Extended activated sludge model no. 1 (ASM1) for simulating biodegradation process using bacterial technology

Ya-jing SONG*, Yue-bo XIE, Doddi YUDIANTO

1. College of Hydrology and Water Resources, Hohai University, Nanjing 210098, P. R. China

2. Civil Engineering Department, Parahyangan Catholic University, West Java 112548, Indonesia

Extended activated sludge model no. 1 (ASM1) for simulating biodegradation process using bacterial technology

Ya-jing SONG*1, Yue-bo XIE1, Doddi YUDIANTO2

1. College of Hydrology and Water Resources, Hohai University, Nanjing 210098, P. R. China

2. Civil Engineering Department, Parahyangan Catholic University, West Java 112548, Indonesia

Phosphorus is one of the most important nutrients required to support various kinds of biodegradation processes. As this particular nutrient is not included in the activated sludge model no. 1 (ASM1), this study extended this model in order to determine the fate of phosphorus during the biodegradation processes. When some of the kinetics parameters are modified using observed data from the restoration project of the Xuxi River in Wuxi City, China, from August 25 to 31 in 2009, the extended model shows excellent results. In order to obtain optimum values of coefficients of nitrogen and phosphorus, the mass fraction method was used to ensure that the final results were reasonable and practically relevant. The temporal distribution of the data calculated with the extended ASM1 approximates that of the observed data.

stream restoration; bacterial technology; extended activated sludge model no. 1 (ASM1); mass fraction; Xuxi River

1 Introduction

Although regulations have been implemented for maintaining the sustainability of the water environment, direct disposal of wastewater into rivers still widely occurs, especially in many developing countries. Various pollutants of high concentration have caused many rivers to lose their self-purification ability, and led to black and smelly river water. Pollutants have also caused the degradation of aquatic habitats. In addition, as rivers play a crucial role in many aspects of daily life, severe pollution in rivers will endanger the supply of clean water.

Research has been conducted to improve the surface water environment. Methods have evolved from the conventional biological filter pond to wetland restoration and standard wastewater treatment (Crites et al. 2006). Constructed wetlands, among the available alternatives, have been found to be the most effective way to treat polluted streams, and cost much less in construction, operation, and maintenance than conventional wastewater treatment plants (Zhou et al. 2007). However, based on the research by the U.S. Environmental Protection Agency, wetlands cannot effectively remove phosphorus from water whenoperating at different scales. It has even been found that the concentration of phosphorus tends to increase at the end of treatment (Richardson and Qian 1999). Similar results were also published by Juang and Chen (2007), who found generally low removal efficiency of nutrient substances by this technology.

In biological treatment, microorganisms (i.e., bacteria) are employed either to degrade pollutants into simple harmless substances or to treat and to restore the surface water (Akgerman et al. 1992; Wilson and Jones 1993; Awadallah et al. 1998; Chen and White 2004; Arzayus and Canuel 2005). In China, bacterial technology has been successfully used in the restoration of polluted lakes (Nie et al. 2008), the quality control of effluent from wastewater treatment plants (Liao et al. 2008), and the restoration of polluted urban streams (Yudianto and Xie 2010). Based on restoration project on the Xuxi River, in Wuxi City, bacterial technology offers an innovative solution for restoring the water quality in urban streams. It also results in better water quality. For instance, the water in the Xuxi River is now clearer and available for aquatic life. This study mainly aimed to investigate the influence of phosphorus in the biodegradation process during the restoration of the Xuxi River. An extended version of the activated sludge model no. 1 (ASM1) was used to study the fate of phosphorus under biodegradation processes.

2 Development of numerical model

2.1 Combination of Streeter-Phelps equation and ASM1

As it is critical and important to have sufficient dissolved oxygen (DO) in rivers for supporting sustainable ecosystems and aquatic life, all problems associated with the reduction of the DO concentration in rivers have become matters of concern since a century ago. Although various quantitative techniques have been used to assess the impacts of pollutants on river systems, significant development of water quality models was truly achieved only after the establishment of the classic BOD-DO models by Streeter and Phelps in the 1920s (Streeter and Phelps 1925).

Presented in a very simple form, the Streeter-Phelps equation in fact is insufficient to illustrate the whole degradation process. Rivers are strongly influenced by advection and dispersion processes, resulting in complex interactions of various constituents in the water. Under the specific condition of the restoration project of the Xuxi River, bacteria play a key role in biodegradation for the restoration of river water quality. The sources and sinks in the Streeter-Phelps equation mainly describe the transformation processes of substances. Therefore, the combination of the Streeter-Phelps equation and ASM1 was considered most appropriate for this study.

2.2 Introduction of phosphorus into model

In general, as various wastewaters are directly discharged into urban streams, the waterquality deteriorates day by day. Some facts show that the deterioration of water quality in urban streamsis mainly caused by high concentrations of total nitrogen and phosphorus contained in domestic, industrial, and agricultural wastewater (Malmqvist and Rundle 2002; Walsh et al. 2005; Garnier et al. 2005).

Although ASM1 is probably still the most widely used model to describe the biodegradation processes in rivers (Yudianto and Xie 2010; Maryns and Bauwens 1997; Stamou 1994) and oxidation ditches (Stamou 1997), it is still limited because phosphorus is ignored in the original ASM1. Due to the role of phosphorus in river water quality and its specific cycling course, it is crucial to involve phosphorus in the biodegradation model. As the original ASM1 only simulates certain biodegradation processes such as carbon oxidation, nitration, and denitrification, the model must be extended to the whole process. With the changing course of phosphorus included, the model can also be used as a tool to predict the possibility of eutrophication occurrence.

Though research has already been done on the simulation of phosphorus with other models from the ASM family, i.e., ASM2, ASM2D, and ASM3 Bio-P, different assumptions are made here to describe the removal process of phosphorus in natural rivers. In stark contrast to the sludge in sewage treatment plants, the sludge in rivers cannot be circulated or discharged. At the same time, there is no unambiguous aerobic or anaerobic zone over distance. Therefore, the storage of phosphate-accumulating organisms (PAOs) is ignored in this research, while its growth has been included in the development of heterotrophic bacteria.

As opposed to nitrogen, phosphorus cannot be transformed into oxidants. As it has the characteristic of cycling between the dissolved form and the solid, ASM1 has to be completed by another process to make it reasonable with phosphorus. In this model, phosphorus is divided into two parts: dissolved phosphorus and particulate phosphorus. When phosphorus enters the water, the dissolved part enhances the growth of active heterotrophic and autotrophic bacteria. On the other hand, the particulate part increases with the decay of bacteria. Due to hydrolysis of phosphorus in the following stage, the particulate part is then transformed into the dissolved part. As the hydrolysis of particulate phosphorus is defined similarly to the hydrolysis of entrapped organic nitrogen, the complete calculation matrix of the whole process is given in Table 1, whereSSis the concentration of readily biodegradable chemical oxygen demand (COD);XHandXAare the concentrations of active heterotrophic and autotrophic bacteria, respectively;SNO3is the concentration of nitrate nitrogen;SNH4is the concentration of ammonium nitrogen;XSis the concentration of slowly biodegradable COD;XPis the concentration of particulate products from COD decay;SNDandXNDare the concentrations of the soluble and particulate degradable organic nitrogen, respectively;Cis the concentration of dissolved oxygen;SPO4andXPO4are the concentrations of dissolved and particulate phosphorus, respectively;SN2is the concentration of nitrogen gas;YHis the heterotrophic yield;μmaxHis the maximum heterotrophic specific growth rate;KOHis the half-saturation

coefficient of oxygen in heterotrophic growth;KNH4is the half-saturation coefficient of ammonium nitrogen in heterotrophic growth;KSis the half-saturation coefficient for heterotrophs;KPO4is the half-saturation coefficient of phosphorus;ηgis the correction factor for anoxic heterotrophic growth;KNO3is the half-saturation coefficient of nitrate nitrogen in anoxic heterotrophic growth;YAis the autotrophic yield;iNBMis the mass ratio of nitrogen to COD in biomass;μmaxAis the maximum autotrophic specific growth rate;KOAis the half-saturation coefficient of oxygen in nitrifying bacterial growth;fPis the fraction of biomass yielding particulate products;kdHis the decay coefficient of heterotrophic bacteria;nHis the decay reaction order of heterotrophic bacteria;kdAis the decay coefficient of autotrophic bacteria;nAis the decay reaction order of autotrophic bacteria;kaNis the ammonification rate by anoxic hydrolysis of heterotrophs;khis the maximum hydrolysis rate constant at 20℃;KXis the half-saturation coefficient of organic in hydrolysis; andηhis the correction factor for anoxic hydrolysis.

2.3 Estimation of model coefficients

In the new extended ASM1, some unknown coefficients should first be established using the conservation principle that electrons and net electrical charges may neither be formed nor destroyed. This is basically the main rule followed by each substance throughout the process. Based on the conservation principle, the equilibrium equation has to be determined to describe each processjand substancec. Mathematically, the complete equations can be written as follows:

wherevjiis the stoichiometric coefficient for componentiin processj, andiciis the conservation factor used to convert the units of componentito the units of the substancec.

In general, each conservation equation containsa prioriinformation and may be applied to each process. Each conservation equation allows the prediction of one stoichiometric coefficient without performing an experiment whenever the other coefficients are known (Henze et al. 1999, 2000). This study uses a method similar to ASM2, ASM2D, and ASM3 to predict the coefficients.

For this extended ASM1, the grey-marked data in Table 1 are the known stoichiometric coefficients of special substances, such as the readily biodegradable COD, the active heterotrophic bacteria, and nitrate nitrogen. Basically, they are used to calculate the unknown coefficients at givenvji. As the values ofiNBM,YH,YA, andfPdo not change with temperature, the values of the kinetics parameters determined under 20℃ are used for further calculation. As each substance is made of five elements, C, H, O, N, and P, for one special substance, the fractions of these elements are supposed to be the same. The complete fraction values for each substance, given by Reichert et al. (2001) in the river water quality model no. 1(RWQM1), are presented in Table 2.

Table 2 Coefficients of various elements for different substances based on mass fraction (Reichert et al. 2001)

As the model is developed based on COD, all coefficients of the organic substances based on COD (Table 3),βj, can then be defined based onαj(jrepresents C, H, O, N, or P) as follows:

Table 3 Coefficients of various elements for COD based on mass fraction

In the extended model, the stoichiometric coefficients of ammonium nitrogen and soluble degradable organic nitrogen can be calculated based on the conservation of nitrogen, while the coefficients of dissolved phosphorus and particulate phosphorus should be defined based on the conservation of phosphorus. For example, in the process of the aerobic growth of heterotrophs, the coefficient ofSNH4can be established from the conservation balance of nitrogen by Eq. (1):

Using the same method, all unknown coefficients can be computed. The complete extended model equations are given in Eqs. (5) through (7), taking dissolved oxygen, dissolved phosphorus, and particulate phosphorus as examples:

whereCsis the saturated dissolved oxygen concentration.

Other equations regarding the coefficients, such asSS,XH,XA,SNO3,SNH4,XS,XP,SND, andXND, can be established with the same method.

Moreover, as described by the advection-dispersion equations, the velocity and longitudinal dispersion coefficient are very important parameters of the hydraulic characteristics of rivers. The longitudinal dispersion coefficient is a key parameter in the river, especially when velocity is low in urban rivers. The reaeration coefficient is one of the most important parameters to express the fate of oxygen. Though there are many methods of giving the expressions of the parameters, the widely used and accepted equations for the average velocityux(Manning 1891), longitudinal dispersion coefficientEx(Wallis and Manson 2005), and reaeration coefficientka(Schnoor 1996) are as follows:

whereWis the width of water surface,His the average water depth,u*is the shear velocity or friction velocity,Ais the wet area of the channel cross-section,Pis the wet perimeter,Sis the channel bed slope,nis Manning’s roughness coefficient, andc′,m, andlare empirical constants depending on the physical and hydraulic conditions of the channel.

3 Model validation

3.1 Xuxi River

The Xuxi River, previously known as the Shaoxiangbang River, is located in theChangnan District of Wuxi City in China. It originates from a gate at the Grand Canal and flows into the ancient canal towards the end of its lower reaches. The total length of the Xuxi River is 1.36 km with an average surface width of 4.5 m and water depth of about 1.4 m. Along its course, the Xuxi River receives wastewater from five refuse transfer stations, five public toilets, four car washing stations, and 43 drain outlets. Based on field investigation, it was calculated that about 10 100 m3of sewage was discharged into the river without treatment every day. For this reason, the water quality of the Xuxi River was deteriorating continuously. Besides, it had a bad odour, and the water was fully covered by algae, which caused very low concentration of DO. No living thing could be found in the Xuxi River.

The Xuxi River flows through a small bridge of the Rong Lane (#1), the Xuxi Bridge (#2), the Xishanxincun Bridge (#3), Wuai Road (#4), and the Xiaomu Bridge (#5) (Fig. 1). Data collected from these five points are used in the new extended model. A weir, which is about 50 cm high, was built on the Xiaomu Bridge (#5) on the lower Xuxi River.

Fig. 1 Sketch of Xuxi River

Without having control of sources of pollution and any artificial oxygenation, selected bacteria (which contain many kinds of aerobic, anaerobic, and facultative anaerobic bacteria) and microbial accelerators (which supply the trace elements to the bacteria) were directly injected into the river to activate the native bacteria in the original water and sediment. About 34 buckets of bacteria, each of which had a volume of 150 kg, were used to restore the quality of the Xuxi River. Water samplings were conducted from August 25 to 31 in 2009.

Fig. 2 shows the time average values of water quality parameters at different locations of the Xuxi River before and after restoration. Compared with the background data, great improvements have been made. After being intensively treated for about two months, the water quality of the Xuxi River has improved from inferior to Grade V to Grade V standards according to theEnvironmental Quality Standards for Surface Waterof China. Without artificial oxygenation, it can be seen that the DO concentration increased by about 0.94 to 2.62 mg/L at the five locations and reached more than 2.0 mg/L for the whole stream (Fig. 2(a)). Similar good results are also given for COD, NH3-N, and TP. As shown in Fig. 2(b), the COD concentration decreased to 8.03 mg/L at the outlet (#5) after restoration,and the average removal rate was larger than 43%. The maximum removal rate of NH3-N reached about 54.6% (Fig. 2(c)). With an average removal rate of 54%, TP had a similar degradation trend (Fig. 2(d)).

Fig. 2 Water quality parameters before and after the restoration of Xuxi River

3.2 Calibration of coefficients

Based on the above conditions, the extended ASM1 should be initially calibrated using values of the kinetics parameters obtained at 20℃. However, the average water temperature observed was 28.3℃. Table 4 shows the values of some key kinetics parameters modified under different temperatures. According to the literature (Jeppsson 1996), the model shows good agreement with the observed data in time only when three key parameters are modified: the heterotrophic maximum specific growth rate (μmaxH), the half-saturation coefficient of ammonium nitrogen in growth of hetertrophs (KNH4), and the half-saturation coefficient of phosphorus (KPO4). The complete calibrated results at 28.3℃ are given in Table 4.

Table 4 Calibrated parameters for extended model at 28.3℃

3.3 Initial and boundary conditions

According to the observed data, the initial conditions for this model were 2.14 mg/L for DO, 10.12 mg/L for COD, 8.34 mg/L for NH3-N, and 0.75 mg/L for TP. Throughout the treatment process, all the boundary conditions were the Neumann conditions. The different dispersion coefficients for the substances were 6xEfor COD, 0.3xEfor active heterotrophic bacteria, 0.25xEfor NH3-N, 0.1xEfor dissolved phosphorus, and 0.01xEfor particulate phosphorus.

3.4 Results and discussion

The results for the calibration of the extended model are in two aspects: time and space. The water quality in the Xuxi River is relatively steady for each day after two months’treatment, and there are only small differences of each substance over distance. In order to avoid the errors from individual data, the averaged data along the distance were used for stimulation in the model. Fig. 3 shows the variation of parameters with time.

Fig. 3 Comparison of temporal distribution of observed and simulated concentrations

The velocity and the dispersion coefficient show the fate of substances in a given water volume. For the transfer process of one special substance, the flow rate and the mean concentration can be used to determine the fate of advection. For dispersion, the dispersion coefficient is one of the most important parameters for determining the distribution of dispersion. Due to solubility, diffusion, settling characteristics, and transfer velocity, the dispersion coefficient for each substance is not constant. The extended model takes this into consideration for different substances to ensure that the calculated data fit well with the observed data.

As presented in Fig. 3(a), the COD concentration first increases and then decreases,which is probably caused by the increase of wastewater discharged into the river. Besides, both ammonium nitrogen and phosphorus need a long time to transform and be released in the water column. Therefore, lower values of dispersion are employed in this extended model to fit the observed data of those particular variables. As presented in Figs. 3(b) and (d), the transformation process of ammonium nitrogen and phosphorus contributes to the degradation of dissolved oxygen (Fig. 3(c)).

The total phosphorus includes the dissolved part and particulate part. The variations of these two parts with time are shown in Fig. 4. From Fig. 4, it is clearly shown that the particulate part of phosphorus increased quickly before it reached the balance state. Comparing Fig. 3(d) with Fig. 4, the trend of the dissolved part shows a greater effect on total phosphorus. Thus, the increasing concentration basically comes from the dissolved part, which, in the end, played a key role in the concentration of total phosphorus in the water volume.

Fig. 4 Concentrations of observed and simulated dissolved and particulate phosphorus

4 Conclusions

(1) As the flow velocity of the Xuxi River is very low, and the hydraulic retention time is long, the biodegradation process may be optimized by enhancing complete mixing.

(2) This extended model uses a new method to describe the transformation process of phosphorus in the river combined with the characteristics of the restoration project of the Xuxi River and hence differs from the assumptions of the ASM-family models. By using the methods of mass fraction and the conservation equation, the parameters are determined.

(3) By taking into account the role of phosphorus in ASM1, the model calibration shows good results. Not only do the concentration curves of the readily biodegradable COD, NH3-N, and DO fit well with the observed data, but the newly added substance, phosphorus, is also well simulated. Through simulation using this extended model, the transformation of all related substances can be better explained.

(4) In simulating the variation of the concentrations with time, because of a possible increase of wastewater discharged into the river, the concentration of COD first increased and then decreased. Both ammonium nitrogen and phosphorus needed a longer time to transform and could be released into the water column. As a consequence, lower values of dispersion are employed in this extended model to fit the observed data of those particular variables. Thetransformation processes of ammonium nitrogen and phosphorus contributes to the degradation of dissolved oxygen.

(5) The temporal distribution of the data calculated with the extended ASM1 shows excellent approximation to that of the observed data.

Akgerman, A., Autenrieth, R., and Bonner, J. 1992.In situremediation through contaminant mobilization by vacuum stripping combined with biodegradation.Journal of Hazardous Materials, 32(2-3), 323-382. [doi:10.1016/0304-3894(92)85101-6]

Arzayus, K. M., and Canuel, E. A. 2005. Organic matter degradation in sediments of the York River estuary: Effects of biological vs. physical mixing.Geochimica et Cosmochimica Acta,69(2), 455-464. [doi:10.1016/j.gca.2004.06.029]

Awadallah, R. M., Soltan, M. E., Shabeb, M. S. A., and Moalla, S. M. N. 1998. Bacterial removal of nitrate, nitrite and sulphate in wastewater.Water Research, 32(10), 3080-3084. [doi:10.1016/S0043-1354(98) 00069-4]

Chen, G., and White, P. A. 2004. The mutagenic hazards of aquatic sediments: A review.Reviews in Mutation Research, 567(2-3), 151-225. [doi:10.1016/j.mrrev.2004.08.005]

Crites, R. W., Middlibrooks, E. J., and Reed, S. C. 2006.Natural Wastewater Treatment Systems. Boca Raton: CRC Press.

Garnier, G. J., Némry, J. N., Billen, G., and Théry, S. 2005. Nutrient dynamics and control of eutrophication in the Marne River system: Modelling the role of exchangeable phosphorus.Journal of Hydrology, 304(1-4), 397-412. [doi:10.1016/j.jhydrol.2004.07.040]

Henze, M., Gujer, W., Mino, T., Matsuo, T., Wentzel, C., Marais, G. V. R., and Van Loosdrecht, M. C. M. 1999. Activated sludge model no. 2D, ASM2D.Water Science and Technology, 39(1), 165-182. [doi:10.1016/ S0273-1223(98)00829-4]

Henze, M., Gujer, W., and Takashi, M. 2000.Activated Sludge Models ASM1, ASM2, ASM2D and ASM3 (Scientific and Technical Report). London: IWA Publishing.

Jeppsson, U. 1996.A General Description of the Activated Sludge Model No. 1 (ASM1): Modelling Aspects of Wastewater Treatment Processes. Lund: Lund Institute of Technology, Lund University.

Juang, D. F., and Chen, P. C. 2007. Treatment of polluted river water by a new constructed wetland.International Journal of Environmental Science and Technology, 4(4), 481-488.

Liao, J., Xie, Y. B., Zong, X. C., and Cao, G. J. 2008. Pilot study on treatment of complicated chemical industrial effluent with CABRM process.Pollution Control Technology, 21(1), 11-15. (in Chinese)

Malmqvist, B., and Rundle, S. 2002. Threats to the running water ecosystems of the world.Environmental Conservation, 29(2), 134-153. [doi:10.1017/S0376892902000097]

Manning, R. 1891. On the flow of water in open channels and pipes. Transactions of the Institution of Civil Engineers of Ireland, 20, 161-207.

Maryns, F., and Bauwens, W. 1997. The application of the activated sludge model no. 1 to a river environment,Water Science and Technology, 36(5), 201-208. [doi:10.1016/S0273-1223(97)00475-7]

Nie, Q. Y., Xie, Y. B., Zhuang, J., and She, L. L. 2008. Cyanobacteria control using microorganism.World Sci-Tech Research and Development, 30(4), 430-432. (in Chinese)

Reichert, P., Borchardt, D., Henze, M., Rauch, W., Shanahan, P., Somlyódy, L., and Vanrolleghem, P. 2001. River water quality model no. 1 (RWQM1), II: biochemical process equations.Water Science and Technology, 43(5), 11-30.

Richardson, C. J., and Qian, S. S. 1999. Long-term phosphorus assimilative capacity in freshwater wetlands: A new paradigm for sustaining ecosystem structure and function.Environmental Science and Technology, 33(10), 1545-1551. [doi:10.1021/es980924a]

Schnoor, J. L. 1996.Environmental Modeling: Fate and Transport of Pollutants in Water, Air, and Soil. NewYork: John Wiley and Sons.

Stamou, A. I. 1994. Modelling oxidation ditches using the IAWPRC model with hydrodynamic effects.Water Science and Technology, 30(2), 185-192.

Stamou, A. I. 1997. Modelling of oxidation ditches using an open channel flow 1-D advection-dispersion equation and ASM1 process description.Water Science and Technology, 36(5), 269-276. [doi:10.1016/ S0273-1223(97)00483-6]

Streeter, H. W., and Phelps, E. B. 1925. A study of the pollution and natural purification of the Ohio River, III: Factors concerned in the phenomena of oxidation and reaeration.Public Health Bulletin No. 146. Washionshon: U.S. Department of Health, Education, and Welfare, Public Health Service.

Wallis, S., and Manson, R. 2005. On the theoretical prediction of longitudinal dispersion coefficients in a compound channel.Water Quality Hazards and Dispersion of Pollutants, 69-84.Springer.

Walsh, C. J., Roy, A. H., Feminella, J. W., Cottingham, P. D., Groffman, P. M., and Morgan II, R. P. 2005. The urban stream syndrome: Current knowledge and the search for a cure.Journal of the North American Benthological Society, 24(3), 706-723. [doi:10.1899/0887-3593(2005)024[0706:TUSSCK]2.0.CO;2]

Wilson, S. C., and Jones, K. C. 1993. Bioremediation of soil contaminated with polynuclear aromatic hydrocarbons (PAHs): A review.Environmental Pollution, 81(3), 229-249. [doi:10.1016/0269-7491(93) 90206-4]

Yudianto, D., and Xie, Y. B. 2010. Influences of limited ammonium nitrogen and water temperature on the urban stream restoration using bacterial technology: View from the perspective of numerical modeling.Journal of Water Resource and Protection, 2(3), 227-234. [doi:10.4236/jwarp.2010.23026]

Zhou, J. B., Jiang, M. M., Chen, B., and Chen, G. Q. 2007. Emergy evaluations for constructed wetland and conventional wastewater treatments.Communications in Nonlinear Science and Numerical Simulation, 14(4), 1781-1789. [doi:10.1016/j.cnsns.2007.08.010]

(Edited by Yun-li YU)

*Corresponding author (e-mail:syj111188@sina.com)

Received May 25, 2011; accepted Dec. 13, 2011