1998. Design Manual of Waste Stabilization Ponds in Mediterranean Countries. Laggon Technology International, Leeds, England. Nelson, K.L., Yang, I.J., 2004. The effects of degradation and consolidation on sludge accumulation in primary wastewater stabilization ponds. In: 6th International Conference on Waste Stabilization Ponds, Avignon, France. Papadopoulos, A., Parisopoulos, G., Papadopoulos, F., Karteris, A., 2003. Sludge accumulation pattern in an anaerobic pond under Mediterranean climatic conditions. Water Res. 37, 634–644. ˜ M.R., Mara, D.D., Sanchez, A., 2000. Dispersion studies in anaerobic ponds: Pena, implications for design and operation. Water Sci. Technol. 42 (10), 273–282. Picot, B., Paing, J., Sambuco, J.P., Costa, R.H.R., Rambaud, A., 2003. Biogas production, sludge accumulation and mass balance of carbon in an aerobic ponds. Water Sci. Technol. 48 (2), 243–250. Salas, J.J., Pidre, J.R., Sardón, N., 2007. Manual de Tecnologías no Convencionales para la Depuración de Aguas Residuales. Capítulo III-Lagunaje. Centro de las Nuevas Tecnologías del Agua (CENTA), Sevilla, ISBN 978-84-611-6884-2. Saqqar, M., Pescod, M.B., 1995. Modelling sludge accumulation in anaerobic wastewater stabilization ponds. Water Sci. Technol. 48 (2), 243–250. Schneiter, R.W., Middlebrooks, E.J., Sletten, R.S., Reed, S.C., 1984. Accumulation, Characterization and Stabilization of Sludges from Cold Regions Lagoons, CRREL Special Report 84-8. U.S. Army Cold Regions Research and Engineering Laboratory (CRREL), Hanover, NH. 2005. Standard Methods for the Examination of Water and Wastewater, 21st ed. American Public Health Association/American Water Works Association/Water Environmental Federation (APHA/AWWA/WEF), Washington, DC, USA. StatSoft, I., 2010. STATISTICA (data analysis software system), 10.0 ed. . U.S. Environmental Protection Agency (EPA), 2007. Method 3051A, Microwave Assisted Acid Digestion of Sediments, Sludges, Soils and Oils. Yu, J.C., Quinn, J.T., Dufournaud, C.M., Harrington, J.J., Roger, P.P., Lohani, B.N., 1998. Effective dimensionality of environmental indicators: a principal component analysis with bootstrap confidence intervals. J. Environ. Manage. 53, 101–119.