Integrated evaluation model for surface water quality a case study in Ho Chi Minh city, Vetnam - Thu Thi Minh Nguyen

Journal of Science and Technology 54 (4B) (2016) 232-239 INTEGRATED EVALUATION MODEL FOR SURFACE WATER QUALITY A CASE STUDY IN HO CHI MINH CITY, VIETNAM Thu Thi Minh Nguyen*, Colin Arrowsmith, David Fraser Royal Melbourne Institute of Technology, GPO Box 2476, Melbourne VIC 3001 Australia *Email: s3455650@student.rmit.edu.au Received: 15th August 2016; Accepted for publication: 10th November 2016. ABSTRACT The quality of water is a measure of a number of characteristics that impact on its condition. Parameters affecting the water quality are classified as having chemical, physical, biological and radiation characteristics. However there is uncertainty as to whether or not a water sample meets set standards in quality if two or more parameters need to be considered. Therefore, this study is to develop an effective evaluation model, adopting fuzzy logic, to assess the quality of water. Using a case study in Ho Chi Minh, Vietnam, a model has been developed and applied the collected surface water monitoring data in 2015. The results obtained from this model are then compared to the results gained from Vietnam national water quality index established in 2011. Keywords: environmental management, surface water quality, water evaluation model, fuzzy logic, water quality index. 1. INTRODUCTION Integrated fuzzy evaluation method has been applied in recent years, This is an assessment approach developed based on fuzzy theory [1], and is able to deal with a number of non-linear relations, simplification, transferring ordinal value to verbal form. It can integrate qualitative and quantitative data, and is capable of dealing with data loss [2]. Evaluation methods based on fuzzy algorithm are applied in many areas such as automatic control, computer science, and environmental management [3, 4, 5 - 7]. In Vietnam, fuzzy logic was applied in many fields of study such as control system [8], fuzzy control [9], computer science [10], GIS system [11, 12], integrated algorithm [13] but not many researches in environmental management field. Ho Chi Minh City is a highly populated and rapidly developing city. Added to this, the city is now facing with the problem of environmental declining especially the surface water system. With dense canals and river system throughout the area, the city has advantages such as good source of water supply, good waterway transportation and water drainage system. However, the surface water of the city has been biological polluted by the industrial factories and lack of appropriate domestic wastewater treatment. The city will be also one of the most suffering cities Integrated evaluation model for surface water quality a case study in HCM city, Vietnam 233 in the world when the sea level rises as a result of greenhouse effect, including heavy floods, urban sanitary, lack of fresh water, etc. Therefore, as a part of the environmental management plan, this research applies the study on fuzzy evaluation, particularly in the field of surface water quality, to build an appropriate model of quality assessment in the city area. Also, water quality index results are to be used for comparison. Vietnam water quality index is the evaluation method of assessing the water quality officially promulgated in 2011. When comparing the two methods, strengths and weaknesses of the fuzzy evaluation model can be pointed out, and a renovation, mitigation and prevention plan for surface water quality could be drawn up for further study. 2. DATA AND METHODS 2.1. Data The dataset applied in the pilot model is the water sampling mean value in 2015, including rivers and canals sites. 26 river sites were set to monitor 15 parameters. The sample is taken on 1st, 8th, 15th and 22nd every month. In each day, there are two period of taking sample, at the highest and lowest tide. The monitoring indicators includes biological oxygen demand, chemical oxygen demand, coliform, E-coliform, pH, total suspended solid, turbidity, dissolved oxygen, heavy metals (cooper, lead, cadmium), salinity, manganese, ammonium, phosphate. For canals, 15 sites have been operated since 2011 and monitored 16 parameters, including biological oxygen demand, chemical oxygen demand, coliform, E-coliform, pH, total suspended solid, turbidity, dissolved oxygen, heavy metals (cooper, lead, cadmium, and chromium), salinity, manganese, ammonium, phosphate [14]. Table 1. The monitoring sites and monitored parameters in 2015. Type of monitoring sites Number of sites Number of parameters River 26 15 Canal 15 16 2.2. Methods 2.2.1. The fuzzy comprehensive analysis method Based on the fuzzy concept developed by Zadeh in 1965 [15, 16], the fuzzy evaluation method comprises of two parts, the membership function and the weight. The membership function is to set up a matrix of membership values of a site, which includes more than two monitored value of different parameters. It is an equation, usually a graph, which transfers a real value (monitored value) into a belonging level (membership value) of particular groups. There are a number of types of fuzzy membership graph with various shapes (linear, curve, etc.) and characteristic (continuous, non-continuous, etc). In practice of fuzzy application, simple and continuous shapes of graph are widely used. Barely determining the membership function type, there are some popular type of graphs applied, including triangular, trapezoidal, Γ- shaped, S-shaped, Gaussian, exponential-like, etc. [17]. Thu Thi Minh Nguyen, Colin Arrowsmith, David Fraser 234 This research applied one of the simplest matrix triangle and trapezoidal graphs [4 – 6, 18 – 20]. The membership graph for each parameter is built based on its characteristic and the standard limit. 2.2.2. Weighting method Weight is the level of importance of a parameter. Considering the water quality based on more than two affecting factors (quality parameters) becomes a multi-criteria decision making problem, and the weight of importance can be evaluated by the multi-criteria assessing methods. The value of weight is more or less due to criterion's importance, which total value of the weights is equal to 1. Some of the methods for assigning weights to a criterion include ordinal [21, 22], outranking [23], or weighting methods [24, 25]. In this study, overweight method is applied. The weight of a parameter is the ratio between mean of the samples and mean of standard values of that parameter. The main principle is the more difference of the samples’ mean compared to the standard value’s mean, the higher the weight [5, 7, 18, 26]. 2.2.3. Vietnam Water Quality Index Water quality index is an algebraic expression of the water quality for decision-makers to set up management plan. The index is a tool for synthesizing and simplifying data and information, and the results gained are easy to understand and use for local authority and people in general. This approach has been developed in different kinds of formula and applied in various areas such as Canada, Chile, England, Taiwan, India, Australia, Malaysia, etc., [27 – 29]. In Vietnam, this method was applied in many studies [30 – 32]. The monitored data will be applied into the formula of Vietnam water quality index 2011 [33], with nine parameters. Final score is in a range from 0 to 100, classified into five groups of quality, from lower B2 to A1, which the higher the score the better the environment. Table 2. Surface water groups of quality. Quality group (from the best to worst) Using purposes according to Vietnam quality standard Transferred to water quality index’s score A1 Good quality, applied for domestic supply and other purposes (A2, B1, B2) 91 – 100 A2 Applied for domestic supply with little treatment and other purpose (B1, B2) 76 - 90 B1 Applied for agricultural purpose or other equivalent purpose (B2) 51 - 75 B2+ Applied for transport purpose only and need further treatment 26 - 50 B2- Applied for transport purpose only and need further treatment 0 – 25 Integrated evaluation model for surface water quality a case study in HCM city, Vietnam 235 3. RESULTS 3.1 Quality results Applying the Water quality index method, the results showed that all the river sites were in low quality, B class. Three out of 25 river sites had received B1 value (Dong Tranh, Nga Bay, and Cai Mep). Most of the river (22/25) and all the canal sites (15/15) were met the lowest quality of water (lower B2) in both lowest and highest tide, which is the sign of risky water pollution in the whole catchment area. The results of the fuzzy model showed the similarities to the water quality index’s score. 24/25 river sites and all the canal sites (15/15) met the lowest value of quality (group B2-). Only one site (river Dong Tranh) met the group B1. 3.2 Pollution type The results from the monitored values of the rivers and canals in the watershed showed that the whole water catchment area had been biological polluted by the very high weighting value of coliform and E-coliform. For all the sites, rivers and canals, all the monitored values from the highest and lowest tide presented that the highest weighted indicator was E-coliform. For rivers, the monitored levels of Coliform were extremely higher than the lowest B2 quality standard. Canal monitoring sites also recorded unusual values of E-Coliform. This can explain the very high weights of these two parameters. Figure 1. The river’s Coliform levels and canal’s E-coliform levels compared to the B2 standard. Considering the location of the extreme values, most of the highest pollution sites were found in the centre of the city, which has the highest population density. Other monitoring sites in the sub urban areas recorded the lowest values compared to others. This is a sign that human, particularly residential activities, was the main source of pollution. Thu Thi Minh Nguyen, Colin Arrowsmith, David Fraser 236 Coliform levels in high tide Coliform levels in low tide E-Coliform levels in high tide E-Coliform levels in low tide Figure 2. Mapping the Coliform and E-coli across the monitoring sites 4. DISCUSSION The results in 2015 showed that the surface water quality of the area, Ho Chi Minh City, Vietnam, was highly biologically polluted. The very high concentration of E-coliform bacteria in the water samples is a sign of pollution of domestic wastewater or other organic waste sources. The two methods, water quality index and fuzzy integrated methods, showed the similarities in results. Water quality index is a comprehensive approach for quality evaluation based on a group of experimental functions. This method is widely applied and results are officially approved in many countries and were formally approved in 2011 in Vietnam. Fuzzy assessing is also a model Legend Monitored values Lowest values Low values High values Highest values Stream lines (rivers, canals, etc.) Population Density (people/km2) 104.439964 - 3869.448730 3869.448731 - 13223.346680 13223.346681 - 32163.755859 32163.755860 - 44937.218750 Integrated evaluation model for surface water quality a case study in HCM city, Vietnam 237 of quality assessing based on two important elements: fuzzy matrix and weights. Although the concept of fuzzy set was long developed, its application on the environmental management and assessment is still being researched. The advantage of the water quality index method is quick assessing, and its result can be recognised by authorities. However, with the limited number of assessed parameters, for example nine parameters regulated in the Vietnam handbook, the environmental quality cannot be totally covered. Fuzzy model, on the other hand, can be applied to an unlimited number of indicators, thus it can comprehensively assess all the relative parameters. Thus, the fuzzy model can comprehensively assess all the relative parameters. 5. CONCLUSION The water quality evaluation model was built based on the logic of fuzzy theory. Quality monitored data from a catchment area in a developing city has been applied for testing the results and compared to other regular method, Vietnam national water quality index. The results of the two methods show parallels and this in turn shows that fuzzy evaluation method is both applicable and practical. Fuzzy evaluation method also has the advantage of handling an unlimited number of quality indicators. Nevertheless fuzzy method should be further researched for more precise membership functions and limiting the uncertainty of the weights. On the other hand, only nine specific parameters are tested in Water Quality Index method, which can limit the result if other quality parameters are applied to evaluate the water. Also, applying experimental equations in assessing the quality of the water could not be applicable for specific region, particularly in a complex and developing city, such as Ho Chi Minh City. Therefore, an integrated method combining Water Quality Index and Fuzzy approach should be further researched to take advantages of the two methods and precise quality evaluation. REFERENCES 1. Matthias Ehrgott, J.R.F., Salvatore Greco - Trends in Multiple Criteria Decision Analysis, Springer US, 2010, p. 429. 2. André Lermontov L. Y., Mihail Lermontov, Maria Augusta Soares Machado - A Fuzzy Water Quality Index for Watershed Quality Analysis and Management, in Environmental Management in Practice, Intech., 2011, p. 24. 3. Meng Lihong C. Y., LI Weihong, ZHAO Ruifeng - Fuzzy Comprehensive Evaluation Model for Water Resources Carrying Capacity in Tarim River Basin, Xinjiang, China, Chinese Geographical Science 19 (1) (2009). 4. Runsheng L. V., B. L., Yanyong X. U. - Fuzzy Comprehensive Evaluation of Groundwater Pollution in Coal Mining Area, IEEE, 2011, p. 4. 5. Shufei Lin Y. Z., Yanwei Zhu - Application of Atmosphere-Environment Quality Assessment Based on Fuzzy Comprehensive Evaluation, in Information Computing and Application, Springer, 2011, p. 8. 6. Zhenxiang Xing Q. F., Dong Liu - Water Quality Evaluation by the Fuzzy Comprehensive Evaluation based on EW Method, in Fuzzy Systems and Knowledge Discovery (FSKD), IEEE, 2011, p. 4. Thu Thi Minh Nguyen, Colin Arrowsmith, David Fraser 238 7. Zhou Zhen-min W. X. C., Zhou Ke - Water Environment Evaluation Based on Fuzzy Comprehensive Method, in Mechanic Automation and Control Engineering (MACE), IEEE, 2011, p. 3. 8. Thuận N.Đ. - Các hệ cơ sở tri thức, in (Presentation). 2007, Đại học Nha Trang. p. 50. 9. Hà N.T.P. - Điều khiển mờ, Đại học Bách khoa Thành phố Hồ Chí Minh, 2005. 10. Thanh N.H., Hà Đ.X. và Hà T.V. - Một phương pháp ra quyết định tập thể dựa trên phân loại dữ liệu mờ: GDM-FC. 2006, Đại học Công nghiệp Thành phố Hồ Chí Minh. p. 13. 11. Quang T. X., Luận T.X. và Yến N.T. H. - Nghiên cứu áp dụng một số phương pháp toán - tin trong xây dựng mô hình đánh giá, dự báo trượt lở đất đá cho các vùng miền núi, Tạp chí Địa chất, Tổng Cục địa chất và Khoáng sản (2008) 7. 12. Hiệp L.Đ. - Mô hình cơ sở dữ liệu mờ trong hệ thống thông tin địa lý (GIS), in Hội thảo ứng dụng GIS toàn quốc, Đại học Bách Khoa Thành phố Hồ Chí Minh, 2011. 13. Lộc V.M. - Xây dựng các thuật toán tích hợp mờ để hình thành hệ trợ giúp quyết định đánh giá học sinh, Tạp chí Bưu chính Viễn thông (2002) 7. 14. Agency E.P. - Report of environmental quality monitoring results in 2011, Department of Natural resources and Environment: Ho Chi Minh City, Vietnam, 2012. 15. Zadeh L.A. - Fuzzy Sets. Information and Control 8 (3) (1965) 338-353. 16. George Bojadziev M.B. (Ed.) - Fuzzy Logic for Business, Finance, and Management, World Scientific Publishing, 2007, p. 253. 17. John H. Holland C.G.L. - Stewart W. Wilson, An Introduction to Fuzzy Sets: Analysis and Design, Massachusetts Institute of Technology, 1998. 18. Shen Jihong F. X. - The fuzzy comprehensive evaluation method based on improved super-standard weight, in International Conference on energy and environmental science, Elsevier Science, 2011, p. 8. 19. Liyun Yang A. L., Hao Bai - Using Fuzzy Theory and Principal Component Analysis for Water Shortage Risk Assessment in Beijing, China, 2010 International Conference on Biology, Environment and Chemistry, 2010, p. 5. 20. Ma L. Y. L., Zhou X. P. - Fuzzy comprehensive evaluation method of F statistics weighting in identifying mine water inrush source, International Journal of Engineering (2010) 6. 21. Palmer M. W. - Ordination Methods - an overview, 2008. 22. Oksanen J. - Major ordination methods, 2004. 23. Salaja Silas K. E. - Elijah Blessing Rajsingh An Analysis on the Effect of Mobility and Load on Fault Tolerant Service Selection Framework for Pervasive Environments, 2012. 24. Gwo-Hshiung Tzeng J. J. H. - Multiple Attribute Decision Making: Methods and applications, Taylor and Francis Group, 2011, 350. 25. Serafim Opricovic, G. H. T. - Extended VIKOR method in comparison with outranking methods, 2007. 26. Ji-hong Zhou C. L. H., Jun-guang Zhao, Ping Li - Water Quality Assessment of Xida Section Based on Fuzzy Comprehensive Evaluation, in Intelligent Information Technology Application Workshops, IEEE, 2009, p. 3. Integrated evaluation model for surface water quality a case study in HCM city, Vietnam 239 27. Rizwan Reza G. S. - Assessment of River Water Quality Status by using Water Quality Index (WQI) in Industrial Area of Orissa, World Applied Sciences, 2010. 28. Ionus O. - Water quality index-Assessment method of the Motru river water quality (Oltenia, Romania), 2010. 29. Simone de Rosemond, D.C.D., Monique Dubé - Comparative analysis of regional water quality in Canada using the Water Quality Index, in Environmental Monitoring and Assessment, Springer, 2009, p. 18. 30. Lãng T. T. - Nghiên cứu chỉ số chất lượng nước để đánh giá và phân vùng chất lượng nước sông Hậu, 2008, p. 8. 31. Hiền P.G. - Nghiên cứu xây dựng chỉ số chất lượng nước (WQI) phục vụ cho công tác quy hoạch sử dụng, Tập san Khoa học và Công nghệ quy hoạch thủy lợi, 2009, p. 3. 32. Technologies, H.C.M.C.D.o.S.a. - Study on surface water partition in Ho Chi Minh City using water quality index WQI and suggestion possible use, 2007, p. 7. 33. Environment, M.o.N.R.a. - National technical standard for surface water quality QCVN 08:2008/BTNMT, The Ministry of Natural Resources and Environment, 2008, p. 44.