Shapeshifting attachment: Exploring multi-dimensional people–place bonds in place-based citizen science

1. Research on citizen science programmes has highlighted that they can foster science content and knowledge gain, enhance pro-environmental behaviour and cultivate civic action among participants. Especially in the case of place-based citizen science, which requires hands-on repeated activity in an out-of-door setting through a scientific lens, evidence suggests that some of these outcomes may be linked to the unique people–place relationships and interactions afforded by such programmes. 2. Even still, studies that empirically examine the influence of place on citizen science participant and programme outcomes are scant. This is due, in part, to the methodological challenges involved in interrogating complex aspects of a person's sense of place—aspects like place attachment—the emotional bonds between people and place. 3. Here, an adapted three-dimensional model of place attachment is proposed as a theoretical framework from which place-based citizen science experiences and outcomes might be empirically examined in depth. The model, which posits personal, social and natural environment dimensions of place attachment is contextu-alized with research findings from the US-based Coastal Observation and


| INTRODUC TI ON
Citizen science is a large and growing tent within the informal science learning community (Bonney et al., 2014). At present, millions of people in the United States alone participate in projects aimed at advancing science knowledge while creating a stimulating and meaningful experience for the participant (Sauermann & Franzoni, 2015;Theobald et al., 2015). Because definitions vary,  have proposed a typology of citizen science focused on key aspects of project outcomes and participant experience. In this schema, projects categorized as 'active citizen science' are those requiring the production of data or information through the actions and thinking of participants (as opposed to more passive forms of engagement where a participant might host an automatic sensor to detect seismic activity in their backyard, for example). While all of active citizen science affords the opportunities for learning, studies indicate that the majority of participants in online projects fail to persist past their first encounter (Sauermann & Franzoni, 2015;Segal et al., 2015), thus limiting learning possibilities. By contrast, hands-on, out-of-doors projects have significantly longer retention times, with most participants continuing past their first encounter and many persisting for years . In this sense, hands-on, out-of-doors citizen science offering participants the chance to learn about the practice of science and the natural history and ecological connectedness of their local environment through continued involvement is akin to the practice of natural history (Fleischner, 2005;Greene, 2005), and further still has roots in traditional ways of being with and knowing place practiced by Indigenous peoples.
A large body of research has catalogued the benefits of hands-on, place-based citizen science (for a thorough review, see Haywood, 2014a). Such outcomes have included enhancements to science literacy, knowledge and understanding (Brewer, 2002;Danielsen et al., 2005;Jordan et al., 2011;Sullivan et al., 2009); increases in 'scientific thinking' and the ability to interpret science information (Braschler et al., 2010;Kountoupes & Oberhauser, 2008); and growth in science skills like observing, measuring and recording data . We hypothesize that the impressive gains seen in many of these programmes may be because this type of citizen science engages participants in an out-of-doors place or geographically situated environmental phenomenon (e.g. the timing of plant budding, bird migration or first snow) that is personally attractive to them. The consistency of data collection over time in a place can help unlock the intricacies of place that hold meaning for the individual, deepening understanding via personalized experience that leads to the formation of a place-based bond. As such, we believe the concept of place is an especially relevant lens through which to interrogate and explain if, how and why engagement through citizen science might yield larger goals. One such 'holy grail' of place-based citizen science and ecological learning initiatives is conservation action and the related concept of ecological literacy, or moving participants from learning to 'application' in the real world (Ballard et al., 2017;Bela et al., 2016;Cooper et al., 2007).

| Place attachment as a dimensional concept
At the most basic level, place attachment (PAT) can be defined as an emotional bond between a person and a place. Place attachment research has provided insight and guidance for land management projects, community development and planning, research on tourism, and environmental stewardship initiatives by exploring attachment to residential spaces and areas (Jorgensen & Stedman, 2001), birthplaces (Nanistova, 1998), 'special places' (Eisenhauer et al., 2000), and tourist or recreational destinations (Vaske & Kobrin, 2001;Warzecha & Lime, 2001). Scholarship on place attachment thus provides a robust lens from which to consider the influence of place on citizen science outcomes. We contend, however, that the relationships between people and place that form or evolve through citizen science are distinct from these other types of interactions.
Although there are well-established examples of citizen science in the 'backyard', many citizen science projects take participants to places they do not own; and yet they are not 'visitors' exploring these places through seldom, singular or one-off interactions either.
Instead, their relationship with place is defined by loose, unofficial and more collective notions of place 'ownership'. While the activities through which citizen scientists engage with the places they study are somewhat self-directed (e.g. participation is voluntary; participants can select a study site), the scientific lens through which participants engage foregrounds both an opportunity to realize a place through patient measurement and observation, as well as put that particular place within an ecological or environmental context simultaneously provided by project data collected by others. In particular, many citizen science projects involve interacting with other individuals to gather, collect or discuss information. Thus, the motivations that drive and maintain repeated, purposeful interactions with a place through citizen science may differ from those driven by 6. Advancing this line of inquiry is an important component of broader efforts to understand how sense of place is altered via place-based citizen science and whether or not that is linked to specific programme outputs or participant outcomes in science knowledge, ecological understanding and civic engagement.

K E Y W O R D S
citizen science, place attachment, place scale, programme outcomes, sense of place residential affiliation or pleasure-seeking. Given these differences and the distinct bonds that form between people and place through citizen science, we propose an adaptation of one theoretical framework within PAT scholarship-the three-dimensional model first presented by Raymond and colleagues (Raymond et al., 2010)-to explore the role of people-place relationships in understanding the personal, societal and scientific benefits of citizen science.
Synthesized and developed into a model and assessment tool of the degree to which rural landholders are attached to their natural resource management region, the integrative three-dimensional PAT model of Raymond et al. (2010) defines attachment as occurring along three primary dimensions or 'poles': personal context, community context and the context of the natural environment; with five central constructs within those three categories ( Figure 1; Table 1).
We use the word dimension instead of pole in this paper because we believe it captures more holistically the nature of place attachment and the intersections among various aspects of the phenomenon.
Individuals may connect to place via one, two or all three dimensions, with the implicit assumption that increasing the number of expressed dimensions and/or the intensity of any/all of them connotes stronger PAT as a whole.
In this paper, we adopt the three-dimensional model as a foundation to guide exploration of place attachment among citizen scientists engaged in place-based interaction where place is not owned, and is divorced from concepts of home (Anton & Lawrence, 2014) or neighbourhood (Corcoran, 2002). Specifically, we posit that maintaining a multi-dimensional approach has the potential to enhance research on how the degree or intensity of attachment distributed among the three dimensions may facilitate exploration of PAT shape and intensity over time, including but not limited to: the existence of multiple, persistent shapes within a cadre of citizen scientists; and the degree to which shape intensifies through equal (or unequal) expansion along each dimension, that is, whether shape shifts.

| Questions of place attachment 'shape' and scale
While scholars can confidently determine if PAT exists, many questions about why, in what ways and to what effect still remain unanswered (Beckley, 2003). Attachment to place often involves common features like the existence of personally relevant meaning (Haywood, 2015), emotionally intense bonds (Manzo, 2003(Manzo, , 2005 and place-protective behaviours (Ramkissoon et al., 2012).
However, the multiple and interrelated reasons for such attachment, the unique character of those bonds and the disparate ways through which that may shape behaviour is less understood. Over the past decade, PAT scholarship has increasingly demonstrated the complex and idiosyncratic nature of the emotional bonds that form between people and place (Altman & Low, 2012). This leads us to explore two issues central to the intersection of PAT and citizen science: • Shape: While research has demonstrated the ability to determine the presence or absence of PAT, whether or not there are certain types or 'shapes' of PAT with uniquely distinguishing characteristics and impacts on behaviour and people-place interactions is an under-developed area of research. Further still, to what extent these hypothesized shapes are malleable over time is not known.
• Scale: As a phenomena where social, psychological and material realities collide, people-place relationships have inherent scalar F I G U R E 1 The three pole model of place attachment as originally presented in Raymond et al. (2010;Figure 1, p. 425), modified to incorporate the five constructs (place identity, place dependence, family bonding, friend bonding and nature bonding) also presented in Raymond et al. (2011;Figure 2, p. 326 Nature bonding TA B L E 1 A summary of the three-dimension categories and five major constructs as defined by Raymond et al. (2010, pp. 426)

Dimension category
Major constructs

Personal context
Place identity: 'Those dimensions of self, such as the mixture of feelings about specific physical settings and symbolic connections to place, that define who we are' Place dependence: 'Functional connection based specifically on the individual physical connection to a setting; for example, it reflects the degree to which the physical setting provides conditions to support an intended use'

Community context
Family bonding: Attachment based on a sense of belongingness or membership in a valued family group based on shared history or experience Friend bonding: Attachment based on a sense of belongingness or membership in a valued friend group based on shared history or experience Natural environment context Nature bonding: 'Implicit or explicit connection to some part of the non-human natural environment, based on history, emotional response or cognitive representation (e.g. knowledge generation)' components. But whether PAT exists-equally or at all-at various geographical scales is relatively unexplored, and is unknown when considering scaling within non-geographical constructs seminal to PAT and citizen science, including social and scientific dimensions.
Because we believe these two issues are particularly significant in the attempt to understand the role of PAT in citizen science, we provide a brief review of both, highlighting existing theoretical frameworks and scholarship that provides a foundation to advance understanding within these areas. We then introduce a citizen science project called COASST (Coastal Observation and Seabird Survey Team) which we use as a case study to explore PAT. Finally, we discuss four emergent research themes that we believe will advance scholarship on PAT in hands-on, out-of-doors citizen science by utilizing critical perspectives revealed through our adapted three-dimensional model framework.

| THE 'shape' of place attachment
Over time, PAT theory has diversified beyond frameworks that understand the phenomena largely as a personal need for connection and fulfilment to those that recognize the many intersecting components of people-place relationships. Williams and Vaske (2003) first developed and validated a two-dimensional PAT model focused exclusively on the personal experience of place that has been utilized widely in the field. Recognizing the complexity and multi-dimensionality of place attachment that extends beyond this 'personal' or individual dimension, several measurement tools now consider both the social and natural components of the phenomenon (Brehm et al., 2006;Scannell & Gifford, 2010a). Scholars more recently have advocated for models that integrate three major dimensions of PATpersonal, social and natural environment (Davenport et al., 2010;Gustafson, 2001). That is, scholars are now recognizing that PAT emerges not just from a personal need to identify with place but also because of the social and ecological connections formed in and with those places. In fact, Scannell and Gifford (2010b) (Kudryavtsev et al., 2012;Lewicka, 2011). While this approach allows examination of the degree to which PAT intensity (i.e. the index value) correlates with other variables, such as the presence of protective behaviour, a willingness to invest resources or effort in the place, or attitudes about the governance of that place; it does not provide a ready mechanism to measure the degree to which the relative strength among PAT dimensions might influence these relationships, nor whether intensity across dimensions varies in concert or independently. Because citizen science projects vary widely in structure (independent to large group data collection), focus (micro-organisms to macro system) and level of participant engagement (passive observation to active analysis of data), it cannot be assumed that the simple presence or absence of place attachment influences motivation to engage in citizen science or participant outcomes in a uniform way. Understanding the nuanced varieties of PAT shape is necessary to interrogate the potential relationship between those shapes and the diverse set of citizen science practices.

| The scale of place attachment
Geographical scale is inherent in the concept of place. However, with few exceptions (e.g. Cuba & Hummon, 1993;Hidalgo & Hernandez, 2001;Lewicka, 2010),   Eanes et al. (2018)). These studies of PAT scaling suggest that an individual's attachment expressed through geographical, social and natural or biophysical lenses is both variable and flexible, and may include or exclude places, people and processes from the immediate to the immense. Because citizen science can occur at the local (e.g. the Lost Ladybug Project) to the global (e.g. eBird) spatial scales, within singular person to population scales of social interaction and identity, and investigate micro-to macro-ecological systems and processes, understanding how PAT may differ 'at scale' may help understand the role of people-place relationships in citizen science.

| The COASST citizen science programme
Housed at the University of Washington, the Coastal Observation and Seabird Survey Team (COASST) is a 20-year-old citizen science programme featuring hands-on, monthly data collection on beaches throughout the Pacific Northwest and Alaska. Data collection modules include beach-cast marine birds and marine debris, respectively.
At present, ~1,000 individuals are actively collecting data. Most participants are non-experts and are local residents. New participants are recruited and trained locally in a single 3-5 hr, expert-led session, after which attendees are invited to sign up for the programme, select a beach of their choice and begin data collection. All participants are encouraged to survey with one or more partners.
Participant training encompasses learning how to demarcate and search the beach (referred to as surveying in COASST), and also how to collect basic data (referred to as evidence in COASST) from any beachcast marine bird carcasses/debris pieces found. In the case of birds, this evidence is then used to make a deduction about the taxonomic identity of the bird. Scientific skills used by COASST participants are aided by the use of specialized tools, including a dichotomous key to beached birds (a step-by-step identification process involving discrete choices about the features of the bird). Discovered carcasses are tagged (to prevent double counting), photographed and left in place.
Debris is bagged, bags are tagged and removed from the beach for post-survey examination. All data are recorded on paper datasheets, which are then digitized and sent to COASST along with photographic evidence. STEM content and skills are reinforced through the continued practice of monthly surveys on the same beach, programme feedback and subsequent learning opportunities, including a range of online and in-person follow-up (Parrish et al., 2007).
In contrast to some hands-on, outdoor programmes in which participants engage for only limited periods, the so-called 'dabblers' (infrequent or inconsistent participants) of Boakes et al. (2016), the majority of COASST participants maintain their involvement in the programme for years. Over 91% of training event attendees choose to sign up to be monthly data collectors; 78% of those go on to conduct at least one survey. Fifty-four percent are still in the programme after 1 year and just over 20% are still active 5 years later (Parrish et al., 2007). Reasons for participant tenacity are varied but appear to include belief that programme-scale data are significant to science and useful in a management or conservation context (Haywood et al., 2016). In fact, COASST data are widely used in scholarly publications and in natural resource management decision-making, relative to issues as diverse as historic use of seabird die-offs as a Native
In 2012  data were divided into new and seasoned participants, we also used these distinctions to investigate whether these populations displayed differences, suggesting that the practice of COASST on site may shift the intensity of PAT.

| A RE VIS ED PAT FR AME WORK FOR CITIZEN SCIEN CE
Our existing survey and interview data indicate that COASST participants feel connected and attached to the places they survey (Haywood et al., 2016), that individuals attach to place for a number of disparate reasons (Haywood, 2019), and that, at the population level, the intensity of a given dimension of attachment is differ-  Table 2). Aspects of residence and ownership are dropped, and a connection through science is added. Each pole is imagined as an axis along which the intensity of attachment could be measured, resulting in a PAT 'shape' indicated here by the red triangle. Shapes could thus apply to a person (e.g. an individual participant in a citizen science programme) or to a group (e.g. a cadre of citizen science participants) Personal pole place iden ty place dependence

Community pole friend/family bonding social rootedness
Natural environment pole nature bonding environment bonding science affinity familiarity or comfort with the place arising out of repeated visitation and/or multiple ways of knowing (Haywood, 2019;Haywood et al., 2016). He et al. (2019) found that both seasoned COASST participants and those new to the programme were motivated to join/ remain based on strong connections to the beach environment and to birds, and through their social connections to friends and family.
Taken together, these findings suggest that evidence exists for all three place attachment dimensions as articulated by Raymond et al. (2010). However, we also found that the five primary constructs included in the original model (Table 1) were not appropriate or sufficient to capture the range of COASST participant responses as they related to PAT within the context of their citizen science activity.
Therefore, and as detailed below, we propose a revised attachment framework for conceptualizing the most prominent constructs of place attachment within the context of out-of-doors, place-based citizen science (Figure 2; Table 2).

| Personal dimension
Survey and interview responses that demonstrated a direct connec- In summary, the constructs of place identity and place dependence appear to be sufficient at encapsulating responses of COASST participants within the personal dimension. While we acknowledge that COASST is only one of many out-of-doors citizen science programmes in which individuals have the opportunity to select a location and repeatedly collect data there, our data to date do not suggest that there are other response groupings beyond these two. As such, in our proposed revised three-dimensional PAT model for citizen science, we do not advocate major adaptations to the two main constructs originally described as defining the personal dimension.

| Community dimension
Survey and interview responses that demonstrated a direct con-  Q4: I'm really a big believer in community and to have all these people, all these various people, working on a project from so many different places-that is a community of people. I've met so many people in (name) county that do this, and they are great-some have even become friends. This is an odd thing that we have in common and so we are very different people but we are just a large family. That is very appealing to me. The community is a huge aspect of my connection to this place.
This seasoned COASST participant indicates that a portion of their place attachment relates to the sense of community felt at the survey site through programme participation, and uses the words community, friend, and family to describe this social connection. Given the difficulty we found in distinguishing between friend and family as separate constructs and the fact that these relationships appear to influence PAT in similar (i.e. positive) ways, we propose adapting the three-dimensional model for a citizen science context to combine separate family and friend bonding constructs into one integrated construct (family-friend bonding) that includes all place-based social connections with others who are present and specifically known to the individual (e.g. family, friends, community members, co-workers), as well as those who are real but imagined (e.g. other participants within the programme).
Finally, we identified a cohort of responses that align with the concept of rootedness highlighted by Raymond et al. (2010)  and some people are more connected to the science.
Other scholars have noted the significance of place some people feel because of connections to a cultural or ethnic heritage intertwined with that place (Billig et al., 2006;Hay, 1998;Mazumdar & Mazumdar, 2004); thus, we propose separating the construct of social rootedness within the community dimension as a distinct element of social bonding. We include in social rootedness a continuum of relationships from individuals once personally known to the participant who have since died (e.g. grandparents) all the way to past communities (personally unknown) to which the individual has a connection (e.g. ancestors or heritage).

| Natural environment dimension
Because the COASST programme is largely focused on natural science data collection in out-of-doors settings, it is not surprising that constructs of the natural environment were frequently identified among participants as key components allowing them to form strong attachments to place. However, we also found significant nuances in participant responses within this dimension, which we term natural environment. Included in this dimension was any response that demonstrated a direct connection/bond/relationship between the COASST participant and survey site, mediated specifically through some aspect of the environment, including the presence of a living or non-living feature or the ability to understand those features up through consideration of the system in its entirety.
For some COASST participants their sense of place attachment was mediated through connection to the biotic components (e.g. birds, wildlife, forest) of that place. We categorized these responses as nature bonding, the construct noted by Raymond et al. (2010) within the natural environment dimension of their original model.
However, we propose refining this construct to explicitly define nature to be some part of the living world other than people.
When asked to report on the meaning found at their survey sites, Finally, and perhaps most unique to the practice of citizen science, we found a significant portion of COASST participants indicated that the practice of environmental or ecological science was an important aspect of their attachment to place. As a citizen science programme soliciting long-term engagement in rigorous data collection by coastal residents, COASST is a priori connecting people to place through science. As a whole, seasoned COASST participants like the one below consistently ranked increased scientific learning and knowledge as one of the most frequently expressed outcomes of programme participation (Haywood, 2015). We propose adding one final construct under the natural environment dimension, science affinity, which we define as a way of connecting to place through the practice of science and more generally knowledge of that place (e.g. via observation, investigation, monitoring or learning). Haywood et al. (2016) found that feelings of place attachment were often linked directly to learning more about the ecology of the place (62%) and seasoned participants reported a greater appreciation for their beach ecosystem through repeated visits and the chance to collect data on the birds there. Science affinity is an aspect of the participant experience that seems to increase with participation: seasoned participants tended to identify themselves more strongly with science, including as members of the 'science team' (He et al., 2019). Furthermore, Haywood et al. (2016) found that 66% of the seasoned participants interviewed perceived the information they were collecting through systematic science was an active way to help protect or conserve their beach and, more broadly, the coastal environment. Such pro-environmental behaviour has been identified as connected to the natural environment dimension (Raymond et al., 2011).

| CON CEP TUALIZING PL ACE AT TACHMENT S HAPE AND TEMP OR AL CHANG E
Changes in place attachment over time have been robustly documented in the place attachment literature and are consistently attributed to variables believed to predict place attachment including residence length (Lewicka, 2005), strength of community ties (Scopelliti & Tiberio, 2010), home ownership (Brown et al., 2003), mobility (Gustafson, 2002) and the presence of certain physical features like green spaces (Bonaiuto et al., 1999). However, given the relative dearth of theory and empirical research on the multiple attachment dimensions and the relative interactions that occur among the constructs therein over time, not much is known about how attachment shape might change along a temporal plane.
Evidence from our research suggests that the shape of PAT is not fixed and may shift as a function of the citizen science activities and place-based experiences individuals accrue over time. He et al. (2019) and Haywood et al. (2016) both demonstrated significant differences in thinking, place interaction and identity between the population of individuals just joining the COASST programme-'new' participantsand those who had engaged in the programme for one to several years-'seasoned' participants. These findings suggest that either PAT shape changes as participants grow into seasoned COASST data collectors, some attachment 'shape profiles' persist in the COASST participant population longer than others, or, most likely, both. We should note, however, that while we believe there is evidence that the strength of attachment among any given participant may vary across the three dimensions highlighted here and that those differences may have real impacts on the evolution of people-place relationships over time and through citizen science, we do not wish to assert that these dimensions should be treated as fully distinct and separate from one another. There is undoubtedly overlap among all three. It is hard to imagine a strong place bond based on components of the natural environment (natural environment dimension), for example, not also influencing a person's sense of identity and belonging with that place (personal dimension). In fact, a strong body of literature exists that challenges such rigid distinctions (Bragg, 1996;Descola, 2013;Vining et al., 2008).
The degree to which PAT shape changes as a function of engagement in long-term citizen science activity, or more simply the possibility that the corps of participants could be categorized by shape along lived experience or other lines, remains unexplored. Finally, it should be noted that the vast majority of COASST participants most prominently hail from western cultural traditions, which undoubtedly influence their notions of identity and perceptions regarding the strength of and separation between the personal, social and natural environment components of a place (and see Quote 6 under social rootedness). As such, our data are not representative of non-Western cultures.

| CON CEP TUALIZING PL ACE AT TACH M E NT S C A LE
Our re-analysis indicates that scale is a crucial concept in PAT, not only defining the geographical or spatial extent individuals real- environment dimensions. We refer to these latter two concepts as social scaling and nature-science 'conceptual grain' (Figure 4).

| Spatial scale
A sense of geographical or spatial scale was apparent along all three dimensions, as highlighted by the excerpts above. At the local scale, COASST participants referred to their survey beach as a possession, as in Quote 3: 'This is my beach, and I think of it as mine'. perhaps akin to the scale of dwelling (Cuba & Hummon, 1993) or home (Anton & Lawrence, 2014 searchers have also found-not only does PAT exist at unique spatial scales but also that these various geographical scales of attachment can influence the nature and scope of 'active engagement in efforts to protect and improve… places' (Ardoin, 2014, pg. 439).

| Social scale
Similar to, but more expansive than Ardoin's (2014) 'cultural' scale of attachment, the object of attachment along the community dimension varied from single known individuals to cultural heritage. We recognize this as social scaling.
At the fine scale, friend and family bonding is most apparent, as in Quote 4, 'She is a wonderful person, … She has definitely become a friend. This is where we get to come together'. However, it was also clear that persons immediately known to the participant were not the only social bonds mediating connection to place. Quote 5 typifies the medium scale as a more conceptual description of persons imagined but not actually known personally: 'I'm really a big believer in community and to have all these people, all these various people, working on a project from so many different places-that is a community of people'. Here the social connections are via the corps of participants across the programme, not known to any individual participant, but imagined by some as people, like themselves, collecting data on their beaches in exactly the same manner and all towards a greater goal (Haywood et al., 2016). Thus, the social sense is of shared experience. The degree to which PAT differs by scale in spatial, social and nature-science arenas and whether or not that changes as a function of engagement in long-term citizen science activity may provide important clues about whether or not and/or how both the science produced through citizen science and the personal outcomes of those that participate (e.g. in the form of conservation action) can 'scale-up' from the hyper local context focused on fine-grain scientific processes to the regional or global context involving complex ecological systems.

| PAT PROFILE AND PROMIS ING RE S E ARCH DIREC TIONS
Our revised theoretical framework based on the three-dimensional model of Raymond et al. (2010), Raymond et al. (2011) arises from experiences of PAT within COASST participants, and provides a robust scaffold from which to develop methods that capture PAT nuance and dimensionality in out-of-doors, repeated activity citizen science.
We hypothesize that shape and scale are inherent parts of place attachment within the citizen science context, and may extend and/ or shift as a person deepens their experience of place via programme participation over time. To encompass the elements of dimensionality, intensity and scale, we propose the concept of a place attachment profile, literally the combination of Figures 3 and 4 with respect to any single person. We believe research on the attachment profile of citizen scientists may increase the potential to understand the ways in which PAT can facilitate and sustain citizen science engagement.
The ability to measure the place attachment profile among citizen science participants, as well as if/how that profile changes over time opens up a host of research avenues from which to explore both the impact of citizen science on participants and the personal dynamics of people-place relationships that may influence citizen science engagement and outputs. Four major research questions have emerged from our work on COASST and PAT that we believe may lead to important theoretical and applied advancements in the field.

Does place attachment help initiate engagement in citizen science?
Among environmental volunteers, Measham and Barnett (2008) have suggested that place attachment is a central motivating factor, connecting people to place via ethics of stewardship and conservation.
With 66% (46 out of 70) of COASST interviewees linking the process of data collection directly to conservation and stewardship of their survey site, programme engagement appears to be a way to protect a valued place for many participants (Haywood et al., 2016). However, there is little research on place attachment as a motivator to begin engagement in citizen science, where environmental ethics may, or may not, be primary to the work. Does a pre-existing attachment to a site motivate individuals to join citizen science programmes offering activities at that site? Does the attachment profile of an individual influence the nature of the programme they select, their data collection accuracy or the duration of their participation? Are particular profiles typical 'entry points' for citizen science, and/or for particular types of citizen science?
2. Does continuing participation in citizen science alter attachment profile?
As noted above, our work with COASST participants to date indicates a demonstrative, multidimensional and multiscalar attachment to place (Haywood, 2015(Haywood, , 2019, and a shift in the motivating factors and situated identity of participants as they move from who they were when they joined the programme to who they are years later as central members of the community of practice (Wenger, 1998)  Several studies have demonstrated that learners in hands-on citizen science projects can master the knowledge and skills needed to successfully perform project tasks at or above a level necessary for the project to deliver scientific outcomes (Dickinson et al., 2010).
Our research to date indicates that non-expert participants in the COASST programme not only learned the science content and skills needed to identify dozens of marine bird species with high accuracy but also that the monthly practice of surveying 'their beach' facilitated the development of species-specific mental models of beach-cast marine bird occurrence that statistically matched long-term regional averages (i.e. the baseline pattern), suggesting more developed critical thinking abilities (Char et al., 2014). Even still, we do not yet know whether the development of critical thinking skills are related to a priori demonstration of PAT-that is, a propensity to learn/think about a place-and/or to the concomitant development of a PAT profile.
Other studies have demonstrated a correlation between place attachment and environmentally responsible behaviours and attitudes (Budruk et al., 2009;Halpenny, 2010;Schultz, 2001), and specifically the link between the nature bonding dimension of PAT and environmental stewardship (Raymond et al., 2011). Our own work has demonstrated that engagement in the COASST programme can influence the sense of stewardship and environmental responsibility felt by participants (Haywood, 2015;Haywood et al., 2016), as is so apparent in Quote 9 (above). Within a citizen science context, could place attachment profile predict the likelihood that pro-environmental behaviour develops or how/when it is expressed?
4. Does participant place attachment profile influence programme outputs? Haywood et al. (2016) demonstrated that almost all (98%) COASST participants readily share their on-the-beach experiences (38%) up through programme-level information (36%) with others, expanding the reach and relevance of the information collected to broader programme-level outputs. Impressively, 20% of participants reported talking with resource managers, politicians or the news media about the data they collect and significance of that information, which has the potential to lead to much larger systemic change. In what ways might these larger programme outcomes be a function of the place attachment profiles of participants? Specifically, could PAT intensity correlate with the likelihood that participants will share information or interpret the significance of data collected for communicators and decision-makers?

| CON CLUS I ON: PAT PROFILE S IG NIFIC AN CE?
There is a growing evidence that active, place-based citizen science participants have a strong sense of place attachment (Haywood, 2019), are capable of learning and thinking at advanced levels for scientific analysis (Trumbull et al., 2000) and exhibit proenvironmental protective behaviours (Halpenny, 2010). However, it is not known the degree to which those abilities are related to each other, whether they grow/shift over time, and the specific dimensions of a place that are implicated therein. Data from the COASST programme highlighted herein demonstrate that place attachment is indeed multidimensional, occurring at multiple and disparate scales, with the potential to impact the citizen science experience.
In order for the informal science learning community to adequately consider the potential role of people-place connection in shaping citizen science engagement, participant outcomes and programme outputs a holistic model of attachment that moves beyond one solitary measure of the affective bond felt with a particular place and one spatial or social scale is required. We believe that using the revised three-dimensional place attachment model and concept of attachment profile presented here as a conceptual and theoretical framework will not only enhance understanding about the role of the citizen science learning environment and what diverse factors influence that process but also how place attachment combined with other experiential elements of citizen science at various scales shape outcomes and impacts. Finally, we suggest that inquiry into these hypothesized relationships has significant potential to reach beyond the practice of citizen science into multiple informal and formal learning environments to enhance both pedagogical and programmatic design and the assessment of critical metrics of science and environmental knowledge and application.

ACK N OWLED G EM ENTS
The authors wish to thank each of the dedicated COASST participants who engaged in this research. This material is based upon work supported by the United States National Science Foundation Advancing Informal Science Learning (AISL) program under Grant Nos. 1322820 and 2031884.

CO N FLI C T S O F I NTE R E S T
None of the authors have any conflicts of interest to declare.

DATA AVA I L A B I L I T Y S TAT E M E N T
Original research audio files, interview transcripts and digital survey responses discussed in this manuscript are kept in a secure cloud-based data management archive, with access limited to senior project personnel and project staff via a secure password and two-step verification. Original raw data cannot be shared without written consent of the research participants and a data use agreement. Aggregate data can be shared upon request, under the terms outlined in the original informed consent process.