Groundwater quality assessment of the quaternary unconsolidated sedimentary basin near the Pi river using fuzzy evaluation technique

Groundwater quality assessment of the quaternary unconsolidated sedimentary basin near the Pi... The present study was carried out to assess the groundwater quality for drinking purposes in the Quaternary Unconsoli- dated Sedimentary Basin of the North Chengdu Plain, China. Six groups of water samples (S1, S2, S3, S4, S5, and S6) are selected in the study area. These samples were analyzed for 19 different physicochemical water quality parameters to assess groundwater quality. The physicochemical parameters of groundwater were compared with China’s Quality Standards for Groundwater (GB/T14848-93). Interpretation of physicochemical data revealed that groundwater in the basin was slightly alkaline. Total hardness and total dissolved solid values show that the investigated water is classified as very hard and fresh water, respectively. The sustainability of groundwater for drinking purposes was assessed based on the fuzzy mathematics evaluation (FME) method. The results of the assessment were classified into five groups based on their relative suitability for portable use (grade I = most suitable to grade V = least suitable), according to (GB/T 14848-93). The assessment results reveal that the quality of groundwater in most of the wells was class I, II and III and suitable for drinking purposes, but well (S2) has been found to be in class V, which is classified as very poor and cannot be used for drinking. Also, the FME method was compared with the comprehensive evaluation method. The FME method was found to be more comprehensive and reasonable to assess groundwater quality. This study can provide an important frame of reference for decision making on improving groundwater quality in the study area and nearby surrounding. Keywords Groundwater quality · Groundwater pollution · Fuzzy mathematics · Physicochemical parameters Introduction may threaten human health and plant growth (Zhu et al. 2014, Hosseini-Moghari et al. 2015). Most water contami- Groundwater is very important in day to day life for people nation originates from the disposal of solid wastes from dif- and society (Shigut et al. 2017). It has not only been used ferent human activities, such as agriculture, construction and for supplying potable water to both urban and rural areas industry, and the disposal of domestic and industrial waste- but also essential for irrigation, economic development, and water is discharged into rivers through the sewer systems protection of environmental and ecological balance (Cheng (Li et al. 2015, Zhang et al. 2015, Kumar et al. 2016). These and Fanhai 2012, Srinivas et al. 2015, Kumar et al. 2015, Al- systems may often leak wastewater into shallow aquifers, Ahmadi 2013). Recently, providing good quality water for directly or indirectly and groundwater supplies from nearby drinking is considered a fundamental requirement for public wells are affected if they are exposed to these wastewater health protection. Consequently, the poor quality of water pollutants (Li et al. 2016, Qishlaqi et al. 2017, Pinto 2015). When groundwater is polluted, its quality cannot be restored by stopping the contaminants from the sources. * Adam Khalifa Mohamed Shallow, unconsolidated aquifers are particularly vulner- adamkh124@yahoo.com able to contamination, which may persist in groundwater Dan Liu for many years or decades (Li et al. 2016, Liu et al. 2010). liudan-swju@163.com Meanwhile, it becomes necessary to monitor the groundwa- Faculty of Geoscience and Environmental Engineering, ter quality regularly and devise ways to protect it (Kaur et al. Southwest Jiaotong University, Chengdu, China 2014, Shigut et al. 2017). In the study area, the shallow aqui- Faculty of Water and Environmental Engineering, Sudan fer in the Quaternary Unconsolidated Sedimentary Basin, University of Science and Technology, Khartoum, Sudan Vol.:(0123456789) 1 3 65 Page 2 of 12 Applied Water Science (2018) 8:65 the largest source of drinking water in the North Chengdu the technique characteristics of the evaluation methods used. plain, China, is, however, constantly impacted by agricul- Therefore, this study will be an essential reference for future ture, industry, mining, and other human activities. This has studies. It will also be useful for the local decision makers in challenged the water resource managers and forced them to regional groundwater management and protection. pay attention to the evaluation of groundwater contamina- tion in the study area based on the recognized assessment methods. Materials and methods At present, many methods are available at home and abroad, and proven to be powerful in water quality assess- Description of study area ment, such as the principal component analysis (Gangopad- hyay et al. 2001), neural network model (Wu et al. 2007), The study area is located between longitudes 103°54′02″ Bayesian discrimination method (Chen et al. 2009), entropy to 104°16′54″ and latitudes 30°40′40″ to 30°57′58″ in the method (Chun-rong and Jun 2011), water pollution index North Chengdu Plain and bounded by many rivers, which method (Liu et al. 2013), grey clustering method (Zhang are tributaries of the Minjiang river and is divided into many 2014), statistical analysis method (Liu et al. 2015) and oth- villages with the total population of over 140,000. The pri- ers. However, these methods cannot directly reflect pol- mary source of drinking water and agricultural irrigation lution characteristics, and the linear relationship between in the study area is groundwater from the Unconsolidated those variables may have some effects on the results (Li Quaternary sediments, which provides water to residents. It et al. 2018). In the same vein, the methods above have their has a sub-tropical humid monsoon climate with four seasons. own merits, but they are less feasible and challenging to Compared to other areas in the same climatic zone, features popularize in the regional groundwater pollution assessment such as low temperature, less sunshine, and rainy weather are due to the complex and changeable environmental problems. more frequent. The mean annual temperature is 10.4 °C, the Furthermore, in all environmental quality assessments, there average temperature of coldest months and the hottest month is uncertainty about environmental risk, because of incon- is 4.6 and 24.4 °C, respectively. The region is dominated by sistency and peculiarities of each groundwater pollutant. NW wind, with a maximum wind speed of 17 m/s and aver- To overcome the shortcomings associated with the above age wind speed of 1.3 m/s. No typhoons are observed. The methods and respond to the call of water resource manag- annual maximum relative humidity is 80%, and the mini- ers, fuzzy mathematics evaluation (FME) method was used. mum relative humidity is 75%. The annual maximum of FME method has a large range of applications which could absolute moisture content is 15.2, and the minimum is 14.3. help in identifying and overcoming any uncertainty regard- Rainfall is the main recharge source of groundwater. The ing the risk of groundwater contamination using member- average annual precipitation is 1134.8 mm. Moreover, the ship functions (Mujumdar and Sasikumar 2002, Ma et al. longest continuous period of rainfall is 28 days. The Long- 2010, Zhang et al. 2012, Kamrani et al. 2016). It has also qiao water plant has been established on the right bank of the been proven effective to deal with complex and changeable Pi river at a distance of 38 m. The daily production capacity environmental problems (Singh et al. 2017), and also con- of the Longqiao water plant is about 10,000–12,000 m /d, trolling the effect of monitoring errors on assessment results and the daily water supply is 8000–10,000 m /d. However, (Ghasemi et al. 2014). the groundwater is exposed to the risk of contamination from This study is considered as the first of its kind to assess different sources. The main pollution sources of groundwater groundwater quality in this region and nearby surrounding. in the study area comes from the domestic sewage water, To show the advantage of FME, it was compared with Com- and industrial wastewater discharged routinely into the Pi 3 3 prehensive Evaluation Method (CEM), which is the most river, which is estimated at 8.4 × 10  m /d and agricultural common method of assessing the quality of groundwater for pollution from using chemical fertilizers and pesticide. It is drinking in China, and recommended by the quality standard believed that most of the wastewater is infiltrated into the for groundwater of China (GB/T 14848-93) (Wu and Sun shallow aquifer in the Quaternary Unconsolidated Sedimen- 2016, Su et al. 2017). This method provides a holistic view tary Basin in the area, because of its shallow depth of the of groundwater quality status and appropriateness for drink- groundwater level (2.0–10 m) and relatively high hydraulic ing purposes by considering various water quality param- conductivity (k = 10–50 m/day). So, it is necessary to evalu- eters based on simple mathematical-numerical tools. There- ate the quality of drinking water in this region since it is fore, the aims of this study are: (1) evaluating groundwater closely linked to people’s health. quality condition and its suitability for drinking purposes in the Quaternary Unconsolidated Sedimentary Basin, (2) iden- tifying the main pollutants which influence the groundwater quality and (3) compare the evaluation results to learn about 1 3 Applied Water Science (2018) 8:65 Page 3 of 12 65 analyzed. pH and total dissolved solids (TDS) were meas- Sample collection and analysis ured in situ using a portable pH and TDS meters because the parameters are likely to change during transport. Water In this study, groundwater samples were obtained from six monitoring wells (S1, S2, S3, S4, S5, and S6) from the shal- sampling methods were according to (Kent and Payne 1988). The samples were analyzed for 19 various physicochemical low aquifer in the Quaternary Unconsolidated Sedimentary Basin near the Pi river, China. These six monitoring wells parameters, include hydrogen ion concentration (pH), total hardness (TH), total dissolved solids (TDS), potassium (K ), are located between the Pi river and Longqiao drinking 2+ 2+ 2+ water supply plant. The locations of groundwater samples sodium (Na ), calcium (Ca ), magnesium (Mg ) sulfates 2− − − ( SO ), chlorides (Cl ), bicarbonates ( HCO ), pot assium are displayed in (Fig. 1). The sampling wells (S1, S2, and 4 3 − − S3) are inside the wall of Longqiao water plant, while the permanganate index (C OD ), nitrate ( NO ), nitrite ( NO ), Mn 3 2 ammonia ( NH ), iron (Fe), manganese (Mn), arsenic (As), sampling wells (S4, S5, and S6) outside the Longqiao water 6+ plant wall. Well (S2) is near the septic tank of the plant chromium (Cr ), and lead (Pb). These parameters were used as index indicators to evaluate the groundwater contamina- workers, and the wells (S1 and S3) are near the manage- ment offices and workers residences, respectively. Whereas, tion risk in the study area. This selection was based on their importance to the water quality and the potential impact on the well (S4) is near the Pi river bridge and the sampling wells (S5 and S6) are in the middle of farms existing in the human health and their pollution potential on groundwater resource in particular (Zhang et al. 2017). Detection methods area. These wells are used as a group in the Longqiao water plant to provide supply drinking water to residents in the for each parameter were as follows: the concentrations of K + 2+ and Na were measured using flame photometer. TH, Ca , Longqiao town and its surrounding area. It was observed 2+ − that the location of wells is exposed to the risk of ground- Mg, Cl , and HCO were analyzed by titrimetric methods. + − − 2− The concentrations of C OD , NH , NO , NO , and SO water pollution resulting from the domestic sewage water Mn 4 3 2 4 and industrial wastewater discharged into the Pi river. Also, were determined using spectrophotometer technique. And 6+ finally, the Fe, Mn, As, Cr , and Pb concentrations were unsuitable use of agricultural chemicals, unhealthy open def- ecation and many more potentially infiltrate into the shallow then measured using atomic absorption spectrometry. In this study, the groundwater suitability for drinking and aquifer. This built the foundation for selecting the location for research samples to evaluate because they have relative domestic purposes was evaluated by complying the values of various groundwater quality parameters according to China’s importance as source of drinking water in the region. The samples were collected after 10 min of pumping and stored Quality Standards for Groundwater (GB/T14848-93), since it is the only way to assess groundwater quality in China by in clean 500-mL glass bottles that were thoroughly washed with detergent and rinsed with deionized water. The samples specifying the classification of groundwater quality, ground- water quality monitoring, evaluation methods and ground- were sent to the laboratory of Environmental Engineering Center of Sichuan Geological Engineering Investigation water quality protection (Chinese 1993). It also corresponds to the methods applied to assess the quality of groundwater and kept in a refrigerator at a temperature below 4 °C until Fig. 1 Location of study area and groundwater samples 1 3 65 Page 4 of 12 Applied Water Science (2018) 8:65 − − − + 6+ in this paper by dividing the groundwater quality into five Cl, COD , NO , NO , NH , Fe, Mn, As, Cr , and Pb), Mn 3 2 4 categories: excellent, good, moderate, poor and very poor. which represent the overall situation of groundwater quality of the Quaternary Unconsolidated Sedimentary Basin. The Principle for fuzzy mathematics evaluation method element u (i =1, 2,…n) is measured value of pollutants that affect the quality of groundwater. Fuzzy mathematics was proposed by Zadeh in 1965 (Mahapatra et al. 2011, Zhang et al. 2009) as a new way to Determining the evaluation standard represent vagueness in everyday life. This method is proven to be capable of dealing with complex systems under uncer- This paper adopts the Chinese national standards of ground- tain and imprecise conditions (Gharibi et al. 2012, Singh water quality (GB/T 14848-93) as the evaluation standard, et al. 2017). Risk assessment of groundwater contamination which was drafted by the Ministry of Geology and Min- could be a challenge because it often involves many ground- eral Resources of the People’s Republic of China. Based on water quality parameters. Fuzzy mathematics can simplify groundwater quality in China and human health requirement this risk assessment process (Zhang et al. 2012). It takes the as well as the objective of water protection, the standard effectual measurement of pollutant concentration compared (GB/T 14848-93) classifies groundwater quality into five with its evaluation criteria. Through accepted linear func- grades (I, II, III, IV, and V). The quality evaluation grades tion, it calculates each pollution element of membership on of water V =(v , v ,…v ) are represented by v (i =1, 2,…m), 1 2 m i the level of groundwater contamination. Afterwards, fuzzy and it is the standard classification value of groundwater matrix can be set, and the weight of each single element for any contamination, which includes five levels: excellent value can be obtained by calculation, which constitutes the (grade I), good (grade II), moderate (grade III), poor (grade weight factors matrix. Last, the membership matrix and the IV), and very poor (grade V). Groundwater with grades I weight factors matrix are multiplied, and the evaluation and II is of excellent and good quality and is suitable for all results can be derived (Agoubi et al. 2016, Feng et al. 2012). uses. Grade III is moderate-quality water, which is generally Figure 2 shows the steps of creating the fuzzy mathematics suitable for drinking, irrigation, and industrial production. model. Grade IV is poor quality water, which is fit only for irriga- tion and industrial production and may be used for drinking Determining the evaluation factors after proper treatment. Grade V groundwater is very poor quality water that cannot be used for any purpose (Chinese Depending on the circumstances of environment, the evalu- 1993). The classification of these grades about each evalua- 2− ation factors set U = (u , u ,…,u ) = (pH, TH, TDS, SO , tion factor is given in Table 1. 1 2 n Table 1 Classification of groundwater quality based on the Chinese national standard (GB/T 14848-93) Parameters Grades I II III IV V pH ≤ 6.5 ≤ 7.0 ≤ 7.5 ≤ 8.0 > 8.5 TH (mg/L) ≤ 150 ≤ 300 ≤ 450 ≤ 550 > 550 TDS (mg/L) ≤ 300 ≤ 500 ≤ 1000 ≤ 2000 > 2000 SO (mg/L) ≤ 50 ≤ 150 ≤ 250 ≤ 350 > 350 Cl (mg/L) ≤ 50 ≤ 150 ≤ 250 ≤ 350 > 350 COD (mg/L) ≤ 1.0 ≤ 2.0 ≤ 3.0 ≤ 10 > 10 Mn NO (mg/L) ≤ 2.0 ≤ 5.0 ≤ 20 ≤ 30 > 30 NO (mg/L) ≤ 0.001 ≤ 0.01 ≤ 0.02 ≤ 0.1 > 0.1 NH (mg/L) ≤ 0.02 ≤ 0.02 ≤ 0.2 ≤ 0.5 > 0.5 Fe (mg/L) ≤ 0.1 ≤ 0.2 ≤ 0.3 ≤ 1.5 > 1.5 Mn (mg/L) ≤ 0.05 ≤ 0.05 ≤ 0.1 ≤ 1.0 > 1.0 As (mg/L) ≤ 0.005 ≤ 0.01 ≤ 0.05 ≤ 0.05 > 0.05 6+ Cr (mg/L) ≤ 0.005 ≤ 0.01 ≤ 0.05 ≤ 0.1 > 0.1 Pb (mg/L) ≤ 0.005 ≤ 0.01 ≤ 0.05 ≤ 0.1 > 0.1 Classification Excellent Good Moderate Poor Very poor Fig. 2 Flow chart of the fuzzy mathematics method 1 3 Applied Water Science (2018) 8:65 Page 5 of 12 65 Determining the weights of factors where each element of X is mapped to a value between 0 and 1. This value is referred to as membership value or degree The weights of factors are important elements in the math- of membership, and it is used to determine the degree of ematical model of FME technique, which reverses the posi- membership of each rating factor. Hence, the fuzzy set A is tion and role of each factor in the measures of comprehen- defined by its MF: sive decision making, and the result of the comprehensive (x)=  (x) , x ∈ X,  (x)∈ [0, 1] . (5) A A A evaluation is directly affected by it. The equations for weight The membership function sets are represented through is as follows (Zhang 2014): triangular, trapezoidal, Gaussian, Pseudo exponential, Sig- moidal and other shapes of fuzzy numbers (Miao et al. 2014, W = , (1) Srinivas et al. 2017). Generally, water quality parameter’s impact is represented by certain range of values, and the firing level of the conclusion is computed as the product of S =  , i ij (2) dismissal levels from the antecedent (Agoubi et al. 2016). j=1 In this study, the triangular membership function is used to normalize the crisp inputs because of its simplicity and where C is the measured values of index i, S is the standard i i computational efficiency and provide an environment more value for index i, n is the grading number of water quality conductive to human-in-the-loop knowledge acquisition standard, and α is the jth sample value under the ith level ij (Mahapatra et al. 2011, Caniani et al. 2015). It can be rep- of classification factor. To make the fuzzy compositional resented mathematically for any of the fourteen ground- operation, the weight of each single factor must be normal- water quality parameters with respect to five classification ized as follows: grades (I, II, III, IV, and V) as follows (Lermontov et al. 2009):when j = 1, W = × , C (3) i ⎧ i 1 x ≤ 𝛼 ij 𝛼 −x ij+1 i=1 𝛼 < x <𝛼 𝜇 (x) = , (6) A ⎨ ij ij+1 𝛼 −𝛼 ij+1 ij 0 x >𝛼 ⎩ ij+1 where W represents the normalized weight of the evaluated index i. Based on the above equation, the weight set of the single when j = 2, 3, 4 factor can be determined A ={ w , w , … w }. 2 i ⎧ 0 x <𝛼 ij−1 𝛼 −x ij+1 Determination of the membership and relation matrix R 𝛼 ≤ x <𝛼 ij ij+1 𝛼 −𝛼 ij+1 ij 𝜇 (x) = x−𝛼 , (7) A ij−1 𝛼 ≤ x <𝛼 ij−1 ij A fuzzy set is completely characterized by its membership 𝛼 −𝛼 ij ij−1 function (MF). The (MF) has been used to assess ground- 0 x <𝛼 ij+1 water quality according to the standards. The level of mem- bership belongs to a type of fuzzy information which over- when j = 5 comes the differences among water index grade standards in different places (Zhang 2014). The FME begins with the 0 x <𝛼 ij−1 concept of a fuzzy set. The fuzzy set describes the relation- x−𝛼 ij−1 𝛼 < x <𝛼 𝜇 (x) = , (8) ship between an uncertain quantity (x) and a membership ⎨ ij 𝛼 −𝛼 ij−1 ij ij−1 function (μ). The elements of fuzzy set membership may 1 x ≥ 𝛼 ⎩ ij be described as a number in the interval [0, 1] (Nasr et al. 2012). The greater the value of membership, the higher the ,  and  are the die ff rent levels of groundwater qual - ij ij−1 ij+1 membership qualifications. When the value of membership ity standards and x is the real measured concentration of is 1, it is subordinated completely, and when the value of each factor. membership is 0, it is subordinated incompletely. The mem- The membership function can be described of five twin bership degree of the fuzzy set is defined over a domain X grades and the fuzzy relationship matrix R is formed as which takes the form: follows: ∶ X → [0, 1], A (4) 1 3 65 Page 6 of 12 Applied Water Science (2018) 8:65 where n denotes the number of indices selected for the r r ⋯ r ⎡ 11 12 1m ⎤ assessment, F is the value of CEM for a given sample, and F ⎢ ⎥ r r ⋯ r 21 22 2m R = . represents the assigned value for the ith index by the quality (9) ⎢ ⎥ ⋮⋮ ⋮ ⎢ ⎥ standard for groundwater (GB/T 14848-93). F is the average ⎣ ⎦ r r ⋯ r n1 n2 nm value of each individual component score of F , and (F ) i i max is the maximum value of the individual component evalu- Fuzzy evaluation of groundwater quality ation score F . Once F is determined, groundwater quality classification can be obtained as per the criteria in Table  3 The fuzzy evaluation matrix B for groundwater quality (Wu and Sun 2016). evaluation is obtained by the compositional process of the weight A and the fuzzy evaluation matrix R of each single Results and discussion grade for I, II, III, IV, and V respectively. The principle of FME method can be described by the mathematical Eq. (10), Groundwater quality parameters with the main purpose of weighting evaluation factors (Hao et al. 2012). The concentration values of groundwater quality parameters obtained from the six wells in the study area were statistically r r ⋯ r ⎡ ⎤ 11 12 1m analyzed. It should be noted that for each well, four samples ⎢ ⎥ r r ⋯ r 21 22 2m ⎢ ⎥ were taken at different periods and there was no apparent B = A × R ={ w , w , … w }× 1 2 i ⎢ ⎥ ⋮⋮ ⋮ change in the analysis results of these samples and, therefore, ⎢ ⎥ ⎣ r r ⋯ r ⎦ it was taken as the average. The results represented as maxi- n1 n2 nm (10) mum, minimum, mean and standard deviation for the main groundwater quality parameters are shown in Table 4. Their From the principle of maximum membership, the grade levels were compared with the acceptable limits recommended of groundwater quality is determined. by the national quality standards for groundwater of China (GB/T 14848-93) to see if they were suitable for drinking. Comprehensive evaluation method As shown in Table 5, the pH of the groundwater is within the range of 6.7–7.5 with the mean of 7.27, indicating that The comprehensive evaluation method (CEM) is based on the groundwater in the study area is slightly alkaline for most the classification of individual component to determine of the groundwater samples. All samples fall within limits the quality category required as in Table  2 to determine (6.5 and 8.5) of quality standard for groundwater of China the individual component evaluation score of Fi, and press (GB/T 14848-93). The average concentrations of water the comprehensive evaluation to determine the value of F. + 2+ 2+ K, Ca, Mg , and HCO were 2.018, 92.34, 17.23, and The evaluation is carried out using the quality standard for 299.0 mg/L, respectively, and no limits set for these param- groundwater of China (GB/T 14848-93) as classified in the eters have been stated of quality standard for groundwater of FME technique. (Chinese 1993). As per standard, CEM can China (GB/T 14848-93). The N a concentration was deter- be computed as follows (Su et al. 2017): mined between 6.00 and 12.00  mg/L with an average of 9.417 mg/L. All samples of groundwater are found below the 2 2 F + F max (11) permissible limit of (GB/T 14848-93). The concentrations of F = (i = 1, 2, 3, … n), TH and TDS range from 260.20 to 355.30 mg/L and 331.80 to 415.40 mg/L, with an average of both 301.50 and 368.97, n respectively. This indicates that the quality of all ground- F = F (i = 1, 2, 3, … n), (12) i=1 water in the study area is classified as very hard to TH and fresh water to TDS, respectively, according to (Todd 1980, Sawyer and McCarty 1967). In this study, the concentra- Table 2 Individual component classification tions of (TH and TDS) are found below the allowable limit Groundwater quality categories I II III IV V of (GB/T 14848-93) for all groundwater samples. At the 2− − − same time, the concentrations of SO , Cl, COD , NO , Mn 4 3 Value for Fi 0 1 3 6 10 − + NO , and NH in groundwater range from 52.83 to 68.20, 2 4 Table 3 Standards of Level Excellent Good Moderate Poor Very poor the groundwater quality classification F < 0.80 0.80~< 2.50 2.50~< 4.25 4.25~< 7.20 > 7.20 1 3 Applied Water Science (2018) 8:65 Page 7 of 12 65 Table 4 Descriptive statistics of Parameters Unit Min Max Mean St.dev Standard (Class III) groundwater quality variables for six sampling wells in the pH – 6.70 7.50 7.27 0.29 6.5–8.5 study area TH mg/L 260.20 355.30 301.50 42.11 ≤ 450 TDS mg/L 331.80 415.40 368.97 36.67 ≤ 1000 K mg/L 1.50 2.30 2.018 0.34 – Na mg/L 6.00 12.00 9.417 2.25 ≤ 200 2+ Ca mg/L 77.15 111.20 92.34 15.22 – 2+ Mg mg/L 15.81 18.85 17.23 1.13 – 2− SO mg/L 52.83 68.20 60.22 6.644 ≤ 250 Cl mg/L 15.96 20.92 17.61 2.304 ≤ 250 HCO mg/L 250.20 347.80 299.00 40.06 – COD mg/L 0.23 0.29 0.25 0.02 3 Mn NO mg/L 1.17 3.96 2.40 0.92 ≤ 20 NO mg/L 0.003 0.005 0.004 0.0006 ≤ 0.07 NH mg/L 0.020 0.68 0.14 0.26 ≤ 0.2 Fe mg/L 0.086 2.23 0.59 0.82 ≤ 0.3 Mn mg/L 0.01 0.21 0.07 0.08 ≤ 0.1 As mg/L 0.0002 0.0002 0.0002 0.00 ≤ 0.05 6+ Cr mg/L 0.002 0.004 0.003 0.001 ≤ 0.01 Pb mg/L 0.005 0.01 0.008 0.002 ≤ 0.05 Note: Standard refers to the quality standard for groundwater developed by China (GB/T14848-93) (Chi- nese 1993) All units of parameters are in mg/L except pH, St.dev standard deviation Table 5 The weight value of groundwater quality parameters in well No. S2 Weight Factors 6+ PH TH TDS SO Cl COD NO NO NH Fe Mn As Cr Pb 4 Mn 3 2 4 W 0.974 0.732 0.297 0.068 0.071 0.039 0.064 0.089 2.363 2.720 0.389 0.005 0.043 0.079 0.123 0.092 0.037 0.009 0.009 0.005 0.008 0.011 0.298 0.343 0.049 0.001 0.005 0.010 15.96 to 20.92, 0.23 to 0.29, 1.17 to 3.96, 0.003 to 0.005 and exceed the permissible limit represented in sites (S2, S4) 0.020 to 0.68 mg/L, respectively, with a cumulative average and for Mn concentration, one sample exceeds the permis- for each concentration of 60.22, 17.61, 0.25, 2.40, 0.004 and sible limit at the site (S2). According to the mean values, the 2+ 2+ + + 0.14 mg/L, respectively. The values of these concentrations dominance of cations is Ca > Mg > Na > K and anions − 2− − for all the groundwater samples collected in this study are is HCO > SO > Cl , which makes the predominant type of 3 4 within the permissible limit of (GB/T 14848-93) for ground- groundwater to be Ca–Mg-HCO type. As mentioned above, water quality except sample S2 of NH which exceeds the the groundwater quality indicators exceeding the standard ratio acceptable limit of 0.2 mg/L according to the national quality in the study area were mainly Fe, Mn, and NH for sampling standards for groundwater (GB/T 14848-93). Finally, the con- points S2 and S4 (Fig. 3). 6+ centrations of Fe, Mn, As, Cr , and Pb in groundwater were 0.086–0.23 mg/L with the mean 0.59 mg/L, 0.01–0.21 mg/L Fuzzy mathematics evaluation results of study area with a mean of 0.07 mg/L, 0.0002–0.0002 mg/L with the mean 0.0002 mg/L, 0.002–0.004 mg/L with a mean 0.003 mg/L and According to the above-mentioned relevant principles of the 0.005–0.01 mg/L with an average of 0.008 mg/L, respectively. FME technique for the assessment of groundwater quality of Among them, Fe and Mn are found to exceed the permissi- the study area, the steps for evaluation results can be listed ble limits of drinking water as set by (GB/T 14848-93) of 0.3 as follows: and 0.1 mg/L, respectively, and remaining concentrations fall within the permissible limits of (GB/T 14848-93) for all groundwater samples. For Fe concentration, two samples 1 3 65 Page 8 of 12 Applied Water Science (2018) 8:65 2.5 Weight of each evaluation factors 2.0 According to Eqs. (1) (2) and (3), the corresponding weight NH Fe values of all wells in the study area were obtained, using the Mn (GB/T14848-93) standard (Table 1) and data from Table 4 for 1.5 the fourteen selected indicators. The weight values for well No. S2 are shown in Table 5, as an example. 1.0 The membership and relation matrix of different evaluation 0.5 factors on various grades 0.0 S1 S2 S3 S4 S5 S6 Based on Eqs. (6) (7) and (8), the degree of membership of Well name each indicator to the groundwater quality grade is calculated. Each indicator is computed to have five levels of membership, Fig. 3 Groundwater quality indicators exceeding the standard ratio in and the fourteen selected indicators can get fourteen sets of the study area numerical values. Accordingly, the corresponding fuzzy rela- tionship matrix R is achieved from the selected indicators of all Evaluation factors wells in the study area which was computed. The membership for well No. S2 is shown in Table 6, as an example. In this study, a FME technique was used to assess the This membership can be expressed by fuzzy relationship groundwater quality of the quaternary unconsolidated matrix, as follows: sedimentary basin near the Pi river according to quality 0 0.20 0.80 0 0 ⎡ ⎤ evaluation parameters and five classifications of groundwa- ⎢ ⎥ 0 0.998 0.002 0 0 ter quality based on the Chinese national standard (GB/T ⎢ ⎥ 0.63 0.37 0 0 0 ⎢ ⎥ 14848-93) (Table 1). The groundwater quality was assessed ⎢ 10 00 0 ⎥ for six wells (S1, S2, S3, S4, S5, and S6) as selected in ⎢ ⎥ 10 00 0 the study area. Of the 19 groundwater parameters analyzed, ⎢ ⎥ 10 00 0 ⎢ ⎥ 14 parameters (pH, TH, TDS, SO , Cl, COD, NO, NO , 4 Mn 3 2 ⎢ ⎥ 10 00 0 6+ NH , Mn, Fe, As, Cr and Pb) were selected due to the R = . ⎢ ⎥ 0.56 0.44 0 0 0 ⎢ ⎥ fact the rest did not have the five-level division as required 0 0 0 0.1 0.9 ⎢ ⎥ in the China’s standard for groundwater quality evaluation ⎢ ⎥ 00 00 1 (GB/T 14848-93). This selection is also based on the fact ⎢ ⎥ 0 0 0.88 0.12 0 ⎢ ⎥ that it is periodically monitored by the Local Environmental 10 00 0 ⎢ ⎥ Protection Department for their vital importance to the water ⎢ ⎥ 10 00 0 quality and potential influence on human health. ⎢ ⎥ 10 00 0 ⎣ ⎦ Table 6 Membership of well Well Parameters I II III IV V No. S2 S2 pH 0 0.20 0.80 0 0 TH 0 0.998 0.002 0 0 TDS 0.63 0.37 0 0 0 SO 1 0 0 0 0 Cl 1 0 0 0 0 COD 1 0 0 0 0 Mn NO 1 0 0 0 0 NO 0.56 0.44 0 0 0 NH 0 0 0 0.1 0.9 Fe 0 0 0 0 1 Mn 0 0 0.88 0.12 0 As 1 0 0 0 0 6+ Cr 1 0 0 0 0 Pb 1 0 0 0 0 1 3 Concentration mg/L Applied Water Science (2018) 8:65 Page 9 of 12 65 Fuzzy evaluation of groundwater quality 0.6 Based on Eq. (10), the fuzzy evaluation matrix B for ground- 0.5 water quality evaluation was obtained by the compositional process of the weight matrix A and the fuzzy evaluation 0.4 matrix R. For example, the fuzzy evaluation matrix B of the 0.3 well No. S2 is as follows: B = (0.076, 0.136, 0.141, 0.036, 0.611). 0.2 Excellent According to the principle of maximum membership to Good Moderate 0.1 determine the groundwater quality level, 0.611 is the maxi- Poor Very poor Final result mum of all five values. Therefore, well No. S2 was found to 0.0 belong to grade V, which is classified as very poor and their S1 S2 S3 S4 S5 S6 Well name water quality cannot be used for any purpose. The degrada- tion in groundwater quality at this well is mainly due to high Fig. 4 Comprehensive assessment of fuzzy technique for assessing concentrations of NH and Fe (see Table 6). Similarly, the groundwater quality groundwater quality grades of the other wells in the study area were obtained. The results are shown in Table 7. From Table 7 and Fig. 4, the results of fuzzy evaluation According to our field investigation in the study, the very matrix B for the other wells showed that the groundwater poor groundwater quality of well No. S2 could be attributed quality of well No. S1 is (0.579, 0.421, 0, 0, 0), and it found to several reasons: to belong to grade I, which is classified as excellent and their water quality is considered to be suitable for vari- 1. In the Basin of Minjiang as Pi river is a tributary of ous purposes. The groundwater quality of wells No. S3 is Minjiang river, sediment has a certain amount of Fe and (0.379, 0.477, 0.144, 0, 0), and categorized as being grade Mn under the natural environmental conditions (Zeng II, which is classified as good, and was also deemed to be Jichuan 2009, Li et al. 2006). Thus, the over-limit ratio suitable for all uses. Finally, the groundwater quality of wells of Fe and Mn is considered natural environment. No. S4, S5 and S6 are (0.264, 0.239, 0.466, 0.032), (0.214, 2. The sample No. S2 was taken from the water plant, 0.318, 0.468, 0, 0), (0.207, 0.319, 0.473, 0, 0), respectively, which is closest to the populated area, and the infiltra- which fall under grade III and is classified as moderate, and tion of the plant domestic wastewater may lead to the their water quality is generally suitable for drinking, irriga- increase of NH content. In essence, the groundwater tion, and industrial production. From the six groundwater quality of the well S2 is mainly influenced by human samples, grade I of groundwater occupies 16.67%, grade activities. II 16.67% and grade III 50% that means about 83.33% of 3. Excessive discharge of industrial waste water and the groundwater samples could be used as drinking water domestic sewage is the main cause of groundwater pol- source. However, S2 is grade V (16.67%), which cannot be lution in the study area. Considering the aforementioned used for drinking. This suggests that the groundwater quality reasons, well No.2 is deemed contaminated and, there- in the study area, in general, is not bad. Through analysis of fore, should be monitored periodically and protected the water indicators, the elevated groundwater quality indi- from the causes of pollution to avoid being consumed cators were mainly NH , Fe, and Mn which have resulted in by local residents. the evaluation of well No. S2 as class V which is very poor. Table 7 The fuzzy evaluation of Well name I II III IV V Results groundwater quality S1 0.579 0.421 0 0 0 I S2 0.076 0.136 0.141 0.036 0.611 V S3 0.379 0.477 0.144 0 0 II S4 0.264 0.239 0.466 0.032 0 III S5 0.214 0.318 0.468 0 0 III S6 0.207 0.319 0.473 0 0 III 1 3 Grade value 65 Page 10 of 12 Applied Water Science (2018) 8:65 Groundwater quality evaluation based on comprehensive evaluation method In the study area, groundwater is a vital source of drinking water for residents. The CEM has been used to assess the qual- ity of groundwater of the Quaternary Unconsolidated Sedi- mentary Basin near the Pi river and to demonstrate the advan- tage of this study. In this method, the same physicochemical water quality parameters that were chosen in the FME method (14 parameters) have been used for six wells (S1, S2, S3, S4, S5, and S6). The results are shown in Table 8. As shown in Table 8 and Fig. 5, F values of all analyzed samples in the S1 S2 S3 S4 S5 S6 Well name study area differ from 0.82 to 7.29, ranging from good quality to very poor quality. The results of the assessment showed Fig. 5 Results of groundwater quality evaluation based on compre- that four groundwater samples (66.67% of all samples) are hensive evaluation method classified as good quality water (grade II) which is suitable for various purposes. Two groundwater samples (33.33% of all boundary water quality index of water quality is, hence making collected samples) are classified as poor and very poor quality the evaluation more comprehensive and reasonable. Whereas, water (grade IV and V), respectively, which are classified as the CEM of water highlights the largest factor of pollution as unsuitable for drinking. The common contaminants in these the index classification is based on the binary logic. Therefore, samples are NH , Fe, and Mn, which are mainly from the natu- they cannot describe the continuity of environmental quality, ral environment, industrial and agricultural activities. Accord- and cannot objectively reflect the influence of the index value ing to the results of the CEM, four groundwater samples (S1, near the water quality grade limit for quality evaluation and S3, S5, and S6) could be used as drinking water source, while classification. (S2 and S4) cannot be used as drinking water source. Despite the FME and CEM generated almost similar results to the holistic picture of groundwater quality in the study area, Comparison between fuzzy mathematics evaluation the fuzzy indicator is recommended as the more useful indica- technique and comprehensive evaluation method tor for the following advantages: 1. Ability to describe a wide variety of non-linear relation- The present study analyzed groundwater quality status of six ships. 2. They tend to be simple since they are based on a set of wells in the Quaternary Unconsolidated Sedimentary Basin local simple models. 3. Fuzzy mathematics can deal with and using FME technique and CEM. Based on the results of the process missing data without compromising the final result. assessment obtained from both methods, the holistic picture of 4. Avoiding artificial precision as well as generating results groundwater quality within the study area was satisfactory and which are more consistent with the ecological complexity of consistent with the actual situation of the study area under the real-world issues. 5. Combining both qualitative and quanti- prevailing conditions. The indices for both methods indicated tative information to express the ecological status of the case that the groundwater quality in the study area was suitable for study, which is a unique capability of fuzzy approach. drinking at ratio 83.34 and 66.67%, respectively. The FME In general, the results revealed that the knowledge-based showed only the well No. 2 to be unsuitable for drinking water models such as FME method were practical and flexible tools (see Table 7), while the CEM showed wells No. 2 and 4 to for incorporating the experts’ attitudes and modeling the cur- be unsuitable for drinking water (see Table  8). This is due rent uncertainties associated with water resources and envi- to the technical difference of both methods. For instance, the ronmental perplexities. However, in the FME technique, the FME approach uses membership degree to describe the limit weight of evaluating indicators is determined by the moni- between different pollution degrees for assessing groundwater toring data compared to groundwater quality standard. As a quality. The FME technique takes into account the impression result, when an abnormal value appears at some evaluating of each assessment factor on the evaluation result and deter- indicator, the condition of overestimating the weight of these mines the major pollutants according to the weights of evalua- indicators would lead to unrealistic evaluation results which tion factors. This reflects how close the actual concentration of Table 8 Groundwater quality Well name S1 S2 S3 S4 S5 S6 classification based on comprehensive evaluation F 0.82 7.29 2.21 4.3 2.35 2.27 method Grade II V II IV II II 1 3 F value Applied Water Science (2018) 8:65 Page 11 of 12 65 may not be in line with the actual situation of the studied area conditions accurately and precisely, specifically for the (Zou et al. 2006). North Chengdu Plain, China. The degree of groundwater pollution risk has a direct con- nection to the water discharge and environmental vulnerabil- Conclusions ity of the region. To improve the status of groundwater and thoughtful scientific planning of groundwater extraction, it Groundwater pollution is a vague concept because there are is necessary to strictly control the industrial wastewater and often no clear-cut boundaries that separate a “polluted” from domestic sewage discharge not only in the North Chengdu an “unpolluted” sample. It is, therefore, necessary to develop Plain but also in other watersheds. a new method based on a fuzzy technique to give solutions Acknowledgements This research was supported by the Fundamental that are robust and have a high level of confidence. Research Funds for the Central Universities (2682015CX020). In this study, six groundwater samples were collected, analyzed and assessed for drinking water quality in the Open Access This article is distributed under the terms of the Crea- Quaternary Unconsolidated Sedimentary Basin near the Pi tive Commons Attribution 4.0 International License (http://creat iveco river. The pH value of the groundwater was slightly alka-mmons.or g/licenses/b y/4.0/), which permits unrestricted use, distribu- tion, and reproduction in any medium, provided you give appropriate line for most of the groundwater samples to basic in nature. credit to the original author(s) and the source, provide a link to the The groundwater is classified as very hard and fresh water Creative Commons license, and indicate if changes were made. based on TH and TDS, respectively. Groundwater quality parameters were compared with the acceptable limits recom- mended by the national quality standards for groundwater of China (GB/T 14848-93). From all groundwater parameters References analyzed, NH , Fe, and Mn were above the permissible lim- its of (GB/T 14848-93). The sequence of the abundance of Agoubi B, Souid F, Kharroubi A, Abdallaoui A (2016) Assessment of 2+ 2+ + + hot groundwater in an arid area in Tunisia using geochemical and major ions is found in the order of Ca > Mg > Na > K − 2− − fuzzy logic approaches. Environ Earth Sci 75:1497 and anions is HCO > SO > Cl . 3 4 Al-Ahmadi ME (2013) Hydrochemical characterization of groundwater The FME technique results show that five of the sam- in wadi Sayyah, Western Saudi Arabia. Appl Water Sci 3:721–732 pled groundwater (S1, S3, S4, S5, and S6) are suitable of Caniani D, Lioi D, Mancini I, Masi S (2015) Hierarchical classifica- tion of groundwater pollution risk of contaminated sites using drinking water directly while the well (S2) is unsuitable for fuzzy logic: a case study in the Basilicata region (Italy). Water drinking unless treated. The common contaminants in these 7:2013–2036 samples are NH , Fe, and Mn, which are mainly from the Chen H, Li X, Liu A (2009) Studies of water source determination natural environment, industrial and agricultural activities. method of mine water inrush based on Bayes’ multi-group step- wise discriminant analysis theory. Rock Soil Mech 30:3655–3659 Thus, authorities should give more attention to the pollution Cheng Y, Fanhai M (2012) GW vulnerability assessment based on of NH , Fe, and Mn to prevent deterioration of good water entropy and fuzzy method. Proc Inst Civ Eng 165:277 quality by taking some effective measures that are required Chinese S. (1993) Quality Standard for Groundwater (GB/T 14848– to enhance the drinking water quality by delineating an 93). AQSIQ (General Administration of Quality Supervision, Inspection and Quarantine of the People’s Republic of China) effective water quality management plan. This method may (in Chinese) play an important role in decision-making for the drinking Chun-rong J, Jun Z (2011) Based on fuzzy weight matter element to water quality assessment because it proved effective in solv - evaluate the water quality of Jialing River in Nanchong, China. ing problems of fuzzy boundaries. So, it is more objective Proc Environ Sci 11:631–636 Feng L, Daohan W, Sijing F, Suzhen Y (2012) Study on the application and scientific and reliable in practice. of fuzzy mathematics in assessing the water quality of Qinghe The FME method was compared with CEM in this study. Reservoir. In: Computer science and electronics engineering Based on the results of the assessment obtained from both (ICCSEE), 2012 international conference on, pp 672–675. IEEE methods, the holistic picture of groundwater quality within Gangopadhyay S, Das Gupta A, Nachabe M (2001) Evaluation of ground water monitoring network by principal component analy- the study area was satisfactory and consistent with the actual sis. Groundwater 39:181–191 situation of the study area under the prevailing conditions. Gharibi H, Mahvi AH, Nabizadeh R, Arabalibeik H, Yunesian M, However, the fuzzy indicator is recommended as a more Sowlat MH (2012) A novel approach in water quality assessment practical indicator for assessing groundwater quality owing based on fuzzy logic. J Environ Manage 112:87–95 Ghasemi E, Amini H, Ataei M, Khalokakaei R (2014) Application of to the introduction of membership degree and weight of each artificial intelligence techniques for predicting the flyrock distance factor to the models. caused by blasting operation. Arab J Geosci 7:193–202 The study can provide an important frame of reference to Hao W, Hanting Z, Wenjuan X, Jinling Z (2012) Evaluation of ground- the government decision-making on improving groundwater water quality using improved fuzzy comprehensive assessment based on AHP. Inter J Appl Sci Eng Res 2:377–384 quality in the study area. Additionally, this study may serve as a guide for future researchers to assess the groundwater 1 3 65 Page 12 of 12 Applied Water Science (2018) 8:65 Hosseini-Moghari S-M, Ebrahimi K, Azarnivand A (2015) Groundwa- Pinto SR (2015) Fuzzy logic based assessment of periodic variation ter quality assessment with respect to fuzzy water quality index of water quality of Nethravathi River in Dakshina Kannada dis- (FWQI): an application of expert systems in environmental moni- trict. Int J Innovative Res Electrical Electron Instrum Control Eng toring. Environ Earth Sci 74:7229–7238 3:321–326 Kamrani S, Rezaei M, Amiri V, Saberinasr A (2016) Investigating the Qishlaqi A, Kordian S, Parsaie A (2017) Hydrochemical evaluation of efficiency of information entropy and fuzzy theories to classifi- river water quality—a case study. Appl Water Sci 7:2337–2342 cation of groundwater samples for drinking purposes: Lenjanat Sawyer CN, McCarty PL (1967) Chemistry for sanitary engineers. Plain, Central Iran. Environ Earth Sci 75:1370 Chemistry for sanitary engineers. McGraw-Hill, New York Kaur BJ, George M, Mishra S (2014) Groundwater quality and water Shigut DA, Liknew G, Irge DD, Ahmad T (2017) Assessment of quality index of Delhi city, India. World Appl Sci J 32:865–871 physico-chemical quality of borehole and spring water sources Kent R, Payne K (1988) Sampling groundwater monitoring wells: spe- supplied to Robe Town, Oromia region, Ethiopia. Appl Water cial quality assurance and quality control considerations, In: Keith Sci 7:155–164 LH (ed) Principles of environmental sampling. ACS Professional Singh AP, Chakrabarti S, Kumar S, Singh A (2017) Assessment of air Reference Book, ACS, Washington, DC, pp 231–246 quality in Haora River basin using fuzzy multiple-attribute deci- Kumar P, Bansod BK, Debnath SK, Thakur PK, Ghanshyam C (2015) sion making techniques. Environ Monit Assess 189:373 Index-based groundwater vulnerability mapping models using Srinivas R, Bhakar P, Singh AP (2015) Groundwater quality assess- hydrogeological settings: a critical evaluation. Environ Impact ment in some selected area of Rajasthan, India using fuzzy multi- Assess Rev 51:38–49 criteria decision making tool. Aquat Proc 4:1023–1030 Kumar P, Thakur PK, Bansod BK, Debnath SK (2016) Assessment of Srinivas R, Singh AP, Sharma R (2017) A scenario based impact the effectiveness of DRASTIC in predicting the vulnerability of assessment of trace metals on ecosystem of river Ganges using groundwater to contamination: a case study from Fatehgarh Sahib multivariate analysis coupled with fuzzy decision-making district in Punjab, India. Environ Earth Sci 75:879 approach. Water Resour Manage 31:4165–4185 Lermontov A, Yokoyama L, Lermontov M, Machado MAS (2009) Su H, Kang W, Xu Y, Wang J (2017) Evaluation of groundwater quality River quality analysis using fuzzy water quality index: Ribeira do and health risks from contamination in the north edge of the Loess Iguape river watershed, Brazil. Ecol Ind 9:1188–1197 Plateau, Yulin City, Northwest China. Environ Earth Sci 76:467 Li X-D, Masuda H, Ono M, Kusakabe M, Yanagisawa F, Zeng H-A Todd D (1980) Groundwater hydrology. Wiley, New York, p 535 (2006) Contribution of atmospheric pollutants into groundwater Wu J, Sun Z (2016) Evaluation of shallow groundwater contamination in the northern Sichuan Basin, China. Geochem J 40:103–119 and associated human health risk in an alluvial plain impacted Li J, Li X, Lv N, Yang Y, Xi B, Li M, Bai S, Liu D (2015) Quantitative by agricultural and industrial activities, mid-west China. Expo assessment of groundwater pollution intensity on typical contami- Health 8:311–329 nated sites in China using grey relational analysis and numerical Wu Q, Xie SH, Pei ZJ, Ma JF (2007) A new practical methodology of simulation. Environ Earth Sci 74:3955–3968 the coal floor water bursting evaluating: the application of ANN Li Y, Zhang Z, Fei Y, Chen H, Qian Y, Dun Y (2016) Investigation vulnerable index method based on GIS [J]. J China Coal Soc of quality and pollution characteristics of groundwater in the 32:1301–1306 Hutuo River Alluvial Plain, North China Plain. Environ Earth Zeng Jichuan L (2009) Evaluation of groundwater resources in a city Sci 75:1–10 guangdong trace elements. Science 12:57–64 (in chinese) Li Z, Zhou B, Teng D, Yang W, Qiu D (2018) Comprehensive evalua- Zhang X-H (2014) A study on the water environmental quality assess- tion method of groundwater environment in a mining area based ment of Fenjiang River in Yaan city of Sichuan Province in China. on fuzzy set theory. Geosyst Eng 21:103–112 IERI Proc 9:102–109 Liu X, Li H, Li S, Jia Q, Sun Y (2010) Fuzzy synthetic evaluation Zhang K, Li H, Achari G (2009) Fuzzy-stochastic characterization of for water quality assessment in Tang River. In: Environmental site uncertainty and variability in groundwater flow and contami- science and information application technology (ESIAT), 2010 nant transport through a heterogeneous aquifer. J Contam Hydrol international conference on, 774–777. IEEE 106:73–82 Liu Y, Zheng B, Fu Q, Luo Y, Wang M (2013) Application of water Zhang B, Song X, Zhang Y, Han D, Tang C, Yu Y, Ma Y (2012) pollution index in water quality assessment of rivers. Environ Hydrochemical characteristics and water quality assessment of Monit China 29:49–55 surface water and groundwater in Songnen plain, Northeast China. Liu Y, Yang Y, Xu C (2015) Risk evaluation of water pollution in Water Res 46:2737–2748 the middle catchments of Weihe River. J Residuals Sci Technol Zhang Y, Li F, Li J, Liu Q, Tu C, Suzuki Y, Huang C (2015) Spatial 12:S133–S136 distribution, potential sources, and risk assessment of trace metals Ma L, Liu Y, Zhou X (2010) Fuzzy comprehensive evaluation method of groundwater in the North China Plain. Hum Ecol Risk Assess of F statistics weighting in identifying mine water inrush source. 21:726–743 Inter J Eng Sci Tech 2:123–128 Zhang Q, Wang S, Yousaf M, Wang S, Nan Z, Ma J, Wang D, Zang F Mahapatra S, Nanda SK, Panigrahy B (2011) A cascaded fuzzy infer- (2017) Hydrochemical characteristics and water quality assess- ence system for indian river water quality prediction. Adv Eng ment of surface water in the northeast Tibetan plateau of China. Softw 42:787–796 Water Sci Tech Water Supply, ws2017237 Miao S, Hammell RJ II, Hanratty T, Tang Z (2014) Comparison of Zhu H-N, Yuan X-Z, Liang J, Liu Y-D, Yin J, Jiang H-W, Huang H-J fuzzy membership functions for value of information determi- (2014) Integrated evaluation system under randomness and fuzzi- nation. Proceedings of the 23rd Midwest Artificial Intelligence ness for groundwater contamination risk assessment in a little and Cognitive Sciences Conference (MAICS) April 26–27, 2014, town, Central China. J Central South Univ 21:1044–1050 Spokane, WA, pp 53–60 Zou Z-H, Yi Y, Sun J-N (2006) Entropy method for determination of Mujumdar PP, Sasikumar K (2002) A fuzzy risk approach for seasonal weight of evaluating indicators in fuzzy synthetic evaluation for water quality management of a river system. Water Resour Res water quality assessment. J Environ Sci 18:1020–1023 38:5-1–5-9 Nasr AS, Rezaei M, Barmaki MD (2012) Analysis of groundwater Publisher’s Note Springer Nature remains neutral with regard to quality using mamdani fuzzy inference system (MFIS) in Yazd jurisdictional claims in published maps and institutional affiliations. Province, Iran. Inter J Comp Appl 59:45–53 1 3 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Applied Water Science Springer Journals

Groundwater quality assessment of the quaternary unconsolidated sedimentary basin near the Pi river using fuzzy evaluation technique

Free
12 pages
Loading next page...
 
/lp/springer_journal/groundwater-quality-assessment-of-the-quaternary-unconsolidated-03W7HvEWCA
Publisher
Springer Berlin Heidelberg
Copyright
Copyright © 2018 by The Author(s)
Subject
Earth Sciences; Hydrogeology; Water Industry/Water Technologies; Industrial and Production Engineering; Waste Water Technology / Water Pollution Control / Water Management / Aquatic Pollution; Nanotechnology; Private International Law, International & Foreign Law, Comparative Law
ISSN
2190-5487
eISSN
2190-5495
D.O.I.
10.1007/s13201-018-0711-0
Publisher site
See Article on Publisher Site

Abstract

The present study was carried out to assess the groundwater quality for drinking purposes in the Quaternary Unconsoli- dated Sedimentary Basin of the North Chengdu Plain, China. Six groups of water samples (S1, S2, S3, S4, S5, and S6) are selected in the study area. These samples were analyzed for 19 different physicochemical water quality parameters to assess groundwater quality. The physicochemical parameters of groundwater were compared with China’s Quality Standards for Groundwater (GB/T14848-93). Interpretation of physicochemical data revealed that groundwater in the basin was slightly alkaline. Total hardness and total dissolved solid values show that the investigated water is classified as very hard and fresh water, respectively. The sustainability of groundwater for drinking purposes was assessed based on the fuzzy mathematics evaluation (FME) method. The results of the assessment were classified into five groups based on their relative suitability for portable use (grade I = most suitable to grade V = least suitable), according to (GB/T 14848-93). The assessment results reveal that the quality of groundwater in most of the wells was class I, II and III and suitable for drinking purposes, but well (S2) has been found to be in class V, which is classified as very poor and cannot be used for drinking. Also, the FME method was compared with the comprehensive evaluation method. The FME method was found to be more comprehensive and reasonable to assess groundwater quality. This study can provide an important frame of reference for decision making on improving groundwater quality in the study area and nearby surrounding. Keywords Groundwater quality · Groundwater pollution · Fuzzy mathematics · Physicochemical parameters Introduction may threaten human health and plant growth (Zhu et al. 2014, Hosseini-Moghari et al. 2015). Most water contami- Groundwater is very important in day to day life for people nation originates from the disposal of solid wastes from dif- and society (Shigut et al. 2017). It has not only been used ferent human activities, such as agriculture, construction and for supplying potable water to both urban and rural areas industry, and the disposal of domestic and industrial waste- but also essential for irrigation, economic development, and water is discharged into rivers through the sewer systems protection of environmental and ecological balance (Cheng (Li et al. 2015, Zhang et al. 2015, Kumar et al. 2016). These and Fanhai 2012, Srinivas et al. 2015, Kumar et al. 2015, Al- systems may often leak wastewater into shallow aquifers, Ahmadi 2013). Recently, providing good quality water for directly or indirectly and groundwater supplies from nearby drinking is considered a fundamental requirement for public wells are affected if they are exposed to these wastewater health protection. Consequently, the poor quality of water pollutants (Li et al. 2016, Qishlaqi et al. 2017, Pinto 2015). When groundwater is polluted, its quality cannot be restored by stopping the contaminants from the sources. * Adam Khalifa Mohamed Shallow, unconsolidated aquifers are particularly vulner- adamkh124@yahoo.com able to contamination, which may persist in groundwater Dan Liu for many years or decades (Li et al. 2016, Liu et al. 2010). liudan-swju@163.com Meanwhile, it becomes necessary to monitor the groundwa- Faculty of Geoscience and Environmental Engineering, ter quality regularly and devise ways to protect it (Kaur et al. Southwest Jiaotong University, Chengdu, China 2014, Shigut et al. 2017). In the study area, the shallow aqui- Faculty of Water and Environmental Engineering, Sudan fer in the Quaternary Unconsolidated Sedimentary Basin, University of Science and Technology, Khartoum, Sudan Vol.:(0123456789) 1 3 65 Page 2 of 12 Applied Water Science (2018) 8:65 the largest source of drinking water in the North Chengdu the technique characteristics of the evaluation methods used. plain, China, is, however, constantly impacted by agricul- Therefore, this study will be an essential reference for future ture, industry, mining, and other human activities. This has studies. It will also be useful for the local decision makers in challenged the water resource managers and forced them to regional groundwater management and protection. pay attention to the evaluation of groundwater contamina- tion in the study area based on the recognized assessment methods. Materials and methods At present, many methods are available at home and abroad, and proven to be powerful in water quality assess- Description of study area ment, such as the principal component analysis (Gangopad- hyay et al. 2001), neural network model (Wu et al. 2007), The study area is located between longitudes 103°54′02″ Bayesian discrimination method (Chen et al. 2009), entropy to 104°16′54″ and latitudes 30°40′40″ to 30°57′58″ in the method (Chun-rong and Jun 2011), water pollution index North Chengdu Plain and bounded by many rivers, which method (Liu et al. 2013), grey clustering method (Zhang are tributaries of the Minjiang river and is divided into many 2014), statistical analysis method (Liu et al. 2015) and oth- villages with the total population of over 140,000. The pri- ers. However, these methods cannot directly reflect pol- mary source of drinking water and agricultural irrigation lution characteristics, and the linear relationship between in the study area is groundwater from the Unconsolidated those variables may have some effects on the results (Li Quaternary sediments, which provides water to residents. It et al. 2018). In the same vein, the methods above have their has a sub-tropical humid monsoon climate with four seasons. own merits, but they are less feasible and challenging to Compared to other areas in the same climatic zone, features popularize in the regional groundwater pollution assessment such as low temperature, less sunshine, and rainy weather are due to the complex and changeable environmental problems. more frequent. The mean annual temperature is 10.4 °C, the Furthermore, in all environmental quality assessments, there average temperature of coldest months and the hottest month is uncertainty about environmental risk, because of incon- is 4.6 and 24.4 °C, respectively. The region is dominated by sistency and peculiarities of each groundwater pollutant. NW wind, with a maximum wind speed of 17 m/s and aver- To overcome the shortcomings associated with the above age wind speed of 1.3 m/s. No typhoons are observed. The methods and respond to the call of water resource manag- annual maximum relative humidity is 80%, and the mini- ers, fuzzy mathematics evaluation (FME) method was used. mum relative humidity is 75%. The annual maximum of FME method has a large range of applications which could absolute moisture content is 15.2, and the minimum is 14.3. help in identifying and overcoming any uncertainty regard- Rainfall is the main recharge source of groundwater. The ing the risk of groundwater contamination using member- average annual precipitation is 1134.8 mm. Moreover, the ship functions (Mujumdar and Sasikumar 2002, Ma et al. longest continuous period of rainfall is 28 days. The Long- 2010, Zhang et al. 2012, Kamrani et al. 2016). It has also qiao water plant has been established on the right bank of the been proven effective to deal with complex and changeable Pi river at a distance of 38 m. The daily production capacity environmental problems (Singh et al. 2017), and also con- of the Longqiao water plant is about 10,000–12,000 m /d, trolling the effect of monitoring errors on assessment results and the daily water supply is 8000–10,000 m /d. However, (Ghasemi et al. 2014). the groundwater is exposed to the risk of contamination from This study is considered as the first of its kind to assess different sources. The main pollution sources of groundwater groundwater quality in this region and nearby surrounding. in the study area comes from the domestic sewage water, To show the advantage of FME, it was compared with Com- and industrial wastewater discharged routinely into the Pi 3 3 prehensive Evaluation Method (CEM), which is the most river, which is estimated at 8.4 × 10  m /d and agricultural common method of assessing the quality of groundwater for pollution from using chemical fertilizers and pesticide. It is drinking in China, and recommended by the quality standard believed that most of the wastewater is infiltrated into the for groundwater of China (GB/T 14848-93) (Wu and Sun shallow aquifer in the Quaternary Unconsolidated Sedimen- 2016, Su et al. 2017). This method provides a holistic view tary Basin in the area, because of its shallow depth of the of groundwater quality status and appropriateness for drink- groundwater level (2.0–10 m) and relatively high hydraulic ing purposes by considering various water quality param- conductivity (k = 10–50 m/day). So, it is necessary to evalu- eters based on simple mathematical-numerical tools. There- ate the quality of drinking water in this region since it is fore, the aims of this study are: (1) evaluating groundwater closely linked to people’s health. quality condition and its suitability for drinking purposes in the Quaternary Unconsolidated Sedimentary Basin, (2) iden- tifying the main pollutants which influence the groundwater quality and (3) compare the evaluation results to learn about 1 3 Applied Water Science (2018) 8:65 Page 3 of 12 65 analyzed. pH and total dissolved solids (TDS) were meas- Sample collection and analysis ured in situ using a portable pH and TDS meters because the parameters are likely to change during transport. Water In this study, groundwater samples were obtained from six monitoring wells (S1, S2, S3, S4, S5, and S6) from the shal- sampling methods were according to (Kent and Payne 1988). The samples were analyzed for 19 various physicochemical low aquifer in the Quaternary Unconsolidated Sedimentary Basin near the Pi river, China. These six monitoring wells parameters, include hydrogen ion concentration (pH), total hardness (TH), total dissolved solids (TDS), potassium (K ), are located between the Pi river and Longqiao drinking 2+ 2+ 2+ water supply plant. The locations of groundwater samples sodium (Na ), calcium (Ca ), magnesium (Mg ) sulfates 2− − − ( SO ), chlorides (Cl ), bicarbonates ( HCO ), pot assium are displayed in (Fig. 1). The sampling wells (S1, S2, and 4 3 − − S3) are inside the wall of Longqiao water plant, while the permanganate index (C OD ), nitrate ( NO ), nitrite ( NO ), Mn 3 2 ammonia ( NH ), iron (Fe), manganese (Mn), arsenic (As), sampling wells (S4, S5, and S6) outside the Longqiao water 6+ plant wall. Well (S2) is near the septic tank of the plant chromium (Cr ), and lead (Pb). These parameters were used as index indicators to evaluate the groundwater contamina- workers, and the wells (S1 and S3) are near the manage- ment offices and workers residences, respectively. Whereas, tion risk in the study area. This selection was based on their importance to the water quality and the potential impact on the well (S4) is near the Pi river bridge and the sampling wells (S5 and S6) are in the middle of farms existing in the human health and their pollution potential on groundwater resource in particular (Zhang et al. 2017). Detection methods area. These wells are used as a group in the Longqiao water plant to provide supply drinking water to residents in the for each parameter were as follows: the concentrations of K + 2+ and Na were measured using flame photometer. TH, Ca , Longqiao town and its surrounding area. It was observed 2+ − that the location of wells is exposed to the risk of ground- Mg, Cl , and HCO were analyzed by titrimetric methods. + − − 2− The concentrations of C OD , NH , NO , NO , and SO water pollution resulting from the domestic sewage water Mn 4 3 2 4 and industrial wastewater discharged into the Pi river. Also, were determined using spectrophotometer technique. And 6+ finally, the Fe, Mn, As, Cr , and Pb concentrations were unsuitable use of agricultural chemicals, unhealthy open def- ecation and many more potentially infiltrate into the shallow then measured using atomic absorption spectrometry. In this study, the groundwater suitability for drinking and aquifer. This built the foundation for selecting the location for research samples to evaluate because they have relative domestic purposes was evaluated by complying the values of various groundwater quality parameters according to China’s importance as source of drinking water in the region. The samples were collected after 10 min of pumping and stored Quality Standards for Groundwater (GB/T14848-93), since it is the only way to assess groundwater quality in China by in clean 500-mL glass bottles that were thoroughly washed with detergent and rinsed with deionized water. The samples specifying the classification of groundwater quality, ground- water quality monitoring, evaluation methods and ground- were sent to the laboratory of Environmental Engineering Center of Sichuan Geological Engineering Investigation water quality protection (Chinese 1993). It also corresponds to the methods applied to assess the quality of groundwater and kept in a refrigerator at a temperature below 4 °C until Fig. 1 Location of study area and groundwater samples 1 3 65 Page 4 of 12 Applied Water Science (2018) 8:65 − − − + 6+ in this paper by dividing the groundwater quality into five Cl, COD , NO , NO , NH , Fe, Mn, As, Cr , and Pb), Mn 3 2 4 categories: excellent, good, moderate, poor and very poor. which represent the overall situation of groundwater quality of the Quaternary Unconsolidated Sedimentary Basin. The Principle for fuzzy mathematics evaluation method element u (i =1, 2,…n) is measured value of pollutants that affect the quality of groundwater. Fuzzy mathematics was proposed by Zadeh in 1965 (Mahapatra et al. 2011, Zhang et al. 2009) as a new way to Determining the evaluation standard represent vagueness in everyday life. This method is proven to be capable of dealing with complex systems under uncer- This paper adopts the Chinese national standards of ground- tain and imprecise conditions (Gharibi et al. 2012, Singh water quality (GB/T 14848-93) as the evaluation standard, et al. 2017). Risk assessment of groundwater contamination which was drafted by the Ministry of Geology and Min- could be a challenge because it often involves many ground- eral Resources of the People’s Republic of China. Based on water quality parameters. Fuzzy mathematics can simplify groundwater quality in China and human health requirement this risk assessment process (Zhang et al. 2012). It takes the as well as the objective of water protection, the standard effectual measurement of pollutant concentration compared (GB/T 14848-93) classifies groundwater quality into five with its evaluation criteria. Through accepted linear func- grades (I, II, III, IV, and V). The quality evaluation grades tion, it calculates each pollution element of membership on of water V =(v , v ,…v ) are represented by v (i =1, 2,…m), 1 2 m i the level of groundwater contamination. Afterwards, fuzzy and it is the standard classification value of groundwater matrix can be set, and the weight of each single element for any contamination, which includes five levels: excellent value can be obtained by calculation, which constitutes the (grade I), good (grade II), moderate (grade III), poor (grade weight factors matrix. Last, the membership matrix and the IV), and very poor (grade V). Groundwater with grades I weight factors matrix are multiplied, and the evaluation and II is of excellent and good quality and is suitable for all results can be derived (Agoubi et al. 2016, Feng et al. 2012). uses. Grade III is moderate-quality water, which is generally Figure 2 shows the steps of creating the fuzzy mathematics suitable for drinking, irrigation, and industrial production. model. Grade IV is poor quality water, which is fit only for irriga- tion and industrial production and may be used for drinking Determining the evaluation factors after proper treatment. Grade V groundwater is very poor quality water that cannot be used for any purpose (Chinese Depending on the circumstances of environment, the evalu- 1993). The classification of these grades about each evalua- 2− ation factors set U = (u , u ,…,u ) = (pH, TH, TDS, SO , tion factor is given in Table 1. 1 2 n Table 1 Classification of groundwater quality based on the Chinese national standard (GB/T 14848-93) Parameters Grades I II III IV V pH ≤ 6.5 ≤ 7.0 ≤ 7.5 ≤ 8.0 > 8.5 TH (mg/L) ≤ 150 ≤ 300 ≤ 450 ≤ 550 > 550 TDS (mg/L) ≤ 300 ≤ 500 ≤ 1000 ≤ 2000 > 2000 SO (mg/L) ≤ 50 ≤ 150 ≤ 250 ≤ 350 > 350 Cl (mg/L) ≤ 50 ≤ 150 ≤ 250 ≤ 350 > 350 COD (mg/L) ≤ 1.0 ≤ 2.0 ≤ 3.0 ≤ 10 > 10 Mn NO (mg/L) ≤ 2.0 ≤ 5.0 ≤ 20 ≤ 30 > 30 NO (mg/L) ≤ 0.001 ≤ 0.01 ≤ 0.02 ≤ 0.1 > 0.1 NH (mg/L) ≤ 0.02 ≤ 0.02 ≤ 0.2 ≤ 0.5 > 0.5 Fe (mg/L) ≤ 0.1 ≤ 0.2 ≤ 0.3 ≤ 1.5 > 1.5 Mn (mg/L) ≤ 0.05 ≤ 0.05 ≤ 0.1 ≤ 1.0 > 1.0 As (mg/L) ≤ 0.005 ≤ 0.01 ≤ 0.05 ≤ 0.05 > 0.05 6+ Cr (mg/L) ≤ 0.005 ≤ 0.01 ≤ 0.05 ≤ 0.1 > 0.1 Pb (mg/L) ≤ 0.005 ≤ 0.01 ≤ 0.05 ≤ 0.1 > 0.1 Classification Excellent Good Moderate Poor Very poor Fig. 2 Flow chart of the fuzzy mathematics method 1 3 Applied Water Science (2018) 8:65 Page 5 of 12 65 Determining the weights of factors where each element of X is mapped to a value between 0 and 1. This value is referred to as membership value or degree The weights of factors are important elements in the math- of membership, and it is used to determine the degree of ematical model of FME technique, which reverses the posi- membership of each rating factor. Hence, the fuzzy set A is tion and role of each factor in the measures of comprehen- defined by its MF: sive decision making, and the result of the comprehensive (x)=  (x) , x ∈ X,  (x)∈ [0, 1] . (5) A A A evaluation is directly affected by it. The equations for weight The membership function sets are represented through is as follows (Zhang 2014): triangular, trapezoidal, Gaussian, Pseudo exponential, Sig- moidal and other shapes of fuzzy numbers (Miao et al. 2014, W = , (1) Srinivas et al. 2017). Generally, water quality parameter’s impact is represented by certain range of values, and the firing level of the conclusion is computed as the product of S =  , i ij (2) dismissal levels from the antecedent (Agoubi et al. 2016). j=1 In this study, the triangular membership function is used to normalize the crisp inputs because of its simplicity and where C is the measured values of index i, S is the standard i i computational efficiency and provide an environment more value for index i, n is the grading number of water quality conductive to human-in-the-loop knowledge acquisition standard, and α is the jth sample value under the ith level ij (Mahapatra et al. 2011, Caniani et al. 2015). It can be rep- of classification factor. To make the fuzzy compositional resented mathematically for any of the fourteen ground- operation, the weight of each single factor must be normal- water quality parameters with respect to five classification ized as follows: grades (I, II, III, IV, and V) as follows (Lermontov et al. 2009):when j = 1, W = × , C (3) i ⎧ i 1 x ≤ 𝛼 ij 𝛼 −x ij+1 i=1 𝛼 < x <𝛼 𝜇 (x) = , (6) A ⎨ ij ij+1 𝛼 −𝛼 ij+1 ij 0 x >𝛼 ⎩ ij+1 where W represents the normalized weight of the evaluated index i. Based on the above equation, the weight set of the single when j = 2, 3, 4 factor can be determined A ={ w , w , … w }. 2 i ⎧ 0 x <𝛼 ij−1 𝛼 −x ij+1 Determination of the membership and relation matrix R 𝛼 ≤ x <𝛼 ij ij+1 𝛼 −𝛼 ij+1 ij 𝜇 (x) = x−𝛼 , (7) A ij−1 𝛼 ≤ x <𝛼 ij−1 ij A fuzzy set is completely characterized by its membership 𝛼 −𝛼 ij ij−1 function (MF). The (MF) has been used to assess ground- 0 x <𝛼 ij+1 water quality according to the standards. The level of mem- bership belongs to a type of fuzzy information which over- when j = 5 comes the differences among water index grade standards in different places (Zhang 2014). The FME begins with the 0 x <𝛼 ij−1 concept of a fuzzy set. The fuzzy set describes the relation- x−𝛼 ij−1 𝛼 < x <𝛼 𝜇 (x) = , (8) ship between an uncertain quantity (x) and a membership ⎨ ij 𝛼 −𝛼 ij−1 ij ij−1 function (μ). The elements of fuzzy set membership may 1 x ≥ 𝛼 ⎩ ij be described as a number in the interval [0, 1] (Nasr et al. 2012). The greater the value of membership, the higher the ,  and  are the die ff rent levels of groundwater qual - ij ij−1 ij+1 membership qualifications. When the value of membership ity standards and x is the real measured concentration of is 1, it is subordinated completely, and when the value of each factor. membership is 0, it is subordinated incompletely. The mem- The membership function can be described of five twin bership degree of the fuzzy set is defined over a domain X grades and the fuzzy relationship matrix R is formed as which takes the form: follows: ∶ X → [0, 1], A (4) 1 3 65 Page 6 of 12 Applied Water Science (2018) 8:65 where n denotes the number of indices selected for the r r ⋯ r ⎡ 11 12 1m ⎤ assessment, F is the value of CEM for a given sample, and F ⎢ ⎥ r r ⋯ r 21 22 2m R = . represents the assigned value for the ith index by the quality (9) ⎢ ⎥ ⋮⋮ ⋮ ⎢ ⎥ standard for groundwater (GB/T 14848-93). F is the average ⎣ ⎦ r r ⋯ r n1 n2 nm value of each individual component score of F , and (F ) i i max is the maximum value of the individual component evalu- Fuzzy evaluation of groundwater quality ation score F . Once F is determined, groundwater quality classification can be obtained as per the criteria in Table  3 The fuzzy evaluation matrix B for groundwater quality (Wu and Sun 2016). evaluation is obtained by the compositional process of the weight A and the fuzzy evaluation matrix R of each single Results and discussion grade for I, II, III, IV, and V respectively. The principle of FME method can be described by the mathematical Eq. (10), Groundwater quality parameters with the main purpose of weighting evaluation factors (Hao et al. 2012). The concentration values of groundwater quality parameters obtained from the six wells in the study area were statistically r r ⋯ r ⎡ ⎤ 11 12 1m analyzed. It should be noted that for each well, four samples ⎢ ⎥ r r ⋯ r 21 22 2m ⎢ ⎥ were taken at different periods and there was no apparent B = A × R ={ w , w , … w }× 1 2 i ⎢ ⎥ ⋮⋮ ⋮ change in the analysis results of these samples and, therefore, ⎢ ⎥ ⎣ r r ⋯ r ⎦ it was taken as the average. The results represented as maxi- n1 n2 nm (10) mum, minimum, mean and standard deviation for the main groundwater quality parameters are shown in Table 4. Their From the principle of maximum membership, the grade levels were compared with the acceptable limits recommended of groundwater quality is determined. by the national quality standards for groundwater of China (GB/T 14848-93) to see if they were suitable for drinking. Comprehensive evaluation method As shown in Table 5, the pH of the groundwater is within the range of 6.7–7.5 with the mean of 7.27, indicating that The comprehensive evaluation method (CEM) is based on the groundwater in the study area is slightly alkaline for most the classification of individual component to determine of the groundwater samples. All samples fall within limits the quality category required as in Table  2 to determine (6.5 and 8.5) of quality standard for groundwater of China the individual component evaluation score of Fi, and press (GB/T 14848-93). The average concentrations of water the comprehensive evaluation to determine the value of F. + 2+ 2+ K, Ca, Mg , and HCO were 2.018, 92.34, 17.23, and The evaluation is carried out using the quality standard for 299.0 mg/L, respectively, and no limits set for these param- groundwater of China (GB/T 14848-93) as classified in the eters have been stated of quality standard for groundwater of FME technique. (Chinese 1993). As per standard, CEM can China (GB/T 14848-93). The N a concentration was deter- be computed as follows (Su et al. 2017): mined between 6.00 and 12.00  mg/L with an average of 9.417 mg/L. All samples of groundwater are found below the 2 2 F + F max (11) permissible limit of (GB/T 14848-93). The concentrations of F = (i = 1, 2, 3, … n), TH and TDS range from 260.20 to 355.30 mg/L and 331.80 to 415.40 mg/L, with an average of both 301.50 and 368.97, n respectively. This indicates that the quality of all ground- F = F (i = 1, 2, 3, … n), (12) i=1 water in the study area is classified as very hard to TH and fresh water to TDS, respectively, according to (Todd 1980, Sawyer and McCarty 1967). In this study, the concentra- Table 2 Individual component classification tions of (TH and TDS) are found below the allowable limit Groundwater quality categories I II III IV V of (GB/T 14848-93) for all groundwater samples. At the 2− − − same time, the concentrations of SO , Cl, COD , NO , Mn 4 3 Value for Fi 0 1 3 6 10 − + NO , and NH in groundwater range from 52.83 to 68.20, 2 4 Table 3 Standards of Level Excellent Good Moderate Poor Very poor the groundwater quality classification F < 0.80 0.80~< 2.50 2.50~< 4.25 4.25~< 7.20 > 7.20 1 3 Applied Water Science (2018) 8:65 Page 7 of 12 65 Table 4 Descriptive statistics of Parameters Unit Min Max Mean St.dev Standard (Class III) groundwater quality variables for six sampling wells in the pH – 6.70 7.50 7.27 0.29 6.5–8.5 study area TH mg/L 260.20 355.30 301.50 42.11 ≤ 450 TDS mg/L 331.80 415.40 368.97 36.67 ≤ 1000 K mg/L 1.50 2.30 2.018 0.34 – Na mg/L 6.00 12.00 9.417 2.25 ≤ 200 2+ Ca mg/L 77.15 111.20 92.34 15.22 – 2+ Mg mg/L 15.81 18.85 17.23 1.13 – 2− SO mg/L 52.83 68.20 60.22 6.644 ≤ 250 Cl mg/L 15.96 20.92 17.61 2.304 ≤ 250 HCO mg/L 250.20 347.80 299.00 40.06 – COD mg/L 0.23 0.29 0.25 0.02 3 Mn NO mg/L 1.17 3.96 2.40 0.92 ≤ 20 NO mg/L 0.003 0.005 0.004 0.0006 ≤ 0.07 NH mg/L 0.020 0.68 0.14 0.26 ≤ 0.2 Fe mg/L 0.086 2.23 0.59 0.82 ≤ 0.3 Mn mg/L 0.01 0.21 0.07 0.08 ≤ 0.1 As mg/L 0.0002 0.0002 0.0002 0.00 ≤ 0.05 6+ Cr mg/L 0.002 0.004 0.003 0.001 ≤ 0.01 Pb mg/L 0.005 0.01 0.008 0.002 ≤ 0.05 Note: Standard refers to the quality standard for groundwater developed by China (GB/T14848-93) (Chi- nese 1993) All units of parameters are in mg/L except pH, St.dev standard deviation Table 5 The weight value of groundwater quality parameters in well No. S2 Weight Factors 6+ PH TH TDS SO Cl COD NO NO NH Fe Mn As Cr Pb 4 Mn 3 2 4 W 0.974 0.732 0.297 0.068 0.071 0.039 0.064 0.089 2.363 2.720 0.389 0.005 0.043 0.079 0.123 0.092 0.037 0.009 0.009 0.005 0.008 0.011 0.298 0.343 0.049 0.001 0.005 0.010 15.96 to 20.92, 0.23 to 0.29, 1.17 to 3.96, 0.003 to 0.005 and exceed the permissible limit represented in sites (S2, S4) 0.020 to 0.68 mg/L, respectively, with a cumulative average and for Mn concentration, one sample exceeds the permis- for each concentration of 60.22, 17.61, 0.25, 2.40, 0.004 and sible limit at the site (S2). According to the mean values, the 2+ 2+ + + 0.14 mg/L, respectively. The values of these concentrations dominance of cations is Ca > Mg > Na > K and anions − 2− − for all the groundwater samples collected in this study are is HCO > SO > Cl , which makes the predominant type of 3 4 within the permissible limit of (GB/T 14848-93) for ground- groundwater to be Ca–Mg-HCO type. As mentioned above, water quality except sample S2 of NH which exceeds the the groundwater quality indicators exceeding the standard ratio acceptable limit of 0.2 mg/L according to the national quality in the study area were mainly Fe, Mn, and NH for sampling standards for groundwater (GB/T 14848-93). Finally, the con- points S2 and S4 (Fig. 3). 6+ centrations of Fe, Mn, As, Cr , and Pb in groundwater were 0.086–0.23 mg/L with the mean 0.59 mg/L, 0.01–0.21 mg/L Fuzzy mathematics evaluation results of study area with a mean of 0.07 mg/L, 0.0002–0.0002 mg/L with the mean 0.0002 mg/L, 0.002–0.004 mg/L with a mean 0.003 mg/L and According to the above-mentioned relevant principles of the 0.005–0.01 mg/L with an average of 0.008 mg/L, respectively. FME technique for the assessment of groundwater quality of Among them, Fe and Mn are found to exceed the permissi- the study area, the steps for evaluation results can be listed ble limits of drinking water as set by (GB/T 14848-93) of 0.3 as follows: and 0.1 mg/L, respectively, and remaining concentrations fall within the permissible limits of (GB/T 14848-93) for all groundwater samples. For Fe concentration, two samples 1 3 65 Page 8 of 12 Applied Water Science (2018) 8:65 2.5 Weight of each evaluation factors 2.0 According to Eqs. (1) (2) and (3), the corresponding weight NH Fe values of all wells in the study area were obtained, using the Mn (GB/T14848-93) standard (Table 1) and data from Table 4 for 1.5 the fourteen selected indicators. The weight values for well No. S2 are shown in Table 5, as an example. 1.0 The membership and relation matrix of different evaluation 0.5 factors on various grades 0.0 S1 S2 S3 S4 S5 S6 Based on Eqs. (6) (7) and (8), the degree of membership of Well name each indicator to the groundwater quality grade is calculated. Each indicator is computed to have five levels of membership, Fig. 3 Groundwater quality indicators exceeding the standard ratio in and the fourteen selected indicators can get fourteen sets of the study area numerical values. Accordingly, the corresponding fuzzy rela- tionship matrix R is achieved from the selected indicators of all Evaluation factors wells in the study area which was computed. The membership for well No. S2 is shown in Table 6, as an example. In this study, a FME technique was used to assess the This membership can be expressed by fuzzy relationship groundwater quality of the quaternary unconsolidated matrix, as follows: sedimentary basin near the Pi river according to quality 0 0.20 0.80 0 0 ⎡ ⎤ evaluation parameters and five classifications of groundwa- ⎢ ⎥ 0 0.998 0.002 0 0 ter quality based on the Chinese national standard (GB/T ⎢ ⎥ 0.63 0.37 0 0 0 ⎢ ⎥ 14848-93) (Table 1). The groundwater quality was assessed ⎢ 10 00 0 ⎥ for six wells (S1, S2, S3, S4, S5, and S6) as selected in ⎢ ⎥ 10 00 0 the study area. Of the 19 groundwater parameters analyzed, ⎢ ⎥ 10 00 0 ⎢ ⎥ 14 parameters (pH, TH, TDS, SO , Cl, COD, NO, NO , 4 Mn 3 2 ⎢ ⎥ 10 00 0 6+ NH , Mn, Fe, As, Cr and Pb) were selected due to the R = . ⎢ ⎥ 0.56 0.44 0 0 0 ⎢ ⎥ fact the rest did not have the five-level division as required 0 0 0 0.1 0.9 ⎢ ⎥ in the China’s standard for groundwater quality evaluation ⎢ ⎥ 00 00 1 (GB/T 14848-93). This selection is also based on the fact ⎢ ⎥ 0 0 0.88 0.12 0 ⎢ ⎥ that it is periodically monitored by the Local Environmental 10 00 0 ⎢ ⎥ Protection Department for their vital importance to the water ⎢ ⎥ 10 00 0 quality and potential influence on human health. ⎢ ⎥ 10 00 0 ⎣ ⎦ Table 6 Membership of well Well Parameters I II III IV V No. S2 S2 pH 0 0.20 0.80 0 0 TH 0 0.998 0.002 0 0 TDS 0.63 0.37 0 0 0 SO 1 0 0 0 0 Cl 1 0 0 0 0 COD 1 0 0 0 0 Mn NO 1 0 0 0 0 NO 0.56 0.44 0 0 0 NH 0 0 0 0.1 0.9 Fe 0 0 0 0 1 Mn 0 0 0.88 0.12 0 As 1 0 0 0 0 6+ Cr 1 0 0 0 0 Pb 1 0 0 0 0 1 3 Concentration mg/L Applied Water Science (2018) 8:65 Page 9 of 12 65 Fuzzy evaluation of groundwater quality 0.6 Based on Eq. (10), the fuzzy evaluation matrix B for ground- 0.5 water quality evaluation was obtained by the compositional process of the weight matrix A and the fuzzy evaluation 0.4 matrix R. For example, the fuzzy evaluation matrix B of the 0.3 well No. S2 is as follows: B = (0.076, 0.136, 0.141, 0.036, 0.611). 0.2 Excellent According to the principle of maximum membership to Good Moderate 0.1 determine the groundwater quality level, 0.611 is the maxi- Poor Very poor Final result mum of all five values. Therefore, well No. S2 was found to 0.0 belong to grade V, which is classified as very poor and their S1 S2 S3 S4 S5 S6 Well name water quality cannot be used for any purpose. The degrada- tion in groundwater quality at this well is mainly due to high Fig. 4 Comprehensive assessment of fuzzy technique for assessing concentrations of NH and Fe (see Table 6). Similarly, the groundwater quality groundwater quality grades of the other wells in the study area were obtained. The results are shown in Table 7. From Table 7 and Fig. 4, the results of fuzzy evaluation According to our field investigation in the study, the very matrix B for the other wells showed that the groundwater poor groundwater quality of well No. S2 could be attributed quality of well No. S1 is (0.579, 0.421, 0, 0, 0), and it found to several reasons: to belong to grade I, which is classified as excellent and their water quality is considered to be suitable for vari- 1. In the Basin of Minjiang as Pi river is a tributary of ous purposes. The groundwater quality of wells No. S3 is Minjiang river, sediment has a certain amount of Fe and (0.379, 0.477, 0.144, 0, 0), and categorized as being grade Mn under the natural environmental conditions (Zeng II, which is classified as good, and was also deemed to be Jichuan 2009, Li et al. 2006). Thus, the over-limit ratio suitable for all uses. Finally, the groundwater quality of wells of Fe and Mn is considered natural environment. No. S4, S5 and S6 are (0.264, 0.239, 0.466, 0.032), (0.214, 2. The sample No. S2 was taken from the water plant, 0.318, 0.468, 0, 0), (0.207, 0.319, 0.473, 0, 0), respectively, which is closest to the populated area, and the infiltra- which fall under grade III and is classified as moderate, and tion of the plant domestic wastewater may lead to the their water quality is generally suitable for drinking, irriga- increase of NH content. In essence, the groundwater tion, and industrial production. From the six groundwater quality of the well S2 is mainly influenced by human samples, grade I of groundwater occupies 16.67%, grade activities. II 16.67% and grade III 50% that means about 83.33% of 3. Excessive discharge of industrial waste water and the groundwater samples could be used as drinking water domestic sewage is the main cause of groundwater pol- source. However, S2 is grade V (16.67%), which cannot be lution in the study area. Considering the aforementioned used for drinking. This suggests that the groundwater quality reasons, well No.2 is deemed contaminated and, there- in the study area, in general, is not bad. Through analysis of fore, should be monitored periodically and protected the water indicators, the elevated groundwater quality indi- from the causes of pollution to avoid being consumed cators were mainly NH , Fe, and Mn which have resulted in by local residents. the evaluation of well No. S2 as class V which is very poor. Table 7 The fuzzy evaluation of Well name I II III IV V Results groundwater quality S1 0.579 0.421 0 0 0 I S2 0.076 0.136 0.141 0.036 0.611 V S3 0.379 0.477 0.144 0 0 II S4 0.264 0.239 0.466 0.032 0 III S5 0.214 0.318 0.468 0 0 III S6 0.207 0.319 0.473 0 0 III 1 3 Grade value 65 Page 10 of 12 Applied Water Science (2018) 8:65 Groundwater quality evaluation based on comprehensive evaluation method In the study area, groundwater is a vital source of drinking water for residents. The CEM has been used to assess the qual- ity of groundwater of the Quaternary Unconsolidated Sedi- mentary Basin near the Pi river and to demonstrate the advan- tage of this study. In this method, the same physicochemical water quality parameters that were chosen in the FME method (14 parameters) have been used for six wells (S1, S2, S3, S4, S5, and S6). The results are shown in Table 8. As shown in Table 8 and Fig. 5, F values of all analyzed samples in the S1 S2 S3 S4 S5 S6 Well name study area differ from 0.82 to 7.29, ranging from good quality to very poor quality. The results of the assessment showed Fig. 5 Results of groundwater quality evaluation based on compre- that four groundwater samples (66.67% of all samples) are hensive evaluation method classified as good quality water (grade II) which is suitable for various purposes. Two groundwater samples (33.33% of all boundary water quality index of water quality is, hence making collected samples) are classified as poor and very poor quality the evaluation more comprehensive and reasonable. Whereas, water (grade IV and V), respectively, which are classified as the CEM of water highlights the largest factor of pollution as unsuitable for drinking. The common contaminants in these the index classification is based on the binary logic. Therefore, samples are NH , Fe, and Mn, which are mainly from the natu- they cannot describe the continuity of environmental quality, ral environment, industrial and agricultural activities. Accord- and cannot objectively reflect the influence of the index value ing to the results of the CEM, four groundwater samples (S1, near the water quality grade limit for quality evaluation and S3, S5, and S6) could be used as drinking water source, while classification. (S2 and S4) cannot be used as drinking water source. Despite the FME and CEM generated almost similar results to the holistic picture of groundwater quality in the study area, Comparison between fuzzy mathematics evaluation the fuzzy indicator is recommended as the more useful indica- technique and comprehensive evaluation method tor for the following advantages: 1. Ability to describe a wide variety of non-linear relation- The present study analyzed groundwater quality status of six ships. 2. They tend to be simple since they are based on a set of wells in the Quaternary Unconsolidated Sedimentary Basin local simple models. 3. Fuzzy mathematics can deal with and using FME technique and CEM. Based on the results of the process missing data without compromising the final result. assessment obtained from both methods, the holistic picture of 4. Avoiding artificial precision as well as generating results groundwater quality within the study area was satisfactory and which are more consistent with the ecological complexity of consistent with the actual situation of the study area under the real-world issues. 5. Combining both qualitative and quanti- prevailing conditions. The indices for both methods indicated tative information to express the ecological status of the case that the groundwater quality in the study area was suitable for study, which is a unique capability of fuzzy approach. drinking at ratio 83.34 and 66.67%, respectively. The FME In general, the results revealed that the knowledge-based showed only the well No. 2 to be unsuitable for drinking water models such as FME method were practical and flexible tools (see Table 7), while the CEM showed wells No. 2 and 4 to for incorporating the experts’ attitudes and modeling the cur- be unsuitable for drinking water (see Table  8). This is due rent uncertainties associated with water resources and envi- to the technical difference of both methods. For instance, the ronmental perplexities. However, in the FME technique, the FME approach uses membership degree to describe the limit weight of evaluating indicators is determined by the moni- between different pollution degrees for assessing groundwater toring data compared to groundwater quality standard. As a quality. The FME technique takes into account the impression result, when an abnormal value appears at some evaluating of each assessment factor on the evaluation result and deter- indicator, the condition of overestimating the weight of these mines the major pollutants according to the weights of evalua- indicators would lead to unrealistic evaluation results which tion factors. This reflects how close the actual concentration of Table 8 Groundwater quality Well name S1 S2 S3 S4 S5 S6 classification based on comprehensive evaluation F 0.82 7.29 2.21 4.3 2.35 2.27 method Grade II V II IV II II 1 3 F value Applied Water Science (2018) 8:65 Page 11 of 12 65 may not be in line with the actual situation of the studied area conditions accurately and precisely, specifically for the (Zou et al. 2006). North Chengdu Plain, China. The degree of groundwater pollution risk has a direct con- nection to the water discharge and environmental vulnerabil- Conclusions ity of the region. To improve the status of groundwater and thoughtful scientific planning of groundwater extraction, it Groundwater pollution is a vague concept because there are is necessary to strictly control the industrial wastewater and often no clear-cut boundaries that separate a “polluted” from domestic sewage discharge not only in the North Chengdu an “unpolluted” sample. It is, therefore, necessary to develop Plain but also in other watersheds. a new method based on a fuzzy technique to give solutions Acknowledgements This research was supported by the Fundamental that are robust and have a high level of confidence. Research Funds for the Central Universities (2682015CX020). In this study, six groundwater samples were collected, analyzed and assessed for drinking water quality in the Open Access This article is distributed under the terms of the Crea- Quaternary Unconsolidated Sedimentary Basin near the Pi tive Commons Attribution 4.0 International License (http://creat iveco river. The pH value of the groundwater was slightly alka-mmons.or g/licenses/b y/4.0/), which permits unrestricted use, distribu- tion, and reproduction in any medium, provided you give appropriate line for most of the groundwater samples to basic in nature. credit to the original author(s) and the source, provide a link to the The groundwater is classified as very hard and fresh water Creative Commons license, and indicate if changes were made. based on TH and TDS, respectively. Groundwater quality parameters were compared with the acceptable limits recom- mended by the national quality standards for groundwater of China (GB/T 14848-93). From all groundwater parameters References analyzed, NH , Fe, and Mn were above the permissible lim- its of (GB/T 14848-93). The sequence of the abundance of Agoubi B, Souid F, Kharroubi A, Abdallaoui A (2016) Assessment of 2+ 2+ + + hot groundwater in an arid area in Tunisia using geochemical and major ions is found in the order of Ca > Mg > Na > K − 2− − fuzzy logic approaches. Environ Earth Sci 75:1497 and anions is HCO > SO > Cl . 3 4 Al-Ahmadi ME (2013) Hydrochemical characterization of groundwater The FME technique results show that five of the sam- in wadi Sayyah, Western Saudi Arabia. Appl Water Sci 3:721–732 pled groundwater (S1, S3, S4, S5, and S6) are suitable of Caniani D, Lioi D, Mancini I, Masi S (2015) Hierarchical classifica- tion of groundwater pollution risk of contaminated sites using drinking water directly while the well (S2) is unsuitable for fuzzy logic: a case study in the Basilicata region (Italy). Water drinking unless treated. The common contaminants in these 7:2013–2036 samples are NH , Fe, and Mn, which are mainly from the Chen H, Li X, Liu A (2009) Studies of water source determination natural environment, industrial and agricultural activities. method of mine water inrush based on Bayes’ multi-group step- wise discriminant analysis theory. Rock Soil Mech 30:3655–3659 Thus, authorities should give more attention to the pollution Cheng Y, Fanhai M (2012) GW vulnerability assessment based on of NH , Fe, and Mn to prevent deterioration of good water entropy and fuzzy method. Proc Inst Civ Eng 165:277 quality by taking some effective measures that are required Chinese S. (1993) Quality Standard for Groundwater (GB/T 14848– to enhance the drinking water quality by delineating an 93). AQSIQ (General Administration of Quality Supervision, Inspection and Quarantine of the People’s Republic of China) effective water quality management plan. This method may (in Chinese) play an important role in decision-making for the drinking Chun-rong J, Jun Z (2011) Based on fuzzy weight matter element to water quality assessment because it proved effective in solv - evaluate the water quality of Jialing River in Nanchong, China. ing problems of fuzzy boundaries. So, it is more objective Proc Environ Sci 11:631–636 Feng L, Daohan W, Sijing F, Suzhen Y (2012) Study on the application and scientific and reliable in practice. of fuzzy mathematics in assessing the water quality of Qinghe The FME method was compared with CEM in this study. Reservoir. In: Computer science and electronics engineering Based on the results of the assessment obtained from both (ICCSEE), 2012 international conference on, pp 672–675. IEEE methods, the holistic picture of groundwater quality within Gangopadhyay S, Das Gupta A, Nachabe M (2001) Evaluation of ground water monitoring network by principal component analy- the study area was satisfactory and consistent with the actual sis. Groundwater 39:181–191 situation of the study area under the prevailing conditions. Gharibi H, Mahvi AH, Nabizadeh R, Arabalibeik H, Yunesian M, However, the fuzzy indicator is recommended as a more Sowlat MH (2012) A novel approach in water quality assessment practical indicator for assessing groundwater quality owing based on fuzzy logic. J Environ Manage 112:87–95 Ghasemi E, Amini H, Ataei M, Khalokakaei R (2014) Application of to the introduction of membership degree and weight of each artificial intelligence techniques for predicting the flyrock distance factor to the models. caused by blasting operation. Arab J Geosci 7:193–202 The study can provide an important frame of reference to Hao W, Hanting Z, Wenjuan X, Jinling Z (2012) Evaluation of ground- the government decision-making on improving groundwater water quality using improved fuzzy comprehensive assessment based on AHP. Inter J Appl Sci Eng Res 2:377–384 quality in the study area. Additionally, this study may serve as a guide for future researchers to assess the groundwater 1 3 65 Page 12 of 12 Applied Water Science (2018) 8:65 Hosseini-Moghari S-M, Ebrahimi K, Azarnivand A (2015) Groundwa- Pinto SR (2015) Fuzzy logic based assessment of periodic variation ter quality assessment with respect to fuzzy water quality index of water quality of Nethravathi River in Dakshina Kannada dis- (FWQI): an application of expert systems in environmental moni- trict. Int J Innovative Res Electrical Electron Instrum Control Eng toring. Environ Earth Sci 74:7229–7238 3:321–326 Kamrani S, Rezaei M, Amiri V, Saberinasr A (2016) Investigating the Qishlaqi A, Kordian S, Parsaie A (2017) Hydrochemical evaluation of efficiency of information entropy and fuzzy theories to classifi- river water quality—a case study. Appl Water Sci 7:2337–2342 cation of groundwater samples for drinking purposes: Lenjanat Sawyer CN, McCarty PL (1967) Chemistry for sanitary engineers. Plain, Central Iran. Environ Earth Sci 75:1370 Chemistry for sanitary engineers. McGraw-Hill, New York Kaur BJ, George M, Mishra S (2014) Groundwater quality and water Shigut DA, Liknew G, Irge DD, Ahmad T (2017) Assessment of quality index of Delhi city, India. World Appl Sci J 32:865–871 physico-chemical quality of borehole and spring water sources Kent R, Payne K (1988) Sampling groundwater monitoring wells: spe- supplied to Robe Town, Oromia region, Ethiopia. Appl Water cial quality assurance and quality control considerations, In: Keith Sci 7:155–164 LH (ed) Principles of environmental sampling. ACS Professional Singh AP, Chakrabarti S, Kumar S, Singh A (2017) Assessment of air Reference Book, ACS, Washington, DC, pp 231–246 quality in Haora River basin using fuzzy multiple-attribute deci- Kumar P, Bansod BK, Debnath SK, Thakur PK, Ghanshyam C (2015) sion making techniques. Environ Monit Assess 189:373 Index-based groundwater vulnerability mapping models using Srinivas R, Bhakar P, Singh AP (2015) Groundwater quality assess- hydrogeological settings: a critical evaluation. Environ Impact ment in some selected area of Rajasthan, India using fuzzy multi- Assess Rev 51:38–49 criteria decision making tool. Aquat Proc 4:1023–1030 Kumar P, Thakur PK, Bansod BK, Debnath SK (2016) Assessment of Srinivas R, Singh AP, Sharma R (2017) A scenario based impact the effectiveness of DRASTIC in predicting the vulnerability of assessment of trace metals on ecosystem of river Ganges using groundwater to contamination: a case study from Fatehgarh Sahib multivariate analysis coupled with fuzzy decision-making district in Punjab, India. Environ Earth Sci 75:879 approach. Water Resour Manage 31:4165–4185 Lermontov A, Yokoyama L, Lermontov M, Machado MAS (2009) Su H, Kang W, Xu Y, Wang J (2017) Evaluation of groundwater quality River quality analysis using fuzzy water quality index: Ribeira do and health risks from contamination in the north edge of the Loess Iguape river watershed, Brazil. Ecol Ind 9:1188–1197 Plateau, Yulin City, Northwest China. Environ Earth Sci 76:467 Li X-D, Masuda H, Ono M, Kusakabe M, Yanagisawa F, Zeng H-A Todd D (1980) Groundwater hydrology. Wiley, New York, p 535 (2006) Contribution of atmospheric pollutants into groundwater Wu J, Sun Z (2016) Evaluation of shallow groundwater contamination in the northern Sichuan Basin, China. Geochem J 40:103–119 and associated human health risk in an alluvial plain impacted Li J, Li X, Lv N, Yang Y, Xi B, Li M, Bai S, Liu D (2015) Quantitative by agricultural and industrial activities, mid-west China. Expo assessment of groundwater pollution intensity on typical contami- Health 8:311–329 nated sites in China using grey relational analysis and numerical Wu Q, Xie SH, Pei ZJ, Ma JF (2007) A new practical methodology of simulation. Environ Earth Sci 74:3955–3968 the coal floor water bursting evaluating: the application of ANN Li Y, Zhang Z, Fei Y, Chen H, Qian Y, Dun Y (2016) Investigation vulnerable index method based on GIS [J]. J China Coal Soc of quality and pollution characteristics of groundwater in the 32:1301–1306 Hutuo River Alluvial Plain, North China Plain. Environ Earth Zeng Jichuan L (2009) Evaluation of groundwater resources in a city Sci 75:1–10 guangdong trace elements. Science 12:57–64 (in chinese) Li Z, Zhou B, Teng D, Yang W, Qiu D (2018) Comprehensive evalua- Zhang X-H (2014) A study on the water environmental quality assess- tion method of groundwater environment in a mining area based ment of Fenjiang River in Yaan city of Sichuan Province in China. on fuzzy set theory. Geosyst Eng 21:103–112 IERI Proc 9:102–109 Liu X, Li H, Li S, Jia Q, Sun Y (2010) Fuzzy synthetic evaluation Zhang K, Li H, Achari G (2009) Fuzzy-stochastic characterization of for water quality assessment in Tang River. In: Environmental site uncertainty and variability in groundwater flow and contami- science and information application technology (ESIAT), 2010 nant transport through a heterogeneous aquifer. J Contam Hydrol international conference on, 774–777. IEEE 106:73–82 Liu Y, Zheng B, Fu Q, Luo Y, Wang M (2013) Application of water Zhang B, Song X, Zhang Y, Han D, Tang C, Yu Y, Ma Y (2012) pollution index in water quality assessment of rivers. Environ Hydrochemical characteristics and water quality assessment of Monit China 29:49–55 surface water and groundwater in Songnen plain, Northeast China. Liu Y, Yang Y, Xu C (2015) Risk evaluation of water pollution in Water Res 46:2737–2748 the middle catchments of Weihe River. J Residuals Sci Technol Zhang Y, Li F, Li J, Liu Q, Tu C, Suzuki Y, Huang C (2015) Spatial 12:S133–S136 distribution, potential sources, and risk assessment of trace metals Ma L, Liu Y, Zhou X (2010) Fuzzy comprehensive evaluation method of groundwater in the North China Plain. Hum Ecol Risk Assess of F statistics weighting in identifying mine water inrush source. 21:726–743 Inter J Eng Sci Tech 2:123–128 Zhang Q, Wang S, Yousaf M, Wang S, Nan Z, Ma J, Wang D, Zang F Mahapatra S, Nanda SK, Panigrahy B (2011) A cascaded fuzzy infer- (2017) Hydrochemical characteristics and water quality assess- ence system for indian river water quality prediction. Adv Eng ment of surface water in the northeast Tibetan plateau of China. Softw 42:787–796 Water Sci Tech Water Supply, ws2017237 Miao S, Hammell RJ II, Hanratty T, Tang Z (2014) Comparison of Zhu H-N, Yuan X-Z, Liang J, Liu Y-D, Yin J, Jiang H-W, Huang H-J fuzzy membership functions for value of information determi- (2014) Integrated evaluation system under randomness and fuzzi- nation. Proceedings of the 23rd Midwest Artificial Intelligence ness for groundwater contamination risk assessment in a little and Cognitive Sciences Conference (MAICS) April 26–27, 2014, town, Central China. J Central South Univ 21:1044–1050 Spokane, WA, pp 53–60 Zou Z-H, Yi Y, Sun J-N (2006) Entropy method for determination of Mujumdar PP, Sasikumar K (2002) A fuzzy risk approach for seasonal weight of evaluating indicators in fuzzy synthetic evaluation for water quality management of a river system. Water Resour Res water quality assessment. J Environ Sci 18:1020–1023 38:5-1–5-9 Nasr AS, Rezaei M, Barmaki MD (2012) Analysis of groundwater Publisher’s Note Springer Nature remains neutral with regard to quality using mamdani fuzzy inference system (MFIS) in Yazd jurisdictional claims in published maps and institutional affiliations. Province, Iran. Inter J Comp Appl 59:45–53 1 3

Journal

Applied Water ScienceSpringer Journals

Published: Apr 20, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off