Sustainable drainage systems: helping people live with water

Sustainable drainage systems or ‘ Suds ’ are increasingly accepted as an effective means of ‘ making space for water ’ , adapting to possible climate change and helping communities become more flood and drought resilient. This study explores potential shifts in perception and attitude through Suds installation, development and habituation. Attitudes and awareness in communities in the USA and UK, where Suds have been in place for some time, were compared and contrasted, examining any evolution of beliefs and practices and wider community resilience. The principal finding was that there existed a lack of understanding about the existence and function of Suds. The paper concludes that consultation regarding solutions during Suds planning and installation, and ongoing dialogue afterwards, could usefully be explored as a means to improve local awareness of and satisfaction with Suds and promote greater understanding of their function. This may in turn encourage behaviour change to improve longer-term functionality of Suds and increase community resilience to flooding and drought.


Introduction
Policies in the USA and the UK now favour more 'sustainable' approaches to flood risk management (EPA, 2007(EPA, , 2013Defra, 2005;Pitt, 2008;Scottish Government, 2003;Werritty et al., 2007). Sustainable drainage systems (Suds), known as 'best management practices' (BMPs) and 'low-impact development' (Lid) in the USA, are surface water drainage systems designed to function in a manner akin to natural drainage. Positive impacts include delivering a sequence of sustainable water management practices, such as reducing surface water runoff, improving water quality and enhancing amenity and biodiversity functions. Suds have been described as 'a catch-all term for a number of different systems' (Jones and MacDonald, 2007) for slowing, filtering or retaining runoff, and putting excess water to use near where it lands rather than dispersing it quickly. Suds can be anything from water butts (rain barrels) and permeable paving, to green roofs, wetlands, ponds, filter and infiltration trenches, swales and rain gardens. Since over 80% of the populations in the USA and the UK live in already developed urban areas (World Bank, 2014), the focus for this research is specifically upon retrofit Suds in the urban environment.
Many urban Suds systems are now well established; evaluation of longer-term performance is possibleand emerging (Davis et al., 2001;Mitchell, 2005). For example, research has explored the deterioration in infiltration at parking sites (Achleitner et al., 2007), identification of the best performing vegetation in rain gardens (Land and Water Constructions, 2006), maintenance regimes for Suds ponds in Scotland (Heal, 2000) and England (Heal et al., 2008), eco-roof performance in the USA (Kurtz, 2010) and green roof storage capacity (Getter et al., 2007). Some research has also been carried out to determine the social benefits of Suds as implicitly measured through house prices over time (Netusil et al., 2014;Zhou et al., 2013).
What is less well understood is how Suds are perceived by their local communities (although some research is emerging, see Shandas et al. (2010)). This should be a central concern, because such perceptions will affect the 'felt' benefits and costs of Suds, willingness to contribute towards maintenance costs and how people interact with them and, therefore, will have an impact on how 'sustainable' they are likely to be (the life expectancy of installations). The opinions of professional bodies and authorities will of course be central to whether and how Suds are implemented, but the research in this paper focuses on the much less studied matter of interactions between Suds and local communities.
The small number of studies conducted around public perceptions and attitudes towards Suds have produced contrasting results. Werritty et al. (2007) found that respondents preferred structural flood defences (e.g. concrete walls or levees), these being supported by over 90% of respondents, with Suds low down the list. Johnson and Priest (2008) found that the insurance industry, media and public remain 'heavily focused' on using structural defences. In contrast, Wendy Kenyon (2007) found that flood walls and embankments were respondents' least favoured options; instead approaches such as regenerating or planting native woodlands were preferred. Similarly overall, Apostolaki and Jefferies (2005) found that respondents preferred 'softer' approaches to managing flood risk, such that attitudes toward ponds and managed rivers were found to be positive because of their amenity, recreation, biodiversity and aesthetic benefits. Awareness of Suds' flood functions, however, was low, with most respondents unaware of their contribution to controlling, filtering and slowing runoff (Apostolaki and Jefferies, 2005).
In the study by Apostolaki and Jefferies (2005), views about Suds were found to be related to awareness of the functions and services they provided: flood control, improved water quality, opportunities for water harvesting and improved amenities, recreation facilities and biodiversity. One of their overriding findings and messages, therefore, was that education and consultation are fundamental to building understanding to enable sustainable strategies to be more effectively pursued, appreciated, managed and maintained. More recently, work by Bastien et al. (2011) found public awareness of the Suds functions of ponds to be much higher than was found by Apostolaki and Jefferies (2005), with close to 75% of people surveyed understanding their contribution to reducing flood risk. They nonetheless again emphasised the value of educational work around Suds ponds; safety, for example, was the most significant public concern around ponds, yet large differences existed between perceived and actual safety levels (see also McKissock et al., 1999).
The disparity between findings in these studies points to the need to establish a more thorough understanding around public perceptions of different Suds approaches. This research addresses the disparity by comparing residents' views and understandings of facilities between two sites: one in the USA and one in the UK. The purpose of these two small, in-depth, qualitative studies was not to produce widely generalisable findings, but rather to help deepen understanding of how people perceive the purpose and functioning of specific urban Suds designed to control surface water flow at source, behaviours around these Suds, and whether this behaviour is likely to facilitate or impede functioning.

Methods
This study focused on one site in the USA and one in the UK (see Figure 1) where Suds have been part of the landscape for the past 8-10 years. This was on the basis that the installations would no longer be seen as 'new' and that any behaviour changes they might encourage from those living in proximity would have occurred. In other words, the devices would be viewed as more natural or 'normal' (Barr et al., 2011;Hargreaves, 2010;Proshansy et al., 1983). The sites were located in the two climatically comparable locales of Bristol (UK) and Portland (Oregon, USA). In Bristol the Suds of interest were permeable paving (and a restored park area), and in Portland they were bioswales; both were installed in 2005-2006. In Bristol, the area was the recipient of a government improvement grant in 2004, which included the installation of permeable paving (see Figure 2). It is close to the city centre and almost entirely residential. The area has an active residents' association and won an award several years ago for the strength of its community activities. Shortly before the permeable paving was installed, the association won another grant to restore a large rubbled area to structured parkland, further increasing water absorbency capacity.
Portland's Bureau of Environmental Services (BES) has been developing sustainable stormwater management facilities since the early 1990s (Reinhardt, 2011). In 1999, the city's stormwater management manual (SWMM) was officially adopted (BES, 2005), and a city-wide 'Green Streets' programme was  (BES, 2007). The stormwater management and 'Green Streets' programmes focus on reducing street flooding and filtering rainfall before it reaches the Willamette and Columbia Rivers, which run through the centre and north boundary of the city. The current programme requires all street improvements to accommodate stormwater management techniques. Part of this programme involves installing bioswales within, or as extensions to, pavements (see Figures 3 and 4) (BES, 2014). In Portland, the study site was located 5 miles (8.05 km) from the city centre along a main street; bioswales were installed when the road was tarmacked and pavements installed.
A semi-structured interview approach (Wengraf, 2001) was adopted to allow respondents to talk freely around 12 key questions, with the interviewer to follow up on main points of interest. Interviews were designed to last 35-45 min. Respondents were informed of the nature and purpose of the research, asked to provide consent acknowledging that anonymised quotes may be used, and assured they could withdraw consent at any time. The Portland State University Office of Research Integrity and University of the West of England Faculty Research Ethics Committee provided approval for use of human subjects in the USA and the UK, respectively (IRB#143039 and FETREC 1314-15).
Potential respondents were initially contacted by attending a residents' association meeting in the UK, and by distributing A5 flyers to households in Portland. In Portland, response to flyers was very poor, and so the researchers adopted an alternative approach and spoke with people outside their houses. These point-of-opportunity interactions tended to be somewhat shorter, but still achieved a mean average of 26 min and proved equally fruitful in terms of the quality and range of data gathered.
Nine individuals were interviewed at each site. Responses were transcribed and analysed using qualitative data analysis software (NVivo). A grounded theory approach was used whereby theory is developed into themes from data collected, presuming as little as possible beforehand (Bryant and Charmaz, 2010;Glaser and Strauss, 2009). The purpose of this research was to investigate individual perceptions and behaviour in depth,

Awareness and understanding
Across both sites results supported the work of Apostolaki and Jefferies (2005) and Bastien et al. (2011): awareness of the existence, and/or purpose and function of Suds tended to be quite limited. A clear lack of awareness that any drainage system had been put in place was in evidence in Bristol, as one respondent commented When you came to our [residents'] meeting … I think it was only me and _____ who had any idea what you were talking about! (respondent no. 11.01.05.2) People were aware the road surface looked different to other roads (see Figure 5), but often had no further insight as to its purpose. Responses suggested the presumed function was to distinguish the area from surrounding streets.
The really brilliant thing about it is that if you are driving down the road, you think 'oh, can I go in there?' … It totally discourages people from driving through the neighbourhood. (respondent no.

11.01.04.2)
A respondent who was new to the neighbourhood said I am [aware] now! Only since we were told about it. I mean, until the residents meeting … I wouldn't be surprised if a lot of the people who lived here when it was put in didn't actually realise it served a drainage purpose. (respondent no. 11.01.04.1) Similarly, a longer-term homeowner responded I see that these streets, they look different to other kinds of streets, but before you came to ask us, I didn't really take any notice of the draining system … what does it, it's the water, is that right? (respondent no. 11.01.02.2) As respondent 11.01.05.2 above notes, those who were involved in negotiating the street improvements tended to be more aware.
I do know about the Suds because it came in when I was here … we were consulted on what went in … So I'm aware that we haven't got flooded when everyone else got flooded. (respondent no. 11.01.05.2) However, a resident who had been around since before the changes but not involved with the residents' association was unaware.
I thought it was mainly just to be flat for people, for wheelchairs, pushchairsmaking it safer. (respondent no. 11.01.01.2) Responses suggest that, unless people had been involved from the beginning or spoken with someone who had, they were likely to be unaware the road was permeable. They would be aware the streets looked different, but not about design function.
In Portland, the bioswales are a more obvious intervention, large visible entities taking up part of the pavement, or acting as an extension to the pavement (see Figures 6 and 7). Awareness of their existence was consequently much higher than in the UK study. Awareness of function was correspondingly somewhat higher, but still not good. For example, two respondents were unaware of their role in water management.  Other respondents were aware of their advertised role in cleaning the water system, but either misunderstood how this worked or did not believe it could function. The bioswales were installed to reduce the amount of stormwater both backing up on the streets, and infiltrating the city's combined sewer system (reducing overflows of untreated sanitary sewage and stormwater into watercourses and greatly improving water quality) (see Figure 8). They also filter pollutants out of stormwater flows (Dingman, 2008;Peters, 1994), further improving water quality downstream. Some respondents were aware of this, with caveats. Weak awareness of purpose and function may affect satisfaction with systems put in place, as well as awareness of appropriate behaviour that could affect functioning.

Understanding and behaviour
The functioning of the Suds systems will be affected by behaviour around them: littering could inhibit water flows, while changes to Suds' vegetation could impact water-cleaning functions. From the authors' interactions, it seemed that some people did not understand how behaviour could affect functioning. First, in Bristol, residents the authors spoke to were surprised to learn about oil's potential effects (oil spillage could impact permeability).
People don't really take care of it because they don't really know … if oil drops on it, you just think 'oh, never mind,' instead of thinking 'oh, we've got to do something about this' … in terms of behaviour, maybe tenants could be informed a bit more. (respondent no. 11.01.02.2) Weeds were another behaviour issue. Almost 10 years' postinstallation, weeds were appearing in soil deposits that had become embedded between the bricks. Best practice would be to carefully scoop out the soil and weeds, but people were unaware of this due to a lack of maintenance information. One resident referred to filling the loose permeable grit between the bricks with cement because he thought he had observed a similar action during installation. Thus, the importance of people's understanding of Suds and potential behavioural impacts will help ensure continued functionality as regards flood risk management and water quality treatment.

Visual appreciation
In Bristol, most residents appreciated the permeable paving simply for its contrasting look to tarmac, suggesting a more 'private' zone, as noted above.
Aesthetically it looks very pleasing, much more pleasing to the eye than a normal street … just the way the bricks have been laid out is quite neat and tidy. ( Therefore, the aesthetic contributions of Suds might be valued differently based on individual perceptions of specific neighbourhood characteristics.

Voice and consultation
There was general support in Bristol for suggestions of future efforts to inform people about the paving and its required maintenance. Some Portland citizens who were more aware of bioswale functions also expressed concern that the city did not do enough to educate others about their structure and function. Others said they had never received any correspondence or invites to negotiation over installation. Informing and educating is a two-way process: Suds designers will need to learn what people want or can cope with and help maintain (or at least not adversely affect) if installations are going to be sustainable and viewed as a positive contribution to stormwater management. Consultation upon solutions, and then on-going dialogue to ensure a continued sense of involvement and ownership, might improve engagement, adoption and so sustainability.
The trees placed in certain bioswales in Portland demonstrate this well.
They look nice, they make the neighbourhood look nice … I think they're goodand the trees too, they look nice. (respondent no.

01.01.02.2)
They're nice, 'cos they've got the little trees coming out of them, and the little bushes, yeah, they're nice … they're just beautiful.
These trees … I told them I was going to put battery acid on them, Since trees were not placed in every bioswale, pre-installation community engagement could have helped to overcome this inconsistency of opinion (see Figure 9). A sense that services were being tailored to needs and desires could encourage engagement and appreciation.

Opinions changing
The process of installations coming to seem more 'normal' and accepted over time would not appear to hold from the interview samples. Those who understood their purpose tended to be initially supportive, and this did not wane in Portland or Bristol. However, interventions or changes since installation could shift attitudes in both directions. For some, appreciation was increased by learning about water-functions.
Well now I learnt what they're for, I think they're even better than I did! I really liked it when I moved in … And now I know it actually serves a purpose, it's even better! (respondent no. 11.01.04.1) When they first came, I didn't have any idea what they were about! … you saw them coming up all over town, big ditches on the street. Whereas for others, in Bristol, opinions about installations worsened over time for some, due to poor behaviour. The paving on one street was lifted 6 months after installation for utilities work (water or gas pipes) and not correctly replaced, creating puddling that a number of respondents commented upon (see Figure 10).

Maintenance
Maintenance is an important issue because beyond behaving appropriately, both the Suds and their responsible local authorities could benefit greatly if local residents were engaged to help with minor maintenance tasks. With both permeable paving and bioswales, weeding and cleaning or removing litter would help functioning. However, many people did not support their involvement with this.
It's the council's responsibility, let them c**k it up themselves.
(respondent no. 11.01.04.1) Figure 9. The standard information notices placed in bioswales A principal concern for many people was that they should not have to pay more in labour or money for Suds maintenance; in the UK it was felt that council tax should cover such provisions, whereas in Portland, water bills were felt to be high enough already. Willingness to pay for or contribute to maintenance will be affected by perceptions of the benefits of Suds systems. This will be related directly to awareness of risk potentialities in both people's own, and neighbouring, communities.

Risk perception
In Bristol, residents the authors spoke to were confident that the area would not flood. With Portland, a couple of respondents were aware of the city's dependence upon the Mount Hood snowpack for part of their water supply. Overall, however, awareness of both flood and drought risk seemed to be dependent upon a heuristic risk evaluation, whereby recent experience and partial knowledge is substituted for 'official' and scientific estimates (Tversky and Kahneman, 1974).

Discussion
The findings from this research serve to confirm the findings of Apostolaki and Jefferies (2005)  that awareness and understanding of Suds' water functions (and indeed their existence) can be low. Education, consultation and clear communication between water management groups and residents are important for raising both awareness and satisfaction, and the quotes in Section 3.3 show that a number of residents were keen for this to happen. Both bioswales and permeable paving were for the most part appreciated for aesthetic or environmental reasons; when water functions were explained, appreciation markedly improved for some respondents. From this small sample, improvement in perceptions of the Suds over time since installation was not observed and, for a few people, opinions actually worsened due to the behaviour of others (such as littering and the mishandling of the permeable paving). The opinions of those who appreciated them improved upon learning about water functions; the opinions of those who did not appreciate them in the first place did not appear to alter. The question remains, if water functions had been explained from pre-installation, whether impressions may have improved over time.

Conclusion
This study has built upon others' findings to investigate public awareness, understanding and preferences around two different Suds systems. The study's main findings point to a lack of awareness and understanding of Suds function at both sites.
In the cases studied, more consultation with residents could have helped to develop greater consensus around the approaches adopted, an understanding of the Suds' operation and an awareness of appropriate behaviour to encourage optimal functioning. Tailoring Suds to community needs could increase the perceived benefits and so willingness to be involved with maintenance; lay maintenance could reduce installation life-cycle costs. Pre-and post-installation engagement could therefore reduce overall costs and help to ensure longer-term functionality. Results suggest that continuing education would be needed to maintain community knowledge, as house occupation changes over time; designing facilities to show their function more clearly and with clearer site information could help to reduce these education costs.
More research is needed to continue deepening the understandings outlined here, working with different sectors of the public and types of Suds. The paper's findings would appear to be widely transferrable, but further research may highlight other considerations. Because of the small sample size, observations noted here should not be interpreted as overarching generalisations, but rather as time-and place-limited findings from two specific case studies.