Bathing water quality analysis, management and policy: an integrated assessment for Ireland

Bathing waters have a major significance for ecosystems and people who derive multiple benefits from them. Bathing sites (BS) are great assets for the local, regional and national economies and development, tourism, health and wellbeing, social and cultural benefits. Their social and cultural benefits include the pleasure, aesthetic and recreational assets, sense of ownership and pride of place for local residents and communities, sports, environmental and coastal education (Vassilopoulos & Koundouri, 2017). Preserving or improving the water quality of BS is regulated with international and national legislation, usually coordinated by central government; regional or local authorities (LAs) are responsible for the relevant monitoring, sampling, actions and measures (Skotak et al., 2012). This can be challenging for many authorities because of human, expertise or financial constraints (Globevnik et al., 2020). The integrated assessment of BS water quality with the investigation of causes and driving factors has received limited research attention compared to its considerable importance for health, ecosystems and wellbeing (Buer et al., 2018; Malcangio et al., 2018). Even rarer are the studies that integrate multiple datasets to analyze the changing status of BS to improve their management and provide policy recommendations (Plana et al., 2018; Quilliam et al., 2019). This article presents such an assessment for the BS of Ireland and contributes to their improved regulatory management. The presented approach can be generalized and is expected to be useful for many countries, especially during the consultation period for the 2006 Bathing Water Directive (BWD). The roadmap of the BWD was developed in the spring of 2021, the public consultation is from 28 October 2021 until 20 January 2022 and the European Commission (EC) adoption is planned for the first quarter of 2023. In this work, various relevant data to BS water quality were collated and integrated into a database, simple statistical analyses were carried out to investigate potential causes, and the current regulations, responsible authorities and progress are discussed to provide useful policy recommendations. The study represents an ongoing and continuous effort for integrated monitoring of the Irish BS water quality, updated after each bathing season. The context and the combination of such elements give an added value beyond the BWD, in terms of local-scale initiatives and simple analyses that can be performed to better manage BS water quality.

The EC with the BWD/2006 requires Member States to monitor and assess the bathing water for at least E. coli and intestinal Enterococci (microbiological), to inform the public about BS quality and to beach management, pollution and its possible sources (EC, 2006). Member States classify bathing water as ‘poor, sufficient, good, or excellent’ depending on the concentrations of these parameters. ‘Poor’ BS are closing for the public (for the entire season or temporarily), and the reasons for the failure to achieve ‘sufficient’ status must be identified. If a BS is classified as ‘poor’ for five consecutive years, bathing is prohibited permanently. The BWD requires a sampling shortly before the start of each bathing season, and no fewer than four samples are considered to be taken and analyzed during the bathing season (max interval between samples is one month). In the event of short-term pollution, one additional sample is required to confirm that the incident has ended, and these samples may be disregarded (replaced by a sample taken 7 days after the end of the incident). The BWD is applied by the Statutory Instrument No. 79/2008-Bathing Water Quality Regulations 2008 in Ireland (148 identified BS). LAs have the primary responsibility for the monitoring of BS, the assessment of pollution impacts and the implementation of management measures designed to minimize pollution. The Environmental Protection Agency (EPA) ensures that LAs carry out their functions under the BWD. Irish Water (IW) is the utility responsible for water supply, sewage and wastewater treatment. The water quality status of each BS based on microbiological measurements is publicly available via www.beaches.ie which summarizes measurements, annual classifications, water quality reports and other information (e.g. amenities, etc.). Information is regularly updated during each bathing season, which runs from 01/06 to 05/09. In June of 2021, the European Environment Agency (EEA) and the EC published the European bathing water quality assessment (2020 data), including reports and maps per Member State (EEA, 2021). A similar managerial structure (hierarchy, coordination and responsibilities) applies in most EU Member States. Having harmonized detection methods and the same requirements for active BS management in Europe were among the main objectives of the BWD (Szewzyk & Knobling, 2007). The importance of transparency, more accurate and frequent monitoring, and the role of LAs have been crucial points for the BWD's implementation (Dizer et al., 2005; Schernewski et al., 2012) and are internationally recognized as fundamental elements for a reasonable BS management (Torres-Bejarano et al., 2018).

Figure 1 shows all BS which received reports and warnings during the summer of 2020. The orange-colored ones were closed for the entire season (Table 2), while the white-colored ones (summarized in the Supplementary material) closed temporarily or received warnings for certain periods. From these orange-colored BS, Merrion was classified as poor for the fifth consecutive year, leading to its permanent closure (declassification), and Clifden will be declassified by the end of 2021, for the same reason. The quality status of most BS is fragile and prone to deterioration or warnings after heavy rainfall events or unexpected discharges and other contamination. The need for more systematic monitoring and tracking of the actions has been noted from the EPA (2020, 2021). The problems' causes are complex and often difficult to be identified with certainty, mostly because of insufficient monitoring. Discharges from urban land use and sewage, wastewater treatment plants (WWTPs), farming activities, overflows after heavy rainfalls, combined with poor infrastructure (pipelines, treatment, misconnections) and combinations of these factors are the usual causes (EPA, 2021).

Scholars have studied the BS quality in Ireland and provided useful insights, and potential solutions to tackle water quality issues. Table 1 provides an overview of the most relevant publications and documents, with their focus, objectives and possible implications for consideration. The international literature has studied similar problems and has tried to improve policy around management of BS: Tiwari et al. (2021) assessed and compared Bathing Water Quality Monitoring Practices in Europe and the United States. The cost burden of the BS management for small communities is relatively high, and also monitoring the quality of BS can result in significant costs for countries with extensive coastlines, so Bonamano et al. (2021) introduced a predictive index based on E. coli to facilitate monitoring, while Palazón et al. (2017) proved how numerical models could reduce the microbiological analyses during bathing seasons.

Irish studies (Table 1) reflect the country's efforts towards a better understanding of the water quality of coastal systems and ways to preserve/improve it. Most studies are examining cases around Dublin Bay, which is reasonable, given the recent urban expansion of the greater area. The importance of monitoring, modeling (and forecasting parameters' behavior) and consideration of more sampling parameters is highlighted by most studies. The research focus has been on the detailed assessment of specific factors, and there are very few cases that aimed to provide insights for regulations based on original research. In the present study, this systematic assessment is supported from a complete database for the deteriorated BS (starting from the bathing season of 2020), investigating potential contamination sources, drivers and actions, and identifying knowledge and informational gaps. Together with the policy review and recommendations, this approach can be useful for practitioners and decision-makers.

To assess the factors that deteriorate bathing water quality, the first step was to form a database with as complete information as possible to describe the situation and ‘map’ each BS (Figure 2), to facilitate the identification of the potential pressures to the BS water quality.

The potential causes for each BS's water quality status that were added to the database following the classification used by most LAs and beaches.ie, and were grouped as:

• Urban & sewage: runoff from urban land cover, runoff from roads and raw sewage discharges.

• Poor network condition: misconnections, leaks, spills and damaged wastewater collection systems.

• Agriculture: runoff from agricultural and pasture activities.

• Stormwater overflow: existing storm events that caused water quality deterioration or expected events.

The above categories are useful for targeting relevant measures (urban or rural drainage, infrastructure works, etc.). For most cases, multiple causes are combined, and it is difficult to distinguish with certainty each cause's magnitude or contribution to the overall pollution. Thus, in each BS we often have at least one entry. According to EPA and beaches.ie., the problems' outline-actions-comments were listed for each BS, collated and will be updated annually.

Following the analyses of Figure 2, the time-series of rainfall were plotted with the monitored water quality below good status, i.e., sufficient and poor status. If the water quality status changed after a rainfall peak, then it is safe to assume that the heavy rainfall caused overflows or runoff from the surrounding land uses, with a direct impact on the water quality. This relation is explored with a simple Pearson correlation coefficient, using a binary variable for the water quality measurements: Poor = 3, Sufficient = 2, Good = 1, Excellent = 0. Thus, the correlation is estimated, for certain dates-events, when the amount (and frequency) of quality measurements was adequate and allowed that estimation. This is an indicative and simplified measure, however useful for the LAs. In general, stormwater overflows (SWOs) are often reported as an issue of concern for many BS, so an extra analysis on the rainfall from their nearest meteorological stations was followed (e.g. trends for daily, monthly, seasonal and annual time steps) (Alamanos et al., 2021a).

The process described in the previous section is an ongoing work, aimed to provide an integrated database useful for monitoring the situation of each BS of interest. It is expected to be expanded and updated in the future. Table 2 is an indicative example of the main features collated in the database.

The BS that were closed for the entire bathing season of 2020 (Table 2) have historically poor water quality status over the last 3–5 years mainly because of raw sewage and poor infrastructure issues, while the contamination of the other BS (Table S.1 – supplementary material) was mostly attributed to SWOs and a combination of other causes. This is an interesting observation on the duration of the water quality problems and their causes. Figure 3 is an example of the process followed for every BS, according to the approach explained in the previous section, and the outputs of Figure 2.

The correlation table-plot of the BS' precipitation data is shown in Figure 4(a), where the station names are shown along the diagonal boxes; the bottom-left part displays a scatterplot of the relationship between each pairwise combination of stations' rainfall (a 1:1 line would be a perfect correlation) and the top-right corner of the matrix displays the Pearson correlation coefficients, as a ‘mirror’ of the bottom-left part.

As mentioned, the rainfall was further analyzed to explore the trends of extreme events, as SWOs often are reported as the main (unexpected) cause for water contamination. Here, a brief extreme analysis is presented, indicatively for the Malahide station, as it can be considered representative of the beaches of Loughshinny, Rush, Portmarnock and Merrion, while it also applies to the Greater Dublin Area (GDA), as Dublin and Dun Laoghaire stations are highly correlated (Alamanos et al., 2021a). Water quality deterioration due to SWOs (Malahide station) came after continuous rainfall events (longer than 1-day storms) or isolated events (1-day storms >10–15 mm in certain cases). Here, the 95th percentile of precipitation on wet days (>10 mm) was 16, while the total precipitation on the days with more than 15 mm rainfall was 2050.5 mm, indicating that the heavy rainfall events can actually cause significant discharges that may degrade the BS' water quality. The maximum annual values of the 5-day total precipitation (Figure 4(b)) show a clear, almost constant increasing trend, which is indicative of the increasing seriousness of the SWO issues. Even this brief analysis, and a more detailed one (Alamanos et al., 2021a), clearly indicated that addressing overflows would not be seen as a rare-occasional issue, but as an increasingly serious phenomenon. This suggestion is in line with EPA's finding on the permanent character of SWOs affecting bathing waters (EPA, 2021).

The main outcome of this process is a detailed description of the situation, the main issues, the actions and their impact so far on the BS' water quality. This is expected to enhance integrated approaches and future analyses, as well as to support the key recommendations for the improvement of each BS' quality status. Table 3 reflects this indicatively for the five BS with permanent warnings, while the findings for the rest of the BS are in the supplementary file.

Although putting this information together is a simple and basic task, it is a fundamental message for many countries' LAs towards the development of databases with as much information as possible, to support the investigation of complex pollution problems and the identification of suitable measures.

Integrated analyses such as the one presented in this work per BS, the consideration of the international and national literature, and the critical evaluation of the causal effect relations for each BS, is a recommended approach for most countries (and especially useful for the LAs). During the process of creating a single database, collating different datasets and findings from the literature, analyzing the situation, progress and providing specific recommendations for each BS, several points for consideration have occurred. These are generalized and are listed below as policy recommendations to improve the regulatory management of the bathing water quality:

• 1.

  Data transparency: It would be good to publish all data (beaches.ie), not only the compliance data to the minimum BWD requirements. These could be enriched by adding also operational/monitoring data from EPA, LAs and IW having thus more transparency:

• • a.

    Adding data about the dates of releases and discharge loads from WWTPs, to encourage IW to continue improving the operation and management of WWTPs and networks.

  • b.

    Once a BS is declassified, water quality data are still reported; however, a bi-monthly year-long sampling with genome sequencing is suggested to identify sources and ways to tackle them by LAs (i.e. septic tanks, farms and animal contamination) or IW (WWTPs). This can give more accurate evidence to EPA for identifying sources of contamination.

• 2.

  Identifying the causes – monitoring parameters and sampling frequency: The determination of pollution sources must be a combinative process, examining the surrounding land uses, the discharges from WWTPs, farming, septic tanks, animal access, etc. close to the BS.

• • a.

    The BWD/2006 requires the monitoring of at least E. coli and intestinal Enterococci; however, monitoring only these two parameters cannot cover all contamination causes. Especially in Ireland, the regulation must be modified to address other pollution sources (agriculture, animal contamination, sediments, pesticides, etc.), so more parameters need to be considered.

  • b.

    A continuous review of the monitored parameters should be carried out and published, together with long-term statistical analyses to show the variation of each cause over time (also relevant with the progress tracking of measures).

  • c.

    More frequent monitoring of water quality measurements during spring and autumn months and after rainfall events would generate a clearer picture of the system as a whole, control for seasonality and detect poor management practices. All the examined BS comply with the minimum monitoring frequency as set by the BWD; however, once a month is not enough to analyze the pollution causes and patterns. The average interval of EU's countries sampling is 13.5 days, the BWD recommends to ‘take additional unscheduled samples in anticipation of short-term pollution’, so perhaps the revised BWD must highlight such issues explicitly.

• 3.

  Short-term pollution: Heavy rainfall leading to runoff-pollution events and SWOs affected at some degree almost all the examined BS.

• • a.

    Sampling data after short-term pollution events are replaced by samples taken 7 days after the end of the incident, according to the BWD. Short-term pollution because of rainfall was proved to be very common in Ireland, so the number of samples' replacement allowed per beach needs to consider that otherwise, it gives a false impression of bathing water quality at certain beaches.

  • b.

    Although early warning systems have been proposed by researchers, there is still limited application. Studying the cause–effect relation of heavy rainfall to water quality detects the rainfall (storm) threshold that can affect a BS above the acceptable quality status. Short-term forecasts that show exceedances of this threshold could be used to warn the public in a timely manner (early warning systems).

  • c.

    SWOs have been found to be a challenging and persistent pressure that needs to be addressed, given the precipitation distribution and trends. IW knows the locations of most SWOs; however, the level of knowledge on discharge volumes, frequencies and duration needs to be further supported and analysed, considering long-term planning (e.g. sustainable urban drainage systems).

• 4.

  Capacity building: LAs require support for enhanced BS monitoring and management, scientific, technological and financial support.

• • a.

    Upgraded WWTPs can significantly improve water quality, and the SWOs must be also addressed.

  • b.

    To achieve integrated management and knowledge for each case study, catchment-scale approaches should be examined, as all EU and national legislations propose. This will facilitate the development of efficient and targeted actions measures. Integrated catchment-coastal models would be highly valuable and easily transferable, as well as ‘whole-of-environment’ approaches combined with stakeholder analyses (Alamanos et al., 2021b).

  • c.

    In case of data limitations, available data sources can be used for various useful datasets. For example, Alamanos & Linnane (2021) listed many free repositories and demonstrated how the user can access such relevant data.

  • d.

    Cooperating with academic and research institutions for monitoring, analysis, or measures actions (e.g. as in Lilliput, where University College Dublin (UCD) assists the LAs with the analysis of microbial source tracking).

  • e.

    LAs to apply measures to improve the bathing waters status, individual actions to help by bringing rubbish home, cleaning up after dogs, reporting pollution, influence others, etc.

Finally, every successful policy action requires stakeholder engagement, and to achieve that, the understanding of its purpose is essential. Cooperation among key stakeholders and the public, through the delivery of educational and informative campaigns, is crucial.

Bathing and coastal waters are valuable ecosystems with multiple environmental and economic impacts, highly appreciated by the public (Hynes et al., 2014). The effort to achieve excellent water quality for BS needs close investigation for each case, identification of the problems' key factors, and cooperation between relevant authorities, researchers and the public. The BWD provides the guidelines and general recommendations; however, each Member State could best tailor them for their needs to address specific issues.

This article presented a work in progress, a continuous enrichment of a database for the monitoring of the water quality of Irish BS. The integrated character of the database is an important element for future efforts; having more (and different) data greatly support the identification of cause–effect relations and measures. Such approaches can be useful for many countries, especially for the LAs involved.

This article, unavoidably, has limitations mainly referring to its narrowed perspective (analyzing only the case of Ireland); however, most recommendations could be considered by other countries facing similar problems. In the case of Ireland, the recommendations of the previous section support the progress of protecting and improving BS water quality: LAs are expected to start enhancing the monitoring and considering more parameters to their investigations. An important point here is to move from the ‘minimum action required to be compliant with the BWD’, to a mindset of greater control of the BS. Moreover, many LAs – not only in Ireland – need to move from the stage of investigation to actions. Such observations, supported by evidence, can be particularly useful for the consultation period of the BWD and similar regulations, regarding their monitoring standards, frequency, samplings replacement, data transparency and simple analyses that can be done to explore data interactions and capacity building.

The author thanks to both reviewers' constructive comments, and the manuscript has been significantly improved.

All relevant data are included in the paper or its Supplementary Information.