Public awareness, behaviours and attitudes towards domestic wastewater treatment systems in the Republic of Ireland

Journal of Hydrology xxx (2013) xxx–xxx

Contents lists available at ScienceDirect

Journal of Hydrology journal homepage: www.elsevier.com/locate/jhydrol

Public awareness, behaviours and attitudes towards domestic wastewater treatment systems in the Republic of Ireland O. Naughton a,⇑, P.D. Hynds b a b

Department of Civil, Structural and Environmental Engineering, Trinity College Dublin, Dublin 2, Ireland Public Health Agency of Canada, 255 Woodlawn Road West, Unit 120 Guelph, Ontario, Canada

a r t i c l e

i n f o

Article history: Available online xxxx This manuscript was handled by G. Syme, Editor-in-Chief Keywords: Socio-hydrology Domestic wastewater treatment Water management Public health

s u m m a r y Numerous studies have highlighted and quantified the role of domestic wastewater treatment systems (DWWTSs) as significant sources of human-specific aquatic contaminants in both developed and developing regions, particularly with respect to private and municipal groundwater supplies. However, from a socio-hydrological perspective, little work has focused on these systems and the potential environmental and human burden posed. This is of particular relevance in the Republic of Ireland, where approximately one third of the population is serviced by DWWTSs. The objective of the current study was to examine levels of awareness and subsequent behavioural tendencies among owners and users of DWWTSs in the Republic of Ireland, particularly in light of recent and future (national and EU) legislative amendments. Structured questionnaires were completed bi-modally with 1106 Irish respondents. Analysis identified a number of significant knowledge gaps which currently exist among DWWTS users in Ireland. These were associated with environmentally inadvisable behavioural practises, potentially leading to increased contamination vulnerability and subsequently, increased human exposure to waterborne contaminants. Household water supply type was significantly associated with DWWTS threat acknowledgement (p = 0.014), with unregulated private groundwater users exhibited the lowest awareness of DWWTS as a potential source of aquatic contaminants despite being the group at greatest risk. A bi-modal clustering approach was employed, with respondents found to fall into one of three distinct ‘‘attitudinal’’ clusters. Future engagement strategies should strive to provide guidance regarding the role of people and their activities within the hydrological cycle. The current study reinforces this conclusion, while providing evidence-based recommendations regarding provision of demographically focused educational strategies; these will further increase environmental policy compliance, and in so doing, decrease the human health and environmental contamination burden posed by DWWTSs. Ó 2013 Elsevier B.V. All rights reserved.

1. Introduction The ability of water systems to meet changing human and environmental needs lies at the core of water security in both developed and developing regions. It is recognised that some of the most critical vulnerabilities in contemporary water systems lie at the intersection between human activities and physical systems (Wheater and Gober, 2011; Sivapalan et al., 2012). The disposal of domestic effluent on-site using domestic wastewater treatment system (DWWTS) is an example of such an interaction. Improperly installed or poorly maintained DWWTSs represent a significant source of nutrients and enteric pathogens, particularly in rural areas; the protection of groundwater resources from such contamination is imperative for the effective management of risks posed to both human health and the environment. Discharge of insuffi⇑ Corresponding author. Tel.: +353 1 8962343; fax: +353 1 677 3072. E-mail address: [email protected] (O. Naughton).

ciently-treated domestic wastewater effluent to aquatic receptors, via direct discharge or baseflow, may lead to excess nutrient enrichment, algal blooms and eutrophication (Gill et al., 2009; Palmer-Felgate et al., 2010: Withers et al., 2011, 2012). It can also lead to waterborne disease; numerous significant outbreaks have been attributed to DWWTS effluent ingress to drinking water sources (Birkhead et al., 1989; Kramer et al., 1996; Karanis et al., 2007; Borchardt et al., 2011). Development in rural areas of the Republic of Ireland has historically followed a dispersed settlement pattern comprising detached private residential dwellings or ‘‘one-off’’ housing (Scott and Murray, 2009). ‘‘One-off’’ dwellings are defined as ‘‘detached dwellings with individual on-site wastewater treatment systems, including septic tanks and other systems’’ (CSO, 2012). Beginning in the mid-1990s, Ireland experienced a period of rapid economic growth, resulting in a dramatic increase in residential construction (Scott et al., 2007). During this period, approximately one in three newly constructed houses were one-off residential dwellings,

0022-1694/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jhydrol.2013.08.049

Please cite this article in press as: Naughton, O., Hynds, P.D. Public awareness, behaviours and attitudes towards domestic wastewater treatment systems in the Republic of Ireland. J. Hydrol. (2013), http://dx.doi.org/10.1016/j.jhydrol.2013.08.049

2

O. Naughton, P.D. Hynds / Journal of Hydrology xxx (2013) xxx–xxx

driven by a cultural predisposition towards rural living and lower development costs, in concurrence with improved infrastructural networks and access to urban centres (Duffy, 2000; Spatial Planning Unit, 2001; Scott and Murray, 2009). While there has been a marked decrease in housing stock growth in recent years, the trend of dispersed single dwellings has continued with one-off houses accounting for one in four of all private residences constructed in the Republic of Ireland since 2006 (CSO, 2012). The scattered settlement pattern has led to the coevolution of rural inhabitants as both a source and receptor of groundwater contamination in the Republic of Ireland, with rural domestic households playing a greater role in the water quality dynamics of the hydrological cycle. The dispersed nature of rural development largely prohibits the use of municipal sewerage schemes, causing an increased reliance on hydrological systems for the disposal and treatment of domestic wastewater. Consequently, treatment and disposal of domestic wastewater is carried out on-site in over 80% of households in rural areas (approximately 438,000), servicing over one third of the population (CSO, 2012). The factors which make sewerage schemes unfeasible in rural areas also commonly apply to treated water supply networks, and so there has been a parallel increase in the use of untreated groundwater as a source of domestic drinking water. Groundwater sources in the Republic of Ireland currently provide a daily drinking water supply to an estimated 26% of the national population, with private groundwater sources supplying approximately 720,000 people or 17% of the population (EPA, 2009; CSO, 2012). The most common form of DWWTS in Ireland consists of a septic tank providing primary treatment, followed by secondary and tertiary treatment via a subsoil percolation system (EPA, 2009). Septic tanks use primarily physical processes to remove suspended sediments and permit limited anaerobic digestion of organic materials (Gill et al., 2009). The percolation area, or soil infiltration system (EN 12566), consists of trenches with slotted pipes and gravel aggregates, receiving wastewater from the septic tank (or other system), subsequently transmitting it to subsoil strata for final treatment and disposal (Beal et al., 2005; EPA, 2009). Where DWWTSs have been sited, designed, installed and maintained in accordance with best practice, they have been shown to provide adequate treatment of domestic wastewater (USEPA, 2002; EPA, 2013). However, poorly maintained systems are potential sources of groundwater contaminants, particularly pathogenic microorganisms, and consequently, waterborne disease outbreaks (Beller et al., 1997; Daly, 2003; Borchardt et al., 2011). Scandura and Sobsey (1997) and Ahmed et al. (2005) have previously reported proximity to domestic septic tank effluent percolation areas as being a cause of groundwater contamination, particularly in areas of high groundwater vulnerability (i.e. thin overlying strata of high permeability subsoils). Borchardt et al. (2011) report that 229 patrons of a new restaurant in northeast Wisconsin were affected during a waterborne norovirus outbreak, which was hydrologically and epidemiologically confirmed as being due to ingress of septic tank effluent to an adjacent private well. Hynds et al. (2012) report that septic tank system location was significantly associated with Escherichia coli presence (p = 0.004) in private wells in diverse hydrogeological settings in the Republic of Ireland. This study identified DWWTS as the hazard source of greatest significance with respect to groundwater contamination in rural areas, particularly during periods of increased antecedent rainfall. As these systems are wholly anthropogenic in nature, this is considered a socio-hydrological issue as they represent a coupled humanwater system whereby human activity and behaviour adversely affects water cycle dynamics (Sivapalan et al., 2012). Private water wells in the Republic of Ireland are primarily located in rural areas lacking municipal sewerage schemes; accordingly, these supplies are situated in areas in which DWWTS are ubiquitously dispersed. Based on data relating to verotoxigenic

E. coli (VTEC) outbreaks in Ireland during 2008, Garvey et al. (2010) suggest that drinking water in households using private wells were a significant factor in an overall outbreak increase. More recently, a surge in VTEC notifications during the period September to November 2011 has been reported in Ireland, with private well usage again highlighted as a significant source (McKeown and Garvey, 2011). During the reporting period 2011, approximately 37% of VTEC notifications were associated with private well exposure (HPSC, 2013). Thus, exposure to private wells represented a significantly increased likelihood of contracting VTEC infection. Ireland has for some years reported the highest levels of VTEC infection in the European Union, with Crude Incidence Rates (CIRs) of 6.2/100,000 in 2011, compared with 0.8/ 100,000 within the EU as a whole (HPSC, 2013). While the high reported levels in Ireland may be a result of differing epidemiological definitions, diagnostic procedures and communicable notification status throughout EU member states, the concurrently high dependency on private groundwater and DWWTSs in Ireland is also likely a significant contributory factor. Recently the issue of DWWTS has come into sharp focus following a ruling by the European Court of Justice (C-188/08, 2009) in October 2009, finding the State (Republic of Ireland) in breach of EU legislation with respect to the regulation of septic tanks and similar systems. Consequently, the Department of the Environment and the Environmental Protection Agency (EPA) have developed and instigated a registration and inspection scheme for DWWTS, as called for under the recent amendment to the Water Services (Amendment) Act 2012 (2/2012). Owners of DWWTS have a duty of care under the Act to ensure appropriate operation and maintenance of their on-site treatment system, and thus not constitute a risk to human health or the environment; the registration and inspection scheme therefore stipulates that DWWTS owners must ‘‘self-inspect’’ their system annually (DoECLG, 2012). However, there is a lack of quantitative information regarding consumer understanding of treatment system processes, maintenance requirements and potential remediation strategies, and so a high level of uncertainty existed pertaining to consumers possessing the necessary tools to adequately inspect and maintain DWWTSs. Previous studies have reported on some of the common inaccurate public perceptions towards DWWTSs, such as the notion that on-site treatment systems are self-maintaining and that septic tank solids need not be regularly emptied or soak-aways inspected for signs of failure (Butler and Payne, 1995; Moelants et al., 2008). The study outlined in this paper for the first time quantifies the levels of awareness amongst Irish DWWTS users on these issues, in the context of adequate system self-inspection as required under national legislation. A key element of the registration and inspection plan is a public engagement strategy to promote best practice relating to the operation and maintenance of DWWTS (EPA, 2013). The objective of the public engagement strategy is to encourage societal changes which will have a positive impact on water quality. A successful public information campaign requires a clear understanding of current public perceptions, activities, and knowledge gaps in relation to DWWTS. The current study examines levels of awareness, attitudes and behaviour among septic tank users in the Republic of Ireland. Subject areas examined include scheduled guidance measures, environmental and public health threats posed by DWWTSs and system design, operation and maintenance. Trends in public perception and behaviour towards the implementation of environmental policy and legislation, both national and international, have been captured and qualified from a socio-hydrological perspective. Results of this study are being used by myriad shareholders including local authorities and governmental agencies to effectively engage with the public prior to and during implementation processes via incorporation of values and perspectives relating to the meaning, value and use of water and wastewater. The study is among the first

Please cite this article in press as: Naughton, O., Hynds, P.D. Public awareness, behaviours and attitudes towards domestic wastewater treatment systems in the Republic of Ireland. J. Hydrol. (2013), http://dx.doi.org/10.1016/j.jhydrol.2013.08.049

O. Naughton, P.D. Hynds / Journal of Hydrology xxx (2013) xxx–xxx

to investigate societal responses to DWWTSs as a primary water stressor and examine the potential of existing and new policy instruments to aid strategic development and integration of the social science contribution to hydrological theory and practice. 2. Materials and methods 2.1. Survey compilation The study was conducted in the Republic of Ireland, which has an area of 70,273 km2. The current population is approximately 4.59 million, of which 38% (1.74 million) reside in categorically rural areas (CSO, 2012).Questions were devised to investigate both current awareness and attitudes among householders in relation to contamination of aquatic ecosystems by DWWTSs, and impending policy strategies which may be implemented to improve awareness and legislative compliance. Furthermore, underlying reasons for current awareness and perceptions were investigated. The developed questionnaire comprised 28 questions, ten of which related to the survey respondent and/or their DWWTS. The remaining eighteen questions referred to the respondents own opinions/ perceptions towards a number of related issues including DWWTS maintenance, operational information, regulation/legislation, remedial works and environmental threats. Surveys were kept as brief as possible, with dichotomous/nominal multiple choice style questions used extensively to minimise completion time and subsequently, improve overall data quality. 2.2. Questionnaire design The questionnaire was designed as a ‘‘formal standardised questionnaire’’, as the overall objective was the collection of quantitative data for subsequent statistical analysis (Rennie and Crosby, 2001; Cohen et al., 2004). In keeping with current best practice, a comprehensive literature review of previous similar studies was undertaken, in order to assess likely behavioural and/or perception-based relationships and highlight potential knowledge gaps among respondents (Cohen et al., 2004). Questionnaires included both dichotomous and nominal forced choice, and scaled/ranking questions. Open-ended questions were not employed to minimise completion time and increase response comparability among cases. The questionnaire was designed with four sections, as follows: Section 1: Personal information and DWWTS details  Demographic questions (age, gender, residential ownership; household size).  Residential DWWTS details (type, location, age, designer).  Residential drinking water source. Section 2: DWWTS inspection and maintenance  Operational/maintenance information presence and source.  Maintenance history.

3

2.3. Survey completion Surveying took place over a 5 month period from August to December 2012. Two survey completion modalities were employed during the study; student-based group surveys and residential web surveys. These methods were chosen as the primary modes of data collection as it was postulated that this would elicit a higher response rate than other methods (such as telephone interviews or postal surveys), while also maximising spatial and demographic representivity within the primarily rural target population. A number of previous studies (Simpson and Hodgins, 2002; Parvez et al., 2006; Celik and Muhammetoglu, 2008; Turbow et al., 2008; Fleming and Bowden, 2009; Purcell and Magette, 2010; Hynds et al., in press) also successfully employing these completion methods. For example, Rhodes et al. (2003) noted that web-based surveys negate the need for interviewer training, allow for rapid access to numerous potential respondents and reduced research costs. A small-scale pilot study (35 respondents) was carried out at the beginning of the research in order to refine both the questionnaire and survey completion methods. Pilot study participants were interviewed with respect to question structure, order and clarity, potential bias, survey length (time) and overall questionnaire structure, in order to refine both the questionnaire and survey completion method (e.g. fatigue effects due to questionnaire length) (Cape, 2010). These data have been excluded from analysis, in keeping with current best practice (Peat et al., 2002). Questionnaire surveys were completed by 1106 private groundwater users, providing the primary data source for analysis. Completion took 12–17 min per respondent, less than the 20 min maximum recommended by Cape (2010) for ensuring response quality. No monetary/financial incentive was offered to respondents. 2.3.1. Group surveys Post-secondary agricultural/horticultural colleges were approached as potential locations for group survey completions, as it was postulated that the majority of attending students would be from rural areas and therefore served by a DWWTS, and also would provide a means of gathering information on attitudes and perceptions among the future custodians of DWWTSs. Overall, four colleges in counties Cork, Galway, Tipperary and Kilkenny agreed to allow group surveys of their students to be undertaken. This took place during the autumn/winter term (October–December 2012) in classes in which >90% of pupils were aged 18–25 years. The objectives of the study were explained to the pupil groups by the researchers on the survey day, and pupils were given the choice of participation in addition to being assured of confidentiality. A similar group-survey approach has previously been successfully employed by Hawton et al. (2002). The sample population was defined as those members of the Irish population, 18 years or over, whose primary domestic residence is served by a DWWTS. Due to the relatively large geographical study area, respondents represent a diverse population of different groundwater source types, hydrogeological units (aquifers) and socio-economic groups.

Section 3: DWWTS registration and remediation  Attitudes towards DWWTS regulation (motives, intent to register, fees).  Financial relief/grants. Section 4: Environmental contamination  Contamination awareness.  Previous contamination events (including contaminant sources).

2.3.2. Web-based surveys An identical internet-based (or ‘‘web’’) survey was initiated and distributed among the rural population via a number of national non-professional interest groups. It was considered that these respondents were likely of an older average demographic, with a significantly higher proportion of female respondents, therefore redressing any potential imbalance/bias resulting from institutional surveying. Repeat response/multiple counting was not permitted, through collation of respondent ID, IP address and survey completion time and date. The web-based survey was hosted on

Please cite this article in press as: Naughton, O., Hynds, P.D. Public awareness, behaviours and attitudes towards domestic wastewater treatment systems in the Republic of Ireland. J. Hydrol. (2013), http://dx.doi.org/10.1016/j.jhydrol.2013.08.049

4

O. Naughton, P.D. Hynds / Journal of Hydrology xxx (2013) xxx–xxx

a website (www.surveymonkey.com), with separate collation links developed for all online survey groups. This approach has previously been employed by Purcell and Magette (2010). Survey modality was statistically evaluated (Section 3.4) to examine whether completion mode affected response patterns. Modality was not found to be a potential source of data bias or response pattern development; therefore, responses to the institutional and web-based surveys were aggregated for study analysis. 2.4. Statistical analysis Responses were categorised, numerically coded and analysed using the Statistical Product and Service Solutions package IBM SPSS Statistics 20 (IBM,2011), including a number of developed pseudo-variables. Pearson chi-square tests of independence were used to determine significance of association between categorical (dichotomous and nominal) variable pairs. Independent samples t-tests (difference of means with equal variance assumed) were used to explore relationships between categorical and scale variables. Non-parametric Mann–Whitney U tests were used to investigate relationships between dichotomous and ordinal (ranked) variables. One-way analysis of variance (ANOVA) was used to determine and quantify associations between continuous and categorical variables, with Bonferroni post hoc multiple comparison analysis employed upon rejection of the global null hypothesis H0. A p-value < 0.05 was used to evaluate all categorical analyses (Blalock, 1960; Agresti, 1996). Agglomerative hierarchical clustering (between groups linkage; squared Euclidean distance method) was used to optimise cluster number, with k-means clustering (iteration and classification) subsequently used to qualify cluster membership. The potential effect of survey modality was investigated using a series of t-tests (not assuming equal variance based upon Levene tests), with the non-parametric Mann–Whitney U test used to examine if modal responses were derived from similar distributions. A similar ‘‘modality investigation’’ has previously been successfully utilised by Purcell and Magette (2010), with similar response numbers. Sample size representivity was calculated based upon the number of households in the Republic of Ireland currently employing a DWWTS, as opposed to ‘‘consumer number’’ (i.e. number of persons served by a DWWTS). Based on the most recent census of the Republic of Ireland (2011), a total of 438,000 households currently use a DWWTS for domestic waste treatment (CSO, 2012). Based on a 95% level of confidence, total sample size of 1106 households and total population of 438,000, the study sample represents a 2.94% confidence interval. There is currently no data available with respect to DWWTS owner demographic strata.

homeowners; home ownership was associated with increasing respondent age (t(1003) = 38.12, p < 0.001). The average respondent reported household size was 4.1 persons/household; comparable with recent Irish census data for rural households (CSO, 2012). Typically, households comprised 3 adults between 18 and 65 years (mean 2.91), with one ‘‘vulnerable resident’’ (0–12 years or >65 years) per household (mean 1.2). Based upon current engineering design criteria (EPA, 2009), this mean household size equates to approximately 615 l/day of effluent for treatment, or 225 m3/annum.

3.1.1. DWWTS details Overall, 91.6% of household representatives (n = 1012) exhibited an awareness of the on-site location of their DWWTS, with homeowners almost three times more likely to be aware of system location (OR = 2.853; v2(1) = 21.016, p < 0.001). Most respondents (95.8%, n = 1048) reported some knowledge of their DWWTS type. Although no gender bias was noted with respect to DWWTS type, a significant trend was noted regarding respondent age and system type awareness (F(6) = 11.606, p < 0.001), with younger respondents exhibiting lower awareness of system type. Traditional septic tank systems were reported by the majority of respondents (87.3%); however 41.8% were unaware of associated secondary or tertiary treatment systems. Septic tank systems with associated percolation area and soak-away were reported by 29.5% and 16% of respondents, respectively. The overall mean DWWTS age was 23.05 years (Std. Dev. 16.33 years). DWWTS system ages ranged from 5 years for reed-bed systems, to 26.9 years for septic tanks (Fig. 1). Septic tanks with associated percolation systems had a mean age of 19 years, while systems including lower specification soak-away systems had a mean age of 28.6 years. The most prevalently reported system designer was a building contractor (32.5%), indicating that a large proportion of DWWTS are installed during construction of the associated dwelling; whilst least prevalent designers were the system manufacturer (7.8%). Overall, 23.4% (n = 261) of respondents were unaware of the system designer. System designer awareness was not statistically associated with either respondent gender or age.

3.1. Respondent and DWWTS information

3.1.2. Discharges to DWWTS Respondents were subsequently asked to report on the nature of discharge to their DWWTS (i.e. wastewater, grey water, rainwater). In all, 94.4%, 66.5% and 15% of respondents reported wastewater, grey water and rainwater/surface drainage entering their system, respectively. No statistical association was found between surface drainage ingress and system type. Discharge of surface drainage to DWWTS was significantly associated with the system designer (excl. ‘‘don’t know’’) (v2(6) = 25.75, p = 0.001); 5.8% of manufacturer designed systems allowed drainage discharges, while these were 16.9%, 14.4% and 15.7% for homeowner-designed, contractor-designed and engineer-designed systems, respectively.

A total of 1106 bimodal surveys were conducted, of which 654 were web-based and 452 institutional face-to-face. The majority (71.4%) of survey respondents were male, with respondents having a mean age of 34.7 years (Range 18–91; Std. Dev. 16.6). Web-based respondents had a mean age of 45.8 years (Range 18–91; Std. Dev. 13.4), while this was 20.2 years (Range 18–60; Std. Dev. 4.5) among the group survey sample. A positive measure of skewness (0.631) associated with respondent age indicates a higher proportion of data-points located left of the mean. Female respondents had a higher mean age (t(1003) = 10.952, p < 0.001), with male and female respondent mean ages of 31.2 years and 43.1 years, respectively. A slight majority of respondents (51.7%, n = 572) were

3.1.3. Household water supply Private wells (38.5%) and public mains supplies (33.6%) accounted for the majority of respondents. Public group water schemes (PuGWS) and private group water schemes (PrGWS) supplied 12% (n = 130) and 13.4% (n = 146) of respondents, respectively. Results suggest a significant association between DWWTS type and water supply type (v2(20) = 59.882, p < 0.001); patterns among bottled water users were found to deviate significantly from mean trends, with higher lack of awareness of DWWTS type (19.2%) compared with the other groups (mean 3.2%) (see Fig. 2).

3. Results

Please cite this article in press as: Naughton, O., Hynds, P.D. Public awareness, behaviours and attitudes towards domestic wastewater treatment systems in the Republic of Ireland. J. Hydrol. (2013), http://dx.doi.org/10.1016/j.jhydrol.2013.08.049

O. Naughton, P.D. Hynds / Journal of Hydrology xxx (2013) xxx–xxx

5

Fig. 1. Composition of respondent reported DWWTS type. Percentages illustrated on first vertical axis (left), while mean reported DWWTS ages are shown on second vertical axis (right).

3.2. DWWTS inspection and maintenance 3.2.1. DWWTS information provision A majority of respondents (70.5%; n = 772) noted that they had not previously been supplied with information regarding system operation/maintenance, while 28.8% (n = 319) reported that they had previously received information; the remaining 0.4% (n = 4) were unaware. No significant association existed pertaining to the provision of information and either gender or age. A relationship was found between home ownership and information provision, home owners were approximately 1.8 times more likely to have received previous guidance (OR 1.851; v2(1) = 20.583, p < 0.001). Information was more often provided to owners of newer systems (t(970) = 7.507, p < 0.001), with a mean system age of 17 years associated with data provision, while this was 25.5 years where no information was provided. This pattern was reflected with respect to system designer (v2(8) = 125.125, p < 0.001). Information was provided more frequently in the case of ‘‘specialistdesigned’’ systems (44% for engineer/architect and 66% for manufacturer) than either homeowner – (22%) or contractordesigned (26%) systems.

3.2.2. DWWTS self-inspection Just under 77% (n = 822) of respondents reported that a visual inspection of their domestic treatment system had been previously undertaken, with 16.2% (n = 173) and 7% (n = 75) of respondents declaring no previous inspection and lack of awareness, respectively. A relationship was noted between the domestic water source type and previous treatment system inspection (v2(10) = 72.73, p < 0.001). While respondents supplied by a private well, PrGWS and PuGWS reported previous system inspections in 83.6%, 84.1% and 76.7% of cases, respectively, these were 67.8% and 61.5% in the case of respondents supplied by municipal mains and bottled water, respectively. Previous provision of data was also found to be significantly related to system inspection (v2(4) = 10.891, p = 0.028); 83% of respondents in receipt of previous information reported previous DWWTS inspection, while 74.6% of respondents not in receipt of information had previously undertaken inspection. Respondents who had carried out previous inspection (n = 822) were asked if any drainage issues (e.g. waterlogging or surface ponding) had been noted, with 20% reporting previous drainage problems in the vicinity of their DWWTS. An association was noted with respect to system age (t (938)=-3.66,

Fig. 2. Primary household drinking water supply type and associated DWWTS type (%) (Private Well n = 418; PrGWS n = 146; PuGWS n = 130; Public Mains n = 365; Bottled n = 26).

Please cite this article in press as: Naughton, O., Hynds, P.D. Public awareness, behaviours and attitudes towards domestic wastewater treatment systems in the Republic of Ireland. J. Hydrol. (2013), http://dx.doi.org/10.1016/j.jhydrol.2013.08.049

6

O. Naughton, P.D. Hynds / Journal of Hydrology xxx (2013) xxx–xxx

p < 0.001); DWWTS with drainage problems were 5 years older than inspected systems with no drainage issues. 3.2.3. DWWTS de-sludging 67.4% of respondents (n = 722) reported that their DWWTS had been previously de-sludged, 22.9% (n = 246) stated that their system had not previously been de-sludged, while the remainder (9.7%) were unaware. System age was significantly associated with previous de-sludging (t(885) = 6.604, p < 0.001); previously desludged systems were typically older (24.7 years vs. 16.7 years). Only 32.6% of respondents reported that their system was de-sludged at least every two years (including respondents indicating no previous de-sludging) (Fig. 3). No significant association was found between de-sludging frequency and system age, system designer or system type. No association was found between surface drainage discharge and either previous de-sludging or de-sludging frequency. 3.3. DWWTS registration and remediation 3.3.1. Regulation of DWWTSs A slight majority of respondents (58.5%, n = 600) agreed that DWWTS should be regulated through a program of monitoring and inspection. There was a notable gender imbalance with respect to regulatory attitudes (v2(1) = 42.438, p < 0.001), with female respondents 2–3 times more likely to agree with active regulation (OR 2.715; 1.998–3.668). There was a significant mean respondent age difference with respect to the same issue (t(933) = 10.053, p < 0.001). Respondents who agreed with the impending regulatory strategies had a mean age of 38.7 years, while those who disagreed had a mean age if 28.3 years. This was further reflected in the age-dependent issue of home-ownership (v2(1) = 80.18, p < 0.001), with homeowners over 3 times more likely to agree with monitoring and inspection of their domestic treatment system (OR 3.218; 2.482–4.173). There was no significant association between system age and attitude towards regulation (p = 0.127). Notably, no significant relationship was found between self-reported respondent DWWTS inspection and respondent attitude/ perception towards regulatory authority inspection. 3.3.2. Monitoring and inspection fees With respect to questions relating to DWWTS registration, monitoring and inspection fees, only data garnered from homeowners was considered for analysis, as it was considered that this sample population was particularly relevant with respect to these

issues. Only 23.1% of homeowners agreed that a charge should be levied by the regulatory authority for monitory and inspection services. No significant difference was found within the homeowner subset with respect to age or gender and the belief that an inspection fee should be incurred. 3.3.3. Intent to register Overall, 70.4% of homeowners (n = 366) reported an intention to register their DWWTS, whilst 15.2% reported that they would not register (n = 79). The remainder of respondents (14.4%, n = 75) intimated that they were undecided at the time of questioning. Female homeowners were over twice as likely to register (OR 2.272; v2(1) = 8.327, p = 0.004). Homeowners previously indicating an agreement with DWWTS regulation were over 5 times more likely to register their DWWTS (OR = 5.006; v2(1) = 46.006, p < 0.001). No significant association was found between homeowner water supply type and intention to register. 3.3.4. Perceived requirement for remedial works In all, 24.8% of respondents (n = 129) noted that they believed some remedial works would be required. System type was associated with the perceived requirement for remedial works (v2(3) = 12.343, p = 0.006); while 27.2% of septic tank owners believed remediation was likely necessary, only 5.9% and 10% of secondary system and reed-bed owners, respectively, shared this sentiment. Further, where homeowners were previously in receipt of DWWTS information, they were almost three times more likely to believe remedial works would not be necessary (OR 2.932; v2(1) = 19.268, p < 0.001). No association was found between either homeowner attitudes towards a monitoring/inspection scheme or intent to register and the belief that their DWWTS would necessitate remedial works. 3.3.5. Grants for remedial works 92% of respondents (n = 936) agreed that a grant should be made available to subsidise necessary remedial works upon system inspection. Respondents who agreed that a grant should be made available for compulsory remediation works were subsequently asked what proportion of the overall cost of remediation should be borne by the regulatory authority (Fig. 4). As shown, approximately 55% of respondents believe that the regulatory should bear >50% of necessary remedial costs, while over 20% believe that 100% of these costs should be borne by the government/authority. 23% proclaimed means-testing as a preferred implementation strategy. Respondents attitudes towards any

Fig. 3. Self-reported mean DWWTS de-sludging frequencies. Respondents indicating no previous de-sludging or lack of knowledge relating to previous de-sludging have been excluded (n = 722).

Please cite this article in press as: Naughton, O., Hynds, P.D. Public awareness, behaviours and attitudes towards domestic wastewater treatment systems in the Republic of Ireland. J. Hydrol. (2013), http://dx.doi.org/10.1016/j.jhydrol.2013.08.049

O. Naughton, P.D. Hynds / Journal of Hydrology xxx (2013) xxx–xxx

7

Fig. 4. Respondent attitudes towards remedial grant percentages (n = 970).

potential grant and subsequently, suitable grant proportion was not found to be demographically linked (age, gender, residential owner, household composition). However, one-way ANOVA showed an association between grant proportion and system age (F(4) = 2.656, p = 0.037), Bonferroni multiple comparisons tests indicate that the more recently the DWWTS has been installed, the more likely the homeowner is to believe that 100% of remedial costs should be borne by the legislative authority. Conversely, owners of traditional septic tank type systems exhibited a higher preference towards increased grant proportions (50–75%, 100%), while owners of secondary treatment systems and reed-bed systems favoured lower grant proportions or means testing (v2(4) = 10.24, p = 0.037). Previous receipt of data was also associated with grant proportion choice (v2(4) = 10.24, p = 0.037); respondents in receipt of previous DWWTS guidance exhibited increased favourability towards lower grant proportions and means testing. Respondents in favour of the impending monitoring/inspection scheme (v2(4) = 45.215, p < 0.001) and respondents who exhibited an intent to register their system (v2(4) = 49.291, p < 0.001) were also more inclined towards lower grant proportions and/or means testing. 3.4. Environmental contamination 3.4.1. DWWTS risk perception In all, 78% of respondents (n = 789) agreed that malfunctioning/ poorly designed treatment systems pose a threat to the environment. While no statistical difference was noted regarding gender and DWWTS threat acknowledgement, a significant mean difference was recorded pertaining to respondent age (F = 5.339, p < 0.001); typically those acknowledging treatment systems as a potential hazard were older than those who did not (36.2 years vs. 28 years). Household size/composition and treatment system age did not have any discernible relationship with ‘‘threat acknowledgement’’, nor did previous provision of guidance. However, household water supply was significantly associated (v2(4) = 12.507, p = 0.014). Households with a private domestic supply (private well or PrGWS) exhibited a mean threat awareness level of 75%, those employing a municipal/public supply (public mains or PuGWS) had a higher mean awareness of 82.9%, while those using primarily bottled water for domestic supply had an awareness level of 64%. Those respondents exhibiting an awareness of the potential environmental threat of DWWTSs were 5.5 times more likely to agree with impending regulatory measures (OR 5.574; 3.994–7.778; v2(1) = 114.998, p < 0.001). Furthermore, those respondents who acknowledged DWWTSs as a potential threat were almost 5 times more likely to register their system

(OR 4.872; 3.403–6.974; v2(1) = 81.839, p < 0.001), and approximately 40% more likely to believe that remedial works would be required upon inspection (OR 1.427; v2(1) = 3.979, p = 0.046). 3.4.2. Drinking water contamination In all, 21.5% of respondents (n = 214) reported that their domestic water source had previously been contaminated. Respondents whose domestic water source had previously been contaminated were asked to report on the contaminant source, where known (Fig. 6). DWWTS Domestic water supply type was found to be significantly associated with previous contamination (v2(4) = 13.8, p = 0.008). While self-reported contamination of private supplies ranged from 16.4% (private wells) to 20.1% (PrGWS), and public systems ranged from 22.6% (PuGWS) to 27% (municipal supplies), those respondents consuming primarily bottled water reported a previous contamination rate of 33.4% (8/24) (Fig. 5). Respondents in receipt of previous DWWTS guidance reported a higher proportion of previous contamination (28.2%), than those who did not (19%) (v2(1) = 10.849, p = 0.004). Respondents whose water supply had previously been contaminated, irrespective of the contaminant source, were approximately 2.5 times more likely to agree with the proposed monitoring and inspection scheme (OR 2.528; v2(1) = 29.86, p < 0.001). They were also approximately 1.6 times more likely to agree with payment of an inspection fee relating to the aforementioned inspection scheme (OR 1.677; v2(1) = 7.742, p = 0.005), and almost twice as likely to acknowledge DWWTSs as a potential environmental threat, irrespective of the contaminant source (OR 1.748; v2(1) = 7.424, p = 0.006). 3.5. Respondent type cluster analysis Clustering was performed on ranked output data from one question developed to capture respondents’ general perception towards the introduction of the septic tank registration and inspection program. Respondents were asked why they believed the new septic tank registration and inspection program was being introduced, and asked to rank four possible reasons in terms of importance. The choices offered were: to protect water quality (Q), to protect human health (H), to generate governmental revenue (R) and to comply with European legislation (C). This question was included within Section 3 of the survey (‘‘DWWTS Registration and Remediation’’). Cluster number was optimised using hierarchical clustering and the ‘‘elbow method’’ (Ketchen and Shook, 1996); it was found that suitable cluster numbers for adequate categorisation of this sample size (max cluster number; 24) and distribution were three and

Please cite this article in press as: Naughton, O., Hynds, P.D. Public awareness, behaviours and attitudes towards domestic wastewater treatment systems in the Republic of Ireland. J. Hydrol. (2013), http://dx.doi.org/10.1016/j.jhydrol.2013.08.049

8

O. Naughton, P.D. Hynds / Journal of Hydrology xxx (2013) xxx–xxx

Fig. 5. Primary household drinking water supply type and self-reported previous contamination (%) (n = 987).

Fig. 6. Primary household drinking water supply type and primary contaminant source where previous contamination has occurred (n = 214).

nine. It was considered that nine clusters would be too numerous for meaningful classification, therefore, three clusters were developed employing k-means cluster analysis (max 100 iterations, convergence criterion 0). Two large clusters (Clusters HQ and RC) and one proportionately small cluster (Cluster M-RQ) were identified and characterised (Table 1) from 945 respondents. The first cluster, labelled HQ, may be described as those respondents who assign higher importance to human health and environmental quality (HQ) with respect to environmental governance. Conversely, cluster RC seems to represent those respondents who exhibit elevated Table 1 Distance between final cluster centres and cluster case numbers (top) and average rank given to reasons for the septic tank registration and inspection program (bottom). Cluster name HQ

M-RQ

RC

HQ M-RQ RC Cluster case number

– 2.872 3.506 527

2.872 – 2.486 90

3.506 2.486 – 328

Environmental quality Human health Governmental revenue Legislative compliance

Average rank 2 2 4 3

2 3 1 4

3 4 1 2

importance to revenue and ‘‘forced’’ compliance (RC) in regard to implementation of legislative tools. Cluster 3 would seem to represent the ‘‘middle ground’’; these respondents, while closer in attitude to cluster HQ respondents’ on issues of environmental quality, risk and human health, also share some a level of agreement with cluster RC respondents on issues of regulation and governance. Consequently, the third cluster has been named ‘‘mixed revenuequality’’ (M-RQ). Cluster membership was assigned to individual respondents (n = 945), and included as an input variable for statistical analysis, in order to examine underlying demographic and attitudinal patterns within mined clusters. Respondent gender, age, household ownership, household size/composition and domestic water supply type were not found to be related with cluster membership. However, respondents DWWTS type was shown to exhibit an association with cluster membership (v2(4) = 22.063, p = 0.015). In cases where drinking water contamination was recorded, analysis showed that the association between cluster membership and contamination source was not statistically significant. Membership of cluster M-RQ ranged from 0% (secondary systems) to 16.6% (respondents unaware of their DWWTS type); membership of cluster HQ ranged from 54.7% (septic tank users) to 72.7% (reed-bed users), while membership of cluster RC ranged from 18.2% (reedbed users) to 35.2% (septic tank users). Overall, owners of conventional septic tank systems were more likely to be members of the

Please cite this article in press as: Naughton, O., Hynds, P.D. Public awareness, behaviours and attitudes towards domestic wastewater treatment systems in the Republic of Ireland. J. Hydrol. (2013), http://dx.doi.org/10.1016/j.jhydrol.2013.08.049

9

O. Naughton, P.D. Hynds / Journal of Hydrology xxx (2013) xxx–xxx

Table 2 Breakdown of cluster responses to questions on the perception of DWWTS as a threat to the environment, the proposed regulation of DWWTS through monitoring and inspection, and the inspection fee levied as part of regulation. Cluster name

HQ M-RQ RC

Threat to environment

Monitoring/inspection

Inspection fee

Yes (%)

No (%)

Yes (%)

No (%)

Yes (%)

No (%)

85 82 71

15 18 29

69.5 50 49.5

30.5 50 50.5

34 20 14

66 80 86

RC clusters, while users of more contemporary secondary and reed-bed systems were more likely members of the HQ cluster. A significant relationship was found with respect to respondent attitudes towards the proposed national monitoring and inspection scheme (v2(2) = 38.803, p < 0.001). Members of clusters M-RQ and RC were equally split between agreement and disagreement with the inspection programme while respondents within cluster HQ were more likely to agree with inspection (Table 2). A significant association was found between cluster membership and attitudes towards inspection fees (v2(2) = 31.277, p < 0.001). While 20% and 14.1% of respondents in clusters M-RQ and RC, respectively, agreed that some form of inspection charge should be incurred by the system owner, this was significantly higher in cluster HQ at almost 34% (n = 143). Respondents’ acknowledgement of malfunctioning DWWTS as environmental contaminant sources resulting in potential human health burden, was also related with respondent cluster membership (v2(2) = 22.281, p < 0.001). While 85% (n = 439) and 82% (n = 72) and of respondents in clusters HQ and M-RQ, respectively, acknowledged DWWTSs as a potential threat to aquatic environments, only 71% (n = 230) of cluster RC made a similar acknowledgement. Analysis of homeowners intent to register with respect to cluster membership resulted in a statistically significant association (v2(4) = 29.96, p < 0.001). While 92% of respondents in cluster HQ (n = 255) reported that they had either already registered their system or planned on registering, 38.6% and 40% of M-RQ and RC cluster members, respectively, reported that they would not register their system or were currently undecided. Respondents associated with cluster RC had a generally negative disposition toward system registration, with 22.7% not planning to register their system.

4. Discussion The prevalence of risk-based regulation in modern environmental regulatory regimes has placed effective risk communication at the centre of environmental and health risk management (Chorus and Bartram, 1999; Rothstein et al., 2006). Effective risk communication is the purposeful exchange of information about risks between interested parties (Covello et al., 1986; Lang et al., 2001); the information to be communicated may relate to the level of risk, the significance or meaning of the risk or the policies aimed at managing or controlling the risk (Covello et al., 1986). The effective communication of environmental risks to the public, such as those posed by DWWTSs, can face many serious difficulties. The issue commonly occurs in a social and political arena characterised by high public interest, but with a lack of trust, confidence or credibility felt by the public towards regulatory authorities (Renn and Levine, 1991). There may be a difference between the public and the regulatory authority regarding in the perceived severity or importance of the risk (Black and Baldwin, 2012a); environmental professionals generally frame risk in terms of expected value of loss whereas the lay public tend to focus on extreme outcomes (Elliott, 2003). These disparities can further decrease trust and inhibit communication between regulators and regulatees (Elliott, 2003). Thus

an understanding of public risk perception, or the subjective assessment of the probability of a specified type of accident occurring and how concerned we are with the consequences (Sjöberg et al., 2004), is key for the effective regulation of environmental risks such as those posed by DWWTSs. Together with the attitudes and perceptions of individuals towards the risks posed by DWWTSs, a key factor which must be considered to aid effective regulation is the capacity of the regulatee to comply (Black and Baldwin, 2012b). Environmental and behavioural knowledge have a significant role in shaping waste management behaviour (Barr, 2007; Schahn and Holzer, 1990). The installation, operation and maintenance of DWWTSs in line with best practice requires a combination of appropriate environmental and behavioural knowledge; ideally individuals will have a general awareness of the environmental risks posed by DWWTSs (environmental knowledge) and also possess an understanding of the actions required to ensure correct functioning (behavioural knowledge). The Department of the Environment and the Environmental Protection Agency (EPA) in Ireland aim to engender this attitude through increased awareness via a series of novel public engagement strategies (EPA, 2013). The primary objective of these engagement strategies is to provide information or knowledge of the appropriate action, be it regular maintenance of the system or replacement if it poses an unacceptable risk, as this is a prerequisite for behaving in an appropriate manner (Barr, 2007). In order for this public information campaign to be successful, a clear understanding of current public perceptions, activities, and knowledge gaps in relation to DWWTS is required. The research presented in this article has identified knowledge deficits limiting the capacity of regulates to comply as well as key public risk perceptions which should be addressed to ensure the effective regulation and sustainability of DWWTSs in Ireland as required under EU legislation. It is also vital that policy makers understand and can detect changes in stakeholder behaviour and perceptions to allow the regulatory approaches to be amended and developed appropriately (Black and Baldwin, 2012a); this research provides baseline data which will facilitate such future regulatory development. To date, dissemination of DWWTS operation and maintenance does not appear to have been widespread; over 70% of respondents claimed they had received no previous information. This lack of appropriate behavioural knowledge is reflected in the reported prevalence of activities at odds with best practice. For example, a key operational deficiency that needs to be addressed is the nature of discharges to DWWTSs. Fifteen percent of respondents reported discharging rainwater and/or surface water run-off to their DWWTS. Best practice dictates that both wastewater (toilet) and grey water (sinks, baths, washing machines) are discharged into the DWWTS, but rainwater and surface water run-off should be diverted to a properly constructed soak-away (EPA, 2009). Excessive hydraulic loading can have a detrimental effect on the operation and efficiency of DWWTSs, potentially causing the discharge of untreated effluent and the blockage of percolation pipes (Butler and Payne, 1995; EPA, 2009). A significant reduction in the risk to human health and the environment would be achieved by specifically addressing this issue in future engagement strategies. Guidelines recommend that visual inspections of the system and de-sludging

Please cite this article in press as: Naughton, O., Hynds, P.D. Public awareness, behaviours and attitudes towards domestic wastewater treatment systems in the Republic of Ireland. J. Hydrol. (2013), http://dx.doi.org/10.1016/j.jhydrol.2013.08.049

10

O. Naughton, P.D. Hynds / Journal of Hydrology xxx (2013) xxx–xxx

of the septic tank are periodically carried out (EPA, 2009). 23% of respondents indicated that either there had been no visual inspection of their DWWTS had taken place or they were unaware of such. Interestingly, there was a significant relationship between source of drinking water and visual inspection. Users of water supply sources most vulnerable to DWWTS contamination, i.e. private wells and private group water schemes, showed the highest selfinspection rate (83.6% and 84.1% respectively), as compared to 67.8% of those on municipal mains. Only 33% of respondents reported that their system was desludged in line with recommended frequencies (EPA, 2009). Highlighting the fiscal benefits of appropriate maintenance of and discharge to DWWTSs to householders (e.g. increased service life) may increase public engagement with the National Inspection Plan and improve public behavioural knowledge. Advances in environmental knowledge will be more difficult to achieve in comparison to changes in behavioural knowledge. Changes relating to environmental knowledge are also likely to take place over a longer period. However, this research has identified some key attitudes and relationships that, if addressed by policy makers, may be used to positively influence public environmental knowledge. A significant majority of respondents (78%) agreed that malfunctioning of poorly designed DWWTSs pose a threat to the environment. Respondents acknowledging DWWTS as an environmental threat were over 5 times more likely to agree with regulatory measures. However, there was a substantial fall off in support for the program of monitoring and inspection, with only a slight majority of respondents (58.5%) agreed that DWWTS should be regulated in this way. This study has identified the issue of household water supply as central to public risk perception and attitude towards regulation, and a primary focus for civic engagement. A significant relationship was found between household water supply type and DWWTS visual inspection, with users of private supplies more likely to carry out an inspection than those on public supplies. Household water supply type was also significantly associated with perception of environmental risk posed by DWWTSs; the mean threat awareness level was higher in households using a public supply (83%) than those with a private domestic supply (75%). Drinking water contamination also exhibited a significant association with the risk perception, irrespective of contaminant source. Respondents were almost twice as likely to acknowledge DWWTSs as a potential environmental threat and approximately 2.5 times more likely to agree with the proposed monitoring and inspection scheme if their water supply had previously been contaminated. This suggests that a greater understanding of the risks posed by DWWTS to drinking water would improve public engagement with inspection and maintenance strategies. In the most recent drinking water surveillance report (EPA, 2012), public water supplies were found to show evidence of microbial contamination in 1.3% of cases, reflecting an 86% reduction in microbial detections since 2005 (EPA, 2012). Private group water schemes were shown to be microbially contaminated in 10.2% of instances, down from 11.6% in 2010. Two recent research studies (Bacci and Chapman, 2011; Hynds et al., 2012) have found that 25–35% of private groundwater sources are intermittently contaminated with faecal indicators, and therefore the level of microbial non-compliance in private supplies is significantly higher than that of public supplies. Small private supplies were the only supply category that did not show an improvement in E. coli compliance in 2011 (EPA, 2012). Users of private wells, accounting for 38% of respondent water supplies and the source type most at risk from DWWTS contamination, reported the lowest contamination rates of all water supply types at 16.4%. Extensive research has been carried out to quantify the hydrological pathways, treatment mechanisms, groundwater vulnerabilities

and risks to human health associated with DWWTSs. However, this has not been translated into a greater public perception or understanding of this risk. Unlike private wells, public supplies are regulated and in the event of source contamination there is a rapid multimedia notification of consumers regarding potential health risks. This effective communication of risk is likely one of the principle drivers underpinning the higher reported contamination rate (27% vs. 16.4%). The higher self-inspection rate among private supply users indicates that the motivation to comply with regulation may exist, but that the appropriate capacity to comply may be lacking, i.e. they requisite knowledge to carry out an adequate visual inspection. Given the inherent difficulties in assessing DWWTSs, a more effective risk mitigation strategy may be to instead promote the inspection of private water supplies. Personal and familial well-being is the most important motivation to personal behaviour (Stern et al., 1993); there is an increased likelihood of pro-environmental behaviours when they are seen to promote health and well-being (Kollmuss and Agyeman, 2002). While it is impractical to monitor private supplies and so provide site-specific information on health risk, a risk communication strategy can create an aware and informed public which recognises the importance of individual behaviour in risk avoidance (Lang et al., 2001). Dissemination of information on how to correctly inspect both DWWTSs and private water supplies, together with more general risk information on the contamination rates of private supplies, may prove effective in improving environmental knowledge amongst private well users, the key demographic in the context of DWWTS public health risks. While this general scientific information exists, there is a need for the development of a socio-hydrological framework to assist in the communication of the risk to private drinking water sources. Results from this study indicate that engagement strategies aimed specifically at the young adult demographic could have significant environmental benefits in Ireland in the long term. Younger respondents exhibited lower awareness and knowledge of DWWTS type, as would be expected given the lower levels of homeownership. Perhaps more important is the relationship between respondent age and risk perception. Respondents acknowledging treatment systems as a potential hazard were typically eight years older than those who did not (36.2 years vs. 28 years). National legislation places a duty of care on DWWTS owners to ensure their system does not constitute a risk to human health or the environment. As homeowners are typically older individuals, this duty of care rarely falls on young adults. However, that does not mean that the younger audience should be ignored. Information from family and friends has been demonstrated as a key influence on individual risk perception (Rundblad et al., 2010; Brown et al., 2011). In a study on the communication, perception and behaviour during a natural disaster involving a ‘Do Not Drink’ notice Rundblad et al. (2010) found that older consumers and those in paid employment were particularly unlikely to read official information leaflets and unlikely to comply with advice. Young adults can be targeted via relatively low-cost methods, such as with social media and informative talks in educational institutions, whereas communication campaigns using traditional media which influence older stakeholders (such as television) may not be feasible in the current economic climate.To ensure the future environmental sustainability of DWWTSs there is a need to educate the younger generation in the potential risks and best practices associated with DWWTSs, with the aim of promoting pro-environmental behaviour in future custodians. In addition to future benefits, environmental behaviour and regulatory compliance among current custodians may also improve due to knowledge transfer along familial and community communication channels. A key consideration in risk communication within any engagement strategy is that the target stakeholders are unlikely to be a

Please cite this article in press as: Naughton, O., Hynds, P.D. Public awareness, behaviours and attitudes towards domestic wastewater treatment systems in the Republic of Ireland. J. Hydrol. (2013), http://dx.doi.org/10.1016/j.jhydrol.2013.08.049

O. Naughton, P.D. Hynds / Journal of Hydrology xxx (2013) xxx–xxx

single audience, but a composite of audiences that have different levels of understanding, education, values and risk perceptions (Lang et al., 2001). Statistical analysis found that respondents fell into three clusters depending on their general perception towards the introduction of the septic tank registration and inspection program. The four choices respondents were asked to rank in terms of importance (environmental quality, human health, governmental revenue and legislative compliance) were designed to gauge the general attitude towards DWWTS regulation. The two main clusters, HQ and RC, represent groups with similar environmental values or underlying orientations toward the physical environment (Barr, 2007). Respondents in cluster HQ believed the protection of public health and water quality were the key drivers behind the introduction of DWWTS regulation, while members of cluster RC perceived revenue generation and compliance with EU legislation to be driving regulation. A third, proportionally smaller cluster composed of individuals that shared some attributes with both larger clusters, believing the introduction of regulation was primarily driven by a mix of revenue generation and water quality protection (M-RQ). Parallels between these clusters can and the regulate type characterisation proposed by Black and Baldwin (2012b); cluster HQ members represent well-motivated regulatees while those in cluster RC less motivated and so less willing to comply. The acknowledgement of malfunctioning DWWTS as a potential threat to human health and the environment fell from 85% in cluster HQ to 71% in cluster RC. A similar decline was seen in support for monitoring and inspection, with 69.5% of cluster HQ members favouring regulation compared to 49.5% in cluster RC. Effective engagement strategies should aim to improve stakeholder motivation to comply with regulation (Black and Baldwin, 2012b), or in effect shift public perceptions and attitudes away from clusters M-RQ and RC towards those of cluster HQ. The attitude that regulation is driven primarily by monetary policy, national and international legislation needs to be addressed in proposed engagement strategies. While the timing of regulatory implementation is at least partially driven by the recent EU ruling against the State, the principle behind the regulation is scientifically justified and required to safeguard public health and the environment. That the hydrological theory governing risks posed by DWWTSs is well understood and documented, but not sufficiently reflected in public attitudes, emphasises the need for a socio-hydrological approach to this issue. Gains in support for pro-environmental behaviour can also be made among cluster HQ members, where respondents give equal weights to the threat posed by DWWTSs to public health and the environment. The perception that DWWTSs do not pose a significant risk to personal health that is prevalent among stakeholders represents a substantial barrier to effective risk management. Results show that were stakeholders perceive a tangible risk to health, generally in the form of drinking water supply vulnerability, they are more likely to actively engage in regulation and exhibit positive environmental behaviour. Greater engagement with active regulation could therefore be achieved across all clusters, and thus the public as a whole, by elucidating and emphasising the specific risks to human health posed by poorly installed and maintained DWWTSs. Cluster membership was not found to be significantly related to respondent gender, age, household ownership, and domestic water supply type, highlighting the diverse mix of audiences which make up each cluster. This research has highlighted the key risk perceptions and audiences associated with DWWTSs in Ireland. The engagement strategy proposed under the DWWTS National Inspection Plan must use this information to determine the appropriate audience, message and communication channel to ensure successful engagement in the risk communication strategies.

11

5. Conclusion The research described in this paper addresses the lack of knowledge on the behaviours and attitudes towards DWWTS in the Republic of Ireland. Analysis identified a number of significant knowledge gaps which currently exist among DWWTS users. These were associated with environmentally inadvisable behavioural practises, potentially leading to increased contamination vulnerability and subsequently, increased human exposure to waterborne contaminants. This study has identified the issue of household water supply as central to public risk perception and attitude towards regulation, and a potential focus for engagement strategies. This research has demonstrated the importance of perceived risk in DWWTS stakeholder behaviour; results show evidence that risk perceptions are related to other judgements, for example, respondents with a greater awareness of potential environmental threats of DWWTS were more supportive of measures to regulate the systems. Results indicate a greater understanding of the risks posed by DWWTS to drinking water would improve public engagement with inspection and maintenance strategies. This may be particularly effective amongst private well users, the group most at risk from DWWTS drinking water contamination, as they reported contamination rates far lower than current evidence indicates. A widespread perception among respondents was that DWWTS poses a significantly smaller threat to public health and the environment than research suggests. Younger respondents were also less likely to view DWWTS as a potential threat to public health and the environment. Hydrological theory quantifies the risks posed by inadequate DWWTSs; insights from social science disciplines are needed on how best to communicate this risk on a national strategic level to reduce risk to human health and the environment. Future engagement strategies should strive to provide guidance regarding the role of people and their activities within the hydrological cycle. The current study reinforces this conclusion, while providing evidence-based recommendations regarding provision of demographically focused educational strategies; these will further increase environmental policy compliance, and in so doing, decrease the human health and environmental contamination burden posed by DWWTSs.

Acknowledgements The authors would like to thank all anonymous respondents for taking the time to complete the questionnaires, and a number of national non-professional interest groups (ESAI, IAH, ICA, NFGWS) for help questionnaire distribution. The authors would also like to thank the staff and students of the Clonakilty, Gurteen, Kildalton and Mountbellew Agricultural Colleges for their time and assistance. The authors would also like to thank two anonymous reviewers for their comments which greatly improved the manuscript. This research has received support from the Environmental Protection Agency STRIVE programme under Project 2013-W-SS10.

References Agresti, A., 1996. Categorical Data Analysis. John Wiley & Sons, Inc. Hoboken, NJ. Ahmed, W., Neller, R., Katouli, M., 2005. Evidence of septic system failure determined by a bacterial biochemical fingerprinting method. J. Appl. Microbiol. 98 (4), 910–920. Bacci, F., Chapman, D. V., 2011. Microbiological assessment of private drinking water supplies in Co Cork, Ireland. Jnl. Water and Health 09.4, 738–751. Barr, S., 2007. Factors influencing environmental attitudes and behaviours: A U.K. case study of household waste management. Environ. Behaviour 39 (4), 435– 473.

Please cite this article in press as: Naughton, O., Hynds, P.D. Public awareness, behaviours and attitudes towards domestic wastewater treatment systems in the Republic of Ireland. J. Hydrol. (2013), http://dx.doi.org/10.1016/j.jhydrol.2013.08.049

12

O. Naughton, P.D. Hynds / Journal of Hydrology xxx (2013) xxx–xxx

Beal, C.D., Gardner, E.A., Menzies, N.W., 2005. Process, performance, and pollution potential: a review of septic tank-soil absorption systems. Aust. J. Soil Res. 43, 781–802. Beller, M., Ellis, A., Lee, S.H., Drebot, M.A., Jenkerson, S.A., Funk, E., 1997. Outbreak of viral gastroenteritis due to a contaminated well. International consequences. JAMA 278, 563–568. Birkhead, G., Janoff, E.N., Vogt, R.L., Smith, P.D., 1989. Elevated levels of Immunoglobulin A to Giardia lamblia during a waterborne outbreak of gastroenteritis. J. Clin. Microbiol. 27, 1707–1710. Black, J., Baldwin, R., 2012a. When risk-based regulation aims low: approaches and challenges. Regulat. Gov. 6, 2–22. Black, J., Baldwin, R., 2012b. When risk-based regulation aims low: a strategic framework. Regulat. Gov. 6, 131–148. Blalock, H.M., 1960. Correlational analysis and causal inferences. Am. Anthropol. 62 (4), 624–631. Borchardt, M., Bradbury, K., Alexander Jr., E., Kolberg, R., Alexander, S., Archer, J., Braatz, L., Forest, B., Green, J., Spencer, S., 2011. Norovirus outbreak caused by a new septic system in a dolomite aquifer. Ground Water 49, 85–97. Brown, S., Barton, M., Nicholls, R., 2011. Coastal retreat and/or advance adjacent to defences in England and Wales. J. Coastal Conserv. 15, 659–670. Butler, D., Payne, J., 1995. Septic tanks: problems and practice. Build. Environ. 30 (3), 419–425. C188/08, 2009. Failure of a Member State to fulfil obligations – directive 75/442/ EEC – waste – domestic waste waters discharged through septic tanks in the countryside – waste not covered by other legislation – failure to transpose. Judgment of the European Court of Justice (Second Chamber) of Complaint Brought by Commission of the European Communities against Ireland. Cape, P., 2010. Questionnaire length, fatigue effects and response quality revisited. In: Advertising Research Foundation Re: Think 2010 Convention. New York. 2010. Celik, E., Muhammetoglu, H., 2008. Improving public perception of tap water in Antalya City – Turkey. J. Water Supply: Res. Technol. – AQUA 57 (2), 109–113. Chorus, I., Bartram, J., 1999. Toxic Cyanobacteria in Water. A Guide to Their Public Health Consequences. Monitoring and Management, E & FN Spon, London. Cohen, L., Manion, L., Morrison, K., 2004. Research Methods in Education, fifth ed. Routledge-Falmer, London, UK. Covello, V.T., von Winterfeldt, D., Slovic, P., 1986. Risk communication: a review of the literature. Risk Abstracts 3 (4), 171–182. CSO, 2012. Census 2011 Profile 4: The Roof over our Heads - Housing in Ireland. Central Statistics Office, Skehard Road, Cork, Ireland. Daly, D., 2003. ‘‘Editorial.’’ Geological Survey of Ireland Groundwater Newsletter 42: 1. DoECLG. 2012. Water Services Acts 2007 and 2012 (Domestic Waste Water Treatment Systems) Regulations 2012. Statutory Instrument No. 223 of 2012. Government Publications Office, Dublin 2, Ireland. Duffy, P., 2000. Trends in nineteenth and twentieth century settlement. In: Barry, T. (Ed.), A History of Settlement in Ireland. Routledge, London. Elliott, M., 2003. Risk perception in environmental decision making. Environ. Pract. 5, 214–222. EPA, 2009. Code of Practice. Wastewater Treatment and Disposal Systems Serving Single Houses (PE < 10). Environmental Protection Agency, Ireland. EPA, 2012. The Provision and Quality of Drinking Water in Ireland – A Report for the Year 2011. Environmental Protection Agency, Ireland. EPA, 2013. National Inspection Plan: Domestic Waste Water Treatment Systems. Environmental Protection Agency, Ireland. Fleming, C., Bowden, M., 2009. Web-based surveys as an alternative to traditional mail methods. J. Environ. Manage. 90, 284–292. Garvey, P., Carroll, A., McNamara, E., McKeown, P., 2010. Epidemiology of verotoxigenic E. coli in Ireland, 2009. Epi-Insight, vol. 11(9), 1–3. Gill, L.W., O’Luanaigh, N., Johnston, P.M., Misstear, B.D.R., O’Suilleabhain, C., 2009. Nutrient loading on subsoils from on-site wastewater effluent, comparing septic tank and secondary treatment systems. Water Res. 43 (10), 2739– 2749. Hawton, K., Rodham, K., Evans, E., Weatherall, R., 2002. Deliberate self-harm in adolescents: self report survey in schools in England. BMJ 2002 (325), 1207. Health Protection Surveillance Centre (HPSC). 2013. Annual Report 2011: Verotoxigenic E. coli. . Hynds, P.D., Misstear, B.D., Gill, L.W., 2012. Development of a microbial contamination susceptibility model for private domestic groundwater sources. Water Resour. Res. 48, W12504. http://dx.doi.org/10.1029/ 2012WR012492. Hynds, P.D., Misstear, B.D., Gill, L.W., 2013. Unregulated private wells in the Republic of Ireland: Consumer awareness, source susceptibility and protective actions. Journal of Environmental Management (JEMA), 127, 278–288. IBM Corp. Released 2011. IBM SPSS Statistics for Windows, Version 20.0. Armonk, NY, IBM Corp. Karanis, P., Kourenti, C., Smith, H., 2007. Waterborne transmission of protozoan parasites: a worldwide review of outbreaks and lessons learnt. J. Water Health 5, 1–38.

Ketchen, D., Shook, C., 1996. The application of cluster analysis in strategic management research: an analysis and critique. Strateg. Manage. J. 17, 441– 458. Kollmuss, A., Agyeman, J., 2002. Mind the Gap: Why do people act environmentally and what are the barriers to pro-environmental behaviour? Environ. Edu. Res. 8 (3), 239–260. Kramer, M.H., Herwaldt, B., Craun, G., Calderon, R., Juranek, D., 1996. Waterborne disease: 1993 and 1994. J. Am. Water Works Assoc. 88, 66–80. Lang, S., Fewtrell, L., Bartram, J., 2001. Risk communication. In: Fewtrell, L., Bartram, J. (Eds.), World Health Organisation (WHO) Water Quality: Guidelines. Standards and Health, IWA Publishing, London, UK. McKeown, P., Garvey, P., 2011. Sustained surge in VTEC cases over last number of weeks. In: Epi-Insight; Disease Surveillance Report of HPSC. Ireland, vol. 12(12), December 2011. Moelants, N., Janssen, G., Smets, I., Van Impe, J., 2008. Field performance assessment of onsite individual wastewater treatment systems. Water Sci. Technol. 58 (1), 1–6. Palmer-Felgate, E.J., Mortimer, R.J.G., Krom, M.D., Jarvie, H.P., 2010. Impact of pointsource pollution on phosphorus and nitrogen cycling in stream-bed sediments. Environ. Sci. Technol. 44, 908–914. Parvez, F., Chen, Y., Argos, M., Hussain, A.I., Momotaj, H., Dhar, R., 2006. Prevalence of arsenic exposure from drinking water and awareness of its health risks in a Bangladeshi Population: results from a large population-based study. Environ. Health Perspect 114, 355–359. Peat, J., Mellis, C., Williams, K., Xuan, W., 2002. Health Science Research: A Handbook of Quantitative Methods. Sage, London. Purcell, M., Magette, W.L., 2010. Attitudes and behaviours towards waste management in the Dublin, Ireland region. Waste Manage. 30 (10), 2003–2012. Renn, O., Levine, D., 1991. Trust and credibility in risk communication. In: Kasperson, R.E., Stallen, P.J.M. (Eds.), Communicating Risks to the Public. Kluwer, Dordrecht, pp. 175–214. Rennie, S.C., Crosby, J.R., 2001. Are ‘‘tomorrow’s doctors’’ honest? Questionnaire study exploring medical students’ attitudes and reported behaviour on academic misconduct. BMJ 322, 274–275. Rhodes, S.D., Bowie, D.A., Hergenrather, K.C., 2003. Collecting behavioural data using the world wide web: considerations for researchers. J. Epidemiol. Health 57, 68–73. Rothstein, H., Irving, P., Walden, T., Yearsley, R., 2006. The risks of risk-based regulation: insights from the environmental policy domain. Environ. Int. 32, 1056–1065. Rundblad, G., Knapton, O., Hunter, P.R., 2010. Communication, perception and behaviour during a natural disaster involving a ‘Do Not Drink’ and a subsequent ‘Boil Water’ notice. a postal questionnaire study. BMC Public Health 10, 641. Scandura, J.E., Sobsey, M.D., 1997. Viral and bacterial contamination of groundwater from on-site sewage treatment systems. Water Sci. Technol. 35 (11/12), 141– 146. Schahn, J., Holzer, E., 1990. Studies of environmental concern: the role of knowledge, gender and background variables. Environ. Behaviour 22, 767–786. Scott, M., Murray, M., 2009. Housing rural communities: connecting rural dwellings to rural development in Ireland. Housing Studies 24 (6), 755–774. Scott, M., Russell, P., Redmond, D., 2007. Active citizenship, civil society and managing spatial change in the rural-urban fringe. Policy Politics 35, 163–190. Simpson, H., Hodgins, E., 2002. Raising groundwater awareness – A rural Ontario case study. In: Proceedings of the Water Environment Federation, vol. 16. Watershed. pp. 1787–1802. Sivapalan, M., Savenije, H.H.G., Blöschl, G., 2012. Socio-hydrology: a new science of people and water. Hydrol. Process. 24, 1270–1276. Sjöberg, L., Moen, B.E., Rundmo, T., 2004. Explaining risk perception: an evaluation of the psychometric paradigm in risk perception research. Rotunde publikasjoner Rotunde no. 84, 2004 Editor: Torbjørn Rundmo, Norwegian University of Science and Technology, Department of Psychology, 7491 Trondheim, Norway. Spatial Planning Unit. 2001 Rural and Urban Roles – Irish Spatial Perspectives. Department of Environment, Heritage and Local Government, Dublin, Ireland. Stern, P.S., Dietz, T., Karlof, L., 1993. Values orientation, gender, and environmental concern. Environ. Behaviour 25 (3), 322–348. Turbow, D.J., Kent, E.E., Jiang, S.C., 2008. Web-based investigation of water associated illness in marine bathers. Environ. Res. 106 (1), 101–109. USEPA. 2002. Onsite Wastewater Treatment Systems Manual. US Environmental Protection Agency, Office of Water, Washington. Wheater, H.S., Gober, P., 2011. Towards a new paradigm of Socio-Hydrology; insights from the Saskatchewan River Basin. American Geophysical Union, Fall Meeting 2011, abstract #H11F-1134. Withers, P.J., Jarvie, H.P., Stoate, C., 2011. Quantifying the impact of septic tank systems on eutrophication risk in rural headwaters. Environ. Int. 37, 644–653. Withers, P.J.A., May, L., Jarvie, H.P., Jordan, P., Doody, D., Foy, R.H., Bechmann, M., Cooksley, S., Dils, R., Deal, N., 2012. Nutrient emissions to water from septic tank systems in rural catchments: uncertainties and implications for policy. Environ. Sci. Policy 24, 71–82.

Please cite this article in press as: Naughton, O., Hynds, P.D. Public awareness, behaviours and attitudes towards domestic wastewater treatment systems in the Republic of Ireland. J. Hydrol. (2013), http://dx.doi.org/10.1016/j.jhydrol.2013.08.049