Business

Adaptation and evaluation of the Canadian Council of Ministers of the Environment Water Quality Index (CCME WQI) for use as an effective tool to characterize drinking source water quality

Description
Protecting drinking source water quality is a critical step in ensuring a safe supply of drinking water. Increasingly, drinking source water protection programs rely on the active participation of various stakeholders with differing degrees of water
Categories
Published
of 9
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  Adaptation and evaluation of the Canadian Council of Ministers of the Environment Water Quality Index(CCME WQI) for use as an effective tool to characterizedrinking source water quality Tim Hurley a, *, Rehan Sadiq b , Asit Mazumder a a Water and Aquatic Sciences Research Program, Department of Biology, University of Victoria, PO Box 3020 Station CSC, Victoria, BritishColumbia, Canada V8W 3N5 b School of Engineering, The University of British Columbia Okanagan, Campus Kelowna, BC, Canada V1V 1V7 a r t i c l e i n f o Article history: Received 24 August 2011Received in revised form1 March 2012Accepted 27 March 2012Available online xxx Keywords: Drinking source water qualityWater quality indexWater quality guidelinesRisk communication a b s t r a c t Protecting drinking source water quality is a critical step in ensuring a safe supply of drinking water. Increasingly, drinking source water protection programs rely on the activeparticipation of various stakeholders with differing degrees of water science knowledge.A drinking source water quality index presents a potential communication and analysistool to facilitate cooperation between diverse interest groups as well as representcomposite water quality. We tested the effectiveness of the Canadian Council of Ministersof the Environment Water Quality Index (CCME WQI) in capturing expert assessments of drinking water quality. In cooperation with a panel of drinking water quality experts weidentified a core set of parameters to reflect common source water concerns. Drinking source water target values were drafted for use in the index corresponding to two basictreatment levels. Index scores calculated using the core parameter set and associatedsource water target values were strongly correlated with expert assessments of waterquality. We recommend a modified index calculation procedure to accommodate param-eters measured at different frequencies within any particular study period. The resulting drinking source water CCME WQI provides a valuable means of monitoring, communi-cating, and understanding surface source water quality. ª 2012 Elsevier Ltd. All rights reserved. 1. Introduction Communication among watershed stakeholders, particularlyin the realm of public education has been identified as a keyfacilitating factor for source water protection (Ivey et al., 2006;Patrick, 2008; Islam et al., 2011). Traditional drinking sourcewater quality assessments have relied on a parameter byparameter assessment of all variables that, either individuallyor through interactive effects, contribute to quality conditions(Chang et al., 1999). Such analysis requires a comprehensiveknowledge of drinking water science to understand and maynot provide a composite measure of source drinking waterquality (de Rosemond et al., 2009). Therefore appropriateanalysis and knowledge translation tools are required tobridge source water quality communication gaps among scientists, policy makers and the public.Developed to integrate, interpret, and communicate envi-ronmental monitoring data, indices have been used to * Corresponding author . Tel.: þ 1 250 472 4789; fax: þ 1 250 472 4766.E-mail addresses:hurleytj@uvic.ca(T. Hurley),rehan.sadiq@ubc.ca(R. Sadiq),mazumder@uvic.ca(A. Mazumder).  Available online atwww.sciencedirect.com journal homepage:www.elsevier.com/locate/watres water research xxx (2012) 1 e 9 Please cite this article in press as: Hurley, T., et al., Adaptation and evaluation of the Canadian Council of Ministers of theEnvironment Water Quality Index (CCME WQI) for use as an effective tool to characterize drinking source water quality, WaterResearch (2012), doi:10.1016/j.watres.2012.03.061 0043-1354/$ e see front matter ª 2012 Elsevier Ltd. All rights reserved.doi:10.1016/j.watres.2012.03.061  successfully characterize water quality status and trends andrelay that information to concerned groups (Cude, 2001; Lumbet al., 2006). A water quality index (WQI) combines themeasuresofseveralwaterqualityvariablesinsuchawayastoproduce a single score that is representative of qualityimpairmentsorsuitabilityofuse(Dunnette,1979).Todate,theapplication of an index to characterize and communicatedrinkingsourcewaterqualitydata hasnot beenfully exploredandmoreimportantly,theeffectivenessofanyresultingindexscores have not been sufficiently verified. The very nature of drinking source water quality may be responsible for the lackof an effective source water quality index. Drinking sourcewatersprovide two primarychallenges to index development:1. Site specificity e Parameters of concern at one location andthus necessitating monitoring may not be of concern else-where and therefore are rarely monitored. If a consistentset of parameters is not monitored on a routine basis at allsites then alternative indices are required that can incor-porate all parameters of concern at all locations.2. Treatment considerations e Since source waters are ulti-mately intended for human consumption, they generallymust undergo some form of treatment. A drinking sourcewater quality index must consider the effects of treatment,bothbeneficial(e.g.microbialinactivation)andunfavorable(e.g. disinfection byproducts), and be adaptable to varioustreatment regimes so as to accurately reflect quality.The Canadian Council of Ministers of the EnvironmentWater Quality Index (CCME WQI) provides a flexible indextemplate adaptable to the site specificity and treatmentconsiderations of drinking source water. The CCME WQI is anobjective-based index that compares measured water qualityvalues to guidelines to produce a score ranging from 0, rep-resenting worst quality, to 100, representing best quality.Practitioners are free to select appropriate parameters andguidelines for their purposes therefore accommodating thesite specific and treatment considerations associated withassessing drinking source water. The CCME provides detailedinformation regarding index calculation and application(CCME, 2001, 2003, 2006b). Index scores are calculated asfollows:CCME WQI ¼ 100 À 0@ ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi F 21 þ F 22 þ F 23 q  1 : 732 1A (1)where: F 1 (scope) represents the percentage of the selectedvariables that do not meet their respective guidelines at leastonce during the time period considered, F 2 (frequency)represents the percentage of individual sample measure-ments(tests)thatdonotmeettheirrespective guidelineinthetime period considered, and F 3 (amplitude) represents theamount by which failed measurements do not meet theirrespective guideline, calculated in 3 steps and scaled toa value between 0 and 100.TheCCMEWQIhasbeenusedtocharacterizethequalityof water for several intended uses including agriculture, theprotection of aquatic life and treated drinking water (Khanet al., 2003, 2004; Lumb et al., 2006). Though previouslyapplied to characterize water intended as a source fordrinking purposes (Khan et al., 2003; Boyaciaglu, 2009;Rickwood and Carr, 2009) a standardized methodology forindex application has not been proposed nor has the effec-tiveness of the resulting index scores in capturing expertunderstanding of quality been tested. Furthermore, consid-eration of how to accommodate unequal sampling regimesamong source water parameters has not been addressed.Automated sampling of select parameters can lead to theiroverrepresentation in the CCME WQI calculation and result ina score bias, an issue that requires attention.In this paper we present a modified version of the CCMEWQI that effectively characterizes drinking source waterquality. In consultation with a panel of drinking water qualityexperts we propose a core set of parameters for use in theindex. Based on available literature, defensible drinking source water target values are drafted for each of the coreparameters. Resulting quality scores calculated using the coreset of parameters are verified against expert assessments of drinking source water quality. Pilot testing of the index iscarriedoutusinghistoricalmonitoringdataandanalternativeindex calculation procedure adopted to reduce the impact of parameter unevenness. Finally a sensitivity analysis is per-formed and challenges to widespread index implementationdiscussed. 2. Materials and methods 2.1. Parameter selection In order to facilitate spatial and temporal comparisons among index scores, a core set of source water parameters wasidentified to which appropriate site-specific parameters of concern could be subsequently added. A comprehensivemulti-step procedure similar to that of Dunnette (1979)wasemployed to identify a suite of priority drinking source waterparameters. Listed in order, the parameter selection proce-dure involved: 1) a review of existing index literature, 2) use of a rejection rationale to produce a screened set of frequentlymonitored source water parameters with significant implica-tions for source water quality and published source waterguidelines/objectives, and 3) a single parameter selectionsurvey e-mailed to a panel of drinking water quality expertsfrom academia, government and industry. Prospective surveyparticipants were identified based upon job title, experience,publishedwork,referrals,andexistingrelationships.Atotalof 34 surveys were e-mailed to individuals who had confirmedtheir willingness to participate. Panelists were asked to iden-tify a set of 10 parameters that should be monitored in sourcewater to identify the most common risks (health risks, treat-ment interference, etc.) prior to next stages: chlorination anddistribution. Respondents were instructed to designateparameters as “include” or “don’t include” and assign a rela-tive significance rating. Freedom was given to add variables if desired. 2.2. Source water target value selection Three published, accessible drinking source water guidelinesets were used to draft drinking source water target values for water research xxx (2012) 1 e 9 2 Please cite this article in press as: Hurley, T., et al., Adaptation and evaluation of the Canadian Council of Ministers of theEnvironment Water Quality Index (CCME WQI) for use as an effective tool to characterize drinking source water quality, WaterResearch (2012), doi:10.1016/j.watres.2012.03.061  the identified core parameters. The three source water qualityguideline documents used were: the European EconomicCommunity’s Drinking Water Abstraction Directive 75/440/EEC amended by Directives 79/869/EEC and 91/692/EEC (Officefor Official Publications of the European Communities, 1975),Chang et al.’s Taiwanese source water quality standards(Chang et al., 1999) and the British Columbia Ministry of theEnvironment’s (BC MOE) water quality guidelines (BC MOE,2010). Drafted target values aimed to reflect a conservativeconsensus among the three guideline documents. If availableguidelines differed substantially, BC MOE guideline valueswere accepted due to the availability of supporting docu-mentation. Source water quality target values were proposedanticipating two levels of treatment:1) chlorination alone2) chlorination preceded by (slow sand) filtration.The proposed source water target values are not intendedto represent a definitive set of source water criteria for the 2respective treatment levels. Instead, the target values providea defensible benchmark against which parameter measurescan be compared and violations quantified. 2.3. Index score validation To supportindex applicability and any inferences drawn fromthe resulting index score, calibration with reference to expertopinion is a commonly used index validation technique(Smith,1990; Khanet al., 2004; CCME,2009). The paneliststhatcompleted the initial parameter selection survey weree-mailed a second survey asking for their assessment of 20water quality scenarios. The scenarios were drafted using actual source water data along with simulated parametervalues. Each scenario was composed of 6 values, with theexceptionofonescenariothatonlyhad5values,for6 e 8oftheselected core parameters. A total of six values for eachparameter were chosen based on CCME recommendations(CCME, 2006b). The 20 water quality scenarios were specifi-cally selected to span a broad range of quality conditions.Expert panelists were asked to provide a rank and score foreach scenario corresponding to a slightly modified CCME WQIscoring system tailored to drinking source water quality(Table 1). Furthermore, the ranks and scores were to be madeconsidering the two different treatment levels for whichguidelines were drafted. Experts were also asked to indicatewhich parameters were most important in deciding on eachrank and score.Average expert scores for each scenario were calculatedand compared to the corresponding index scores. The arith-meticmeanofallexpertnumericscoreswasusedtorepresentthe central tendency of the sampled panel. Simple linearregression was used to assess the relationship between indexand mean expert scores under both treatment scenarios. Theresultingregressionlinewascomparedtoatheoretical1:1line( y ¼ x ) to gauge index effectiveness in characterizing drinking source water quality. 2.4. Exploration of factor weights In its present Eq.(1)root mean additive aggregation form, thefactors that compose the CCME WQI are unweighted/equallyweighted. No weights are assigned to the individual constit-uent parameters as the relative significance of the individualparameters is considered to be addressed in their corre-sponding guideline/target values (CCME, 2001). However,assigning different weightings to the factors F 1 , F 2 and F 3 hasnot been explored.Using Excel Solver software (Frontline Systems Inc., NorthLake Tahoe, Nevada) alternative factor weightings wereinvestigated. Factor values for each of the 20 scenarios werecalculated using standard CCME factor formulations (CCME,2001, 2003, 2006b). Factor weightings in Eq.(2)were then optimized so as to minimize the deviation between the meanexpertscoreofeachscenarioandtheresultingweightedindexscore. Scenarios in which optimized weights were heavilyskewed toward 1 factor, assigning weights of near zero to theother 2 were excluded from the analysis. Normality wasverified using the Shapiro e Wilk test before testing for differ-ences in mean optimized factor weightings under eachtreatment level.CCME WQI weighted ¼ 100 À 0@ ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi w 1 F 21 þ w 2 F 22 þ w 3 F 23 q ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi w 1 þ w 2 þ w 3 p  1A (2) 2.5. Index application and sensitivity analysis Index scores calculated using historical water quality moni-toring data were used to investigate the relative contributionof each factor and parameter to resulting scores. Seasonalscores (Winter: December e February, Spring: March e May,Summer: June e August, Fall: September e November) werecalculated for a total of 47 sites located in 3 Canadian prov-inces. Data spanned various time periods between 1990 and2009. Scores were tabulated separately for the two treatmentlevel target value sets. Due to inconsistent monitoring programs, scores could not be calculated for all seasons at allsites. A total of 186 scores were calculated under each treat-ment scenario. British Columbia water quality data wereprovided by the Water and Aquatic Sciences Research Table 1 e Source water quality index scoring system. Rank Score Interpretation Excellent 95.0 e 100.0 Water quality meets all criteriafor use as a source of drinking waterGood 80.0 e 94.9 Water quality rarely or narrowlyviolates criteria for use as asource of drinking waterFair 65.0 e 79.9 Water quality sometimesviolates criteria, possibly by awide margin, for use as a sourceof drinking waterMarginal 45.0 e 64.9 Water quality often violates criteriafor use as a source of drinking water by a considerable marginPoor 0 e 44.9 Water quality does not meet anycriteria for use as a sourceof drinking water water research xxx (2012) 1 e 9 3 Please cite this article in press as: Hurley, T., et al., Adaptation and evaluation of the Canadian Council of Ministers of theEnvironment Water Quality Index (CCME WQI) for use as an effective tool to characterize drinking source water quality, WaterResearch (2012), doi:10.1016/j.watres.2012.03.061  Program at the University of Victoria (3 sites) as well as theBC MOE Environmental Monitoring System Web Reporting (6 sites). Alberta data were collected from the Alberta Envi-ronment’s publicly accessible River Network Station WaterQuality Data program (26 sites). Manitoba values wereprovided by request from Manitoba Water Stewardship’sWater Management Section (12 sites).The effect of the number of samples of each parameter onindex score was investigated using Pielou’s evenness index(Pielou, 1966). Developed as an index of biodiversity, Pielou’sevenness index measures how close in number the pop-ulationsofeachspeciesinacommunityaretoeachother.Theindex is bounded by 0 and 1, with a value of 1 representing a community in which all species are equally abundant anda value of 0 representing a community dominated by a singlespecies.Adaptedtothesourcewaterqualitycontext,thewaterquality parameters represent the different species while theirrespectivenumberofsamplesareequivalenttotheabundanceof each species. The remainder of the sensitivity analysis wasbased on the approach of Rickwood and Carr (2009). 3. Results 3.1. Parameter selection A total of 24 completed surveys were returned (response rateof 71%). The majority of respondents recommended thatturbidity, pH, total organic carbon (TOC), Escherichia coli ,nitrate, total coliforms and iron be included among the vari-ables used to assess general source water quality conditions.These parameters meet the recommendation that sevenparameters be used when calculating CCME WQI scores(CCME, 2006b), this recommendation being more conservativethan the 4 parameter absolute minimum. Pesticides were alsoidentified as an important consideration however the varietyof pesticides species and inconsistency in their monitoring resulted in the exclusion of pesticides from the set of coreparameters. The inclusion of temperature, a very frequentlymonitored parameter with aesthetic and treatment efficacyimplications, along with the expert recommendation thatnitrate and nitrite be considered together so as to allow forcomparisons to guideline values produced a core set of 8representative source water quality parameters (Table 2). E.coli was identified as the most important parameter, followedbyturbidity.Ironreceivedthelowestmeansignificancerating. 3.2. Source water target value selection Proposed source water target values are presented inTable 2.Target values are presented for two anticipated levels of treatment:1) chlorination alone2) chlorination preceded by (slow sand) filtration.Several of the target values correspond with treateddrinking water quality guidelines. For example, the treatmentlevel 1 turbidity target value of 0.3 NTU corresponds with theUnited States Environmental Protection Agency (USEPA) T   a    b    l   e    2    e     C   o   r   e   p   a   r   a   m   e    t   e   r   s   s   e    l   e   c    t   e    d    b   y   s   u   r   v   e   y   p   a   n   e    l    i   s    t   s   a   n    d   c   o   r   r   e   s   p   o   n    d    i   n   g   p   r   o   p   o   s   e    d   s   o   u   r   c   e   w   a    t   e   r    t   a   r   g   e    t   v   a    l   u   e   s .    P   a   r   a   m   e    t   e   r   s   a   r   e    l    i   s    t   e    d    i   n   o   r    d   e   r   o    f    t    h   e   p   e   r   c   e   n    t   o    f   s   u   r   v   e   y   s    t    h   a    t    d   e   s    i   g   n   a    t   e    d    t    h   e   p   a   r   a   m   e    t   e   r   a   s    “    i   n   c    l   u    d   e    ” .    M   e   a   n   s    i   g   n    i    fi   c   a   n   c   e   r   a    t    i   n   g   s   w   e   r   e   c   a    l   c   u    l   a    t   e    d    b   a   s   e    d   o   n   e   a   c    h   e   x   p   e   r    t    ’   s   r   e    l   a    t    i   v   e   s    i   g   n    i    fi   c   a   n   c   e   r   a    t    i   n   g .    T   a   r   g   e    t   v   a    l   u   e   s   r   e    f   e   r    t   o    t   w   o    t   r   e   a    t   m   e   n    t    l   e   v   e    l   s   :    1    )   c    h    l   o   r    i   n   a    t    i   o   n   a    l   o   n   e   a   n    d    2    )   s    l   o   w   s   a   n    d    fi    l    t   r   a    t    i   o   n   a   n    d   c    h    l   o   r    i   n   a    t    i   o   n .     C   o   r   e   p   a   r   a   m   e    t   e   r    (   u   n    i    t   s    )    P   e   r   c   e   n    t    “    i   n   c    l   u    d   e    ”    M   e   a   n   s    i   g   n    i    fi   c   a   n   c   e   r   a    t    i   n   g    (    1    ¼     l    i    t    t    l   e   s    i   g   n    i    fi   c   a   n   c   e ,    5    ¼    v   e   r   y   s    i   g   n    i    fi   c   a   n    t    )    T   r   e   a    t   m   e   n    t    l   e   v   e    l    1   :   c    h    l   o   r    i   n   a    t    i   o   n   a    l   o   n   e    T   r   e   a    t   m   e   n    t    l   e   v   e    l    2   :   s    l   o   w   s   a   n    d    fi    l    t   r   a    t    i   o   n   a   n    d   c    h    l   o   r    i   n   a    t    i   o   n    T   r   e   a    t   m   e   n    t    l   e   v   e    l    1   :   c    h    l   o   r    i   n   a    t    i   o   n   a    l   o   n   e    T   r   e   a    t   m   e   n    t    l   e   v   e    l    2   :   s    l   o   w   s   a   n    d    fi    l    t   r   a    t    i   o   n   a   n    d   c    h    l   o   r    i   n   a    t    i   o   n    M   a   x    i   m   u   m   a   c   c   e   p    t   a    b    l   e   v   a    l   u   e    M    i   n    i   m   u   m   a   c   c   e   p    t   a    b    l   e   v   a    l   u   e    M   a   x    i   m   u   m   a   c   c   e   p    t   a    b    l   e   v   a    l   u   e    M    i   n    i   m   u   m   a   c   c   e   p    t   a    b    l   e   v   a    l   u   e    p    H    9    2    3 .    4    1    8 .    5    6 .    5    8 .    5    6 .    5    T   u   r    b    i    d    i    t   y    (    N    T    U    )    8    8    4 .    2    9    0 .    3    5    T   o    t   a    l   o   r   g   a   n    i   c   c   a   r    b   o   n    (   m   g    /    L    )    8    3    4 .    2    5    4    4         E  .      c      o        l        i     (    C    F    U    /    1    0    0   m    L    )    7    9    4 .    9    5    1    0    1    0    0    N    i    t   r    i    t   e   a   n    d    N    i    t   r   a    t   e   a   s    N    (   m   g    /    L    )    7    5    3 .    3    3    1    0    1    0    T   o    t   a    l   c   o    l    i    f   o   r   m   s    (    C    F    U    /    1    0    0   m    L    )    6    3    4 .    0    0    1    0    0    1    0    0    I   r   o   n    (   m   g    /    L    )    5    0    2 .    3    3    0 .    3    0 .    3    T   e   m   p   e   r   a    t   u   r   e    (          C    )    4    6    3 .    1    8    1    5    1    5 water research xxx (2012) 1 e 9 4 Please cite this article in press as: Hurley, T., et al., Adaptation and evaluation of the Canadian Council of Ministers of theEnvironment Water Quality Index (CCME WQI) for use as an effective tool to characterize drinking source water quality, WaterResearch (2012), doi:10.1016/j.watres.2012.03.061  standard against which 95% of treated water samples ina month must be lower (USEPA, 2009). A 0.3 NTU target valuewas selected to better align index and expert scores and toreflect the importance that panelists place on turbidity valueswhen assessing drinking source water quality. 3.3. Index score validation A total of 11 of the 24 panelists that completed the parameterselection survey responded to the second survey (a 46%response rate). A close association between expert and indexscoreisclear(Fig.1).ThemeanabsolutedeviationoftheCCMEWQIfrom theexpertscoreswas 7.17 Æ 1.11for treatment level1 and 7.48 Æ 1.14 for treatment level 2. In both cases, theregression line characterizing the relationship between themean expert rating and the CCME WQI score (Treatment Level1: CCME WQI ¼ 1.026(Mean expert rating) À 0.501, r 2 ¼ 0.772,  p < 0.001, df  ¼ 18; Treatment Level 2: CCME WQI ¼ 1.105(Meanexpert rating) À 8.493, r 2 ¼ 0.774, p < 0.001, df  ¼ 18) was notsignificantly different ( P < 0.05) from a theoretical 1:1 rela-tionship ( y ¼ x ). 3.4. Exploration of factor weights Mean optimized factor weights under both treatment levelsare presented inTable 3. A total of 4 scenarios under treat-ment level 1 and 5 scenarios under treatment level 2 wereexcluded due to optimized weights of near zero for 2 of the 3factors. Factor weight optimization reduced the differencebetween the Eq.(2)index value and its corresponding meanexpertscoreto withinat least8.0 Â 10 À 7 pointsforallincludedscenarios. Neither treatment level 1 nor treatment level 2optimized factor weights provided evidence for the applica-tion of unequal weights to the 3 index factors (TreatmentLevel 1 ANOVA; F (2,42) ¼ 1.449, p ¼ 0.246; Treatment Level 2ANOVA; F (2,42) ¼ 0.881, p ¼ 0.422). 3.5. Index application and sensitivity analysis Data limitations only permitted index calculations to be madeusing a total of six parameters (pH, turbidity, total organiccarbon, E. coli , nitrate and nitrite as N, and temperature).Several sites monitored dissolved organic carbon (DOC)instead of total organic carbon. DOC makes up greater than80% of TOC in natural waters (Owen et al., 1995; Larsen et al.,2011) however, to accommodate these sites, a regressionequation was computed to describe the relationship betweenDOC and TOC at 6 of the 47 sites. The regression equation wasthen used to convert DOC measures to TOC estimates.Immediate concerns became apparent in regards to thenumber of measures of each parameter. All 47 sites share theissue of an unequal number of samples among the six coreparameters analyzed. Pielou’s evenness index values rangedfrom 0.80 to 1 with a mean value of 0.97. This unequalnumber of samples results in a score bias, the magnitude anddirection of which depends on the behavior of the parametermeasures relative to their target values. Overrepresentedparameters with a high proportion and/or magnitude of failures will drive down the score while an overabundanceof a parameter with a low probability and/or magnitude of failure will artificially inflate the score. In order to reduce theeffect of deviation from parameter evenness we adopteda modified CCME WQI in which factors F 2 and F 3 werecalculated on a parameter by parameter basis before being averaged. Factor F 1 was calculated using the originalformulation. Original and modified factor formulas are pre-sented inTable 4. Of note, the relationship between waterquality scenario scores calculated using the modified CCMEWQI formulation and mean expert scores was the same asthat depicted inFig. 1. Of the 20 scenarios evaluated by theexpert panel, 19 had Pielou’s evenness index scores of 1. Thesingle scenario that strayed from perfect evenness had anevenness index score of 0.995 due to one parameter having 5measures instead of the standard 6 common to all otherparameters.Turbidity was the parameter that most frequently violatedsource water target values for both treatment levels (Fig. 2).The mean proportion of turbidity measures that violated thetreatmentlevel1targetvalueof0.3NTUwas0.94 Æ 0.01acrossall sites and seasons. The treatment level 2 target value of 5NTU reduced the mean turbidity violation rate to approxi-mately0.55 Æ 0.02 for each calculatedscore. TOC target values 01020304050607080901000 10 20 30 40 50 60 70 80 90 100    C   C   M   E   W   Q   I   S  c  o  r  e Mean Expert Score (n=11) 01020304050607080901000 10 20 30 40 50 60 70 80 90 100    C   C   M   E   W   Q   I   S  c  o  r  e Mean Expert Score (n=11) A B Fig. 1 e Mean expert scores (  n [ 11) for each of the 20 water quality scenarios plotted against calculated CCME WQI indexscores. Expert assessments were made considering two different treatment levels: A) Treatment Level 1 e chlorination onlyand B) Treatment Level 2 e slow sand filtration and chlorination. The mean expert scores for each treatment level areplotted against CCME WQI scores calculated using the corresponding treatment level target values. Error bars represent the95% confidence intervals. The continuous solid line represents a theoretical line with gradient  [ 1 corresponding toa perfect match between index and expert score. water research xxx (2012) 1 e 9 5 Please cite this article in press as: Hurley, T., et al., Adaptation and evaluation of the Canadian Council of Ministers of theEnvironment Water Quality Index (CCME WQI) for use as an effective tool to characterize drinking source water quality, WaterResearch (2012), doi:10.1016/j.watres.2012.03.061
Search
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks