Introduction

Forests provide a range of important ecosystem services, including the provision of clean water (Ellison et al. 2017), which is of growing importance in an era of rapid climate change (Furniss et al. 2011). Past research has shown that forest management can enhance ecosystem services and improve forest health by reducing wildfire risk, reducing invasive species spread, mitigating climate change, and facilitating healthier stands of trees, among many other benefits (Hirsch et al. 2001; Muzika 2017; Oliver et al. 1994; Ontl et al. 2020). However, some aspects of forested land management can impair the health of downstream water bodies, particularly when forest best management practices (BMPs) are not followed (Anderson and Lockaby 2011). Specific BMPs designed to protect water quality may differ depending on location, but generally include maintaining sufficient riparian buffers during timber harvesting, appropriate drainage from logging road surfaces, appropriately sizing and aligning culverts and temporary bridges, as well as other similar practices (Wilcox et al. 2019). Understanding forest management decisions and what factors influence an individual’s willingness to implement BMPs is critical for maintaining water quality and high level provisioning of ecosystem services.

Decisions about BMP implementation on public land are often made by foresters or other conservation professionals, but decisions about BMP implementation on privately owned forestland are likely to be made by the individuals that own those forests. These private forest owners control about 34% of the forestland in the U.S. and about 55% of the forestland in the Northern U.S., making their decisions around forest management particularly important to understand (Butler et al. 2021; Butler and Ma 2011; Liknes et al. 2010). Many of these private forest owners are considered family forest owners (FFOs), defined as individuals, families, and other types of unincorporated groups that own forest land (Butler 2008). Many FFOs work with consulting conservation professionals, such as foresters, to assist with management decisions. Previous research has demonstrated that trust in these conservation professionals can impact a FFO’s decision to adopt specific management practices (Ambrose-Oji et al. 2020; Deuffic and Ní Dhubháin 2020; McFarlane et al. 2012). Studies assessing FFOs’ perspectives on stream water quality have found that clean water is seen as an important goal by landowners across demographic and political spectrums, although it is rarely a primary goal for management (Dutcher et al. 2004; Evans and Jensen-Ryan 2017; Mancheva 2018; Wiskind 2003). Further, many landowners often do not report seeing a connection between their management activities and clean water (Dutcher et al. 2004; Mancheva 2018). For natural resource managers seeking to promote the adoption of BMPs, several studies have shown that riparian forest incentive programs intended to protect water quality – and to confer benefits to the landowner – are generally viewed in a more positive light than regulations (Butler et al. 2022; Dutcher et al. 2004; Evans and Jensen-Ryan 2017; Kooistra et al. 2018; Mancheva 2018). Research also suggests that landowners are often more comfortable with regulations when these regulations are perceived as not imposing unwanted restrictions on landowner behavior, and perceived to be enforced uniformly and fairly (Quartuch and Beckley 2014). One study found that landowners often were not aware of existing water quality regulations, and that they were more likely to be opposed to changing management practices if they perceived that there was something to gain economically from not making changes (Kooistra et al. 2018). In summary, FFOs have been shown to care generally about water quality, but are more reluctant to implement changes if there is a perceived financial constraint, or if there is doubt about the connection between changing a management practice and overall water quality improvement.

Like water quality concerns, climate change is often viewed by FFOs as an issue that does not connect to their forest management actions and is rarely incorporated into management activities intentionally, although there is evidence of incidental adoption of climate change resilient management practices (Boag et al. 2018; Grotta et al. 2013; Hengst-Ehrhart 2019). For example, some factors that have been cited by forest landowners as limiting their willingness to manage for climate change are a lack of place-based educational opportunities regarding management strategies (Boag et al. 2018; Grotta et al. 2013) and the perception that the impacts of climate change were years or decades away (Grotta et al. 2013; Hengst-Ehrhart 2019). Furthermore, even when FFOs are willing to change their management behaviors, large landowners often find implementing adaptive management for climate change to be both cost and effort-prohibitive on a property-wide scale, resulting in less robust changes being made on a limited number of forest stands (Hengst-Ehrhart 2019). Taken together, the existing studies on forest landowner perspectives toward managing for water quality and climate change indicate attitudes ranging from slight interest to ambivalence, which has not translated into widespread, targeted management actions. To this end, research on what drives FFOs who are actively managing their land for either water quality or climate change to do so could be particularly useful for increasing the adoption of forestland BMPs.

Fig. 1
figure 1

Map of Vermont counties (Franklin, Chittenden, Addison) in the Lake Champlain basin where forested land owned by interviewees was located

Full size image

The Vermont Portion of the Lake Champlain Basin

This study focuses on the Vermont portion of the Lake Champlain Basin in the Northeastern United States, an area where basin-wide, non-point source water quality challenges converge with significant forest land cover (Fig. 1). 78% of the state is forested and sits at the border of two forest systems where hardwoods (maple/beech/birch) meet softwoods (spruce/pine) (VT DFPR, 2022). Of the forested land in Vermont, 80% is privately held, while the remaining 20% is publicly held. The forests in Vermont provide many services, including the production of wood products, tourism and recreation, air and water filtration, carbon sequestration, flood protection, wildlife habitat, and the protection of biodiversity (PSC & Frederick 2020; VTDPFR, 2022). They also play a critical role in protecting the water quality of Lake Champlain, which currently does not meet Clean Water Act goals (U.S. EPA, 2016; Zia et al. 2022). While phosphorus (P) contributions from agriculture, urban lands, and stream bank erosion are the primary drivers of non-point source impairment in the lake, forested land is estimated to contribute roughly 11% of the total P load, a product of the forest’s low loading rates but high spatial coverage. The estimated reduction in loading from forest land for most lake segments need only be modest (5%) to meet targets set by the US Environmental Protection Agency (U.S. EPA, 2016); however, in the most heavily impaired lake segments, phosphorus reductions of 40 to 50% from forested lands are required.

There are a number of existing policies and programs in Vermont that encourage forest management practices that support improved water quality outcomes. For example, a set of voluntary “acceptable management practices” for timber harvesting have been in effect since 1987, and were revised in 2018 (VT DPFR, 2018). These guidelines are for practices such as constructing and removing temporary bridges, disposing of logging slash, smoothing and mulching skid trails/temporary roads after project completion, and leaving a buffer between streams and an operation (Wilcox et al. 2019). There are also opportunities for landowners to help preserve ecosystem services through conservation easements, and through Vermont’s Use Value Appraisal Program, also known as “Current Use.” Current Use allows landowners with more than 25 acres of forested land to apply for a tax break, as long as they maintain a forest management plan and use their land for harvesting timber (Vermont DFPR, 2010). This program ensures a base supply for the state’s timber industry while also improving forest health, since landowners must work with a professional forester to develop a forest management plan in order to maintain enrollment. Although Current Use facilitates interactions between landowners and foresters, these interactions tend to be infrequent and transactional (E. Tapper, personal communication, January 3rd, 2024).

To better understand the impact of existing programs and to inform effective outreach and policies, more work is needed to understand if individual forest owners in the basin are aware of the impact their actions can have on water quality, as well as to identify barriers to BMP implementation. We sought to address this gap by asking the following research questions:

  1. Q1.

    How do landowner attitudes and beliefs about water quality, quantity, and climate change influence their forest management decisions?

  2. Q2.

    Which factors are associated with family forest landowners’ moving past the precontemplation stage of adopting BMPs during timber harvesting?

Theoretical Framing: Transtheoretical Model of Behavioral Change and Trust

To address these research questions, we used the Transtheoretical Model of behavioral change (TTM) as a theoretical frame for our work. We also examined trust as a necessary precursor to the processes of change outlined in the TTM. The TTM is a stage-based model that has been used by past researchers to identify why individuals make or do not make behavioral changes over time. Originally proposed by Prochaska and Norcross (1979) and revised by Prochaska and DiClemente (1983), this theory asserts that behavioral change occurs during five sequential stages: precontemplation, contemplation, preparation, action, and maintenance, along with a sixth stage of relapse (adapted for non-health topics as “recidivism”) that applies when an individual reverts to a prior stage. During the precontemplation stage, individuals have yet to consider changing their behavior. In contemplation, individuals begin the initial stages of considering whether they will make a behavioral change, while the preparation stage is the first point in which individuals intend to change their behavior and begin planning for this change. In the action and maintenance stages, individuals make the behavioral change or sustain it, respectively. Although this pathway is sequential, individuals will sometimes stagnate at a specific stage or move back stages as they progress toward behavioral change. The TTM is a useful framework for examining individual habits, long-term decision-making, and interventions targeted at habits and decision-making (Prochaska and DiClemente 1983). The framework has been used widely to assess behavior change for health interventions, including substance abuse and physical activity (Hutchison et al. 2009). It has also been used increasingly in environmental contexts, such as to assess interventions for promoting carbon footprint reduction and the adoption of agricultural best management practices (Bamberg 2013; Doran et al. 2022; Gatersleben and Appleton 2007; He et al. 2010; Mair and Laing 2013; Quartuch et al. 2021). Despite increasing use, the TTM has been criticized for a variety of reasons, including the arbitrariness of its partitions between stages (Bamberg 2013). Nevertheless, it provides a useful lens for thinking about what drives individual behavior change.

As individuals progress through the stages of change, cognitive, affective, and evaluative processes – known as processes of change – influence the individual’s progress (Prochaska and DiClemente 1983). These processes fall into two groups: (1) experiential processes, and (2) behavioral processes. Examples of the former include consciousness raising (efforts to seek new information) and self- and environmental re-evaluation (assessment of the behavior with respect to the individual’s values and to their physical/social environment, respectively). In contrast, reinforcement management (reward by oneself or others for making the change) is an example of a behavioral process, as is helping relationships (accepting the support of others during behavioral change). Social support is a term that is sometimes used interchangeably with helping relationships (Quartuch et al. 2021), but has also been used to differentiate between a helping relationship with a member of one’s social network versus a professional (Gibbison and Johnson 2012). Some processes of change are considered more relevant to specific stages of change than others (see Fig. 2). Experiential processes can be promoted through education and communication campaigns, while behavioral processes may also require the support of broader environmental conditions and social networks/capital.

Additionally, self-efficacy – the degree of confidence or skill an individual has in making behavioral change – is also important in progressing through TTM stages (Bandura 1977, 1982). As individuals move through each successive stage, self-efficacy is generally thought to increase. High self-efficacy has been found to lead to behavior change in TTM studies and other decision-making models, as well as in multiple environmental contexts, including forest legacy planning (Doran et al. 2020, 2022; Quartuch et al. 2021; Richens et al. 2018).

Central to the utility of the TTM is the idea that individuals within each stage of behavioral change will respond more readily to interventions tailored to their needs at that stage, as opposed to a one-size-fits-all approach (He et al. 2010; Velicer and Prochaska 2008). The assessment of behavior change interventions by TTM stage can lead to useful stage-specific conclusions. For example, Gatersleben and Appleton (2007) assessed the perceptions of bicycle commuting held by people in different TTM stages, finding that people in the precontemplation stage can be brought closer to behavioral change just by increasing their awareness of the positive impact bicycle commuting can have on emissions (the process of change known as “consciousness raising”), while different strategies were needed to message toward individuals in other stages of change. Similarly, other studies have found that many interventions intended to encourage environmentally responsible behaviors are primarily reaching individuals already engaging in behavioral change, potentially preventing recidivism in these populations but not reaching individuals in early stages (He et al. 2010; Mair and Laing 2013).

In their early work on the TTM, Prochaska and DiClemente (1982) note that a trusting relationship with a professional is a precondition for any successful intervention. Behavioral interventions necessarily involve communication about a desired behavior, and trust in a communicator (e.g. conservation professionals) allows individuals to act without in-depth knowledge of a given topic (Brossard and Nisbet 2006). Trust can be thought of as a psychological state in which an entity (a trustor) accepts some level of vulnerability (i.e., risk) based on a positive expectation of another entity (a trustee) (Hardin 2002). Stern and Coleman (2015) identify four forms of trust relevant to collaborative natural resource management: rational trust, affinitive trust, procedural trust, and dispositional trust. Rational trust is based on the likelihood of a positive outcome, and stems from past performance; affinitive trust is rooted in shared values and emotional affinity for a trustee; procedural trust is engendered by the presence of control systems – rules or procedures – that protect the trustor; and dispositional trust reflects a trustor’s baseline tendency to be either trusting or skeptical. Distrust – the expectation of a negative outcome of accepting risk – is also multidimensional and has corresponding antecedents (Coleman and Stern 2018a; Stern and Coleman 2015). We suggest that each of the different forms of trust (or distrust) are likely to be important precursors to the different TTM processes of change.

Lastly, while the TTM has been applied widely in other fields, few studies have used the TTM to explore behavior change in the context of forest management. One notable exception is Quartuch et al. (2021), who used the TTM to explore legacy planning intentions of FFOs in the US Northern Forest region. This study found that most forest owners are engaging in or contemplating legacy planning actions, but few are engaged in the advanced levels of this planning across all TTM stages. The present study expands on this limited existing research by applying the framework to better understand which factors lead to the adoption of water quality BMPs, and by examining the forms of trust that act as precursors to the TTM processes of change.

Fig. 2
figure 2

Source adapted from Doran et al. (2022); Prochaska and Velicer (1997); Mair and Laing (2013)

Stages and processes of change for the Transtheoretical Model of behavioral change.

Full size image

Methods

Using the TTM as our theoretical lens, we conducted qualitative analysis of semi-structured interviews with twenty-eight FFOs in Vermont. Interview subjects were recruited using landowner contact information provided by the county foresters for each county in the study area. All interviewees had a history of interaction(s) with a forester at some point; some interviewees regularly worked with a forester while others did so infrequently and only when required by a conservation easement or the Current Use program. For example, some landowners interacted with a forester every ten years, when they were required to update their forest management plan. Interviews were conducted in person on the property of interest or by phone. The interview protocol was developed based on questions asked in Version 6 of the National Woodland Forest Owner Survey, which is a paper survey conducted by the US Forest Service (Butler et al. 2021). The semi-structured protocol allowed for flexibility in the order that questions were asked, and allowed for follow-up questions. However, question topics were consistent across the interviews. The protocol incorporated the following topics related to our research questions:

  1. Q1.

    How do landowner attitudes and beliefs about water quality, quantity, and climate change influence their forest management decisions?

    • Questions regarding interviewees’ attitudes and beliefs about water quality and flooding.

    • Questions regarding interviewees’ attitudes and beliefs about forest management.

  2. Q2.

    Which factors are associated with family forest landowners’ moving past the precontemplation stage of adopting BMPs during timber harvesting?

    • Questions about the ways in which interviewees manage their land.

    • Questions about interviewees’ experience with extreme events such as floods.

    • Questions about from whom and where interviewees’ receive information about land management.

Qualitative analysis was completed in three phases. In Phase 1, advanced undergraduate students in a mixed methods research class were familiarized with academic content related to forest management, forest ownership patterns, water quality, and human behavior. These students also learned how to interpret peer reviewed articles and archival documents. Lastly, students received seven weeks of extensive training on mixed methods research, including constructivist research and qualitative coding. Under the supervision of one of the authors, the students developed a list of a priori qualitative codes, including concern for water quality and management strategies to address water quality. The students then analyzed the interview transcripts to first assign these a priori codes to the text and then create and apply emergent codes, including subcodes nested within the a priori codes. They also coded for emergent themes that were descriptive of ways that interviewees managed their woodlots. Each interview was coded by three students. To ensure intercoder reliability, the course instructor ran coding comparison queries in NVivo 12 software. Any instances in which there was less than 90% agreement between two coders triggered a class discussion about the most appropriate application of codes. Finally, the course instructor extensively reviewed all coding completed by the students. For example, there were several instances in which students coded text about the risk of extreme flood events as concern for water quality, and the instructor added a second code about concern for climate change.

In Phase 2, two researchers conducted multiple iterative rounds of coding using the TTM as a theoretical frame, adding the TTM stages and processes of change to the a priori set of codes. In this analysis phase, interviewees were assigned a stage within the TTM (precontemplation, contemplation, preparation, action, maintenance, and recidivism). If an interviewee demonstrated more than one stage, multiple codes were applied. Coding in this order (i.e., applying the theoretical frame after the students had already coded for emergent themes) ensured that themes that fell outside of the TTM framework were not overlooked. Lastly, the researchers ran coding comparison queries again, and instances in which there was less than 90% agreement between the two coders triggered a discussion about the most appropriate application of codes. For example, coding comparison led to a discussion between researchers about the length of time an interviewee needed to have adopted a given practice for to be assigned the maintenance stage. Relevant literature often describes individuals as being in maintenance if they have adopted a given practice for at least six months (Doran et al. 2022; Prochaska and DiClemente 1983). However, much of that work addresses actions that require frequent management, such as smoking. In the context of forest management, as with other natural resource decisions (Doran et al. 2022), six months is likely an inappropriate timeline, given that implementation of forest management actions happen far less frequently. Thus, we conceptualized the difference between action and maintenance as individuals who had newly begun to adopt a practice versus individuals who had been implementing a practice over an extended timeline.

During Phase 2 of qualitative coding, trust emerged as a salient theme outside of the initial a priori coding. As a result, a 3rd phase of qualitative coding occurred, which involved two researchers applying a trust theory framework by coding for each of the four forms of trust (and distrust) described above. For example, we coded discussions about trusting an information source (e.g. a forester) based on competence or capacity as rational trust. Similarly, we coded discussions about trusting an information source based on likeability as affinitive trust. As with the first two phases, the researchers ran coding comparison queries and instances in which there was less than 90% agreement between two coders triggered a discussion about the most appropriate application of codes.

Results

The majority of respondents (26/28, ∼93%) were over the age of 60. Six of the respondents (∼21%) were enrolled in the Current Use program, and one individual reported having a conservation easement on their land. Our results revealed that interviewees were managing their land with the following priorities in mind: maple sugaring, recreation, wildlife habitat, and water quality. While all of the interviewees had some history of interaction with a forester, fourteen (50%) interviewees reported that they worked especially closely with a forester when it came to managing their woodlot. These close relationships involved consulting with the forester multiple times per year about work done on their woodlots, such as erosion control measures or logging, and engaging with outreach efforts led by the forester, such as workshops and forest walks. These relationships contrasted with the other fourteen individuals, who interacted with a forester only when it was time to have their management plan updated or when they were harvesting timber. For many individuals the plans are revised every ten years, which limited the depth of the interactions with foresters. These fourteen individuals described their relationships with foresters as either neutral or distrustful. For example, one interviewee described a neutral relationship when they discussed the way they work with their forester:

Under the terms of the easement, we have to let them know in advance when we’re logging. And then when we’re all done, the county forester comes in and sees what we’re doing to make sure we’re up to doing it the way the state would like, which we are.

Another interviewee described a distrustful relationship between themselves and their county forester when they explained, “I feel like the [county forester] doesn’t want you to [harvest timber] yourself. They want a big contractor to come in and do it… and I feel they don’t trust me that I can do it myself.” Thus, our sample was split between individuals who worked regularly with a foresters and those who did so only when they were required to.

Q1. How do Landowner Attitudes and Beliefs About Water Quality, Quantity, and Climate Change Influence Their Forest Management Decisions?

Nineteen (∼68%) interviewees expressed some level of concern for water resources in Vermont. Those concerns fell into two overarching categories: (1) extreme weather events, and (2) water quality in lakes and streams. Many interviewees who expressed concern about extreme weather events referenced Tropical Storm Irene, an event that caused substantial damage throughout Vermont in 2011. For example, one interviewee stated, “After that storm I drove down [road name] every day and then I drove through that [road name] corridor where there was so much flooding…I think about how all those water systems are interconnected and how intense changes can happen quickly.” Another interviewee stated:

We did have a huge event, what they call Tropical Storm Irene. It widened and deepened that stream… rather substantially and that whole Central Valley part of Vermont was impacted very, very drastically… we’re very concerned about harvesting anything within what they call the Riparian zone, in river and stream areas. But it was pretty, pretty evident how catastrophic events… can literally alter the landscape.

Another interviewee expressed concern about increasing extreme weather events when stating: “And one of the things we’re seeing, we’re getting… big downpours, at least in eastern Chittenden County.” Interviewees who expressed concern about water quality articulated an awareness of the degradation of Lake Champlain and its tributaries. For example, one interviewee stated, “Every summer I help with collecting water samples from the Lewis Creek, and they get sent to the state lab and analyzed… we know that we have a range of pollutants in our beautiful state. So yeah, I’m concerned about that and about Lake Champlain.” These and other statements demonstrate participants’ general awareness of water resource issues in Vermont.

Twelve (∼43%) of the interviewees clearly articulated a connection between the way they manage their forest and water quality in Vermont lakes and streams. For example, when asked about why they made a specific harvesting decision, one interviewee stated, “I’m aware that trees stop erosion, and with erosion comes the carrying of pollutants and other things into lakes, particularly in Lake Champlain. So, property owners who have forests along the lake, if they were to cut down the tree, there would be more erosion and more stuff flowing into the lake.” When asked why they chose not to harvest trees on a certain hillside, another interviewee explained their desire to protect water quality this way:

Well, basically there is terrific runoff from the mountains coming down to Lake Champlain and all you need to do is look at the Winooski River now. And because we have much more significantly strong rains and things and more of them, it’s exacerbated… there has been a history of logging in all of the forests up there and when they log in the winter they historically have skidded the logs down to where they could be grabbed later in the year, and those skids are perpendicular to the hill. So they are wonderful routes for water to cascade down, And there are still lots of remains of those skids.

The interviewee went on to describe how they are leaving as many trees as possible on that hillside, in order to slow down the runoff. Less than half of interviewees were able to articulate a connection like this one, despite nineteen (∼68%) interviewees expressing concern for water resources.

Conversely, sixteen (∼57%) of the interviewees explicitly articulated that they did not see this connection. For example, when asked about the connection between water quality and the management of their own forest, one interviewee said, “Yeah, [water quality] is a lot less related to this particular forest, but it is certainly a part of Missisquoi watershed which leads directly into Lake Champlain.” Often, other interviewees would simply answer in the negative when asked if they believed the management of their forest had an impact on water quality; common responses to this question included “no” and “not really”.

Very few landowners (6/28, 21%) discussed specific strategies to manage their land for water quality. These few landowners discussed both the management of roads and trails and the implementation of BMPs during timber harvests. For example, one interviewee stated, “Last year we put in probably close to 40 water bars across some of the trails.” Another interviewee described their efforts this way:

I’m trying to work on the runoff areas that have been created by the road, in a small way because I don’t have big machines – I don’t have a lot of things to work with, but using stones and rocks and things that are available to try to mimic what they’re doing now on the state roads, which is to set up rocks and things that just slow down everything.

Even when landowners were asked about it directly, no clear trend emerged to suggest that experiencing extreme weather events impacted their management decisions. For example, when one interviewee was asked if experiencing extreme weather events changed the way they manage their forest land, the interviewee responded, “no I can’t say that it has.” In short, the majority of landowners’ attitudes and beliefs about water quality, quantity, and climate change did not, in our interview sample, appear to influence forest management decisions. Few interviewees were taking steps to manage their woodlots for water quality, despite overall concern about water quality.

Q2. Which Factors are Associated with Family Forest Landowners’ Moving Past the Precontemplation Stage of Adopting BMPs During Timber Harvesting?

Qualitative coding using the TTM revealed that at least one interviewee fell primarily into each TTM category, with the exception of “Maintenance” (see Table 1). Several interviewees articulated some elements of multiple stages of the TTM categories, however, all were strongly associated with only one stage, and thus were ultimately assigned to that primary stage. The TTM was applied in reference to adopting BMPs during timber harvesting, specifically constructing and removing temporary bridges, disposing of logging slash, smoothing and mulching skid trails/temporary roads after project completion, and leaving a buffer between streams and an operation. The majority of interviewees (15/28, ∼54%) fell into “Precontemplation”. Five (∼18%) of the landowners were in the “Contemplation” stage and could articulate actions that they would like to take in order to start managing their woodlots for water quality. For example, one interviewee in this category stated, “… I know that we need to do more about conserving riparian corridors because that’s a really good way of keeping… erosion from happening… and soil and phosphorus out of the rivers. So that’s a good step we can all work towards.” One (∼4%) of the interviewees was preparing to act, and thus was assigned “Preparation”. This interviewee described an erosion problem that they were planning to take steps to address. Six (21.4%) of the landowners were in the “Action” stage and had begun managing their woodlot for water quality. These landowners were the same six that articulated specific strategies to manage their land for water quality. For example:

…if you fell too many trees you allow too much runoff and you degrade the streams rather significantly. And so it’s really important not to harvest close to the streams.

We did not assign any interviewees to Maintenance, because none of our interviewees indicated that they had been managing their woodlots for water quality for an extended period of time. One (∼4%) of the interviewees fell into “Recidivism” by describing actions taken immediately after Tropical Storm Irene in 2011 to install water bars, but not acting again after this initial work. The interviewee stated that this work was conducted as part of their participation in a grant program. They explained, “[I had] a bunch of water bars put in on our road…then also at that time we did some TSI [timber stand improvement] work and that was also part of a grant… that’s kind of it in a nutshell.” The interviewee explained that they were not considering other actions (i.e., if they had not taken that action following Tropical Storm Irene, they would have been assigned precontemplation).

Table 1 Number and percentage of interviewees assigned to stages of the Transtheoretical Model
Full size table

Additionally, qualitative analysis showed evidence of three of the TTM processes of change. Specifically, interviewees expressed sentiments demonstrating consciousness raising, environmental re-evaluation, and helping relationships. Fourteen (50%) of the individuals described becoming aware of the connection between their land and water quality and were therefore designated as having undergone consciousness raising. Fourteen (50%) of the individuals had also experienced helping relationships, specifically from a forester who explained the connection between their parcel and water quality. Finally, nine (∼32%) of the interviewees discussed repeated consideration and re-consideration of the connection between forested land and water quality. These individuals were assigned the code of environmental re-evaluation.

The concept of trust emerged as closely linked with helping relationships during the second round of qualitative coding. The third round of qualitative coding revealed that rational trust specifically was consistently associated with helping relationships. For example, multiple landowners discussed trusting their forester because the forester had demonstrated sophisticated knowledge of ecology and management. One interviewee described how their forester had deep and extensive knowledge of the ecosystem and emphasized that by stating, “I do whatever [forester’s name] recommends.” Similarly, multiple interviewees discussed trusting information that came from a local conservation organization founded and run by a trusted forester; the interviewees described the organization in terms of their reputability and competence within the field of forest management. These helping relationships – and associated rational trust – were not concentrated among individuals in any specific stage of the TTM, and generally appeared to occur across all stages of the model that were represented in our interviews.

Discussion

Our work provides a valuable example of the application of the TTM in a forestry context. Previous work by Quartuch et al. (2021) used the TTM to examine legacy planning among FFOs, but the majority of TTM scholarship has occurred within the field of public health (Hutchison et al. 2009; Velicer 2008). Our study corroborates Quartuch et al.’s (2021) findings that consciousness raising is an important process of change in early TTM stages, that helping relationships are critical throughout the TTM stages, and that trust may be an important precursor for establishing those helping relationships.

Among our interviews, helping relationships were evident in the early, middle, and late stages of the TTM. This finding is consistent with results described by Gibbison and Johnson (2012) – who found that social supports – used by Gibbison and Johnson to describe helping relationships within an individual’s social network – were important for both initiating and maintaining a new behavior. This finding is also supported by one of the few studies to use the TTM in a natural resources context (Quartuch et al. 2021), where the authors found that helping relationships are important throughout the stages of forest legacy planning decisions, and especially in early stages of change. However, in the original framing of the TTM, helping relationships were primarily associated with later stages of change (Prochaska and DiClemente 1983). Our work provides further evidence that helping relationships may be critical throughout the stages of the TTM in a natural resources context, possibly through supporting an individual’s perceived self-efficacy or self-identity.

Additionally, rational trust was associated with helping relationships within our interviews. Rational trust involves a calculation about accepting risk based on the likelihood of a positive outcome and stems from knowledge of prior performance or perceptions of competence (Coleman and Stern 2018b). In our results, interviewees described following forester advice about management because those foresters were competent at their jobs, which suggests that rational trust may be an important precursor to helping relationships. This finding is partially supported by Quartuch et al.’s (2021) work, which also found that trust was often a precursor for helping relationships, but did not break down this finding into different forms of trust. As the role of trust in the processes of change was an emergent (rather than an a priori) theme in our work, we did not specifically investigate the connection between other forms of trust and the processes of change. However, we theorize that other forms of trust could be precursors to some of the processes of change, including helping relationships. For example, affinitive trust (i.e. trust built on social connectedness) is possibly a precursor for helping relationships, self-re-evaluation and social liberation, particularly when those relationships and processes occur across disparate groups of forest landowners. This suggestion is supported by previous research that demonstrates the importance of trust (including affinitive trust) for successfully conveying messages across organizational boundaries (Coleman and Stern 2018a).

Although the majority of interviewees in our study were generally concerned about water quality, their concern was not necessarily specific to their own parcel. Many of our interviewees expressed strong concern about the impairment of Lake Champlain but were unaware of how the management of their forest could support restoration efforts. This disconnect generally implies that there is space for outreach and education to facilitate such a connection, which would provide a pathway for moving landowners from the precontemplation stage to contemplation.

We also posit that helping relationships might be critical for consciousness raising (i.e., making the connection between one’s actions and water quality). Around half of our interviewees demonstrated consciousness raising, and those that did often reported the role of a forester (i.e., a helping relationship) as important for providing information. This provides further support for past studies (Gibbison and Johnson 2012; Quartuch et al. 2021) that suggest that helping relationships are important throughout the stages of the TTM, including during early stages of the model. We encourage future researchers using the TTM for natural resource questions to explore the linkages between the various processes of change, as this may serve to deepen the theory and its applicability within the field of forestry and beyond.

This potential association between helping relationships and consciousness raising – when coupled with the connection between helping relationships and rational trust – points toward an opportunity for messaging on the part of trusted professionals to FFOs about the connection between forest management and water quality, and the role that landowners can play in addressing water quality within their watersheds. Conservation professionals may be able to leverage the rational trust derived from their professional reputations to effectively communicate the importance of making this connection. We encourage foresters, University Extension agents, state or federal agency staff, and other consulting natural resource professionals to consider developing outreach programs that are designed to teach FFOs about the connection between forest management and water quality, the importance of this connection at both the parcel and watershed level, and to emphasize the efficacy of specific actions in addressing collective water quality problems. In some cases, natural resource professionals have already begun this work. For example, The University of Vermont recently launched a program called “The Watershed Forestry Partnership” which “facilitates research, communication, collaboration, and implementation of forest restoration and management practices that protect water resources in the Lake Champlain basin” (LCSG, 2022). This program, and those like it, has the potential to afford researchers the opportunity to evaluate the efficacy of outreach and interventions more fully in the context of the TTM stages and processes of change.

At the same time that we encourage education and outreach, we suggest that more research is needed to fully understand the specific types of education, outreach, and support that might be most effective for forest landowners in each TTM stage. Not all education and outreach programs are equally impactful (Ardoin et al. 2020; Heimlich and Ardoin 2008), and we posit that certain approaches are likely to have varying degrees of efficacy based on forest owner characteristics (Andrejczyk et al. 2016; Butler et al. 2022; Holt et al. 2021), and for addressing the relevant processes of change at play in moving forest landowners to later stages of change. While we demonstrate that consciousness-raising-based outreach from trusted sources could be effective, more research studying the effectiveness of outreach programs that are already in place is needed. Specifically, future research could seek to understand which types of outreach and education are currently being provided by foresters, and if this outreach is provided to all of the landowners they work with or just a subset. Evidence from other fields suggests that certain types of targeted outreach and education may serve to move landowners from the precontemplation stage to contemplation and action (Gatersleben and Appleton 2007; He et al. 2010; Mair and Laing 2013), but continued research is needed to understand how forest landowners in particular move through those stages.

Lastly, there is substantial opportunity for future research to explore how the structure of the TTM can be adapted and applied to forestry research. In particular, the traditional framing of the maintenance stage as “maintaining a change for more than six months” may be unsuitable, as many forestry BMPs could be perceived by landowners as one-time or rare practices – including culvert replacements, riparian buffer size increases, or the installation of road drainage infrastructure, as was the case in this study. Given that the TTM was developed to address daily health habits like smoking, more work is needed to understand how it applies outside the field of public health. Forest management projects happen on timelines counted in decades, so longitudinal data is needed parse out what constitutes “action” versus “maintenance”, as well as the role that self-efficacy and decisional balance play in long timescale practices.

Limitations

Our study was geographically limited, and therefore not generalizable. However, single case studies from specific geographic areas are important contributions because they provide the basis for later meta-analysis across multiple cases and because they can provide context about useful educational interventions to natural resource practitioners within the study area.

It is also important to note that we were not able to address all of the relationships that may influence landowner decision-making. While many landowners in our interviews mentioned both public and private foresters as a helping relationship that changed their understanding of water quality BMPs for timber harvesting, peer networks (Hamunen et al. 2015), family (Aguilar et al. 2017), and neighboring landowners (Clarke et al. 2021) have all been shown by previous research to have an impact on landowner decision-making. Despite its limitations, our study suggests that there is an opportunity for outreach that highlights the connections between forest management and water resources to educate forest landowners in precontemplation about the impact management choices can have on water quality.

Conclusions

Our research found that within the Lake Champlain Basin, FFOs are concerned about water quality and climate change, but often do not perceive a connection between their land and these issues – and act on reducing the impacts of their land management even less often. Further, our results point towards rational trust as a potentially important factor in encouraging FFOs to adopt forest management practices. From a management perspective, these findings highlight the need for outreach on the part of trusted professionals to increase awareness of the connection between forestland and water resource issues. For scholars, this work builds on the few existing studies that use the Transtheoretical Model of Behavioral Change in forestry contexts specifically, and environmental contexts more generally (Bamberg 2013; Doran et al. 2022; Gatersleben and Appleton 2007; He et al. 2010; Mair and Laing 2013; Quartuch et al. 2021). Outside of the United States, smallholder and family-owned forestland is a prevalent land type with a significant global body of research (Mayer 2019), highlighting the need for decision-making frameworks that can be applied globally. Future research should test the TTM in similar contexts outside of the U.S. to further explore its relevance. While there are limitations to the TTM as a theoretical framework, this study shows that it can be an effective way to analyze behavioral change in contexts beyond public health.