The purpose of this research was to (1) identify an existing water quality index methodology that could be used in Abu Dhabi, United Arab Emirates, and (2) apply the water quality index to Abu Dhabi’s waters to communicate water quality conditions and track changes in conditions over time. The available monitoring data were reviewed to establish the types and frequencies of data available to populate an index, and a literature search was performed to identify and review existing water quality ind in use around the world. The Canadian Council of Ministers of the Environment Water Quality Index was selected and customized for use in Abu Dhabi. Using the Canadian Council of Ministers Water Quality Index, two different data aggregation methods were investigated for developing a single Emirate-wide Water Quality Index score. Finally, an alternative method for calculating indices when there are fewer than four parameters was investigated. Three Water Quality Indexs were developed for use in Abu Dhabi: Eutrophication Index, Microbial Index, and Heavy Metals (in sediment) Index. The Canadian Council of Ministers Water Quality Index methodology was found to be relatively easy to use and flexible as a building block for tailored indices, which are used to communicate marine water quality conditions to various stakeholders, including scientists, managers, policymakers, and the public.

Introduction

Water quality indices (WQIs) are used around the world to convey information about the condition of marine or freshwaters where potential environmental degradation from development, urbanization and other changing land uses may impact water quality. A 2014 global review of WQIs found 554 published articles on the subject, describing 97 different WQIs (Alves et al., 2014). The first WQI was developed in 1965 in Germany (Lumb et al., 2011). Since that time, WQIs have evolved to incorporate new statistical and water quality understanding (Alves et al., 2014). Differences among WQIs include ways in which data are summarized (e.g. using different statistical methods or mathematical equations), and how the resulting values are interpreted (Lumb et al., 2011). No single WQI has been found to be universally suitable (Tyagi et al., 2013; Lumb et al., 2011).

WQIs are used to summarize diverse and complex data in a simple and consistent manner that is easy to interpret and provide a means to communicate the status of water quality to the target audience (Lumb et al. 2011). Although specific details can be lost during the process of summarizing data, WQIs provide the ability for a variety of audiences to understand and track water quality – a powerful tool that can help policymakers and managers put in place measures to protect these important resources (Cude, 2001). WQIs focus on aspects of water quality that have been determined to be important to a jurisdiction and incorporate perceived or known stressors. A jurisdiction may develop a WQI de novo or customize an existing WQI, depending upon the particular needs and resources of stakeholders and managers.

In 2012, marine WQIs were developed for Abu Dhabi coastal waters for use as a mechanism to communicate the condition of the Emirate’s marine waters to various stakeholders, including scientists, managers, policymakers, and the public. Abu Dhabi is one of seven emirates in the United Arab Emirates and has approximately 700 km of coastline along the Arabian Gulf (SCAD, 2015) (Fig. 1), which includes four existing and proposed Marine Protected Areas (MPAs) and over 30 recreational beaches. Continuous water quality monitoring began in 2006 to characterize conditions at coastal sites including urban and protected areas. These monitoring data were available for development of marine WQIs.

The purpose of this research was to (1) identify an existing WQI methodology that could be used for assessing marine water quality in Abu Dhabi, and (2) apply the marine WQI to Abu Dhabi’s waters to effectively communicate water quality conditions and track changes in conditions over time.

Methodology

Available monitoring data

The available monitoring data were reviewed to understand the types and frequencies of data available to populate a WQI. Water and sediment samples were collected from 18 coastal sites located throughout Abu Dhabi City and in the Western Region of Abu Dhabi Emirate (Fig. 1). Samples were collected monthly at sites near Abu Dhabi City (Sites 1-11, 13-14, and 23). Samples were collected every 3 months at sites located outside of the city, in the Western Region and marine protected areas (sites 12 and 15-19).

Temperature, salinity, pH, dissolved oxygen (DO), and chlorophyll a were measured from surface and bottom waters using a Hydrolab-DS 5. A Secchi disc was used to measure water transparency. Grab samples were collected for laboratory analysis. Parameters included nitrate, phosphate, ammonia, fecal coliforms and enterococci in water. Sediments were collected using a Peterson Grab Sampler and analyzed for metals (including cadmium, copper, nickel, lead, zinc, mercury) in the laboratory. These parameters were chosen for the routine monitoring as being reflective of possible stressors in Abu Dhabi waters such as industrial and municipal effluents, disturbance from development, and port traffic. A subset of parameters were used in the WQIs and are described below. Detailed sampling methodology, including the full list of parameters analyzed, and laboratory methods used in routine monitoring can be found in Appendix 1 (available at https://webaddress/).

Index selection

A brief literature search was performed to identify and review existing WQIs in use around the world (Appendix 2: available at http://webaddress/). Other papers (e.g. Lumb et al., 2011; Shah and Joshi, 2017; Tyagi et al., 2013; Alves et al., 2014) have more extensively reviewed WQIs and were used to inform our review. The initial review included WQIs that were currently being used and maintained by a jurisdiction, and/or that were being used in similar geographic and climatic areas as Abu Dhabi. After the initial review, relevant WQIs were further reviewed using the following criteria: publicly available, applied to water quality data (fresh or marine), and currently used by a government or independent entity to summarize and communicate water quality results. Several WQIs were selected (i.e. CCME, 2001; Brown et al., 1970; Water Research Center, 2014; Schaffelke et al., 2011; Cude, 2001; Brown, 2016; U.S. EPA, 2015) for potential application in Abu Dhabi. These WQIs were evaluated a third time using the following selection criteria: flexibility and ease of use; ability to stay relevant into the future (e.g. regularly used and maintained by an entity); and results that are easy to interpret.

The CCME WQI is a methodology that allows a user to develop a customized index using a downloadable calculator tool. The CCME WQI methodology was selected for use in Abu Dhabi Emirate because it best met the final selection criteria based on (1) its flexibility and ease of use with respect to the number and types of parameters that can be included, the time period it can accommodate, and the user-entered guideline values against which data are evaluated; (2) it is available free of use to the public and is regularly maintained and updated by the CCME; and (3) the resulting index score is easy to understand and communicate to multiple audiences.

The CCME WQI1 calculates scores based on 1) the number of parameters that do not meet their respective guideline values (i.e. the “Scope”); 2) how frequently they exceed2 the guideline values (i.e. the “Frequency”); and 3) by how much the guideline values are exceeded (i.e. the “Amplitude”) (CCME, 2001). The final index score is a number between 1 and 100. The user may assign ranges of this final score to a descriptive water quality condition category. This calculator includes only the statistical approach; thus, it can be applied to any jurisdiction, not just Canada. Any jurisdiction using the CCME calculator identifies their own parameters, guideline values, how data are aggregated (e.g. by site, by year) and, if desired, condition ranges. A detailed description of how the index is calculated is provided in Appendix 3 (available at https://webaddress/).

Application of the Canadian Council of Ministers of the Environment Water Quality Index to Abu Dhabi coastal waters

Several user-based decisions are required before applying the WQI calculator to a dataset. Users must determine which parameters to include and specific guideline values for each. The guideline value determines whether a data point “passes” or “fails” a test. And finally, users must determine how the data will be aggregated (or disaggregated) in the calculator, which could affect the final index score that is calculated and the corresponding water quality condition rating.

Identifying parameters

The CCME WQI calculator can accommodate as many water quality parameters as desired, but the more parameters included, the less meaningful the index results become due to dilution of the results. That is, including less relevant parameters may diminish the signal provided by the index. Further, including parameters that are indicative of different issues in a single index may prevent certain trends from becoming visible. The CCME guidance available at the time (CCME, 2001) indicated a minimum of four parameters is ideal, though the guidance has been updated to indicate a minimum of eight parameters is ideal, so long as they are relevant to a particular site or geographic context (CCME, 2017). The Abu Dhabi database includes more than 20 parameters in water and sediment. As a first step, the critical concerns to track and hence the relevant parameters to include were determined. Three critical concerns were identified: potential eutrophication from excess nutrients, microbial contamination from periodic releases of untreated sewage, and sediment heavy metal contamination from industrial activities. A separate WQI was created for each concern: Eutrophication Index, Microbial Index, and Heavy Metals Index. From the suite of 20 parameters routinely measured, DO, chlorophyll a, nitrate, phosphate and ammonia were identified as being most relevant to the Eutrophication Index. Fecal coliforms and enterococci were the only relevant parameters available for the Microbial Index, and lead, mercury, zinc, nickel, copper and cadmium were used for the Heavy Metals Index.

Because the Microbial Index only had two parameters, which is less than the ideal number, a modified version of the calculator was used, which omits the F1 “scope” variable. This version was created by the CCME for use in situations where fewer than four parameters are used. Working with CCME directly, we were able to accommodate this situation using a modified version of the calculator.

Identifying guideline values

After determining which parameters to include in the WQIs, guideline values were identified for each parameter in the calculator. Typically, guideline values can be taken from a jurisdiction’s water quality standards, guidelines, or limits. However, at the time, ambient marine water quality limits for the Emirate had not been established. Further, little research existed particularly related to effects of changes in water quality in Abu Dhabi. This limited knowledge base dictated that guidelines from other jurisdictions were reviewed (see Wael Suleiman paper this issue). The review included 1) the context of the general geological and physical characteristics of the local environment; 2) whether the guidelines were science-based and peer reviewed; and 3) the basis for the guideline (e.g. to protect aquatic health or public health). For guidelines relevant to Abu Dhabi, factors considered included whether the jurisdiction had characteristics such as oligotrophic marine waters with high salinity and temperature, and carbonate sediments. Guideline values were identified for each parameter included in each of the three WQIs: Eutrophication, Microbial, and Heavy Metals.

Calculation approaches

Because the WQI calculator is flexible and can accommodate different ways of calculating the index scores, users are able to test how different aggregation of data affects the index score results. For example, calculating a score for each sampling site produces a different picture of overall water quality than does calculating a single score for the entire Emirate. Different audiences may require producing both types of scores: site-specific and Emirate-wide. However, the more data are aggregated within the calculator – e.g. to produce an Emirate-wide score – the less meaningful the score becomes because geographic differences are masked. Two different aggregation methods were investigated for developing a single Emirate-wide score for each of the 3 WQIs: the “non-averaging” method and the “averaging” method. The non-averaging method involved using the index calculator itself as a tool to produce a single score for each index by aggregating the data across all sites for the full year into a given index calculation. The averaging method involved the user taking the average of all the site-specific index scores for the year across the entire Emirate to produce a mean Emirate index score.

Communication of index

Using the CCME index calculator, a single score between 1 and 100 is calculated for each desired set of data (e.g. each monitoring station each year), and the ranges of these scores may be gathered into a condition category by the user. The condition categories for the Abu Dhabi WQIs used a familiar traffic light color scheme where red corresponds to “poor”, yellow to “fair”, and green to “good” water quality condition (Table 1), which is generally easily understood by stakeholders.

Results and discussion

Development of the Environment Agency Abu Dhabi indices

The CCME WQI framework and calculations were used to calculate the Abu Dhabi WQIs. Three indices relevant to the Emirate were developed: the Eutrophication Index, which tracks issues related to increased nutrients from population and growth-related contamination; the Microbial Index, which tracks contaminants related to protection of public health during recreational water use; and the Heavy Metals Index, which tracks contaminants in marine sediments and is indicative of longer-term contamination trends.

Eutrophication

The Eutrophication Index was designed to identify areas of the coast where nutrient over-enrichment may lead to excessive planktonic algal growth and possible harmful algal blooms (Boesch, 2019; Glibert and Burkholder, 2006). This index is based on five parameters that are indicative of nutrient over-enrichment: nitrate - N, phosphate - P, ammonia, DO, and chlorophyll a (U.S. EPA, 2001). The guideline values for each of these five parameters included in the Eutrophication Index were developed based on historical monitoring data from Abu Dhabi Emirate, including data collected from the pristine MPAs; and consideration of international standards from countries with sedimentary, oceanographic, and climatic conditions similar to Abu Dhabi (see Wael Suleiman paper this issue) (Appendix 4, Table 1: available at https://webaddress/).

Microbial

The Microbial Index was designed to identify coastal areas where bacterial contamination may be high enough to cause an increased risk of illness to swimmers through direct contact from recreational use and includes two indicators of fecal contamination for marine waters: enterococci and fecal coliforms. These two parameters are used worldwide as indicators of fecal contamination because their presence in sufficient concentrations in marine waters has been tied to increases in human illnesses from contact with the contaminated water, including gastroenteritis, respiratory illness, and eye, ear, and skin infections (Cabelli, 1983; Fleisher et al., 1996). The human health-based guideline values for these parameters were adopted from the United States Environmental Protection Agency (U.S. EPA, 2012a) and the World Health Organization (WHO, 2003) (Appendix 4, Table 2: available at https://webaddress/).

Heavy metals

The Heavy Metals Index was designed to indicate the level of contamination in sediments based on concentrations of six heavy metals: cadmium, copper, nickel, lead, zinc, and mercury. These metals were chosen because they are potentially more toxic than other metals (e.g. manganese, iron) (ATSDR, 2016). Although some metals occur naturally in marine sediments, high concentrations of metals above the guideline value or the presence of toxic metals such as mercury and cadmium can indicate anthropogenic sources of contamination at specific locations. Guideline values for each metal were developed based on a review of international best practices for sediment quality criteria and a review of historical data collected by EAD from the MPAs in the Western Region of Abu Dhabi (Appendix 4, Table 3: available at https://webaddress/).

Data aggregation

In addition to communicating results about a specific monitoring site, the indices must be able to summarize and communicate the general water quality condition across the entire Emirate. Two methods of calculating the aggregated WQI scores for the Emirate across all sites were investigated: the non-averaging method and the averaging method. With the non-averaging method, the index is calculated using raw data from all the sites at each sampling time for the whole year. Using the averaging method, the index score is calculated for each site for the year and then the resulting scores are averaged across sites. These two approaches were compared to the scores from each site as well as the raw data to ascertain which approach better represented observations. The illustrative results for the Heavy Metals Index using data from years 1-4 were compared using both scoring methods (Fig. 2).

The non-averaging method consistently produced lower index scores than the averaging method. In each year for the Heavy Metals Index, the scores produced by the different methods fell into two different categories of water quality condition.

In Fig. 3, we further explored the difference between the two aggregation approaches shown in Fig. 2. Fig. 3 shows the number of sites scoring within each condition category (i.e. good, fair, poor) each year for the Heavy Metals Index. The results in Fig. 2 show that using the non-averaging method to produce overall Emirate scores resulted in an overall Heavy Metal Index score in the “Fair” condition category all four years, whereas the averaging method resulted in overall Heavy Metals Index scores in the “Good” condition category. The averaging method of aggregating sites agrees better with the site-specific results in Fig. 3, showing most sites scored in the “Good” condition category all four years.

The non-averaging method likely produces lower overall Emirate index scores than the averaging method because as more sites are aggregated, the frequency of failed tests (F2 score) increases, resulting in a lower overall WQI score, even though at any given site perhaps only a few tests failed and the water quality at each site individually is better than the aggregated score shows. As a result, the averaging method was found to better agree with the site-specific data and therefore was used to produce the overall Emirate index scores for Abu Dhabi Emirate.

An additional analysis was performed to investigate how the calculator would handle datasets with fewer than the recommended four parameters (Appendix 5, available at https://webaddress/).

Conclusions

The CCME WQI methodology can easily be applied to a variety of uses to track trends, identify water quality problems, and communicate to stakeholders. Users must take a few steps before applying the WQI calculator to water quality datasets, including the following:

  • Determine the purpose of the WQI(s) and which parameters are relevant to include for each one

  • Identify guideline values for each parameter included

  • Determine at what level of aggregation the index scores should be calculated, depending on the purpose and the intended audience, and keeping in mind the limitations of aggregation.

The index development process and associated analyses performed here shed some light on the strengths and limitations of the CCME WQI methodology. Key findings from this process include the following:

  • The CCME WQI methodology is flexible and allows users to choose the parameters, guideline values, level of data aggregation, and period of time over which the index is calculated. The calculator is also regularly maintained and updated by the CCME.

  • Data aggregated in the calculator by site show how water quality varies across the Emirate while data aggregated across sites present a single value intended to represent the overall Emirate, thereby masking any localized geographical differences in water quality. These differently aggregated WQI scores may be useful for conveying information to different audiences (e.g. policy makers, general public, scientists).

  • Overall water quality using Emirate-wide WQI scores is also dependent on how data are aggregated. The averaging method, in which users take the mean of the site-specific WQI scores to produce Emirate-wide WQI scores, agrees better with the site-specific WQI scores and raw data than the non-averaging method, which combines raw data across all sites for the year in the calculator to produce Emirate-wide scores.

  • When fewer than four parameters are available to include in the WQI, the modified version of the calculator, which omits the F1 variable (“Scope”), produces index scores that agree better with the raw data used to calculate the scores (discussed in Appendix 5, available at https://webaddress/).

The three WQIs - Eutrophication Index, Microbial Index, and Heavy Metals Index – now allows Abu Dhabi to track the impacts of pollution abatement initiatives on the state of marine water quality over time in a consistent and understandable way. The CCME WQI methodology was found to be relatively easy to use and flexible as a building block for its tailored WQIs, which are used to communicate marine water quality conditions to various stakeholders, including scientists, managers, policymakers, and the public.

Acknowledgements and funding

The authors gratefully acknowledge the financial support of the Environment Agency—Abu Dhabi (EAD).

Supplemental material

Supplementary material for this article can be accessed on-line at the publisher’s website.

Notes

1

The downloadable index calculator, a technical guidance document, and a user’s manual are available at the CCME’s Web site at http://www.ccme.ca/ourwork/water.html?category_id=102.

2

The calculator accommodates both guidelines that are not to be exceeded (e.g. metals, toxins, bacteria) as well as those that should be exceeded (e.g. dissolved oxygen) or should fall within a range (e.g. pH).

References

Alves, M., Teresa, F. and Nabout, J.,
2014
.
A global scientific literature of research on water quality indices: Trends, biases and future directions
.
Acta Limnologica Brasiliensia
;
26
(
3
),
245
53
. doi:
ATSDR (Agency for Toxic Substances and Disease Registry)
,
2016
.
Priority List of Hazardous Substances
. Available at: https://www.atsdr.cdc.gov/spl/
Boesch, D.,
2019
.
Barriers and bridges in abating coastal eutrophication
.
Front. Mar. Sci
.;
6
doi:
Brown, D.,
2016
. Oregon water quality index: Background, analysis and usage. Laboratory and Environmental Assessment Program, Hillsboro, OR. Available at https://www.oregon.gov/deq/FilterDocs/wqmreportingmethodsF.pdf [accessed 20 July 2018].
Brown, R.M., McLelland, M.I., Deininger, R.A., and Tozer, R.G.,
1970
.
A water quality index – do we dare?
Water & Sewage Works
, Oct,
339
343
.
Cabelli, V.,
1983
. Health effects criteria for marine recreational waters. EPA 600/1-84-004. U.S. Environmental Protection Agency.
CCME (Canadian Council of Ministers of the Environment)
,
2001
.
CCME
Water Quality Index 1.0 Technical Report. Available at http://www.ccme.ca/files/Resources/calculators/WQI%20Technical%20Report%20(en).pdf [accessed 15 Mar 2017].
CCME (Canadian Council of Ministers of the Environment)
,
2017
.
CCME
Water Quality Index User’s Manual 2017 Update. Available at https://www.ccme.ca/files/Resources/water/water_quality/WQI%20Manual%20EN.pdf [accessed 03 Jan 2020].
Cude, C.,
2001
.
Oregon water quality index: A tool for evaluating water quality management effectiveness
.
J Am Water Res Assoc;
37
(
1
),
125
37
. doi:
Fleisher, J.M., Kay, D., Salmon, R.L., Jones, F., Wyer, M.D., and Godfree, A.F.,
1996
. Marine waters contaminated with domestic sewage: nonenteric illnesses associated with bather exposure in the United Kingdom.
Glibert, P.M., Burkholder J.M.,
2006
. The Complex Relationships Between Increases in Fertilization of the Earth, Coastal Eutrophication and Proliferation of Harmful Algal Blooms. In: Granéli E., Turner J.T. (Eds)
Ecology of Harmful Algae. Ecological Studies (Analysis and Synthesis)
, vol
189
.
Springer
,
Berlin, Heidelberg
Lumb, A., Sharma, T.C., Bibeault, J.F.,
2011
.
A review of genesis and evolution of water quality index (WQI) and some future directions
.
Water Qual Expo Health
;
3
,
11
24
. doi:
Schaffelke, B., Carleton, J., Doyle, J., Furnas, M., Gunn, K., Skuza, M., Wright, M., Zagorskis, I.,
2011
. Reef Rescue Marine Monitoring Program – Final Report of AIMS Activities 2010/2011 Inshore Water Quality Monitoring. Available at http://elibrary.gbrmpa.gov.au/jspui/bitstream/11017/2880/1/MMP_Inshore_Water_Quality_Monitoring_Report_2010_11.pdf [accessed 15 Mar 2017].
Shah, K. and Joshi, G.,
2017
.
Evaluation of water quality index for River Sabarmati, Gujarat, India
.
Appl Water Sci
7
,
1349
58
. doi:
Statistics Centre Abu Dhabi (SCAD),
2015
. Explore Abu Dhabi Through Statistics, 2015. Abu Dhabi, UAE. Available at https://www.scad.ae/Release%20Documents/Explore%20Abu%20Dhabi-english.pdf [accessed 20 July 2018].
Suleiman, W., Anbiah, R., Al Raisi, A., Al Hosani, S., Barber, M., Whaley, G.,
2020
.
Development of ambient marine standards using long-term data: An example from Abu Dhabi
.
Aquat. Ecosyst. Health Mgmt
.
23
(
2
),
Tyagi, S., Sharma, B., Singh, P. and Dobhal, R.,
2013
.
Water quality assessment in terms of water quality index
.
Am J Water Res
1
(
3
),
34
38
.
U.S. EPA (United States Environmental Protection Agency)
,
2001
. Nutrient Criteria Technical Guidance Manual: Estuarine and Coastal Marine Waters. Office of Water, Washington, DC. Report available at https://www.epa.gov/sites/production/files/2018-10/documents/nutrient-criteria-manual-estuarine-coastal.pdf [accessed 08 Jan 2020].
U.S. EPA (United States Environmental Protection Agency)
,
2012a
. Recreational Water Quality Criteria: 2012 Recreational Water Quality Criteria. (U.S. EPA online information). Web site: http://water.epa.gov/scitech/swguidance/standards/criteria/health/recreation/index.cfm. (accessed 17 Sept 2015).
U.S. EPA (United States Environmental Protection Agency)
,
2015
. National Coastal Condition Assessment 2010. Office of Water and Office of Research and Development, Washington, DC. Report available at https://www.epa.gov/sites/production/files/2016-01/documents/ncca_2010_report.pdf [accessed 20 July 2018].
WHO (World Health Organization)
,
2003
. Guidelines for Safe Recreational Water Environments. Volume 1: Coastal and Fresh Waters. ISBN: 92 4 154580 1. (WHO online information). Web site: http://www.who.int/water_sanitation_health/bathing/srwe1/en/. (accessed 17 Sept 2015).

Supplementary data