1 23 Environmental Science and Pollution Research ISSN 0944-1344 Volume 25 Number 22 Environ Sci Pollut Res (2018) 25:21915-21926 DOI 10.1007/s11356-018-2295-5 Anthropogenic activities impact on atmospheric environmental quality in a gas-flaring community: application of fuzzy logic modelling concept Olayiwola Akin Akintola, Abimbola Yisau Sangodoyin & Foluso Oyedotun Agunbiade 1 23 Your article is protected by copyright and all rights are held exclusively by Springer- Verlag GmbH Germany, part of Springer Nature. This e-offprint is for personal use only and shall not be self-archived in electronic repositories. If you wish to self-archive your article, please use the accepted manuscript version for posting on your own website. You may further deposit the accepted manuscript version in any repository, provided it is only made publicly available 12 months after official publication or later and provided acknowledgement is given to the original source of publication and a link is inserted to the published article on Springer's website. The link must be accompanied by the following text: "The final publication is available at link.springer.com”. RESEARCH ARTICLE Anthropogenic activities impact on atmospheric environmental quality in a gas-flaring community: application of fuzzy logic modelling concept Olayiwola Akin Akintola1 & Abimbola Yisau Sangodoyin2 & Foluso Oyedotun Agunbiade3 Received: 22 November 2017 /Accepted: 9 May 2018 /Published online: 24 May 2018 # Springer-Verlag GmbH Germany, part of Springer Nature 2018 Abstract We present a modelling concept for evaluating the impacts of anthropogenic activities suspected to be from gas flaring on the quality of the atmosphere using domestic roof-harvested rainwater (DRHRW) as indicator. We analysed seven metals (Cu, Cd, Pb, Zn, Fe, Ca, and Mg) and six water quality parameters (acidity, PO4 3−, SO4 2−, NO3 − , Cl −, and pH). These were used as input parameters in 12 sampling points from gas-flaring environments (Port Harcourt, Nigeria) using Ibadan as reference. We formu- lated the results of these input parameters into membership function fuzzy matrices based on four degrees of impact: extremely high, high, medium, and low, using regulatory limits as criteria. We generated indices that classified the degree of anthropogenic activity impact on the sites from the product membership function matrices and weight matrices, with investigated (gas-flaring) environment as between medium and high impact compared to those from reference (residential) environment that was classified as between low and medium impact. Major contaminants of concern found in the harvested rainwater were Pb and Cd. There is also the urgent need to stop gas-flaring activities in Port Harcourt area in particular and Niger Delta region of Nigeria in general, so as to minimise the untold health hazard that people living in the area are currently faced with. The fuzzy methodology presented has also indicated that the water cannot safely support potable uses and should not be consumed without purification due to the impact of anthropogenic activities in the area but may be useful for other domestic purposes. Keywords Atmospheric pollution . Domestic roof-harvested rainwater . Fuzzy logic . Gas flaring, anthropogenic activities Introduction The flaring of associated gases with crude oil is a common practice in crude oil exploration in Niger Delta region of Nigeria. Though gas flaring has technically been outlawed in Nigeria since 1984, it is still being practice today because there are sometimes exemptions granted to the crude oil ex- ploratory companies. Also, enforcement of the law is too weak and attached penalty is either not clearly stated or too light to serve as deterrent. Hence, despite the long-standing laws against gas flaring in Nigeria, the burning of natural gas dur- ing crude oil extraction and shifting of deadlines to end the practice continue, with serious economic and health conse- quences for people living nearby and the nation in general. It has been estimated that gas flaring costs Nigeria up to US$2.5 billion annually beside the roaring toxic flare effects on the quality of the atmosphere, the health, and livelihoods of in- habitants of the area (Roderick 2005). On a global scale, gas flaring contributes significantly to climate change, thus affect- ing communities all over the world. Port Harcourt, Nigeria, is reputed for oil exploration with oil wells scattered around the city. The city also hosts a crude oil refinery. The presence and activities of many multinational companies that are operators of the Nigerian National Petroleum Corporation (NNPC) joint venture and their oil fields all around Port Harcourt axis, combined with the activ- ities around the refinery, are predisposing factors for pollution from gas flaring and other oil exploration activities. Responsible editor: Philippe Garrigues * Olayiwola Akin Akintola akinbolanle97@yahoo.co.uk 1 National Horticultural Research Institute, Jericho Reservation Area, Idi-Ishin, P.M.B.5432, Dugbe Post Office, Ibadan, Nigeria 2 Department of Agricultural and Environmental Engineering, Faculty of Technology, University of Ibadan, Ibadan, Nigeria 3 Department of Chemistry, Faculty of Sciences, University of Lagos, Akoka, Lagos, Lagos State, Nigeria Environmental Science and Pollution Research (2018) 25:21915–21926 https://doi.org/10.1007/s11356-018-2295-5 Author's personal copy http://crossmark.crossref.org/dialog/?doi=10.1007/s11356-018-2295-5&domain=pdf mailto:akinbolanle97@yahoo.co.uk The gas-flaring activity can be further associated with the complications of inadequate good quality water supply especially for people living in developing countries which is one of the goals of United Nation Sustainable Development Goals. Flaring of associated natural gas has negative effects on the viability of alternative water supply in form of domestic roof-harvested rainwater (DRHRW) (Thomas 2000). Several households in less developed climes harvest rainwater and channel it for po- table uses with the assumption of its purification by the water cycle. This assumption may not be safe because flaring of natural gas results in serious increase in con- centrations of SOx, NOx, and CO2 gases discharged into the atmospheric environment with very serious negative effects on the quality of harvested rainwater. Moreover, metals are associated with crude oil in wet natural gas wells and the reaction vessels used for the combustion. This can be discharged with the particulate matters (soot) generated by gas flaring and contaminates DRHRW. The interfering effects have not been adequately covered by studies from developing nations due to limited analytical facilities. The use of DRHRW as good index for accessing extent of air pollution has been documented (Akintola et al. 2013, 2016). Thus, the application of DRHRW to evaluate impact of gas-flaring activities is being proposed here. This evaluation of impact of gas-flaring activities re- quires the application of novel modelling concept such as fuzzy logic and fuzzy set theory modelling concept and not just the conventional method of reporting concentra- tion of individual parameters. The merits of this ap- proach are already extensively documented in literature (Adebowale et al. 2008; Agunbiade et al. 2011; Akintola et al. 2016). This includes the consideration of synergis- tic effects of the multiple exposure risk parameters among others and will help in facilitating appropriate mitigation policy formulation particularly against gas flaring and related climatic change effects. Fuzzy logic was introduced in 1965 from artificial intelligence (Zadeh 1965) and is being applied to model chaotic, biological, other animate systems, and various environ- mental compartments (Sadegh-Zadeh 2001; Onkal-Engin et al. 2004; Adebowale et al. 2008; Agunbiade et al. 2011; Akintola et al. 2016). The application of fuzzy logic in evaluating the quality of DRHRW is relatively few (Akintola et al. 2016). A search of literature (Scopus and Web of Knowledge) did not give any result of its application in evaluating impact of gas flaring. This study was therefore aimed at evaluating the degree of pollution impact by gas-flaring emissions relative to less polluted residential community using DRHRW quality as model input and Fuzzy Comprehensive Assessment method. Materials and methods Sampling and sampling site description This study was conducted in gas-flaring city of Port Harcourt (4.75° N, 7.0° E), Nigeria, with comparative reference sites in Ibadan (7.43° N, 3.92° E), a largely residential environment (Fig. 1). Domestic roof-harvested rainwater samples were col- lected at 12 locations per city, in three replicates for three consecutive years in some pre-selected locations in Port Harcourt and Ibadan, based on the prevailing activities in the area. This was used as indicator of atmospheric pollution impact. In Port Harcourt area, the oil companies operating in the region are in the habit of gas flaring. The airborne pollutants from gas-flaring activities were presumed to have dissolved into the rainwater samples collected in Port Harcourt axis. Sample collections in the Port Harcourt axis were carried out in areas where people dwell (Fig. 1), some kilometres from the oil rigs and the gas-flaring point. This was informed by the need to mea- sure direct human exposure impact tendencies and not just the degree of pollution by gas flaring. The possibility of other anthropogenic sources of airborne pollutants in the Port Harcourt metropolis informed the use of Ibadan as control. Ibadan is largely residential and the likely sources of gaseous pollutants are the incineration of solid wastes, vehicular emission, and fumes from electricity generating sets similar to what may be obtainable in Port Harcourt metropolis. The degree of pollution, the level of impact, the extent of contamination of rainwater obtained from atmospheric precipitation, and the human exposure risk in these two cities are not expected to be the same, based on the level of industrial and crude oil exploration activity in Port Harcourt. All sampling containers were high-density polyethylene (HDPE) and were pre-treated. They were previously washed with detergent and soaked in 0.1MHNO3 for 24 h. They were then rinsed with demineralized water. In order to eliminate possible contamination from the ground, the rainwater sam- ples were collected at 1–2 m above the ground surface. These were transported within 24 h after collection in iced-crest to laboratory for analysis and kept in the refrigerator until analysed. Preparation of samples and its handlings were done according to standard procedures (John-De-Zuane 1990; APHA 1995; WHO 2004), with metals in the samples pre- served using 5 mL conc. HNO3 in 1 L of water sample (John- De-Zuane 1990; APHA 1995). Chemical analysis Chemical analysis of DRHRWwas carried out in this section. The samples were digested to liberate metals in organic matrix 21916 Environ Sci Pollut Res (2018) 25:21915–21926 Author's personal copy or adsorbed to particulate matter associated with hydrocarbon combustion. DRHRW sample (50 mL) and 3 mL of conc. HNO3 were heated for 1 h (at 80 °C) and made up to 100 mL in volumetric flask and kept in pre-treated bottles for analysis. Seven metals (Cu, Cd, Pb, Zn, Fe, Ca, and Mg) were quantified in the digested water sample with flame atom- ic absorption spectrometer (FAAS), Bulk scientific 205A model using direct air-acetylene flame method. Analysis qual- ity assurance involved triplicate analysis for precision mea- surement; blank analysis was used to correct for interference and inter-laboratory confirmation of instrumental analysis re- sult of some samples (Alberta Environment 2006). Other pa- rameters used to assess air quality and gas-flaring effects were as follows: acidity, as measure of acid rain tendency, deter- mined by t i t r ime t ry ; phospha te de t e rmined by vanadomolybdophosphoric acid method at 470 nm in a spec- trophotometer; sulphate, as measure of SOx, by turbidimetry method at 420 nm; and nitrate, as indicator of NOx, by ultra- violet spectrophotometric screening method at 220 and at 275 nm (APHA 1995). pH was measured with Hanna pH meter and chloride was determined by argentometric titration. Univariate data exploratory analysis plot, box, and whisker was used to present the spread of the results of chemical anal- ysis using SPSS 17 software while Fuzzy Comprehensive Assessment (FCA) was used for degree of gas-flaring impact evaluation. Fuzzy Comprehensive Assessment principle The detailed explanation of the principle of FCA and its appli- cation in environmental studies can be obtained in literature (Silvert 2000; Lu and Lo 2002; McKone and Deshpande 2005; Ocampo-Duque et al. 2006). The steps taken in the com- putation of fuzzy membership function and fuzzy algorithm in this study have been reported in our previous application of the concept (Adebowale et al. 2008; Agunbiade et al. 2011; Akintola et al. 2016) with its equations summarised below. θa ¼ 1 when 0≤x≤a b−xð Þ b−að Þ when a < x < b 0 when x≥b 2 64 3 75 ð1Þ θb ¼ 0 when x≤a or x≥c x−að Þ b−að Þ when a < x < b 1 c−xð Þ c−bð Þ when x ¼ b when b < x < c 2 666664 3 777775 ð2Þ Fig. 1 Map of the sampling sites. The twelve (12) sampling points in both Ibadan (residential) and Port Harcourt (gas flaring) are represented as dots and numbered 1–12 on the map Environ Sci Pollut Res (2018) 25:21915–21926 21917 Author's personal copy θc ¼ 0 when x≤b or x≥d x−bð Þ c−bð Þ when b < x < c 1 d−xð Þ d−cð Þ when x ¼ c when c < x < d 2 666664 3 777775 ð3Þ θd ¼ 0 when x≤c x−cð Þ d−cð Þ when c < x < d 0 when x≥d 2 64 3 75 ð4Þ where θa–θd are the membership function for the degree of impacts a, b, c, and d respectively; x represents the field data; a–d are the limit criteria for the degree of impact based on WHO (2004) regulatory standard for each parameter (Table 1). The membership functions were then formulated into a membership matrix Xk. The classification to which the harvested water belongs was obtained with Fuzzy Comprehensive Assessment infer- ence Eqs. (5–7) (Chang et al. 2001; Onkal-Engin et al. 2004; Shen et al. 2005). k j ¼ ∑m 1 wiθij ð5Þ ∑m 1 wi ¼ 1 ð6Þ kp ¼ max k j � � ð7Þ with θij being membership function;Wi the weight associated with each impact measurement parameter obtained from the ratio of the concentration of the parameter to the overall weight (Eqs. 8 and 9) (Shen et al. 2005; Adebowale et al. 2008): ai kð Þ ¼ ci kð Þ=Si ð8Þ Wi kð Þ ¼ ai kð Þ=∑ai kð Þ ð9Þ where ci(k) is the concentration of ith parameters in monitoring site k and Si is the average limits used in formulating the membership function for each parameter. Equations (8) and (9) are used to obtain the weight (Wi(k)) and expressed as Pk which is the weight matrix and the product of the membership matrixXk = (θij)nxm and the weight matrixPk = (Wi(k))1xm using the fuzzy algorithm (Eq. 10) is interpreted as the matrix index classification of the degree of impact gas flaring on the sites. X kPk ¼ k1 k2 ⋮ k j 2 664 3 775 ð10Þ The calculations were carried with Microsoft Excel 2016. Table 1 The limits used in the membership formulations (a, b, c, d) based on the regulatory standard limits in relation to the degree of pollution impact on investigated sites Parameter Classifications Low impact (a) mg L−1 Medium impact (b) mg L−1 High impact (c) mg L−1 Extremely high impact (d) mg L−1 Pb 0.01 0.05 0.10 0.50 Cu 1.0 2.0 5.0 10.0 Cd 0.005 0.01 0.05 0.25 Fe 2.0 10.0 20.0 50.0 Mg 100 250 500 1000 Ca 50 100 250 500 Zn 2.0 5.0 10.0 20.0 Chloride 100 250 500 750 pH > 7 7.0 7.5 8.0 9.0 pH < 7 7.0 6.5 6.0 5.0 Sulphate 100 250 500 1000 Nitrate 0.2 0.5 1.0 2.0 Acidity 10 25 50 100 Phosphate 0.2 0.5 1.0 2.5 21918 Environ Sci Pollut Res (2018) 25:21915–21926 Author's personal copy Results Quality parameters In Fig. 2, we present the results of concentrations, median, and the spreads of the parameters. The bar within the box repre- sents the median (which is a measure of central tendency), and the ends of the whiskers and the edges of the box represent the concentration range (minimum and maximum) as well as the upper and lower quartiles respectively. Modelling pollution impact Figure 3a, b presents the fuzzy membership function matrixes and the corresponding weight matrixes respectively for inves- tigating gas-flaring environment (Port Harcourt). This was computed for the seven metals and six quality parameters in 12 sampling points. In Fig. 4a, b, the fuzzy membership func- tion matrixes and the corresponding weight matrixes respec- tively for reference environment (Ibadan) are presented. These were also computed for the seven metals and six quality pa- rameters in 12 sampling points. The product of the member- ship functions and their weight were used to evaluate the de- gree of impact based on the class to which each parameter belongs. Fuzzy algorithm results Figure 5a, b is the results of fuzzy product moment algorithm of Port Harcourt and Ibadan environments respectively. The DRHRW in these environments were impacted with metals from both natural and anthropogenic sources. Discussion Quality parameters In Fig. 2a, the highest metal concentration, the widest range, and the highest median were observed and quartile Fig. 2 Box and whisker presentation of the metals and other quality parameters investigated, where the x-axis is the metal/quality parameter’s concentration (mg L−1, except for pH that is dimensionless). a Box and whisker presentation of Cd, Cu, and Pb in Ibadan (residential) and Port Harcourt (gas-flaring) environs. b Box and whisker presentation of Zn, Ca, Mg, and Fe in Ibadan (residential) and Port Harcourt (gas-flaring) environs. c Box and whisker presentation of sulphate, pH, and Cl in Ibadan (residential) and Port Harcourt (gas-flaring) environs. d Box and whisker presentation of phosphate, acidity, and nitrate in Ibadan (residential) and Port Harcourt (gas flaring) environs Environ Sci Pollut Res (2018) 25:21915–21926 21919 Author's personal copy concentrations were attributable to Cu compared to other metals presented in the figure. The median concentration of Cu in gas-flaring site (Port Harcourt) is moderately higher than the domestic/background site (Ibadan). The observed Cu concentration may be attributed to a number of factors of which gas flaring may be a part. The possibility of contribu- tion from the alloys of the flaring pipeline or other related sources is suggested. More importantly, Pb and Cd median concentrations observed indicate higher values in gas-flaring sites than residential si tes (Fig. 2a). The median Sample Point 1 1 1 1 1 1 1 1 1 0 1 1 0 1 0 0 0 0 0 0 0 0 0.60 0 0 0.75 0 0 0 0 0 0 0 0 0 0.40 0 0 0.25 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Sample Point 2 0 1 0 1 1 1 1 1 0.46 1 0.99 0 1 0.40 0 0.80 0 0 0 0 0 0.54 0 0.01 0.99 0 0.60 0 0.20 0 0 0 0 0 0 0 0 0.01 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Sample Point 3 0.50 1 0 1 1 1 1 1 0 1 1 0 1 0.50 0 0.40 0 0 0 0 0 0.86 0 0 0.52 0 0 0 0.60 0 0 0 0 0 0.14 0 0 0.48 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Sample Point 4 0.25 1 0 1 1 1 1 1 0 1 0.98 0.17 1 0.75 0 1 0 0 0 0 0 1 0 0.02 0.83 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Sample Point 5 0 1 0 1 1 1 1 1 0.34 1 1 0.18 1 0 0 0.80 0 0 0 0 0 0.66 0 0 0.82 0 1 0 0.20 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Sample Point 6 0 1 0 1 1 1 1 1 0 1 1 0 1 1 0 0.60 0 0 0 0 0 0 0 0 0.86 0 0 0 0.40 0 0 0 0 0 1 0 0 0.14 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Sample Point 7 0 1 0.50 1 1 1 1 1 0 1 1 0.61 1 0.80 0 0.50 0 0 0 0 0 0.80 0 0 0.39 0 0.20 0 0 0 0 0 0 0 0.20 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Sample Point 8 0 1 0 1 1 1 1 1 0.86 1 1 0 1 0.20 0 0 0 0 0 0 0 0.14 0 0 0.86 0 0.80 0 0.88 0 0 0 0 0 0 0 0 0.14 0 0 0 0.12 0 0 0 0 0 0 0 0 0 0 Sample Point 9 0 1 1 1 1 1 1 1 0 1 0.99 0.16 1 0.40 0 0 0 0 0 0 0 0 0 0.01 0.84 0 0.60 0 0 0 0 0 0 0 0.87 0 0 0 0 0 0 0 0 0 0 0 0 0.13 0 0 0 0 Sample Point 10 0.75 1 0 1 1 1 1 1 0 1 1 0.16 1 0.25 0 1 0 0 0 0 0 0.74 0 0 0.84 0 0 0 0 0 0 0 0 0 0.26 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Sample Point 11 0.50 1 0.50 1 1 1 1 1 0.26 1 1 0.17 1 0.50 0 0.50 0 0 0 0 0 0.74 0 0 0.83 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Sample Point 12 0 1 0.50 1 1 1 1 1 0 1 1 0 1 0 0 0.50 0 0 0 0 0 0 0 0 0.93 0 0.93 0 0 0 0 0 0 0 0.03 0 0 0.07 0 0.07 0 0 0 0 0 0 0 0.97 0 0 0 0 a Fig. 3 a Fuzzymembership function arranged into evaluationmatrix for Port Harcourt, Nigeria. bCorresponding fuzzy weight matrix for Port Harcourt, Nigeria, arranged along the pattern of evaluation matrix (a) 21920 Environ Sci Pollut Res (2018) 25:21915–21926 Author's personal copy concentrations of Pb and Cd in the site associated with gas flaring are higher than the upper quartile concentration Pb and Cd in residential area. This implies that there are generally higher Pd and Cd contents in the water samples collected in atmosphere with gas flaring than the residential environment. Hence, there are strong indications that anthropogenic impacts suspected to be connected to high-level gas flaring and crude oil refining activities in Port Harcourt are contributing to low- ering the quality of the atmosphere and the harvested water than the residential environment. The samples from Port Harcourt city were collected several kilometres distance from refineries and oil rigs scattered around the city and the obtain- ed Pb and Cd being high indicates a high tendency of air dispersion of these pollutants and widespread exposure risk from the polluted air. In comparison with residential samples (Ibadan), the Cd concentration in the gas-flaring region (Port Harcourt) despite the distance to polluting source was of a median value Cd of 50 μg L−1 compared to 20 μ gL−1 (in Ibadan) which is 2.5-fold levels of Cd pollution between res- idential and gas-flaring environments. It is logical to postulate that the concentration will be several orders of magnitude higher close to the refinery, flaring, and oil installation sites. Factors that can affect these concentrations include direction and speed of wind, particulate matter concentration, pH, and rainfall index. The concentrations of Cd and Pb in almost all of the roof-harvested water from Port Harcourt were higher than the WHO recommended standard for drinking water of 3 and 10 μg L−1 for Cd and Pb respectively (WHO 1992, 2004). This implies that Port Harcourt is significantly impacted with Cd and Pb pollutants. The main emission source of Pb into the atmosphere has been reported to be the combustion of gasoline (Carreras et al. 2009). Cd is usually considered a highly mobile heavy metal in regard to moving from soil to plant (Jia et al. 2010); hence, there is the possibility of Cd- contaminated rainwater causing accumulation of Cd in plant tissues and its transfer to man via food chain along- side the exposure risk of application of Cd- and Pb- contaminated rainwater for domestic purposes. From Fig. 2b, no definite trend was observed for these metals even though their spreads in both cities are similar. Generally, the concentrations of these four metals are higher than the concentrations of the much more toxic Cd and Pb. Zinc and Cu are known to be essential for all plants (Kabata- Pendias and Pendias 2001; Zhang et al. 2011), though they can have toxic effects on plants at over dosage. Summarily, the concentrations of metals in Fig. 2b as well as that of Cu in Fig. 2a are considered moderate because of their beneficial bio- chemical functions. Calcium, Mg, and chloride average concentrations are within the USEPA (1993) and WHO (2004) recommended standard of 75, 500, and 250 mg L−1 for Ca, Mg, and chloride respectively. Values of Ca for Ibadan and Port Harcourt re- spectively range from 0.43 to 0.96 and from 0.85 to 1.23 mg L−1 while that of Mg range from 0.69 to 0.98 and from 0.72 to 0.99 mg L−1, and Cl− contents are in the range 13.3–31.8 and 16.6–30.7 mg L−1 respectively for Ibadan and Port Harcourt. All the samples examined for Fe recorded values above specification for drinking water standard of 0.3 mg L−1 by WHO and USEPA standard specification for drinking water. They however fall within the alternative ex- tended limit of 1.0 mg L−1. Concentration of Fe for Ibadan and Port Harcourt ranges from 0.39 to 1.13 and from 0.35 to 1.13 mg L−1 respectively. P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 Pb 0.059 0.277 0.109 0.172 0.342 0.193 0.292 0.234 0.350 0.098 0.162 0.413 Cu 0.034 0.022 0.015 0.034 0.016 0.040 0.023 0.008 0.025 0.040 0.028 0.021 Cd 0.000 0.208 0.291 0.214 0.205 0.271 0.146 0.391 0.000 0.244 0.162 0.095 Fe 0.019 0.013 0.009 0.005 0.019 0.009 0.009 0.005 0.016 0.023 0.028 0.006 Mg 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.000 0.001 0.001 0.001 0.000 Ca 0.003 0.002 0.002 0.002 0.002 0.002 0.002 0.001 0.002 0.003 0.003 0.001 Zn 0.090 0.020 0.035 0.029 0.021 0.018 0.023 0.022 0.037 0.028 0.038 0.026 Chloride 0.020 0.016 0.016 0.023 0.015 0.017 0.026 0.011 0.023 0.017 0.029 0.011 pH 0.287 0.180 0.180 0.214 0.175 0.178 0.246 0.139 0.237 0.248 0.276 0.019 Sulphate 0.009 0.005 0.004 0.005 0.004 0.006 0.007 0.002 0.007 0.007 0.007 0.004 Nitrate 0.091 0.071 0.056 0.088 0.035 0.040 0.050 0.027 0.089 0.050 0.000 0.048 Acidity 0.370 0.175 0.270 0.193 0.153 0.221 0.154 0.148 0.197 0.220 0.243 0.171 Phosphate 0.017 0.012 0.010 0.021 0.011 0.014 0.021 0.011 0.015 0.021 0.022 0.012 b Fig. 3 (continued) Environ Sci Pollut Res (2018) 25:21915–21926 21921 Author's personal copy Modelling pollution impact In Port Harcourt (Fig. 3a), Pb was detected in every sampling point while Cd was detected in 83.3% of the samples. Lead and Cd accounted for more than 30% of the contamination load in 92% of the sampling points (Fig. 3b). The combination of these two metals as expressed in the weight matrixes in Fig. 3b is as high as 51 and 63% of contamination loads in sam- pling points 12 and 8 respectively. Comparatively, from fuzzy matrixes in Ibadan (residential environment, Fig. 4a), the wa- ter quality was generally between the low and medium impact classifications (θa and θb respectively), except in sampling point 5 that are contaminated with Pb and Cd in sampling points 1 and 6 that had membership in high impact classifica- tion. The classifications of Pb and Cd in the sites that raise concern are the ones with medium and high membership (θb and θc) classification. Cd and Pb are found in these unaccept- able medium and high membership classifications in almost Sample Point 1 Pb Cu Cd Fe Mg Ca Zn Chloride pH Sulphate Nitrate Acidity Phosphate 1 1 0 1 1 1 1 1 0.26 1 1 0 1 0 0 0.40 0 0 0 0 0 0.74 0 0 0.86 0 0 0 0.60 0 0 0 0 0 0 0 0 0.14 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Sample Point 2 Pb Cu Cd Fe Mg Ca Zn Chloride pH Sulphate Nitrate Acidity Phosphate 0 1 1 1 1 1 1 1 0.86 1 0.80 0 1 0.60 0 0 0 0 0 0 0 0.14 0 0.20 0.34 0 0.40 0 0 0 0 0 0 0 0 0 0 0.66 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Sample Point 3 Pb Cu Cd Fe Mg Ca Zn Chloride pH Sulphate Nitrate Acidity Phosphate 0.50 1 1 1 1 1 1 1 0.86 1 0.80 0 1 0.50 0 0 0 0 0 0 0 0.14 0 0.20 0.86 0 0 0 0 0 0 0 0 0 0 0 0 0.14 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Sample Point 4 Pb Cu Cd Fe Mg Ca Zn Chloride pH Sulphate Nitrate Acidity Phosphate 0.75 1 1 1 1 1 1 1 0 1 0.99 0 1 0.25 0 0 0 0 0 0 0 0.74 0 0.01 0.86 0 0 0 0 0 0 0 0 0 0.26 0 0 0.14 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Sample Point 5 Pb Cu Cd Fe Mg Ca Zn Chloride pH Sulphate Nitrate Acidity Phosphate 0 1 0.50 1 1 1 1 1 0.66 1 0.99 0.09 1 0 0 0.50 0 0 0 0 0 0.34 0 0.01 0.91 0 0.98 0 0 0 0 0 0 0 0 0 0 0 0 0.02 0 0 0 0 0 0 0 0 0 0 0 0 Sample Point 6 Pb Cu Cd Fe Mg Ca Zn Chloride pH Sulphate Nitrate Acidity Phosphate 0.25 1 0 1 1 1 1 1 1 1 1 0 1 0.75 0 0.80 0 0 0 0 0 0 0 0 0.86 0 0 0 0.20 0 0 0 0 0 0 0 0 0.14 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Sample Point 7 Pb Cu Cd Fe Mg Ca Zn Chloride pH Sulphate Nitrate Acidity Phosphate 0 1 0.25 1 1 1 1 1 0.80 1 1 0.17 1 0.80 0 0.75 0 0 0 0 0 0.20 0 0 0.83 0 0.20 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Sample Point 8 Pb Cu Cd Fe Mg Ca Zn Chloride pH Sulphate Nitrate Acidity Phosphate 1 0.5 1 1 1 1 1 1 0 1 1 0.18 1 0 0.5 0 0 0 0 0 0 0 0 0 0.82 0 0 0 0 0 0 0 0 0 0.77 0 0 0 0 0 0 0 0 0 0 0 0 0.23 0 0 0 0 Sample Point 9 Pb Cu Cd Fe Mg Ca Zn Chloride pH Sulphate Nitrate Acidity Phosphate 1 1 1 1 1 1 1 1 0.34 1 0.64 0 1 0 0 0 0 0 0 0 0 0.66 0 0.36 0.99 0 0 0 0 0 0 0 0 0 0 0 0 0.01 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Sample Point 10 Pb Cu Cd Fe Mg Ca Zn Chloride pH Sulphate Nitrate Acidity Phosphate 0.25 1 0.50 1 1 1 1 1 1 1 0.80 0 1 0.75 0 0.50 0 0 0 0 0 0 0 0.20 0.54 0 0 0 0 0 0 0 0 0 0 0 0 0.46 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Sample Point 11 Pb Cu Cd Fe Mg Ca Zn Chloride pH Sulphate Nitrate Acidity Phosphate 1 1 1 1 1 1 1 1 0 1 1 0 1 0 0 0 0 0 0 0 0 0.34 0 0 0.98 0 0 0 0 0 0 0 0 0 0.66 0 0 0.02 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Sample Point 12 Pb Cu Cd Fe Mg Ca Zn Chloride pH Sulphate Nitrate Acidity Phosphate 0.75 1 1 1 1 1 1 1 0 1 0.82 0 1 0.25 0 0 0 0 0 0 0 0.60 0 0.18 0.86 0 0 0 0 0 0 0 0 0 0.40 0 0 0.14 0 0 0 0 0 0 0 0 0 0 0 0 0 0 a Fig. 4 a Fuzzy membership function arranged into evaluation matrix for Ibadan, Nigeria. b Corresponding fuzzy weight matrix for Ibadan, Nigeria, arranged along the pattern of evaluation matrix (a) 21922 Environ Sci Pollut Res (2018) 25:21915–21926 Author's personal copy all the sampling sites in Port Harcourt. Owing to the toxicity of Cd and Pb, it would have been preferred that their classifica- tion is low as against medium or high. Notable are the completely unacceptable classifications of Pb in Port Harcourt sampling point 5 (θc = 1) and point 12 (θc = 0.93 and θd = 0.07) which fall predominantly into high impact and extremely high impact category. Moreover, Cd in Port Harcourt sampling point 8 raise exposure risk with a high impact and extremely high impact categories (θc = 0.88 and θd = 0.12 respectively). Cd and Pb are expected to be at low impact classification in Ibadan, being a residential area. These metals were however found with membership functions in medium and high impact in some of the sampling sites (Fig. 4a). In sampling point 1, Pb was distributed between medium impact (θb = 0.60) and high impact (θc = 0.40). A similar Pb result was observed in point 7 with 80% (θb = 0.80) membership functions in medium and 20% (θc = 0.20) in high impact. There was a worrisome Pb pollution observed in the rainwater harvested from sampling point 7 (Fig. 4a) with concentration in the borderline of high impact (98%) and extremely high impact (2%). Similarly, Cd was observed in point 1 to be spread between 40% (θb = 0.40) membership functions in medium and 60% (θc = 0.60) in high impact which can cause toxic exposure risk. The result of Cd pollution in sampling point 6 was also a distribution between medium (θb = 0.80) and high impact (θc = 0.20). In all the other sampling site in Ibadan, both Pb and Cd were either in low impact classification (i.e. not detected or less than regula- tory standard for drinking water) or spread between low im- pact and medium impact with larger membership function in the low impact. The contributions of Pb and Cd pollution observed in some of the sampling sites are attributable to Point 1 Point 2 Point 3 Point 4 Point 5 Point 6 Point 7 Point 8 Point 9 Point 10 Point 11 Point 12 0.343 0.450 0.207 0.000 0.244 0.548 0.210 0.000 0.203 0.466 0.329 0.000 0.282 0.710 0.000 0.000 0.211 0.405 0.383 0.000 0.147 0.546 0.317 0.000 0.329 0.563 0.108 0.000 0.207 0.194 0.552 0.047 0.245 0.306 0.416 0.031 0.298 0.637 0.064 0.000 0.431 0.568 0.000 0.000 0.177 0.207 0.402 0.213 Point 1 Point 2 Point 3 Point 4 Point 5 Point 6 Point 7 Point 8 Point 9 Point 10 Point 11 Point 12 0.234 0.425 0.552 0.375 0.329 0.414 0.455 0.400 0.403 0.485 0.356 0.366 0.238 0.288 0.406 0.507 0.265 0.507 0.499 0.292 0.594 0.374 0.425 0.483 0.427 0.288 0.043 0.116 0.384 0.080 0.048 0.239 0.003 0.140 0.222 0.151 0.000 0.00 0.000 0.000 0.000 0.000 0.000 0.102 0.000 0.000 0.000 0.000 a b Fig. 5 Results of fuzzy algorithm of each of the site performed by the product moment of the fuzzy and weight matrices for a Port Harcourt and for b Ibadan, Nigeria P2= P3= P4= P5= P6= P7= P8= P9= P10= P11= P12= Pb 0.000 0.243 0.161 0.112 0.392 0.160 0.240 0.000 0.000 0.167 0.000 0.110 Cu 0.023 0.087 0.047 0.052 0.013 0.016 0.019 0.016 0.040 0.022 0.024 0.065 Cd 0.352 0.035 0.000 0.000 0.107 0.240 0.160 0.170 0.000 0.126 0.063 0.000 Fe 0.010 0.007 0.012 0.018 0.013 0.015 0.015 0.032 0.018 0.012 0.024 0.016 Mg 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 Ca 0.002 0.001 0.002 0.002 0.002 0.002 0.002 0.003 0.002 0.002 0.003 0.002 Zn 0.029 0.027 0.038 0.042 0.022 0.018 0.046 0.054 0.040 0.022 0.060 0.021 Chloride 0.015 0.012 0.019 0.015 0.016 0.025 0.018 0.030 0.028 0.019 0.034 0.024 pH 0.216 0.185 0.253 0.275 0.170 0.207 0.190 0.310 0.311 0.195 0.327 0.268 Sulphate 0.006 0.005 0.004 0.006 0.004 0.004 0.005 0.007 0.007 0.005 0.009 0.008 Nitrate 0.067 0.089 0.137 0.114 0.073 0.061 0.102 0.087 0.195 0.107 0.097 0.140 Acidity 0.251 0.289 0.306 0.319 0.168 0.227 0.180 0.253 0.322 0.305 0.320 0.313 Phosphate 0.025 0.020 0.021 0.042 0.200 0.025 0.024 0.039 0.036 0.017 0.041 0.032 b Fig. 4 (continued) Environ Sci Pollut Res (2018) 25:21915–21926 21923 Author's personal copy anthropogenic sources which are largely due to the effects of increasing domestic and commercial activities in Ibadan. The environment though largely residential still have some small- and medium-scale business activities which operate on self- generated electricity using gasoline and diesel-powered elec- tricity generators. Comparatively, the values of the impact’s membership func- tions and weights of Pb and Cd were higher in Port Harcourt (gas-flaring) environment than that of Ibadan (residential). Lead and Cd in gas-flaring area of Port Harcourt, Nigeria, were majorly in the medium and high impact classifications. Relatively higher impacts of Pb and Cd in Port Harcourt area are attributable to anthropogenic sources which are strongly associated but not limited to gas-flaring activity in the oil in- stallations around Port Harcourt; industrial activity in the Port Harcourt refinery; and activities of other industries within Port Harcourt City. Therefore, it is not safe for Port Harcourt city residents to depend on DRHRW as source of potable water without treatments due to potential human exposure risk and its health implications. Lead and Cd have been implicated in some serious medical conditions such as disruption of enzyme activities in human, lung cancer, impaired intellectual develop- ment in infancy, hypertension, peripheral vascular disease, in- creased adult mortality, cognitive decline in older people, low bone mineral density, and osteoporosis (Alfven et al. 2000; Low et al. 2000; Canfield et al. 2003; Alfven et al. 2004; Needleman 2004; Navas-Acien et al. 2004; Bulut and Baysal 2006; Schober et al. 2006; Silvera and Rohan 2007; Navas- Acien et al. 2007; Whitfield et al. 2010). Hence, there is the need for continuous monitoring of the activities of the industries in the area. There are also needs for regular evaluation of contents of the gaseous effluents from the refineries. More importantly, stiffer measures towards end- ing gas-flaring activities in Port Harcourt and Niger Delta region of Nigeria must be introduced so as to ensure strict compliance with international standards for the discharge of industrial effluents. In general, Cu, Fe, Mg, Ca, and Zn contribution to the overall pollution load is minimal. They also fall within the low impact classification (Figs. 3a and 4a). Hence, they are of safer concentrations, especially since they are beneficial metals to human biochemical systems. Contribution of Cu, Fe, Mg, Ca, and Zn put together was less than 10% in Port Harcourt environment with the exception of location 1 (14.7%) (Fig. 3b), while that of Ibadan environment put to- gether was less than 12.5% in all locations (Fig. 4b). In a similar manner, chloride, sulphate, and phosphate all com- bined contributed less than 6% pollution load in Port Harcourt environment (Fig. 3b) and less than 8.5% of the total pollution load in Ibadan environment (Fig. 4b). The overall contribution of Cu, Fe, Mg, Ca, Zn, chloride, sulphate, and phosphate, relative to the total pollution loads in both Port Harcourt and Ibadan environment, were less than 13 and 20% respectively with the exception of location 1 in Port Harcourt where the overall contribution is 19.3%. In Port Harcourt environment (Fig. 3b), the acidity and pH were in the range of weight contribution range of 14.8% (sam- ple point 8) to 37% (sample point 1) and 1.9% (sample point 12) to 28.7% (sample point 1) respectively. Comparatively, acidity and pH contributed 16.8% (sample point 5) to 32.2% (sample point 9) and 17.0% (sample point 5) to 32.7% (sam- ple point 11) respectively in Ibadan (Fig. 4b) environment. The fuzzy matrix placed the pH in Ibadan at the low impact classification with few cases in medium impact classification but the pH of Port Harcourt samples was majorly (58.3%) in the medium impact, 16.7% each in the low impact and high impact, and 8.3% in the extremely high impact class. The results obtained for acidity could be attributable to the emis- sions of SOx, NOx, etc. from industries, the crude oil refining activities, and the flaring of associated gas around Port Harcourt area. Natural gas contains acidic gases while there are non-hydrocarbon components (nitrogen and sulphur com- pounds) also in crude oil. These components may be respon- sible for the emissions of SOx, NOx, etc. which when reacted with water molecules in the atmosphere produce H2SO4 and HNO3 responsible for the acidity and low pH witnessed. Moreover, the low pH values of the water sample will keep the metals in the DRHRW in their ionic more toxic forms (Campbell 1996). Lastly, reduced pH and increased acidity of rainwater will increase corrosion rate of the roofing mate- rials, increase metal leaching from such roof, and cause high metal concentration into the DRHRW making it more unsafe for potable uses (Akintola et al. 2016). Fuzzy algorithm results For Port Harcourt environment, 75% of the samples collected (nine sampling points) were in the medium impact class with their maximum membership function in this region (Fig. 5a). These nine sampling points also havemembership value in the high impact class while the three remaining sampling points belong to the high impact class with their max kj in this class. In addition, all the 12 sampling points have membership func- tions in the low impact class ranging from 14.7% (sampling point 6) to 34.3% (sampling point 1). Sampling points 8, 9, and 12 also have membership functions (4.7, 3.1, and 21.3% respectively) in the extremely high impact class. Hence, we can conclude that the water quality of the DRHRW from Port Harcourt is between the medium and high impact class. Generally, membership functions in the range of 17.7– 43.1% were in the low impact class; 20.7–71.0% membership classified the water as medium impact; 6.4–55.2% were ob- tained for the high impact water system and 3.1–21.3% were in the extremely high class, in contrast to the water samples from Ibadan (residential environment) that had only one sam- ple with membership in extremely high impact class (Fig. 5a). 21924 Environ Sci Pollut Res (2018) 25:21915–21926 Author's personal copy Expectedly, the algorithm product of the reference environ- ment (Ibadan) classified the environment asmajorly in the low impact class (23.4–55.2%), with 23.8–59.4 and 0.3–42.7% in the medium and high impact class respectively. We have only one sampling point (sampling point 8) in extremely impact class (Fig. 5b). Four out of the 12 sampling points (sampling points 2, 3, 8, and10) are predominantly in the low impact category. Although the maximum values Kj for seven sam- pling points were in the medium impact class, they also have significant membership in the low impact region. Hence, sam- ples from the reference environment can be accurately classi- fied as between low and medium impact classification. Conclusions The results of this study indicate that there are negative im- pacts of anthropogenic activities on Port Harcourt air quality and its resultant rainwater. Gas flaring was considered to be a significant contributor. Ibadan (residential) environment was also found to beminimally contaminated with metal generated mainly from various anthropogenic activities such as domestic and commercial activities. The additional impacts of anthro- pogenic activities such as industrial gaseous effluents, the ef- fects of oil refining and exploration activities, and the impact of gas flaring in Port Harcourt environment make the harvest- ed rainwater unsafe for domestic use without treatment and call for continuous monitoring and stringent regulatory activ- ities in the area. There are indications of potential exposure hazards to the environment and to the dense population of people living in these areas if there is direct application of DRHRW for potable uses, especially for Port Harcourt resi- dents. Unfortunately, in order to meet their daily water needs, residents of these communities rely on DRHRW to augment water from shallow wells scattered around homes due to the collapse of public water supply. Findings from this research indicated that Cd and Pb are priority metals that must be con- trolled in the investigated areas. The contributions of Cu, Fe, Mg, Ca, and Zn are also observed but the quantities observed were still within safe range recommended by regulating au- thorities. It is suggested that public and private partnership should be put in place to provide potable water and minimise dependence on the direct use of impacted DRHRW. Also, there are needs for closer monitoring of industries, especially those involved in oil exploration and their gaseous effluent discharges by relevant government agencies to ensure strict adherence to international standards of treatments of industrial effluents before discharge into the atmosphere. There is also the urgent need to stop gas-flaring activities in Port Harcourt area and reduce the contribution of greenhouse gases in par- ticular within the Niger Delta region of Nigeria and to mini- mise the untold health hazard that people living in the area are constantly faced with. References Adebowale KO, Agunbiade FO, Olu-Owolabi BI (2008) Fuzzy compre- hensive assessment of metal contamination of water and sediments in Ondo estuary, Nigeria. Chem Ecol 24(4):269–283 Agunbiade FO, Awe AA, Adebowale KO (2011) Fuzzy logic-based modelling of the impact of industrial activities on the environmental status of an industrial estate in Nigeria. Toxicol Environ Chem 93(10):1856–1879 Akintola OA, Sangodoyin AY, Agunbiade FO (2013) Evaluation of en- vironmental pollution effects on domestic roof-harvested rainwater in southern part of Nigeria using impact indices. Water Pract Technol 8(2):244–255 Akintola OA, Sangodoyin AY, Agunbiade FO (2016) Fuzzy logic model- ling of the effects of pollution on domestic roof-harvested rainwater quality in residential and industrial environments. Arch Environ Contam Toxicol 71(1):113–121. https://doi.org/10.1007/s00244- 016-0278-4 Alberta Environment (2006). Guidelines for quality assurance and quality control in surface water quality programs in Alberta. Prepared by Patricia Mitchell environmental consulting for Alberta Environment, W0603. 67pages Alfven T, Elinder CG, CarlssonMD, Grubb A, Hellstrom L, Persson B et al (2000) Low-level cadmium exposure and osteoporosis. J Bone Miner Res 15:1579–1586 Alfven T, Elinder CG, Hellstrom L, Lagarde F, Jarup L (2004) Cadmium exposure and distal forearm fractures. J Bone Miner Res 19:900– 905 APHA –American Public Health Association, (1995). Standard methods for the examination of water and wastewater, 19th ed. Washington, DC Bulut Y, Baysal Z (2006) Removal of Pb (II) from wastewater using wheat bran. J Environ Manag 78:107–113 Campbell, P.G.C. (1996). Interactions between trace metals and aquatic organisms: a critique of the free ion activity model. Metal speciation and bioavailability in aquatic systems. (Eds. Andre Tessier and David R. Turner), New York: John Wiley &Sons, Incorporated, 3, pp 45–102 Canfield RL, Henderson CR Jr, Cory-Slechta DA, Cox C, Jusko TA, Lanphear BP (2003) Intellectual impairment in children with blood lead concentrations below 10 microg per deciliter. N Engl J Med 348:1517–1526 Carreras HA, Wannaz ED, Pignata ML (2009) Assessment of human health risk related tometals by the use of biomonitors in the province of Córdoba, Argentina. Environ Pollut 157:117–122 Chang NB, Chen HW, Ning SK (2001) Identification of river water qual- ity using the fuzzy synthetic evaluation approach. J Environ Manag 63:293–305 De Zuane John PE (1990) Handbook of drinking water quality: standards and control. John Wiley & Sons, New York, USA Jia L, WangW, Li Y, Yang L (2010) Heavy metals in soil and crops of an intensively farmed area: a case study in Yucheng City, Shandong Province, China. Int J Environ Res Public Health 7:395–412. https:// doi.org/10.3390/ijerph7020393 Kabata-Pendias A, Pendias H (2001) Trace elements in soils and plants (3rd ed.). CRC Press, Boca Raton, FL Low KS, Lee GK, Liew SC (2000) Sorption of cadmium and lead from aqueous solutions by spent grain. Process Biochem 36: 59–64 Lu RS, Lo SL (2002) Diagnosing reservoir water quality using self- organizing maps and fuzzy theory. Water Res 36:2265–2274 McKone TE, Deshpande AW (2005) Can fuzzy logic bring complex environmental problems into focus? Environ Sci Technol 39:42A– 47A Environ Sci Pollut Res (2018) 25:21915–21926 21925 Author's personal copy https://doi.org/10.1007/s00244-016-0278-4 https://doi.org/10.1007/s00244-016-0278-4 https://doi.org/10.3390/ijerph7020393 https://doi.org/10.3390/ijerph7020393 Navarro Silvera SA, Rohan TE (2007) Trace elements and cancer risk: a review of the epidemiologic evidence. Cancer Causes Control 18:7– 27 Navas-Acien A, Selvin E, Sharrett AR, Calderon-Aranda E, Silbergeld E, Guallar E (2004) Lead, cadmium, smoking, and increased risk of peripheral arterial disease. Circulation 109:3196–3201 Navas-Acien A, Guallar E, Silbergeld EK, Rothenberg SJ (2007) Lead exposure and cardiovascular disease—a systematic review. Environ Health Perspect 115:472–482 Needleman H (2004) Lead poisoning. Annu Rev Med 55:209–222 Ocampo-Duque W, Ferré-Huguet N, Domingo JL, Schuhmacher M (2006) Assessing water quality in rivers with fuzzy inference sys- tems: a case study. Environ Int 32:733–742 Onkal-Engin G, Demir I, Hiz H (2004) Assessment of urban air quality using fuzzy synthetic evaluation. Atmos Environ 38:3809–3815 Roderick, P. (2005). Gas flaring in Nigeria: a human right, environmental and economic monstrosity. https://www.foe.co.uk/sites/default/files/ downloads/gas_flaring_nigeria.pdf(downloaded on Aug. 3, 2016) Sadegh-Zadeh K (2001) The fuzzy revolution: goodbye to the Aristotelian Weltanschauung. Artif Intell Med 21:1–25 Schober SE, Mirel LB, Graubard BI, Brody DJ, Flegal KM (2006) Blood lead levels and death from all causes, cardio vascular disease, and cancer: results from the NHANES III mortality study. Environ Health Perspect 114:1538–1541 Shen G, Lu Y, Wang M, Sun Y (2005) Status and fuzzy comprehensive assessment of combined heavy metal and organo-chlorine pesticide contamination in the Taihu Lake region of China. Journal. Environ Manag 76:355–362 Silvert W (2000) Fuzzy indices of environmental conditions. Ecol Model 130:111–119 Thomas T (2000) Where is rainwater harvesting going?Waterlines 18(3): 7–10 USEPA (US Environmental Protection Agency) (1993) Water quality standards handbook, 2nd edition, EPA-823-B-93-002 and EPA- 823-B-94-006. Office of Water, Washington, DC, USA Whitfield, J.B., Dy, V., McQuilty, R., Zhu, G., Heath, A.C., Montgomery, G.W., Martin, N.G. (2010). Genetic effects on toxic and essential elements in humans: arsenic, cadmium, copper, lead, mercury, sele- nium, and zinc in erythrocytes Environ Health Perspect 118, 776– 782 (2010). doi:https://doi.org/10.1289/ehp.0901541 World Health Organisation, WHO (1992).Cadmium environmental health criteria, 134.http://www.inchem.org/ (Accessed 27th of May 2007) World Health Organisation, WHO (2004) Guidelines for drinking water quality, third edition. Vol.1 Recommendations. World Health Organization, Geneva Zadeh LA (1965) Fuzzy sets. Inf Control 8:338–353 Zhang H, Cui B, Zhang K (2011) Heavy metal distribution of natural and reclaimed tidal riparian wetlands in south estuary, China. J Environ Sci 23(12):1937–1946 21926 Environ Sci Pollut Res (2018) 25:21915–21926 Author's personal copy https://www.foe.co.uk/sites/default/files/downloads/gas_flaring_nigeria.pdf https://www.foe.co.uk/sites/default/files/downloads/gas_flaring_nigeria.pdf https://doi.org/10.1289/ehp.0901541 http://www.inchem.org/ Anthropogenic... Abstract Introduction Materials and methods Sampling and sampling site description Chemical analysis Fuzzy Comprehensive Assessment principle Results Quality parameters Modelling pollution impact Fuzzy algorithm results Discussion Quality parameters Modelling pollution impact Fuzzy algorithm results Conclusions References